Apparatus, systems, methods, and improvements formachine processing of language, for example such using sound / meaning associations

Abstract

Systems and methods evaluate the meanings of whole words bound within the meanings of their parts by analyzing and identifying features and components of words (such as letters and syllables), using sound information, for example phoneme class structure, to find sound correlation and using synonymity to find meaning correlation between words (or word parts) and a data set of known / identifiable sound / meaning associations called morphemes to find morpheme‑word‑parts that optimally describe an input word. Various means are described of compiling morphemes into a data set. The systems and methods further include analyzing the semantic functions of word parts and their modifying interrelationships, and using identifications made by one or more analyses. The systems and methods can evaluate or improve understanding of word meanings. The systems and methods are useful in researching sound / meaning associations in language generally and for improving language processing generally by the incorporation of more semantic information.

Description

APPARATUS, SYSTEMS, METHODS, AND IMPROVEMENTS FOR MACHINE PROCESSING OF LANGUAGE, FOR EXAMPLE SUCH USING SOUND / MEANING ASSOCIATIONSA portion of the disclosure of this patent document contains material which is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure, as it appears in the Patent and Trademark Office patent file or records, but otherwise reserves all copyright rights whatsoever.CROSS-REFERENCE TO RELATED APPLICATIONSThis claims priority to US application Serial Number 63 / 731,057 filed March 26, 2024, US application Serial Number 63 / 712,539 filed October 28, 2024, and US application Serial Number 63 / 770,509 filed March 12, 2025, the contents of each and all of which are incorporated herein by reference.BACKGROUNDModem civilization owes everything to three inventions arising out of ancient Greece: phonemes, morphemes, and logic. First, the ancient Greeks used symbols to record their own speech sounds, or “phonemes” (the Greek word cpcovf] means “speech sound, voice”) as letters. (The Greek phonemes are so cleverly conceived that it is thought that they can be used to record any human language.) The result — alphabetic writing — makes these sounds tangible, permanent, objective. Second, the Greeks used consistent sets of letters to record recurring spoken syllables, thus inventing morphemes (the Greek word popcpij means “form, shape”). Greek morphemes bind sound to meaning and enable concept formation, metaphor, and abstract thought. Third, ancient Greek thinkers from Homer to Aristotle consciously crafted these morphemes (phoneme class structures with distinct semantic values) into new words, thoughts, ideas. They made comparisons, distinctions, and connections, tethering one concept to another in a causal chain of logic (the Greek word yo means “reason, speech, word”). A profound corpus of literature, history, mathematics, science, law, and philosophy results from these Greek phonemes, morphemes, and logical reason. The Greek corpus enables flourishing civilizations — not only that of Greece (Athens in 5thand 4thcentury B.C.E.) but of Rome, Europe, and America. The Greeks themselves fully appreciated and analyzed the beautiful utility of their language toolset.However, during the Greco-Roman transition (and again in the Anglo-Norman period), ancient Greek morphemes were mis-heard, re-spelled, truncated, merged, and generallymisunderstood by laymen transcribing Greek into distorted Latin forms and into other new alphabets and languages. Add to these confusions the disadvantages of oral culture, for the many dark centuries intervening between the burning of the library at Alexandria and the Renaissance. Add to these conditions, today’s near-erasure of Greco-Roman culture. Is it possible that those same symbols, sounds, and meanings persist today, not only in Europe-derived languages, for example English, but perhaps even in languages only indirectly touched by those wide-ranging and seafaring adventurers? What if those Greek inventions, with the help of this one, could bring on another renaissance?SUMMARYExploration of the relationship between sound and meaning promotes further innovation in communication. Apparatus, systems, and methods are described for evaluating language using sound and also using both sound and meaning in the content of language, for example using digital apparatus, systems, and methods, for example digital computers and processors and their associated components. In some configurations, such apparatus, systems, and methods allow recognition and identification of concepts, metaphors and abstractions in language using such apparatus, systems, and methods. In some configurations, the apparatus, systems, and methods evaluate words and word parts to identify characteristics of the word and of its word parts, to better evaluate words and their meanings, for example based on the characteristics of the words and words parts, for example their semantic characteristics, among other things, discussed more fully herein. In one example, the apparatus, systems, and methods use sound correlation to evaluate words. In examples herein, such apparatus, systems, and methods receive input of words or word parts and evaluate the one or more words or word parts for one or more sound correlations, in preferred examples sound correlations with sound class structures, and in examples described herein sound class structures that can be revealed to have associated semantic values. For example, such apparatus, systems, and methods may receive input of one or more words or word parts and evaluate the one or more words or word parts for one or more identifiable morphemes, or one or more identifiable phoneme class structures with associated semantic value(s), providing for example a sound and meaning correlation. In such an example, evaluating a phoneme class structure can lead to associating a meaning with the input word or word part, for example where a data set includes sound and meaning information, such as can be included in a morpheme lexicon. In a further example, apparatus, systems, and methods receive input of one or more symbols, sounds, and meanings and evaluate one or more symbols, sounds, and meanings for one or more identifiable morphemes, or one or more identifiable phonemeclass structures with associated semantic values. Additionally, or alternatively, apparatus, systems, and methods receive input, for example in any of the forms described herein (including without limitation words, symbols, sounds, and meanings), and evaluate the sound and meaning of the input. Evaluating the sound and meaning provides the ability to recognize and / or identify one or more concepts, metaphors, abstractions, idioms, figures of speech, connotations, cognates, and / or other associations relating to or represented by the input, which can improve machine processing of language. In some examples, the apparatus, systems, and methods can evaluate the sound and meaning of words input there to. In a further example, the apparatus, systems, and methods can evaluate semantic and phoneme class characteristics of words input to the system.In any of the examples of apparatus, systems, and methods described herein, the apparatus, systems, and methods can be configured to analyze at least one text string, for example a word or part of a word, for example a word that may be from one or any of a plurality of languages. In a number of the examples herein, the at least one text string can include at least one unit representing an element of the text string, and in a familiar example the at least one unit representing an element of the text string can be a letter forming a part of a word. The apparatus, systems, and methods are also configured to compare the at least one unit to a class of data in a data set containing a plurality of such classes. In some examples, the comparison is to classes of data in a data set in which those classes of data are associated with units of a string, and in at least one example the unit in the data set is part of a morpheme, and the classes are phoneme classes, for example where each unit in the morpheme has a phoneme class associated with it. In some examples of an organization of the classes, members of such classes are non-overlapping, and in other examples some of the members may belong to more than one class, for example sibilants and aspirates may both include a letter H. In the present examples, each class contains representations of a selected characteristic of a plurality of units found in language, and a plurality of such classes contain respective pluralities of representations of characteristics of units found in language. In some examples, each class contains representations of sounds corresponding to a plurality of units found in language, and a plurality of such classes contain respective pluralities of representations of sounds of the units in language, in some configurations. In preferred examples, the selected characteristic of sounds are selected phonemes, each class is a phoneme class, and each phoneme class contains respective representations of phonemes. In another set of examples, the selected characteristic of sounds is pronunciation, or the physical manner of sound production, namely the position of one or more parts of the mouth, especially in relation to how or whether the breath is allowed to escape the oral cavity,and each class contains representations of pronunciations, selected according to a desired criterion. Other sound classes can be defined, additionally or alternatively.In any of the examples of apparatus, systems, and methods described herein, the apparatus, systems, and methods configured to compare described in the preceding paragraph can be configured to determine a desired correlation between at least one text string, for example the at least one text string compared in the preceding paragraph, with a string in the data set. In the examples described herein, the purpose of the comparisons is to find similarities between the at least one text string and one or more strings in the data set. In some examples, a desired correlation is identity, across a text string, between a class associated with a respective unit and a class of a respective unit of a string in the data set, for the entire string. For example, a desired correlation can be identity between respective phoneme classes associated with the text string and respective phoneme classes of a respective unit of a string in the data set. In another configuration, a desired correlation can be an approximate identity between the string of classes associated with the string of units in a text string, and the respective classes associated with units of a string in the data set. In a further configuration of apparatus, systems, and methods configured to determine a desired correlation, data representing whether or not a desired correlation was found can be saved, preserved, set aside or otherwise associated with the at least one unit or the at least one text string.In any of the apparatus, systems, and methods described herein, the apparatus, systems, and methods can be configured to analyze at least one text string having at least one unit representing an element of the text string and to compare the at least one unit of the at least one text string to a phoneme class of at least one unit of a string in a data set containing at least a plurality of phoneme classes. In some configurations, at least one phoneme class has at least two phonemes in the class. In a further configuration, there are at least nine phoneme classes, and, in another configuration, there are phoneme classes associated with the respective ones of vowel, labial, dental, velar, liquid, nasal, semivowel, sibilant, and / or aspirate. Other phoneme class differentiations can be selected, and, additionally or alternatively, other sound classes with respective sound differentiations can be selected / assembled and used as part of a data set for improved evaluation of language using machines.In some examples of the apparatus, systems, and methods described herein, the apparatus, systems, and methods can be configured to analyze at least one text string that represents a word part and / or a word. The word part and / or the word includes at least one unit representing an element, for example a letter, of the word part or the word, and the at least one unit is compared to a phoneme class of at least one unit of a string in the data set, wherein the data set contains at least a plurality ofstrings, each with associated phoneme classes. Where the at least one unit finds an equivalent in a phoneme class associated with a data set string, the phoneme class is noted, for example by being saved, preserved, set aside, or otherwise associated with the at least one unit, for example so that the information can be associated with the word part or the word, as the case may be. In the context of the examples described herein, finding an equivalent can be an exact match or an exact correspondence, or can be an approximate match, as determined by the desired criteria set by the user. For ease of discussion and presentation / explanation, “user” shall mean herein a designer, user, researcher, analyst, developer, curator, administrator, or other person or entity or machine or agent (for example, an A.I. agent), whether or not involved in designing, maintaining, hosting, operating, managing, executing, carrying out, or using the system, or processing or reviewing the results of the system, whether human or machine, or single or multiple or a combinations of such users, though not every type of possible “user” will be involved from beginning to end. In one or more configurations of the apparatus, systems, and methods described herein, any one or more of them may be configured to analyze a text string representing a word from one or more of a plurality of languages.In any of the examples described herein, including but not limited to the apparatus, systems, and methods configured for comparing at least one unit to a phoneme class of at least one unit of a string in a data set containing at least a plurality of phoneme classes, apparatus, systems, and methods are configured for comparing a letter to a letter unit of at least one string in a data set. In some configurations, at least one unit of the text string represents a letter, and the apparatus, systems, and methods are configured for comparing the letter to a letter unit of at least one string in the data set. In a further configuration, the apparatus, systems, and methods are also configured for comparing the letter to a phoneme class of a unit of at least one string in the data set, and in another configuration, the apparatus, systems, and methods are also configured for comparing a phoneme class to a letter unit of at least one string in the data set. Comparing a letter to a phoneme class or a phoneme class to a letter can be done in a number of ways, and in some configurations, the letter can have associated with it a phoneme class, a phoneme class representation, or a phoneme class designation, and the comparing is carried out with the phoneme classes or the representations of the phoneme classes. In another set of configurations, the apparatus, systems, and methods are also configured for comparing a phoneme class corresponding to the letter to a phoneme class of a unit of at least one string in the data set.In any of the examples described herein, the apparatus, systems, and methods can be configured for analyzing at least one text string having at least one unit representing an element ofthe text string, wherein the at least one unit represents a phoneme class, and further comparing the phoneme class to a phoneme class of a unit of at least one string in the data set. In some configurations, the data set includes phoneme classes, for example in the form of data representing a plurality of phoneme classes, and letters or representations of letters forming word parts and / or words. Also, in any of the examples described herein, the data set may include if desired morphemes or data representing morphemes. Morphemes are helpful to find connections between sound characteristics of words to find like words, which may be useful for example for words that may have diverged in spelling through language differences, custom, or pronunciation changes over time. Also, in any of the examples described herein, the apparatus, systems, and methods are configured for comparing a letter to a letter unit of at least one string in the data set, a letter to a phoneme class of a unit of at least one string in the data set, a phoneme class to a letter unit of at least one string in the data set, a phoneme class to a phoneme class of a unit of at least one string the data set, and / or comparing a unit with at least one phoneme class in the data set representing at least a part of a morpheme, and in at least one configuration comparing to all morphemes in the data set. In any one or more of the examples described herein, the apparatus, systems, and methods are configured for comparing a unit in a text string with all letters and / or all phoneme classes in a string in the data set. In at least one configuration, the apparatus, systems, and methods are configured for analyzing at least one text string having at least one unit representing an element of the text string, wherein the at least one unit represents a letter and comparing the letter to at least one letter in the data set and also comparing a representation of the letter to at least one class, in one example a phoneme class, in the data set. In the example in the preceding sentence, the representation of the letter can be the letter itself or a class, for example a sound class, for example a phoneme class, in which the letter can be found. In this example, if an acceptable match is not found for the letter, the letter or a representation of the letter can be compared to a class in the data set, for example a sound class, for example a phoneme class, to find a possible acceptable match. The criteria for an “acceptable match” can be determined by a user. In at least one configuration for this example, comparison on a specific level, for example letter-to-letter, can be done first followed by a comparison on a more general level, for example letter-to-class, class-to-class, and / or class-to-letter. While most of the examples of a “class” presented herein are directed to a phoneme class, it is understood that other classes can be defined as desired by the user, according to their interests or needs.In any of the examples described herein, the apparatus, systems, and methods can be configured for evaluating at least one unit representing an element of a text string and transforming the unit to, or assigning, a phoneme class to which the unit in the text string can belong. In someexamples, the apparatus, systems, and methods are configured for transforming the unit into data representing a list of letters belonging to a phoneme class. In such configurations, a unit from an input that does not show a desired correlation with data from a data set can be compared to one or more classes represented in the data set to see if a desired correlation between the representation of the transformed unit and data in the data set can be found. For example, if a desired exact letter match or a desired similar letter match is not found, for example after comparing all data in the data set, the unit can be transformed to or assigned a sound class, for example a phoneme class, which can then be compared to data in the data set, for example to see if there is an acceptable correlation or an acceptable amount of correlation to result in a conclusion for a designation of an exact class match or a class similarity, for example a phoneme class match or similarity. Such information can then be used to determine a desired correlation, and to save, preserve, set aside or otherwise associate the correlation between the at least one unit and the at least one text string. Comparing on a unit-to-unit basis and also on a class-to-class basis can improve the quantity and quality of the analysis of words by machine processors.In any of the examples described herein, the apparatus, systems, and methods can be configured for comparing a unit from an input string wherein a unit represents a sound symbol and wherein the unit is compared to a class of at least one string in the data set, in one example a phoneme class of at least one string in the data set. Also, in any of the examples described herein, the apparatus, systems, and methods can be configured for determining a desired correlation that represents a desired sound correlation. In at least one example, the apparatus, systems, and methods are configured for comparing a unit from an input string to a letter, and / or a class of at least one unit in the string in the data set, in one example a phoneme class, and further including comparing data of the unit representing a sound symbol to a phoneme class of at least one string in the data set.In any of the examples described herein, the apparatus, systems, and methods can be configured for evaluating a unit representing an element of a text string and for determining a desired correlation that is one or more of an exact letter match, a letter similarity, a sound class match, and a sound class similarity. A letter similarity and a sound class similarity can be based on criteria set by a user. In at least one configuration, an exact letter match is based on evaluating letters or data representing letters. A letter similarity is based on evaluating letters or data representing letters, and where the units representing the text string are not exactly identical to a data set string while being sufficiently close as defined by the user to save or otherwise keep track of the result of comparing. A sound class match is based on evaluating sound classes corresponding to units of a text string, and they are exactly identical. In some examples where the sound class is a phonemeclass, a phoneme class match is based on evaluating phoneme classes corresponding to units of a text string. A sound class similarity is based on evaluating sound classes of the text string and where the sound classes of the text string are not exactly identical to a data set string while being sufficiently close as defined by the user to save or otherwise keep track of the result of comparing. In an example where the sound class is a phoneme class, phoneme class similarity is based on evaluating phoneme classes of the text string and where the phoneme classes of the text string are not exactly identical to a data set string while being sufficiently close to keep track of the result.In any of the examples described herein, the apparatus, systems, and methods can be configured for comparing a unit of a text string to data in a data set, for example a data set containing at least a plurality of strings with a plurality of phoneme classes and determining a desired correlation between the text string unit and the data in the data set. In some configurations, the apparatus, systems, and methods are configured for generating correlation data that has data that the user determines provides the desired information from the comparing for the requirements as they have determined. The correlation data can take a number of forms, and in the examples described herein, it comprises data representing the text string and data representing a string in the data set for which a desired correlation was determined. The correlation data can be saved or otherwise stored, attached, or otherwise linked to the unit of the text string, or tracked in a manner desired by the user to allow ready availability of the correlation data with the unit of the text string and the relevant data in the data set. In at least one configuration, the apparatus, systems, and methods are configured for generating a correlation pair with the input string and the data string and any classes associated with those strings, for example sound classes, and by way of further example, phoneme classes. In another configuration, they are configured for generating a correlation pair with an input word and a data set string and their associated data, for example class information and correlation information. In a further configuration, they are configured for generating a correlation pair with a letter or letters and / or with a phoneme class or phoneme classes of the text string that correlate with the string in the data set, for example a morpheme, and saving with a letter(s) or phoneme class(es) from the string in the data set. In at least one example, generating a correlation pair includes generating a correlation pair of a word-part and a morpheme, and their unit-to-unit correlations. In at least one example, generating a correlation pair includes generating a correlation pair with a representation that correlates with the text string and at least the correlated data from the data set, for example at least one keyword representing a morpheme from the data set. In at least one example, the representation is one or more of a dictionary word defining the text string or a thesaurus synonym of the dictionary word, and wherein the apparatus, systems, and methods areconfigured for generating a correlation pair with either the word and / or the synonym and with a keyword representing a morpheme from the data set and / or a thesaurus synonym of the keyword. In at least one example, the representation is one or more of a dictionary word defining the text string and / or a thesaurus synonym of the dictionary word, and wherein the apparatus, systems, and methods are configured for generating a correlation pair with the word and / or the synonym and with a keyword representing a morpheme from the data set and / or a thesaurus synonym of the keyword.In another example of apparatus, systems, methods, and assemblies for language processing, an information collection, for example a database, data set, or other organized aggregation of data, and in one example a morpheme lexicon, includes representations of a plurality of morphemes. In one example, each of a plurality of morphemes includes at least one literal spelling, a sound sequence and at least one meaning. In one configuration, the at least one literal spelling is a spelling, for example in a selected language, for example that of the user, of the morpheme with which it is associated. The morpheme may be a morpheme corresponding to a word-part, in which case the spelling is of a representation of a word-part. For a number of morphemes in the data set, the morphemes are less than complete words. In another configuration, the at least one literal spelling can be in a Greek alphabet or a Roman alphabet. In another configuration of the information collection, each morpheme is differentiated from all other morphemes by one or more of its sound sequence, for example a phoneme class structure, the at least one meaning, and the at least one literal spelling. In a preferred embodiment, the morphemes are different in both their literal spelling and meaning, but in one configuration at least one morpheme is identical to another morpheme in either literal spelling or meaning. In some configurations, a morpheme is cross-referenced to at least one other morpheme in the information collection. Two morphemes may be cross-referenced by sound similarity and / or meaning similarity.In any or all of the foregoing examples of the information collection, the at least one meaning corresponds to a meaning of the morpheme. In some configurations, the at least one meaning of a morpheme is associated with at least one whole word, for example a word that includes a word-part that includes or incorporates the respective morpheme. In one configuration, data associated with the morpheme includes at least two words, each of which have a word-part having both a sound sequence that is the same as the phoneme class structure associated with the morpheme and having the same meanings to a selected degree of synonymity. In some examples, the word-parts are spelled with letters belonging to the same respective phoneme classes and in the same order. The morpheme meaning can be represented by a morpheme keyword, and the meaning of a morpheme in at leastone example is other than a proper noun, and in at least some examples the respective meanings are syntax-neutral. And in preferred examples, none of the morphemes are proper nouns.In any or all of the foregoing examples of the information collection, the sound sequence is a sequence of sound characteristics, and they may correspond to elements of the at least one literal spelling. In preferred examples, the sound sequence is a phoneme class structure, for example a sequence of phoneme classes, and in some examples, there is a phoneme class for each of the elements of the at least one literal spelling. In some configurations, the phoneme class structure includes at least one element selected from the group of vowels, labials, dentals, velars, liquids, nasals, semi-vowels, sibilants, and aspirates. In one configuration, there is an identification of a phoneme class structure for each entry in the information collection.In any or all of the foregoing examples of the information collection, a morpheme can be associated with at least a part of a word represented by a sound element. In one example, the part of the word is represented by a phoneme class associated with a letter of the word. In another example, a word-part is represented by a phoneme class structure associated with a syllable of the word.In any or all of the foregoing examples of the information collection, the information collection can include a plurality of data elements corresponding to each morpheme. Such data elements can provide additional useful information about the respective morpheme, and the data elements can be searchable, even in different languages. The data elements may be based on characteristics of words, for example characteristics of words in which the morpheme can be found. In at least one example, the data elements include one or more of a category, a negation status, syllable count, position, and attested examples. There may be additional data elements, for example cross-reference data linking to other morphemes related in sound or meaning. The information collection may also contain representations indicating whether or not the associated morpheme has a semantic value. For example, the presence or absence of data in the information collection may be used as an indication whether or not the associated morpheme has a semantic value. In one example, some morphemes in the data set may not be defined, and the presence or absence of a meaning identifies whether or not the associated morpheme has a semantic value. In another example, attested examples include words having at least word-parts with the same meaning as the morpheme and containing a phoneme class structure corresponding to the morpheme. Attested examples may include individual words or phrases, and may include if desired examples whose meaning is an idiomatic expression or figure of speech. Attested examples in some configurations may include proper nouns, and in some other configurations none of the attested examples are proper nouns. Additionally, attested examples can have a source language, and the source language of at least oneof the attested examples may be an ancient language. In one example, the ancient language is Greek. Additionally, attested examples can include at least one attested example in a second ancient language, and in some examples the second ancient language may be Latin. One or more of the attested examples can be in a modern language, and attested examples can include attested examples in both an ancient language and a modern language. In one preferred configuration, attested examples are from both English and non-English modern languages. In another example of data in the information collection, data can include a position of a morpheme that belongs to at least one of a prefix, root, suffix or combination thereof.In any or all of the foregoing examples of the information collection, the meaning of a respective morpheme associated with a modifying word-part modifies another word-part in the word in which the morpheme is contained. In one example, the morpheme associated with the modifying word-part represents a first word-part in the word and modifies a second word-part in the word. A morpheme used as a modifier can be a definite modifier or a relative modifier, and a modifier can also be classified as quantitative or qualitative. In some examples, a modifier can be a definite quantitative modifier, a definite qualitative modifier, a relative quantitative modifier or a relative qualitative modifier.In any or all of the foregoing examples of the information collection, the information collection can be organized or indexed according to sound characteristics. For example, a sound characteristic corresponding to the respective morpheme can be a sound class, for example a phoneme class, and the data can be organized or indexed according to such a sound class, for example according to phoneme class. In the example of a data set of morphemes, the morpheme may contain a sound sequence, preferably a phoneme class structure, and the data can be organized or indexed according to phoneme class structure. For example, data can be indexed according to a phoneme class corresponding to a portion of the sound sequence for the morpheme, and in one example indexed according to the phoneme class structure of each morpheme. In further examples, each morpheme in the information collection includes a representation of a phoneme class structure corresponding to the morpheme, and has at least one syllable with a beginning phoneme, a middle vowel, and an ending phoneme.In any or all of the foregoing examples of the information collection, the information collection can be used alone for processing words such that a word is processed in conjunction with the information collection according to the requirements of the user to produce an output. Additionally or alternatively, the information collection can be used in combination with any one or more of the apparatus, systems, methods, and assemblies described herein in any desiredcombination, or in combination with any one or more of apparatus, systems, methods, or assemblies other than those described herein, and which may be suitable for language processing.Apparatus, systems, methods, and assemblies for analyzing characteristics of a word, including the elements of a word in terms of semantic function, for example the semantic function of each word-part of a word can help to evaluate words and develop information about the word and its meaning, thereby helping to make language processors more accurate. In one example, the semantic function of a word-part, for example where the word-part is part of a word having multiple word-parts, is evaluated as a function of the word-part’s position in the word. Additionally or alternatively, the semantic function of the word-part is evaluated as a function of the location of letters in the word, for example the location of letters of each word-part relative to the locations of all of the letters in the word. In preferred configurations, the semantic function of the word-part is evaluated based on both the absolute position of the word-part in the word and as a function of an ordinal location of the letters of each word-part relative to all of the letters in the word. Each of these features separately or together with one or more of the other features can help to better evaluate word-parts as well as the words of which they are a part. Additionally, each of these features can be better evaluated in conjunction with word-parts and their associated word when they are evaluated against morphemes in a data set, for example morphemes in a Morpheme Lexicon. These help to more thoroughly evaluate semantic features of word-parts, and therefore of words.In the foregoing example for analyzing a semantic function of a word-part, the semantic function of each of multiple word-parts of a word can be evaluated as a function of the position of each respective word-part in the word. Each word-part can also be evaluated as a function of a relative position of each respective word-part relative to other word-parts in the word. Relative position is distinct from and gives more information than actual position. For example, knowing the actual position, for example that a word-part is the first or last of multiple parts, only rules out a suffix in the first position or a prefix in the last position. It does not tell what word-part is in the first or last position, for example whether it is a prefix or a root, or for another example whether it is a root or a suffix. If a word-part is found to have a relative position known for prefixes, for example a position preceding an already identified prefix or root, the word-part can be assigned that position datum, and it is assumed to have a semantic function in the word and therefore assigned at least initially such a function. Suffixes can be treated similarly except that some suffixes will not be semantically defined, and these might not have a semantic function, but otherwise will be assigned the semantic function of modifying affix. Additional or alternative to evaluating word-part positions relative to other word-parts, it may also be useful to differentiate word-parts based on their positionsin the whole word. Evaluating relative positions in the word can help to indicate that a word-part is modifying another word-part.In another example of apparatus, systems, methods, and assemblies for evaluating characteristics of a word, for example a semantic function of a word-part, a semantic function can be evaluated for each of a plurality of word-parts of a word, which develops additional information about the word-parts, for example whether a word-part is a modifying word-part based on its position relative to other word-parts. For example, a word-part that is the last word-part of a word can be either a suffix or a root, and then evaluated to more confidently determine which. Similarly, a word-part that is the first word-part of the word can be either a prefix or a root and evaluated to more confidently determine which. Furthermore, intermediate word-parts can be either a prefix or a root and evaluated to more confidently determine which. Once the position of the word-part is known relative to the other word-parts, their respective semantic functions can be evaluated to more confidently determine the contribution of each, for example as a non-modifying root, modifying root or modifying affix, and how each affects the other. Evaluating whether a word-part is a modifying word-part or a non-modifying word-part helps to determine the contributions of the word-parts to the overall meaning of the word. In one example, a word-part that is at a beginning or an end of a word may be a modifying word-part, modifying a following word-part or a preceding word-part, respectively. Additionally, a word-part that does not modify any other word-part may be a non-modifying root, and a word-part that is modified by another word-part may be a root or another modifying word-part.In any of the foregoing examples for analyzing characteristics of a word, for example a semantic function of a word-part, an example of evaluating the word-part includes identifying word-parts that are semantically defined and word-parts that are semantically undefined. For example, whether a word-part is semantically defined can be determined by the presence or absence of at least one keyword in a data set representing a meaning associated with the word-part. If a representation of the word-part can be found in the desired data set, and it is associated with a meaning, it can be concluded that the word-part is semantically defined. If a word-part cannot be identified as having a meaning, the word-part can be identified as semantically undefined. A useful data set for evaluating possible meanings of word-parts includes a data set of morphemes, which may have one or more keywords associated with each morpheme, and if the word-part can be associated with the morpheme and has an associated keyword in the data set, the word-part is semantically defined, for example by way of the keyword associated with the morpheme.In a further example of apparatus, systems, methods, and assemblies for evaluating characteristics of words and word-parts, a word-part is evaluated for determining whether or not the word-part is a non-modifying root. In one configuration, the word-part is evaluated to determine if it is a root or an affix, and other word-parts of the word are also evaluated for the same. If the word has only a single root, the word-part identified as a root can be identified as a non-modifying root. If the word contains multiple word-parts that can function as roots, each word-part that is a root is evaluated to determine if the word-part modifies another word-part. Each root modifying another root following it is identified as a modifying root, and the remaining root having no root following it is identified as the non-modifying root. In one example, each word-part is evaluated for whether the word-part modifies another word-part, and if so which other word-part is being modified. The information is then used to identify each root word-part as either a modifying root or a non-modifying root. In any examples for which a root word-part is identified as a modifying word-part, the modifying word-part can be assigned a type as being relative, definite, qualitative or quantitative, and if desired as being relative qualitative, relative quantitative, definite qualitative or definite quantitative.In any of the foregoing examples for analyzing a semantic function of a word-part, word-parts of the word can be evaluated relative to each other. For example, the word-part that is determined to have a semantic function can be evaluated to see if it is a modifier or a non-modifier. For example, the word-part that has a semantic function is evaluated for whether or not it is a non-modifying root, a modifying root, or a modifying affix. In situations where a word-part is an affix, the word-part would be evaluated to determine its position, for example as a prefix or as a suffix. For example, for a word-part that is a prefix, a modifying root, or a semantically defined suffix, the word-part is a modifier, and the word-part will be modifying another of the word-parts in the word.Additionally, in any of the foregoing examples for analyzing a semantic function of a word-part, and where the word-part is evaluated as being a modifying word-part, the word-part is further analyzed to see if the modifier word-part is one or more of relative, definite, qualitative or quantitative. In some examples, the modifying word-part may be one of relative qualitative, relative quantitative, definite qualitative, or definite quantitative. Identification of these modifier types helps to provide more detail about the word being evaluated. In other examples, at least one category keyword may be selected to be associated with the word-part.In any of the foregoing examples for analyzing a semantic function of a word-part, the word-part can be evaluated for being associated with a negation status, and assigning the negationstatus to the word-part. In some examples, the word-part can be associated with a morpheme, and the morpheme may have associated with it a negation status, and the morpheme negation status is assigned to the word-part. In one example, the morpheme is associated with a Morpheme-Word-Part of a word.Additionally or alternatively to any one or more of the foregoing examples for evaluating a semantic function of a word-part, each word-part can be evaluated for whether or not the word-part is a non-modifying root, modifying root, or modifying affix. In one configuration, a modifying root and a modifying affix can be further identified as at least one of relative, definite, qualitative or quantitative, and in a further configuration as relative qualitative, relative quantitative, definite qualitative or definite quantitative. These can help in evaluating each word-part and their contribution to the meaning of the word as a whole.Also additionally or alternatively to any one or more of the foregoing examples for evaluating a semantic function of a word-part, the word-parts can be evaluated for identifying roots in the word, including by finding the total number of roots in the word and evaluating which root of multiple roots is a final root, to be identified as a non-modifying root. Any other root in the word that is not the non-modifying root would be classified as a modifying root. Conversely, if there is only one root in the word, the root would be classified as a non-modifying root. The root or roots would then be evaluated further in conjunction with any affix to help in further evaluating the meaning of the word.In any one or more of the foregoing examples for evaluating a semantic function of a word-part, data can be received from a data set to help in evaluating the word-part. In one example, data can be retrieved from the data set for the word-part that represents a position, for example an attested position, associated with the word-part. In one configuration, the position data retrieved from the data set is associated with position data for a word-part in the data set, which for example may be a morpheme corresponding to the word-part. In one example, the association of the word-part and the morpheme is represented by a Morpheme- Word-Part pair, for example that was developed through a Morpheme Lexicon. Data in the data set associated with one or more attested positions associated with a respective word-part should be consistent with the semantic function of the respective word-parts implied by their relative position in the word, and when they are consistent the assignment of the semantic function to the word-part has a higher level of reliability. For example, the position is either one or a combination of a prefix, root or suffix position. Additionally, further data may also include a classification for the modifier as being relative, definite, qualitative or quantitative.In a further example of evaluating a word, the word or word part can be evaluated for negation status. In one example, a word, word part and / or keyword can be evaluated against a data set, for example a list of known words having a negative connotation, and the word and / or word part assigned a negation status if the word is found on the list. In a further example of evaluating a word, a word and its word parts can be evaluated as a function of the expected divisions of the word into word parts, for example syllables, relative to a set of morphemes. In one example, the word parts are evaluated against morphemes and differences noted, for example additional letters or missing letters, to assist in evaluating the word parts. In one configuration, phoneme class structures of respective word parts are evaluated relative to morphemes in a data set to evaluate the correctness of the phoneme class structures and the division of the word into word parts.In any one or more of the foregoing examples for evaluating a semantic function of a word-part, the semantic function assigned to a word-part may be evaluated in the context of those assigned to other word-parts in the word. Where the assignments are internally consistent, the assignments can be accepted, but where they are potentially inconsistent, one or more of the assignments can be modified and all the word-parts reevaluated until they are internally consistent, if possible. For example, two word-parts identified as roots would be further evaluated to determine which of the root word-parts modifies the other of the root word-parts, which would then be identified as a modifying root, and the other identified as the non-modifying root. The process can continue until the assignments for the respective word-parts are acceptable according to any criteria established by the user.Alternatively or additional to any one or more of the foregoing examples for evaluating a semantic function of a word-part, a modifying word-part can be assigned a semantic function of a type of modifier through a data set lookup process. In one example, the word-part’s meaning, is evaluated against a list of Category Keywords in a data set, which may provide additional information for more completely classifying the word-part. In another configuration, the word-part is evaluated by identifying a category for the word-part by looking for the word-part in a data set. In one example of identifying such a category, the system can identify a category in a data set having a Category Assignment, for example by matching the word-part with a word-part in the data set and retrieving associated category information from the data set. In examples described herein, the data set is a Morpheme Lexicon and includes category information. For example the word-part can be evaluated against morphemes in the Morpheme Lexicon, and a match will identify the morpheme for which associated category data can be assigned to the word-part. The category information represented in the data set, for example the Morpheme Lexicon, can be one or more of relative,definite, qualitative or quantitative, and may be further specified as relative qualitative, relative quantitative, definite qualitative or definite quantitative.In any one or more of the foregoing examples for evaluating a semantic function of a word-part, the word-part can be semantically defined by identifying at least one keyword in a data set representing a meaning. In one example, the word-part can be evaluated against the data set wherein the data set includes a plurality of morphemes, and the word-part can be matched to a morpheme in the data set. The word-part can be associated with the morpheme and / or a keyword in the data set associated with a morpheme for semantically defining the word-part. In one configuration, the keyword represents a meaning associated with the morpheme.In another example of evaluating a word, each word-part of the word can be evaluated to identify one or more roots and any prefixes and suffixes. The suffixes are evaluated to determine which are semantically defined, and the remaining word-parts are evaluated to identify the type of each of one or more roots, for example based on their relative positions in the word. If there is a single root, the root would be identified as a non-modifying root. If there are multiple roots, the relative positions of the roots will identify which is a modifying root and which root is being modified, and they can be identified as such.In a further example of a method of improving the evaluation of the elements of a word, a data set can be used which contains words and word-parts, and a word can be assigned at least one modifier type to help in evaluating words and word-parts. In one example, the modifier can be at least one of quantitative, qualitative, definite, or relative. Additionally, or alternatively, the modifier type can be defined as definite qualitative, definite quantitative, relative quantitative or relative qualitative. Additionally, the data set can include a plurality of keywords, for example category keywords, associated with each modifier type, which can help further evaluate words. The keywords can be arranged in a hierarchy to further help in evaluating words.In any of the foregoing examples, the apparatus, systems, and methods can be those for language processing, some examples of which may be referred to by the acronyms Al (artificial intelligence) or NLP (natural language processing). They may be used for any of these applications, including but not limited to lexical analysis tools, word analyzers, text-to-speech, speech-to-text, and other like communications processes through technological media, for example digital computers and processors and their associated components and / or applications, for example digital forms of books, magazines, newspapers, film, video, over the air broadcasts, streaming, LP and CD products and other recording forms, for example tape and digital storage, digital tools and their associated components and accessories, for example computers and their accessories, telephones, smart phones,network devices, hearing aids, user interface devices, for example keyboards, microphones and cameras, online media, for example online forums and data streaming platforms, antennae, radios, ham radios, telegraphs, semaphores, transmitters, receivers and transceivers, tuners, VR glasses, medications equipment and repeaters, and satellites, controllable or programmed equipment, for example robots including VLAMs, and digital or other computerized processes, for example software coding, live streaming, writing, translating, language instruction, interdisciplinary communication and scholarly research. In any of the foregoing, apparatus, systems, and methods can be implemented in a way that can more accurately or reliably represent human communication through a digital form of language.TERMSAs used herein, the following terms have the stated meanings:MORPHEME: a unit of language consisting of at least one syllable, with a Phoneme Class Structure and semantic value.MORPHEME-WORD-PART: a part of a word, and a Morpheme, paired.PHONEME CLASS: a set of Phonemes, or a set of letters or sound symbols corresponding to Phonemes and representing the same set of speech sounds in written form, produced in a similar manner (by some combination of the lips, teeth, tongue, nasal cavity, throat).PHONEME: a speech sound that is the whole of or part of a syllable.PHONEME CLASS STRUCTURE: the structure of a syllable in terms of its beginning, middle, and ending Phoneme units.TOKEN: a set of parameters and associated data.WORD-PART (when capitalized): a word-part correlated with a Morpheme.Brief Descriptions of the DrawingsFig. 