JP2004287696A - Retrieval system and retrieval program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a retrieval system capable of efficiently retrieving a desired document. <P>SOLUTION: This system comprises a document database for recording document data that are an object of retrieval; a retrieval database for recording retrieval data; and a morphological analysis dictionary preliminarily recording morphological analysis data, in which a word is extracted by morphologically analyzing the document data recorded in the document database. This system further comprises a morphological analysis means for specifying, with respect to the word, an affirmative form or negative form; a document registration means for recording the extracted word, information for specifying the affirmation or negation, and information for specifying the document in association with each other in the retrieval database; an input means for inputting a question sentence; a retrieval condition forming means for morphologically analyzing the question sentence and forming a retrieval condition based on the resulting word and information for specifying the affirmation or negation; a retrieval means for retrieving a document matched to the retrieval condition from the retrieval database; and a display means for displaying the retrieval result by the retrieval means. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a search system and a search program capable of efficiently searching for desired document data.
[0002]
[Prior art]
When searching for documents, etc., usually a text-based sentence is divided for each word by a method called morphological analysis and registered in the database in advance, and at the time of search, a question sentence is morphologically analyzed and divided for each word A system that obtains accurate results in a short time by performing a search has become mainstream. This morphological analysis technology is indispensable for document retrieval. In addition, there are various approaches to a search method of a search engine in search of search accuracy. Normally, morphological analysis is performed to extract noun phrases, verbs, etc., and the result is displayed depending on the presence or absence of the word. In recent trends, syntax analysis is performed to analyze even the meaning.
As a system for performing syntax analysis, a morphological analysis system that uses syntax rules to perform syntax analysis and considers the structure of a sentence is known (Patent Document 1). There is also known a system that improves the quality of morphological analysis by registering similar words and words on a text basis (Patent Document 2).
[0003]
[Patent Document 1]
Japanese Patent Application Laid-Open No. 08-050588 [Patent Document 2]
JP-A-11-039347
[Problems to be solved by the invention]
However, in the systems disclosed in Patent Literatures 1 and 2, words and words can be matched as similar words, but words having the same meaning (for example, garbled characters) and sentences (for example, characters cannot be displayed). There is a problem that can not be matched with. Further, if the sentence “characters cannot be displayed” is registered in the dictionary for morphological analysis, documents containing the words “characters” and “display” cannot be searched, so that there is a problem that search accuracy is reduced.
[0005]
The present invention has been made in view of such circumstances, and has as its object to provide a search system and a search program capable of efficiently searching for a desired document.
[0006]
[Means for Solving the Problems]
According to the first aspect of the present invention, there is provided a document database for recording document data to be searched, a search database for recording search data, a morphological analysis dictionary in which morphological analysis data is recorded in advance, With reference to the analysis dictionary, a word is extracted by performing a morphological analysis on the document data recorded in the document database, and a morphological analysis unit for specifying a positive form or a negative form for the word, and the morphological analysis unit The extracted words, information for specifying affirmation or denial, information for specifying a document, and document registration means for recording the information in the search database in association with each other, input means for inputting a question sentence, and the morphological analysis means A search that morphologically analyzes a question sentence and creates search conditions based on the obtained words and information that identifies affirmation or negation A matter creating means, search means for a document matching the search condition to search from the search database, and further comprising a display means for displaying the search result in the search unit.
[0007]
According to a second aspect of the present invention, there is provided a document database for recording document data to be searched, a search database for recording data for search, a morphological analysis dictionary in which data for morphological analysis is recorded in advance, With reference to the analysis dictionary, a word is extracted by performing a morphological analysis on the document data recorded in the document database, and a morphological analysis unit for specifying a positive form or a negative form for the word, and the morphological analysis unit Document registration means for associating the extracted words, information specifying affirmation or negation, and information specifying a document and recording them in the search database, a similar word dictionary in which similar words are recorded in advance for each word, The question sentence is morphologically analyzed by the input means for inputting the question sentence and the morphological analysis means, and the obtained word is identified as affirmative or negative. First search condition creation means for creating a search condition based on information to be searched, and similar word acquisition means for obtaining a similar word of a word obtained as a result of morphological analysis of the question sentence with reference to the similar word dictionary A second search condition creating means for creating a search condition based on the similar words obtained by the similar word acquisition means; and a document which matches each of the first search condition and the second search condition. A search unit for searching the search database; and a display unit for displaying the document data searched by the search unit.
