JP2007257149A - Document processing apparatus and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a document processing apparatus which uses a method of representing concepts represented by a plurality of words and can use the relationships in meaning between the concepts as a dictionary, and which is suitable for increasing the accuracy of statistical analysis in text data analysis or the accuracy of document searches. <P>SOLUTION: The document processing apparatus 1 expresses concepts contained in a document as combinations of tokens and intent representations extracted from clause information contained in a document, and expresses the relationships between a plurality of concept representations based on information about the relationships between clauses. The document processing apparatus 1 includes a document data acquisition part 41 for acquiring a set of document data; a language processing part 42 for performing a language analysis process on the document data to extract concept representations contained in the document data; a concept representation display part 44 and a display 60 for displaying the concept representations; a concept representation thesaurus dictionary storage part 30 for storing a concept representation thesaurus dictionary; and a dictionary search part 43 for searching the concept representation thesaurus dictionary for concept representations. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a document processing apparatus and a document processing method.

  In recent years, a large amount of documents (text, text information) can be collected and stored, and text processing techniques and document processing techniques for utilizing such information have been developed. Conventional document processing technologies include document search technology that searches for documents that match the user's intention, document classification technology that classifies a large number of document sets into several subsets, and extracts important phrases from documents or document sets. There are a keyword extraction technique and a document summarization technique based on the keyword extraction technique.

  Recently, text mining technology for analyzing large amounts of text data such as questionnaire data and call center data has been developed. In order to analyze text data, these technologies classify them into categories based on the concepts and topics contained in the text, discover new ideas and knowledge, and use the number of texts belonging to the categories to find statistical trends and Analyze the relationship.

  As such text data analysis technology, there is one that performs analysis by performing statistical processing based on language analysis information. In this case, the problem is the processing of text including different expressions (words) but having the same meaning. For example, in the analysis of call center data, if you want to process text data that includes the content "customer is angry", instead of processing only text data that includes the word "angry", "get angry" Text data including different expressions, such as “coming to the head”, “prone”, “disturbed by wrinkles”, “pushing angry hair”, but having the same meaning must also be processed as text data having the same content. Otherwise, problems arise in the accuracy and accuracy of statistical analysis.

  There are synonym dictionaries, related word dictionaries, thesaurus dictionaries, and the like as techniques for handling information on words that are different words but have the same meaning. These are words and phrases that are semantically classified and organized. The aforementioned text data analysis technique also uses a thesaurus dictionary or a category dictionary. However, the thesaurus dictionaries and category dictionaries used in the text data analysis technique described above describe the relationship between words (or compound words) to the last, and are “angry” as expressions that represent the same meaning as “angry”, It was not possible to describe and use expressions expressing one meaning with a plurality of words, such as “coming to the head”, “disturbing the habit”, and “pushing angry hair”. For this reason, there is an inaccuracy in the statistical numerical value such as “the number of text data including a certain content”, and there is a problem in the accuracy and accuracy of the result of the statistical analysis based on such information.

On the other hand, a concept expression method capable of handling concepts expressed by a plurality of words has been proposed (see, for example, Patent Document 1). When this concept expression method is used, phrases expressed by a plurality of words can be handled as one concept expression. By constructing a thesaurus dictionary using this concept expression method, expressions such as `` get angry '', `` get angry '', `` coming to the head '', `` get angry '', `` disturb on the heels '', `` poke angry hair '' It can be described as conceptual expressions having the same meaning, and is expected to improve the problem. Further, when applying the concept expression method and analyzing the text set, the user is free to search for the concept expressions included in the document set and to generate the document category using the necessary concept expressions. (For example, refer to Patent Document 2). In addition, there is a technique in which the concept expression method is applied to extend target data to multi-field data, or an existing category definition can be reused (for example, see Patent Document 3).
JP 2005-063030 A JP 2004-334602 A JP 2004-303198 A

  However, in the conventional document processing apparatus and document processing method, synonymous expressions and related expressions are searched based on a thesaurus dictionary to which the above-described concept expression method is applied, and a plurality of different concept expressions are handled as one concept expression. There is room for further improvement in terms of performing more practical document processing in consideration of user convenience.

  The present invention was made in order to solve the conventional problem, using a concept expression method for expressing a concept represented by a plurality of words, and using a semantic relationship between concepts as a dictionary, An object of the present invention is to provide a document processing apparatus and a document processing method suitable for improving accuracy of statistical analysis in text data analysis or improving accuracy of document search.

  The present invention has a concept expression thesaurus dictionary composed of concepts expressed by the concept expression method described in Patent Document 1, and searches for synonymous expressions and related expressions based on the concept expression thesaurus dictionary, and makes a plurality of different expressions. The concept expression is handled as one concept expression.

  The document processing apparatus of the present invention according to claim 1 expresses a concept included in a document by a combination of a token extracted from the phrase information included in the document and an intention expression, and further includes a plurality of information based on relationship information between phrases. A document processing apparatus that expresses a relationship between concept expressions and processes document data based on a concept expression based on a combination of the token and the intention expression and a relationship between the concept expressions, and acquires a set of document data to be processed Document data acquisition means, language processing means for performing language analysis processing on the document data acquired by the document data acquisition means, and extracting concept expressions included in the document data, and concepts extracted by the language processing means Conceptual expression for storing a conceptual expression display means for displaying an expression, and a conceptual expression thesaurus for registering a conceptual expression expressing the concept included in the document A thesaurus dictionary storage means, and has a configuration including a dictionary search means for searching the concept expressed in the concept representation thesaurus.

  A document processing apparatus according to a second aspect of the present invention is the document processing apparatus according to the first aspect, wherein the conceptual expression specifying means for specifying a conceptual expression included in a set of document data acquired by the document data acquiring means, and the conceptual expression The concept expression specified by the specifying means is defined as a category definition, a document group is generated from a set of document data based on the category definition, and the document category is generated by the document category generating means. And a document category display means for displaying the document category.

  According to a third aspect of the present invention, there is provided the document processing apparatus according to the first or second aspect, wherein the concept expression extraction extracts one concept expression based on a synonymous concept expression group registered in the concept expression thesaurus dictionary. And a concept expression display means for displaying the synonymous expression group concept expression extracted by the concept expression extracting means.

  A document processing apparatus according to a fourth aspect of the present invention is the document processing apparatus according to any one of the first to third aspects, wherein the syntactic conceptual expression and the synonym registered in the conceptual expression thesaurus dictionary are not semantically. Synonym related concept expression specifying means for specifying concept expressions included in a set of document data to search related concept expressions, and synonyms for displaying search results of the synonymous concept expressions and the related concept expressions Related concept expression display means, the dictionary search means based on the designation of the synonym related concept expression designation means, the synonymous concept expression and the related concept expression registered in the concept expression thesaurus dictionary And the synonym related concept expression display means displays the search result of the dictionary search means.

  According to a fifth aspect of the present invention, there is provided the document processing apparatus according to any one of the first to fourth aspects, for searching for an extended concept expression obtained by extending a concept expression registered in the concept expression thesaurus dictionary. An extended concept expression specifying means for specifying a concept expression included in a set of document data, and a specified concept expression registered in the concept expression thesaurus dictionary based on the specification of the extended concept expression specifying means. Extended concept expression search means for searching and collecting the one synonym concept expression from the set of document data, and the concept expression display means includes the extended concept expression search means. The search result is displayed.

