A search system for word group input

Abstract

The present invention relates to the field of intelligent program technology, and discloses a search system for word group input, including a vocabulary definition module, a word collocation search module, a word association module, a related word combination module, an online search module, a vocabulary interpretation module, a synonym search module and a feedback adjustment module; the vocabulary definition module is used to set the word group priority requirements according to the situation, and give priority to generating work-related words in the work situation, and give priority to generating life-related words in the life situation; the word collocation search module is used to combine the input of the word, the situation requirements set by the user, the vocabulary interpretation module and the synonym search module. The present invention first searches and preliminarily screens words with similar meanings through the word collocation search module, and then performs priority matching through the related word combination module after association selection through the association module, so as to ensure that the input phrase is more suitable for the use situation.

Description

Technical Field

[0001] The present invention relates to the field of intelligent program technology, and more specifically to a search system for word group input. Background Art

[0002] Word group input needs to further output associative words that meet the dependency and relevance based on the input words and combined with the contextual environment frequently used by the user.

[0003] The existing patent discloses a method and device for intelligent word grouping, and a device for intelligent word grouping (publication number CN108073292A), wherein the method specifically includes: obtaining the user's input content; obtaining the words to be grouped corresponding to the input content, and the part of speech of each word to be grouped; determining the part of speech collocation score between adjacent words in the word grouping path corresponding to the words to be grouped according to the preset part of speech collocation rules and the part of speech of each word to be grouped; wherein the preset part of speech collocation rules are used to describe the collocation relationship between parts of speech. The technology disclosed in this patent lacks the combination of the contexts frequently used by users, and the similarity of the meaning of the output words, the degree of association of grammatical features, and the phonology are not taken into consideration; when outputting associative words, there is no consideration of whether the associated words and the selected words are dependent on each other, and whether the usage environment is similar. Summary of the invention

[0004] The main technical problem solved by the present invention is to provide a search system for word group input, which solves the problems in the above-mentioned background technology.

[0005] To solve the above technical problems, according to one aspect of the present invention, more specifically, a search system for word group input includes a vocabulary definition module, a word collocation search module, a word association module, a related word combination module, an online search module, a vocabulary interpretation module, a synonym search module and a feedback adjustment module;

[0006] The vocabulary definition module is used to set the word group priority requirements according to the scenario, giving priority to generating work-related vocabulary in the work scenario and giving priority to generating life-related vocabulary in the life scenario;

[0007] The word collocation search module is used to search and preliminarily form a vocabulary set that meets the requirements based on the input words and in combination with the scenario requirements set by the user, the input of the vocabulary interpretation module and the synonym search module;

[0008] The word association module is used to output a preliminary vocabulary set according to the output of the word collocation search module, and select the closest words from the vocabulary set through multimodal learning; and also associate and generate an associative vocabulary set based on the closest words;

[0009] The associated word combination module is used to screen the associated word set and match the associated words with the closest words, so as to combine and replace the associated words with the closest words;

[0010] The online search module is used to perform online search based on the closest words and associated words.

[0011] The vocabulary interpretation module is used to interpret the closest words produced and the associated words;

[0012] The synonym search module is used to compare the closest word with the input character or word, and also compare the difference between the closest word and the associated word;

[0013] The feedback adjustment module is used to feed back the closest words and associated words to the word interpretation module and the synonym retrieval module.

[0014] Furthermore, the word combination search module specifically includes: a search definition module, a word combination optimization module and a word combination learning module;

[0015] The search definition module is used to define the priority of vocabulary search according to the usage scenario, so as to output a vocabulary set that is more in line with the actual situation;

[0016] The word group optimization module is used to optimize the logic and grammar of the word group to ensure that the composed words are both in line with the situational needs and logical;

[0017] The word group learning module is used to learn the input vocabulary and the defined usage scenarios, and the usage scenario environment is not limited to one.

[0018] Furthermore, the word collocation search module performs preliminary screening of the retrieved words according to the word meaning similarity characteristics, grammatical feature correlation degree and phonological influence degree of the words, including:

[0019]

[0020] In the formula, W represents the correlation degree of words with mutual collocation attributes after retrieval and screening, c represents the similarity characteristics of the retrieved word meanings, and s represents the correlation degree of the grammatical characteristics of the retrieved words; and:

[0021]

[0022] In the formula, d represents the correlation between phonological and grammatical features, and m represents the phonological influence of the searched word;

[0023] When W ≥ 2.3, it means that the retrieved words meet the initial screening requirements of the input words.

