Methods and systems for understanding a meaning of a knowledge item, using information associated with the knowledge item

Abstract

"METHODS AND SYSTEMS TO UNDERSTAND A MEANING OF A KNOWLEDGE ITEM, USING INFORMATION ASSOCIATED WITH THE KNOWLEDGE ITEM". Systems and methods for determining the meaning of a knowledge item using related information are described. In one aspect, the knowledge item is received, the related information associated with the knowledge item is received, at least one related meaning based on the related information is determined and a knowledge item meaning for the knowledge item based at least , in part, the related meaning is determined. Some algorithms and types of related information useful in the realization of such systems and methods are described.

Description

Descriptive report of the Invention Patent for: * # * METHODS AND SYSTEMS TO UNDERSTAND A MEANING OF A KNOWLEDGE ITEM, USING INFORMATION ASSOCIATED WITH THE KNOWLEDGE ITEM * # *.Field of the InventionThe invention refers, in general, to knowledge items. More particularly, the invention relates to methods and systems for understanding the meaning of knowledge items, using information associated with the knowledge item.Background of the InventionTwo items of knowledge are sometimes associated with each other through manual techniques orautomated. Knowledge items are anything physical or non-physical that can be represented by symbols and can be, for example, key words, nodes, categories, people, concepts, products, phrases, documents and other units of knowledge. The itemsknowledge can take any form, for example, a single word, a term, a short phrase, a document or some other structured or unstructured information. The documents include, for example, web pages of various formats, such as HTML, XML, XHTML; Portable Document Format (PDF) files; and word processor andapplication program document files. For example, a knowledge item, such as the content of a document, can be matched with another knowledge item, such as a keyword or advertisement. Similarly, a knowledge item, such as a document, can be associated with another document containing related content, so that the two documents can be viewed as being related.An example of the use of knowledge items is in Internet advertising. Internet advertising can take many forms. For example, a website publisher may allow advertising for a fee on their web pages. When the publisher wants to display an advertisement on a network page to a user, a facilitator can provide an advertisement for the editor to display on the network page. The facilitator can select advertising through a variety of factors, such as demographic information about the user, the category of the web page, for example, sports or entertainment, or the content of the web page. The facilitator can also match the content of the web page with an item of knowledge, such as a keyword, from a list of keywords. An advertisement associated with the matched keyword can then be displayed on thenetwork. A user can manipulate a mouse or other input device and * # * click * # * on advertising to see a network page on the advertiser's network website that offers good goods or services for sale.In another example of Internet advertising, the current matched keywords are displayed on a publisher’s network page in Related Links or similar section. Similar to the example above, the content of the web page is matched with one or more keywords, which are then displayed in the Related Links section, for example. When a user clicks on a particular keyword, the user may be directed to a search results page, which may contain a mix of advertisements and regular search results. Advertisers make an offer on the keyword to have their ads appear on a search results page for the keyword. A user can manipulate a mouse or other input device and * # * click * # * the advertisement to see a web page on the advertiser's web site offering articles or services for sale.Advertisers want the content of the network page to be closely related to advertising, because a user viewing the network page is more likely toclick on advertising and buy the articles or services being offered, if they are highly relevant to what the user is reading on the web page. The publisher of the web page also wants the content of the advertising to match the content of the web page, because the publisher is often compensated if the user clicks on the advertising and a mismatch could be offensive to the advertiser or publisher , in the case of sensitive content.Various methods have been used to match keywords with the content. Most of these methods have involved a form of text matching, for example, matching keywords with words contained in the content. The problem with text matching is that words can relate to multiple concepts, which can lead to content mismatch with the keyword.For example, the term * # * apple * # * (apple) can relate to at least two concepts. Apple can refer to the fruit or the computer company of the same name. For example, a web page may contain a story about Apple Computer and the keyword most frequently used on the web page, in this case * # * apple * # *, could be chosen to represent the web page. In thisFor example, it is desirable to display an advertisement for Apple Computer and not apple, the fruit. However, if the highest bidder on the keyword * # * apple * # * is an apple seller and if the keyword * # * apple * # * is matched with the web page, advertising about apples, the fruit, will be displayed on the page of the network that deals with Apple, the computer company. This is undesirable, because a web page reader about a computer company is also probably not interested in buying apples.Mismatching knowledge items, such as keywords, with content can result in irrelevant advertisements being displayed for content. Therefore, it is desirable to understand the meaning of knowledge items.summaryEmbodiments of the present invention comprise systems and methods that understand the meaning of knowledge items using related information. One aspect of an embodiment of the present invention comprises receiving a knowledge item and