A document processing apparatus according to a sixth aspect of the present invention provides the document processing apparatus according to any one of the second to fifth aspects, wherein a new conceptual expression is generated based on the document category generated by the document category generating means. It has a configuration provided with dictionary registration means for registering in the thesaurus dictionary.
In the document processing method of the present invention according to claim 7, the concept included in the document is expressed by a combination of a token extracted from the phrase information included in the document and the intention expression, and a plurality of information is based on the relationship information between the phrases. A document processing method for processing document data by using a concept expression thesaurus that expresses a relationship between concept expressions and registers a concept expression by a combination of the token and the intention expression and a relationship between the concept expressions, and is a processing target. A document data acquisition procedure for acquiring a set of document data; a language processing procedure for performing a language analysis process on the document data acquired by the document data acquisition procedure and extracting a concept expression included in the document data; and the language A concept expression display procedure for displaying the concept expression extracted by the processing procedure, and a dictionary search procedure for retrieving the concept expression from the concept expression thesaurus dictionary , Is the execution.

  A document processing method according to an eighth aspect of the present invention is the document processing method according to the seventh aspect, wherein the conceptual expression specifying procedure for specifying a conceptual expression included in a set of document data acquired by the document data acquiring procedure, and the conceptual expression The concept expression specified by the specifying means is defined as a category definition, a document group is generated from a set of document data based on the category definition, and the document category is generated by the document category generating means. And a document category display procedure for displaying the document category.

  A document processing method according to a ninth aspect of the present invention is the document processing method according to the seventh or eighth aspect, wherein one conceptual expression is extracted based on a synonymous conceptual expression group registered in the conceptual expression thesaurus dictionary. A procedure and a concept expression display procedure for displaying the synonym expression group concept expression extracted by the concept expression extraction procedure.

  The document processing method of the present invention according to claim 10 is semantically synonymous with synonymous concept expressions and synonyms registered in the concept expression thesaurus dictionary according to any one of claims 7 to 9. In order to search for related concept expressions, a synonym related concept expression designation procedure for designating concept expressions included in a set of document data, and a synonym for displaying the synonymous concept expressions and search results of the related concept expressions A related concept expression designation procedure, and in the dictionary search procedure, based on the designation of the synonym related concept expression designation procedure, the synonymous concept expression and the related concept registered in the concept expression thesaurus dictionary Expressions are searched, and in the synonym related concept expression display procedure, the search result of the dictionary search procedure is displayed.

  A document processing method according to an eleventh aspect of the present invention is the document processing method according to any one of the seventh to tenth aspects, wherein an extended concept expression obtained by extending a concept expression registered in the concept expression thesaurus dictionary is retrieved. The specified concept representation registered in the concept representation thesaurus dictionary based on the extended concept representation designation procedure for designating the concept representation included in the document data set and the designation of the extended concept representation designation procedure. And an extended concept expression search procedure for searching for the one synonymous concept expression from the set of document data, and executing the extended concept expression search procedure in the concept expression display procedure. Showing search results for.

  A document processing method according to a twelfth aspect of the present invention is the document processing method according to any one of the seventh to eleventh aspects, wherein the new concept expression is converted to the concept expression based on the document category generated by the document category generation procedure. A dictionary registration procedure for registering in the thesaurus dictionary is executed.

  The present invention provides a document data acquisition unit that acquires a set of document data to be processed, and performs a language analysis process on the document data acquired by the document data acquisition unit to extract a concept expression included in the document data A conceptual expression thesaurus that stores a conceptual expression thesaurus that registers a conceptual expression that expresses a concept included in the document, and a conceptual expression display that displays the conceptual expression extracted by the language processing means. By providing storage means and dictionary search means for searching a concept expression from the concept expression thesaurus dictionary, the semantic relation between concepts represented by a plurality of words, rather than the conventional semantic relation between words, is provided. Relationships can be used as a conceptual thesaurus dictionary, improving the accuracy of statistical analysis in text data analysis, or document search accuracy Is suitable above, it is possible to provide a more practical document processing apparatus.

  Hereinafter, a document processing apparatus and a document processing method according to an embodiment of the present invention will be described with reference to the drawings.

[Concept expression method]
Examples of dependency analysis and concept expression applied to the embodiment of the present invention are shown in FIG. 1, FIG. 2, FIG. 3, and FIG. Here, FIG. 1 is an example of a dependency analysis result in document processing. FIG. 2 is an example of a conceptual expression that can be generated from the document data of FIG. FIG. 3 is an example of a conceptual expression representing the relationship between basic units. FIG. 4 is an example of an extended search result of concept expressions. The concept expression is described in detail in the above-mentioned Patent Document 1.

  The concept representation described in Patent Document 1 is based on clauses or inter-phrase relation information obtained as a result of language analysis of document data (text data). As the language analysis, for example, morphological analysis and phrase dependency analysis can be used. Morphological analysis analyzes words contained in document data. Dependency analysis analyzes clauses included in document data, and analyzes clauses in a relationship between dependency and dependency as a relationship between clauses. For example, when the document data “software installation cannot be executed normally” is analyzed in language, information as shown in FIG. 1 can be obtained.

  In FIG. 1, “self” represents an independent word, and “attached” represents an attached word. Autonomous words are part-of-speech words such as verbs, adjectives and nouns, and adjuncts are part-of-speech words such as particles and auxiliary verbs. A normal phrase is composed of one independent word and zero or one or more attached words. Depending on the analysis method, there may be a result that a single phrase includes a plurality of independent words, but here, an analysis method for generating a phrase so that the phrase always includes only one independent word. Shall be used.

  The concept expression is expressed by a basic unit of the concept expression and a relation expression between the basic units. The basic unit of conceptual expression is expressed using tokens and intention expressions. A token is a word that expresses one meaning by itself, and an independent word can be used. For example, in FIG. 1, “software”, “install”, “normal”, and “execute” are tokens. The token expression can use the token notation, or can be converted to a representative token notation. On the other hand, the intention expression is an expression representing the addition of meaning by a word in a phrase, and the intention added to the phrase is analyzed by extracting a specific expression pattern of the word. For example, the expression “~ not (auxiliary verb)” or “~ z (auxiliary verb)” means “cancellation”, the expression “can do (auxiliary verb)” means “possible”, “~ tai (auxiliary verb)” "Can be said to have added the meaning of" request "to the phrase. For example, intention expressions “possible” and “cancellation” are extracted from the phrase “cannot be executed” in FIG. The intention expression can be expressed as, for example, “(+ cancellation)” and “(+ possible−cancellation)”, where “+ XX” indicates that the intention expression is added and “−XX” indicates that It means that no intention expression is added.

  The basic unit of concept expression is expressed by token only, intention expression only, or a combination of token and intention expression. The basic unit of the concept expression by the combination of the token and the intention expression corresponds to, for example, “execution (+ possible + cancellation)” of “concept expression 7” in FIG. Here, the combination of a token and an intention expression means that a token specified in a certain phrase is included and the specified intention expression is added to the phrase.

  The relationship between basic units indicates that there is a strong semantic relationship between basic units. This semantically strong relationship basically means being included in a clause having a dependency relationship. Assuming that the relationship between basic units is represented by “⇒”, for example, the concept representation “software → installation” of “concept representation 8” in FIG. 2 is expressed as “software” in the two clauses in the dependency relationship. "Means that the installation clause includes" installation "(" install software "). By using the phrase dependency relationship as the relationship between basic units, it is not possible to simply specify the co-occurrence appearance relationship in the document data like the logical expression “software & install” of words generally used in document search etc. It can be specified that the basic unit appears in the document data with a semantically strong relationship.