[0008]
According to a third aspect of the present invention, there is provided a document database for recording document data to be searched, a search database for recording data for search, a morphological analysis dictionary in which data for morphological analysis is recorded in advance, With reference to the analysis dictionary, a word is extracted by performing a morphological analysis on the document data recorded in the document database, and a morphological analysis unit for specifying a positive form or a negative form for the word, and the morphological analysis unit Document registration means for associating the extracted words, information specifying affirmation or negation, and information specifying a document and recording them in the search database, a similar word dictionary in which similar words are recorded in advance for each word, The question sentence is morphologically analyzed by the input means for inputting the question sentence and the morphological analysis means, and the obtained word is identified as affirmative or negative. First search condition creation means for creating a search condition based on information to be searched, and similar word acquisition means for obtaining a similar word of a word obtained as a result of morphological analysis of the question sentence with reference to the similar word dictionary Morphological analysis of the similar words by the morphological analysis means, and second search condition creation means for creating a search condition based on the obtained word and information specifying affirmative or negative, and the first search A search unit for searching the search database for a document that satisfies each of the condition and the second search condition; and a display unit for displaying the document data searched by the search unit.
[0009]
According to a fourth aspect of the present invention, in the search system, the search system calculates the importance of the word obtained by the morphological analysis unit, and associates the importance of the word with each word and records the importance in the search database. Means, and a counting means for sorting the search results in order of importance, wherein the display means displays the search results sorted in order of importance by the counting means.
[0010]
The invention according to claim 5 includes a document database for recording document data to be searched, a search database for recording data for search, and a morphological analysis dictionary in which data for morphological analysis is recorded in advance. A search program operating in a search system, wherein a word is extracted by morphologically analyzing document data recorded in the document database with reference to the morphological analysis dictionary, and a positive or negative form of the word is extracted. A document extraction process for associating a word extracted by the morphological analysis process, information for specifying affirmative or negative, and information for specifying a document and recording the information in the search database; And a morphological analysis of the question sentence by the morphological analysis processing. Or a search condition creating process for creating a search condition based on the information specifying the negation, a search process for searching the search database for a document that meets the search condition, and a search result in the search process is displayed. The display processing is performed by a computer.
[0011]
According to a sixth aspect of the present invention, there is provided a document database for recording document data to be searched, a search database for recording search data, a morphological analysis dictionary in which morphological analysis data is recorded in advance, Is a search program that operates in a search system including a similar word dictionary in which similar words are recorded in advance, and by performing morphological analysis on document data recorded in the document database with reference to the morphological analysis dictionary. A word is extracted and, for the word, a morphological analysis process for specifying a positive or negative form, and the word extracted by the morphological analysis process, information for specifying positive or negative, and information for specifying a document are associated with each other. A document registration process for recording in the search database, an input process for inputting a question sentence, and the morphological analysis process. A first search condition creating process for creating a search condition based on the morphological analysis of the written question sentence and the obtained word and information specifying affirmative or negative, and referring to the similar word dictionary, A synonym acquisition process for obtaining a synonym of the word obtained as a result of morphological analysis of the word, a second search condition creation process for creating a search condition based on the synonym obtained by the similar word acquisition process, And causing the computer to perform a search process of searching the search database for a document that meets each of the first search condition and the second search condition, and a display process of displaying the document data searched by the search process. Features.