1 is a schematic diagram of first and second systems that can operate separately or in combination, wherein each system is one that can receive, evaluate, display and / or output a characterization of a word part, wherein the characterization arises from one or more of a sound analyzer, a semantic analyzer, word structure analyzer, semantic function analyzer, word part modifier analyzer, or any one or more of the modules described herein.Fig. 2 is a schematic block diagram illustrating a general process by which each of the first and second systems can operate either separately or in combination.Fig. 3 is a schematic block diagram illustrating a general process by which an exemplary system can operate as a server for clients to evaluate word parts.Fig. 4 is a schematic flow diagram representing how one or more word parts of a word can be evaluated according to any one or more selected criteria.Fig. 4A is a schematic block diagram of a data set, for example a morpheme lexicon, for use with any of the systems, apparatus, methods and / or modules described herein.Fig. 4B is a schematic flow diagram similar to Fig. 4 representing how one or more word parts of an example word can be evaluated according to any one or more selected criteria.Fig. 5 is a schematic diagram illustrating methods and apparatus for evaluating word parts, and various applications of such methods and apparatus for evaluating word parts.Fig. 6 is a schematic diagram illustrating several examples of combinations of components for evaluating word parts.Fig. 7 is a schematic diagram of a flow process including a plurality of component processes for processing words.Figs. 8A-B is a schematic block diagram of a system for processing words illustrating exemplary modules, one or more of which can be used for processing words, and exemplary data sets, one or more of which can be used for processing words.Fig. 9 is a schematic diagram of an example process and components for evaluating words.Fig. 10 is a schematic diagram of characteristics of words that can be processed with one or more processes for processing words, including with the components of Fig. 9.Fig. 11 is a schematic diagram representing processing for correlations between sound, meaning and instances illustrated in Fig. 10 and such as can be used to develop a morpheme lexicon, for example toward finding lowest common denominators and morphemes, which resultant lexicon can be used with one or more of the plurality of component processes of Fig. 9.Fig. 12 is a schematic diagram representing correlation of sounds and meanings in word parts, for example morphemes, such as can be used during processing of one or more of the plurality of the component processes of Fig. 9, and which can be used in building a morpheme lexicon.Fig. 13 is a schematic diagram of a process for evaluating words, which can also be used in a process such as that illustrated in Figs. 7 and 9, and which can also be used for generating a data set for use in evaluating words.Fig. 14 is a schematic diagram of a combination of processes for processing words, the result of which can be used as part of the method illustrated in Figs. 7 and 9, and which can also be used for generating a data set for use in evaluating words.Fig. 15 is a schematic diagram of a process for evaluating words, which can also be used in a process such as that illustrated in Figs. 7 and 9, and which can be used for generating a data set for use in evaluating words.Fig. 16 is a schematic diagram illustrating applications of a data set such as that illustrated in Figs. 13 and 15, and which can also be used in a process such as that illustrated in Figs. 7 and 9.Fig. 17 is a schematic diagram of possible forms of input for one or more of the processes such as those illustrated in Figs. 7 and 9.Fig. 18 is a schematic diagram of possible options for manual interventions that can occur in processes for processing words, including one or more of the processes illustrated in Figs. 7 and 9.Fig. 19 is a schematic diagram of an overview of parts of the flow process of Figs. 7 and 9.Fig. 20 is a schematic diagram of possible processes that can be part of a plurality of component processes for processing words, including with the plurality of component processes illustrated in Fig. 9, for example lookup processes.Fig. 21 is a schematic diagram of one of the processes illustrated in Fig. 20.Fig. 22 is a schematic diagram of another of the processes illustrated in Fig. 20.Fig. 23 is a schematic diagram of possible processes that can be part of the plurality of component processes for processing words, including for a method such as that illustrated in Fig. 9, for example word splitter and generator.Fig. 24 is a schematic diagram of one of the processes illustrated in Fig. 23.Fig. 25 is a schematic diagram of one of the processes illustrated in Fig. 24 illustrating additional examples.Fig. 26 is a schematic diagram of a process for sound analysis.Fig. 27 is a schematic diagram of a component process for processing words, including one that can be part of the method illustrated in Fig. 9.Fig. 28 is a schematic diagram of a sound correlation process.Fig. 29 is a schematic diagram of a sound correlation process with additional examples.Fig. 30 is a schematic diagram of an example of sound correlation.Fig. 31 is a schematic diagram of an example of sound correlation with additional examples.Fig. 32 is a schematic diagram of an additional example of sound correlation.Fig. 33 is a schematic diagram of a process for possible use with sound correlation.Fig. 34 is a schematic diagram of a process for possible use with sound correlation, including for example sound classes.Fig. 35A is a schematic diagram of a process for possible use with sound correlation using sound classes.Fig. 35B is a schematic diagram of a process for possible use with sound correlation using sound classes.Fig. 36 is a schematic diagram of a process for possible use with sound correlation using sound classes with a method alternative to that illustrated in Figs. 35 A and B.Fig. 37 is a schematic diagram of a process for possible use with sound correlation using sound classes.Fig. 38 is a schematic diagram of examples of sound correlation.Fig. 39 is a schematic diagram of processes for possible sound correlation.Fig. 40 is a schematic diagram of examples of sound correlation.Fig. 41 is a schematic diagram of component processes for processing words, including that can be part of the method illustrated in Fig. 9.Fig. 42 is a schematic diagram of a component process for processing words, including one that can be part of the method illustrated in Fig. 9, with additional components.Fig. 43 is a schematic diagram of a combination of component processes for processing words, including ones that can be part of the method illustrated in Fig. 9.Fig. 44 is a schematic diagram of the combination of Fig. 43 illustrating additional detail.Fig. 45 is a schematic diagram of a process for semantic analysis.Fig. 46 is a schematic diagram of a process for semantic analysis with additional examples.Fig. 47 is a schematic diagram of a process for semantic analysis illustrating possible output.Fig. 48 is a schematic diagram of a process for a candidate combiner.Fig. 49 is a schematic diagram of the process of the candidate combiner of Fig. 48 with additional examples.Fig. 50 is a schematic diagram of a process for a solution ranker.Fig. 51 is a schematic block diagram of a system for evaluating words, for example for characterizing words, for example assigning categories, illustrating exemplary modules, one or more of which can be used for categorizing words, and exemplary data sets, one or more of which can be used for categorizing words.Fig. 52 is a schematic diagram of a process for evaluating words, which can also be used in a process such as that illustrated in Figs. 7 and 9, and which can also be used for categorizing words.Fig. 53 is a schematic diagram illustrating examples of structures of modifiers in words.Fig. 54 is a schematic diagram of a process for generating an output, which can also be used in a process such as that illustrated in Fig. 9, and which can be used for tokenizing data.Fig. 55 is a schematic diagram of a process for generating output, which can also be used in a process such as that illustrated in Fig. 9, and which can be used for generating output.Fig. 56 is a schematic diagram of an example of a token, such as that produced in Figs. 54 or 55.Fig. 57 is a schematic diagram illustrating a possible output method, which can also be used in a process such as that illustrated in Fig. 9.Fig. 58 is a schematic diagram of a process for editing a token, which can also be used in a process such as that illustrated in Fig. 9.Fig. 59 is a schematic diagram of a process for editing a token, which can also be used in a process such as that illustrated in Fig. 9.Fig. 60 is a schematic diagram of a process for editing a token, which can also be used in a process such as that illustrated in Fig. 9.Fig. 61 is a schematic diagram of a process for editing a token, which can also be used in a process such as that illustrated in Fig. 9.Fig. 62 is a schematic diagram representing parts of a token.Fig. 63 is a schematic diagram representing parts of a token.Fig. 64 is a schematic diagram representing parts of a token.Fig. 65 is a schematic diagram representing methods and systems and their possible outputs.DESCRIPTIONThis specification taken in conjunction with the Figures sets forth examples of apparatus and methods incorporating one or more aspects of the present inventions in such a manner that any person skilled in the art can make and use the inventions. The examples provide the best modes contemplated for carrying out the inventions, although it should be understood that various modifications can be accomplished within the parameters of the present inventions. Additionally, the Description includes headings. The headings and overall organization of the present description are for the purpose of convenience only and are not intended to be limiting in any way.Various benefits will become apparent with consideration of the description of the examples herein. However, it should be understood that not all of the benefits or features discussed with respect to a particular example must be incorporated into an apparatus, system, or method in order to achieve one or more benefits contemplated by these examples. Additionally, it should be understood that features of the examples can be incorporated to achieve some measure of a given benefit eventhough the benefit may not be optimal compared to other possible configurations. For example, one or more benefits may not be optimized for a given configuration in order to achieve cost reductions, efficiencies or for other reasons known to the person settling on a particular product configuration or method.Examples of a number of configurations of apparatus, systems and methods are described herein, and some have particular benefits in being used together. However, even though these apparatus and methods are considered together at this point, there is no requirement that they be combined, used together, or that one be used with any other. Additionally, it will be understood that a given apparatus, system, or method could be combined with others not expressly discussed herein while still achieving desirable results.Apparatus, systems, and methods for language processing take a number of forms, and the present apparatus, systems, and methods described herein can be used to improve language processing by including one, a plurality, or all of the processes described herein. A number of the processes described herein can be implemented separately or individually or in various combinations with existing apparatus, systems, and methods for language processing, and a number of the processes described herein can be used to further improve language processing in a manner, for example described herein. The apparatus, systems, and methods described herein can be implemented together with desirable results, but it is understood that one or more of the apparatus, systems, and methods described herein can be omitted while still producing desirable results. Therefore, even though the apparatus, systems, and methods described herein are presented according to one possible sequence or combination of processing, individual or combinations of processes can be rearranged or omitted, and the resulting remaining process or processes can still be used to produce desirable results. The following is a description of possible processes for language processing that can be used individually as described or in combination in the sequence or sequences as described herein.Apparatus, systems, and methods for language processing can include a number of components that can be carried out separately or in combination. Example components include one or more of the following: an input verifier, manual intervention / custom data set upload, lookup processes, including one or more of a semantic lookup and a negation lookup, keyword tool, word splitter, vowel analyzers, converter modules, string generator, semantic analyzer, tokenizer, token editor, categorizers, a sound evaluator, a semantic evaluator, candidate combiner, solution ranker, and one or more data sets for example dictionaries and / or a morpheme lexicon. Any one or more ofthese can be used separately with existing processes or with the processes herein, or used in combination as illustrated and described herein in conjunction with those illustrations.In some combinations of apparatus, systems, and methods for language processing, a Semantic Analyzer can include several of the components described herein. In examples described herein, a Semantic Analyzer can include a process for sound correlation and a process for meaning correlation. The process for sound correlation and the process for meaning correlation can operate in conjunction with an accessible data set, which in some examples can be a Morpheme Lexicon, while in other examples can be a data set configured by the user. It is noted that in other examples described herein, either or both of the process for sound correlation and the process for meaning correlation can be stand-alone processes and operate independently and separately.In an example of products, methods, systems and software, apparatus 100 for processing language (Fig. 1) such as for processing one or more words and one or more word parts of such words includes a number of components, which will be understood as capable of taking a number of forms and which can be implemented on a number of devices, which in turn may be co-located or distributed while still capable of interacting with each other and carrying out the functions described herein. It is also understood that apparatus can also be implemented in fewer devices or a single device than those represented in Fig. 1. For purposes of illustration, an exemplary apparatus 100 includes a first digital device or computer 102, which may take any form of digital device such as those described herein or currently known or known in the past, or to be developed in the future, including those for processing language, stationary or portable, co-located or distributed. The digital device allows a user (not shown but represented by the apparatus for use by the user) to use the digital device to process one or more words and one or more word parts of the respective word in accordance with any one or more of the processes described herein. “User” has been previously defined. For purposes of discussion, the user referred to with reference to Figs. 1-6 is a generic user, and the user can be the same individual or agent operating for their own personal use, as part of employment, for research, for services provided to others, or for any other purpose for which processing language is useful. The user can represent multiple users for one or more of the same purposes, while the present examples of a user for the digital device 102 will refer to the user as opposed to multiple users unless otherwise indicated for purposes of illustration. Furthermore, the user can be processing a word or word part on a single digital device or on multiple digital devices, the processing and results for which can remain on a single digital device or multiple digital devices while still being available for further processing, review and / or revision. The present description will have the user associated with a single digital device unless otherwise indicated, it being understoodthat any number of combinations and permutations of devices and methods can be used. Additionally, while the description herein applies to any form of language processing, examples will be given in the context of processing a word and word parts thereof, with the understanding that the teachings herein can then be applied to processing of one or more words in any form whatsoever, including documents, databases, word streams of any type and any form, as would be understood upon considering the present description.The user provides to the digital device or computer 102 an input 104 (FIGS. 1-2), which in the present examples will be an input string in the form of a word with one or more word parts, where at least one of the word parts will be processed 105 (FIG. 2) and with which the word can also be processed. Input string can be any string of characters or symbols representing a word or word part. Processing can include any of the processes described herein and processes carried out by any components described herein, including the modules, and may include comparisons and evaluations of words and word parts, as well as saving, displaying, and outputting or otherwise making information, for example results, available to users and / or other systems / processes. The user can directly provide the input 104, for example through any apparatus or methods for providing input for processing words, including those described herein, request the input from another device for example a portable device 106 or otherwise receive the input for processing, for example from a second digital device or computer 102'. The second digital device may be associated with a second user, another device of the first user, a service provider, including for example any of those used in the past, currently, or in the future for processing words, for example search engines, data centers, artificial intelligence facilities, research centers or any other individuals or organizations involved in processing words. The input can take any of the forms described herein, and in one example one or more words in any number of forms, for example the user’s native language, another language, electronically by a file or other means, as well as any other form used in the past, currently, or in the future used for receiving or otherwise inputting words for processing. In the present example, the first digital device or computer 102 includes the desired instructions for processing the word and its word parts, which instructions may be stored on non-transitory machine-readable media for example 108, or may be stored elsewhere in association with the digital device or computer 102 and / or on the portable device 106. The digital device identifies a word part, which may be taken in any order or identified in any desired way and which will be called a first word part for purposes of illustration, it being understood that word parts can be evaluated in any order or separately in accordance with any instructions or as designed. The word part can then be evaluated or analyzed in accordance with any one or more of the processes described herein, including those associated with any one or more ofthe modules discussed further herein. The analysis or evaluation produces a result which associates with the word part one or more characterizations 110. The one or more characteristics provide information about the word part to help understand not only the word part, and its contribution to the word, but also to help understand the word as a whole. The one or more characteristics may be output 112 to a display such as on the digital device or computer 102 or the portable device 106, or output to a separate media 114. Additionally or alternatively, the one or more characterizations may be transmitted to the second digital device or computer 102', for example through a network device 116 over a network 118, which may be any conventional network or other suitable communications method.The analysis or evaluation of the word part and the word of which it is a part can be done in any of the ways described herein with any of the systems or apparatus described in accordance therewith or other suitable systems or apparatus, including conventional systems and apparatus that may be modified for use with one or more of the elements described herein. The analysis may include any one or more of the following, which is not an exhaustive list and is given as one or more examples for analyzing or evaluating the word part and / or the word of which it is a part. The analysis or evaluation can be done separately, or one or more together either serially in suitable order or in parallel, and the resulting characterizations provided as desired, either separately or together when multiple analyses or evaluations are carried out. Each word part can be evaluated for vowels, vowel sequences, presence or absence of diphthongs, sound class structure and in preferred examples phoneme class structure (PCS, in 105, 110), sound correlation relative to other word parts for example word parts or word part strings in data sets, including in preferred examples morphemes, semantic or meaning correlation, and word part categories, for example one or more of prefix, root, suffix, semantically defined, semantically undefined, modifying, non-modifying, modified, unmodified, for example a non-modifying root, relative, definite, quantitative, qualitative, definite quantitative, definite qualitative, relative quantitative, and relative qualitative. The word, which is formed in part by the word-part, can be evaluated for a meaning or a keyword that can be associated with a meaning of the input word, a negation characteristic, possible forms of the word in a hyphenated form, syllable count, possible candidate combinations of word parts which can indicate one or more forms or meanings associated with the word or its inflected forms, single-vowel strings, multiple-vowel strings, stand-alone vowels, silent vowels, doubled vowels, vowel sequences, syllabic vowel sequences, syllabic stand-alone vowels, diphthongs, synonyms, degree of synonymity between a meaning of the first word part and strings or words from a data set. With any one or moreof these characterizations of the word part, the characterizations can be output from the first digital device, for example saved, displayed, transmitted or otherwise made available to the user."Text", "word", "word part", "letter", “sound class”, “morpheme”, and “phoneme class” in their singular and plural forms are used throughout the description and may include such items in human recognizable forms as well as digital or other representations thereof. In the context of these terms and language processing, “input” and “input string” are used interchangeably herein.In an example of such analysis or evaluation of a word part, under the control of a processor or digital processing circuitry, the word part can be associated with a sound structure, for example a sound structure associated with the letters of the word part, such as a phoneme class structure (PCS), which represents a way of characterizing the word part. In this example, the sound structure characterizes the word part based on letters, but the characterization of the letters through the sound classes associated with the letters generates representations that can be used, for example in machine processing, to evaluate sound characteristics of the letters, word parts and also of words. The sound characterization corresponding to the word part can then be output for purposes as described herein or any other desired purpose. Such output may be preceded by optional postprocessing 120, which may be any additional processing after an analysis or evaluation of the word part, and in the present example where the word part is analyzed or evaluated for a sound class, postprocessing may include using or modifying the sound class characterization for additional processing, such as to identify possibly related words, meanings or for other purposes. In one example of postprocessing of a word part characterized with sound classes, the sound class structure can be compared, under the control of a processor or digital processing circuitry, to data representing sound classes in one or more data sets, for example data set 122. In some examples, the data set can include representations or strings of sound classes, along with any associated data, for example other words or word parts, and the word part sound classes can be compared to sound class strings in the data set, for example to identify potentially similar word parts or words represented in the data set. In one example, the sound classes are phoneme classes, and a phoneme class structure characterizing the word part is compared to phoneme class structures in the data set, and any matches (to a desired degree of similarity) identify potentially similar word parts (and their associated words). Such potentially similar word parts from the data set can be associated with the word part, for example for further processing, for example using the methods described herein. For example, potentially similar word parts can be ranked and output, for example to allow a user to evaluate the results and possibly modify the results or the processes that led to the results, as well as to use the results to search for additional potentially similar word parts and their associated words. As discussed more fully below,preferred examples described herein use a morpheme lexicon as the data set or one of the data sets 122. Such postprocessing or any other postprocessing can then produce an output 112. Additionally, any preprocessing 123 of a word or word part can be carried out either before the input string is received, or after receipt of the input string and before the word part is analyzed or evaluated in step 105.Items illustrated in dashed lines represent optional items. Items represented in “chain” of dash and dot are inserted solely for ease of illustration and discussion.The first digital device or computer 102 and its associated components, if any, can operate as a standalone device for processing words and word parts, for example as described in conjunction with Fig. 2, and the processing for producing word part characterizations can use any one or more of the components and processes described herein. Such standalone apparatus and process discussed with respect to Fig. 2 is represented by the left side of Fig. 1 combined with the process described with respect to Figure 2. The standalone nature of the first digital device or computer 102 is represented by the dashed line 124 in Fig. 1. However, in some configurations, the user can use the first digital device or computer 102 in conjunction with the second digital device or computer 102’, for example as referenced above, and may carry out one or more of the same processes, with output, as it can with solely the first digital device or computer 102. In some examples, the second digital device or computer 102 and its associated components, if any, can carry out all of the processes and functions and results as the first digital device or computer 102, receiving input 104’ through the second digital device from the first user over a suitable network 118, with the input received at a network device 116’. Any input can be processed, displayed or otherwise applied to the portable device 106’ as desired and / or processed in the second digital device or computer 102’ under the control of a processor or digital processing circuitry, for example in accordance with instructions that may be stored on non-transitory machine-readable media 108’. In one example, the word part of the word can be associated with a sound structure, for example a sound structure associated with the letters of the word part, for example a phoneme class structure (PCS), one representation of a way of characterizing the word part. The sound characterization corresponding to the word part can then be output, for example in a token, for purposes described herein or any other desired purpose. Such output can be provided, for example over the network 118 and received by the user or other entity associated with the first digital device or computer 102, and the output may include any desired data, for example one or more of the word, word part, sound class information, for example phoneme class structure, and any other possibly associated data, for example any one or more of: vowel identifications, vowel sequences, presence or absence of diphthongs, sound correlation relative toother word parts for example word part strings in data sets, including morphemes, semantic or meaning correlations, word part categories for example one or more of prefix, root, suffix, semantically defined, semantically undefined, modifying, non-modifying, modified, unmodified, for example a non-modifying root, relative, definite, qualitative, and quantitative, and output data associated with the word can include one or more of representations of meanings, keywords, negation characteristics, word forms for example hyphenation, syllable count, and possible candidate combinations of word parts indicating such characterizations as one or more forms or meanings associated with the word, single-vowel strings, multiple-vowel strings, standalone vowels, silent vowels, doubled vowels, vowel sequences, syllabic vowel sequences, syllabic stand-alone vowels, synonyms, degree of similarity for example degree of synonymity between a meaning of the word and words or strings from a data set. Any one or more of these forms of output may be useful to a user or other system for viewing the results and / or allowing the user or other system to further process data associated with the output. Such output may be preceded by optional postprocessing 120’, which may be any additional processing after an analysis or evaluation of the word part, and in the present example where the word part is analyzed or evaluated for a sound class, postprocessing may include using or modifying the sound class characterization for additional processing, for example to identify possible related words, meanings or for other purposes. An example of postprocessing of a word part characterized with sound classes includes comparing the sound class structure under the control of a processor or digital processing circuitry to data representing sound classes in one or more data sets, for example data set 122’, which may but need not be identical to the data set 122. In one example where the second digital device is part of an enterprise or commercial system, the data set 122’ will be configured for and contains expanded information beyond what would be desired by a single user, as would be understood by one skilled in the art. For example, where the second digital device or computer 102’ represents a host for a multiple-user enterprise, the data set 122’ is expanded to accommodate the multiple users as well as any additional processing appropriate for such an enterprise. Additionally, the second digital device or computer 102’ will be further associated with additional components for example server components and associated instructions stored on appropriate media such as 108’ and / or 114’, as necessary. In such an example, the second digital device or computer 102’ and its associated components and processes would serve as the primary apparatus and associated methods, and the host with the second digital device or computer 102’ can include the instructions for and process as many of the functions for analyzing and evaluating word parts as desired, represented on the right side of the dashed line 124 in Fig. 1, and any user associated with the first digital device or computer 102 or similar would beanalogous to one or more clients as would be understood by those skilled in the art. Additionally, any preprocessing 123' of a word or word part can be carried out either before the input string is received, or after receipt of the input string and before the word part is analyzed or evaluated in step 105'. In other examples, structures and functions can be shared as desired between the first digital device or computer 102 and the second digital device or computer 102', and / or with any additional digital devices as desired.In some examples, the second digital device or computer 102' will be a host with associated processes, including processes for analyzing or evaluating a word and word parts, for example one or more of those described herein, and for example with one or more modules such as those described herein, with any additional components desired for carrying out host functions. In some examples, the host can be a search engine, a user accessible data source, a content publisher, a database manager, cloud-based resource manager, for example a host server facility or farm, a content generator or aggregator or agent such as ChatGPT, or a means of accessing such an agent such as Alexa, Siri, or the like. In one example, the host can carry out desired processes for evaluating the word part, with or without any preprocessing 123' of the word and / or word part, and with or without any postprocessing of the word and / or word part. The host can, for example in accordance with instructions that may be stored on non-transitory machine-readable media, the word part of the word can be associated with a sound structure, for example a sound structure associated with the letters of the word part, for example a phoneme class structure. The sound characterization corresponding to the word part can then be output for purposes described herein or any other purpose as desired, and in some examples, the phoneme class structure may be used to identify possibly related words, meanings or for other purposes. In one example, the word part and the associated sound classes can be used to compare with sound class structure under the control of a processor or digital processing circuitry to data representing sound classes in one or more of the data sets 1147122'. Any matches to data in the data set, to the desired degree of similarity, can provide information about the word part and therefore the word, including one or more possibly related meanings, and / or other words (for example as described herein attested words) that may have the same or similar sound class structure therein and possibly the same or similar or related meaning as the input word part. The characterized word part and any additional data such as data developed through comparisons with data in the one or more data sets can be output or processed further in postprocessing steps. Output with desired content in a desired form can be provided to one or more users and / or one or more systems, for example in a token, and may include one or more of the word,word part, sound class structure and / or phoneme class structure, meaning, and other possibly related words.In other examples, the second digital device or computer 102' will be a background digital device with associated processes, including processes for analyzing or evaluating a word and word parts, for example one or more of those described herein, and for example with one or more modules such as those described herein, with any additional components desired for carrying out processing functions of the type discussed herein. For example, the background digital device can be a search engine, artificial intelligence device or facility, Internet traffic monitor or other system for monitoring and / or collecting data streams (for example where the input string would be received without action from a user or other entity outside the second digital device or computer 102’), server or data facility or farm, or commercial or enterprise system directed to other functions with a second digital device or computer 102' receiving data from any of the foregoing for its own analytical purposes without returning output or results to any of the foregoing systems or facilities. In one example, the background digital device can carry out desired processes for evaluating the word part, with or without any preprocessing 123' of the word and / or word part, and with or without any postprocessing of the word and / or word part. The background digital device can, for example in accordance with instructions that may be stored on non-transitory machine-readable media, the word part of the word can be associated with a sound structure, for example a sound structure associated with the letters of the word part, for example a phoneme class structure. The sound characterization corresponding to the word part can then be used for the purposes described herein or any other purpose as desired, and in some examples, the phoneme class structure may be used to identify possibly related words, meanings or for other purposes. In one example, the word part and the associated sound classes can be used to compare with sound class structure under the control of a processor or digital processing circuitry to strings of data representing sound classes in one or more of the data sets 1147122'. Any matches to data in the data set, to the desired degree of similarity, can provide information about the word part and therefore the word, including one or more possible related meanings, and / or other words that may have the same or similar sound class structure therein and possibly the same or similar or related meaning as the input word part. The characterized word part and any additional data such as data developed through comparisons with data in the one or more data sets can be output or processed further in postprocessing steps. Output with desired content may include one or more of the word, word part, sound class structure and / or phoneme class structure, meaning, and other possibly related words, as well as other data developed during preprocessing, the referenced word part analysis and / or postprocessing. The output can be used todevelop and / or train artificial intelligence models, develop, maintain and / or modify databases, evaluate or summarize documents, perform data analysis on large data stores including text or representing words in a manner convertible to text such as transcription data for podcasts or videos, develop, train or modify business development models, analyze trends or other information represented by data streams or other forms of data available to the second digital device, develop, modify or evaluate advertising, for example targeted or other focused advertising, as well as other uses.Other examples of systems and processes with which the first and / or second digital devices can be implemented are presented in three columns in Fig. 5. Such examples are illustrative and not exclusive, but demonstrate the utility and application for the apparatus, systems and methods described herein.Any of the systems and apparatus described herein can use any one or more of the methods and modules described herein to achieve the desired word part evaluation and characterization, with any preprocessing and any postprocessing as desired, to produce the desired result with or without output. Several of the processes and modules will be described in conjunction with Fig. 4 while it is understood that any one or more of the processes or modules described can be omitted or substituted by other processes and modules, and / or supplemented with any one or more additional processes and modules. However, the present description provides examples of several processes and modules 400 that can provide useful results in word evaluation and for characterizing word parts, which description can be supplemented or substituted with any one or more additional details from the more detailed descriptions of the processes and modules described elsewhere herein. In one example, one or more words and / or one or more word parts of a word 402 will be received, for example at an input, for evaluation 404, if any, of the word and / or the word part and characterization 406 of one or more of the word parts after any preprocessing that may occur, as represented by the ellipsis 408. Following one or more of the characterizations of the word part, the word, word part and any data or results of processing can be provided for postprocessing and / or output 410.In one example of processing a word and / or a word part, the word and / or word part can be evaluated for possible identification of sound classes 412 for the individual letters of the word / word part, and appropriate sound classes can be assigned or associated with each letter. The sound classes can be used to evaluate each word part in accordance with a preferred sound class identification established for the system. In one example, the letters can be evaluated for assignment of a sound class using a letter to sound class converter module 802 (Figure 8A). In preferred examples, the sound class identification uses a phoneme class definition set 414, such as those described herein.The module 802 is preferably a letter-to-phoneme class converter module using the phoneme class definition set. When the letters are assigned or otherwise associated with an appropriate phoneme class, the associated phoneme classes can be used during any additional processing, for example in preferred configurations during sound correlations with existing data sets.In another example of processing a word and / or a word part, the word can be evaluated for identifying a possible meaning 416 of the word. The meaning can be developed in a number of ways, one of which is to determine whether or not an assumed or known meaning was provided by a user or another system at the time the word was presented for input. Another way for identifying a possible meaning or meanings for the word is to evaluate the word against the contents 418 of one or more data sets. The input word can be evaluated, for example, by using a semantic lookup module 804, which in turn may use one or more word data sets 806 (Fig. 8A), Any possible meanings determined as possible candidates, to a desired degree of similarity, for possible meanings of the word can be associated with the word for use in further processing.A further example of processing a word and / or a word part may include evaluating the word 420 for the possible existence of a keyword representing a meaning for the input word. Any candidate keywords can be associated with the input word for use in further processing. A system for identifying possible keywords may include a keyword module 808, which may include additional elements or processes for carrying out the keyword review, and which in turn may use a keyword store 810.Any one or more of the evaluations for sound class, including for phoneme class identification, and meaning or keyword evaluations can be done separately or together, as represented by the dashed loop 422, serially or in parallel as may be appropriate for the evaluations. Additionally, these evaluations can be substituted or supplemented with one or more other processes and / or modules, as may be appropriate, as represented by the ellipses 408a. A number of other processes and modules for evaluating the word and / or word part are described herein, including for example in conjunction with the block diagram of Figs. 8 A and 8B.Additional or alternative to the word / word part evaluation, the word part can be evaluated by one or more processes with one or more modules for characterizing 406 one or more features or aspects of the word part. Various characterizations of the word part can help to define the word part, which helps to define the word, and may also help to provide additional words or concepts to explain the meaning of the word. In one example, the word part is evaluated for a sound class structure 424, for example based on the sound classes assigned to letters in a syllable or otherwise associated with the word part. The sound class structure can be used to characterize the word part, and in preferredexamples the sound class structure can be used to identify possible meanings of the word part. In preferred examples, the sound class structure 424 is based on a phoneme class 426 definition set, which is based on understood relationships between a phoneme class and a letter it represents. Accommodation of phoneme classes assigned to letters in the word part provide a phoneme class structure (PCS) that can help evaluate the word part by means beyond just the letters of the word part. The sound classes, and in the preferred examples the phoneme classes, and their structure as determined by the letter sequence of the word part, can be used to evaluate the word part against other data to produce additional characterizations based on the sound class or phoneme class structure. Additionally, a sound correlation module 812 and procedures similar to those carried out by the sound correlation module can be used to provide additional characterizations for the word part.In an additional example for characterization of the word part, the sound class structure of the word part, for example the phoneme class structure 426 can be used to evaluate the word part against the content of a data set, for example strings in the data set 428. The data set can include a number of items of data, and, in the illustrated example of a morpheme lexicon 428A in FIG. 4A, includes one or more of data set words and / or word parts 430, meanings 432 or data representing meanings, for example keywords 434, morphemes 436, attested examples 438, and any additional associated data that may be provided by the designer, as may be appropriate, as represented by the ellipses 408b. The word part can be further characterized by comparing 440 the sound class structure of the word part, in the present example the phoneme class structure 426 to strings representing a phoneme class structure 442 in the data set 428. In the case of an exact match or a match to a desired degree of similarity, the sound class structure of the word part allows the system to characterize the word part as being sufficiently similar to or associated with the data in the data set 428 corresponding to the same or substantially similar phoneme class structure 442. Such a characterization allows the word part to be associated with a meaning through its sound class structure, where there is a correlated sound class structure and meaning represented in the data set 428. Through the comparison 440, the sound class structure can be used to assign one or more meanings to the word part. Additionally, such characterization allows the word part to be associated with the morpheme 436, its meaning 432 and morpheme keyword 434 and any attested examples 438 associated with the identified data set phoneme class structure 442. This allows the system, for example any of the digital devices, to evaluate a word part not only on its letters but also on the sound associated with the word part through the sound class structure of the word part. In preferred examples described herein, the dataset 428 is a morpheme lexicon, represented schematically at 814 in Fig. 8B and described more fully herein in other sections.In an additional or alternative example for characterizing the word part, meaning information associated with the word part can be evaluated against meaning data in the data set 428 to develop additional meaning characterization for the word part. In one example, meaning information 416 associated with the word can be compared 446 to meaning information in the data set 428 to find a match, to a desired degree of similarity or synonymity. For example, any one or more of the meaning information associated with the word, for example the meaning 416, one or more definition keywords 418 or other keywords 420 associated with the input word part can be compared to similar information associated with strings in the data set 428. For example, if the meanings or keywords of the word part are sufficiently similar to the meaning in the data set 428 (see, for example, meaning 432 in the morpheme lexicon 428A in Fig. 4A), or morpheme keyword 434, the word part can be further characterized by the meaning information associated with the matching string in the data set 428, including a corresponding morpheme 436. The word part can be additionally characterized by any additional associated data for the morpheme 436. Additional processing or additional steps can be carried out by other processes and / or modules described herein, for example a semantic analyzer module 816 in Fig. 8B.Any one or more of the characterizations for the word part including the sound class comparison or sound correlation and the meaning comparison or meaning correlation can be done separately or together, as represented by the dashed loop 448, serially or in parallel as may be appropriate for the characterizations. Additionally, these characterizations can be substituted or supplemented with one or more other processes and / or modules, as may be appropriate, as represented by the ellipses 408c. A number of other processes and modules for evaluating the word and / or word part are described herein, including for example in conjunction with the block diagram of Figs. 8 A and 8B.Additional or alternative to the word / word part evaluation and the word part characterization, the word part can be categorized 450, either alone or in combination with one or more of the other word parts of the word, for example without limitation as a modifier or non-modifier, having or lacking a semantic function, and / or definite, relative, quantitative, and qualitative, and in some examples with categories that can be hierarchical. Categorizing the word part can help to further define the word part, including for its function in the word and relative to other word parts. For example, the word part can be categorized 452 according to the structure of the word part in the word, for example as a function of an ordinal location of the letters of each word part compared toall of the letters in the word. In one example of characterizing a word part based on structure, the word part can be identified as to whether or not it is a prefix, a root, or a suffix, and / or whether or not the word part is a modifier or modified. The word part can be analyzed and categorized as a function of one or more of an absolute position of each respective word part in the word, a relative position of each respective word part relative to other word parts in the word, at least one position associated with the word part, as well as the ordinal location of the letters. The word part can be categorized with a semantic function based on the foregoing.Additional or alternative to categorizing the word part according to the functional and / or structural role of the word part in the word, determining 454 whether or not the word part is a modifier or is modified helps to further categorize the word part. For example, determining whether the word part is a modifier or a non-modifier can help to identify whether or not the word part is a non-modifying root. Additionally, determining whether the word part has more than one root can help to properly categorize the word part as between a prefix and a non-modifying root. Furthermore, categorizing 454 the word part as a modifier or a non-modifier allows the system and a process to further categorize a modifying word part, which can provide a significant amount of information about the word part. In a number of examples, a modifying word part can be further categorized 456 as definite, relative, quantitative or qualitative, and possibly as definite quantitative, definite qualitative, relative quantitative, or relative qualitative, thereby providing further information about the word part. Additionally, a modifying word part may be further categorized with a hierarchical categorizing scheme.Any one or more of the systems, apparatus, methods and modules described herein can use a data set of the type that is described herein as a morpheme lexicon. In various examples, the morpheme lexicon is a data set stored on non-transitory machine-readable media, wherein when one or more representations of the data are accessed by a digital device, for example one having at least one processor for evaluating the data, wherein the at least one processor can access the data and place the data in circuits for evaluating the data, for example circuits 126 / 126' (Fig. 1), which may be registers, comparators, or other digital components on or associated with the digital devices or computers 102 / 102' modifying their circuits for processing, evaluating, comparing or otherwise using one or more representations of one or more elements of data in the data set. In various examples, one or more items of data from the morpheme lexicon will be compared to input data for helping to evaluate and / or characterize the input data. In one example, the morpheme lexicon 428A (Fig. 4A) includes data corresponding to a plurality of morphemes 436 and sound class information for the respective morpheme, in the present example phoneme class structures 442, characterizingthe respective morphemes. The morpheme lexicon also preferably includes at least one morpheme keyword 434 representing a meaning of the respective morpheme, and may also include additional data for each morpheme (represented by the ellipses 408b), for example meaning information 432 that can take a number of forms, and / or attested examples 438. In some examples, not every morpheme will include each and every item of additional data that is associated with other morphemes. Additional data associated with respective morphemes, also called associated data, may also include one or more of position data, syllable count, negation status, category / categorizer data, a morpheme’s literal spelling, cross references to other morphemes, and / or any other data that may be considered relevant by the designer. In preferred examples, the phoneme class structure is comprised of sequences corresponding to the letters of the word part, and include vowels, labials, dentals, velars, liquids, nasals, semi vowels, sibilants and aspirates. In the present examples, the morpheme lexicon can be used by either or both of the first and second digital devices 102 / 102' and by any users including personal users, commercial and enterprise users such as hosts as well as background digital devices and others in a wide variety of industries and applications.Some of the processes described with respect to Fig. 4 can be supplemented with one or more additional processes or modified as desired. In one example of application of several processes to an example word “deciduous” (Fig. 4A), the word 458 can be evaluated 460 with one or more processes. In the present example, the word is processed using instructions stored on non-transitory machine-readable media adapted for execution by one or more processors on any of the digital devices for evaluating the word. In the illustrated example, the word is evaluated, for example using a semantic lookup 462, for whether or not a meaning can be associated with the word. In one example, the system can search for whether or not the word 458 “deciduous” has a meaning in any available data sets 464, which can be one or more reference data sets, for example dictionaries, thesaurus, synonym data sets, or other data sources available for evaluating words for a meaning. For “deciduous,” the semantic lookup 462 in searching available data sets 464 finds a possible meaning associated with “deciduous” that is “fall,” which is saved or output 462A for further use or availability for any desired processes. Alternatively or additionally, the system can search for whether or not the word 458 “deciduous” has associated with it or can be associated through further processing with a keyword representing a meaning for the word. For example, in a keyword lookup 466 process, the system can review Input data for possible keywords that have already been associated with “deciduous,” which input keyword, for example “from,” can be saved or output 466A for further use or availability for any desired processes. Additionally or alternatively, the input keyword can be used to search for other keywords that may indicate a meaning for the word“deciduous,” which may be saved or output as desired. Other possible lookups may be applied, including for example without limitation, those lookups described with respect to the semantic lookup module 804.Additionally or alternatively, the word part evaluation 460 may evaluate 468 whether or not the word 458 “deciduous” or any part thereof has a negative meaning or connotation. In one example, the word “deciduous” is evaluated for any data associated with it, such as an input keyword, results from any other processes, for example the semantic lookup 462 and / or keyword lookup or keyword tool 466, or word parts, and such data compared to one or more data sets 464 to see if any of the data is associated with a negative meaning or connotation. In one example, the data set 464 may include a list of words or word parts that are associated with negative meanings or connotations, and if any of the data compared matches with data in the data set for a negative meaning or connotation, the word “deciduous” can be assigned a suitable negation status. In the present example, the word “deciduous” is assigned 468A a negation status of "negative" or "neutral", because it contains or is similar to a term on a data set that is a list of negating terms. The negation status can be saved or output for further use or availability for any desired processes. Other possible negation lookup processes may be applied, including for example without limitation, those lookups described with respect to a negation lookup module 818 (Fig. 8A).Additionally or alternatively, the word part evaluation 460 may evaluate 470 the word 458 “deciduous” for possible division or splitting into word parts, which may or may not be identical to word divisions associated with syllables of words based on common dictionaries or other data sources. In one example, the evaluation 470 evaluates the word “deciduous” by letter type, in the present example vowels and consonants, and evaluated according to letter sequence criteria, for example that might be included in an appropriate data set, for example data set 464. In one example, the word 458 string is evaluated for one or more of a consonant-vowel-consonant (CVC) form, consonant-vowel form, vowel-consonant-vowel form, or vowel-consonant syllables. For example, these sequences are more likely to find similar forms among morphemes, such as the morphemes included in the morpheme lexicon, which may increase the probability of finding relevant information when the word or word parts are compared to data in a data set such as a morpheme lexicon or similar. Additionally or alternatively, the word splitter process 470 can evaluate vowel forms, including vowel sequences. In one example, a vowel sequence that can be classified as a diphthong can be helpful in identifying appropriate locations to split a word, and other vowel forms can also be helpful, for example doubled vowels and syllabic stand-alone vowels. In the present example, the word splitter process 470 consonant-vowel forms and the vowel sequences and splits“deciduous” as “DE:CID:UOUS,” which is saved or output 470A for further use or availability for any desired processes. Other possible word splitter processes may be applied, including for example without limitation those word splitter processes described with respect to one or more of a word splitter module 820, vowel analyzer module 822, vowel sequence module 824 or diphthong lookup module 826.Other evaluation processes can be applied or used for evaluating the word and / or a word part as represented by the ellipses 408.In one such additional evaluation process, the word 458 is evaluated for its letters and possible conversion or assignment of equivalences for the letters, for example in a letter converter 472. The letter converter may use a data set such as an assignment or equivalency data set that may be included in the data set 464 and which evaluates each letter of the word 458 “deciduous” for a sound representation. Assigning or relating a sound to each letter can help to define or provide information about the meaning of the word. In preferred examples herein, each letter is assigned a phoneme class, for example labial, dental, vowel, velar, liquid, nasal, semi-vowel, sibilant, or aspirate. In the present example, the letter converter 472 converts 472A the letters of the word to "dental -vowel: velar-vowel-dental: vowel-vowel-aspirate", which is saved or output for further use or availability for any desired processes. One or more of these processes for evaluating a word or word part may be supplemented or replaced by any number of other evaluation processes, including without limitation a verifier module 828, a hyphenation lookup module 830, a syllable count module 832, a string generator module 834, a sound class to letter converter module such as a phoneme class to letter converter module 836 and / or a string to sound class converter module such as a string to phoneme class structure converter module 838 (Fig. 8A).Word and word part evaluation may occur in any number of sequences or one or more processes may be done in parallel. However, in one example of a possible sequence, the letter converter module 472, which may be used to convert letters to phoneme classes, can be completed before the word splitter 470, and the word splitter can use the phoneme class structure of various letter combinations to help identify preferred letter combinations for splitting the word.Further examples of characterizing a word part, such as word parts for the input word 458 “deciduous” evaluate the word parts and characterize 473 the word parts and associate additional information to the word parts based on the evaluation. For example, the word parts of the word 458 are evaluated for the sound information 474 as against data in one or more data sets 476 having sound information associated with words or word parts and with meaning information, which in preferred examples is a morpheme lexicon. This example will be described in the context ofphoneme classes assigned to letters of the word, for example from the letter converter 472, and a morpheme lexicon such as that described with respect to Fig. 4A. However, it is understood that comparisons can be carried out with other sound characteristics for a word part and with other data sets having sound information associated with words and / or their meanings. The phoneme class structure of a first word part obtained from the letter converter 472 by way of 472B is compared to phoneme class structures in the morpheme lexicon 476, and where there is an exact match an association is made between the word part and any one or more items of information from the data set that is associated with the matching phoneme class structure. In preferred examples, the items of information from the data set include an identification of a morpheme and a meaning associated with the morpheme, for example by way of a morpheme keyword, a definition or other meaning associated with the matching phoneme class structure. Identifying a match in a phoneme class structure to a desired degree of similarity allows meaning information to be associated with the word part without further analysis, for example by other processes. Any meaning data associated with the data set phoneme class structure provides significant information to help provide a meaning for the word part, which will also help with determining the meaning for the word. Other data from the data set may also include attested examples, and characterizations of attested examples corresponding to the morpheme, such as related categories, syllable count, word part position in the word and / or negation status. The association of these items of information from the data set can be made digitally or they can be stored or otherwise made available for processing in association with the input word part. Also in preferred examples, the phoneme class structure of the word part may be compared to the phoneme class structures in the morpheme lexicon 476 and