  In the clause dependency relationship, when a certain clause becomes a dependency clause, there is only one reception clause, but a plurality of dependency clauses can relate to the same one reception clause. For example, “Phrase 4” in FIG. 1 is a receiving clause of “Phrase 2” and “Phrase 3”. For this reason, there are two ways of expressing the relationship between basic units in the concept expression: when expressing the inter-phrase relationship of receiving clauses having a plurality of dependency clauses, and when not expressing them.

  1) When not expressing the inter-phrase relationship of receiving clauses having a plurality of dependency clauses, the concept representation is a simple one-dimensional list representation of the basic unit. This corresponds to “concept expression example 1” in FIG. 3, for example.

2) When expressing the inter-phrase relationship of receiving clauses having a plurality of dependency clauses, the concept representation is a tree representation of the basic unit. This corresponds to “concept expression example 2” in FIG. 3, for example.
In the case of 1), the concept expression is simple and easy for the user to understand and easy to perform operations such as expansion of the expression, but there is a problem that the complicated phrase dependency relation structure cannot be expressed. In the case of 2), a complicated clause dependency relation structure can be expressed, but it is considered complicated and difficult to understand for the user and difficult to operate. Although both 1) and 2) can be used, the following embodiments will be described using the concept expression method 1) that is easy for the user to understand and operate.

  Further, there is a concept retrieval method for extracting a concept expression obtained by extending a designated concept expression from document data based on the above-described concept expression. For example, when “installation → execution” of “concept expression 9” in FIG. 2 is specified, “concept expression 10” to “concept expression 12”, “concept” in FIG. The “concept expression 20” is extracted from the “expression 17” and is changed from “extended concept expression 1” to “extended concept expression 7” shown in FIG. That is, a concept expression including a desired basic unit is extracted as an extended concept from the concept expression (shown in FIG. 2) included in the document data of FIG.

[First Embodiment]
FIG. 5 shows a document processing apparatus according to the first embodiment of the present invention. This is provided with a concept expression thesaurus dictionary to which the concept expression method described in Patent Document 1 is applied, and synonymous expressions and related expressions are searched based on the concept expression thesaurus dictionary, and a plurality of different concept expressions are converted into one concept expression. Are treated as

  5, the document processing apparatus 1 includes a document data structure storage unit 10 that stores a document data structure generated by a document data structure generation unit 430 (shown in FIG. 11) in the language processing unit 42, and a language processing unit 42. A concept expression storage unit that assigns a unique identification number (ID) to the concept expression extracted by the concept expression extraction unit 440 (shown in FIG. 11) and stores and manages it together with information on the number of appearance documents and the appearance frequency, for example. 20, a concept expression thesaurus dictionary storage unit 30 that stores a concept expression thesaurus dictionary that represents a semantic relationship between the above-described concept expressions, a document processing unit 40 that performs various processes on input document data, and a user A specification receiving unit 50 for receiving and inputting a specification related to document processing based on the operation of the document, and a display 60 for displaying a document processing result and the like. It is a configuration that.

  In FIG. 5, a document processing unit 40 includes a document data acquisition unit 41 for inputting document data (document set data) to be processed, a language processing unit 42 for extracting concept expressions and searching for extended concept expressions. A dictionary search unit 43 for searching for synonymous expressions and related expressions of a specified concept expression using a concept expression thesaurus dictionary; and a concept expression display unit 44 for displaying the search result of the dictionary search unit 43 on the display 60, for example. ,have.

  Here, the document data acquisition unit 41 assigns a unique ID to the document set data and passes it to the language processing unit 42. The document set data may be input by a user operation, or text stored in an electronic file or database may be read by a scanner or the like.

  The concept expression thesaurus dictionary storage unit 30 stores and manages the concept expression thesaurus dictionary. This conceptual expression thesaurus dictionary can express, for example, synonyms, broad meanings, narrow meanings, relations and anomalies as semantic relations between the above-described conceptual expressions, but in this embodiment, conceptual expressions that represent synonymous expressions and related expressions. Use a thesaurus dictionary. Synonymous expressions are expressions that express the same meaning. For example, expressions such as “get angry”, “get angry”, “coming to the head”, and “disturbing the moth” are synonymous expressions. can do. Related expressions are expressions that are not the same meaning but that are semantically related.For example, for expressions such as “anger”, expressions such as “laugh”, “cry”, “surprise”, and “joy” are related expressions. can do.

  The components of the concept expression thesaurus dictionary are shown in FIGS. 6, 7, 8, and 9. FIG. In the concept representation thesaurus dictionary, a concept is represented by a list of one or more basic units of concept representation. This expression corresponds to “concept expression example 1” in FIG. As shown in FIG. 6, the basic unit of the concept expression is expressed using one token information (notation, part of speech) and zero or one or more intention expression (intention tag) lists. As shown in FIG. 7, the concept expression list includes a concept expression used in the concept expression thesaurus dictionary and an identification number (concept expression ID) assigned to the concept expression. The concept expression thesaurus dictionary manages concept expressions based on a concept expression list.

  In the concept expression thesaurus dictionary, a list of synonymous expression groups (synonymous expression group list) representing the same meaning is registered. As shown in FIG. 8, the synonym expression group list includes a synonym expression group ID, a synonym expression group label, and a belonging concept expression ID list.

  A related expression list is registered in the concept expression thesaurus dictionary. As shown in FIG. 9, the related expression list is a list of concept expression IDs and concept expression IDs assigned to concept expressions that are semantically related to the concept expression specified by the concept expression ID (related concept expression ID list). And a list of synonym expression group IDs (related synonym expression group ID lists) assigned to synonym expression groups that are semantically related to the concept expression specified by the synonym expression group ID.

  The concept expression thesaurus dictionary storage unit 30 stores information on synonymous expression relations and related expression relations between concept expressions in the concept expression thesaurus dictionary based on the above-described intention expression list, concept expression list, synonym expression group list, and related expression list. I manage.

  Further, in FIG. 5, when the conceptual expression “angry” is designated through the designation receiving unit 50, for example, the dictionary search unit 43 stores the conceptual expression list of FIG. 7 registered in the conceptual expression thesaurus dictionary, FIG. Based on the synonymous expression group list and the related expression list shown in FIG. 9, the synonymous expressions and related expressions are searched, and the synonymous expressions (synonymous concept expressions) and related expressions (related concept expressions) shown in FIG. 10 are acquired. Here, regarding the related expression, not only the related concept expression but also the information of the related synonymous concept expression group may be acquired.

  Next, the configuration of the language processing unit 42 is shown in FIG.

  In FIG. 11, a language processing unit 42 performs a morphological analysis and dependency analysis on document data, and a document data structure that converts the data structure of document data based on the analysis result of the language analysis unit 420. A generation unit 430, a concept expression extraction unit 440 that extracts a concept expression included in the document data using the data structure generated by the document data structure generation unit 430, and a concept expression obtained by extending the specified concept expression as document data. And an extended concept expression search unit 450 that extracts from.

  Here, when performing morphological analysis, the language analysis unit 420 analyzes words included in the document data of the document data set. When dependency analysis is performed, sentences and phrases included in the document data are analyzed, and phrases having a relationship between dependency and dependency are analyzed as a relationship between phrases. For example, morphological analysis and dependency analysis are performed on the sentence “Software installation cannot be executed normally”, and the analysis result shown in FIG. 1 is acquired.