receiving related information associated with the knowledge item. This related information by including a variety of information, such as related documents andrelated data. Another aspect of an embodiment of the present invention comprises determining at least one related meaning based on the related information and determining a meaning for the knowledge item based at least in part on the meaning of the related information. A variety of algorithms using the related meaning can be applied to these systems and methods. Additional aspects of the present invention are addressed to computer systems and computer-readable media, having characteristics related to the preceding aspects.Brief Description of DrawingsThese and other features, aspects and advantages of the present invention are best understood when the following Detailed Description is read with reference to the attached drawings, where:Figure 1 illustrates a block diagram of a system according to an embodiment of the present invention; Figure 2 illustrates a flowchart of a method ofaccording to an embodiment of the present invention; andFigure 3 illustrates a flowchart for a sub-routine of the method shown in figure 2.Detailed Description of Specific EmbodimentsThe present invention comprises methods and systems for understanding the meaning of knowledge items using the knowledge item itself, as well as information associated with the knowledge item. Reference will now be made in detail to the exemplary embodiments of the invention, as illustrated in the attached text and drawings. The same reference numbers are used for all drawings and in the description below for a reference to the same or equal parts.Various systems according to the present invention can be constructed. Figure 1 is a diagram illustrating an example system in which exemplary embodiments of the present invention can operate. The present invention can operate, and be realized, in other systems as well.The system 100 shown in figure 1 includes multiple client devices 102a-n, server devices 104, 140 and a network 106. The network 106 shown includes the Internet. In other embodiments, other networks, such as an intranet, can be used. In addition, the methods according to the present invention can operate on a single computer. Each of the client devices 102a-n shown includes computer-readable media, such as random access memory (RAM) 10 8, onshown embodiment coupled to a processor 110. Processor 110 executes a set of computer executable program instructions stored in memory 108. Such processors may include a microprocessor, an ASIC, and state machines. These processors include or may be in communication with media, for example, media that can be read on a computer, which store instructions that, when executed by the processor, cause the processor to perform the steps described here. Computer-readable media embodiments include, but are not limited to, an electronic, optical, magnetic or other storage or transmission device capable of providing a processor, such as the processor in communication with a sensitive input device touch, with instructions that can be read on a computer. Other examples of suitable media include, but are not limited to, a floppy disk, CD-ROM, magnetic disk, memory chip, ROM, RAM, an ASIC, a configured processor, all optical media, all tape media magnetic or other magnetic media, or any other medium from which a computer processor can read instructions. Also, several other forms of media that can be read on a computer can transmit or conductinstructions for a computer, including a router, a private or public network, or another wired and wireless transmission or channel device. The instructions can comprise code from any computer programming language, including, for example, C, C ++, C #, Visual Basic, Java and JavaScript.Client devices 102a-n may also include various external or internal devices, such as a mouse, CD-ROM, keyboard, display, or other input or output devices. Examples of 102a-n client devices are personal computers, digital assistants, personal digital assistants, cell phones, mobile phones, smart phones, radio call equipment, digital clipboards, laptop computers, a processor-based device and similar types of systems and devices . In general, a client device 102a-n can be any type of processor-based platform connected to a network 106 and which interacts with one or more application programs. Client devices 102a-n shown include personal computers running a browser application program, such as Internet Explorer ™, version 6.0 from Microsoft Corporation, Netscape Navigator ™,version 7.1 from Netscape Communications Corporation, and Safari ™, version 1.0 from Apple Computer. Through client devices 102a-n, users 112a-n can communicate with each other over network 106 and with other systems and devices coupled to network 106.As shown in figure 1, server devices 104, 140 are also coupled to network 106. The document server device 104 shown includes a server that runs a knowledge item agent application program. The server device 140 shown includes a server that runs a content agent application program. Similar to client devices 102a-n, each of the server devices 104, 140 includes a processor 116, 142 coupled to a memory that can be read on computer 118, 144. Each server device 104, 140 is described as a system single computer, but can be implemented as a network of computer processors. Examples of server devices 104, 140 are servers, large computers, networked computers, a processor-based device and similar types of systems and devices. Client processors 110 and server processors 116, 142 can be any of several computer processors as wellknown, such as Intel Corporation's processorsSanta Clara, California and Motorola Corporation of Schaumburg, Illinois.Memory 118 of document server device 104 contains a knowledge item processor application program, also known as knowledge item processor 124. Knowledge item processor 124 determines a meaning for knowledge items. The meaning can be a contextual representation and can be, for example, a vector of weighted concepts or groups or groupings of words. Knowledge items can be received from other