[0012]
According to a seventh aspect of the present invention, there is provided a document database for recording document data to be searched, a search database for recording search data, a morphological analysis dictionary in which morphological analysis data is recorded in advance, Is a search program that operates in a search system including a similar word dictionary in which similar words are recorded in advance, and by performing morphological analysis on document data recorded in the document database with reference to the morphological analysis dictionary. A word is extracted and, for the word, a morphological analysis process for specifying a positive or negative form, and the word extracted by the morphological analysis process, information for specifying positive or negative, and information for specifying a document are associated with each other. A document registration process for recording in the search database, an input process for inputting a question sentence, and the morphological analysis process. A first search condition creating process for creating a search condition based on the morphological analysis of the written question sentence and the obtained word and information specifying affirmative or negative, and referring to the similar word dictionary, A morphological analysis of the similar word by the morphological analysis process, and a search based on the obtained word and information identifying affirmative or negative. A second search condition creation process for creating a condition, a search process for searching the search database for a document that meets each of the first search condition and the second search condition, and a document searched for by the search process A display process for displaying data is performed by a computer.
[0013]
In the invention according to claim 8, the search program calculates importance of words obtained by morphological analysis processing, and associates the importance of words with each word and records the importance in the search database. The computer is further made to perform a process and a tallying process of sorting the search results in order of importance, and the display process displays the search results sorted in order of importance by the tallying process.
[0014]
According to the present invention, the affirmative / negative flag is registered in the search database at the time of morphological analysis at the time of document registration, so that it is possible to prevent a document having a completely opposite meaning from being found. In particular, at the time of trouble handling processing, document data having a completely opposite meaning is unnecessary, so that desired document data can be efficiently found. Further, since a document containing a similar word of a word obtained as a result of morphological analysis of a question sentence is searched, it is easy to find desired document data. When displaying the search result, the search result using the morphological analysis result and the search result using the similar word are stored in the search result storage unit, and are displayed after the importance is rearranged. As a result, compared to the case where the search using the morphological analysis result and the search using similar words are executed separately, the effect that the document with high importance is more likely to be displayed at the top is obtained. Can be
[0015]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, a search system according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing the configuration of the embodiment. In this figure, reference numeral 1 denotes a document database in which a plurality of document data to be searched are recorded. Reference numeral 2 denotes a document registration unit that registers search data in the search database 3 in order to search for document data recorded in the document database 1 at high speed. Reference numeral 4 denotes a morphological analysis dictionary in which dictionary data for morphologically analyzing a sentence is recorded. Reference numeral 5 denotes a morphological analysis unit that performs morphological analysis of document data recorded in the document database 1 with reference to the morphological analysis dictionary. Reference numeral 6 denotes a weighting dictionary in which dictionary data for weighting important words included in the document data is recorded. Reference numeral 7 denotes an importance calculating unit that refers to the weight dictionary 6 and weights words included in the document data recorded in the document database 1. Reference numeral 8 denotes an input unit for inputting a question sentence for finding desired document data, and is constituted by a keyboard or the like. Reference numeral 9 denotes a search condition creation unit that creates search conditions for efficiently searching for an answer to the question sentence input from the input unit 8. Reference numeral 10 denotes a similar word dictionary in which similar words are recorded in advance for each word. Reference numeral 11 denotes a search unit that performs a search according to the search condition created by the search condition creation unit 9. Reference numeral 12 denotes a search result storage unit that temporarily stores the search results obtained by the search unit 11. Reference numeral 13 denotes a counting unit that counts search results stored in the search result storage unit 12. Reference numeral 14 denotes a display unit that displays a search result, and is configured by a display or the like.
[0016]
Here, a business using the search system shown in FIG. 1 will be described. The document database 1 records documents that explain how to deal with troubles such as computers and computer peripherals, and how to deal with questions (phenomena when troubles occur in devices) input from the input unit 1. Is used to efficiently find out from the document data recorded in the document database 1. In the following description, in order to provide user support for a device, a user support representative obtains a response to a trouble received by telephone, fax, or e-mail using the search system shown in FIG. Will be described as an example.