  The document data structure generation unit 430 converts each document data into a predetermined data structure based on the analysis result of the language analysis unit 420. Here, as shown in FIG. 12, it is converted into a tree structure of “document”, “sentence”, and “sentence”. In addition, various pieces of information are assigned to each component of the data structure (shown in FIG. 13). Further, as shown in FIG. 14, a word list is generated in which unique identifiers are assigned to the words included in the document data. Here, the document data structure generation unit 430 can also calculate the part-of-speech information of the word list, the appearance frequency in the whole, or the number of appearing documents. The word list and the data structure are stored and managed by the document data structure storage unit 10.

  As shown in FIG. 13, the “document” (corresponding to FIG. 13A) as a component of the data structure manages a sentence ID list included in the document data, and “sentence” is the own sentence. It manages IDs and clause ID lists included in one's own sentence. Similarly, “Phrase” manages its own phrase ID, a word ID list of words included in the phrase, a related phrase ID list, and a received phrase ID. Here, the word ID is a word ID in the word list of FIG. The related phrase ID list is an ID of a related phrase that receives the relevant phrase. As in the above example, since a plurality of clauses can be related clauses for one received clause, they are managed by the related clause ID list. The received phrase ID is an ID of a received phrase that is a related phrase. A dependency clause can take only one receiving clause. Here, it is also possible to hold the type of dependency relationship as information managed by the clause. For example, whether the modification is a continuous modification or a continuous modification. In addition, the type of relationship can be described by the type of particles that connect phrases.

  In addition, the intention tag list in FIG. 13 is data of intention tags representing additional meanings for words or phrases. Here, an intention tag representing an additional meaning is given to a word or phrase in the document data structure from an attached word expression in the phrase. This intention tag is used to add a meaning to a clause as a tag when an attached word or the like in the clause represents a specific additional meaning. For example, the meaning tags “cancellation”, “request”, “possible”, and “question” are added to a phrase when a predetermined word appears in the phrase. Multiple intention tags may be attached to one clause. An example of the intention tag ID in the intention tag list is shown in FIG. In the concept expression, for example, an intention tag is represented by an expression “(+ cancellation + possible)”. An intention tag can be used alone or as a conceptual expression, or can be used in a form added to a word such as “execution (+ possible + cancellation)”.

  In FIG. 11, when the concept expression extraction unit 440 uses the document data structure to extract the concept expression included in the document data, the concept expression of a specific pattern is previously stored in the document data as preprocessing. All types included are extracted, and the user can search for the basic concept expression based on the extracted information of all types. Here, it is assumed that all types of five types of “1 word”, “2 words”, “3 words”, “1 word (+ intention tag)”, and “intention tag” are calculated in advance. When extracting the concept expression, the number of documents in which the concept expression appears or the appearance frequency is also calculated. The concept expression extraction unit 440 also extracts the concept expressions registered in the concept expression list of the concept expression thesaurus dictionary and calculates the number of documents in which the concept expressions appear or the appearance frequency. The concept representation extracted by the concept representation extraction unit 440 is stored in the concept representation storage unit 20. Note that the method of extracting a concept expression used in the present embodiment is described in detail in Patent Document 1.

  The extended concept expression search unit 450 uses the concept expression storage unit 20 to add a basic unit of a concept expression composed only of tokens to the concept expression specified by the user, and specify the pattern. The concept expression of the pattern to which the intention tag of the basic unit of the concept expression at the end of the generated concept expression is added is extracted. For example, when the concept expression “installation → execution” is designated for the document data “software installation cannot be executed normally” in FIG. 1, “extended concept expression 1” to “extended concept expression 4” in FIG. To extract.

  Further, the extended concept expression search unit 450 can perform an extended concept expression search by designating a synonym expression group concept expression (claim 5). When the synonym expression group concept expression is designated, an extended concept expression search is performed for all the concept expressions belonging to the synonym expression group, and the information is combined into one to perform an extended search. For example, if the synonym expression group concept expression “get angry (synonymous)” is specified as the target concept expression, the affiliation concept expressions are “anger”, “speak”, “head ⇒ come”, “障 ⇒ fail”, “belly ⇒ stand” Perform extended concept expression search for.

  Here, FIG. 16 shows an example of an extended search in which one basic unit of concept expression is added.

  In FIG. 16, (A) shows the result of the extended concept expression search for the belonging concept expression of the synonymous expression group concept expression. (B) is the result of rearranging the positions of the basic units of the conceptual expression to which the result of (A) is added, by tokens. (C) is obtained by replacing the belonging concept expression of the synonym expression group concept expression with the synonym expression group label. (D) is a summary of the same conceptual representation in (C). Here, (D) is extracted as a result of the extended search by the extended concept expression search unit 450. If the appearance frequency is required, the sum of the appearance frequencies is calculated when the same concept expression is put together.

  Next, FIG. 17 shows an example of an extended search for adding an intention tag of a concept expression basic unit at the end of a specified concept expression.

  In FIG. 17, (A) shows the result of the extended concept expression search for the belonging concept expression of the synonymous expression group concept expression. (B) is the result of rearranging the result of (A) with the added intention tag. (C) is obtained by replacing the belonging concept expression of the synonym expression group concept expression with the synonym expression group label. (D) is a summary of the same conceptual representation in (C). Here, (D) is extracted as a result of the extended search by the extended concept expression search unit 450. If the appearance frequency is required, the sum of the appearance frequencies is calculated when the same concept expression is put together.

  Further, FIG. 18 shows a search result when the extended search of FIGS. 16 and 17 is combined.

  As described above, the extended concept expression search unit 450 performs an extended search for adding one basic unit of a concept expression and an extended search for adding an intention tag of a concept basic unit at the end of a specified concept expression. The search results of the final extended concept expression in FIG. 18 are acquired by combining both search results.

  Next, the configuration of the concept expression extraction unit 440 is shown in FIG.

  In FIG. 19, the concept expression extraction unit 440 includes a token extraction unit 441, an intention expression extraction unit 442, a concept expression basic unit relationship extraction unit 443, a concept expression generation unit 444, and a concept expression basic unit generation unit 445. ,have. The conceptual expression basic unit generation unit 445 receives the token and the intention expression from the token extraction unit 441 and the intention expression extraction unit 442, respectively, and generates a basic unit of the concept expression based on them.

  Here, the concept expression generation unit 444 extracts the synonym expression group of the concept expression thesaurus dictionary as one concept expression (synonym expression group concept expression) and registers it in the concept expression storage unit 20. This synonym expression group concept expression is a collection of concept expressions belonging to the corresponding synonym expression group (for each synonym expression group label), and the frequency of occurrence of the synonym expression group concept expression is the frequency of the concept expression to which it belongs. It is expressed as the sum of appearance frequencies (or set sum). For example, the synonym expression group concept expression shown in FIG. 20 is generated based on the information in the concept expression list in FIG. 7, the synonym expression group list in FIG. 8, and the related expression list in FIG.

  The document processing apparatus 1 configured as described above will be described with reference to FIG.

  In the document processing unit 40, the dictionary search unit 43 checks whether the concept expression specified via the specification receiving unit 50 is registered in the concept expression list of the concept expression thesaurus dictionary (S101, S102). For example, the concept expression ID is acquired (S103), and the process proceeds to S104.