where there is a partial match, to a desired degree of similarity, an association may also be made between the word part and any one or more items of information from the data set associated with the data set phoneme class structure. Partial matches may be useful, for example, when an exact match has not been identified. Other comparison criteria can also be used, alternatively or additionally as desired by the designer. Specifically with respect to the word “deciduous”, the sound analyzer process 474 finds an exact match between the letters "DE" in the input word part and the morpheme "DE", and the morpheme "DE" is saved or output 474A for further use or availability for any desired processes. Additionally, the phoneme class structure "dental vowel" corresponding to the input word part letters "DE" may also match or show similar matches to morpheme's Ml and M2, which as seen in Table D correspond to morphemes meaning "two" and "ten", both of which contain the "dental -vowel" phoneme class structure. These morpheme's and their associated data can also be saved or output 474A for further use or availability for any desired processes. Possible additional processes mayinclude ranking the three morpheme's in order of preference, eliminating one or more, or doing more sound analysis to locate additional candidate morpheme's to help define the input word part and therefore the word “deciduous”. Similar sound analysis 474 would be carried out for each of the additional word parts, in the present example "CID" and "UOUS", and the results saved or output 474A for further use or availability for any desired processes. Additionally or alternatively, one or more elements of other sound analyses or sound correlations described herein can be carried out, including with the sound correlator module 812 and phonemic search processes.The word part and / or its word can be characterized by semantic analysis, additional or alternative to any sound analysis, which can provide meaning information for the input word part. In the present example, a semantic analyzer process 476 can evaluate any keywords associated with the input string of the word “deciduous” and compare them to words in a data set. In preferred examples, the semantic analyzer process accesses the data set 476, in the preferred examples a morpheme lexicon, and compares any input keywords or keywords associated with the input word through other processes either preceding or concurrent with the semantic analysis. In the present example, any meanings associated with the input word, and keywords from the keyword tool 466 and / or the semantic lookup 462 are received from those modules or are available to the semantic analyzer 476 from the previous processes along with any other meaning information are compared to meaning information from the data set, for example the morpheme lexicon 476. For any exact matches, and for any desired matches to the desired degree of similarity, the morphemes and morpheme keywords are then associated with the input word and thereafter available for further processing in association with the word “deciduous.” The meaning "fall" developed in the semantic lookup can be determined to be sufficiently similar to a morpheme keyword "down", corresponding to a morpheme with the following spellings: “KO ”, "cat", "cath", "cast", and "cid". In some examples, meaning similarity can also be characterized for an input word by comparing with inflections of words in the morpheme lexicon, also to a desired degree of similarity or synonymity. Any meanings that are the same or similar, to a desired degree of similarity, for example a desired degree of synonymity, and the corresponding morpheme or morpheme's are associated with the input word and the results saved or output 476A for use or availability for any desired processes. Additionally or alternatively, one or more elements of other semantic analysis or semantic search process described herein can be carried out, including with the semantic analyzer module 816 and the semantic search process described herein.Each word part of the input word 458 “deciduous” can also be characterized, alternatively or additionally, by one or more categories for each word part per se as well as word parts relative toeach of the other word parts in the word. In the present example, a categorizer process 478. The categorizer process can carry out one or more of any number of evaluations to assign characteristics to the word parts. In the present examples, each of the word parts in “deciduous” are evaluated for various categories, and any results saved or provided as output for further use or availability. In one example, each of the word parts, as identified in the word part evaluation 460, for example, is evaluated. Each may be evaluated for a word part structure, modifying characteristics, and possible category keywords to be associated with any of the word parts. In the example of evaluating a word part for structure, for example as a function of an ordinal location of the letters of each word part compared to all of the letters in the word. For “deciduous”, "DE" is evaluated 478B and determined to be a prefix, for example as being the first word part in the word. Additionally, the ending part "UOUS" is not associated with any root and would be categorized as a suffix, leaving "CID" as a root. Further categorization can also assign modifying functions to the word parts. Because prefixes and suffixes are always modifiers, they can be categorized as "modifying", and "CID" would then be classified as "modified" and also as a "non-modifying root". These categories assigned to the various word parts can then be saved or output 478A for use or availability for any desired processes. Additionally, the word parts can be further categorized for their characterization as modifiers and non-modifiers. For example, two word parts in a word (a word other than "“deciduous”) that are both classified as roots can be further categorized as a modifying root and a non-modifying root, the final root in order would be the non-modifying root. In additional examples, word parts can be categorized as definite, relative, quantitative or qualitative, and further as definite quantitative, definite qualitative, relative quantitative, and relative qualitative, for example based on the morphemes with which the word parts are associated through the sound analyzer process 474, where the morpheme lexicon includes category information. For example, the morpheme "from" has a category of "relative qualitative", and the morpheme "down" also has assigned a category of "relative qualitative". These categories can be assigned to the corresponding word parts "DE" and "CID", respectively. These categories for the word parts also can be saved or output 478A for use or availability for any desired processes.Additionally or alternatively, further modifying categories can be used to characterize word parts, for example based on keywords and their meanings, for example to a desired degree of similarity. For example, the morpheme lexicon 476 may include category keyword data which may produce a match to a desired degree of similarity with any one or more keywords or meaning words associated with the input, and such category keyword data can then be associated with the word part. The morpheme category keyword data can be hierarchical, and a level at which the input informationmatches the morpheme lexicon data provides further information for categorizing the word part. In another example, using a different external data source than the morpheme lexicon, any keywords associated with the word part from processes prior to the categorizer can be used to search a Category Keyword data set, for example one developed through searching external data sources, for example dictionaries and encyclopedias, for modifier keywords, and any such modifier keywords found in the Category Keyword data set matching, to a desired degree of similarity, keywords associated with the word part and therefore can be associated with the respective word part. The categorization for the words parts also can be saved or output 478A for use or availability for any desired processes.Additional word part characterization can be carried out on the word parts of “deciduous” as desired, as represented by the ellipses 408. Postprocessing, if any, and any output processes 480, can then be completed. The process can continue for all "h" word parts until all word parts in the word and the word itself have been evaluated as desired.At any point in the process, for example when a word part or word has been evaluated and characterized to the desired extent, any or all of the data input and data collected can be output 482 or made available in any form desired by the user or implemented by the designer. In one example, any or all of the data and associated input can be included in a token and saved, transmitted, or otherwise made available. In preferred examples, the output and any token would include one or more of the results of sound correlation, including any phoneme class structures, associated morpheme's or morpheme data and correlated meanings.A token with output or other word-related information can be generated in a number of ways. In one example, a system and process 500 (Fig. 5) can operate on an input string received 502 by a processor-based system 504 and is subject to an evaluation process 506 that includes one or more of assigning sound classes to elements of the input string, for example phoneme classes, one or more meanings, for example keywords, vowel analysis that may include diphthong lookup and / or standalone vowel, syllabic stand-alone vowel, syllabic vowel sequence, or string generation based on morphemes as they are assembled in the words. The input string is alternatively, or additionally then evaluated for one or more of sound correlation using a data set 507 (e.g. a morpheme lexicon), including with one or more of sound classes, for example phoneme classes in a phoneme class structure, a data set including sound classes, such as phoneme class structures and morphemes, meaning correlation using a data set 507, and / or categorization using a categorizer for further characterizing a word and / or word part as non-modifying root, and / or modifier type of definite, relative, quantitative or qualitative. In one embodiment, the word part evaluation is carried out insoftware 508 resident on or downloadable to a digital device, such as a computer, portable device such as a tablet, smart phone, or an embedded system, or other device having a processor, any one or more of which may include the software used to operate the processor-based system 504 for assigning sound classes, vowel analysis, string generation, sound correlation, meaning correlation and / or word part characterization, and that may also include optional additional processing, for example word verification, negation lookup, hyphenation lookup, word splitting, syllable counting, string generation, input string to phoneme class structure conversion, combination of candidates, ranking of solutions, characterization of semantic function, and token editing, as well as other word analysis functions described herein, all as represented generally at 510 and discussed more fully elsewhere. Also as represented by 512, the process can be applied in a number of manners, platforms and applications, for example without limitation any artificial intelligence function, word searching, evaluation and modification, language processing, robotics, health care, linguistics, search engines, publishing, translation, database and data center management, among others. A token is generated 514 from the input string dependent on the evaluation of the word parts and the data developed therein, and is stored 516 in digital memory or is transmitted 518 to another device by a modem or otherwise. The result 520 is that the input string is evaluated and data developed and the input string is characterized in one or more ways, including transformed to, among other things, sound classes, for example phoneme class structures, morphemes and categories, by the digital system, and also held by (i.e. stored by) the digital system. In one embodiment, the storage can occur on processorbased system 504. In another embodiment, the storage can occur on a second and / or third through "z" digital systems (i.e. represented by 522). Therefore, as an example, the token can be generated at any time on a first digital system, transmitted by modem or otherwise to a second digital system (e.g., as a form of search query, text generation request, text analysis request, and / or text verification request, or the like), and stored on or by the second digital system. Similarly, as an example, the token can be generated at any time on a first digital system, and transmitted by modem or otherwise to a second digital system, and a third, fourth or "z-th" digital system may acquire or otherwise record the token, for example for storage or other processing, for example with a background digital device.Generation, use and exchange of output information, whether or not in the form of a token, such as is described generally in Fig. 5 can be applied in a number of situations. Figure 6 illustrates a number of those situations where a digital device (e.g., a desktop or other stationary computer, tablet, smart phone, host devices, or servers, and the like) is evaluating an input string and modifying part or all of the data in the input string based on the evaluation and providing output with all or part of the data, as modified / supplemented, for example in a token. The output may includecharacterizing information, for example sound and meaning data 602 derived as part of the result of the evaluation of the Input string based on sound and / or meaning data from a sound / meaning data set 604 in a data set such as Data Sets “a” 606. Alternatively or additionally, the output may include additional characterizing information 602 for the input string as part of the result of the evaluation of the input string based on data, rule sets, lookup tables, and the like in one or more Data Sets 1, 2...“ / ” in the Data Sets “a” 606. The characterizing information may include sound class information, such as phoneme classes and phoneme class structures, morphemes, characterizations of the word or word parts of the input as a non-modifying root, and / or a modifier type of definite, relative, quantitative or qualitative, as well as any number of other characterizations. Any or all of such information can be included as output, including in the form of a token, as indicated at 602. A number of system configurations are represented in Fig. 6, including users on the left and recipients / providers on the right, solely for purposes of illustration, and a background recipient in the lower center. A first user (User 1) is illustrated as having two different devices, corresponding to User 1 A and User 1 B, indicated generally as 608, and additional users are indicated by the ellipses 610 and User "k" A represented at 612. Device "A" for User k is illustrated in expanded detail and will be discussed in more detail for purposes of illustration.In one example of a system such as that discussed with respect to Fig. 5, User k is a standalone digital device with all components and instructions and data self-contained, whether integrated, co-located or otherwise available for use as a self-contained system. Processing instructions and data for use in evaluating word parts in the various examples as described herein are either embedded with components of the system, included on a data set “DS bl” 614 or otherwise available for operation as a self-contained system. Instructions for processing may be stored on non- transitory machine-readable media, including for example DS bl 614 or elsewhere associated with the digital device of User k, and likewise with any data used by the system. In the present example, the data used by the system includes any desired data sets useful for evaluating word parts, and may include dictionaries, a thesaurus, rule sets, lookup tables, and the like, as well as a sound and meaning data set, having sound class information, for example phoneme classes, and meaning information, for example keywords and / or attested examples, for example one having some or all of the data of a morpheme lexicon as described herein. In this example, the digital device of User k receives and evaluates an input string, for example the input word "deciduous", at an input 616 and modifies part or all of the data in the input string based on evaluation of all or part of the input string for providing output by 618 with all or part of the data as modified, for example in a token. The output includes one or more of sound and meaning data, additional characterizing data 602developed through the evaluation, for example sound class information, for example phoneme classes and phoneme class structures, morphemes, identification of a non-modifying root and / or modifier type of definite, relative, quantitative or qualitative, as well as any number of other characterizations. Output 619 to User k may take any number of forms, for example through various user interfaces 620 such as a display, a text string or other format, for example for review, editing or other modifications, or the like, whether during or after processing, or the output may be stored in the data set 614. Any output can be used to generate a token or edited, if desired, with a tokens generator / editor 622. In this example, User k can carry out the evaluation of the word and word parts without interacting with any other digital device, user, recipient / provider or other third-party device.In another example of a system such as that discussed with respect to Fig. 5, User k is also a standalone digital device with all components and instructions self-contained as described above, but one or more of the data sets and sound / meaning data, for example a morpheme lexicon, can be located otherwise than in the first example, and such data communicated 624 between the User k 612 and the Data Sets "a" 606. The double arrows for the communication 624 represent exclusive data communication only between User k and the Data Sets "a". In this example, User k accesses the Data Sets 606 for processing any word provided at the input 616, for example by the individual user or another system associated with User k and any results and modified data are output to the user as described in the first example above, and which may include generating one or more tokens for display, review, editing or other action by the user, and may be stored on the User k digital device or in association with the Data Sets 606. In this example as well, User k can carry out the evaluation of the word and word parts without interacting with any other user, recipient / provider or other third- party device. Additionally, these two examples of User k can represent a single person, a commercial enterprise or organization, agency or other entity using the systems and methods for their own use without consciously sharing the input words and / or any output.In a further example of a system such as that discussed with respect to Fig. 5, second digital devices can be implemented in a manner similar to that described herein, and may include devices such as those associated with recipient / providers. Such a recipient / provider may provide services to users or otherwise interface with users to provide goods, services, information or other benefits to users and / or other third parties, whether by agreement, passively, unsolicited, surreptitiously or otherwise. Second digital devices may include a recipient / provider 1 626 through any number of devices represented by ellipses 628 to recipient / provider "m" 630. While each recipient / provider can operate as a standalone user, business or other enterprise, such as was described with respect to User k, each of the recipient / providers are described for purposes of illustration as interfacing with one ormore third parties in the form of users, or otherwise. While each recipient / provider is illustrated as being approximately identical to each other, each may be different from another based on their own models of user, business, enterprise or other entity, as may be unique to their respective situations. The discussion of the examples herein focus on possible similarities or overlap based on apparatus, systems and methods described herein. For example, recipient / provider 1 operates with a digital device configured for receiving input 632 such as input words and / or data from one or more data sets, for example Data Sets "a" 606 described previously or a similar data set, such as one having conventional data sets as well as having a form of sound / meaning data set, including sound class information associated with words or word parts, as well as meanings, and also possibly including attested examples and associated data such as any or all of that described herein with respect to a morpheme lexicon. The recipient / provider 1 digital device 626 may receive the input words and may access data from the data sets 606, for example over a network 636, for evaluating the input words and developing characterizations for those input words, which may be included for output at 634. Tokens may also be generated or edited 638 and output, for example over a modem or other suitable device. The double lines for the communication 636 represent an example of exclusive data communication only between the recipient / provider 1 and the Data Sets "a" 606, but the form of the communi cation / interacti on with third parties may be otherwise in other examples. The output includes any one or more of sound and meaning data 602 and any additional characterizing data 602 developed through the evaluation, for example sound class information, for example phoneme classes and phoneme class structures, morphemes, identification of a non-modifying root and / or modifier type of definite, relative, quantitative, or qualitative, as well as any number of other characterizations. The output can be provided internally within the organization, or output to the Data Sets "a" 606.In another example of a system such as that discussed with respect to Fig. 5, the recipient / provider 1 digital device 526 can operate in a manner similar to that discussed herein wherein the recipient / provider 1 receives input by 632 from a user, such as User k over a network 640. The input, as with any of the examples herein, may be in any number of forms, including but not limited to a word, a document or other file containing one or more words, instructions for input (such as to obtain data for evaluation), or other forms. The recipient / provider 1 accesses data from the Data Sets "a" 606 over the network 636, and evaluates the input relative to the data to provide characterizations 634 for output. Alternatively, data may be distributed in various amounts between two or more of User k, Data Sets 606 and recipient / provider 1. A token can be generated 638 and transmitted to User k, or edited for further processing before transmitting to User k. The output, suchas the token, includes any one or more of sound and meaning data 602, and any other additional characterizing data 602 developed through the evaluation, for example sound class information, for example phoneme classes and phoneme class structures, morphemes, identification of a nonmodifying root and or modifier type of definite, relative, quantitative or qualitative, as well as any number of other characterizations. The data might also be provided internally within the organization and / or saved in the Data Sets "a". Alternatively or additionally, Data Sets "a" data can be stored and accessed on a media such as DS b2 642, and the output in any or all forms can also be saved to the media 642. In this example, recipient / provider 1 may operate as a host, search engine, artificial intelligence platform, agent, language processing platform and / or translation platform, word processing and search platform, verification service, linguistics service, or other services.In a further example of a system such as that discussed with respect to Fig. 5, the recipient / provider "m" 630 may include all of the structures and functions discussed herein, including for example the structures and functions of recipient / provider 1, and operates with a digital device configured for receiving input 644 such as input words and / or data from one or more data sets, for example Data Sets "a" 606 described previously or a similar data set, such as one having conventional data sets as well as having a form of sound / meaning data set, including sound class information associated with words or word parts, as well as meanings, and also possibly including attested examples and associated data such as any or all of that described herein with respect to a morpheme lexicon. The recipient / provider "m" may receive the input words and access data from the data sets over a network 646, which in some examples may be the exclusive input and output represented by the dual lines, but in the present example the recipient / provider "m" communicates with other devices in addition to the Data Sets "a" 606. In the present example, the recipient / provider "m" receives input such as input words and / or word parts from User k and evaluates the input words with the data from the Data Sets "a" and develops characterizations for those input words, which may be included for output at 648. The output 602 may include any one or more of sound and meaning data and any additional characterizing data 602 developed through the evaluation, for example sound class information, for example phoneme classes and phoneme class structures, morphemes, identification of a non-modifying root and / or modifier type of definite, relative, quantitative, or qualitative, as well as any number of other characterizations. Alternatively or additionally, Data Sets "a" data can be stored and accessed on a media such as DS bx 650 so that Data Sets "a" is not the exclusive repository of the Data Sets 1, 2 throughand sound / meaning data, and the output in any or all forms can also be saved 649 to the media 650. In the present example, however, there may be instances in which the output is only a subset of the outputordinarily developed by the recipient / provider, or none of the characterizations are output to User k, but instead the evaluations of the input are processed by the recipient / provider "m" for evaluating possible characterizations of the input, and processing the results of the evaluations for another purpose or function, which purpose or function is carried out for the benefit of User k or for another purpose. In one example, the recipient / provider "m" is a robotic device which evaluates the input, for example as instructions for a desired action or function, processes the input and determines the best fit for the desired action or function, and carries out such action or function. Robotics is an example of such a process and output, while other examples of applications of the type described with respect to recipient / provider "m" may include language processing systems, error checking systems, medical platforms, publishing platforms, database and data center management platforms, as well as other platforms. In such examples, these platforms evaluate input data including sound information associated with the input data to correlate the sound information with meaning information, and / or to develop characterizations of the input words for purposes of determining a best fit for a desired action or function requested by User k, either separately or in conjunction with any actions with recipient / provider "m", and the recipient / provider "m" can carry out further processing for additional functions or purposes.In another example of a system such as that discussed with respect to Fig. 5, a recipient 652 can operate in a manner similar to that discussed herein and wherein the recipient may operate in a manner similar to that described with respect to User k, recipient / provider 1 and / or recipient / provider "m", including with the same or similar components and processes with the following exceptions. In the present example, the recipient 652 receives words and / or word parts or representations thereof and applies some or all to an input 654 for processing, including for evaluating the words and word parts for identifying characteristics such as those described herein. For example, the words and / or word parts can be evaluated for any one of more of sound features and meanings for correlating the sounds with the meanings, including identifying sound class information associated with the words and word parts, including phoneme class information and phoneme class structures and may output the identified characteristics through 656. The characterizations may include any or all of sound and meaning data, sound class information, for example phoneme classes and phoneme class structures, morphemes, identification of a nonmodifying root and / or modifier type of definite, relative, quantitative, or qualitative, as well as any number of other characterizations. Additionally, any or all of the characterizations can be included in a token, which can be edited, output or saved 658. Any or all data can be saved internally or on separate media such as “DS by” 670. The recipient 652 may also optionally be different than otherexamples in having a separate Data Sets "b" 672 having any or all of the contents of Data Sets "a" 606, including sound and meaning data, for example a morpheme lexicon. Alternatively, the recipient 652 can access either or both Data Sets "a" (connection not shown) and Data Sets "b". While the recipient 652 can operate in a manner similar to User k, recipient / provider 1 and / or recipient / provider "m" and receive input from entities requesting evaluation of input words, the present example of recipient 652 has recipient receiving input data passively or without any action from third parties other than such third parties transmitting data to other third parties or among themselves, which receipt of data is indicated by the dashed line 674, representing passive monitoring and accumulation of data over the networks 624, 636, 640, 646 and 676, as represented by the dashed ovals. The recipient 652 operates in the background without engagement initiated by any third party. Recipient 652 may represent, among other things, a data aggregator, data processor, artificial intelligence platform, or marketing or advertising platform, such as for targeted advertising or marketing. In the example of advertising or marketing, recipient 652 receives data without any request from third parties, but may provide data to third parties on its own initiative, for example after processing or developing characterizations of input words, and possible additional postprocessing for additional information.SYSTEM SCHEMATICThe following is a general description of the method of the system’s word analysis, assuming that all or most of the individual elements described herein are included, and is premised on the idea that a word meaning develops from the meaning of a word’s parts, as a summation of components. The “word” can be a word input by a user or otherwise receive for evaluation by any system desired for evaluating words. Features of whole words that can be analyzed by the system include various components that have an impact on the sound, meaning, and structure of a word. For example, features that are analyzed, in no particular order but moving from left to right in the diagram include: a) a word’s letters, and their number, b) a word’s negation status, and c) keywords that can be associated with a word’s meaning. These are features that can be easily analyzed and that provide useful information. This system description illustrates a number of elements that contribute to the whole, and each of a number of the items are unique and individually contribute beneficial attributes to the system, which it is submitted produces a system that is greater than the sum of each of the parts. The elements considered separately, as discussed herein, provide benefits individually that can apply not only to the present system but also any other system, for example a system for analyzing words.a) A word’s letters can be divided into types of letters, for example vowel and consonant types. At this level, or at a different level if desired, both vowels and consonants may belong to and be further analyzed by or converted to their sound class, for example a phoneme class. An additional feature of letter types may include types of letters or letter combinations that indicate syllables. Such syllables may be classified as “word-parts” of words. The component letters of such word-parts may have phoneme classes which, in sequential combination, form a phoneme class structure. The phoneme class structures of word-parts can be compared, for example, with phoneme class structures in other language elements, for example other word-parts or, for example morphemes (for examples morphemes in a data set, for example a morpheme lexicon) to evaluate sound correlations between word-parts of a word and morphemes that may be contained in said word. All of the foregoing data collected and analyzed about a word’s letters and the syllables they form, or derivable from a word’s letters, can be saved or otherwise made available future processes. b) Additionally, a word’s negation status can be separately analyzed and saved, for example by a Negation Lookup. As used herein, “saved” refers to any form of saving, storing, attaching, recording or otherwise, or transmitting for such, to allow ready availability of the data, whether availability to the systems or for the methods described herein or other systems or methods for machine processing of language. c) Additionally, a word’s keywords can be collected, analyzed, and / or saved, for example by a Semantic Lookup. Keywords of words can be used to find language elements, for example the same or similar words and their associated data to help in evaluating the word. For example, a keyword can be compared, for example, with keywords of morphemes to establish meaning correlations between a word and morphemes that may be contained in said word. Those morphemes with both sound correlation and meaning correlation provide evidence that one or more morphemes corresponding to a word-part are components of a given word. Such a morpheme can be associated with the respective word-part to form a morpheme-word-part.A set of morpheme-word-parts associated with a word can be further analyzed in terms of a) their individual position(s) in words and b) negation status. Positions (a) of a morpheme-word-part may include, for example, absolute position in a word, position relative to the letters in a word or relative to other morpheme-word-parts in the same word, and / or position associated with the morpheme component of the morpheme- word-part, for example a morpheme’s attested position(s) in other words. Negation status (b) of a morpheme- word-part may relate to a morpheme’s attested negation status in other words. It is possible for the system to compare the negation status of a morpheme- word-part to the negation status of the whole word it belongs to. If both are “negative,”for example, it may be true that the morpheme-word-part is the part that accounts for the word’s negative meaning. Negation is one type of modification that might be included in a semantic function analysis of morpheme-word-parts, but there may be others. The system can identify various semantic functions of morpheme-word-parts, for example whether each is for example a modifier or non-modifier, or a root or affix, and how the parts modify or are modified by one another. This semantic function analysis provides definite, accurate information about the word’s structure, which information can be used in machine processing of language. Additionally, the system may further analyze some parts, for example various categories (and sub-categories, to the desired level of specificity) of modifiers.Since the meanings of whole words are closely bound to the meanings of word-parts, analyzing and understanding all the components of words is relevant to incorporating word meaning into semantic language processing. In the description herein, “meaning” refers to what is signified by a word or word-part or morpheme, for example, one or more or a combination of an entity or action, or quality thereof. “Semantic” has to do with the meanings signified by words (as distinct from the rules by which words can be arranged or combined with other words). The systems and methods discussed herein relating to semantic processing are directed to the meanings of words.In one configuration of an overall system 700 (Fig. 7), one or more words 702 are received as input, and a first of the words is evaluated 704 for various characteristics, and the letters and word parts of the word are evaluated 706 for various characteristics. At a first level, the word is evaluated 708 for a negation status, and letters 710 are evaluated 712 for various letter types, for example whether they are vowels or consonants. At a second level, the letters are evaluated for and assigned 714 a sound class, for example a phoneme class, and the word is evaluated 716 for word parts, for example by splitting the word or by syllabification. At a third level, a meaning of the word is evaluated 718 for a possible meaning or meanings, for example by receiving a keyword associated with the word representing a meaning associated with the word or by accessing an external data source, for example a dictionary or the like to evaluate a possible meaning or meanings for the input word. Additionally, the word parts and the phoneme classes for each of the word parts are assembled 720 into respective sound class structures, for example phoneme class structures for the word parts. At a fourth level, a data source, for example a morpheme lexicon or the like, can be searched 722 for possibly related words or word parts, for example morphemes, having the same or similar meaning to the input word. On the word part side, the letters and / or phoneme class structures can then be used to search 724 for the same word part in the same or similar data set, for example the morphemelexicon, to identify sound correlations in and potential candidates from the data set that might have the same or similar meaning as the word part and / or input word.At a fifth level, some or all of the information about the word parts, including sound information and any associated meaning information, and some or all of the information about the meaning or meanings of the word are assembled 726 to allow any meaning information from the process to be associated with the input word, as word / word part-meaning associations, for example as a morpheme-word-part or plurality of morpheme-word-parts. The word / word part-meaning associations can then be output for the user or other system for the results, or may be further processed as desired. In one example of further processing, each word / word part-meaning association can be evaluated for accuracy, for example by comparing the evaluation 704 for the word and the evaluations 706 for the letters and word parts, and ranking or assembling the results in an evaluated order of relevance or accuracy.At a sixth level, each word / word part-meaning association, in the preferred examples each morpheme-word-part group (as well as any desire data arising from any previous processing), is evaluated 728 for its position in the word relative to other word parts, and each word part is evaluated 730 for a possible negation status. At a seventh level, each word / word part-meaning association is evaluated 732 for any semantic function, and at an eighth level each word / word partmeaning association is evaluated 734 for its possible status as a modifier, and evaluated 736 as to whether or not the word part is a root and if so if it is a non-modifying root. Additionally, at a ninth level, any modifying word parts can be evaluated 738 and assigned a category for example a category associated with a category keyword and / or a modifier type, for example one or more of definite, relative, quantitative, or qualitative. As represented in Fig. 7, the levels, including the fifth through ninth levels develop and record information regarding the meaning and structure of words and / or word parts. Once system processing is complete, any combination of the foregoing identifications can be included as the output 740, which output in preferred examples illuminates the meaning and structure of the Input word in terms of its component word-parts. Such information, including any information developed at one or more individual levels on either or both of the word part or word sides of the process, can be tokenized, and elements of the token can be edited if desired.Each level and each side of the process represented by Fig. 7 can be carried out separately, and need not be part of the overall process as described and illustrated, but alternatively can be incorporated in other processes and systems for evaluating words and word parts. Any one orcombination of systems and processes for language processing associated with a level or side of the process represented by Fig. 7 can improve processing of words and word parts.Examples of a language processing apparatus, system, and method, for example the combination designated 800 (Figs. 8A-8C), for evaluating one or more words and / or word parts can have a number of configurations. It is understood that systems may contain other modules not described herein, which may include conventional elements such as authentication systems, network management tools and the like. Additionally, the present apparatus, systems and methods may be implemented on a single device or a network of devices, including cloud-based computer implementations, with associated operating systems, memory and communications capabilities, and operations can be controlled through hardware or through computer programs installed on non- transitory computer storage and executed by processors to perform the functions described herein. The various stores are implemented using non-transitory computer-readable storage devices along with suitable database management systems for data storage, access and retrieval.In one example, they include one or more apparatus, systems and methods represented by modules 801A for evaluating and characterizing words and word parts. In one example, they include apparatus, systems and methods for one or more of sound correlation, meaning correlation, sound and meaning correlation, word categorization and / or word part categorization. Various modules and data sets are described separately and they can be implemented separately, but one of more of them can be combined as desired, and the modules and data sets represented in Figs. 8A-8C are well- suited to being used together for evaluating and characterizing words and word parts. The apparatus, systems and methods described with respect to each module and / or data set represent their respective examples, and additional examples of such modules are described elsewhere, in conjunction with various illustrations. However, one skilled in the art would be able to implement configurations of a module as described with respect to such module and other examples with or without additional or different features give in other examples herein.In one example, sound correlation can be carried out with a sound correlation module 812 (Fig. 8B) in conjunction with a data set, for example a morpheme lexicon 814, which data set can be part of the system or accessible thereto. In another example, the meaning correlation can be carried out with a semantic analyzer module 816 also in conjunction with a data set, for example morpheme lexicon 814. In a further example, evaluating and categorizing of words and / or word parts can include one or more of a semantic lookup module 804, a keyword tool module 808, a negation lookup module 818, a word splitter module 820, a vowel analyzer module 822 vowel sequence analyzer module 824 and / or a diphthong lookup module 826, a letter to phoneme class convertermodule 836 and / or a phoneme class to letter converter module 836 and / or a string to phoneme class structure converter module 838, any one of those three of which can be used to evaluate sound characteristics of words and word parts, a word structure analyzer module 840, a semantic function analyzer module 842, and / or a modifier analyzer module 844, which may include any one or more of a word part modifier analysis module, a non-modifying root analysis module, and / or a modifier type module. One or more of these modules, taken singly or in various combinations, help to evaluate word parts and their words, which can help to more clearly and more thoroughly evaluate the word parts and their words, their meanings and structures both alone and in comparison to other possibly similar words or word parts. Such evaluations and characterizations can improve any number of word systems, platforms and methodologies.Processing apparatus, systems and methods may also include one or more of an input module 846 for receiving or obtaining input such as input words and / or input word parts. The input can be received in any number of forms and datatypes, representing words and / or word parts. Additionally, input data can be received in one or more packets or data streams, electronic files, or otherwise. The systems include conventional hardware elements for operations, including network interfaces and protocols the input devices for data entry, and output devices for display, printing or other presentations of data.Optionally, a manual intervention module 848 can be provided so one or more of users or administrators can view or revise data and results at various stages of processing, as well as output and token information. Any revisions can be applied and one or more processes repeated and presented again to the user or administrator. Additionally, the module can be figured to modify or customize data sets and / or data within data sets.Optionally, a custom data set upload module 848 can be provided so one or more of users or administrators can append and / or replace one or more data sets, for example to update a morpheme lexicon data set or to append a language support reference lexicon. Any updates to data sets can be applied in an ongoing manner or to the next input following the upload, as specified by the system designer.The apparatus, systems and methods, optionally, can include a module 828 for verifying the input for processing. The verifier can evaluate whether a particular input word is acceptable based on established or selectable verification criteria. A verifier module could flag elements of input data that would not be processed, such as input from a language not supported by the system, extraneous characters such as punctuation or numbers, slang or vulgarities and / or words that are capitalized.An optional semantic lookup process can be included, such as with module 804. Semantic lookup can look for meaning words that may represent the meaning of an input word, and the meaning word can be used to evaluate the results at various points in the evaluation process. The module 804 can compare the input word to words appearing in one or more data sets 806 to determine if one or more meanings can be associated with the input word, and if so the meaning(s) would be designated a lookup keyword to be associated with the input word. The potential meanings can be evaluated to a desired degree of similarity, for example a degree of synonymity, and corresponding lookup keywords can be saved. In some examples, the module 804 can use first and second data sets for obtaining meaning candidates, one data set being in a first language and another data set being in a second language. Other lookup sources and techniques can be implemented, alternatively or additionally.The apparatus, systems and methods can also include, optionally, a module 808 for labelling potential meaning candidates for being designated lookup keywords. The module can estimate the reliability of candidates to be selected as keywords, for example input keywords and lookup keywords. Labelling can allow a weight or preference to be applied to candidates for keywords according to desired criteria, which may be set by the user or designer, such as the relative reliability of the reference data set from which keywords were obtained. For example, candidate meanings appearing in two or more data sets, for example reference data obtained from two different sources, can be labelled differently than one found the only in data associated with a single source.Optionally, a negation lookup process, for example that associated with a negation lookup module 818, can be used to characterize an aspect of the input word. Meaning correlation can be improved by evaluating a negation characteristic of the word or any of its word parts. The negation lookup module can operate in conjunction with a negation list 854 evaluating the negation status of words and / or word parts. The input word or word parts can be compared to entries in the negation list, where the entries are selected according to known occurrences of words or word parts that negate, reverse or diminish a meaning of another word part.Additionally, a hyphenation lookup process such as that provided by a hyphenation lookup module 830 can be used to identify how conventional data sets separate a word into word parts. The module can use data in the word data sets 806 to identify the hyphenation applied to an input word, to identify possible syllables and possible syllable count, as others identify them.Additionally, the apparatus, systems and methods may optionally evaluate a word for splitting into word parts such as may be done with the word splitter module 820. In preferred embodiments, the word splitter module evaluates a word as a function of common morphemeconfigurations and assigns possible word part combinations accordingly. For example, the module can evaluate the word for a consonant-vowel-consonant or "CVC" form, as well as other forms of morphemes as candidates for word parts for the word.The apparatus, systems and methods may also include, optionally, a process for analyzing vowels, for example one or more of the vowel analyzer module 822, the vowel sequence analyzer module 824 and the diphthong lookup module 826, which can help better characterize expected syllables in a morpheme universe. Additionally, if desired, one or more of the modules 822, 824 and 826 can be used in conjunction with the word splitter module to more accurately identify possible syllables for the input word. One or more of the modules 822, 824 and 826 can be used in conjunction with a conversion store 852 and a diphthong list 854 for analyzing vowels and vowel sequences in a word. In one example, the vowel analyzer module 822 identifies vowels in the word, and identifies single vowels and multiple vowels. The module 822 can also divide single vowels into terminal "E" vowels and the remainder as stand-alone vowels, and also identifies multiple vowels into either doubled vowels or vowel sequences. These can help to identify higher probability syllables in the word. The module 824 can further evaluate the vowel sequences by transliterating the vowel sequence into equivalent Greek letters using the conversion store 852 core conversion instructions to determine whether any portion of the transliterated string exactly matches letters of a diphthong, for example by comparing to entries in the diphthong list 854 using the module 826. Therefore, the modules 822, 824 and 826 either separately or together improve machine recognition of word parts in words.With or without all of the foregoing modules, the apparatus, systems and methods may also provide, optionally, a syllable count, such as that provided by the syllable count module 832. The syllable count module can count the number of syllables using existing techniques, for example a hyphenation lookup, or a technique focused on morphemes and expected divisions between word parts of a word. The module can use a number of criteria included as part of a data store or list for evaluating the word for syllable count, for example as may be represented in a data store 856 for syllable counting criteria. Syllable counting may evaluate vowels and consonants and their relative positions, as well as possible rule sets as to which are syllables, such as the occurrences of "Y", terminal "E", rule sets regarding stand-alone vowels, syllabic vowel sequences, and the like as well as any other known common occurrences of consonants and vowels in words, as desired by the designer.Additionally, the apparatus, systems and methods optionally may include sound and letter conversions, such as may be applied with the letter to phoneme class converter module 802, thephoneme class to letter converter module 836, and the string to phoneme class structure converter module 838. One or more of the modules 802, 836 and 838 can operate in conjunction with a sound / phoneme class store 858, for example to convert letters to phoneme classes and / or phoneme classes to letters, and to convert identified syllables or word parts to a phoneme class structure. The phoneme class structure can be easily evaluated or compared to similar data or data converted to phoneme class structures to make associations with meanings in a data set, for example in the preferred configurations a morpheme lexicon, to establish sound and meaning correlations for word parts. The phoneme class structures can also be easily tokenized for one or more of processing, transmission, review or editing.If desired, sound analysis can be used to improve the evaluation of words for their meanings, and may include one or more of sound analysis such as may be carried out with a sound analyzer module 860 and sound correlation such as may be carried out with the sound correlator module 812. In one example, the sound analyzer module 860 can compare letters in a word part of an input word with letters in a data set, for example a data set similar to the morpheme lexicons described herein, for a match or a similarity, to a desired degree of similarity. Letter matches and letter similarities between an input word part and a portion of a word in the data set indicate a sound match or sound similarity, respectively. With a sound match or a desired sound similarity, for example, a phoneme class match or similarity, the input word part can be associated with data in the data set corresponding to the data set match, and any or all of the data associated with the data set match. In some examples, the data associated with the data set match may include meaning information, which may establish a sound / meaning correlation, as well as optionally additional information depending on the content of the data set that produced the data set match. In the example of the morpheme lexicon 814 described herein, the additional data includes one or more of a morpheme, a morpheme meaning such as a morpheme keyword, one or more literal spellings of a morpheme, position data, syllable count data, negation status data, attested examples, correlations to other morphemes, one or more category identifications, category keywords as well as other desired information. In an example that uses processes such as one or more of those carried out with the sound correlator module 812, or in addition to the sound analyzer module 860, the apparatus, systems and methods can compare a phoneme class structure associated with a word part of an input word with phoneme class structures in a data set, for example a data set similar to the morpheme lexicons described herein, for a match or a similarity, to a desired degree of similarity. Sound class matches and sound class similarities between an input word and a portion of a word in the data set indicate a sound match or sound similarity, respectively. With a sound match or a desired sound similarity, forexample, a phoneme class match or similarity, for example a phoneme class structure match, the input word can be associated with data in the data set corresponding to the data set match, and any or all of the data associated with the data set match. In some examples, the data associated with the data set match may include meaning information, which may establish a sound / meaning correlation, as well as optionally additional information depending on the content of the data set that produced the data set match. In the example of the morpheme lexicon described herein, the additional data includes that described above corresponding to the sound analyzer module 860. Using sound correlations based on sound class information, preferably phoneme class structures, finds a wider and deeper meaning correlation for words, providing more information for the user and for use in a wide variety of other systems.If desired, semantic or meaning analysis can be used to improve the evaluation of words for their meanings, and if optionally combined with one or more other processes, such as processes associated with one or more of the other modules in Fig. 8A and Fig. 8B, it can provide additional meaning information for word analysis, and depending on the form of the database, it can help to provide sound and meaning correlation, for example between an input word part and data set data, and in the preferred examples, between an input word part and morpheme data from a morpheme lexicon. In one example using processes such as any of those associated with the sound correlator module 812, any meaning words associated with the input word, in some examples keywords such as lookup keywords and / or input keywords, can be compared to meaning words in one or more data sets, to a desired degree of similarity, for example to a desired degree of synonymity. In one example, meaning words are compared, and in other examples additionally and / or alternatively synonyms of meaning words are compared, and in further examples additionally and / or alternatively inflections of meaning words are compared. Other methods of comparing meanings between an input word and a data set word can be used, additionally or alternatively.Other methods of comparing meanings or finding meaning correlations between an input word and a data set word can be used, additionally or alternatively. Such meaning correlation can help to expand the evaluation of words in processing. For example, in a variant of the semantic analyzer module 816, the semantic analyzer module 816 can be applied identically as a meaning correlator module 862 carrying out any one or more of the processes described herein for semantic analysis or associated with the semantic analyzer module 816, and when carried out in conjunction with a data set having sound information any results establishing a meaning correlation can also provide a sound correlation for any data set entries having both meaning and sound information. For example, in data sets having entries of both sound and meaning, for example sound class data andmeaning data, finding a match or similarity with meaning can associate the input word or input word part with the sound