[0017]
First, an operation of creating a search database will be described with reference to FIGS. First, the document registration unit 2 reads the document data recorded in the document database 1 and passes it to the morphological analysis unit 5. In response, the morphological analysis unit 5 performs a morphological analysis on the document data with reference to the morphological analysis dictionary 4, and outputs frequently appearing independence part of speech (noun, verb, adjective, etc.) and affirmative / negative flag to the document registration unit 2. return. At this time, the morphological analysis unit 5 counts the frequency of words appearing in the document data, and returns it to the document registration unit 2. In response, the document registration unit 2 registers the analysis result in the search database 3.
[0018]
Normally, when a sentence is morphologically analyzed, an independent word and an adjunct word can be acquired, and each inflected form can be acquired. Here, at the time of registration in the database 3 for search, it is classified into any one of a verb, an adjective, an adjective verb, an inflected noun, and a noun including an idiom. In addition, the inflected part of speech (part of speech) is not registered in principle. However, if there is "cancellation", the "affirmative / negative flag" is set to "1".
[0019]
Next, the document registration unit 2 instructs the importance calculation unit 7 to calculate the importance. In response to this, the importance calculation unit 7 refers to the weighting dictionary 6 to weight the words that appear in the document, calculates the importance, and registers it in the search database 3. The importance calculation used here is performed using the well-known tfidf method, and the weighting is performed by extracting a word to be weighted in advance and registering the weighting value in the weighting dictionary 6.
[0020]
The table structure of the search database 3 created by this operation will be described with reference to FIG. The search database 3 includes three types of tables: a page table, a keyword table, and a word table. The page table includes a “page ID” which is a unique number for each document data, a “full address” of a directory where the document data is stored or an “address” which is a URL, and a “title” which is a title of the document of the document data. , "Text" which is the text of the document data, and "classification ID" for narrowing down when used in the search.
[0021]
The keyword table is linked to the page table by “page ID” and linked to the word table by “word ID”. “Cost” is a value indicating how many times this keyword has appeared in this document data. “Weighting value” is a value given by the weighting dictionary 6. “Importance” is a value obtained by the tfidf method.
[0022]
The word table includes a “word” obtained by morphological analysis, a “word ID” assigned to each word, a “frequency of use” indicating the number of documents in which this word is used in all document data, It consists of the obtained “part of speech” and the “negative affirmation flag”.
[0023]
Next, an operation of obtaining an answer to the input question will be described with reference to FIG. First, when the support person inputs a question sentence from the input unit 8, the input unit 8 reads the input question sentence (step S1). Here, it is assumed that the user has input as a question sentence that "garbled characters have started to appear". The input unit 8 transfers the input question sentence to the search condition creating unit 9.
[0024]
Next, the search condition creating unit 9 passes the received question sentence to the morphological analysis unit 5. In response, the morphological analysis unit 5 refers to the morphological analysis dictionary 4 and morphologically analyzes the received question sentence (step S2). As a result of this analysis, “noun: garbled”, “adjective verb: selfish (positive form)”, and “general verb: awake (positive form)” are obtained as the analysis results. Subsequently, the morphological analysis unit 5 returns the analysis result obtained here to the search condition creation unit 9.
[0025]
Next, the search condition creating unit 9 creates a search condition based on the morphological analysis result (Step S3). Here, the search condition is “noun: garbled & adjective verb: selfish (negative affirmation flag = 0) & general verb: awake (negative affirmation flag = 0)”. The search condition creating unit 9 transfers the created search condition to the search unit 11.
[0026]
Next, the search unit 11 searches the search database 3 using the search condition received from the search condition creation unit 9 (step S4). As a result of this search, the search unit 11 obtains the page ID of the document data including the garbled noun, the adjective verb (positive form), and the occurrence of the general verb (positive form). If there is a plurality of corresponding document data, a plurality of page IDs can be obtained in the search unit 11. Subsequently, the search unit 11 stores the obtained search result in the search result storage unit 12 (Step S5). Then, the search unit 11 notifies the search condition creation unit 9 that the search has been completed.