  Next, the dictionary search unit 43 includes the concept expression ID acquired in S103 in the belonging concept expression ID list for the synonym expression groups registered in the synonym expression group list based on the synonym expression group ID acquired in S104. (S105), and if it is included, the synonym expression group ID is acquired (S106).

Thereafter, the dictionary search unit 43 performs the processes of S105 and S106 for all the synonym expression groups registered in the synonym expression group list (S107 and S108).
Next, the related expression search procedure for the document processing apparatus 1 configured as described above will be described with reference to FIG.

  In the document processing unit 40, the dictionary search unit 43 checks whether the concept expression designated via the designation receiving unit 50 is registered in the concept expression list of the concept expression thesaurus dictionary (S201, S202). For example, the concept expression ID is acquired (S203), and the process proceeds to S204.

  In S204, the dictionary search unit 43 searches the related expression list for the same concept expression ID as the concept expression ID acquired in S203. If found, the related concept expression ID list and the related synonym expression group ID corresponding to the concept expression ID are found. A list is acquired (S205, S206).

  Next, when there is an unprocessed related synonym expression group ID in the related synonym expression group ID list, the dictionary search unit 43 adds the concept expression ID belonging to the related synonym expression group ID to the related concept expression ID list ( S207, S208). Thereafter, when the processes of S206 to S208 are performed on all the synonym expression groups in the related synonym expression group ID list, the synonym expression search is ended.

  Next, a procedure for generating the synonym expression group conceptual expression for the document processing apparatus 1 configured as described above will be described with reference to FIG.

  In the concept expression extraction unit 440, the concept expression generation unit 444 acquires the belonging concept expression ID list of one synonym expression group registered in the synonym expression group list of the concept expression thesaurus dictionary (S301).

Next, the concept representation generation unit 444 checks whether the concept representation in the belonging concept representation ID list is registered in the concept representation storage unit 20 (S302). If one or more concept expressions in the belonging concept expression ID list are registered, the synonym expression group corresponding to the concept expression is registered in the concept expression storage unit 20 as the synonym expression group concept expression (S303).
Thereafter, the concept representation generation unit 444 performs the processing from S301 to S303 for all the synonym representation groups registered in the synonym representation group list of the concept representation thesaurus dictionary (S304).

  Here, examples of the concept expression display of the concept expression display unit 44 are shown in FIGS.

  The concept expression display unit 44 displays the concept expression extracted by the language processing unit 42 on the screen of the display 60 based on the user's designation received and input via the designation receiving unit 50. In this screen, the user searches the concept expression by referring to the extracted concept expression or referring to the extended concept expression search result by using the basic concept expression browser 61 of FIG. 24, a basic concept expression display area 65, extended concept expression display areas 66 and 67, a category list display area 62, a workspace 63, and a text browser 64 are set in the basic concept expression browser 61. ing. The basic concept expression display area 65 displays the basic concept expression, the number of appearing documents indicating the number of document data containing the basic concept expression, and the part of speech of the basic concept expression. In the extended concept expression display areas 66 and 67, among the concept expressions displayed in the basic concept expression display area 65, an extended concept expression generated based on the concept expression designated by the user is displayed. In the category list display area 62 and the workspace 63, a category list of document categories generated by the document category generation unit 92 (shown in FIG. 29) is displayed. In the workspace 63, document categories are displayed graphically. The text browser 64 displays the document data distributed to the document categories generated by the document category generation unit 92. In addition, when a target category is input to the target category designation input field 68 based on a user's designation operation, conceptual expressions included in a document set composed of document data belonging to this category are displayed. . Further, the concept expression display unit 44 displays only the label on the screen of the display 60 as the concept expression of the synonym expression group as shown in FIG. Similarly, as shown in FIG. 26, the label of the synonymous expression group and the belonging concept expression are displayed.

  According to the document processing apparatus 1 according to the first embodiment of the present invention, the concept included in the document is expressed by a combination of the token extracted from the phrase information included in the document and the intention expression, and further, the inter-phrase interval A set of document data to be processed in a configuration in which a relationship between a plurality of concept expressions is expressed based on the relationship information, and the document data is processed based on a concept expression based on a combination of a token and an intention expression and the relationship between the concept expressions. A document data acquisition unit 41 (corresponding to document data acquisition means), and a language processing unit 42 that performs language analysis on the acquired document data and extracts a concept expression included in the document data ( Language expression means), a concept expression display unit 44 and a display 60 (corresponding to concept expression display means) for displaying the concept expression extracted here, a sentence A concept expression thesaurus dictionary storage unit 30 (corresponding to the concept expression thesaurus dictionary storage means) for storing a concept expression thesaurus dictionary for registering concept expressions expressing the concepts included in the dictionary, and a dictionary for retrieving concept expressions from the concept expression thesaurus dictionary And a search unit 43 (corresponding to dictionary search means). This configuration corresponds to an embodiment of the present invention according to claims 1 and 7.

  With this configuration, it is possible to use a semantic relationship between concepts expressed by a plurality of words, rather than a conventional semantic relationship between words, as a concept expression thesaurus dictionary. Useful for improving accuracy or document search accuracy. For example, the dictionary retrieval unit 43 retrieves the concept expression extracted by the language processing unit 42 from the concept expression thesaurus dictionary, whereby a document related to the synonymous concept expression can be easily acquired. Moreover, since the concept expression display unit 44 displays the concept expression extracted from the document data on the screen of the display 60, the user can easily confirm the concept expression, and usability is improved.

  Further, according to the present embodiment, a concept expression extraction unit 440 (corresponding to a concept expression extraction unit) that extracts one concept expression based on synonymous concept expression groups registered in the concept expression thesaurus dictionary, and a concept It has a configuration provided with a concept expression display unit 44 and a display 60 (corresponding to concept expression display means) for displaying the synonymous expression group concept expression extracted by the expression extraction unit 440. This configuration corresponds to an embodiment of the present invention according to claims 3 and 9.

  With this configuration, the search accuracy of documents corresponding to concept expressions belonging to the synonym expression group is increased. Moreover, since the conceptual expression display unit 44 displays the synonymous expression group conceptual expression on the screen of the display 60, the usability of the user is improved.

  In addition, according to the present embodiment, the designation receiving unit 50 for designating a concept expression included in a set of document data in order to search for an extended concept expression obtained by extending the concept expression registered in the concept expression thesaurus dictionary. , A concept expression display unit 44 and a display 60 (corresponding to an extended concept expression specifying means), and on the basis of this specification, the designated concept expression registered in the concept expression thesaurus dictionary is searched to obtain one synonym concept expression And an extended concept expression search unit 450 (corresponding to an extended concept expression search unit) for searching for the one synonymous concept expression from a set of document data to be processed. The search result of the concept expression search unit 450 is displayed on the screen of the display 60. This configuration corresponds to an embodiment of the present invention according to claims 5 and 11.

  With this configuration, the concept expression thesaurus dictionary is used to perform an extended concept expression search based on the user's specification, and the result of the extended concept expression search is displayed on the screen of the display 60, thereby improving user convenience.