class data from the entry in the data set. For example, for data sets having some of the data of any morpheme lexicon described herein having sound data, such as phoneme class structures, the meaning correlator module 862 can identify entries in the data set with the same or similar meanings, thereby identifying sound information associated with the meanings, thereby establishing a possible sound correlation between an input word part and an entry in a data set.Any of the apparatus, systems and methods described herein can include processes for combining candidates similar to those associated with a candidate combiner module 864. The candidate combiner uses data from earlier processes, namely input data and any data resulting from any of the prior processes to which the candidate combiner module 864 has access, and arranges any candidate words and / or word parts developed through any of the evaluations or analyses in order. In one example, they are arranged in order of the word parts in the input string, and then reordered according to criteria established by the designer and / or the user. For example, candidates that are incomplete, illogical (e.g., not conforming to established word or language criteria / usage) or fail to meet one or more standards established by the designer / user are eliminated or downgraded, and the remainder are saved as possible solutions, for example in a candidate store 866.The apparatus, systems and methods described herein can also include a solution ranker process similar to any of those associated with a solution ranker module 866, which ranks or prioritizes any results which provide multiple candidates corresponding to the input word. Ranking can be by various criteria, and in the present example, any data associated with the input word, including data developed during any processing of the input word, is evaluated against one or more data sets, and the rankings of the candidates are saved. In one example, the input word and the data developed during processing is evaluated against a morpheme lexicon, and the candidates ranked according to how closely each candidate conforms to an associated entry in the morpheme lexicon. Various criteria can be used, including the extent to which the input word and any input data associated with the input word matches an entry in the morpheme lexicon.The apparatus, systems and methods described herein can include processes that may be similar to any of those associated with a categorizer module 870 for characterizing words and their word parts with categories, to help provide further information about the input word. The categorizer module 870 may characterize a word’s structure and identify the various word parts, characterize one or more word parts as to semantic function, for example, a non-modifying root or for example modifiers, identify the type of modifier associated with the word part, if any, for example, to enable identifying a word part as lacking a semantic function, and / or identify a modifier for a word part asone or more of definite, relative, quantitative, and qualitative. Characterizing words and word parts with categories such as these help to provide further information about the meaning of the word and relevance of word parts in the meaning. Categorizing words and word parts can be carried out according to a number of criteria, some of which may be saved in a data store, such as categorizer, logical premises data store 872. The categorizer module 870 can include one or more processes, including for example one or more of those associated with a word structure analyzer such as the module 840, a semantic function analyzer such as the module 842, and / or a modifier analyzer module 844. The input word can be analyzed structurally by comparing the absolute position of each word part in the word and their positions relative to one another against expectations represented by one or more elements of data in a data set, for example data representing expected positions of word parts and their possible influences on adjacent word parts, and / or whether or not the word part appears as a prefix, root or suffix. Additionally, identifying multiple roots in a word will suggest identifying the root that is not final in a sequence of roots is a modifying root and the last root in sequence is a non-modifying root. Additionally or alternatively, the semantic function analyzer module 842 can be used to identify whether or not the word part has a semantic function, and in the preferred examples of analysis with a morpheme lexicon, the morpheme associated with the word part is checked for the existence of a meaning, for example a morpheme keyword, lacking which the word part would be characterized as lacking a semantic function.The modifier analyzer module 844, whether or not used in conjunction with the modules 840, 842 and / or 870, can evaluate the word and word parts to identify its status as a modifier, identify the modifier type, and / or identify equivalent or similar words or word parts serving the same or similar modifying function. In one example, a word part can be characterized as a non-modifying root or a word part can be characterized as a modifier of the type of one or more of definite, relative, quantitative and qualitative. In another example, meanings associated with the input word, either by way of input meanings or by way of prior comparison processes, can be compared to meanings in one or more data sets 806 to identify possible characterizations that can be applied to the input word. In some examples, a hierarchical arrangement in the data set being used to compare with the meanings can provide additional possibly related information to be associated with the word part, and therefore the word.In preferred embodiments, the apparatus, systems and methods described herein may also include tokenizer processes such as one or more of those associated with a tokenizer module 874. One or more tokens are generated containing one or more elements of data, which may include for example one or more of the input word, phoneme class structures associated with word parts of theinput word, one or more meanings, one or more morphemes, any sound / meaning correlations, and any additional characterizations associated with the input word and / or its word parts, for example those developed through categorizer functions, for example any non-modifying root, and / or modifier types of one or more of definite, relative, quantitative, and qualitative, and if desired any keywords. A token can include related tokens, for example related tokens containing data for respective word parts. An output token can be generated with a token for the word and related tokens for respective word parts. Tokens can be saved in a token store 876, and output can be saved in an output store 878 or transmitted.If desired, processes can be applied such as those similar to ones applied with a token editor module 880. The token editor allows a user, designer or processor to edit a token as desired, and the token and its data, as revised, can be used for additional word analysis, for example, for purposes of research, data development, system development, and the like. Items such as the phoneme class structure of word part, letters of word part, semantic function of word part, modifier type of word part, meaning of word part, relative values of data associated with the word part, and any other data associated with the word part can be edited singly or in combination. The token, as revised, can then be saved or transmitted.Any one or more of the components of the apparatus, systems and methods can output, transmit or otherwise make their data available for use. Processes similar to those associated with an output module 882 can be included in any desirable configuration. System output may include any data or representations received as input or developed during processing, which may be provided at any point of the processing as well as when processing is complete. Output includes saving any or all of the data to the output store 878 and / or applying any or all of the data to a display or other device or devices for review and / or further processing.Any one or more of the modules can operate in conjunction with or otherwise have access to various data stores 801B. As will be described herein, various of the modules can use one or more conventional word data sets 806, such as dictionaries, thesaurus, encyclopedias, synonym tables, antonym tables, and / or etymological dictionaries, or less traditional data sets such as neural maps. One or more of such data sets can be used with one or more of the semantic lookup module 804, the hyphenation lookup module 830, as well as possibly others of the modules. The choice and content of possible conventional word data sets to be used can be based on the requirements and / or preferences of the designer.Additionally, any of the processes described herein can use any suitable media for permanently or temporarily storing data, including any of the data received as input as well as anydata developed or identified during processing of the input. Such data can be saved on a media variously identified as an array 884 or storage for provenance data, and can be any conventional storage device or system, including as discussed herein. Additionally, any one or more of the items identified herein as data stores or otherwise as storage or components for saving data can be combined on one or more devices, distributed in the cloud, on servers, or otherwise as would be understood by one skilled in the art.In one configuration of apparatus, systems, and methods for processing words and word parts, a beneficial data set is one in which sounds of words and / or word parts are associated with respective meanings, and possibly with examples, for example example words in which the sounds are found and which have meanings that are sufficiently similar (e.g., to a desired degree of similarity at the discretion of the data set designer) to the meaning of the word or word part associated with the sound. In the preferred examples herein, such a sound / meaning association data set is a morpheme lexicon 814. In the examples described herein, the morpheme lexicon 814 includes sound information, for example sound class information and in the preferred examples phoneme class information, in entries in the morpheme lexicon wherein the sound information is also represented in words or word parts included for that entry. In one example, the meaning is represented by one or more morpheme keywords. Optionally, the morpheme lexicon can also include one or more example words or phrases, termed herein attested examples, which contain the sound represented by the sound information for the entry and which have a meaning sufficiently similar to the meaning represented by the entry. In one example, the morpheme lexicon includes entries wherein the entry includes a phoneme class structure, corresponding to the input word or word part, a meaning, for example as represented by a morpheme keyword, and one or more example words or phrases, for example attested examples. Additionally, if desired, the sound and meaning data set can include additional data associated with the entry, which associated data can include one or more of position information, negation status, syllable count, category or category type, cross references to other entries, and / or keywords associated with the category or category type, termed herein category keywords. Other data and / or relationships can be included in a morpheme lexicon. The morpheme lexicon 814 can be used in conjunction with one or more of a sound correlator such as the module 812, a semantic analyzer such as the module 816 and / or one or more of the other categorizer functions (e.g. for example as described in conjunction with categorizer 870, analyzer 840, semantic function analyzer 842 and modifier analyzer 844) described in conjunction with Fig. 8B.In another system for analyzing words and word parts (Fig. 8C), process modules 8C02 are used to analyze or evaluate words and word parts for developing and improving a sound / meaning data set, such as the morpheme lexicons described herein, represented generally in Fig. 8C at 8C04. The morpheme lexicon assembly 8C04 is a data store that may be on a conventional media but accessible during design, maintenance and improvement for secure reading, writing and organization of data on the media, which is a machine-readable non-transitory storage device. As noted elsewhere, the morpheme lexicon can be co-located with a user's digital device and / or accessible otherwise if located remotely over a network or similar communications system.The morpheme lexicon 8C04 includes a plurality of entries, each entry of which will be described by way of examples, though each entry is not necessarily populated with the same categories of information in every instance. The description of an entry will suffice to provide information for assembling a usable morpheme lexicon, and in a preferred embodiment one which has data of several types, where the data for a given data type for a given entry represents the lowest common denominator for that entry. In the preferred embodiments, the morpheme lexicon includes meaning data represented by morpheme keywords, and each morpheme keyword preferably represents a lowest common denominator of meaning for the respective entry. Additionally, the entry includes a morpheme having a meaning whose morpheme keyword represents the lowest common denominator of various possible meanings for that morpheme. A further item of data for an entry is a phoneme class structure, which preferably represents a lowest common denominator of sound, for example phoneme class structures for the entry. Moreover, each entry preferably includes one or more examples (e.g., termed attested examples herein) of words or phrases having at least a word part sounding similar to the phoneme class structure, preferably having a phoneme class structure identical to the phoneme class structure of the entry, and having a meaning the same as or sufficiently similar to the meaning of the entry. Finding the lowest common denominator for morpheme keywords and for phoneme class structures relating thereto helps to optimize and form a consolidated data set for the morpheme lexicon.In one example of apparatus, system and methods for adding entries to a morpheme lexicon, revising a morpheme lexicon and / or generating a morpheme lexicon, the system can include a keyword module 8C06, in the illustrated configuration a keyword lowest common denominator reducer module, for evaluating keywords for the optimal meaning based on a sample of instances, each of whose parts have a sound that is sufficiently similar, to a desired degree of similarity, to others of the instances, for example where the sounds are represented by phoneme class structures. A morpheme instance finder module 8C08 evaluates morpheme keyword candidates based onkeywords identified by the reducer module 8C06 for a given phoneme class structure to find the instances having word parts of the indicated sounds having meanings sufficiently close to the identified meaning candidate or meaning candidates, or to one another. The phoneme class structure lowest common denominator reducer module 8C10 identifies candidate phoneme class structures that represent instances having word parts sufficiently similar in sound and meaning to each other for the phoneme class structure to constitute a lowest common denominator among possible phoneme class structures of instances having sounds and meanings sufficiently similar to justify being included as an entry in the morpheme lexicon. In a preferred embodiment, a sufficient number of instances are selected having portions with sufficiently similar sounds, in the present example phoneme class structures, and sufficiently similar meanings to justify serving as an entry in the morpheme lexicon. In one configuration, evaluation of the phoneme class structures is iterated sufficiently, and preferably in coordination with iteration of candidate meanings for the respective instances sufficiently to identify lowest common denominators for a phoneme class structure and a respective meaning for a morpheme to be selected to improve the quality of the morpheme lexicon with reduced overlap between entries while optimizing the number of entries in the morpheme lexicon.With a given morpheme lexicon, additional instances from new words can be evaluated for finding a lowest common denominator of sound, for example phoneme class structure, together with a lowest common denominator of meaning, for example a morpheme keyword so that an additional morpheme can be identified and added as an entry with the phoneme class structure and morpheme keyword. Alternatively or additionally, an existing entry can be reviewed in the context of the existing phoneme class structure, morpheme keyword and attested examples (i.e., the instances selected for the morpheme entry with the respective phoneme class structure and morpheme keyword), and modified and / or updated in one of the three entries, for example to improve or expand the morpheme lexicon. In one example, an existing morpheme lexicon can be supplemented with words from other languages, words from earlier generations or ancient sources to improve or expand the morpheme lexicon.The one or more computers and other digital devices associated with the system for evaluating words do not need to be physically proximate to each other or in the same machine farm. Thus, the computers logically grouped as a machine farm may be interconnected using a local-area network (LAN) connection or a wide-area network (WAN) connection (e.g., such as the Internet or a metropolitan-area network (MAN) connection). For example, a machine farm may include computers physically located in different continents or different regions of a continent, country,state, city, campus, or room. Data transmission speeds between computers in the machine farm can be increased if the computers are connected using a LAN connection or some form of direct connection.Management of the computers and other digital devices may be de-centralized. For example, one or more computers may comprise components, subsystems and circuits to support one or more management services. In one of these embodiments, one or more computers provide functionality for management of dynamic data, including techniques for handling failover, data replication, and increasing robustness. Each computer may communicate with a persistent store and, in some embodiments, with a dynamic store.A computer may include a file server, application server, web server, proxy server, appliance, network appliance, gateway, gateway, gateway server, virtualization server, deployment server, secure sockets layer virtual private network (“SSL VPN”) server, or firewall or other digital device that can execute instructions. In one embodiment, the computer may be referred to as a remote machine or a node. In one embodiment, the computer may be referred to as a cloud.The system and its components, such as a computer, digital device, memory, imaging module, and calibration module, may include hardware elements, such as one or more processors, logic devices, comparison devices or circuits. For example, the system and its components may include a bus or other communication component for communicating information and a processor or processing circuit coupled to the bus for processing information. The hardware elements can also include one or more processors or processing circuits coupled to the bus for processing information. The system also includes main memory, such as a random access memory (RAM) or other dynamic storage device, coupled to the bus for storing information, and instructions to be executed by the processor. Main memory can also be used for storing position information, temporary variables, or other intermediate information during execution of instructions by the processor. The system may further include a read only memory (ROM) or other static storage device coupled to the bus for storing static information and instructions for the processor. A storage device, such as a solid state device, magnetic disk or optical disk, can be coupled to the bus for persistently storing information and instructions.The system and its components, such as a computer, digital device, memory, imaging module, and calibration module, may include, e.g., computing devices, desktop computers, laptop computers, notebook computers, mobile or portable computing devices, tablet computers, smartphones, personal digital assistants, or any other computing device.According to various embodiments, the processes described herein can be implemented by the system or hardware components in response to the one or more processors executing an arrangement of instructions contained in memory. Such instructions can be read into memory from another computer-readable medium, such as a storage device. Execution of the arrangement of instructions contained in memory causes the system to perform any one or more of the illustrative methods and processes described herein. One or more processors in a multi-processing arrangement may also be employed to execute the instructions contained in memory. In alternative embodiments, hard-wired circuitry may be used in place of or in combination with software instructions to effect illustrative embodiments. Thus, embodiments are not limited to any specific combination of hardware circuitry and software. To provide for interaction with a user, embodiments of the subject matter described in this specification can be implemented on a computer having a display device, e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor, for displaying information to the user and a keyboard and a pointing device, e.g., a mouse or a trackball, by which the user can provide input to the computer or other user interface devices. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback, e.g., visual feedback, auditory feedback, or tactile feedback; and input from the user can be received in any form, including acoustic, speech, or tactile input.The network or networks may be any type or form of network and may include one or more of the following: a point-to-point network, a broadcast network, a wide area network, a local area network, a telecommunications network, a data communication network, a computer network, an ATM (Asynchronous Transfer Mode) network, a SONET (Synchronous Optical Network) network, a SDH (Synchronous Digital Hierarchy) network, a wireless network and a wireline network. In some embodiments, the network may include a wireless link, such as an infrared channel or satellite band. The topology of the network may include a bus, star, or ring network topology. The network may include mobile telephone networks utilizing any protocol or protocols used to communicate among mobile devices, including advanced mobile phone protocol (“AMPS”), time division multiple access (“TDMA”), code-division multiple access (“CDMA”), global system for mobile communication (“GSM”), general packet radio services (“GPRS”) or universal mobile telecommunications system (“UMTS”). In some embodiments, different types of data may be transmitted via different protocols. In other embodiments, the same types of data may be transmitted via different protocols.Specifically contemplated implementations can feature instructions stored on non-transitory machine-readable media. Such instructional logic can be written or designed in a manner that hascertain structure (architectural features) such that, when the instructions are ultimately executed, they cause the one or more general purpose machines (e.g., a processor, computer or other machine) to behave as a special purpose machine, having structure that necessarily performs described tasks on input operands in dependence on the instructions to take specific actions or otherwise produce specific outputs. “Non-transitory” machine-readable or processor-accessible “media” or “storage” as used herein means any tangible (i.e., physical) storage medium, irrespective of how data on that medium is stored, including without limitation, random access memory, hard disk memory, EEPROM, flash, storage cards, optical memory, a disk-based memory (e.g., a hard drive, DVD or CD), server storage, volatile memory and / or other tangible mechanisms where instructions may subsequently be retrieved and used to control a machine. The media or storage can be in standalone form (e.g., a program disk or solid state device) or embodied as part of a larger mechanism, for example, any of the types of computers referenced herein, including a laptop computer, portable device, server, network, printer, or other set of one or more devices. The instructions can be implemented in different formats, for example, as metadata that when called is effective to invoke a certain action, as Java code or scripting, as code written in a specific programming language (e.g., as C++ code), as a processor-specific instruction set, or in some other form or language; the instructions can also be executed by a single, common processor or by different processors or processor cores, depending on embodiment. Throughout this disclosure, various processes will be described, any of which can generally be implemented as instructions stored on non-transitory machine-readable media. Depending on product design, such products can be fabricated to be in saleable form, or as a preparatory step that precedes other processing or finishing steps (i.e., that will ultimately create finished products for sale, distribution, exportation or importation). Also depending on implementation, the instructions can be executed by a single computer and, in other cases, can be stored and / or executed on a distributed basis, e.g., using one or more servers, web clients, or application-specific devices. Each function mentioned in reference to the various Figures herein can be implemented as part of a combined program or as a standalone module, either stored together on a single media expression (e.g., single floppy disk) or on multiple, separate storage devices. Throughout this disclosure, various processes will be described, any of which can generally be implemented as instructional logic (e.g., as instructions stored on non-transitory machine-readable media), as hardware logic, or as a combination of these things, depending on embodiment or specific design. “Module” as used herein refers to a structure dedicated to a specific function; for example, a “first module” to perform a first specific function and a “second module” to perform a second specific function, when used in the context of instructions (e.g., computer code), refers to mutually-exclusive code sets. When used in the context of mechanical or electromechanical structures (e.g., an “encryption module,” it refers to a dedicated set of components which might include hardware and / or software). In all cases, the term “module” is used to refer to a specific structure for performing a function or operation that would be understood by one of ordinary skill in the art to which the subject matter pertains as a conventional structure used in the specific art (e.g., a software module or hardware module), and not as a generic placeholder or “means” for “any structure whatsoever” (e.g., “a team of draft horses”) for performing a recited function. “Erasure Coding” if used herein refers to any process where redundancy information is stored, such that the data can be recovered if a memory device or unit of memory is off-line or otherwise inaccessible. “Redundancy information” is used to describe additional data used to permit recovery of data where the underlying data is incomplete or reflects error (e.g., it can encompass an erasure code, used to permit recovery of data when a storage device is inaccessible, as well as error code correction or “ECC” functions). “RAID” as used herein refers to a redundancy scheme that is tolerant to one or more devices or storage locations being offline or otherwise inaccessible, for example, encompassing a situation where “m” page-sized segments of data are received and where “n” physical pages in respective structural elements are used to store data (i.e., where n>m) to permit recovery of the original n segments notwithstanding that at least one of the m respective structural elements is offline; for example, the term RAID as used below is not limited to a disk-based scheme. “Compression” refers to a process where data of a first size is encoded in a manner so that the data can be recovered or nearly recovered from storage of data of a second, smaller size — it includes situations where data can be losslessly recovered as well as situations where there is some loss, such as where data is quantized to enable more efficient storage. “Compression” also should be understood to encompass the functions of decompression, which generally occur using the same structural elements referenced below, but in reverse order as part of a reverse process; for example, a memory controller can be asked to decompress data in association with a data read and thus return a larger quantum of data than natively represented by a logical block address (LB A) range specified by the data read. “Deduplication” refers to a process where attempted storage of identical pages of data at separate memory locations is detected, and where that storage is consolidated into one page or other instance of the data.Similarly, while operations are depicted in the drawings in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. In certain circumstances, multitasking and parallel processing may be advantageous. Moreover, theseparation of various system components in the embodiments described above should not be understood as requiring such separation in all embodiments, and it should be understood that the described program components and systems can generally be integrated in a single software product or packaged into multiple software products.At least one aspect of the present disclosure is directed to a system for evaluating characteristics of words and word parts, for example in processing of language. In some embodiments, the system includes a computer, a memory, and means for comparing words and word parts. The computer is configured to compare a unit of a text string to a sound class of a string in a data set containing at least a plurality of sound classes and determine a desired correlation between them, for example a class match or a class similarity. The computer is configured to compare a unit of a text string to a phoneme class of a string in a data set containing at least a plurality of phoneme classes and determine a desired correlation between them, for example a class match or a class similarity. The computer is configured to compare a unit of a text string to a phoneme class of a string in a morpheme lexicon containing at least a plurality of phoneme classes and determine a desired correlation between them, for example a class match or a class similarity. The computer is configured to compare a unit of a text string to a phoneme class of a string in a data set containing at least a plurality of phoneme classes selected from the group of phoneme classes of vowels, labials, dentals, velars, liquids, nasals, semivowels, sibilants, and aspirates. The computer is configured to compare a unit of a text string to a phoneme class structure of a string in a data set containing at least a plurality of phoneme class structures and determine a desired correlation between them, for example a class match or a class similarity. The computer is configured to generate a correlation pair with either of a letter or a Phoneme Class of the text string that correlates with a morpheme, and with a letter or Phoneme Class from the string in a morpheme lexicon and generating a token with the data, and in one example, the token includes a morpheme. The computer is configured to evaluate a possible sound class correlation of an input string with a string from a data set and evaluate a possible semantic correlation of the input string with the string from the data set. The computer is configured to compare a source keyword associated with a text string to at least one data set keyword and determine a correlation between the source keyword and the at least one data set keyword. The computer is configured to analyze a semantic function of a word-part of a word as a function of an absolute position of the word part in the word, and as a function of an ordinal location of the letters of each word-part in comparison to all of the letters in the word. The computer is configured to evaluate whether the word part is associated with an attested position in an entry in a data set, for example an entry in a morpheme lexicon. The computer is configured to evaluate aposition of a word part relative to other word parts in the word. The computer is configured to evaluate whether a word-part is a non-modifying root, for example where the word-part is a word part corresponding to a morpheme. The computer is configured to evaluate whether a modifying root and modifying affix of a word are one of a relative qualitative, relative quantitative, definite qualitative, or definite quantitative type of modifier. The computer is configured to evaluate one of a word and a word part of the word by comparing data associated with the word and the word part of the word with a category keyword in a data set.BASIC SYSTEM, OVERVIEWIn examples described herein (see Fig. 9), a system 900 can include a Semantic Analyzer 902 that evaluates whether an Input 904 word correlates according to desired criteria to a data set 906 string, for example to a Morpheme, or multiple Morphemes, found within the data set. It can help to enable and / or improve machine processing of text, words, etc., of language, and help users in the same way.A Semantic Analyzer can include one or more of a process 908 for sound correlation between an input and one or more strings in a data set and a process 910 for meaning correlation between the input and one or more strings in the data set. In examples described herein, sound correlation can include a sound class-to-sound class comparison. Letter- to-letter comparisons can also be done, and letter-to-letter comparisons and / or the sound class-to-sound class comparisons can be done before, after, or at the same time as a process for meaning correlation. Any results can then be saved as desired.In some examples described herein, a Semantic Analyzer 902 includes a process for sound correlation between an input and one or more strings in a sound and meaning data set 906 by doing sound class-to-sound class comparisons, and in some examples, the Semantic Analyzer does a sound correlation through phoneme class-to-phoneme class comparisons. In other examples, sound correlation is done through comparing sound symbols from an input 904 with sound features in strings from the data set. Any of the sound correlations can be done separately and followed by one or more processes for meaning correlation and saving any results. Additional processes can be carried out before (912), during, and / or after (914) the meaning correlation.In preferred embodiments the Semantic Analyzer evaluates correlation that may include forms of both 1) sound correlation and 2) meaning correlation. Additionally, pairs of correlated Word-Parts and Morphemes are 3) combined in a sequence (if there are multiple) to create at least one possible solution and 4) the solution(s) is quantified. Examples of such embodiments aredescribed in more detail below (processes 1-4, one or more of which can be operated together or separately).1) A sound correlation may be sought between a part of the word and a Morpheme to form a pair. Such a sound correlation could include, for example, a likeness or similarity to a desired degree between the letters or sound symbols of part of the Input word and the component units of a Morpheme. Additionally, or alternatively, sound correlation may be quantified in terms of an amount or a quality of correlation between the Phoneme Classes of the letters or sound symbols of part of the Input word and the component units of the Morpheme.2) A meaning correlation could then be sought by comparing a meaning of the Input word to a meaning of a Morpheme, where both the Input word and the Morpheme each have an associated Keyword or multiple Keywords that represent their meaning(s). In a Basic Semantic Analyzer 902 System, the Input Keyword or Lookup Keyword, may be either provided by the user during word Input (see, for example, Figs. 42 and 43), or retrieved via a lookup, respectively (see, for example, Fig. 21). In an Expanded Semantic Analyzer System discussed below, Keywords associated with an Input may be obtained via one or more Lookups and an optional Keyword Tool. A Morpheme’s Keyword(s) may be retrieved via lookup in its data set, for example, a Morpheme Lexicon that includes Morphemes and Morpheme Keywords (definition terms).3) Additionally, post processing after the Semantic Analyzer may arrange the correlated Morpheme-Word-Part pairs in a sequence(s), for example aligning them by the letter or sound symbol order of the Input word. There may be only one such sequence possible, or multiple, any of which may be a solution(s). If there are multiple solutions, they may be saved or displayed as Output (listed in some order, for example alphabetically) or further processed, for example for scoring, ranking, or other action as desired.4) Additionally, post processing after the Semantic Analyzer may attempt to quantify one or more solutions to either confirm that a single possible solution is a desirable one, or to assess which of multiple solutions is the most highly correlated. One way of making such a quantification is by scoring each solution by assessing each solution’s overall Sound and Meaning correlation to the Input, as well as its quality (in terms of logical structure and completeness) and letter count, and then, if there are multiple solutions, ranking them by their overall scores. The criteria for the assessment may be set by the user as desired.In a Basic embodiment of the Semantic Analyzer System, the user or processor Input 904 to the Semantic Analyzer system is a word and, optionally, includes an associated Keyword (see forexample Figs. 45-46 and accompanying description). The Output is a solution or set of solutions that identifies the correlated Morpheme or Morphemes to go with the Input.Example: an Input word “deciduous”The following is an illustration of possible processing for evaluating an input word, but is not intended to be exhaustive or the only combination of evaluation and results obtainable through any system or process described herein. The same is true for any of the word or sound symbol examples herein. In preferred embodiments of the Basic System, an Input String is entered in a way as described herein. Additionally, the user enters a Keyword to be associated with the Input word, for example, “falling.” (Alternatively, an Input Keyword can be retrieved through lookup in a dictionary of the Input source language — in this case, English.) The System includes in at least one configuration four processes that occur (in this example, though other variations are possible) in the following sequence (corresponding to the processes 1-4 above): Phonemic Search, Semantic Search, Candidate Combiner, and Solution Ranker. A Phonemic Search process compares the Input String to entries in a Morpheme Lexicon data set, letter for letter. A Letter Match sound correlation is found between a part of the Input String and the following Morphemes: de (from), dec (ten), id (unique), du (two), and uous (an adjectival suffix). Next, the Phonemic Search process also converts the Input String to a sequence of Phoneme Classes and finds Class Match sound correlation to a Morpheme tech / dec (make, construct) and a Class Similarity (with one extra unit) sound correlation with kata / cid (down, separate). A second process, Semantic Search, can pare down these Preliminary Candidates to Final Candidates by searching for Morpheme Keywords that match the string of the Input Keyword (whether it was manually entered or obtained by lookup for example in a dictionary) “falling.” Semantic Search approves the suffix “uous” (it would not benefit from a meaning correlation since this Morpheme happens to have no associated Morpheme Keyword) and searches for either an Exact or a Single or Double Synonym meaning correlation for the remaining candidates. In this case, it is less likely that there would be evidence for a meaning correlation between the Input String’s Keyword and the meanings of the Morphemes dec (ten), tech / dec (make, construct), id (unique), and du (two), whereas the meanings of the Morphemes de (from) and kata / cid (down, separate) would be expected to be confirmed as Final Candidates, given their meaning correlation (“falling down” and “down, separating from” being like ideas). Another process, Candidate Combiner, discovers in this example that a possible combination of these Final Candidates is: de-cid-uous. Since other combinations do not meet Combining Criteria, Solution Ranker accepts the Candidate Combiner’s one Solution and the system saves or displays it as theOutput: the word “deciduous” is comprised of three Morpheme-Word-Parts, namely de (from), kata / cid (down, separate), and uous (adjectival suffix).Phonemic search can provide improved results, for example over word searches alone, giving the user and other processes results that would not have been available without the phonemic search. Additionally, using a Morpheme Lexicon, for example one described herein can also provide improved results over other data sets.EXPANDED SYSTEM, OVERVIEWIn an Expanded embodiment of the Semantic Analyzer System 900 as described above, the user or processor Input 904 is a word that is pre-processed 912 by one or more optional additional components that may either operate as a discrete process or processes or be appended to or accessible by the system to improve the performance of the Semantic Analyzer and its components, including but not limited to the following:• Verifier 914 (see also 828 in Fig. 8A);• Semantic Lookup 916 (see also 804 in Fig. 8A and Fig. 21);• Negation Lookup 916 (see also 818 in Fig. 8A and Figs. 20 and 22);• Hyphenation Lookup 916 (see also 830 in Fig. 8A and Fig. 20); and / or• Keyword Tool 916 (see also 808 in Fig. 8A and Figs. 20 and 21).The above components are enabled by associated data sets 918 that can either be appended to or accessible by the system as defaults or appended and customized by a user, described more fully below. (See Input, Method, Custom Data Sets.) Such data sets, generally referred to generically as data sets or as “reference data sets” or “system data sets,” are distinct from sound and meaning data sets 906, for example the Morpheme Lexicon data set, used for sound correlation and / or meaning correlation.Pre-processing enables the Expanded System to improve correlations and scoring of solutions. For example, inclusion of a Verifier 914 enables improved Input. Inclusion of a Word Splitter 920 process can improve the efficiency and accuracy of Sound Correlation between Word-Parts and Morphemes, analyzing the location of syllables in the Input word so as to better align these with Morpheme syllables. Inclusion of one or more Lookups 916 and a Keyword Tool 916, which together may find and curate a larger set of Keywords associated with the Input, helps correlate Input Keywords with Morpheme Keywords, improving Semantic Correlations between the Input word and Morphemes. Inclusion of processes of the type referenced herein as Word Splitter 920, Negation Lookup 916, and / or Hyphenation Lookup 916 enables scoring solutions by whetherany of the solutions accord with one or more of the evaluated Negation Status and Syllable Count of the Input.In some embodiments of the Expanded system that include, for example, the above pre-processing components, the system could operate in the following order: the system receives an Input 904, a Verifier 914 checks that the Input meets criteria (in at least one example, according to a Verification Criteria data set), and alerts or queries the user if not. When criteria are met, the system forwards the Input to Lookups 916 to gather data relating to the Input word. Lookups might for example collect and / or evaluate a Keyword or Keywords associated with the Input, a Negation Status, and / or a Syllable Count. A Word Splitter 920 could identify the types and, if desired, the sound classes, for example Phoneme Classes of letters or sound symbols of the Input word, and thereby make a Syllable Estimate (in at least one example, according to Syllable Counting Criteria). A String Generator 922 could create from these identified possible Syllables a set of searchable Strings that the Semantic Analyzer 902 (described in the Basic System above, and in more detail below) might compare one by one to items in a data set, for example Morphemes in a data set, which can improve the quality of the results. Other orders of operation are also possible.Data collected or calculated by system processes is saved, for example to a media such as an Array 884. As presented herein, an Array may appear as a table or two-dimensional arrangement of data, for ease of illustration and understanding the examples, but the Array and the data represented in the Array can take any number of forms, as would be understood by those skilled in the art and / or those skilled in information technology. Also, the Array and data represented in the Array need not be located in the same location or in physical proximity, but may be widely distributed but still accessible according to the criteria of the user. In some examples, Semantic Analyzer compares pre-processing data saved in the Array with data in the system’s data set 906, in preferred examples a Morpheme Lexicon, to find combinations of Morphemes that correlate according to desired criteria to the Input 904 with respect to both sound and meaning. It sends its Solution 926 (Output Identifications) as Strings to the Tokenizer 928 and / or the Categorizer 930 for post-processing. (See Figs. 54 and 51-52.)In a further embodiment of the System, the Input is a word that is pre-processed 912, then sent to the Semantic Analyzer 902, then post-processed 914 by one or more optional additional components that may either operate as one or more discrete processes or operate on the Solution(s), including but not limited to the following:• Categorizer 930 (see Figs. 51-52, and see also 870, 840, 842, and 844 in Fig. 8B)Tokenizer (see Fig. 54, and see also 874 in Fig. 8B)• Token Editor (see Fig. 58, and see also 880 in Fig. 8B)The Output of such a system may include either Output Identifications 932, or cognate-finding Output Tokens 934, or both. (See Fig. 55.)In some embodiments of the further System that include, for example, all of the above post-processing components, the Solution(s) (strings with embedded data) is input into a Categorizer 930 process that identifies semantic functions of Word-Parts in the Solution. This Categorizer data is then added to the relevant Output Identifications 932 that were obtained from the Semantic Analyzer’s 902 Solution. The same Solution (or the Solution plus Categorizer data) is input into a Tokenizer 928 process that creates Output Tokens of each of the Solution Strings. These enable the user to find similarities between Objects in the Token (see Fig. 58) and other words (or other types of data) in a data set, to search for example a pre-designated data set , for example a dictionary already uploaded into the system. After one or more of the post-processing components have added their data, an Output is displayed or otherwise made available including that data in the form of Output Identifications (which may include Categorizer data), and cognate-finding Output Tokens (which may include Categorizer data) for the Input Word and each of its parts. In some examples, these Output Tokens can then be edited using a Token Editor, which the system may make available to the user along with Output results.Expanded Example: an Input word “deciduous”The word “deciduous” is Input by any of many optional methods. This Input String is verified by the Verifier’s 914 Criteria as having no disallowed characters (for example hyphens or accents or capital letters) and thus is forwarded to Lookups for processing. (If it had been misspelled “desiduous,” at least one embodiment would query the user whether the system should correct the spelling.)A Primary Semantic Lookup 916 in an English dictionary finds Lookup Keywords associated with the Input: falling, down, from, leaves, disappearing. A Secondary Semantic Lookup 916 may find similar Lookup Keywords in a Latin dictionary under “deciduus” (the closest match to the Input String). A Negation Lookup 916 finds there is a negation in a Definition String of deciduous (“disappear”) and also a negation in the Input String (“de”) but since there is no positive truncated form (there being no such word as “ci duous”) the word’s negation status can’t be confirmed and remains “unknown.” A Hyphenation Lookup 916 looks in a pronouncing dictionary for a matching entry and locates the syllabification “de-cid-u-ous,” thus the Syllable Count is 4.The Input String is divided by the Word Splitter 920 into syllables, first into vowel or consonant letter types: deciduous = consonant, vowel, consonant, vowel, consonant, vowel, vowel,vowel, consonant (CVCVCVVVC) according to Letter Identifying Criteria. Then the vowels are logically identified as certain types based on the identified letter types preceding and following each. In the current example, “deciduous,” the letter types would be further identified as: consonant, Stand-Alone Vowel, consonant, Stand-Alone Vowel, consonant, Vowel Sequence, consonant. Next, Vowel Sequences found are further analyzed — in this case “uou” is transliterated into ancient Greek letters as von. The latter two letters match an ancient Greek diphthong, thus “ou” can be identified as a Syllabic Vowel Sequence, leaving a remainder, “u,” which is identified as a Stand-Alone Vowel. Syllable Counting Criteria determines possible divisions of syllables and, whether they are divided as de or dec, ci or cid or id, du or u, the total Syllable Estimate of the strings is “4.” Depending on the embodiment, the String Generator can create a variety of strings from these, possibly something like: de, dec, deci, ci, cid, cidu, i, id, idu, du, duous, u, uous, ous. Syllable Counting Criteria could, in at least one example, ensure that every syllable includes a vowel. This would help steer the system away from misleading correlations and false positives, for example a correlation between “ec” and a Morpheme “ek” (out, away from), which would orphan the preceding “d” consonant.Next, the Semantic Analyzer 902 retrieves data saved from pre-processing to analyze the word. Phonetic Search might find several sound correlations, including among them, for example, the following:DE is a Letter Match with a Morpheme “DE” from the Morpheme Lexicon.DEC is a Letter Match with “DEC”ID is a Letter Match with “ID”DU is a Letter Match with “DU”DE (one letter missing) is a Letter Similarity with “DEC” DECI (one letter added) is a Letter Similarity with “DEC” UOUS is a Letter Similarity with “IOUS” (one letter different) DEC is a Class Match with “TECH”CID is either a Class Match or a Class Similarity (one letter missing) with “KAT(A),” depending on how “kata” is listed in the Morpheme LexiconDE Letter Match with “DE” (from) has an Exact Match KeywordDEC Letter Match with “DEC” (ten) is not confirmed by a Keyword match ID Letter Match with “ID” (unique) is not confirmed by a Keyword match DU Letter Match with “DU” (two) is not confirmed by a Keyword match DE Letter Similarity with “DEC” (ten) is not confirmed by a Keyword match DECI Letter Similarity with “DEC” (ten) is not confirmed by a Keyword matchUOUS Letter Similarity with “IOUS” is confirmed because it has no associated Keyword DEC Class Match with “TECH” (make, construct) is not confirmed by a Keyword match CID Class Match or Class Similarity with “KAT(A),” depending on how “kata” is listed in theMorpheme Lexicon (down, separate) has an Exact or Single Synonym or Double Synonym Match Keyword, depending on dictionaries used.Letter Match, Letter Similarity, Class Match and Class Similarity are discussed in more detail elsewhere herein.The Semantic Analyzer’ s Meaning Search could confirm that two of the Preliminary Candidate Morphemes, “de” (from, away) and “kat(a)” (down, separate), have associated Morpheme Keywords (included in the Morpheme Lexicon) correlated to Input Keywords and / or Lookup Keywords (falling, down, from) or, depending on the dictionary used for Semantic Lookups, synonyms of these, whereas there is no evidence for the other Morpheme possibilities: “dec” (ten) or “id” (unique). Therefore the two semantic matches (“de / de” and “kat(a) / cid”) are promoted to Final Candidates and the others set aside. The word-part “uous” matches a suffix, which, being semantically undefined (because its semantic value is unknown, for example as indicated by the absence of a meaning keyword in its entry in the morpheme lexicon data set), needs no semantic confirmation and can be promoted to a Final Candidate.Candidate Combiner (see Fig. 42) discovers a possible combination of these Final Candidates is: de-cid-uous. Since other combinations do not meet Combining Criteria, Solution Ranker (see Fig. 42) accepts the Candidate Combiner’s one Solution.If there were multiple possible Solutions, Solution Ranker might still favor this combination because it is confirmed by pre-processing data saved to the Array 884, for example, data respecting letter count, negation status (inclusion of the negative Morpheme “de” does not contradict the “unknown” negation status of the word), syllable count / estimate (found by lookup of the Morphemes in the Lexicon, whose entry data includes a syllable count), sound correlation, meaning correlation, and Quality (the Solution being both logical and complete, because the prefix “de” precedes the root “kat(a)” and the suffix is after the root and there are no letters unaccounted for).The Categorizer 930 evaluates the word structure with attested and relative positions of “de,” “kat(a)” and “uous,” in comparison with Logical Premises to determine that “de” is a prefix modifying “kat(a),” which being the only root is therefore the non-modifying root, and further evaluates the modifier as having a relative qualitative sense (a direction, away from). The suffix “uous,” being semantically undefined, is ignored in the semantic function analysis. Thus a literaldefinition of the word, which could be included as part of the Output, might be rendered as “having the quality of falling or separating down from.”Tokenizer enables the creation of four Output Tokens 934, one for the whole word and one for each part, in this case: “deciduous,” “de,” “kat(a),” and “uous.” The most likely to yield interesting cognate forms would be the root, “kat(a),” which could find words, for examples: catapult, cataract, cede, coincide, decision, suicide. Categorizer 930 identifications (“de” as a relative qualitative modifier, “kat(a)” as a non-modifying root, and “uous” as an adjectival suffix) could also be included in their respective Output Tokens and in the Output Token for the whole word.Output includes Output Identifications 932 as well as Output Tokens 934 for the whole word and each part. The user might for example select an Output Token and perform a search on it, for example by clicking on the Morph eme-W ord-Part, to search for example a pre-designated data set , for example a dictionary already uploaded into the system. Or, to edit any Token, the user could select the Token Editor and the desired Token. Sample Output and Output Tokens:The word “deciduous” literally means “having the quality of falling down or separating from” and is comprised of three Morpheme-Word-Parts, namely a prefix de (from) that is a relative qualitative modifier, a root cid (down, separate, from the Latin form of the Greek preposition Kara, meaning down), and uous (adjectival suffix). It may have a slight negative connotation. deciduous, de, kat(a) / cid, uousMORPHEMESMorphemes identify causal relationships between the literal spellings, sounds, and meanings of words. A morpheme integrates these as one sound-meaning unit. Additionally, a morpheme can distill the related concepts in a collection of words sharing sound and meaning. To date, there has not been an attempt to integrate sound with meaning, or even to use sound by itself or word meaning by itself to improve machine “reasoning,” the focus having been on sentences rather than individual words and word-parts. Anything dependent on concepts — abstract thought, logic and evidence, metaphor, reasoning, persuasion, meaning-making and meaning-lookup in reference works, language and translation between languages, and specific categories of research and knowledge creation, for example literature, science, medicine — can be enhanced when morphemes are included in the process, for example given access to morphemes in an organized data set of morphemes,examples of which Morphemes and Morpheme Lexicon are described further below. (See Morpheme Lexicon.)Morpheme as a Trio of CharacteristicsA morpheme identifies a sound / meaning association that exists in an assembly of letters and their corresponding meaning, and which may be demonstrated by reference to one or more words (described herein as attested examples). The components of a word are usually represented by a literal spelling made up of a set of letters. Because letters are useful for silent tasks, for example writing, reading, editing, and language processing applications, it is easy to overlook the fact that letters also represent sounds. The sounds are used in oral communications, and the sounds representing the letters forming syllables convey meaning, typically on a syllable basis, and the meanings of the syllables combine to produce a meaning of the word. These sounds can be leveraged to discover and compare other features of words and word-parts — including their meanings. As will be demonstrated in the methods and examples described herein, a word’s meaning is a summation of the meanings of its component word-parts. Analyzing words and pairing a word’s component parts to a word’s component morphemes unlocks the meaning of the word as a whole. Additionally, analyzing words through their sounds and meanings allow easier processing of words in language processing.One benefit of using morphemes in language processing is that each morpheme, for example those implemented in a morpheme lexicon as described herein, is a condensation that can encapsulate both a range of sounds (represented by a sound structure, for example a phoneme class structure, which encompasses many possible literal spellings) and a wide umbrella of meanings (which may range from literal concretes through conceptual abstractions and metaphors). The morphemes in one example of a proposed Morpheme Lexicon are implemented, in one example configuration, as an association of a trio of characteristics (see Fig. 10), wherein the trio