[0024] Furthermore, the word association module specifically includes: a semantic representation module, a multimodal learning module, a vocabulary perception module and a word group dependency processing module;

[0025] The semantic representation module determines the common features and common interpretations of the vocabulary set according to the preliminary vocabulary set output by the word collocation search module, and performs multimodal learning based on the common features and common interpretations;

[0026] The multimodal learning module, in combination with a predefined learning program, collects the words that best fit the user's work and life scenarios, and these words have the same characteristics as the vocabulary set output by the word collocation search module;

[0027] The word group dependency processing module matches the collected words that best fit the user's work and life scenarios with the input words in terms of relevance and dependency, and outputs the word set whose relevance and dependency meet the requirements as an associative output.

[0028] Furthermore, the associated word combination module specifically includes: a feature matching module and a vocabulary association module;

[0029] The vocabulary association module is used to divide the degree of association between the initially formed vocabulary set and the associative vocabulary set;

[0030] The feature matching module matches and outputs the composed vocabulary set with the associative composed vocabulary set according to the association degree division made by the vocabulary association module.

[0031] Furthermore, the word association module will perform word association based on the commonalities of the initially screened words. The associated words are matched based on their dependencies with the screened words, similar usage environments, and semantic composition that meets the requirements. At this time, there are:

[0032]

[0033] Where K represents the matching index between the search word and the associated word, v represents the dependency between the search word and the associated word, and n represents the similarity between the usage contexts of the search word and the associated word; and:

[0034]

[0035] In the formula, u represents the influence of the situational environment on the semantic use, and r represents the degree of semantic consistency between the search word and the associated word.

[0036] The larger the value of K is, the higher the priority of the associated output words is, and the associated words are output first.

[0037] Furthermore, the feedback adjustment module will also feed back the associative words with the highest priority to the synonymous search module for retrieval, and the words retrieved by the synonymous search module will be corrected and output to the word collocation search module.

[0038] Furthermore, after the vocabulary interpretation module interprets the vocabulary closest to the associated vocabulary, the synonym search module needs to search and compare the two again, and the comparison results are output to the word collocation search module and the word association module for learning.

[0039] The beneficial effects of the search system for word group input of the present invention are:

[0040] 1. The present invention first searches and preliminarily screens words with similar meanings through a word collocation search module, and then performs priority matching through an association module after association selection, thereby ensuring that the input phrases are more suitable for the usage scenario.

[0041] 2. The present invention uses a word collocation search module to preliminarily screen the searched words according to the word meaning similarity characteristics, grammatical feature correlation degree and phonetic influence degree of the words, which can effectively improve the matching degree of the searched words.

[0042] 2. The present invention uses a word association module to perform vocabulary association based on the commonalities of the preliminarily screened words. The associated vocabulary is matched with the screened words because they have dependencies, similar usage environments, and their semantic components meet the requirements. This can greatly improve the relevance of the produced associated words. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The present invention is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0044] Figure 1 It is a structural schematic diagram of the present invention. DETAILED DESCRIPTION

[0045] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0046] like Figure 1 As shown, according to one aspect of the present invention, a search system for word group input is provided, comprising a vocabulary definition module, a word collocation search module, a word association module, a related word combination module, an online search module, a vocabulary interpretation module, a synonym search module and a feedback adjustment module;

[0047] The vocabulary definition module is used to set the priority requirements of word grouping according to the situation. In the work situation, work-related words are given priority, and in the life situation, life-related words are given priority.

[0048] The word collocation search module is used to search and preliminarily form a vocabulary set that meets the requirements based on the input words and in combination with the scenario requirements set by the user, the input of the vocabulary interpretation module and the synonym search module;

[0049] The word association module is used to output a preliminary vocabulary set based on the output of the word collocation search module, and select the closest words from the vocabulary set through multimodal learning; at the same time, it also associates and produces an associative vocabulary set based on the closest words. The word collocation search module first retrieves and preliminarily screens words with similar meanings, and after the association module associates and selects, the associated word combination module gives priority matching, which can ensure that the input phrases are more suitable for the usage scenario.

[0050] The associated word combination module is used to screen the associated word set and match the associated words with the closest words, so as to combine and replace the associated words with the closest words;

[0051] An online search module, used for performing online search based on the closest words and associated words;

[0052] The vocabulary explanation module is used to explain the closest words produced and the associated words;

[0053] A synonym search module is used to compare the closest word with the input character or word, and also compare the difference between the closest word and the associated word;

[0054] The feedback adjustment module is used to feed back the closest words and associated words to the vocabulary interpretation module and the synonym retrieval module.