devices connected to network 106, such as, for example, server device 140.Knowledge item processor 124 can also match a knowledge item, such as a keyword, to an article, such as a web page, located on another device connected to the network 106. Articles include documents, for example , web pages of various formats, such as HTML, XML, XHTML, Portable Document Format (PDF) files, and word processor document files, database and application program, audio, video or any other information of any kind that is made available on a network (such as the Internet), a personal computer or aother means of computing or storage. The embodiments described here are generally described in relation to documents, but can operate on any type of article. Knowledge items are usually anything physical or non-physical that can be represented by symbols and can be, for example, keywords, nodes, categories, people, concepts, products, phrases, documents and other units of knowledge. Knowledge items can take any form, for example, a single word, a term, a short sentence, a document or some other structured or unstructured information. The embodiments described here are generally described in relation to the keywords, but the embodiments can operate on any type of knowledge item.Memory 144 of server device 140 contains a content agent application program, also known as content agent 146. In one embodiment, content agent 146 receives a matched keyword from knowledge item agent 124 and associates a document , like an advertisement, with it. The advertisement is then sent to a website of the requester's network and placed in a frame on a network page, for example. In one embodiment, content agent 146 receivesrequests and returns content, such as advertisements, and correspondence is handled by another device.The knowledge item agent 124 shown includes an information finder 136, a knowledge item processor 135 and a meaning processor 136 In the embodiment shown, each comprises a memory resident computer code 118. The knowledge item processor 135 receives a keyword and identifies information known about the keyword. Known information can include, for example, one or more concepts associated with one or more terms analyzed from the keyword. A concept can be defined using a grouping or a set of words or terms associated with it, where the words or terms can be, for example, synonyms. For example, the term 'apple' can have two concepts associated with it - fruit and the company of computers - and thus each can have a grouping or set of related words or terms. A concept can also be defined by various other information, such as, for example, relationships with related concepts, the intensity of relationships with related concepts, parts of speech, common usage, frequency of use, the breadth of the concept and other usage statistics concept in language.The information locator 136 identifies and retrieves related information associated with keywords. In the shown embodiment, the related information could include related documents and additional related data. Related documents could include the text of advertisements and the target network website of advertisers who have made an offer on a keyword. Additional related data could include other keywords purchased by advertisers, search results on a keyword from a search agent, cost per click of advertisers and data related to the success rate of advertising. Some of that information can be obtained, for example, from server device 140. Information processor 136 processes related information located by information locator 134 to determine at least one related meaning for the related information located. This related meaning and the information known about the keyword is then passed to the meaning processor 137. The meaning processor 137 uses the information known about the keyword and the related meaning to determine the meaning of the word. -key. Note that other functions and features of the information finder 134, theknowledge item processor 135, information processor 136 and meaning processor 137 are described in more detail below.The server device 104 also provides access to other storage elements, such as a knowledge item storage element, in the example shown a meaning database 120. The knowledge item database 120. The knowledge database knowledge item can be used to store knowledge items, such as keywords and their associated meanings. The server device 140 also provides access to other storage elements, such as a content storage element, in the example shown a content database 148. The content database can be used for storing item related information knowledge, for example, documents and other data related to knowledge items. Data storage elements can include any or a combination of methods for data storage, including, without limitation, arrays, hash tables, lists and pairs. Similar types of data storage devices can be accessed by the server device 104.It shouldto benoticed that the present invention canunderstand systems having a different architecture than the one shown in figure 1. For example, in some systems according to the present invention, information locator 134 may not be part of knowledge item agent 124 and may perform its operations offline line. The system 100 shown in figure 1 is for example only, and is used to explain the exemplary methods shown in figures 2 to 3.Various methods according to the present invention can be performed. An exemplary method in accordance with the present invention comprises receiving a knowledge item, receiving related information associated with the knowledge item, determining at least one related meaning based on the related information and determining a meaning of a knowledge item. knowledge based, at least in part, on the related meaning of the related information. Related information can be associated with the knowledge item in any way and determined to be related in any way. Related information can comprise related articles and related data. Some examples of related articles include advertising by an advertiser who made an offer for a knowledge itemand a network page associated with advertising. 