[1002] includes 1) a standardized sound description, in the present example that is a sequence of phoneme classes or Phoneme Class Structure of a morpheme, or the PCS(M)

[1004] , 2) at least one meaning description, in the present example that is a morpheme keyword, or the KW(M)

[1006] , and 3) at least one literal spelling, in the present example arising from actual words (here referred to as “instances,”

[1008] and elsewhere — after selected for inclusion in a morpheme data set — referred to as “attested examples”) in which the morpheme is present. As illustrated herein with examples, a morpheme is associated with the foregoing trio of characteristics that can be considered as defining each morpheme uniquely.In a preferred example of a morpheme, and as illustrated with a number of the morphemes described herein, any two of the elements of the trio of characteristics (the phoneme class structure of a morpheme, the meaning of a morpheme, and the literal spellings and meanings of ’’instances” of the morpheme) can be used to derive the remaining characteristic, for a given morpheme

[1002] :1) Given “instances” and a common meaning, a sound description

[1004] may be created for a morpheme having that common meaning (see Fig. 11).2) Given “instances” and a common sound, a shared meaning may

[1006] be derived for a morpheme having that common sound (see Fig. 11).3) Given a meaning and a sound of a morpheme, one can collect “instances”

[1008] (if they exist) having the specified meaning and the specified sound in common (see Fig. 12). A morpheme represents in a preferred configuration a set of words (or word-part components of words) in “lowest common denominator” (LCD) form, a sound / meaning association whose LCD of sound is its phoneme class structure, the PCS(M), and whose LCD of meaning is the KW(M), whose values are discovered by a process detailed further below. To elaborate on the LCD analogy, in a math example, fractions i and 5 / 8 cannot be directly compared because they are in different units, so an LCD, 8, is found and the first fraction converted to 4 / 8 and since 4 / 8 < 5 / 8, ’A < 5 / 8. Analogously, an alphabetic sequence KITT is hard to compare with CHEET. As units belonging to an alphabet, these letters seem equally distinct: KITT CHEET. But a sound commonality does exist between the two syllables, and can be found when both strings are converted into sequences of phoneme classes: the related units are velar units K and Ch, vowel units I and EE, and dental units TT and T. By comparing the phoneme classes rather than the letters of KITT and CHEET, a “velar, vowel, dental” pattern — the phoneme class structure they share — can be seen. Analogously, a word “kitten” whose meaning is “a young cat” and a word “cheetah” whose meaning is “a carnivorous quadruped” have no definition terms in common, but both share keywords (or synonyms of keywords, to some degree) with a word “cat.” Just as LCDs can be used to compare numbers, a sound description and / or meaning description can also be used as a kind of LCD for comparing words. The description below details how the morpheme is a distillation of or “lowest common denominator” of a set of words (or word-parts), along with their spellings, sounds, and meanings, providing a means of connecting a robust set of words from a wide range of contexts. Such a distillation is desirable because it allows more comprehensive processing of language. Today’s training material for LLMs (Large Language Models) is limited to the literal usage examples and meaning connections that have already been made (by human writers), and ignores additionalmeaning connections that may not have yet been made but are logically possible. These can be made accessible by using the sound and meaning association of the morphemes in the proposed data set.Deriving LCDs, of Sound and / or Meaning, from InstancesFor example, Fig. 11 illustrates how a set of words that are potential instances

[1104] of a potential morpheme can be evaluated. A sound comparison process

[1106] discovers whether they contain a common sound, indicative of a PCS(M). A meaning comparison process

[1108] determines whether they contain a common meaning, indicative of a KW(M). The two evaluation processes shown in Fig. 11 are to an extent independent, but can be used to cross-check one another, and may be used iteratively to improve both sets of results. For example, it is important to ensure that the set of “instances” distilled to a common sound overlaps with the set of “instances” that distill a common meaning. Any potential “instances” that seem to fall outside of either the LCD of sound or the LCD of meaning can be discarded.In the sound comparison process

[1106] , the literal spellings of the potential “instances,” words 1, 2,. . .n

[1110] are compared to find units they have in common, which units may be either letters or sound symbols in common (literal correlations like those illustrated in Table B below) or sound classes, which in preferred examples are phoneme classes, of letters or sound symbols (structural correlations like those illustrated in Table C below). The commonality might not be across the whole word, rather might occur in a syllable part of each word instance (as in “kitf ’ of “kitten” and “cheet” of “cheetah”). The common parts, if found, are sounds 1, 2,. . .n

[1112] and are reduced to a “lowest common denominator” of sound

[1114] equivalent to the PCS(M)

[1116] , (The description in Phoneme Class Sequences (Morphemes) below illustrates this process of finding common parts, with examples in Table G.)In the meaning comparison process

[1108] , the meanings of the potential “instances” are compared to find whether words 1 , 2, ... n

[1110] share keywords (or synonyms, to the desired “degree of synonymity,” discussed below) and, if shared meanings 1, 2,...n

[1118] are found, these may be reduced to a “lowest common denominator” of meaning

[1120] equivalent to the KW(M)

[1122] , As has been shown, the “instances” give rise to the Phoneme Class Structure and Keyword that express the sound and meaning association made by a morpheme

[1002] , (The logical entwinement

[1010] of a morpheme’s trio of characteristics is described in more detail in the subsections below after considering some common terminology to be used.)“Degree of synonymity” is used herein as a phrase representing similarities of meaning among words, for example Meaning Correlation when evaluating an input word with a meaningrepresentation from a data set. Degree of synonymity can be used in a number of applications in evaluating words and / or word parts to more easily quantify, for example for machine processing, the existence or absence of meaning similarity, to the desired degree of the designer. For example of one set of criteria, an “exact” match would be a first degree of synonymity, a “single synonym” match (for example a match of a word to a synonym found in a thesaurus lookup) would be a second degree of synonymity, and a “double synonym” match would be a third degree of synonymity (for example a synonym of one word matches a synonym of another), and so on, to the desired degree of synonymity.Overview of terminologyIn the equations and examples below, for simplicity there are often three of each unit (whether the unit is an “instance,” letter, keyword, meaning, phoneme class, word), but the same logic applies even when there are more or fewer than three of a unit. Also, any member of a set of letters to be converted, {Li, L2, L3,. . ,Ln}, may in the description herein be understood to refer to either a single alphabet letter (for example letter F or letter E) or a multi-letter sequence (for example “Ph”). (For example, as illustrated in Table G further below, either the letter C or the multi-letter sequences Ch, Ck, Gh can represent the velar phoneme, and letter E or EA or El can represent the vowel phoneme.) In a word, a letter or multi-letter sequence can represent the respective phonemes of the beginning, middle, and end of a syllable, whereas in a morpheme those same components can be represented in a more general way by a sound class, for example a phoneme class (PC). A phoneme class structure (PCS), then, can describe a syllable as a sound sequence. Letter sequences in words, then, may be converted (as discussed below, see Conversion Instructions) to their respective phoneme classes and expressed as a PCS. Further, the set of respective phoneme classes of the set of respective letters that are included in a set of words can describe, for example, the phoneme class structure common to a set of words that are “instances” of a morpheme with the same sound, or a phoneme class sequence of a morpheme, or PCS(M), as detailed below.Morpheme Phoneme Class Structure1) A first (in no particular order) of the foregoing trio of characteristics of a morpheme is its PCS, or the PCS(M)

[1004] , The PCS is preferably a “lowest common denominator” representation of the sound of a morpheme, including the sound common to its individual “instances.” It represents a standardized form that allows for machine comparison of sounds without the need for actual sound production or audio file analysis. If the PCS is known (for example, given in the Morpheme Lexicondata set), it can be used with either member of the trio to find the third member of the trio. If the PCS is not known, it can be derived if the other two trio members are known. (See Fig. 11.)To calculate a PCS(M) that is a PCS for a morpheme M and comprised of a sequence PCi, PC2, PC3,. . PCn, given KW(M) and a collection of “instances” of the morpheme, namely words ii, iz, i.3. .. , one can find the shared one syllable (see below) of these words that, for example, includes a string whose first, second, third,. . .final letters are LI, L2, L3,...Ln, and one can find via lookup on a Table of Phoneme Classes the phoneme classes of common to (or in some examples, preponderating in) those letters, which belong to the following letter sets: PC(Ll({ii, i2, is. . . })) (that is, the first letter of each of the words), PC(L2({ii, iz, i3... })), PC(L3({ii, iz, i3... })),... PC(Ln({ii, i2, i3... })). The sequence of the phoneme class common to or predominating in the letters in each set is equivalent to the sequence of phoneme classes that describe the morpheme’s sound: PC{Llii, LL2, Lli3... } = PCi of PCS(M), and similarly for the next letters in the sequence. And the resulting sequence PCi, PC2, PC3,. . PCn = PCS(M). Alternatively, one may think of this relationship as PCS({a, b, c,. . n}) = PCS(M), or PCS({ii, iz, i3. . . }) = PCS(M), where a PCS present in or preponderating in the word “instances” collected identifies the morpheme’s PCS. The PCS(M) can then be used, in conjunction with the known value KW(M), to find additional “instances” of words containing the same morpheme. Additionally, calculating a PCS provides a “lowest common denominator” that allows the user to make connections to sounds in many more words.To calculate which sequence of letters is the shared one syllable (for example, in word “instances” that may contain M but have multiple syllables), one can find all the phoneme classes of all letters in all the word “instances” collected, and then identify those phoneme class sequences with the highest preponderance. (For example, in Table G further below, the one shared syllable happens to be the first syllable in each of three word “instances,” decor, docent, and doctor.)For example, as previously illustrated in Fig. 11, if the PCS is unknown, it can be derived from the letter sequence held in common by multiple “instances” of the morpheme. For example, Table G below illustrates multiple “instances” of a morpheme “8SIK deik” (with a common meaning “show”), which include a sequence of the same phoneme classes — dentals, vowels, and velars — spelled in various ways. (The PCS does not include the interpolated letter N found in only two of the many examples sharing a common meaning. When the interpolation is excluded, it can be seen that the PCS is otherwise common to all the “instances.”) Therefore, using the known “instances” and the known common meaning, one can derive the PCS of a morpheme.Morpheme Keyword2) A second of the foregoing trio of characteristics of a morpheme is its KW, or KW(M)