  Although not shown in the above-described embodiment, the document processing apparatus 1 of the present invention is a personal computer (for example, a storage unit such as a CPU, RAM, ROM, and HDD, a communication unit, a display unit such as a liquid crystal display, a keyboard, A personal computer having the above-mentioned document data acquisition means, language processing means, concept expression display means, concept expression thesaurus dictionary storage means, dictionary search means, concept expression extraction means, and extended concept The same effect can be obtained by functioning as an expression designating means and an extended concept expression searching means. Further, the processing procedure shown in FIGS. 21 to 23, the aforementioned document data acquisition means, language processing means, concept expression display means, concept expression thesaurus dictionary storage means, dictionary search means, concept expression extraction means, extended concept expression designation means, You may make it memorize | store the program for making the said personal computer perform the process sequence which an extended concept expression search means performs in the said memory | storage part.

[Second Embodiment]
Next, a document processing apparatus according to the second embodiment of the present invention is shown in FIG. This is different from the first embodiment in that a concept expression thesaurus is used to search for synonymous expressions and related expressions of a specified concept expression and display a search result. In addition, the same code | symbol is provided to the same structure as 1st Embodiment, and description is abbreviate | omitted partially.

  In FIG. 27, the dictionary search result display unit 45 displays the synonymous expression and the related expression of the concept expression searched by the dictionary search unit 43 based on the designation of the concept expression received and input via the designation receiving unit 50. 60 screens are displayed.

  Here, FIG. 28 shows an example of a dictionary search dialog.

  For example, a concept expression (for example, “get angry”) is input into the search target concept expression input field 71 of the related expression search screen 70 displayed on the display 60 via the designation receiving unit 50 based on a user operation, and a search button. When 72 is input, the dictionary search unit 43 searches for a synonym expression and a related expression from the concept expression thesaurus dictionary as described above. The dictionary search result display unit 45 displays search results in predetermined search result fields 73 and 74 based on the search results of the dictionary search unit 43. In the related expression search screen 70 of FIG. 28A, the synonymous expression groups of the related expression search results are displayed only with labels. In the related expression search screen 70 of FIG. 28B, the synonymous expression group of the related expression search result is displayed with the label and the belonging concept expression. The designation of the concept expression is not limited to the present embodiment, and may be performed by selecting the concept expression displayed on the display 60.

  According to the document processing apparatus 1 according to the second embodiment of the present invention, a synonymous conceptual expression registered in the conceptual expression thesaurus dictionary and a conceptual expression that is not synonymous but semantically related are searched. Therefore, a designation receiving unit 50 for designating concept expressions included in a set of document data, a dictionary search result display unit 45 and a display 60 (corresponding to synonym related concept expression designating means), the synonymous concept expressions, A dictionary search result display unit 45 and a display 60 (corresponding to synonym related concept expression display means) for displaying the search result of the related concept expression. The dictionary search unit 43 is connected via the designation receiving unit 50. Based on the designation, the synonymous concept expression and the related concept expression registered in the concept expression thesaurus dictionary are searched, and the dictionary search result display unit 45 includes the dictionary search unit 4. The results of the search has a configuration which is adapted to be displayed on the screen of the display 60. This configuration corresponds to an embodiment of the present invention according to claims 4 and 10.

  With this configuration, synonymous expressions and related expressions of the concept expression designated by the user are retrieved from the concept expression thesaurus dictionary and displayed on the screen of the display 60, so that the user convenience is improved.

  Although not shown in the above-described embodiment, the document processing apparatus 1 of the present invention is a personal computer (for example, a storage unit such as a CPU, RAM, ROM, and HDD, a communication unit, a display unit such as a liquid crystal display, a keyboard, The same effect can be obtained even if the personal computer functions as the above-mentioned synonym related concept expression designation means and synonym related concept expression display means.

[Third Embodiment]
Next, FIG. 29 shows a document processing apparatus according to the third embodiment of the present invention. This is different from the first embodiment in that, for example, a document category is generated by the document category generation unit 92 and the generated document category is displayed by the document category display unit 94. In addition, the same code | symbol is provided to the same structure as 1st Embodiment, and description is abbreviate | omitted partially.

  Here, the document category is a document group having a category definition as a reference for the belonging document and having a document collected based on the definition as a member. FIG. 30 shows information for the document category storage unit 80 to store and manage this document category. As shown in FIG. 30, the document category is managed by category ID, category definition, belonging document ID list, number of belonging documents, and category label. The category label may be requested from the user. The category definition is managed separately from the category ID, the belonging document ID list, the number of belonging documents, and the category label.

  29, the document processing unit 40 includes a document category generation unit 92 that generates the document category, a dictionary registration unit 91 that registers the document category generated by the document category generation unit 92 in the thesaurus dictionary, and a document category generation. A document category storage unit 80 for storing the document category generated by the unit 92, a document category display unit 94 for displaying the document category generated by the document category generation unit 92 on the display 60, and a document including the specified conceptual expression A document data generation unit 93 that generates data, and a document data display unit 95 that displays the document data generated by the document data generation unit 93 on the display 60 are provided.

  Here, the document category generation unit 92 generates three types of document categories: a concept expression category, a composite category, and a data ID designation category.

  First, the concept expression category is a category that defines a concept expression. The user can generate a document category by designating a concept expression from a predetermined screen displayed on the display 60 by the concept expression display unit 44. For example, when a categorization operation is performed by the user from the predetermined screen and a concept expression is designated and input via the designation receiving unit 50, the document category generation unit 92 sets the designated concept expression as a category definition, A document category having a document having field data including the conceptual expression as a member is generated. The document category generation unit 92 extracts document data including a designated conceptual expression based on the document data structure stored in the document data structure storage unit 10, and sets the document data as a document belonging to the document category.

  The composite category is a category that defines a logical expression of an existing category. The user can create a composite category by performing a predetermined operation on a predetermined screen displayed on the display 60 by the concept expression display unit 44. For example, a composite category can be generated by the user performing an operation on the category list 62 and the workspace 63 shown in FIG. 24 and specifying a combination of existing categories. The combination of categories can be expressed by a logical operation expression of category IDs of existing categories. In the present embodiment, it is assumed that “composite AND category”, “composite OR category”, and “composite NOT category” can be generated as composite categories.

  The composite AND category is a category that takes an AND of designated category IDs. For example, when “category 1”, “category 3”, and “category 5” are designated, the category definition of the composite AND category is “1AND3AND5”, and the belonging text is a product set of the belonging text sets of the three categories. .

The compound OR category is a category that takes OR of designated category IDs. For example,
When “Category 1”, “Category 3”, and “Category 5” are designated, the category definition of the composite OR category is “1OR3OR5”, and the belonging text is a union of the belonging text sets of the three categories.

  The composite NOT category is a category that takes a NOT of a specified category ID. For example, when “category 1”, “category 3”, and “category 5” are specified, the category definition of the composite NOT category is “NOT (1OR3OR5)”, and the belonging text is the sum of the belonging text sets of the three categories. Denial of set.

  Furthermore, the data ID designation category is a category that defines a document ID list. In this embodiment, a user cannot create arbitrarily. However, the data ID designation category may be generated by the user designating the document ID. However, in this case, the reuse of the category definition is effective only within the same target document set, and it is necessary to prevent reuse and re-application to other contrast document sets. In order to make all the definitions reusable, it is better that the user cannot create the data ID designation category.

  In FIG. 29, when the document category display unit 94 displays the document category generated by the document category generation unit 92, for example, a list display such as the category list 62 shown in FIG. Graphical display.

  The document data display unit 95 displays, for example, the data browser 64 shown in FIG. 24 on a predetermined screen of the display 60 when displaying the document data including the concept expression designated by the user. Here, the concept expression extraction unit 440 of the language processing unit 42 extracts document data including the designated concept expression based on the document data structure stored in the document data structure storage unit 10, and the document data display unit 95 displays the document data extracted by the concept expression extraction unit 440 on the data browser 64 on the display 60 screen.