0 item fromknowledge can be, for example, a keyword. An example of related data comprises cost per click data and success rate data associated with advertising. In one embodiment, the meaning of knowledge item may comprise a weighted vector of related concepts or groupings of words.In one embodiment, the knowledge item is processed after being received to determine any associated known concepts. A concept can be defined by a grouping or group of words or terms. A concept can still be defined by various other information, such as, for example, relationships with related concepts, outside of relationships with related concepts, parts of speech, common use, frequency of use, the breadth of the concept and other statistics about the use of the concept in language. In one embodiment, determining the meaning of the knowledge item comprises determining which of the associated concepts represents the meaning of the knowledge item.In one embodiment, the knowledge item comprises a plurality of concepts and the related meaning comprises a plurality of concepts and the determination of the meaning of the knowledge item comprises the establishment of a probability for eachconcept of knowledge item that the knowledge item will be solved in part for the concept of knowledge item, determining the intensity of a relationship between each concept of knowledge item and each concept of related meaning and adjusting the probability for each concept knowledge item based on intensities. In one embodiment, the knowledge item has a plurality of concepts and a plurality of related meanings are determined, where each related meaning has a plurality of concepts. A knowledge item meaning determination involves establishing a probability for each knowledge item concept that the knowledge item will be resolved in part for the knowledge item concept and establishing a probability for each related meaning concept of that the knowledge item will be resolved in part to the concept of related meaning.Figures 2 to 3 illustrate an example method 200 in accordance with the present invention in detail. This exemplificative method is provided by way of example, as there are a variety of embodiments of the methods according to the present invention. The method 200 shown in figure 2 can be performed or performed in another way byany of several systems. The method 200 is described below as performed by the system 100 shown in figure 1 by way of example, and several elements of the system 100 are referenced in the explanation of the example method of figures 2 to 3. The method 200 shown provides an understanding of the meaning of a keyword using information associated with the keyword.Each block shown in figures 2 to 3 represents one or more steps performed in example method 200. With reference to figure 2, in block 202, example method 200 begins. Block 202 is followed by block 204, where a keyword is received by knowledge item agent 124. The keyword can be received, byexample, from an external database over network 106, such as the content database 14 8 or can be received from other sources.Then, in block 206, the keyword is processed by the knowledge item processor 135 to determine known information about the keyword. For example, the keyword can have one or more concepts associated with it. Each concept can have an associated grouping or group of words. A concept can also be defined by various other information, such as, for example, relationships with related concepts, the intensityof relationships with related concepts, parts of speech, common usage, frequency of use, the breadth of the concept and other statistics on concept usage in language.For example, for the term apple there may be two possible concepts associated with it. The first concept of apple, fruit, can be defined in relation to related words or concepts, such as fruit, food, pie and eating. The second concept of the apple, the computer company, can be defined through relationships with related words or concepts, such as computer, PC and technology. A keyword can be a short phrase, in which case the phrase can be separated by the knowledge item processor 135, for example, in individual terms. In this example, knowledge item processor 135 can further determine concepts associated with each term. In some embodiments, the keyword will have no information associated with it.Block 206 is followed by block 208 where related information associated with the keyword is identified by information locator 134 and received by information processor 136. Related information can include documents, such as text from advertisements and network sites of advertisers who made an offer for a keyword, search results on the networkabout the keyword itself and related data, such as other offers for the keyword made by the advertisers, the cost per click that the advertisers associated with the keyword are paying, the number of times a user has purchased an item after click through associated advertising to a site in the advertiser's network. This related information can be located from a variety of sources, such as, for example, server device 140, advertiser network sites and search agents.Block 208 is followed by block 210, where at least one related meaning is determined from the related information by information processor 136. For example, for each individual related document a meaning could be determined or a global meaning for all documents could be. be determined. For example, if the documents include the text of five advertisements associated with the keyword, a related meaning for each advertisement could be determined or the meanings of all five advertisements could be combined to provide a related global meaning. In one embodiment, documents are processed to determine a vector of weighted concepts contained in the documents. The concepts vectorweighted can represent the meaning of the document. For example, if the advertising refers to Apple Computers sales, the meaning of that advertising can be fifty percent computers, thirty percent Apple Computers and twenty percent sales. The related data can be used, for example, to adjust the weights of the meanings of individual documents or of the related global