[1006] , A KW(M) is preferably a “lowest common denominator” representation of the meaning of a morpheme and its “instances,” as illustrated in Fig. 11. If the meaning of the morpheme, KW(M), is known (for example, given in the Morpheme Lexicon data set), it can be used with either member of the trio to find the third member of the trio. If KW(M) is not known, it can be derived if the other two trio members are known, as set forth above. Alternatively, the KW(M) may be “known” but of uncertain reliability (for example, listed incorrectly in some other data set), in which case the other members of the trio can be used to find “instances” whose keywords either confirm its correctness (through having some degree of synonymity) or demonstrate some preponderating alternate meaning. Keywords of “instances” (or “instance” keywords) are the source of Morpheme Keywords, and can be collected by a process analogous to Semantic Lookup below. (Extracting keywords from a definition string in a dictionary is standard in language processing.)To derive a KW(M) from a given collection of “instances” that share a common PCS, for example from “instances” that are words {ii, , is. .. }, one can find a set of meanings (via lookup in a data set) that are {KW(ii), KW(i2), KW(i3),.. . } and find also synonyms of the set of these keywords (to the desired degree of synonymity), and, among the resulting set {KW(ii)i, KW(ii)2, KW(ii)3,.. ., KW(i2)i, KW(i2)2, KW(i2)3,. . ., KW(i3)i, KW(i3)2, KW(i3)3,. .. }, one can find the most preponderating (see below) meaning, or KW(M). When such a preponderating KW({ii, i2, i3.. . }) is assigned as a KW(M), it can, if desired, be used in conjunction with the known phoneme class structure value, PCS(M), to find additional “instances” of words containing the same morpheme. Additionally, deriving KW(M) from KW({ii, i2, i3. . . }) provides a “lowest common denominator” that allows the user to make connections to meanings in many more words. (The process of distilling meaning from a set of meanings may either be done with the help of a machine process or manually, or a combination, and may include iteration and fine-tuning for optimal results.)To elaborate on what is meant by a common PCS, where PCS(M) and its word “instances” ii, i2, i3. .. are given, one can assume the “instances” belong to the same morpheme and that PCS({ii, i2, i3.. . }) = PCS(M), as was shown earlier above. To elaborate on what is meant by a preponderating meaning, for example, if a set of 20 word “instances” are collected with the same PCS, and ten “instances” share a keyword X to a desired degree of synonymity of meaning correlation, and five share a keyword Y to a desired degree of synonymity of meaning correlation, and the remainder have five different keywords, keyword X is the most preponderating, and keyword Y has a possibledesired degree of synonymity. (See also Table J further below for examples of degrees of synonymity.)For example, in Table G further below, if the meaning is unknown, it can be derived from meanings common to collected like-sounding “instances.” For example, the meaning “show” is common to the collected like-sounding “instances,” and may be preferred to the narrower meaning “teach / instruct” that is common to a few but not others of the “instances.” Therefore, using the known “instances” and the known common PCS, one can derive the meaning of a morpheme.Morpheme “Instances ”3) A third of the foregoing trio of characteristics of a morpheme is a set of “instances”

[1008] of words that contain the morpheme. “Instances” of morphemes (included, in a preferred example of a morpheme lexicon data set, in an “attested examples” column) are words (or, in some cases, phrases), evaluated by a process illustrated in Fig. 11 described above, that include a similar sound and have a similar meaning. In fact, the array of literal spellings and meanings in real words are what give rise to the sound / meaning association that a morpheme identifies. If “instances” are given, they can be used with either member of the trio to find the third member of the trio. If “instances” are not given, they can be collected if the other two trio members are known, as follows.“Instances” can be found by comparing the given sound PCS(M) to sequences of letters included in words, for example words in a data set, whose meanings resemble the given meaning KW(M) to some degree of synonymity. If in an example (see Fig. 12) a PCS(M)

[1202] is comprised of three phoneme classes PCi, PC2, PC3

[1204] (in other examples, fewer or more phoneme classes are possible), one can find via lookup on a table of phoneme classes

[1206] the set of letters belonging to those three phoneme classes: {L(PCi)}, {L(PC2)}, and {L(PC3)}

[1208] , Then one can search a word data set

[1210] for a string

[1212] whose component letters include a sequence of one letter (which, as described above, may be a multi-letter unit) from each set, or one from {L(PCi)}

[1214] , one from {L(PC2)}

[1216] , and one from {L(PC3)}

[1218] , in that order; and find via lookup in the same or another data set, for each of the word results { ii, i2, is. .. }

[1220] , keywords for each: {KW(ii), KW(iz), KW(i3),... }

[1222] , to the desired degree of synonymity. Then each set of keywords {KW(ii)i, KW(ii)2, KW(ii)3,. .. }, {KW(i2)i, KW(i2)2, KW(i2,... }, {KW(i3)i, KW(i3)2, KW(is)s, .. . } can be compared

[1224] to the given KW(M)

[1226] , and those example words whose keyword set includes at least one keyword with a meaning similar to KW(M), where such equivalence is to some degree of synonymity, can be saved as “instances”

[1228] of the morpheme.For example if PCi = {B, P, Ph}, PC2 = {A, E, I, O, U}, and PCs = {C, CC, G, K, Ch, GG, Gh, Q} then the strings sought would include for example the string “P, E, G.”For example, if “instances” are unknown but a morpheme with a known PCS and KW is given, for example, to populate Table G further below with word “instances” of the morpheme “deik,” words can be sought that contain the known PCS of that morpheme — a sequence of a dental letter, {D, DD, T, TT, or Th}, a vowel letter or letters, {A, E, I, O, U or any syllabic vowel sequence}, and a velar letter, {C, CC, CK, CH, GG, GH, K, Q}, and those words that also have a given meaning synonymous with “show” (to the desired degree of synonymity) can be considered “instances” of that morpheme. Therefore, using the known PCS and KW of a morpheme, one can derive “instances” of that sound and meaning association.Note that determining whether a word is an “instance” of a Morpheme is made easier by establishing some criteria, for example the criteria that a keyword associated with the “instance” matches the KW(M) (to some desired degree of synonymity) increases the likelihood that the morpheme is actually a “meaningful” component in a word “instance” and not merely an incidental occurrence of letters. In preferred examples, lack of meaning correlation (to some preferred degree of synonymity) excludes a word from being an “instance” of a morpheme. For example, although the letters D, E, C, appear in sequence in the word “deciduous,” they do not have the same meaning as the Morpheme with that PCS, “8SK dec,” which means “ten.” Therefore “deciduous” would not by this criteria be considered an “instance” of the Morpheme “8SK dek.” The foregoing example illustrates how such criteria, sound correlation combined with meaning correlation, function as a two-factor authentication that an “instance” is an “instance” of a morpheme.As discussed above, a morpheme associates a sound with a meaning, or PCS(M) t= KW(M). This association arises from the fact that certain sets of words, for example {ii, i2, is,. .. }, share sounds and have similar meanings — a fact discoverable by comparison of the respective phoneme classes of their letters, or PCS({ii, i2, is,... }), and of their keywords, KW({ii, i2, is,... }). Where the sounds and meanings of a set of words are consistent with the sound and meaning of a morpheme, such a set of words can be said to be “instances” of that morpheme, expressed in a logic equation as:Thus the morpheme and its “instances” have a sound and meaning association, illustrated in Fig. 10, expressed in a logic equation as:PCS(M) KW(M) {I}(M)As a corollary, given words that happen to include the sound of a morpheme (rather than given known “instances” already associated with a particular morpheme), one may expect some of the words to have meanings akin to the same morpheme. And given words that happen to have meanings akin to a morpheme, one may expect some of those words to also share a sound in common. In sum, sound commonality makes meaning commonality (to some degree of synonymity) more likely. And, conversely, meaning commonality is often accompanied by sound commonality. These facts, expressed as follows, can be used to make predictions (see the concluding paragraph below):If PCS({Wi, W2, W3, ...Wn }) = PCS(M), it is likely KW({ Wi, W2, W3,. .. Wn }) = KW(M).If KW({Wi, W2, W3,.. . Wn }) = KW(M), it may be true that for some of {Wi, W2, W3,... Wn } PCS({Wi, W2, W3,. .. Wn }) = PCS(M).As a corollary, where both the sounds and meanings of a set of words are consistent with one another, it is possible that the sound and meaning association is explained by the existence of a morpheme (but see also Assembling Morpheme Data for other considerations in the selection of morphemes for inclusion in a data set):Since PCS(M) KW(M) {I}(M), then if both PCS({Wi, W2, W3,. . .Wn }) = PCS(M?) and KW({W1, W2, W3,...Wn }) = KW(M?), {Wi, W2, W3,...Wn }(M) = {I}(M?).An additional benefit of the logical relationship of the trio of characteristics described herein is that, given a literal spelling and a meaning of any term, for example given a term spelled with letters Tl, T2, T3,.. Tn, and a meaning KW(T), one can use PC({T1, T2, T3,. ..Tn}) found via lookup to calculate PCS(T), and use the PCS(T) as the sound of a posited morpheme to see whether there are additional word “instances” with sound PCS(T) and meaning KW(T).The trio of relationships expressed in the foregoing will help identify morphemes by what may be considered their most important characteristics — PCS(M), KW(M), and the literal spellings and meanings of collected “instances” of words containing such morphemes. Using this trio to populate the Morpheme Lexicon data set enables a new and efficient way to use machine processing to collect and compare words by their sounds, letters, spellings, and meanings, which should have many implications for future language processing applications. As one example, these relationships can be used to make predictions about the semantic and sound relationships between any two or more words.MORPHEME LEXICONExamples of a language processing apparatus, system and method, for example the combination designated 1218 and 1520 (Figs. 13, 15 and 16) as a morpheme lexicon, for use in evaluating one or more words and / or word parts and in assigning characterizations in the form of sound / meaning characteristics, negation status, and categories can take a number of configurations. In one example, they include systems and methods represented by the morpheme lexicon, for example illustrated in Table D with some sample entries and data. It is understood that the systems and methods represented by 1218 and 1520 may be instantiated with modules not described herein, some or all of which may include conventional elements. One or more of the apparatus, systems and methods represented by 1218 and 1520 may be individually or together implemented as desired, or additionally or alternatively they may be combined with any one or more of the other apparatus, systems and methods described herein. They individually and together help to provide information about a word and its word parts.An information collection, for example a database or data set of words or word-parts, forming a collection of language features having two or more of the elements of the trio of characteristics improves language processing. The Morpheme Lexicon is such a data set forming a collection of language features called morphemes, whose sounds and meanings can be useful for analyzing words to improve language processing. The data set includes representations of data, for example a plurality of morphemes. While the data set can be assembled from one or more of a variety of data collections using one or more of a variety of processes, a preferred Morpheme Lexicon will include the morpheme, relevant sound information, and relevant meaning information. For each of at least some of the plurality of morphemes, and in some examples a majority of the morphemes in the data set, the morpheme includes representations of at least one literal spelling of the morpheme, a sound sequence, and at least one meaning. Additionally, a further preferred Morpheme Lexicon will also include a literal spelling and attested examples. Such features will allow better processing of words by language processors. However, these features for a Morpheme Lexicon are not exclusive of other information, and as presented in other subsections below, additional associated data for respective morphemes can further improve language processing (see for example, Morpheme Data). In preferred configurations, the data set has one or more of the characteristics of the Morpheme Lexicons described herein.A morpheme is a unit of language of at least one syllable, with a Phoneme Class Structure and semantic value. Morpheme, from the Greek word popcpr] morphe, literally means “form.” (For a further elaboration of morphemes as language features, see the previous discussion Morpheme as aTrio of Characteristics. For the previous discussion of how sound is represented by Phoneme Classes and their Structures, see Phonemic Search.) A morpheme lexicon, including any of those described and illustrated herein, can improve language processing to more accurately evaluate words and to expand results (see Fig. 39

[3902] and

[3904] ) available through language processing.Morpheme SoundsTo make a sound, a morpheme’s syllable will contain at least one vowel, the “tone” of the sound or intonation. The sound of a morpheme can be represented by a sound sequence presented 1) as a general pattern of vowels and consonants, or 2) by the particular sound classes that make up the syllable, or 3) literally, by using a set of letters (or sound symbols) that make up the syllable. The most common pattern in morphemes of such a sound sequence is a pattern of beginning phoneme, a middle vowel, and ending phoneme, that is the (usually three-letter) consonant-vowel-consonant (CVC) pattern representing respectively the syllable’s beginning, intonation, and ending. The CVC pattern is characteristic of “Indo-European roots” and common to some examples herein that are referred to as Root Morphemes, both of which tend to be monosyllabic. (Other patterns are possible because the vowel sometimes has a double function, serving as both the middle of the syllable and the beginning, or the middle of the syllable and the ending, or in rare examples the vowel is the only part of the syllable.)The characteristic CVC pattern of vowels and consonants can be further helpful to processing by noting that they belong to particular sound classes, for example Phoneme Classes, for example the vowel class and the eight consonant classes described herein. And each of these sound classes can be further organized by grouping with particular letters, for example single letters or multi-letter combinations (for example, the labial class might include B, F, P, Ph, and V). It is noted that while the most common pattern of vowels and consonants in morphemes is usually a three-letter sequence, orthographic changes for certain phonemes (a single Greek letter < is represented by two letters PH in English, and in later English one letter F), interpolations (of a letter L, N, R, or S), and truncations or elisions (necessitated when certain letters are juxtaposed) may cause some morphemes to have fewer or more than the usual three letters. By assigning sound classes to elements of a word data set, in the present example phoneme classes in the Morpheme Lexicon, language processors can easily process sounds of words and expand the amount of information that can be developed during the processing of words.The sequence of Phoneme Classes that describe a syllable creates what herein is called a Phoneme Class Structure (PCS). In the preferred embodiments, Phoneme Class Structure (PCS) is asound description of a morpheme, and is one of the trio of identifying characteristics of a morpheme (as described previously and which can be found by methods analogous to those described in the subsection titled Morpheme as a Trio of Characteristics). The benefits arising from the PCS expression of a morpheme as opposed to the general CVC form and the particular literal spellings has been discussed previously. (See Phoneme Class Sequences (Morphemes).) The Phoneme Class Structure of a Root Morpheme is fairly stable even when affixes are appended before and / or after the Root, which makes Root Morphemes relatively easy to recognize and differentiate from other morphemes. The Phoneme Class Structure can be included as data associated with a Morpheme entry in the Morpheme Lexicon data set, or calculated separately for a list of letters for a supported alphabet, for example using conversion instructions. (See Conversion Instructions.) In the Morpheme Lexicon examples shown in the tables herein, the Phoneme Class Structure is the organizing feature of the data set and appears in column one. Note: the illustrations and examples discussed herein presume a morpheme lexicon that supports English-language Input (for example, the Morpheme Keywords are in English), but the lexicon can be designed to support any one or any combination of languages. (See Supported Languages.)Table A, below, gives examples of possible Root Morpheme entries in typical monosyllabic CVC pattern and that have been selected to be included in the Morpheme Lexicon data set (see Table D further below).TABLE A: Sample Entries in a Basic Morpheme LexiconWhile CVC is a reliable pattern, not all morphemes follow the CVC pattern. Affix-position morphemes can have one, two, or even three syllables, resulting in a less consistent vowel / consonant pattern for affixes. Perhaps due to pronounceability (a function of sound for speech), affixes (unlike roots) often follow a VCV or CV or V or VC pattern — presumably to append more smoothly to either side of a root in what would otherwise be a CVC pattern. For example, when a prefix, root, and suffix are combined in a word, the effect is more pleasing and pronounceable if consonants alternate with vowels. For example, the alternating vowels and consonants of “oh, la la la” (V CV CV CV) is easier to say than the awkwardly clustered consonants in “strum strum strum” (CCCVC CCCVC CCCVC). Since there are probably more Root Morphemes than Affix Morphemes and Root Morphemes tend to have a CVC pattern, in preferred examples the set of Morphemes included in the Morpheme Lexicon will include a preponderance of CVC patterns, but other forms may be selected. (Examples of Affix Morphemes are included with the Root Morpheme examples in Table D, below.)Morpheme MeaningsA morpheme is representative of and encompasses an array of meanings for a class of objects (the “instances” it represents and in which it is found for example as a syllable), and is by nature more conceptual in meaning than the meanings of the concrete entities, actions, or qualities described by the attested examples of the morpheme. Morpheme meanings can be included in a data set, for example the lexicons described herein, and associated with their respective sound data (for example phoneme class structures and literal spellings). As illustrated in Table A above, the data set can be populated with meaning data that is either a single word or a plurality of words, as the case may be, selected according to criteria established by the user. The meaning data, also referred to herein as Morpheme Keywords, are found by way of processes , for example those described above in Morpheme as a Trio of Characteristics and described in more detail below (see Assembling Morpheme Data and see also Morpheme Data). The meaning data can be used as one entry point into the data associated with the respective morpheme in the data set.In preferred examples, morpheme meanings are distilled into one or a few keywords chosen with reference to reliable related information. In one example, the morpheme meanings developed from any selected sources are evaluated or compared to each other to more clearly differentiate morphemes from one another, for example to reduce possible overlap between morphemes selected for the data set and their meanings with a goal of enhancing distinctness between the morphemes. In a further example, morpheme meanings are chosen with reference to the meanings of at least two “instances” containing the morpheme, which “instances” are found by methods analogous to thatdescribed above in the context of the trio of characteristics of morphemes. (See Morpheme as a Trio of Characteristics; for a more detailed discussion of “instances,” see Assembling Morpheme Data, below.) In a preferred example, the meaning of the morpheme represents information that is the “lowest common denominator” available, for example based on a high frequency meaning associated with or common to such “instances,” providing a more concise and more accurate data set for processing.As examples of morphemes having a conceptual rather than particular meaning, see the Morpheme Keywords column of Table A, above, for example: “want, show, down, with” are broad terms, applicable to many contexts. The more abstract Morpheme Keyword meaning is preferred so that it can represent or encompass the more concrete meanings of the same morpheme’s attested examples.Assembling Morpheme DataThe contents of a morpheme lexicon besides the morpheme, sound information, meaning information, and preferably spelling information, includes in most of the examples described herein, additional associated data, for example attested examples. The contents can be developed in a number of ways (herein, the “morpheme selection process”).A possible morpheme selection process is a “top-down” approach (see Fig. 13) in which the system designer conducts research on one or more existing data collections that are possible sources of candidates for inclusion in a morpheme lexicon, for example as discussed herein or in other ways that would be understood after reviewing the present examples. Any one or a combination

[1302] of data collections can be used. Possible data collections can include scholarly reference works , for example lexicons

[1304] and etymological dictionaries

[1306] and lists of “Latin,” “Greco-Roman,” “Proto-Indo-European,” or “Indo-European” roots, morphemes, stems, bases — for example a list of roots

[1308] , Some data collections may be lists and others may include entries that either explicitly name or make reference to roots, prefixes, and / or suffixes of words. The contents of any or all of such data collections can be evaluated and used to populate part or all of the Morpheme Lexicon data set. For example, the contents may include literal spellings

[1312] and / or meanings

[1314] of words or parts of words that may relate to potential candidate morphemes. The system designer can assemble this data

[1310] as desired to build a set of candidate morphemes. For example, meanings supplied by a data collection can be assessed as possible morpheme keywords, or as secondary keywords. For example, a data collection may associate a particular literal spelling (in a word or a word-part) with a meaning, and this association can be saved as a possible attested example andsecondary keyword of a morpheme being considered for inclusion. A secondary keyword is one associated with an attested example. The system designer might further evaluate each morpheme candidate as to its phoneme class structure. For example, once a PCS of a morpheme has been identified, it can be used to find further examples that can then populate the lexicon, as illustrated in Fig. 12 described above. The system designer might further evaluate each morpheme candidate as to possible other additional associated data, for example position, syllable number, negation status or other data

[1316] , as desired. Other data collections or other system resources and descriptions (for example Morpheme as a Trio of Characteristics, Conversion Instructions, or Negation Lookup) may be helpful in deciding how to optimally populate such data. The data collection(s) thus serve as one possible starting point, and the system designer makes further evaluation, selection, and fine-tuning to the desired level of detail, to assemble and populate the Morpheme Lexicon

[1318] ,Once the data collection or collections have been selected, the process of selecting candidates for the morpheme data set may involve several steps of evaluation and selection, though the steps, their order and their frequency may vary depending on the process and the data set(s) involved. As discussed above, a morpheme need not be restricted by any one literal spelling, rather it may be selected because it expresses a general sound / meaning association, and which may be more fundamental than the meaning of the words it may represent. As illustrated in Fig. 14, the data collections

[1402] can be used to identify potential candidates of morpheme “instances,”

[1404] whose meanings

[1406] may be obtained from entries in the same or another data set. These “instances” and meanings can be a starting point for evaluating candidate entries for the lexicon

[1408] , As discussed earlier, “instances” can be evaluated to see whether they have meanings in common (which in this example are given in the data set) or sounds in common (see Fig. 11 [1112-16]), for example, sounds expressing the Phoneme Class Structure

[1412] of a morpheme

[1414] , “Instances” with common meanings that contain common sound structures can be included

[1416] in association with their respective morpheme in the lexicon.The “instances” selected for inclusion help to show a relationship between the morpheme and words in usage that include the morpheme, as illustrated in Table A above showing the mutual dependency of morpheme meaning and “instances.” Once sounds are associated with a general pattern of vowels and consonants (a word or word-part), evaluating a meaning or meanings for such an association uses additional criteria. As a morpheme candidate is evaluated, “instances” may be identified from data collections that correspond to the morpheme candidate, in which case such “instances” may be included as attested examples (described more fully below) if the morpheme candidate is ultimately selected for the Morpheme Lexicon. If the “instances” are determined to notcorrespond to the morpheme candidate, they may be included with a different morpheme candidate as an attested example after further evaluation, or omitted from the Morpheme Lexicon.In one example of using a scholarly source, for example a data collection of ancient Greek words, one could examine in a lexicon all un-prefixed forms of ancient Greek words and find among these, sets of words that share a sound and meaning association, to find Root Morphemes, followed by an examination of the affixes most commonly added to those roots in Greek words (namely, numbers and prepositional prefixes). Other possible data collections include etymological dictionaries

[1302] and existing reference collections or lists of “Latin,” “Greco-Roman,” “Proto-Indo-European,” or “Indo-European” roots, morphemes, stems, or bases

[1303] ,In a more general example, using one or more data sets to assemble a morpheme lexicon, a data set of words in a modern dictionary, for example the Oxford English Dictionary, can be converted to Phoneme Class strings (see Fig. 33 for an example of converting a word part string to a Phoneme Class string) and keywords (wherein the keywords can be derived or selected from the definition strings in the dictionary). The Phoneme Class strings and the keywords can be compared to entries (or parts of entries) in a digitized ancient Greek-English lexicon (see Fig. 12 for an example of comparing Phoneme Class strings to words in a data set), to find common Phoneme Class Structures and common keywords that represent potential morphemes that associate those sounds and meanings. (See also Fig. 15.)Another possible process of building a morpheme lexicon is by a “bottom-up” induction method. For example, existing word entries in reference works may be analyzed for recurring letter strings that may be morpheme candidates. For example, candidate prefix morphemes are easiest to identify because they are attached to the beginning of multiple words whose entries (in dictionaries, for example) are listed alphabetically by first letter. Prefixes, once identified, could be truncated from their entries and then the remaining letters of that entry manually or otherwise evaluated as to whether each is a candidate root morpheme (or other candidate prefix morpheme). Additionally, the same truncated (non-prefixed) forms of the words can be found in the same dictionary listed by the first letter of the resulting root (or second prefix). The same process could be repeated starting from the ends of words to find candidate suffix morphemes.In the foregoing example for identifying prefixes, a data collection of ancient Greek words could be analyzed for prefixed and un-prefixed forms of words, which task is an easy starting point because the prefixes begin the word and the words are listed alphabetically. Any prefixes noted can be included as candidate prefix morphemes (for examples, numbers and prepositions). Additionally or alternatively, the same prefixed and un-prefixed forms of words can be further analyzed to findsets of words that share a sound and meaning association and if a set is found having such an association, these words can be attested examples for a candidate root morpheme. Additionally or alternatively, an examination can be made of the suffixes added to those roots in Greek words. Any or all of the foregoing can be considered for possible inclusion in a morpheme lexicon. As an example of prefixed and un-prefixed root forms, a set of words starting with the same four-letter string “anti” begins the prefixed forms avriPaivcn antibaino, avnOsco antitheo, avruiAsoi antipleo, whose corollary, un-prefixed root forms are 0aivo) baino, Osco theo, TIFSOJ pleo.Fig. 15 illustrates another possible inductive process for compiling a morpheme lexicon, for example, one might transliterate words in a modern dictionary

[1502] , for example the Oxford English Dictionary, to Phoneme Class strings and keywords (from the definition strings)

[1504] , compare transliterated strings

[1506] to entries (or parts of entries) in a digitized ancient Greek-English lexicon of Greek entries with English definition strings

[1508] , for example Beekes’ Etymological Dictionary of Greek and compare the keywords of the English entries

[1510] to the keywords

[1512] in the definition string of the Greek entries to find meaning correlations to some degree of synonymity (using a thesaurus

[1514] if desired) to find common Phoneme Class Structures

[1516] that represent potential morphemes

[1518] to include in the Morpheme Lexicon

[1520] , This comparison process has the benefit of tailoring the morphemes to an expected language of Input — in this example, English. A like comparison process can be used with dictionaries of other languages to support a different expected language of Input, for example for separate morpheme lexicons or for one morpheme lexicon that supports both languages. For example for a single morpheme lexicon, meanings and / or attested examples, discussed further below, may include words from both languages.Where a Phoneme Class Structure of a dictionary word matches that of an entry in the lexicon, the two can be evaluated or saved to be evaluated for possible similar meanings, in ways described herein (for example for an exact match or to a desired degree of synonymity).In another example of generating a morpheme lexicon, a variety of possible rulesets might be implemented in order to compile the entries of a morpheme lexicon from other word sources, for example where a specialized data set, for example an ancient Greek-English lexicon, is not available (for example, for the desired language) or has already been used. The rules in the ruleset will determine the types of morphemes included and their effectiveness in analyzing the meanings of input words. Such rules and the resulting entries and the associated data can then be fine-tuned to enhance the capabilities of the Morpheme Lexicon.Additionally or alternatively, a Morpheme Lexicon might be compiled by finding, in entries or parts of entries in one or more data collections, a “lowest common denominator” of sound (for example, a shared Phoneme Class Structure) and of meaning (for example, a common keyword, shared to some degree of synonymity). In a basic example for illustration, which uses the induction method, four words can be found in the Oxford Classical Greek Dictionary that share the literal string “apyvp” argur and share the Keyword “silver,” so the sound of “argur” can be considered to mean “silver” and all four considered to be attested examples sharing one literally (letter-by-letter) correlated morpheme.TABLE B: Attested Examples, Literal CorrelationsFor example, to expand on the literal correlations in Table B above, words can be sought that sound similar, for example correlate structurally with a phoneme class structure. For example, words having the vowel-liquid-velar pattern of the syllable “arg” can be evaluated to see if they are in any way common to the other words in Table B. Table C below shows possible results of such an expanded search, and includes additional words whose meanings now range more widely while still having the same phoneme class structure. Therefore, the phoneme class structure has led to additional words attesting to a potential candidate morpheme for inclusion in the Morpheme Lexicon. Note that as the set of attested examples expands, the meaning broadens. Whereas the original set of word meanings are “of or having to do with silver”, once the set of words is expanded to include phoneme class structure matches, the meaning broadens to include the mining of silver, its use as money in accounting, debt, credit, the work it enables to be performed, its influence on power, status, competition, and government, and the defense and security of wealth generally. The commonality or “lowest common denominator” of sound can therefore be a clue by which broad or fundamental commonality of meaning may be traced, and which can be used to develop candidates for a suitable data set, for example the preferred Morpheme Lexicons described herein.TABLE C: Attested Examples, Structural CorrelationsTherefore, regardless of the specific method used to assemble the Morpheme Lexicon, the method preferably uses the relationship between words and their word-parts: as words are a sum of the meanings of their parts, the respective meanings of word-parts can be used as clues to word meaning, and vice versa. Candidates for whole-word meanings and word-part meanings may come from the same data set, or combined from different data sets. Using methods described above for processing words, for example Verifier and / or Word Splitter, whole words and word-parts from the data sets may be evaluated to find “instances.” By reviewing the “instances” and their meanings, which may be represented by keywords from definitions of words that are “instances,” candidate meanings for each word-part can be evaluated. Likewise, sound information including sound characteristics of those “instances” can be calculated or derived to produce a most-likely Phoneme Class Structure based on the most common word-part configurations. Those “instances” containing the same Phoneme Class Structure and also having sufficiently similar meanings can be selected as attested examples to represent a morpheme candidate. The common meaning(s) can be evaluated for selecting a Morpheme Keyword that is the best meaning for the morpheme that is consistent with the keywords of the attested examples. (The meanings of attested examples can optionally be included as “secondary keywords” and supplement Morpheme Keywords.) Other possible “instances,” wordparts having the same Phoneme Class Structure, can also be collected (from the same or other data set, concurrently or at another time) and evaluated for a possible common meaning to possibly expand the attested examples, and / or more closely resolve or fine-tune the Morpheme Keyword meaning.The process of evaluation of whole words, word-parts, and “instances” to find common Phoneme Class Structures and common meanings that are candidate morpheme sound / meaning associations can be continued as desired until the user / designer is satisfied with the results, for example taking into account the considerations described herein. Other word-parts of the evaluated words can be processed in a similar way, and other words containing those respective word-parts can likewise be evaluated. With sufficient “instances” containing a Phoneme Class Structure and having sufficiently common meanings, a morpheme can be selected, a literal spelling assigned, a Morpheme Keyword selected, and attested examples associated. Moreover, once the Phoneme Class Structure has been identified to a sufficient confidence level, it can then be used to search and evaluate other words having word-parts with the same Phoneme Class Structure, for example by a process described earlier (see Fig. 12). Additionally, for a given morpheme, the attested examples can be evaluated further to provide associated data, for example position data, syllable count, negationstatus, category, secondary keywords, and any other desired associated data for possible inclusion in the Morpheme Lexicon.Additionally, because definitions of potentially related words may not be identical, criteria are used for comparing meanings, which in the present examples include a selected degree of synonymity between them. If the “instances” have meanings equal, to the desired degree of synonymity, to the estimated meaning of the morpheme, and if they include a syllable having the same phoneme class structure as the morpheme candidate, the “instances” will be selected as attested examples for the morpheme, and a Morpheme Keyword will be selected that is the best meaning for the morpheme that is consistent with the attested examples and their keywords.After a desired evaluation, in preferred configurations, as discussed in Morphemes, above, morphemes are selected to populate the lexicon with data that includes, for each entry, a morpheme that is a component in two or more attested examples that share both a Phoneme Class Structure (preferably included in the data set with the morpheme) and a meaning (where the morpheme meaning is represented by the shared meaning, which is a correlation to some degree of synonymity set by the user / system designer). Such a Morpheme Lexicon is an improvement over other systems because it uses sound / meaning associations among words and word-parts to populate the Morpheme Lexicon, and the resulting contents can enable improved language processing.The process for evaluating keywords of “instances” and a possible Morpheme Keyword may involve multiple cycles of evaluation. A meaning correlation process, which can be used for comparing keywords of “instances” by various degrees of synonymity, has already been described (see the discussion of how keywords are correlated in Morpheme as a Trio of Characteristics, Morpheme Keyword, and see also the analogous process in Semantic Search). The same meaning correlation process can be used for comparing keywords of “instances” to one another. However, there may be several ways to evaluate keywords of “instances” for selecting a final Morpheme Keyword. In one example, when populating the data set with morphemes, morpheme meanings may be selected from among keywords of “instances” (and keywords of synonyms thereof to the desired degree of synonymity), as determined by the user / designer, and the Morpheme Keyword selected based on a statistical frequency of a meaning occurring in keywords of “instances” (for one example of such a process, see Keyword Tool described herein). Alternatively, instead of using a higher frequency keyword, there may be a more abstract, “umbrella” term encompassing what the keywords of “instances” have in common. The more abstract Morpheme Keyword meaning if selected will then encompass the more concrete meanings of the keywords of the “instances” that become the morpheme’s attested examples. The process of distilling meaning for a MorphemeKeyword from a set of secondary keywords may either be done with the help of a machine process or manually, or a combination, and may benefit from iteration and fine-tuning. As an example of fine-tuning, after finding the broad meaning “depth” for the morpheme “ ot9 bath,” this morpheme could be compared with another entry that happens to have the same phoneme class structure: “TW.Q path” (feeling). If the system designer decided the two morphemes are similar enough that this broad meaning encompasses both sets of attested examples, they could be combined into one entry. Alternatively they could be cross-referenced, as shown in Table D below.In a number of examples of a Morpheme Lexicon, the Morpheme Lexicon can be considered a dynamic data set. It can be routinely or continually modified based on review and analysis of additional external data sets or data sources, regardless of size and content, and / or based on evaluating the results produced by processes, for example, those described herein, using the then- instituted Morpheme Lexicon. Additionally, forms of the Morpheme Lexicon can be evaluated and expanded or enhanced even before the Morpheme Lexicon is considered sufficiently complete for normal use. For example, the Morpheme Lexicon can be developed sufficiently to a desired usable level or desired content, and then the Morpheme Lexicon can be tested or evaluated by running any one or more of the processes described herein, evaluating the results and making desired adjustments to continue building the Morpheme Lexicon until it is considered sufficiently complete for intended use, for example, those uses described herein. External data sets can be input into one or more processes described herein that use the Morpheme Lexicon, and results evaluated for purposes of modifying or expanding the Morpheme Lexicon. Examples of methods for testing or evaluating the Morpheme Lexicon are discussed below with respect to research methods with the Morpheme Lexicon. (See Morpheme Lexicon Research, Method.)As the Morpheme Lexicon is being populated, or after the Morpheme Lexicon is put into service, Morpheme Keywords can be modified (in number or in content) as the meanings of more “instances” of a Morpheme are evaluated for the first time, or as changes are made to which subset of all the collected “instances” are selected for inclusion as “attested examples.” If new “instances” reinforce the existing meaning, a previously assigned Morpheme Keyword could remain the same. But if new “instances” diverge in meaning, either the then-current subset of “instances” can remain associated with the original morpheme and a second or multiple subsets of “instances” can be assigned to new morphemes, or a new more abstract Morpheme Keyword can be selected to encompass both subsets. In other words, the “instances,” once collected, can be evaluated and sorted and selected for inclusion in the lexicon as attested examples, as desired, either into one Morpheme entry or a set of related Morpheme entries.For an example of further evaluation of a Morpheme Keyword, if a set of attested examples “puppy, kitten” are assigned to an imaginary morpheme XYZ and given a meaning “baby mammal,” a possible new “instance,” “chick,” may be found with the same sound. Then, XYZ’s meaning might need to be broadened. More “instances” should be sought as evidence for this broader possible meaning. If the same sound as “puppy, kitten,” is found in, say, “duckling, tadpole,” the meaning of XYZ may be revised to “baby animal.” But if “chick” is the only “instance” that is a non-mammal, it may be an outlier to which the meaning data should not be force-fit. If so, “chick” could be discarded (or set aside to await better confirming evidence, or incorporated into another morpheme) and the meaning of XYZ kept as “baby mammal.” The attested examples selected for inclusion, and the Morpheme Keywords that encompass their meanings, can be both malleable, and can be fine-tuned together.Morpheme DataThe following discussion of Morpheme Data provides information about the content of data associated with a morpheme in a Morpheme Lexicon. As noted elsewhere, an exemplary Morpheme Lexicon is illustrated as a table with rows corresponding to each morpheme with columns and cells containing its associated data for ease of illustration. However, the architecture for a Morpheme Lexicon and its data can be designed in any of a number of desired ways by the user / designer. Therefore, the reference to cells, rows and columns is not to be taken as a limitation of the form of the Morpheme Lexicon. Additionally, a more detailed discussion of how the Morpheme Data can be used by various systems, and the benefits of including such data, is discussed elsewhere.The data in the columns of each Morpheme entry row is of a variety of types and meant to address a wide range of questions one could have when evaluating individual word parts of a text string, especially in disambiguating one Morpheme from another. The proposed content allows easy comparison between Morpheme entries (for example, during population of the data set) and aids the analysis of morphemes found in words (during system operation, once the data set is compiled). The presently-included data arises from an examination of the features of a morpheme, for example, all its component sounds, rather than merely the first letter, and their relationships to other sounds and sound classes, rather than only literal correspondences. The organizing device of a Phoneme Class Structure additionally provides a common format for the expression of what is otherwise difficult to describe in machine-interpretable terms: the sound of a syllable. The objective, detailed measurements of morpheme features described herein allow machine processing to yield consistent answers that can then be further fine-tuned. (For ease of understanding, the discussion hereinillustrates a Morpheme Lexicon as a spreadsheet of rows of Morpheme entries with associated columns of data, but it is understood that a Morpheme Lexicon can be implemented in a number of ways.) While the present description of exemplary Morpheme Lexicons includes data sets having a variety of associated data types, some of which may be omitted or abbreviated, it has been found that the morpheme, its sound information, meaning and at least some examples are useful to improved language processes, and illustrate useful relationships. The existence of these relationships can be used to benefit language processes. Moreover, these relationships and addition of data in one or more of the additional groups of associated data for such data sets enhances the quality of language processes.If the data for morpheme entries are thought of as columns in a spreadsheet, then one task in populating the data set is to choose the content of the associated data, which may help in choosing the number of columns (or morpheme features) to analyze, and another task if desired is to have sufficient data / columns that can be populated that allow one morpheme to be sufficiently distinguished from another. Morpheme data could include but are not limited to any or all of the following data types associated with the respective morpheme, and is numbered here to accord with the columns in the sample Morpheme Lexicon illustrated in Table D, below:1) general sound data, for example, a phoneme class structure common to the attested examples2) particular sound data, for example at least one literal spelling of a word-part of an attested example3) at least one meaning, for example a Morpheme Keyword encapsulating the keyword meanings of the attested examples4) attested position(s), for example prefix, root, or suffix (or combination thereof) corresponding to positions where the morpheme is found in the attested...