[0055] In this embodiment, the word collocation search module specifically includes: a search definition module, a word group optimization module and a word group learning module;

[0056] A search definition module is used to define the search priority of vocabulary according to the usage scenario, so as to output a vocabulary set that is more suitable for the actual situation;

[0057] The word group optimization module is used to optimize the logic and grammar of word groups to ensure that the words formed are both in line with the situational needs and logical;

[0058] The word group learning module is used to learn the input vocabulary and the defined usage scenarios, so that the usage scenario environment is not limited to one.

[0059] The word collocation search module performs preliminary screening of the retrieved words based on the similarity of the word meanings, the degree of correlation of grammatical features, and the degree of influence of phonology.

[0060]

[0061] In the formula, W represents the correlation degree of words with mutual collocation attributes after retrieval and screening, c represents the similarity characteristics of the retrieved word meanings, and s represents the correlation degree of the grammatical characteristics of the retrieved words; and:

[0062]

[0063] In the formula, d represents the correlation between phonological and grammatical features, and m represents the phonological influence of the searched word;

[0064] When W ≥ 2.3, it means that the retrieved words meet the initial screening requirements of the input words.

[0065] When searching for synonyms, when "book" is input for similar word search, another similar word retrieved includes "calligraphy". The similarity of the meanings of the two words is c=21%, the degree of phonological influence of the two words is m=65%, and the degree of grammatical feature association of the two words is s=34%. Then:

[0066]

[0067] And there are:

[0068]

[0069] Since W=1.6, it means that the retrieved words do not meet the initial screening requirements of the input words.

[0070] The word collocation search module can be used to preliminarily screen the searched words according to the word meaning similarity characteristics, grammatical feature correlation degree and phonetic influence degree, which can effectively improve the matching degree of the searched words.

[0071] In this embodiment, the word association module specifically includes: a semantic representation module, a multimodal learning module, a vocabulary perception module, and a word group dependency processing module;

[0072] The semantic representation module determines the common features and common interpretations of the vocabulary set according to the output of the collocation search module, and performs multimodal learning based on the common features and common interpretations;

[0073] The multimodal learning module, combined with a pre-defined learning program, collects the most suitable vocabulary for the user's work and life scenarios, and these vocabulary have the same characteristics as the vocabulary set output by the word collocation search module;

[0074] The word group dependency processing module matches the collected words that best fit the user's work and life scenarios with the input words in terms of their relevance and dependency, and outputs the word sets whose relevance and dependency meet the requirements as associative words.

[0075] In this embodiment, the associated word combination module specifically includes: a feature matching module and a vocabulary association module;

[0076] A vocabulary association module is used to divide the degree of association between the initially formed vocabulary set and the vocabulary set formed by association;

[0077] The feature matching module matches and outputs the composed vocabulary set with the associated vocabulary set according to the association degree division made by the vocabulary association module.

[0078] In this embodiment, the word association module will perform word association based on the commonalities of the initially screened words. The associated words are matched if they are dependent on the screened words, have similar usage environments, and their semantic components meet the requirements. At this time, there are:

[0079]

[0080] Among them, K represents the adaptation index between the search word and the associated word, v represents the dependency between the search word and the associated word, and n represents the degree of semantic consistency between the search word and the associated word; and there are:

[0081]

[0082] In the formula, u represents the influence of the situational environment on the semantic use, and r represents the degree of semantic consistency between the search word and the associated word.

[0083] The larger the value of K is, the higher the priority of the associated output words is, and the associated words are output first.

[0084] The word association module associates the word "homework" with "book". The semantic coincidence degree between the search word and the associated word is r=78%, and the semantic coincidence degree between the search word and the associated word is n=83%. Then we have:

[0085]

[0086] And the dependency between the search word and the association word is v=87%, then:

[0087]

[0088] From the above calculation, it can be known that the search word "book" and the association word "homework" have a relatively high degree of consistency, and its output priority is as high as 78%.

[0089] The word association module will perform vocabulary association based on the commonalities of the initially screened words. The associated words will be matched based on their dependencies on the screened words, similar usage environments, and semantic composition that meets the requirements. This can greatly improve the relevance of the produced associated words.

[0090] In this embodiment, the feedback adjustment module also feeds back the associative words with the highest priority to the synonymous search module for retrieval, and the words retrieved by the synonymous search module are corrected and output to the word collocation search module;

[0091] After the vocabulary interpretation module interprets the vocabulary closest to the associated vocabulary, the synonym search module needs to search and compare the two again, and the comparison results are output to the word collocation search module and the word association module for learning.