[0027]
Next, the search condition creation unit 9 refers to the similar word dictionary 10 and obtains words obtained as a result of morphological analysis of the question sentence (here, “noun: garbled”, “adjective verb: selfish (positive form)” , "General verb: awake (positive form)") (step S6). In the similar word dictionary 10, as shown in FIG. 4, similar words are recorded in association with each other. In this example, “characters cannot be displayed” is recorded in advance for “garbled characters”. The search condition creating unit 9 searches for the similar words obtained here in “text” recorded in the page table of the search database 3 (step S7). As a result of this search, the search unit 11 obtains a page ID of the document data including the corresponding similar word (here, “characters cannot be displayed”). If there is a plurality of corresponding document data, a plurality of page IDs can be obtained in the search unit 11. Subsequently, the search unit 11 stores the obtained search result in the search result storage unit 12 (Step S8). Then, the search unit 11 notifies the counting unit 13 that the search has been completed.
[0028]
Next, the tallying unit 13 refers to the search database 3 and assigns the page ID to the important document based on the importance and weighting value in the keyword table having the page ID recorded in the search result storage unit 12. The search results are totalized by rearranging them in the order of data (step S9). Then, the counting unit 13 displays the sorted document data on the display unit 14 (Step S10).
[0029]
By this operation, the document data having a high possibility of containing an answer to the question “garbled characters are arbitrarily generated” are sequentially displayed on the display unit 14. The support person looks at the document data and sends an answer to the user who sent the question.
[0030]
Next, an operation of further performing morphological analysis on the obtained similar words will be described with reference to FIG. In this case, in each step shown in FIG. 3, the search operation in step S7 is replaced with steps S71 and S72.
The search condition creation unit 9 transfers the sentence “characters cannot be displayed” obtained by referring to the similar word dictionary 10 to the morphological analysis unit 5. In response, the morphological analysis unit 5 performs a morphological analysis (step S71). Then, the morphological analysis unit 5 returns the morphological analysis result to the search condition creating unit 9. Here, “character” and “display (negation)” are obtained as the morphological analysis results. Then, the search condition creating unit 9 creates a search condition based on the analysis result, and transfers the created search condition to the search unit 11. Then, the search unit 11 searches the search database 3 using the search condition received from the search condition creation unit 9 (Step S72).
[0031]
As described above, by performing morphological analysis on similar words, it is possible to use the search database 3 without changing it, so that desired document data can be efficiently searched.
[0032]
The search condition creating unit 9 may add a search condition for shortening the search time by referring to a stop word dictionary for preventing a predetermined word from being searched. This makes it possible to register words that do not need to be checked intentionally, thereby preventing the processing time for words appearing in many documents from being lengthened.
[0033]
As described above, since the affirmative / negative flag is registered in the search database 3 during the morphological analysis at the time of document registration, it is possible to prevent a document having a completely opposite meaning from being found. In particular, at the time of troubleshooting, document data having a completely opposite meaning is not necessary, so that desired document data can be efficiently found. Further, since a document containing a similar word of a word obtained as a result of morphological analysis of a question sentence is searched, it is easy to find desired document data. When displaying the search result, the search result using the morphological analysis result and the search result using the similar word are stored in the search result storage unit 12, and the display is performed after the importance is rearranged. As a result, the possibility that a document with higher importance is displayed at a higher position than in a case where a search using a morphological analysis result and a search using a similar word are executed separately is increased.
[0034]
Note that the input unit 8 and the display unit 14 illustrated in FIG. 1 may be provided in a terminal device via the Internet or the like. In this way, if the user who purchases the computer equipment inputs a question using a personal computer at home and displays the answer on the user's personal computer, a user support system using the Internet is constructed. It is possible.
[0035]
Alternatively, a question sentence input from the input unit 8 is totaled, a word having a high frequency of appearance is selected from the question sentence, and the selected word is used in the “word” field of the similar word dictionary 10 shown in FIG. Good. Then, by extracting similar words for these words from a term dictionary or the like and registering them in the similar word dictionary 10, the similar word dictionary 10 can be easily created.