  Further, the document data display unit 95 can display on the screen of the display 60 documents belonging to the category designated by the user on the screen displayed on the display 60 by the document category display unit 94. Here, based on the category information (shown in FIG. 30) stored in the document category storage unit 80, the document data generation unit 93 selects document IDs belonging to the category (category definition) specified from the belonging document ID list. Based on the detected document data structure stored in the document data structure storage unit 10, the document data corresponding to the detected document ID is extracted. The document data display unit 95 displays the document data extracted by the document data generation unit 93 on the display 60 screen.

  A dictionary registration procedure for the document processing apparatus 1 configured as described above will be described with reference to FIG. Here, a case where a compound OR document category among the document categories is registered in the concept expression thesaurus dictionary is shown.

  In the document processing unit 40, the dictionary registration unit 91 inputs a document category to be registered designated through the designation receiving unit 50 (S401).

  Next, the dictionary registration unit 91 determines whether the designated document category is a compound OR category (S402), and if it is not a compound OR category, the process ends. If the designated document category is a compound OR category, the dictionary registration unit 91 inputs the registration type designated via the designation receiving unit 50 (S403).

  Next, the dictionary registration unit 91 determines whether to register the document category as “related expression” or “synonymous expression” based on the designated registration type (S404). Here, when registering as “synonymous expression”, the process proceeds to S405, and when registering as “related expression”, the process proceeds to S410.

  In S405 (synonymous expression registration), the dictionary registration unit 91 acquires concept expressions that are category definitions for all basic concept expression categories among categories that are constituent members of the target document category, and stores them in the concept expression thesaurus dictionary. The synonymous expression group including the acquired concept expression is searched.

  Next, the dictionary registration unit 91 inputs information on the synonym expression group to be registered, which is designated via the designation receiving unit 50 (S406). Furthermore, based on the designation of the synonym expression group to be registered, the dictionary registration unit 91 registers the document category information to be registered as a new synonym expression group or searched by a synonym expression group search. It is determined whether to register in any one of the existing synonymous expression groups (S407). Here, when registering as a new synonym expression group, it progresses to a new synonym expression group generation process (S408), and when registering as an existing synonym expression group, it progresses to a synonym expression registration process (S409).

  In the new synonym expression group generation process (S408), the dictionary registration unit 91 creates a new synonym expression group in the concept expression thesaurus dictionary. At this time, the category label may be input as the category label, or may be designated by the user. Nothing is registered in the belonging concept expression ID list.

  In the synonym expression registration process (S409), the dictionary registration unit 91 acquires and acquires concept expressions that are category definitions for all basic concept expression categories among categories that are constituent members of the document category to be registered. Whether the registered concept expression is registered in the concept expression list of the concept expression thesaurus dictionary, and if it is registered, the concept expression ID is acquired, and a unique ID is assigned to the unregistered concept expression, and a new concept is obtained. It registers in the concept expression list of the expression thesaurus dictionary, and acquires the registered concept expression ID.

  Further, in the synonym expression registration process (S409), the dictionary registration unit 91 adds the acquired concept expression ID to the belonging concept expression ID list of the synonym expression group to be registered. Here, if there is an ID overlap in the belonging concept expression ID list, it is eliminated.

  When it is determined that the document category is registered as “related expression” (S404), in the related expression registration process (S410), the dictionary registration unit 91 selects all of the categories that are constituent members of the document category to be registered. The concept expression which is a category definition is acquired for the basic concept expression category of, and it is checked whether the acquired concept expression is registered in the concept expression list of the concept expression thesaurus dictionary, and if it is registered, the concept expression ID is acquired. If not registered, a unique ID is assigned to the concept expression and registered in the concept expression list of the concept expression thesaurus dictionary, and the registered concept expression ID is acquired.

  Further, in the related expression registration process (S410), the dictionary registration unit 91 acquires all the acquired concept expression IDs in the related concept expression ID list of the corresponding concept expression ID in the related expression list of the concept expression thesaurus dictionary. All the concept expression IDs except for the self are added. Here, if there is an ID overlap in the related concept expression ID list, it is eliminated.

  According to the document processing apparatus 1 according to the third embodiment of the present invention, the designation receiving unit 50 and the concept for designating the concept expression included in the set of document data acquired by the document data acquiring unit 41. A document that forms an expression display unit 44 (corresponding to concept expression designating means) and a concept expression designated here as a category definition, generates a document group based on the category definition from a set of document data, and sets it as a document category A category generation unit 92 (corresponding to a document category generation unit), a document category display unit 94 for displaying the document category generated by the document category generation unit 92, and a display 60 (corresponding to a document category display unit). It has a configuration. This configuration corresponds to an embodiment of the present invention according to claims 2 and 8.

With this configuration, the user designates a conceptual expression as a desired category definition from the screen of the display 60 via the designation receiving unit 50, and confirms a document group generated based on the category definition on the screen. Therefore, user convenience is improved.
Further, according to the present embodiment, a dictionary registration unit 91 (corresponding to a dictionary registration unit) that registers a new concept expression in the concept expression thesaurus dictionary based on the document category generated by the document category generation unit 92 is provided. It has a configuration. This configuration corresponds to an embodiment of the present invention according to claims 6 and 12.

  With this configuration, the document category information generated by the document category generation unit 92 can be registered in the concept expression thesaurus dictionary, so that the accuracy of text data analysis and document search using the concept expression thesaurus dictionary is improved.

  Although not shown in the above-described embodiment, the document processing apparatus 1 of the present invention is a personal computer (for example, a storage unit such as a CPU, RAM, ROM, and HDD, a communication unit, a display unit such as a liquid crystal display, a keyboard, The same effect can be obtained even if the personal computer is made to function as the above-described concept expression specifying means, document category generation means, document category display means, and dictionary registration means. It is. Further, the storage unit stores a program for causing the personal computer to execute the processing procedure shown in FIG. 31, the processing steps performed by the above-described concept expression specifying unit, document category generation unit, document category display unit, and dictionary registration unit. You may do it.