Claims

CLAIMS1. A method of analyzing at least one text string wherein the at least one text string includes at least one unit representing an element of the text string, the method comprising: comparing the at least one unit of the at least one text string to a sound class of at least one unit of a string in a data set containing at least a plurality of sound classes; determining a desired correlation between the at least one text string with a string in the data set; and saving data representing whether or not a desired correlation was found during the step of determining.

2. The method of claim 1 wherein the at least one text string represents a word-part and comparing the at least one unit includes comparing part of a word.

3. The method of claim 1 wherein the at least one text string represents a word and comparing the at least one unit includes comparing at least part of a word.

4. The method of any one of the preceding claims 1-3 wherein the data set includes a plurality of Phoneme Classes and wherein comparing includes comparing the at least one unit of the at least one text string to a Phoneme Class.

5. The method of claim 4 wherein each Phoneme Class includes a plurality of Phonemes and wherein comparing includes comparing the at least one unit of the at least one text string to a Phoneme Class containing the plurality of Phonemes.

6. The method of any one of the preceding claims 1-5 wherein determining a desired correlation between the at least one text string with the string in the data set includes identifying similarities between the at least one text string with the sound class.

7. The method of claim 6 wherein identifying similarities between the at least one text string with the sound class includes finding a similarity meeting a desired criterion and saving the at least one text string, the sound class, and the characteristic of the desired criterion.

8. The method of claim 6 wherein identifying similarities between the at least one text string with the sound class includes determining when a unit in the at least one text string is in the same sound class as the sound class of the string of the data set.

9. The method of claim 8 wherein determining when a unit in the at least one text string is in the same sound class as the sound class of the string of the data set includes determining whether the text string unit and the data set string unit are one of the classes selected from the group of Phoneme Classes of vowels, labials, dentals, velars, liquids, nasals, semivowels, sibilants, and aspirates.

10. The method of any one of the preceding claims 1-9 wherein the at least one text string represents a word from at least one of a plurality of languages and comparing the at least one unit includes comparing part of a word from at least one of the plurality of languages.

11. The method of any one of the preceding claims 1-10 wherein comparing the at least one part of the text string includes comparing to a plurality of strings in the data set.

12. The method of any one of the preceding claims 1-11 wherein comparing the at least one part of the text string includes comparing to all strings in the data set.

13. The method of any one of the preceding claims 1-12 wherein at least one unit of the text string represents a letter and wherein comparing includes comparing the letter to a letter unit of at least one string in the data set.

14. The method of any one of the preceding claims 1-12 wherein the at least one unit of the text string represents a letter and wherein comparing includes comparing the letter to a Phoneme Class of a unit of at least one string in the data set.

15. The method of any one of the preceding claims 1-12 wherein the at least one unit of the text string represents a Phoneme Class and wherein comparing includes comparing the Phoneme Class to a letter unit of at least one string in the data set.

16. The method of any one of the preceding claims 1-12 wherein the at least one unit of the text string represents a Phoneme Class and wherein comparing includes comparing the Phoneme Class to a Phoneme Class of a unit of at least one string in the data set.

17. The method of any one of the preceding claims 1-16 wherein the comparing includes comparing the unit with at least one of a group of all letters and all Phoneme Classes in the data set.

18. The method of any one of the preceding claims 1-17 wherein the comparing includes comparing the unit with at least one Phoneme Class in the data set representing at least a part of one Morpheme.

19. The method of any one of the preceding claims 1-18 wherein the comparing includes comparing a letter to data in the data set representing a Phoneme Class.

20. The method of any one of the preceding claims 1-18 wherein the comparing includes comparing a word-part to data in the data set representing at least one Phoneme Class.

21. The method of any one of the preceding claims 1-20 wherein the comparing includes comparing a letter to data in the data set representing a Morpheme.

22. The method of any one of the preceding claims 1-21 wherein the comparing includes comparing a word-part to data in the data set representing a Morpheme.

23. The method of any one of the preceding claims 1-22 wherein the comparing includes comparing the unit with at least one string representing a Morpheme in the data set.

24. The method of any one of the preceding claims 1-22 wherein the comparing includes comparing the unit with data representing a plurality of morphemes in the data set.

25. The method of any one of the preceding claims 1-24 wherein the comparing includes comparing the unit with strings representing all of the morphemes in the data set.

26. The method of any one of the preceding claims 1-25 wherein the unit represents a sound symbol and the comparing includes comparing the sound symbol to a Phoneme Class of at least one string in the data set.

27. The method of any one of the preceding claims 1-26 wherein determining the desired correlation includes determining a desired sound correlation.

28. The method of any one of the preceding claims 1 -27 wherein determining the desired correlation includes determining a correlation that is in the group of an exact letter match, a similar letter match, an exact Phoneme Class match, and a similar Phoneme Class match.

29. The method of claim 28 wherein determining the desired correlation includes determining a correlation that is, for all compared units, an exact letter match between letters.

30. The method of claim 28 wherein determining the desired correlation includes determining a correlation that is, for all except no more than one of the compared units, an exact letter match between letters, and the all except for no more than one is a single letter, in any position in a sequence of letters, that is added, missing, or different.

31. The method of claim 28 wherein determining the desired correlation includes determining a correlation that is an exact letter match between letters except for no more than two letters in the compared units, wherein two letters in any position in a sequence of letters are transposed with one another.

32. The method of claim 28 wherein determining the desired correlation includes determining a correlation that is, for all compared units, an exact Phoneme Class match between a Phoneme Class representing respective letters of the compared units and a Phoneme Class of respective compared units in at least one string in the data set.

33. The method of claim 28 wherein determining the desired correlation includes determining a correlation that is, for all except no more than one of the compared units, an exact Phoneme Class match between a Phoneme Class representing respective letters of the compared units and a Phoneme Class of respective compared units in at least one string in the data set, and theall except for no more than one is a single Phoneme Class, in any position in a sequence of units, that is added, missing, or different.

34. The method of claim 28 wherein determining the desired correlation includes determining a correlation that is, for all except no more than two of the compared units, an exact Phoneme Class match between a phoneme class representing respective letters of the compared units and a Phoneme Class of respective compared units in at least one string in the data set, wherein two Phoneme Classes in any position in a sequence of Phoneme Classes are transposed with one another.

35. The method of claim 28 wherein determining the desired correlation that is a similar Phoneme Class match includes determining a correlation between Phoneme Classes of the letters in the at least one text string and a Morpheme in the data set wherein no more than a single Phoneme Class, in any position in a sequence of units, is added, missing, or different.

36. The method of claim 28 wherein determining the desired correlation that is a similar Phoneme Class match includes determining a correlation between Phoneme Classes of the letters in the at least one text string and a Morpheme in the data set wherein all Phoneme Classes are identical except for no more than two single Phoneme Classes, in any position in a sequence of units, transposed with one another.

37. The method of any one of the preceding claims 1-36 wherein the comparing includes comparing a first text string and further including comparing a second text string.

38. The method of any one of the preceding claims 1-37 further including generating correlation data containing data representing the text string and data representing the string in the data set to which the unit was compared during the comparing step.

39. The method of claim 38 wherein generating a correlation pair includes saving the pair with either of a letter or a Phoneme Class of the text string that correlates with a Morpheme, and with a letter or Phoneme Class from the string in the data set.

40. The method of claim 39 wherein generating a correlation pair includes generating a correlation pair with a representation that correlates with the text string and at least one keyword representing a Morpheme from the data set.

41. The method of claim 40 wherein the keyword is one or more of a dictionary word, a synonym, generating a correlation pair includes generating a correlation pair.

42. The method of any one of the preceding claims 1-41 further including transforming the unit to a Phoneme Class.

43. The method of any one of the preceding claims 1-42 further including transforming the unit into data representing a list of letters belonging to a Phoneme Class.

44. The method of any one of the preceding claims 1-43 further including generating a correlation of data representing information resulting from the determining of a desired correlation.

45. The method of claim 44 wherein the saving data includes saving data representing one or more of the data selected from the group of:- an ordered sequence of units making up the text string,- historical data that has been collected for the word-part, including for a word to which the word-part belongs, and for the individual units of the word-part, wherein the historical data represents one or more of the data selected from the group of a location of syllable divisions relative to the word-part, an Input word string of which the word-part is a part, at least one semantic value for an input word, a negation status for an input word, positions of units within the text string for an input word string, letter types or sound symbol types of units in the text string,- an ordered sequence of data making up correlated data from the data set,- data set data representing data for which a desired correlation was found, including data representing a Morpheme, including one or more of the data selected from the group of a sequence of Phoneme Classes, data representing a semantic value or a syntactic value, and if a semantic value, at least one keyword, attested examples containing the Morpheme, a value representing a number of syllables in the Morpheme, a position of the Morpheme in an attested example,- pairing data representing a type of correlation found between the word-part and data in the data set, including one or more of the data selected from the group of an alignment value corresponding to units in the text string, relative to data in the data set, a sound correlation type.

46. A system configured to carry out the method of any one of the preceding claims 1-45.

47. Apparatus including a processor configured to carry out the method of any one of the preceding claims 1-45.

48. A method of analyzing at least one text string comprising evaluating a possible sound class correlation of an input string with a string from a data set and evaluating a possible semantic correlation of the input string with the string from the data set.

49. The method of claim 48 wherein evaluating a possible sound class correlation of an input string with a string from a data set includes evaluating a possible sound class correlation of the input string with a string from a morpheme lexicon.

50. The method of any one of the preceding claims 48-49 further including the steps of any of the processes selected from a group of verifying the input string, looking up a semantic value, looking up hyphenation and / or pronunciation data, looking up or evaluating a negation status, splitting the input string, combining possible candidates, scoring possible candidates, ranking possible candidates, and outputting possible candidates.

51. Apparatus, systems, or methods according to any of the examples and embodiments described herein.

52. The method of any one of the preceding claims 1-51 wherein the sound class includes a plurality of items wherein each of the plurality of items the sound class sounds similar.

53. The method of any one of the preceding claims 1-52 wherein the plurality of sound classes includes a sound class of vowels.

54. The method of any one of the preceding claims 1-53 wherein the sound class is a phoneme class and including vowel, labial, dental, velar, liquid, nasal, semivowel, sibilant, and / or aspirate phoneme classes.

55. The method of any one of the preceding claims 1-54 wherein the at least one text string is associated with at least one keyword.

56. The method of any one of the preceding claims 1-55 wherein the at least one text string represents a concept, and wherein proper nouns (for example, a particular entity, for example a name of a person or a place or institution) are excluded.

57. The method of any one of the preceding claims 1-56 wherein the at least one text string is written using either the letters of an alphabet or the sound symbols of a syllabary type of script.

58. The method of claim 57 wherein the at least one text string is written using letters or sound symbols from a different alphabet or script than the one used in writing the string in the data set.

59. The method of claim 57 wherein the at least one text string is written using letters or sound symbols that belong to one of multiple alphabets and / or multiple scripts used in writing the string in the data set.

60. The method of any one of the preceding claims 1-59 wherein the at least one text string can be in any one of multiple languages.

61. The method of any one of the preceding claims 1-60 wherein the string in the data set includes at least one syllable.

62. The method of any one of the preceding claims 1-61 wherein the string in the data set has an identifiable pattern of consonants and vowels.

63. The method of claim 62 and the identifiable pattern includes one or more of the group of consonant-vowel-consonant, consonant-vowel, vowel-consonant-vowel, vowel-consonant.

64. The method of claim 63 and the data set includes at least the consonant-vowel-consonant pattern.

65. The method of any one of the preceding claims 1-64 wherein the string in the data set has an identifiable structure of consonants and vowels, phoneme classes of which consonants and vowels are, when combined into a particular sequence (phoneme class structure) associated with one or more meanings.

66. The method of any one of the preceding claims 1-65 wherein the string in the data set is applicable to one or multiple languages.

67. A method of analyzing at least one keyword associated with a text string and an associated plurality of sound classes for the text string, the method comprising: comparing the at least one source keyword associated with the text string to at least one data set keyword associated with a string in a data set that contains a plurality of strings and at least one data set keyword associated with respective ones of the strings in the data set; determining a desired correlation between the source keyword and the at least one data set keyword; and saving data representing whether or not a desired correlation was found during the step of determining, and if so, saving the association between the text string, the source keyword, the data set keyword, and the sound classes for the text string.

68. The method of analyzing of claim 67 wherein the comparing includes comparing to at least one data set keyword associated with respective ones of the strings in the data set wherein the strings in the data set include respective associated sound classes.

69. The method of analyzing according to any one of the preceding claims 67-68 wherein the text string is a word input by the user.

70. The method of analyzing according to any one of the preceding claims 67-69 wherein the source keyword has a meaning.

71. The method of analyzing according to any one of the preceding claims 67-70 wherein the source keyword has a conceptual meaning.

72. The method of analyzing according to any one of the preceding claims 67-71 wherein the source keyword is associated by being retrieved from lookup in at least one reference work in which the text string is an entry, and is part of a definition string of that entry.

73. The method of analyzing according to any one of the preceding claims 67-72 wherein the source keyword is associated by being retrieved from lookup in at least one reference work in which a synonym, found via another lookup, of the text string is an entry, and is part of the definition string of that entry.

74. The method of analyzing according to any one of the preceding claims 67-73 wherein the source keyword is associated by being retrieved from lookup in at least one reference work in which an antonym, found via another lookup, of the text string is an entry, and is part of the definition string of that entry.

75. The method of analyzing according to any one of the preceding claims 67-74 wherein the source keyword is associated with the text string by being the result of a Negation Lookup process.

76. The method of analyzing according to any one of the preceding claims 67-75 wherein the at least one source keyword is associated with the text string by being input by the user along with the text string.

77. The method of analyzing according to any one of the preceding claims 67-76 wherein the comparing compares letter-by-letter.

78. The method of analyzing according to any one of the preceding claims 67-77 wherein the comparing compares sound symbol -by-sound symbol.

79. The method of analyzing according to any one of the preceding claims 67-78 wherein the comparing compares each one of a possible plurality of source keywords to each one of a possible plurality of data set keywords, until both sets of keywords are exhausted.

80. The method of analyzing according to any one of the preceding claims 67-79 wherein the string in the data set is a morpheme.

81. The method of analyzing according to any one of the preceding claims 67-80 wherein the data set keyword is a Morpheme Keyword associated with a Morpheme entry in a Morpheme Lexicon.

82. The method of analyzing according to any one of the preceding claims 67-81 wherein the data set keyword is associated with the string in the data set by being retrieved from lookup in at least one reference work in which one of the attested examples associated with a Morpheme entry in the Morpheme Lexicon is an entry, and is part of the definition string of that entry.

83. The method of analyzing according to any one of the preceding claims 67-82 wherein the desired correlation is an exact letter-by-letter or sound symbol-by-sound symbol match.

84. The method of analyzing according to any one of the preceding claims 67-83 wherein the desired correlation is an exact or partial letter-by-letter or sound symbol-by-sound symbol match, wherein the definition of partial is specified by the system designer as a certain number or percentage of correlated letters, sound symbols, or syllables.

85. The method of analyzing according to any one of the preceding claims 67-84 wherein the desired correlation is an exact or partial letter-by-letter or sound symbol-by-sound symbol match between the source keyword, or any inflected form of that source keyword.

86. The method of analyzing according to any one of the preceding claims 67-85 wherein the desired correlation is an exact or partial letter-by-letter or sound symbol-by-sound symbol match between the source keyword, or any inflected form of that source keyword, or a synonym of that source keyword, and the data set keyword.

87. The method of analyzing according to any one of the preceding claims 67-86 wherein the desired correlation is an exact or partial letter-by-letter or sound symbol-by-sound symbol match between the source keyword, or any inflected form of that source keyword, or a synonym of that source keyword, or any inflected form of that source keyword, and the data set keyword.

88. The method of analyzing according to any one of the preceding claims 67-87 wherein the desired correlation is an exact or partial letter-by-letter or sound symbol-by-sound symbol match between the source keyword, or any inflected form of that source keyword, or a synonym of that source keyword or having some desired range of degrees of separation of synonymity specified by the designer, and the data set keyword.

89. The method of analyzing according to any one of the preceding claims 67-88 wherein the desired correlation is for a specified number or range of numbers, specified by the designer, of matches between a source keyword and a data set keyword.

90. The method of analyzing according to any one of the preceding claims 67-89 wherein the desired correlation is for a specified type, specified by the designer, of match between a source keyword and a data set keyword.

91. The method of analyzing according to any one of the preceding claims 67-90 wherein the desired correlation is for a specified type, specified by the designer, of match, including a specific degree or range of degrees of synonymity between a source keyword and a data set keyword.

92. The method of analyzing according to any one of the preceding claims 67-91 wherein the desired correlation is for a specified combination of a type and number or range of numbers, specified by the designer, of matches between source keywords and data set keywords.

93. The method of analyzing according to any one of the preceding claims 67-92 wherein the correlation process uses a standard edit distance algorithm.

94. The method of analyzing according to any one of the preceding claims 67-93 wherein the correlation process uses reference works, for example, one or more dictionaries and thesauruses in one or more languages.

95. The method of analyzing according to any one of the preceding claims 67-94 wherein saving data includes saving data representing one or more of the following: whether or not a desired correlation was found during the step of determining, the text string whose source keyword was correlated, the string in the data set whose data set keyword was correlated, the particular source keyword(s) and data set keyword(s) correlated, the number of correlated keywords and their type(s) of correlation(s) including the presence or absence of inflection and / or including the degree of synonymity.

96. The method of analyzing according to any one of the preceding claims 67-95 along with analyzing the text string according to the method of analyzing at least one text string of any one of the preceding claims 1-95.

97. A method of analyzing a semantic function of each part of a word that has multiple word-parts and each word-part has one or more respective letters, the method comprising: analyzing the semantic function of each of the multiple word-parts of the word as a function of an absolute position of the word part-in the word, and as a function of an ordinal location of the letters of each word-part in comparison to all of the letters in the word; and processing results of the analyzing.

98. The method of claim 97 further including identifying whether a word-part is associated with an entry in a data set that is one or more of a prefix, a root, or a suffix, and if so, assigning it a value of a semantic word-part.

99. The method of claim 97 further including identifying whether a word-part is associated with an entry in a data set that is one or more of a prefix, a root, or a suffix, and if a prefix or a root assigning it a value of a semantic word-part and if a suffix that has an associated semantic value in a data set, assigning it a value of a semantic word-part, and if it is a suffix that has no associated semantic value in a data set, assigning it a value of a semantically undefined word-part.

100. The method of any one of the preceding claims 97-99 wherein an entry in a data set is a morpheme in a morpheme lexicon.

101. The method of any one of the preceding claims 97-100, the method comprising: identifying for each word-part which word-parts are semantically defined and which undefined.

102. The method of any one of the preceding claims 97-101 wherein whether a word-part is semantically defined is determined by the presence or absence of at least one keyword in a data set representing a meaning.

103. The method of 102 wherein a word-part is associated with a morpheme and the at least one keyword in a data set representing a meaning is a morpheme keyword.

104. The method of claim 98 further determining whether a word-part is a modifier or a non-modifier.

105. The method of any one of the preceding claims 97-104 further comprising evaluating whether a word-part is non-modifying root.

106. The method of evaluating a word-part comprising evaluating whether the word-part is a non-modifying root.

107. The method of any one of the preceding claims 97-106 further comprising evaluating the total number of roots in the word, and evaluating which root of multiple roots is the final and therefore non-modifying root or, if there is only one root in the word, evaluating is as the non-modifying root.

108. The method of any one of the preceding claims 97-107 further comprising identifying a word-part that is a non-modifying root.

109. The method of any one of the preceding claims 97-108 further determining whether a word-part has a semantic function of a non-modifying root, a modifying affix, or a modifying root.

110. The method of claim 109 further comprising determining that a word-part is a non-modifying root comprises evaluating each of the word-parts’ status as a root or prefix or suffix, evaluating the total number of roots in the word, and evaluating which root of multiple roots is the final and therefore non-modifying root or, if there is only one root in the word, evaluating is as the non-modifying root.

111. The method of claim 110 further identifying each root in a word, if there are multiple roots in a word, as being either the non-modifying root or a modifying root.

112. The method of any one of the preceding claims 97-111 further including identifying a word-part that is a modifying word-part and determining the word-part it directly modifies.

113. The method of claim 112 where determining is based on a set of Logical Premises or position data114. The method of any one of the preceding claims 97-113 wherein the semantic function is one of a non-modifying root, modifying root, or modifying affix and wherein analyzing is analyzing a non-modifying root, modifying root, or modifying affix.

115. The method of any one of the preceding claims 97-114 wherein a word-part whose semantic function is either a modifying root or a modifying affix is further analyzed, and wherein analyzing further includes analyzing such modifiers to be at least one of relative, definite, qualitative, or quantitative.

116. The method of claim 115 wherein analyzing such modifiers includes analyzing such modifiers to be one of a relative qualitative, relative quantitative, definite qualitative, or definite quantitative type of modifier.

117. The method of any one of the preceding claims 97-116 further comprising retrieving a negation status of a morpheme associated with a respective Morpheme-Word-Part of the word that is either a modifying root or a modifying affix.

118. The method of claim 117 further comprising retrieving data associated with the Input word relating to a negation status, position, or syllable count.

119. The method of any one of the preceding claims 97-118 wherein analyzing the semantic function includes retrieving from a data set a position associated with the word-part to be analyzed.

120. The method of claim 119 wherein the position of the word-part is associated with the word-part through its association with a morpheme.

121. The method of claim 120 wherein the morpheme has an associated position in a data set.

122. The method of any one of the preceding claims 97-121 wherein the word-part is associated with a morpheme, which has associated with it at least one position in a data set.

123. The method of any one of the preceding claims 120-122 where the association of the word-part and the morpheme is a Morpheme-Word-Part pair and the data set is a Morpheme Lexicon.

124. The method of any one of the preceding claims 97-123 wherein analyzing the semantic function of each word-part is analyzing the semantic function of each of the multiple word-parts of the word as a function of one or more of an absolute position of each respective word-part in the word, a relative position of each respective word-part relative to other word-parts in the word, at least one position associated with the word-part, and the ordinal location of the letters of each word-part relative to the letters in the word; andassigning to each word-part a semantic function.

125. The method of any one of the preceding claims 97-124 wherein analyzing the semantic function of each word-part is analyzing the semantic function of each of multiple Morpheme- Word-Parts of the word as a function of one or more of an absolute position of each respective word-part in the word, a relative position of each respective word-part relative to other word-parts in the word, at least one attested Position associated with the morpheme of the respective Morpheme-Word-Part of each respective word-part, and the ordinal location of the letters of each Morpheme-Word-Part relative to the letters in the word; and assigning to each Morpheme- Word-Part a semantic function.

126. The method of any one of the preceding claims 97-125 further including evaluating the semantic function of the word-parts for compatibility with a set of Logical Premises.

127. A method of analyzing a semantic function of each part of a word that has multiple word-parts and each part has one or more respective letters, the method comprising: analyzing the semantic function of each of the multiple parts of the word as a function of a relative position of each word-part, compared to an absolute position of another of the multiple word-parts; and processing results of the analyzing.

128. The method of claim 127 further determining whether a word-part is a modifier or a non-modifier.

129. The method of any one of the preceding claims 127-128 further comprising evaluating whether a word-part is a non-modifying root.

130. The method of any one of the preceding claims 127-129 wherein the semantic function is at least one of a non-modifying root, modifying root, or modifying affix and wherein analyzing is analyzing at least one of a non-modifying root, modifying root, or modifying affix.

131. The method of claim 130 wherein a word-part whose semantic function is a modifying root or a modifying affix and further analyzing such modifiers to be at least one of relative, definite, qualitative, or quantitative.

132. The method of claim 131 wherein the modifying root and modifying affix are one of a relative qualitative, relative quantitative, definite qualitative, or definite quantitative type of modifier.

133. The method of any one of the preceding claims 127-132 evaluating the total number of roots in the word, and evaluating which root of multiple roots is the final and therefore non-modifying root or, if there is only one root in the word, evaluating is as the non-modifying root.

134. The method of any one of the preceding claims 127-133 further comprising identifying a word-part that is a non-modifying root.

135. The method of any one of the preceding claims 127-134 further determining whether a word-part has a semantic function of a non-modifying root, a modifying affix, or a modifying root.

136. The method of claim 135 further comprising determining that a word-part is a non-modifying root comprises evaluating each of the word-parts’ status as a root or prefix or suffix, evaluating the total number of roots in the word, and evaluating which root of multiple roots is the final and therefore non-modifying root or, if there is only one root in the word, evaluating is as the non-modifying root.

137. The method of claim 136 further identifying each root in a word, if there are multiple roots in a word, as being either the non-modifying root or a modifying root.

138. The method of any one of the preceding claims 127-137 further including identifying a word-part that is a modifying word-part and determining the word-part it directly modifies.

139. The method of claim 138 where determining is based on a set of Logical Premises or position data.

140. The method of any one of the preceding claims 97-139 wherein the position data is located in a data set.

141. The method of any one of the preceding claims 97-140 wherein the data set is a morpheme lexicon.

142. The method of claim 141 wherein analyzing such modifiers includes analyzing such modifiers to be one of a relative qualitative, relative quantitative, definite qualitative, or definite quantitative type of modifier.

143. The method of any one of the preceding claims 127-142 further comprising retrieving a negation status of a morpheme associated with a respective Morpheme-Word-Part of the word that is either a modifying root or a modifying affix.

144. The method of claim 143 further comprising retrieving the negation status from a Negation Lookup process.

145. The method of any one of the preceding claims 142-144 further comprising retrieving a negation status of a word-part from a Negation Lookup process.

146. The method of any one of the preceding claims 97-145 further including retrieving from an existing data set for at least one of the multiple word-parts at least one attested position associated with that word-part; and processing results of the retrieving.

147. The method of claim 146 wherein retrieving is retrieving from a data set a position corresponding to a morpheme corresponding to at least part of the word.

148. The method of any one of the preceding claims 146-147 wherein retrieving a position is retrieving a position of a word-part of a word in the data set.

149. The method of any one of the preceding claims 146-148 wherein the position is either one or a combination of a prefix, root, and suffix.

150. The method of any one of the preceding claims 146-149 wherein the semantic function is at least one of a non-modifying root, modifying root, or modifying affix, and wherein analyzing is analyzing at least one of a non-modifying root, modifying root, or modifying affix.

151. The method of claim 150 wherein a modifying root or modifying affix are one of relative or definite, or one of qualitative or quantitative, and wherein analyzing is analyzing a category of a word-part that is a modifier is based on whether it is a relative, definite, qualitative, or quantitative type of modifier.

152. The method of any one of the preceding claims 97-126 in combination with any one of the preceding method claims 127-146 in combination with any one of the preceding method claims 147-151.

153. The method of claim 152 wherein the semantic function is at least one of a non-modifying root, modifying root, or modifying affix and wherein analyzing is analyzing at least one of a non-modifying root, modifying root, or modifying affix.

154. The method of claim 153 wherein the modifying root or the modifying affix are at least one of relative, definite, qualitative, or quantitative, and wherein analyzing is analyzing a semantic function of a word part based on whether it is a relative, definite, qualitative, or quantitative modifier.

155. The method of any one of the preceding claims 137-154 further including determining whether at least one attested position associated with each of the respective word-parts is consistent with the semantic function of the respective word parts implied by the order of the respective word parts in the word.

156. The method of claim 155 wherein determining includes reviewing the semantic function assigned to a word part in the context of the semantic function assigned to other word parts in the word and modifying one or more of the foregoing until all semantically defined word parts in the word have been assigned a semantic function that is consistent with the semantic functions of the other word-parts, and confirming such assignments.

157. The method of any one of the preceding claims 155-156 wherein consistent is consistent with a set of Logical Premises and position data.

158. The method of any one of the preceding claims 97-157 further including either identifying each word-part by a confirmed semantic function or, for any whose assigned semantic function is contradicted by their attested position in a data set, identifying both the assigned semantic function and the attested position of each word-part.

159. The method of any one of the preceding claims 97-158 wherein semantically defined word-parts are those whose associated morphemes have at least a morpheme keyword.

160. The method of any one of the preceding claims 97-159 further comprising identifying a category of a word-part by a semantic lookup process followed by a comparison to a list of Category Keywords.

161. The method of any one of the preceding claims 97-160 further comprising identifying a category of a word-part by a lookup in a data set.

162. The method of claim 161 wherein the data set is a Morpheme Lexicon, and identifying a category is by lookup of a Category Assignment in the Morpheme Lexicon.

163. The method of any one of the preceding claims 97-162 further including providing data and identifications as categorization data.

164. The method of any one of the preceding claims 97-163 wherein a word-part is associated with a morpheme and analyzing is analyzing the word-part relative to the morpheme.

165. The method of any one of the preceding claims 97-164 wherein analyzing the semantic function includes analyzing by a separate semantic analyzer process.

166. The method of any one of the preceding claims 97-165 wherein a semantic function is at least one of to represent a root meaning, or to directly modify the meaning of a root, or to indirectly modify the meaning of a root by directly modifying another modifier that either directly or indirectly modifies a root.

167. The method of any one of the preceding claims 97-166 wherein the word-part is assigned a category that belongs to the set of: relative qualitative, relative quantitative, definite qualitative, and definite quantitative.

168. The method of any one of the preceding claims 97-167 further including evaluating the word-parts of a word based on one or more of the following Logical Premises: a word can include one non-modifying root and any number of word-parts that are modifiers modifiers are word-parts that include modifying affixes and modifying rootsmodifiers directly modify one word-part, but if part of a sequence of modifiers may indirectly modify one or more additional word-parts a modifier can itself be modified by another modifier word-parts modify other word-parts from the outside toward the center of the word — the “center” being either a root that is the only root in the word, or the final root if there are multiple roots in the word (for example, in a word which has both at least one prefix and at least one suffix, modification moves inward from the affix ends of the word toward the central non-modifying root) a word has at least one root and may have zero, one, or more affixes a word-part is either a root or an affix a word-part is either a non-modifying root or a modifier a root is either a non-modifying root or a modifier there will be no root word-part after a suffix or before a prefix; and if there is at least one prefix and at least one suffix in a word, the root must be in a position between them a root that is a modifier can only modify another root it precedes and otherwise cannot act as a modifier a root that is the only root in the word, or the final root if there are multiple roots in the word, is the non-modifying root in a word the non-modifying root in a word is modified by any affix that may follow or precede it, and / or by any root that may precede it an affix is always a modifier an affix is either a prefix (which precedes a root) or a suffix (which follows a root) an affix can modify a root or another affix an affix that is a prefix modifies what it precedes an affix that is a suffix modifies what it follows a prefix modifies a word-part it immediately precedes a suffix modifies a word-part it immediately follows a suffix in the proposed data set may or may not be semantically defined, and if semantically defined it modifies the meaning of what it follows and if semantically undefined, it may for the purpose of categorization be assumed to have no semantic function there is one and only one non-modifying word-part in a word, and it is a root, and the rest of the word-parts are modifiers a word has either zero or at least one prefix, followed by at least one root, followed by either zero or at least one suffixa semantic function is either to be a root or to modify the meaning of another word-part a modifier modifies the meaning of another word-part in either a definite, relative, quantitative, or qualitative manner the meaning of a word derives from the sum of meanings of a word’s parts.

169. The method of any one of the preceding claims 97-168 wherein a word-part is semantically defined by at least one keyword in a data set representing a meaning.

170. The method of claim 169 wherein a word-part is associated with a morpheme and the at least one keyword in a data set representing a meaning is a morpheme keyword.

171. A method of analyzing a word comprising identifying one or more roots and any prefixes and any suffixes, determining which suffixes are semantically defined, identifying those suffixes and the prefixes from such identification as modifying affixes, determining the position of the one or more roots by their relative position(s), and, if multiple determining which is the final and identifying it as the non-modifying root and the other(s) as one or more modifying roots, or if it is the only root identifying it as the non-modifying root.

172. A method of analyzing a semantic function of a word-part of a word, the method comprising evaluating the word-part for the presence of a modifier and accessing a data set to evaluate what type of modifier the word-part is.

173. The method of claim 172 wherein evaluating the presence of a modifier comprises identifying the presence of at least one keyword in a data set representing a meaning associated with the word-part and assigning a modifier type to the word-part.

174. The method of claim 173 wherein the modifier type is one or more of relative, definite, qualitative or quantitative.

175. The method of claim 173 wherein the modifier type is at least one of relative qualitative, relative quantitative, definite qualitative or definite quantitative.

176. A data set containing words and word-parts, the data set comprising at least one modifier type associated with a word in the data set wherein the modifier type is at least one of quantitative, qualitative, definite, or relative.

177. The data set of claim 176 where the modifier type is at least one of definite quantitative, definite qualitative, relative quantitative, or relative qualitative.

178. The data set of any one of the preceding claims 176-177 wherein each modifier type is associated with a plurality of keywords.

179. The data set of claim 178 wherein the plurality of keywords are category keywords.

180. The data set of any one of the preceding claims 178-179 wherein each of the plurality of keywords is associated with a hierarchy of keywords.

181. A method of evaluating a word, the method comprising: evaluating at least a first word-part of the word relative to another word-part of the word and identifying that the first word-part modifies the another word-part; and identifying that the first word-part is at least one of a quantitative, qualitative, definite, or relative type of modifier.

182. The method of claim 181 further comprising identifying the first word-part as one of definite quantitative, definite qualitative, relative quantitative, or relative qualitative type of modifier.

183. The method of any one of the preceding claims 181-182 further including identifying at least one category keyword for the first word-part.

184. The method of claim 183 further including identifying a hierarchy of keywords associated with the first word-part.

185. An information collection, for example, a database, the collection comprising: representations of data including: a plurality of morphemes, and for each of at least some of the plurality of morphemes: at least one literal spelling, a sound sequence, and at least one meaning.

186. The information collection of claim 185 wherein the information collection is on a single server.

187. The collection of any one of the preceding claims 185-186 wherein representations of data are either machine readable or human readable.

188. The collection of any one of the preceding claims 185-187 wherein the at least one literal spelling is a spelling comprised of at least one of letters of the Greek alphabet or letters of the Roman alphabet.

189. The collection of any one of the preceding claims 185-188 wherein the at least one literal spelling is a spelling of a representation of only the morpheme.