[0036]
Further, a program for realizing the function of the processing unit in FIG. 1 is recorded on a computer-readable recording medium, and the program recorded on this recording medium is read into a computer system and executed to execute the document search process. May go. Here, the “computer system” includes an OS and hardware such as peripheral devices. The “computer system” also includes a WWW system provided with a homepage providing environment (or display environment). The “computer-readable recording medium” refers to a portable medium such as a flexible disk, a magneto-optical disk, a ROM, and a CD-ROM, and a storage device such as a hard disk built in a computer system. Further, the “computer-readable recording medium” refers to a volatile memory (RAM) inside a computer system serving as a server or a client when a program is transmitted through a network such as the Internet or a communication line such as a telephone line. In addition, programs that hold programs for a certain period of time are also included.
[0037]
Further, the above program may be transmitted from a computer system storing the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in the transmission medium. Here, the "transmission medium" for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line. Further, the program may be for realizing a part of the functions described above. Furthermore, what can implement | achieve the function mentioned above in combination with the program already recorded on the computer system, and what is called a difference file (difference program) may be sufficient.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of an embodiment of the present invention.
FIG. 2 is an explanatory diagram showing a table structure of a search database 3 shown in FIG.
FIG. 3 is a flowchart showing a search operation of the search system shown in FIG. 1;
4 is an explanatory diagram showing a table structure of a similar word dictionary 10 shown in FIG.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Document database, 2 ... Document registration part, 3 ... Search database, 4 ... Morphological analysis dictionary, 5 ... Morphological analysis part, 6 ... Weighting dictionary, 7 ... Importance calculation part, 8 ... Input part, 9 ... Search condition Creation unit, 10 ... Similar word dictionary, 11 ... Search unit, 12 ... Search result storage unit, 13 ... Total unit, 14 ... Display unit

Claims (8)

A document database for recording document data to be searched;
A search database that records search data;
A morphological analysis dictionary in which data for morphological analysis is recorded in advance,
With reference to the morphological analysis dictionary, while extracting words by morphologically analyzing the document data recorded in the document database, for the words, a morphological analysis unit that specifies a positive or negative form,
Document registration means for associating the word extracted by the morphological analysis means, information specifying affirmation or negation, and information specifying a document and recording the information in the search database,
Input means for inputting a question sentence,
A morphological analysis of the question sentence by the morphological analysis means, based on the obtained word and information specifying affirmation or negation, a search condition creating means for creating a search condition,
Search means for searching the search database for documents matching the search condition;
Display means for displaying a search result in the search means. A document database for recording document data to be searched;
A search database that records search data;
A morphological analysis dictionary in which data for morphological analysis is recorded in advance,
With reference to the morphological analysis dictionary, while extracting words by morphologically analyzing the document data recorded in the document database, for the words, a morphological analysis unit that specifies a positive or negative form,
Document registration means for associating the word extracted by the morphological analysis means, information specifying affirmation or negation, and information specifying a document and recording the information in the search database,
A similar word dictionary in which similar words are recorded in advance for each word,
Input means for inputting a question sentence,
Morphological analysis of the question sentence by the morphological analysis means, a first search condition creating means for creating a search condition based on the obtained word and information specifying affirmative or negative,
Referring to the similar word dictionary, a similar word acquiring means for obtaining a similar word of a word obtained as a result of morphological analysis of the question sentence,
A second search condition creation unit that creates a search condition based on the similar words obtained by the similar word acquisition unit;
Search means for searching the search database for documents matching each of the first search condition and the second search condition;
Display means for displaying the document data searched by the search means. A document database for recording document data to be searched;
A search database that records search data;
A morphological analysis dictionary in which data for morphological analysis is recorded in advance,
With reference to the morphological analysis dictionary, while extracting words by morphologically analyzing the document data recorded in the document database, for the words, a morphological analysis unit that specifies a positive or negative form,
Document registration means for associating the word extracted by the morphological analysis means, information specifying affirmation or negation, and information specifying a document and recording the information in the search database,
A similar word dictionary in which similar words are recorded in advance for each word,
Input means for inputting a question sentence,
Morphological analysis of the question sentence by the morphological analysis means, a first search condition creating means for creating a search condition based on the obtained word and information specifying affirmative or negative,
Referring to the similar word dictionary, a similar word acquiring means for obtaining a similar word of a word obtained as a result of morphological analysis of the question sentence,