meaning. Alternatively, the meaning of a document could be groupings of related words.Block 210 is followed by block 212, where the meaning of the keyword is determined based on the meaning or related meanings by the meaning processor 137. The meaning processor 137 receives the meaning or related meanings from the information processor 136 and the processed keyword of knowledge item processor 135. For example, in block 212, the meaning processor will receive the keyword apple and its two related concepts from the knowledge item processor and will receive the related meaning of advertising for Apple Computers information processor 136. A variety of methods could be used to determine the meaning of the keyword based on the meaning or meaningsrelated information received from the information processor 136.For example, the related meaning can be used as a clue to determine the best concept to be associated with the keyword in order to provide meaning to the keyword. Where the related meaning is, for example, fifty percent computer, thirty percent Apple Computers, and twenty percent sales, the relationship between the weighted concepts of the related meaning and the keyword concepts could be used to indicate that the apple key will be associated with the concept of the computer company. Alternatively, the related meaning or meanings and related data can be used to develop a new meaning for the keyword.Any one or more of a variety of related information can be used to determine the meaning of a keyword. Examples of related information that can be used to determine the meaning of a keyword include, without limitation, one or more of the following:. The text of advertisements associated with advertisers who currently have made an offer on the knowledge item.. The network page or destination network pages foradvertising.. The text of advertisements from advertisers who have in the past made an offer regarding the keyword.. Other keyword-related offers made by advertisers who currently made an offer related to the keyword..Search results on a search agent's keyword.. The number of people who purchased an article, after viewing advertising, from an advertiser's network site, which is associated with the keyword.There are a variety of other related information that can be included and these are just examples. In addition, this related information can be given different weights, depending on some of the information. For example, the text of advertisements from current advertisers may be weighted more than the text of advertisements from previous advertisers associated with the keyword. Also, items associated with the highest cost-per-click advertiser can be more weighted, based on cost-per-click.Figure 3 illustrates a subroutine 212 for performing method 200 shown in figure 2. Subroutine 212 determines the meaning of the keyword using arelated meaning or related meanings. An example of the subroutine is as follows.The subroutine starts at block 300. At block 300, the probabilities for each set of words associated with the keyword are established. For example, in one embodiment each keyword can comprise one or more terms and each term can have one or more concepts associated with it. For the purposes of this example, the keyword comprises a single term with at least two related concepts. In block 300, each concept associated with the keyword is given an a priori probability that the keyword will be resolved for it. This a priori probability can be based on information contained in a network of interconnected concepts and / or on data previously collected on the frequency of each term being solved for the concept.Block 300 is followed by block 302, where the strength of the relationship is determined between the keyword concepts and the concepts of the meaning or related meanings. For example, in one embodiment, the related meaning can be understood from a weighted set of concepts. An intensity is determined for the relationship between each keyword concept and each concept of related meaning. The weight ofeach related meaning concept can be used to adjust the intensity of the relationship between the related meaning concepts and the keyword concept. The intensity can reflect the probability of co-occurrence between concepts or a measure of the connection between the two concepts, which can be derived from ontological data.Block 302 is followed by block 304, where the intensities computed in block 302 are used to adjust the probability that the keyword will be resolved for each of its associated concepts. For example, the intensities determined for the relationship between each keyword concept and each concept of related meaning are used to adjust the likelihood of each keyword concept being considered. In one embodiment, after the odds for the keyword concepts have been adjusted, the odds are normalized to one. The steps that occur in blocks 302 and 3 04 can be repeated a number of times in order to reinforce the impact of the relationship intensities on the probabilities.In one embodiment, the keyword can comprise multiple concepts and multiple related meanings can each comprise multiple concepts. In this embodiment, the meaning of the keyword canbe determined by establishing a probability for each keyword concept where the keyword would be solved in part for the keyword concept and a probability for each concept of related meaning that the keyword would be solved in part for the concept of related meaning. These probabilities can be established in the manner described with respect to figure 3.Now returning to figure 2, block 212 is followed by block 214, in which the meaning of the keyword is associated with the keyword and stored. The keyword and its associated meaning could be stored together, for example, in knowledge item database 120, or they could be stored separately, in separate databases.Although the above description contains many specificities, these specificities should not be construed as limitations on the scope of the invention, but merely as examples of the modalities shown. Those skilled in the art will see many other possible variations that are within the scope of the invention.