190. The collection of any one of the preceding claims 185-189 wherein the at least one meaning corresponds to a meaning of the morpheme.

191. The collection of any one of the preceding claims 185-190 wherein the at least one meaning is represented by at least one whole word.

192. The collection of any one of the preceding claims 185-191 wherein the sound sequence is a sequence of sound characteristics corresponding to elements of the at least one literal spelling.

193. The collection of any one of the preceding claims 185-192 wherein the sound sequence is a phoneme class structure.

194. The collection of claim 193 wherein the phoneme class structure includes a phoneme class for each of the elements of the at least one literal spelling.

195. The collection of any one of the preceding claims 193-194- wherein the phoneme class structure includes at least one element selected from the group of vowels, labials, dentals, velars, liquids, nasals, semivowels, sibilants, and aspirates.

196. The collection of any one of the preceding claims 185-195 wherein the phoneme class structure is a sequence of phoneme classes.

197. The collection of any one of the preceding claims 185-196 wherein the data set is organized as a function of sound classes.

198. The collection of any one of the preceding claims 185-197 and the sound classes are phoneme classes.

199. The collection of any one of the preceding claims 185-198 wherein the information collection associates a phoneme class structure with a plurality of literal spellings arising from a plurality of contexts of the group of alphabets or scripts, languages, historical time periods, connotations, positions of the syllable within a word, positions of the word within a sentence, and levels of formality or informality.

200. The collection of any one of the preceding claims 185-199 wherein the data set is an organization of a set of morphemes wherein at least a plurality of the morphemes is less than complete words.

201. The collection of claim 200 wherein the plurality of the morphemes are single syllables forming less than a complete word.

202. The collection of any one of the preceding claims 185-201 wherein the collection includes a plurality of data elements for each morpheme wherein each of the plurality of data elements is searchable.

203. The collection of claim 202 wherein each of the plurality of data elements is searchable in different languages.

204. The collection of any one of the preceding claims 202-203 wherein the plurality of data elements for a morpheme are associated with the morpheme logically based on characteristics of words.

205. The collection of claim 204 wherein some of the data elements for a morpheme are associated with the morpheme as a function of meaning.

206. The collection of any one of the preceding claims 185-205 wherein the data set is an organization that is other than alphabetical (by first letter).

207. The collection of any one of the preceding claims 185-206 wherein the data set is indexed other than by alphabetically according to a first letter of the at least one literal spelling of each morpheme.

208. The collection of any one of the preceding claims 185-207 wherein the data set is indexed according to a sound characteristic representing a portion of the sound sequence.

209. The collection of claim 208 wherein the data set is indexed according to a phoneme class representing a portion of the sound sequence.

210. The collection of any one of the preceding claims 185-209 wherein the data set is indexed according to a phoneme class structure representing the sound sequence.

211. The collection of claim 210 wherein the phoneme class structures for a preponderance of the morphemes in the data set describes a sound that is of one syllable, with a beginning phoneme, a middle vowel, and an ending phoneme.

212. The collection of claim 211 wherein one or more of the phonemes that is a consonant in the phoneme class structure of a morpheme may be a consonant blend in one or more attested examples of the morpheme.

213. The collection of claim 211 wherein one or more of the phonemes that is a consonant in the phoneme class structure of a morpheme is a consonant doubling in one or more attested examples of the morpheme.

214. The collection of claim 211 wherein one or more of the phonemes that is a vowel in the phoneme class structure of a morpheme may be a vowel sequence in one or more attested examples of the morpheme.

215. The collection of claim 211 wherein one or more of the phonemes that is a vowel in the phoneme class structure of a morpheme may be a vowel doubling in one or more attested examples of the morpheme.

216. The collection of claim 210 wherein the phoneme class structures for some of the morphemes in the data set describe a sound that is of one syllable, with a beginning phoneme that is a vowel, and an ending phoneme that is a consonant.

217. The collection of claim 216 wherein one or more of the phonemes that is a consonant in the phoneme class structure of a morpheme may be a consonant blend in one or more attested examples of the morpheme.

218. The collection of claim 216 wherein one or more of the phonemes that is a consonant in the phoneme class structure of a morpheme may be a consonant doubling in one or more attested examples of the morpheme.

219. The collection of claim 216 wherein one or more of the phonemes that is a vowel in the phoneme class structure of a morpheme may be a vowel sequence in one or more attested examples of the morpheme.

220. The collection of claim 216 wherein one or more of the phonemes that is a vowel in the phoneme class structure of a morpheme may be a vowel doubling in one or more attested examples of the morpheme.

221. The collection of claim 210 wherein the phoneme class structures for some of the morphemes in the data set describe a sound that is of one syllable, with a beginning phoneme that is a consonant and an ending phoneme that is a vowel.

222. The collection of claim 221 wherein the beginning phoneme that is a consonant in the phoneme class structure may be a consonant blend of a morpheme in one or more attested examples of the morpheme.

223. The collection of claim 221 wherein the beginning phoneme that is a consonant in the phoneme class structure of a morpheme may be a consonant doubling in one or more attested examples of the morpheme.

224. The collection of claim 221 wherein the ending phoneme that is a vowel in the phoneme class structure of a morpheme may be a vowel sequence in one or more attested examples of the morpheme.

225. The collection of claim 221 wherein the ending phoneme that is a vowel in the phoneme class structure of a morpheme may be a vowel doubling in one or more attested examples of the morpheme.

226. The collection of claim 210 wherein the phoneme class structures for the preponderance of morphemes in the data set describe a sound that has the form consonant-vowel-consonant.

227. The collection of claim 226 wherein one or more of the phonemes that is a consonant in the phoneme class structure of a morpheme may be a consonant blend in one or more attested examples of the morpheme.

228. The collection of claim 226 wherein one or more of the phonemes that is a consonant in the phoneme class structure of a morpheme may be a consonant doubling in one or more attested examples of the morpheme.

229. The collection of claim 226 wherein the phoneme that is a vowel in the phoneme class structure of a morpheme may be a vowel sequence in one or more attested examples of the morpheme.

230. The collection of claim 226 wherein the phoneme that is a vowel in the phoneme class structure of a morpheme may be a vowel doubling in one or more attested examples of the morpheme.

231. The collection of any one of the preceding claims 185 -230 wherein the data set includes an explicit identification of a phoneme class structure for each entry in the data set.

232. The collection of claim231 wherein the data set is organized by the first phoneme class of the phoneme class sequence describing the sound of a morpheme entry.

233. The collection of any one of the preceding claims 185-232 further including at least one word associated with each morpheme, wherein at least a part of the word is represented by a sound element associated with the at least part of the word.

234. The collection of claim 233 wherein the at least a part of the word is represented by a phoneme class associated with a letter of the word.

234. The collection of claim 233 wherein the at least a part of the word is represented by a phoneme class associated with at least one letter of the word.

235. The collection of any one of the preceding claims 233-234 wherein the at least a part of the word is represented by a phoneme class structure associated with a syllable of the word.

236. The collection of any one of the preceding claims 185 -235 wherein each morpheme is represented by a phoneme class structure.

237. The collection of claim 236 wherein the collection is indexed according to phoneme class structure.

238. The collection of any one of the preceding claims 185-237 further including data associated with each morpheme including one or more selected from the group of phoneme class structure, literal spelling(s), meaning(s), category(ies), negation status(es), syllable count(s), position(s), attested examples.

239. The collection of claim 238 further including a category associated with each morpheme and wherein the category identifies a function for any that are semantically defined.

240. The collection of claim 238 wherein attested examples include words evaluated as both having the same meaning and containing a word-part that is identical to the morpheme entry in phoneme class structure.

241. The collection of any one of the preceding claims 238-240 wherein attested examples include phrases containing within them a word evaluated as both having the same meaning and containing a word-part that is identical to the morpheme entry.

242. The collection of claim 238 further including attested examples whose meaning is an idiomatic expression or “figure of speech.”243. The collection of claim 240 wherein a meaning of an attested example is evaluated against a meaning of the morpheme and a similarity of meaning of a word or word-part in the attested example and the meaning of the morpheme is a function of at least one of a number of correspondences between and a degree of synonymity between one or more of the keywords of a morpheme and one or more of the keywords from the definition string for the word or word-part in the attested example.

244. The collection of claim 243 wherein at least one attested example is not a proper noun.

245. The collection of claim 243 wherein none of the attested examples is a proper noun.

246. The collection of claim 238 wherein the collection includes associated data in the form of a negation status that identifies whether the morpheme has a frequency of occurrence in negation status in the form of often, sometimes, or rarely associated with a negative sense.

247. The collection of claim 238 wherein the collection includes associated data in the form of a syllable count that identifies a number of syllables in the morpheme.

248. The collection of claim 238 wherein the collection includes associated data in the form of a position of the morpheme that belongs to a set including prefix, root, suffix, or a combination thereof.

249. The collection of any one of the preceding claims 212-248 wherein the morpheme has a source language and wherein the attested examples include at least one in the morpheme’s source language.

250. The collection of any one of the preceding claims 212-249 wherein the attested examples have a source language and the source language of at least one of the attested examples is in an ancient language.

251. The collection of claim 250 wherein the ancient language of at least one attested example is Greek.

252. The collection of any one of the preceding claims 212-251 wherein attested examples further include at least one attested example in a second ancient language.

253. The collection of claim 252 wherein the second ancient language is Latin.

254. The collection of any one of the preceding claims 212-253 wherein the attested examples further include at least one attested example in a modern language.

255. The collection of any one of the preceding claims 212-254 wherein attested examples include attested examples in at least one ancient and at least one modern language.

256. The collection of any one of the preceding claims 212-255 wherein attested examples include attested examples from both English and non-English modern languages.

257. The collection of any one of the preceding claims 212-256 further including associated data for respective ones of morphemes shared by a preponderance of the attested examples for the respective morpheme.

258. The collection of any one of the preceding claims 185-257 wherein a literal spelling of the morpheme is attested by its being a portion of at least two words, the portions having both a sound sequence that is the same phoneme class structure and the same meanings to a selected degree of synonymity.

259. The collection of any one of the preceding claims 185-258 wherein the sound sequence is a phoneme class structure that is attested by being in a portion of at least two words having the same meaning to a selected degree of synonymity and spelled with letters belonging to the same respective phoneme classes and in the same order.

260. The collection of any one of the preceding claims 185-259 wherein the at least one meaning is represented by at least one morpheme keyword.

261. The collection of any one of the preceding claims 185-260 wherein the at least one meaning is derived from a meaning of at least one word in an ancient language containing the morpheme.

262. The collection of claim 261 wherein the ancient language is Greek.

263. The collection of any one of the preceding claims 185-262 wherein the at least one meaning of a preponderance of the entries is other than a proper noun.

264. The collection of any one of the preceding claims 185-262 wherein the at least one meaning of a preponderance of the entries is a conceptual meaning referring to any one or more of a class of objects that are concrete entities, concrete actions, or qualities of either concrete entities or concrete actions.

265. The collection of claim 264 wherein the conceptual meaning is attested by the respective morpheme being present as a portion of at least two words having the same meaning and the same phoneme class structure.

266. The collection of claim 265 wherein the respective conceptual meanings of the words which are the two attested examples are keywords within a definition string corresponding to a dictionary entry for said words.

267. The collection of any one of the preceding claims 185-266 wherein the meaning of the preponderance of the entries is syntax-neutral.268 The collection of any one of the preceding claims 185-267 and the meaning of some of the entries modifies at least one other morpheme in the word in which it is contained.

269. The collection of any one of the preceding claims 185-268 wherein meanings of a preponderance of the morphemes are based on the meaning of at least one ancient Greek word.

270. The collection of claim 269 wherein the category identifies whether a respective morpheme that is used as a modifier is definite or relative type of modifier.

271. The collection of claim 270 wherein the category identifies whether the morpheme, used as a modifier, is quantitative or qualitative type of modifier.

272. The collection of any one of the preceding claims 185-271 wherein each of the plurality of morphemes in the collection is differentiated by one or more of characteristics that are selected from a set that includes a sound sequence, the at least one meaning, and at least one literal spelling.

273. The collection of claim 272 wherein the set includes a phoneme class structure.

274. The collection of any one of the preceding claims 185-273 wherein the morphemes in the collection are not identical to other morphemes in the collection with respect to both a literal spelling and a meaning.

275. The collection of claim 274 wherein at least one morpheme is identical to another morpheme with respect to either literal spelling or meaning.

276. The collection of any one of the preceding claims 185-275 wherein the plurality of morphemes in the collection are a first plurality of morphemes and wherein a second plurality of morphemes in the collection are not identical to morphemes in the first plurality of morphemes with respect to both phoneme class structure and a meaning.

277. The collection of claim 276 wherein at least a first morpheme in the first plurality of morphemes is identical to a second morpheme in the second plurality of morphemes with respect to either phoneme class structure or meaning.

278. The collection of any one of the preceding claims 185-277 wherein one morpheme in the collection is linked with a second morpheme in the collection.

279. The collection of claim 278 wherein the link represents a sound similarity.

280. The collection of claim 278 wherein the link represents a sound similarity and a meaning similarity.

281. The collection of any one of the preceding claims 185-280 wherein each of some of the plurality of morphemes correlate respective ones of at least one Greek morpheme with at least one Latin morpheme with the same sound structure and meaning to a desired degree of synonymity.

282. The collection of any one of the preceding claims 185-281 wherein the collection is stored on a non-transitory machine-readable media.

283. An apparatus comprising instructions stored on non-transitory machine-readable media, the instructions adapted for execution by at least one processor for evaluating an input string, the instructions when executed causing the at least one processor to: identify a first word part; evaluate the first word part for one or more of sound data representing the word part, a semantic function of the word part, a type of modifier for the word part, and whether or not the word part is a non-modifying root, or one of a definite, relative, quantitative, or qualitative modifier.

284. The apparatus of claim 283 wherein the sound data represents sound class structure, and wherein the instructions when executed cause the at least one processor to associate a first sound class structure to the first word part.

285. The apparatus of any one of the preceding claims 283-284 wherein the sound data represents phoneme class structures, and wherein the instructions when executed cause the at least one processor to associate a first phoneme class structure to the first word part.

286. The apparatus of any one of the preceding claims 283-285 wherein the first word part includes a plurality of letters, and wherein the instructions when executed cause the at least one processor to associate a sound class to each letter of the first word part.

287. The apparatus of any one of the preceding claims 283-286 wherein the semantic function of the word part is represented by the presence of a meaning in a data set associated with the word part, and wherein the instructions when executed cause the at least one processor to associate with the first word part an indicator representing a determination that there is a meaning associated with the word part.

288. The apparatus of claim 287 wherein the meaning associated with the word part is found in a data set, and wherein the instructions when executed cause the at least one processor to associate a location in the data set with the first word part.

289. The apparatus of any one of the preceding claims 283-288 wherein a modifier type is one of definite quantitative, definite qualitative, relative quantitative, and relative qualitative, and wherein the instructions when executed cause the at least one processor to associate with the first word part a modifier type of one of definite quantitative, definite qualitative, relative quantitative, or relative qualitative.

290. A method of evaluating a word part of a word, comprising receiving a word or word part at an input; identifying a first word part from the input; evaluating the first word part for one or more of sound information, a semantic function of the word part, a type of modifier for the word part, and whether or not the word part is a non-modifying root, or one of a definite, relative, quantitative or qualitative modifier.

291. The method of claim 290 wherein evaluating the first word part for sound information includes evaluating the first word part for sound class information.

292. The method of any one of the preceding claims 290-291 wherein evaluating the first word part for sound information includes evaluating the first word part for phoneme class information.

293. The method of any one of the preceding claims 290-292 wherein evaluating the first word part for sound information includes identifying letters in the word part and assigning a sound class to each letter.

294. The method of any one of the preceding claims 290-292 wherein evaluating the first word part for sound information includes assigning to a letter in the word part a sound class of vowel, labial, dental, velar, liquid, nasal, semivowel, sibilant, or aspirate.

295. The method of any one of the preceding claims 290-294 wherein evaluating the first word part for a semantic function of the word part includes evaluating the first word part for whether or not it has a defined meaning.

296. The method of 295 wherein evaluating the first word part for a semantic function of the word part includes identifying an entry in a data set corresponding to the word part as having a meaning.

297. The method of claim 296 wherein evaluating the first word part by identifying an entry in a data set includes determining whether or not a keyword is in the data set corresponding to the first word part.

298. The method of claim 297 wherein evaluating the first word part by identifying an entry in a data set includes determining whether or not a definition keyword is in the data set corresponding to the first word part.

299. The method of any one of the preceding claims 290-298 wherein evaluating the first word part includes evaluating the first word part to determine if the first word part is associated with a morpheme.

300. The method of claim 299 wherein determining if the first word part is associated with a morpheme includes determining whether or not data associated with the first word part is one of a phoneme class structure is a match with or similar to a phoneme class structure associated with the morpheme, a meaning that is the same as a meaning of the morpheme to a specified degree of synonymity, and a meaning that is the same as a meaning associated with an attested example of the morpheme to a specified degree of synonymity.

301. The method of any one of the preceding claims 290-300 wherein evaluating the first word part for a type of modifier includes evaluating the first word part for one or more of a prefix, root, suffix, modified, unmodified, definite quantitative, definite qualitative, relative quantitative or relative qualitative.

302. The method of any one of the preceding claims 290-301 wherein evaluating the first word part includes evaluating a semantic correlation for the first word part and determining one or more of an exact synonym, a single synonym, a double synonym.

303. The method of any one of the preceding claims 290-302 further including generating a token with one or more of a representation of the word part, a representation of at least one sound class associated with the word part, a meaning, a keyword, a sound class structure, a phoneme class structure, a morpheme, a morpheme keyword, a modifier type that is one of definite, relative, quantitative, or qualitative.

304. The method of any one of the preceding claims 290-303 further including transmitting a token with one or more of a representation of the word part, a representation of at least one sound class associated with the word part, a meaning, a keyword, a sound class structure, a phoneme class structure, a morpheme, a morpheme keyword, a modifier type that is one of definite, relative, quantitative, or qualitative.

305. The method of any one of the preceding claims 290-304 wherein the method is embodied as a method of operating a mobile device application, and wherein performing the evaluating of the word part function includes wirelessly transmitting a token to a destination using a network device.

306. The method of any one of the preceding claims 290-304 wherein the method is embodied as a method of operating the at least one processor on a server, and wherein performing the evaluating of the word part includes wirelessly transmitting a token to a destination using a network device.

307. The method of claim 306 further including receiving a word or word part including the first word part from a user along with a request from the user to evaluate the first word part.

308. The method of any one of the preceding claims 290-307 further including carrying out a semantic lookup for the word.

309. The method of any one of the preceding claims 290-308 further including searching for a keyword that can be associated with the word to a desired degree of synonymity.

310. The method of any one of the preceding claims 290-309 further including evaluating a possible negation status of the first word part.

311. The method of any one of the preceding claims 290-310 further including evaluating possible hyphenations for the word.

312. The method of any one of the preceding claims 290-311 further including splitting the word into syllables.

313. The method of any one of the preceding claims 290-312 further including identifying vowels in the word.

314. The method of any one of the preceding claims 290-313 further including identifying any vowel sequences in the word.

315. The method of any one of the preceding claims 290-314 further including identifying any diphthongs in the word.

316. The method of any one of the preceding claims 290-315 further including identifying a number of syllables in the word.

317. The method of any one of the preceding claims 290-316 further including generating a string for the word including the first word part.

318. The method of any one of the preceding claims 290-317 further including converting letters in the first word part to respective phoneme classes.

319. The method of any one of the preceding claims 290-318 further including evaluating the first word part by comparing the first word part to a plurality of strings of phoneme classes in a data set.

320. The method of any one of the preceding claims 290-319 further including evaluating the first word part by comparing representations of the first word part to representations in a data set.

321. The method of claim 320 wherein representations in the data set include phoneme class structures wherein evaluating the first word part includes evaluating the first word part to phoneme class structures in the data set.

322. The method of any one of the preceding claims 320-321 wherein representations in the data set include a representation of a keyword and further including evaluating a word part keyword to the data set keyword.

323. The method of any one of the preceding claims 319-322 wherein the data set includes a plurality of morphemes, and further including associating a morpheme with the first word part.

324. The method of any one of the preceding claims 319-323 further including evaluating a keyword associated with the input word against meaning keywords in a data set.

325. The method of claim 324 further including evaluating the keyword associated with the input word against keywords in the data set to a desired degree of synonymity.

326. The method of any one of the preceding claims 290-325 further including combining a plurality of candidate solutions from a data set.

327. The method of any one of the preceding claims 290-326 further including ranking a plurality of candidate solutions from a data set in order of relevance.

328. The method of any one of the preceding claims 290-327 further including evaluating a plurality of word parts in the word for assigning one or more categories to each of the plurality of word parts.

329. The method of claim 328 further including evaluating each of the plurality of word parts for one or more of prefix, root, suffix, semantically defined, semantically undefined, modifying, non-modifying, modified, unmodified, relative, definite, quantitative, qualitative, definite quantitative, definite qualitative, relative quantitative or relative qualitative.

330. The method of any one of the preceding claims 290-329 further including generating a token comprising one or more of the word, the first word part, a morpheme-word-part, a representation of a meaning, a sound class structure, negation status, syllable count, synonyms, antonyms, diphthong, single-vowel string, multiple-vowel string, standalone vowel, silent vowel, doubled vowel, vowel sequence, syllabic vowel sequence, syllabic stand-alone vowel, position, attested examples, degree of synonymity, and ranking.

331. An apparatus for carrying out any of the methods of any one of the preceding claims 290- 330.

332. The apparatus of claim 331 wherein the apparatus includes at least a processor on a device of a user who input the word.

333. The apparatus of claim 332 wherein the device includes a network device and wherein the device is configured to send a token including the first word part and at least one result of the step of evaluating.

334. The apparatus of claim 333 wherein the device is configured to send a token with a phoneme class structure.

335. The apparatus of claim 331 wherein the apparatus includes at least a processor on a device of a host that received an input word from a user.

336. The apparatus of claim 335 wherein the apparatus is configured to send a token including the first word part and at least one result of the step of evaluating.

337. The apparatus of claim 336 wherein the device is configured to send a token with a phoneme class structure.

338. An apparatus comprising instructions stored on non-transitory machine-readable media, the instructions adapted for execution by at least one processor for evaluating an input string, the instructions when executed causing the at least one processor to: identify a first word part in response to a query from a user; evaluate the first word part for one or more of sound data representing the word part, a semantic function of the word part, a type of modifier for the word part, and whether or not the word part is a non-modifying root, or one of a definite, relative, quantitative or qualitative modifier; and output results from the evaluation for transmitting to the user information corresponding to the results.

339. The apparatus of claim 338 wherein the sound data represents sound class structures, and wherein the instructions when executed cause the at least one processor to transmit the data representing sound class structures.

340. The apparatus of any one of the preceding claims 338-339 wherein the sound data represents phoneme class structures, and wherein the instructions when executed cause the at least one processor to transmit the phoneme class structures and the first word part to the user.

341. The apparatus of any one of the preceding claims 338-340 wherein the instructions when executed cause the at least one processor to generate a token containing the results from the evaluation and transmitting the token to the user.

342. A method of evaluating a word part of a word, comprising: receiving from a user a word or word part at an input; identifying a first word part from the input;evaluating the first word part for one or more of sound information, a semantic function of the word part, a type of modifier for the word part, and whether or not the word part is a non-modifying root, or one of a definite, relative, quantitative or qualitative modifier; and preparing for output to the user information representing results of the evaluating.

343. The method of claim 342 further including generating a phoneme class structure representing the first word part, and preparing for output to the user representations of the phoneme class structure.

344. The method of any one of the preceding claims 342-343 further including evaluating sound information for the first word part and evaluating at least one meaning associated with the first word part and associating the sound information with the meaning, and preparing for output to the user representations of the sound information and the meaning.

345. An apparatus for evaluating one or more word parts of a word, the apparatus comprising: a processor for receiving or storing and executing instructions; an input and an output in communication with the processor and wherein the input and the output are configured for receiving one or more word parts of a word at the input for operation by the processor and receiving data from the processor to the output for a user; and instructions stored on non-transitory machine-readable media, which upon execution, evaluate a first word part for sound information and meaning information, and wherein the evaluation of the first word part for sound information includes identifying sound class information representing the first word part.

346. The apparatus of claim 345 wherein the stored instructions evaluate the first word part for phoneme class information.

347. The apparatus of claim 346 further including comparing the phoneme class information with data set phoneme class information, and identifying a data set word part having the same phoneme class information as the first word part, and associating the first word part with the data set word part.

348. The apparatus of claim 347 wherein identifying a data set word part having the same phoneme class information includes identifying a morpheme and associating the first word part with the morpheme.

349. The apparatus of claim 348 further including outputting to the user the first word part and the morpheme.

350. The apparatus of any one of the preceding claims 345-349 wherein the stored instructions evaluate the first word part for one or more characterizations of the first word part includesevaluating the first word part for one or more of a definition keyword, negation status, syllable count, single-vowel strings, multiple-vowel strings, stand-alone vowels, silent vowels, doubled vowels, vowel sequences, syllabic vowel sequences, syllabic stand-alone vowels, diphthongs, synonyms, degree of synonymity between a meaning of the first word part and words from a data set, prefix, root, suffix, semantically defined, semantically undefined, modifying, non-modifying, modified, unmodified, relative, definite, quantitative, qualitative, definite quantitative, definite qualitative, relative quantitative, and relative qualitative.

351. The apparatus of claim 350 wherein the stored instructions include instructions for outputting to the user one or more of the characterizations.

352. An apparatus comprising instructions stored on non-transitory machine-readable media, the instructions adapted for execution by at least one processor for evaluating an input string, the instructions when executed causing the at least one processor to: receive input from a user and identify a first word part in the input; evaluate the first word part for one or more of sound data representing the word part, a semantic function of the word part, a type of modifier for the word part, and whether or not the word part is a non-modifying root, or one of a definite, relative, quantitative or qualitative modifier; and present to the user results from the evaluation.

353. The apparatus of claim 352 wherein the sound data represents sound class structures, and wherein the instructions when executed cause the at least one processor to transmit the data representing sound class structures.

354. The apparatus of any one of the preceding claims 352-353 wherein the sound data represents phoneme class structures, and wherein the instructions when executed cause the at least one processor to transmit the phoneme class structures and the first word part to the user.

355. The apparatus of any one of the preceding claims 352-354 wherein the instructions when executed cause the at least one processor to generate a token containing the results from the evaluation and transmitting the token to the user.

356. The apparatus of any one of the preceding claims 352-355 further including instructions when executed cause the results to be stored on the user’s device.

357. The apparatus of any one of the preceding claims 352-356 further including instructions when executed cause the results to be modified when the at least one processor receives editing instructions from the user.

358. A method of evaluating a word part of a word, comprising: receiving from a user a word or word part at an input;identifying a first word part from the input; evaluating the first word part for one or more of sound information, a semantic function of the word part, a type of modifier for the word part, and whether or not the word part is a non-modifying root, or one of a definite, relative, quantitative or qualitative modifier; and preparing for output to the user information representing results of the evaluating.

359. The method of claim 358 further including generating a phoneme class structure representing the first word part, and preparing for output to the user representations of the phoneme class structure.

360. The method of any one of the preceding claims 358-359 further including evaluating sound information for the first word part and evaluating at least one meaning associated with the first word part and associating the sound information with the meaning, and preparing for output to the user representations of the sound information and the meaning.

361. The method of any one of the preceding claims 358-360 further including displaying and storing the results on the device of the user.

362. The method of any one of the preceding claims 358-361 further including editing the results based on input from the user.

363. An apparatus for evaluating one or more word parts of a word, the apparatus comprising: a processor for receiving or storing and executing instructions; an input and an output in communication with the processor and wherein the input and the output are configured for receiving one or more word parts of a word at the input for operation by the processor and receiving data from the processor to the output for a user; and instructions stored on non-transitory machine-readable media, which upon execution, evaluate a first word part for sound information and meaning information, and wherein the evaluation of the first word part for sound information includes identifying sound class information representing the first word part.

364. The apparatus of claim 363 wherein the stored instructions evaluate the first word part for phoneme class information.

365. The apparatus of claim 364 further including comparing the phoneme class information with data set phoneme class information, and identifying a data set word part having the same phoneme class information as the first word part, and associating the first word part with the data set word part.

366. The apparatus of claim 365 wherein identifying a data set word part having the same phoneme class information includes identifying a morpheme and associating the first word part with the morpheme.

367. The apparatus of claim 366 further including outputting to the user the first word part and the morpheme.

368. The apparatus of any one of the preceding claims 363-367 wherein the stored instructions evaluate the first word part for one or more characterizations of the first word part includes evaluating the first word part for one or more of a definition keyword, negation status, syllable count, single-vowel strings, multiple-vowel strings, stand-alone vowels, silent vowels, doubled vowels, vowel sequences, syllabic vowel sequences, syllabic stand-alone vowels, diphthongs, synonyms, degree of synonymity between a meaning of the first word part and words from a data set, prefix, root, suffix, semantically defined, semantically undefined, modifying, non-modifying, modified, unmodified, relative, definite, quantitative, qualitative, definite quantitative, definite qualitative, relative quantitative, and relative qualitative.

369. The apparatus of claim 368 wherein the stored instructions include instructions for outputting to the user one or more of the characterizations.

370. A method of analyzing a semantic function of each part of a word that has multiple wordparts and each word-part has one or more respective letters, the method comprising identifying a word-part as semantically defined by identifying at least one keyword in a data set representing a meaning for the word-part.

371. A computer-implemented method comprising: storing at least one word and at least one word part of the at least one word in memory; using at least one processor to: identify a first keyword representing a possible meaning for the word; identify at least one division for the word that represents a beginning or an ending of the at least one word part; compare at least one unit of the at least one word part to a sound class of at least one unit of a string in a data set containing at least a plurality of sound classes; determine a desired correlation between the at least one word part and the data set string containing the at least one unit; compare a data set keyword to the first keyword and determining a desired correlation between the at least one word and the first keyword; andstore at least one of the desired correlation between the at least one word part and the data set string and the at least one word and the first keyword.

372. The method of claim 371 wherein using the at least one processor to compare at least one unit of the at least one word part to a sound class includes using the at least one processor to compare at least one unit of the at least one word part to a phoneme class.

373. The method of any one of the preceding claims 371-372 wherein using the at least one processor to compare the data set keyword to the first keyword includes using the at least one processor to compare a morpheme keyword to the first keyword.

374. The method of any one of the preceding claims 371-373 further including comparing the first keyword to data associated with at least one of a plurality of attested examples from the data set, such as a secondary keyword.

375. The method of any one of the preceding claims 371-374 wherein using the at least one processor to determine a desired correlation between the at least one word part and the data set string containing the at least one unit includes using the at least one processor to compare the first keyword to a morpheme keyword corresponding to the data set string containing the at least one unit.

376. The method of any one of the preceding claims 371-375 wherein using the at least one processor to compare at least one unit of the at least one word part to a sound class of at least one unit of a string in a data set includes using the at least one processor to compare at least one unit of the at least one word part to a sound class of at least one unit of a string in a morpheme lexicon.

377. The method of any one of the preceding claims 371-376 further comprising using the at least one processor to identify a characterization of one or more of the word and the at least one word part.

378. The method of any one of the preceding claims 371-377 further comprising using the at least one processor to identify a modifier of one or more of the word and the at least one word part.

379. The method of any one of the preceding claims 371-378 further comprising using the at least one processor to identify a non-modifying root of the word.

380. The method of any one of the preceding claims 371-379 further comprising using the at least one processor to identify a modifier type of one or more of definite, relative, quantitative, and qualitative.

381. The method of any one of the preceding claims 371-380 further comprising using the at least one processor to generate a token containing data representing the sound class.

382. The method of claim 381 wherein further comprising using the at least one processor to generate a token containing data representing a phoneme class structure.

383. The method of claim 382 wherein further comprising using the at least one processor to transmit the token containing data representing a phoneme class structure.

384. The method of any one of the preceding claims 371-383 further comprising using the at least one processor to identify a negation status for the at least one word.

385. The method of any one of the preceding claims 371-384 wherein using the at least one processor to compare at least one unit of the at least one word part to a sound class of at least one unit of a string in a data set containing at least a plurality of sound classes includes using the at least one processor to compare at least one phoneme class structure of the word part to a one phoneme class structure in the data set.

386. The method of any one of the preceding claims 371-385 wherein using the at least one processor to compare a data set keyword to the first keyword and determining a desired correlation between the at least one word and the first keyword includes using the at least one processor to compare the first keyword to a morpheme keyword in the data set.

387. The method of any one of the preceding claims 371-385 wherein using the at least one processor to compare includes using the at least one processor to compare to data in a morpheme lexicon containing representations of a phoneme class structure, a negation status, category data and at least one attested example.

388. An apparatus comprising: at least one processor; memory to store words and word parts and sound classes; and instructions that when executed are to cause the at least one processor to execute the steps of any one of the preceding claims 371-387.

389. An apparatus to store in memory words and word parts and sound classes, said apparatus comprising instructions stored on non-transitory machine-readable media, the instructions when executed to cause at least one processor to execute the steps of any one of the preceding claims 371- 387.

390. A method of evaluating at least one text string of a word having a word part, wherein the at least one text string includes at least one unit representing an element of the text string, the method comprising: using a negation lookup module to identify a negation status for at least one of the word and the word part; converting the at least one text string to a phoneme class structure;comparing the phoneme class structure of the at least one text string to a phoneme class structure of a string in a data set; comparing a meaning of a word containing the at least one text string to a meaning in an entry of the data set; evaluating the word part for a semantic function; evaluating the word part for a modifier type; and identifying an entry in the data set having the same phoneme class structure as the phoneme class structure of the text string and saving the phoneme class structure and data corresponding to the entry in the data set.

391. The method of claim 390 further including evaluating the word with one or more of a semantic lookup module, a keyword tool module, a word splitter module, a diphthong lookup module, and evaluating the word part with one or more of a semantic function analyzer module, and a modifier analyzer module.

392. The method of any one of the preceding claims 390-391 further including tokenizing the word, the word part and the phoneme class structure.

393. The method of any one of the preceding claims 390-392 further including editing a token containing the word, the word part or the phoneme class structure.

394. The method of any one of the preceding claims 390-393 wherein the data set is a morpheme lexicon, and wherein comparing the phoneme class structure of the at least one text string to a phoneme class structure of a string in the data set includes comparing the phoneme class structure of the at least one text string to a phoneme class structure of a string in the morpheme lexicon.

395. The method of any one of the preceding claims 390-394 further including evaluating at least one of the word and the at least one text string using at least one of a categorizer module, a word structure analyzer module, a semantic function analyzer module, and a modifier analyzer module.

396. A method of evaluating a word, the method comprising: accessing a data set containing a phoneme class structure, a keyword representing a meaning of a second word or a word part and a plurality of attested examples, and comparing the phoneme class structure to a phoneme class structure of an input.

397. The method of claim 396 further including comparing an input word to at least one of the keyword and data associated with an attested example.

398. The method of any one of the preceding claims 396-397 further including generating a token containing the phoneme class structure.

399. An apparatus comprising at least one processor, memory to store words and word parts and phoneme class structures, and instructions that when executed are to cause the at least one processor to execute the steps of any one of the preceding claims 1-45, 48-175, 181-184, 290-330, 342-344, 358-362, 370-387, and 390-398.

400. An apparatus to store in memory words and word parts and phoneme class structures, the apparatus comprising instructions stored on non-transitory machine-readable media, the instructions when executed to cause at least one processor to execute the steps of any one of the preceding claims 1-45, 48-175, 181-184, 290-330, 342-344, 358-362, 370-387, and 390-398.

Images