It is a figure showing the dependency analysis result which concerns on the 1st Embodiment of this invention. It is a figure showing the concept expression which concerns on the 1st Embodiment of this invention. It is a figure explaining the structure of the concept expression which concerns on the 1st Embodiment of this invention. It is a figure showing the concept expression expansion search which concerns on the 1st Embodiment of this invention. 1 is a block diagram of a document processing apparatus according to a first embodiment of the present invention. It is a figure explaining the structure of the basic unit of the concept expression which concerns on the 1st Embodiment of this invention. It is a figure showing the concept expression list which concerns on the 1st Embodiment of this invention. It is a figure showing the synonymous expression group list which concerns on the 1st Embodiment of this invention. It is a figure showing the related expression list which concerns on the 1st Embodiment of this invention. It is a figure showing the synonymous expression and related expression which concern on the 1st Embodiment of this invention. It is a block diagram showing the detail of the language processing part which concerns on the 1st Embodiment of this invention. It is a figure showing the document data structure concerning the 1st Embodiment of this invention. It is a figure showing the information which the component of the document data structure which concerns on the 1st Embodiment of this invention manages. It is a figure showing the word list which concerns on the 1st Embodiment of this invention. It is a figure showing the intention tag list which concerns on the 1st Embodiment of this invention. It is a figure explaining the extended concept expression search (basic unit addition) which concerns on the 1st Embodiment of this invention. It is a figure explaining the extended concept expression search (intention tag addition) which concerns on the 1st Embodiment of this invention. It is a figure showing the extended concept expression search which concerns on the 1st Embodiment of this invention. It is a block diagram showing the detail of the concept expression extraction part which concerns on the 1st Embodiment of this invention. It is a figure showing the concept expression of the synonym expression group which concerns on the 1st Embodiment of this invention. It is a flowchart of the synonym expression search process which concerns on the 1st Embodiment of this invention. It is a flowchart of the related expression search process which concerns on the 1st Embodiment of this invention. It is a flowchart of the production | generation process of a synonym expression group conceptual expression which concerns on the 1st Embodiment of this invention. It is a figure showing the display screen which concerns on the 1st Embodiment of this invention. It is a figure showing the display screen of the conceptual expression (only label) which concerns on the 1st Embodiment of this invention. It is a figure showing the display screen of the concept expression (a label and a affiliation concept expression) which concerns on the 1st Embodiment of this invention. It is a block diagram of the document processing apparatus which concerns on the 2nd Embodiment of this invention. It is a figure showing the dictionary search dialog which concerns on the 2nd Embodiment of this invention. It is a block diagram of the document processing apparatus which concerns on the 3rd Embodiment of this invention. It is a figure showing the category information which concerns on the 3rd Embodiment of this invention. It is a flowchart of the dictionary registration process which concerns on the 3rd Embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Document processing apparatus 10 Document data structure storage part 20 Conceptual expression storage part 30 Conceptual expression thesaurus dictionary storage part 40 Document processing part 41 Document data acquisition part 42 Language processing part 43 Dictionary search part 44 Conceptual expression display part 50 Specification reception part 60 Display

Claims (12)

A concept included in a document is expressed by a combination of a token and an intention expression extracted from clause information included in the document, and a relationship between a plurality of concept expressions is expressed based on relationship information between clauses. A document processing apparatus for processing document data based on a concept expression by a combination of the above and a relationship between the concept expressions,
Document data acquisition means for acquiring a set of document data to be processed;
Language processing means for performing language analysis processing on the document data acquired by the document data acquisition means, and extracting a concept expression included in the document data;
Concept expression display means for displaying the concept expression extracted by the language processing means;
A concept expression thesaurus dictionary storing means for storing a concept expression thesaurus dictionary for registering a concept expression expressing the concept included in the document;
Dictionary search means for searching for a concept expression from the concept expression thesaurus dictionary;
A document processing apparatus comprising: The document processing apparatus according to claim 1,
A concept expression specifying means for specifying a concept expression included in a set of document data acquired by the document data acquiring means;
A document category generation unit that generates a document group based on the category definition from a set of document data, using the concept representation designated by the concept representation designation unit as a category definition;
Document category display means for displaying the document category generated by the document category generation means;
A document processing apparatus comprising: The document processing apparatus according to claim 1 or 2,
A concept expression extracting means for extracting one concept expression based on synonymous concept expression groups registered in the concept expression thesaurus dictionary;
Concept expression display means for displaying the synonymous expression group concept expression extracted by the concept expression extraction means;
A document processing apparatus comprising: The document processing apparatus according to any one of claims 1 to 3,
Synonymous related concepts for specifying conceptual expressions included in a set of document data in order to search synonymous conceptual expressions and conceptual expressions that are not synonymous but semantically related, registered in the conceptual expression thesaurus dictionary An expression designating means; and a synonym related concept expression display means for displaying a search result of the synonymous concept expression and the related concept expression,
The dictionary search means searches for the synonymous concept expression and the related concept expression registered in the concept expression thesaurus dictionary based on the designation of the synonym related concept expression designation means, and displays the synonym related concept expression display The document processing apparatus is characterized in that the means displays the search result of the dictionary search means. The document processing apparatus according to any one of claims 1 to 4, wherein:
In order to search for an extended concept expression obtained by extending the concept expression registered in the concept expression thesaurus dictionary, an extended concept expression specifying means for specifying a concept expression included in a set of document data, and the extended concept expression specification An extension for searching for the specified syntactic concept registered in the concept expression thesaurus dictionary based on the designation of the means, collecting it into a synonymous concept expression, and searching for the synonymous concept expression from the set of document data A concept expression search means,
The document processing apparatus, wherein the concept expression display means displays a search result of the extended concept expression search means. The document processing apparatus according to any one of claims 2 to 5,
A document search apparatus comprising: a dictionary registration unit that registers a new concept expression in the concept expression thesaurus dictionary based on the document category generated by the document category generation unit. A concept included in a document is expressed by a combination of a token and an intention expression extracted from clause information included in the document, and a relationship between a plurality of concept expressions is expressed based on relationship information between clauses. A document processing method for processing document data using a concept expression thesaurus that registers a concept expression by a combination of and a relationship between concept expressions,
A document data acquisition procedure for acquiring a set of document data to be processed;
A language processing procedure for performing a language analysis process on the document data acquired by the document data acquisition procedure and extracting a concept expression included in the document data;
A concept expression display procedure for displaying the concept expression extracted by the language processing procedure;
A dictionary search procedure for searching a concept expression from the concept expression thesaurus dictionary;
The document processing method characterized by performing. The document processing method according to claim 7.
A concept expression specifying procedure for specifying a concept expression included in a set of document data acquired by the document data acquiring procedure;
A document category generation procedure in which the concept expression specified by the concept expression specifying means is a category definition, a document group is generated based on the category definition from a set of document data,
A document category display procedure for displaying the document category generated by the document category generation means;
The document processing method characterized by performing. In the document processing method of Claim 7 or Claim 8,
A concept expression extraction procedure for extracting one concept expression based on synonymous concept expression groups registered in the concept expression thesaurus dictionary;
A concept expression display procedure for displaying the synonym expression group concept expression extracted by the concept expression extraction procedure;
The document processing method characterized by performing. The document processing method according to any one of claims 7 to 9,
Synonymous related concepts for specifying conceptual expressions included in a set of document data in order to search synonymous conceptual expressions and conceptual expressions that are not synonymous but semantically related, registered in the conceptual expression thesaurus dictionary An expression designation procedure, and a synonym related concept expression designation procedure for displaying a search result of the synonymous concept expression and the related concept expression,
In the dictionary search procedure, the synonym related concept representation and the related concept representation registered in the concept representation thesaurus dictionary are searched based on the designation of the synonym related concept representation designation procedure, and the synonym related concept representation display In the procedure, a search result of the dictionary search procedure is displayed. The document processing method according to any one of claims 7 to 10,
An extended concept expression designation procedure for designating a concept expression included in a set of document data in order to search for an extended concept expression obtained by extending the concept expression registered in the concept expression thesaurus dictionary, and the extended concept expression designation An extension that searches the specified syntactic concept registered in the conceptual expression thesaurus dictionary based on the procedure specification, collects it into one synonymous conceptual expression, and searches the synonymous conceptual expression from the set of document data A concept expression search procedure, and
In the concept expression display procedure, a search result of the extended concept expression search procedure is displayed. The document processing method according to any one of claims 7 to 11,
A document processing method for executing a dictionary registration procedure for registering a new concept expression in the concept expression thesaurus dictionary based on the document category generated by the document category generation procedure.

Images