[0092] Of course, the above description is not a limitation of the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by ordinary technicians in this technical field within the essential scope of the present invention also fall within the protection scope of the present invention.

Claims

1. A search system for word group input, characterized in that: It includes vocabulary definition module, word collocation search module, word association module, associated word combination module, online search module, vocabulary explanation module, synonym search module and feedback adjustment module; The vocabulary definition module is used to set the word group priority requirements according to the scenario, giving priority to generating work-related vocabulary in the work scenario and giving priority to generating life-related vocabulary in the life scenario; The word collocation search module is used to search and preliminarily form a vocabulary set that meets the requirements based on the input words and in combination with the scenario requirements set by the user, the input of the vocabulary interpretation module and the synonym search module; The word association module is used to output a preliminary vocabulary set according to the output of the word collocation search module, and to select the closest words from the vocabulary set through multimodal learning; and also to associate and generate an associative vocabulary set according to the closest words; The associated word combination module is used to screen the associated word set and match the associated words with the closest words, so as to combine and replace the associated words with the closest words; The vocabulary interpretation module is used to interpret the closest words produced and the associated words; The synonym search module is used to compare the closest word with the input character or word, and also compare the difference between the closest word and the associated word; The feedback adjustment module is used to feed back the closest words and associated words to the word interpretation module and the synonym retrieval module; The online search module is used to search for composed words and associated words; The word collocation search module performs preliminary screening of the retrieved words based on the word meaning similarity characteristics, grammatical feature correlation degree and phonological influence degree of the words, including: In the formula, Indicates the relevance of words with matching attributes after retrieval and screening. Represents the similarity feature of the retrieved word meaning, Indicates the degree of relevance of the grammatical features of the searched words; and has: In the formula, Indicates the correlation between phonological and grammatical features, Indicates the degree of phonological influence of the search term; when , it means that the retrieved words meet the initial screening requirements of the input words.

2. The search system for word group input according to claim 1, characterized in that: The word collocation search module specifically includes: a search definition module, a word group optimization module and a word group learning module; The search definition module is used to define the priority of vocabulary search according to the usage scenario, so as to output a vocabulary set that is more in line with the actual situation; The word group optimization module is used to optimize the logic and grammar of the word group to ensure that the composed words are both in line with the situational needs and logical; The phrase learning module is used to learn the input vocabulary and the defined usage scenarios, so that the usage scenario environment is not limited to one.

3. The search system for word group input according to claim 1, characterized in that: The word association module specifically includes: a semantic representation module, a multimodal learning module, a vocabulary perception module and a word group dependency processing module; The semantic representation module determines the common features and common interpretations of the vocabulary set according to the preliminary vocabulary set output by the word collocation search module, and performs multimodal learning based on the common features and common interpretations; The multimodal learning module, in combination with a predefined learning program, collects the words that best fit the user's work and life scenarios, and these words have the same characteristics as the vocabulary set output by the word collocation search module; The word group dependency processing module matches the collected words that best fit the user's work and life scenarios with the input words in terms of relevance and dependency, and outputs the word set whose relevance and dependency meet the requirements as an associative output.

4. The search system for word group input according to claim 3, characterized in that: The associated word combination module specifically includes: a feature matching module and a vocabulary association module; The vocabulary association module is used to divide the degree of association between the initially formed vocabulary set and the associative vocabulary set; The feature matching module matches and outputs the composed vocabulary set with the associative composed vocabulary set according to the association degree division made by the vocabulary association module.

5. The search system for word group input according to claim 4, characterized in that: The word association module will perform word association based on the commonalities of the initially screened words. The associated words are matched if they are dependent on the screened words, have similar usage environments, and their semantic components meet the requirements. At this time, there are: in, Indicates the matching index between the search term and the associated term. Indicates the degree of dependency between the search term and the associated term. Indicates the similarity between the usage contexts of the search term and the associated term; and: In the formula, Indicates the influence of situational environment on semantic usage, Indicates the degree of semantic consistency between the search term and the associated term; when The larger the value is, the higher the priority of the associated output words is, and the higher the priority of the associated words is output.

6. The search system for word group input according to claim 1, characterized in that: The feedback adjustment module will also feed back the associative words with the highest priority to the synonymous search module for retrieval, and the words retrieved by the synonymous search module will be corrected and output to the word collocation search module.

7. The search system for word group input according to claim 1, characterized in that: After the vocabulary interpretation module interprets the vocabulary closest to the associated vocabulary, the synonym search module needs to search and compare the two again, and the comparison results are output to the word collocation search module and the word association module for learning.

Images