Claims

CLAIMS 1) A method characterized by the fact that it includes: to receive a knowledge item; Receive related information associated with the knowledge item; to determine at least one related meaning based on the related information; and To determine a meaning for a knowledge item based at least in part on related meaning. 2) Method, according to claim 1, characterized in that the knowledge item is a keyword. 3) Method, according to claim 1, characterized in that the related information comprises related articles. 4) A method according to claim 3, characterized in that the articles comprise an advertisement from an advertiser who has made an offer on an item of knowledge. 5) Method according to claim 4, characterized in that the articles still comprise a web page associated with the advertisement. 6) Method according to claim 5, characterized in that the related information still comprises related data. 7) Method according to claim 6, characterized in that the related data comprises cost-per-click data associated with the advertisement. 8) A method according to claim 1, characterized in that the reception of the knowledge item still comprises the processing of the knowledge item to determine any associated known concepts. 9) A method according to claim 1, characterized in that the knowledge item comprises a plurality of associated concepts and the determination of the meaning of the knowledge item comprises determining which of the associated concepts represents the meaning of the knowledge item. 10) Method, according to claim 1, characterized in that the knowledge item comprises a plurality of concepts and the related meaning comprises a plurality of concepts and the determination of the meaning of the knowledge item comprises: to establish a probability for each knowledge item concept that the knowledge item should be solved for the knowledge item concept; to determine the intensity of the relationship between each concept of a knowledge item and each related concept of meaning; and Adjust the probability for each knowledge item concept based on the intensities. 11) Method, according to claim 1, characterized in that the knowledge item comprises a weighted vector of concepts. 12) Method, according to claim 1, characterized in that the meaning of the knowledge item comprises clusters of related words. 13) Method, according to claim 1, characterized in that the knowledge item comprises a plurality of concepts, a plurality of related meanings, each related meaning comprising a plurality of concepts, and the determination of the meaning of the knowledge item comprises: to establish a probability for each knowledge item concept that the knowledge item should be solved, in part, for the knowledge item concept; and to establish a probability for each related meaning concept that the knowledge item should be solved, in part, for the related meaning concept. 14) Computer-readable media containing program code characterized by comprising: program code for to receive one item knowledge; Program code for receiving related information associated with the knowledge item; program code to determine at least one related meaning based on related information; and Program code to determine a meaning of a knowledge item based at least in part on related meaning. 15) Computer-readable media according to claim 14, characterized in that the knowledge item is a keyword. 16) Computer-readable media according to the Claim 14, characterized by the fact that the related information comprises related articles. 17) Computer-readable media according to claim 16, characterized in that the articles comprise an advertisement from an advertiser who has made an offer on a knowledge item. 18) Computer-readable media according to claim 17, characterized in that the articles still comprise a web page associated with the announcement. 19) Computer-readable media according to claim 18, characterized in that the related information still comprises related data. 20) Computer-readable media according to claim 19, characterized in that the related data comprises cost-per-click data associated with the advertisement. 21) Computer-readable media according to claim 14, characterized in that the program code for receiving the knowledge item still includes program code for processing the knowledge item to determine any associated known concepts. 22) Computer-readable media according to claim 14, characterized in that the knowledge item comprises a plurality of associated concepts and program code for determining the meaning of the knowledge item comprises program code to determine which of the associated concepts represents the meaning of the knowledge item. 23) Computer-readable media according to claim 14, characterized in that the knowledge item comprises a plurality of concepts and the related meaning to understand a plurality of Understanding concepts and determining the meaning of a knowledge item: Program code to establish a probability for each knowledge item concept that the knowledge item should be solved for the knowledge item concept; Program code to determine the intensity of the relationship between each knowledge item concept and each related meaning concept; and Program code to adjust the probability for each knowledge item concept based on intensities. 24) Computer-readable media according to claim 14, characterized in that the knowledge item comprises a weighted vector of concepts. 25) Computer-readable media according to claim 14, characterized in that the meaning of the knowledge item comprises clusters of related words. 26) Computer-readable media according to claim 14, characterized in that the knowledge item comprises a plurality of concepts, a plurality of related meanings, each related meaning comprising a plurality of concepts and the Determining the meaning of a knowledge item involves: Program code to establish a probability for each knowledge item concept that the knowledge item should be solved, in part, for the knowledge item concept; and Program code to establish a probability for each related meaning concept that the knowledge item should be solved, in part, for the related meaning concept.