Morphological analysis of the similar words by the morphological analysis means, a second search condition creation means for creating a search condition based on the obtained word and information specifying affirmation or negation,
Search means for searching the search database for documents matching each of the first search condition and the second search condition;
Display means for displaying the document data searched by the search means. The search system comprises:
Importance calculating means for calculating the importance of the word obtained by the morphological analysis means, recording the importance of the word in the search database in association with each word,
Aggregation means for sorting search results by importance,
Further comprising
4. The search system according to claim 1, wherein the display unit displays the search results sorted by importance in the totaling unit. A document database for recording document data to be searched;
A search database that records search data;
A search program that operates in a search system including a morphological analysis dictionary in which data for morphological analysis is recorded in advance,
With reference to the morphological analysis dictionary, while extracting words by morphologically analyzing the document data recorded in the document database, for the words, a morphological analysis process for specifying a positive or negative form,
A word extracted by the morphological analysis process, information for specifying affirmation or negation, and document registration processing for associating information for specifying a document and recording the information in the search database;
Input processing for inputting a question sentence,
A morphological analysis of the question sentence by the morphological analysis process, based on the obtained word and information specifying affirmation or negation, a search condition creation process of creating a search condition,
A search process for searching the search database for a document that matches the search condition;
A search program for causing a computer to perform a display process for displaying a search result in the search process. A document database for recording document data to be searched;
A search database that records search data;
A morphological analysis dictionary in which data for morphological analysis is recorded in advance,
A search program that operates in a search system including a similar word dictionary in which similar words are recorded in advance for each word,
With reference to the morphological analysis dictionary, while extracting words by morphologically analyzing the document data recorded in the document database, for the words, a morphological analysis process for specifying a positive or negative form,
A word extracted by the morphological analysis process, information for specifying affirmation or negation, and document registration processing for associating information for specifying a document and recording the information in the search database;
Input processing for inputting a question sentence,
A first search condition creation process for creating a search condition based on the obtained word and information specifying affirmation or negation, by morphologically analyzing the question sentence by the morphological analysis process;
Referring to the similar word dictionary, a similar word acquisition process of obtaining a similar word of a word obtained as a result of morphological analysis of the question sentence,
A second search condition creation process for creating a search condition based on the similar words obtained by the similar word acquisition process;
A search process for searching the search database for a document that matches each of the first search condition and the second search condition;
A search program for causing a computer to perform a display process of displaying document data searched by the search process. A document database for recording document data to be searched;
A search database that records search data;
A morphological analysis dictionary in which data for morphological analysis is recorded in advance,
A search program that operates in a search system including a similar word dictionary in which similar words are recorded in advance for each word,
With reference to the morphological analysis dictionary, while extracting words by morphologically analyzing the document data recorded in the document database, for the words, a morphological analysis process for specifying a positive or negative form,
A word extracted by the morphological analysis process, information for specifying affirmation or negation, and document registration processing for associating information for specifying a document and recording the information in the search database;
Input processing for inputting a question sentence,
A first search condition creation process for creating a search condition based on the obtained word and information specifying affirmation or negation, by morphologically analyzing the question sentence by the morphological analysis process;
Referring to the similar word dictionary, a similar word acquisition process of obtaining a similar word of a word obtained as a result of morphological analysis of the question sentence,
A second search condition creating process for creating a search condition based on the obtained word and information specifying affirmative or negative, by morphologically analyzing the similar words by the morphological analysis process;
A search process for searching the search database for a document that matches each of the first search condition and the second search condition;
A search program for causing a computer to perform a display process of displaying document data searched by the search process. The search program,
Calculating the importance of the word obtained by the morphological analysis processing, and the importance calculation processing of recording the importance of the word in association with each word in the search database;
An aggregation process that sorts search results by importance,
To the computer further,
8. The search program according to claim 5, wherein the display process displays search results sorted in order of importance by the aggregation process.

Images