JP7050533B2 - Editing device, editing method and editing program - Google Patents

Description

The present invention is a technique for editing text data input by a user and used for searching.

In general, a search device that searches for related documents from a set of one or more documents has a function of inputting search conditions such as a search formula and a search sentence and extracting documents corresponding to the entered search conditions from the document set as search results. Have.

Since a general search device performs a search based on the word specified in the input search condition, it is determined that only the document containing the same word as the instructed word corresponds to the search condition. Therefore, documents containing synonyms or similar words of the specified word may be omitted from the search results.

Therefore, in order to prevent documents containing synonyms and synonyms from being leaked from the search results, a function is desired in which documents containing synonyms and synonyms are also included in the search results.

For example, Patent Document 1 discloses a document search device that enables a search with high search accuracy when synonyms are extracted from a set of documents to be searched and keywords in the search conditions are expanded. ing.

Patent Document 2 discloses a synonym extraction system that extracts synonyms from documents having synonyms that are valid only in a group of documents related to a specific case, such as a proposal or a specification for constructing an information system.

Further, in Patent Document 3, based on the search condition in which the constituent elements and vocabulary of the structured document stored in the structured document database are specified, the specified constituent elements and vocabulary and their occurrence positions in the document structure are described. A structured document search method that can easily search for a structured document similar to the above is disclosed.

Non-Patent Document 1 discloses a method for extracting a consent expression. Non-Patent Document 2 discloses a method for acquiring a superior / inferior relationship between words.

By using the above technique, it is possible to include documents including synonyms and synonyms of the words specified in the search conditions in the search results.

Here, in a search device such as Patent Document 1, generally, text data including a search condition is manually input. The search device acquires the text data, generates a search condition using the words included in the text data, and executes a search for the document corresponding to the search condition. The searcher may refer to the documents contained in the search results and perform further searches.

Japanese Patent Application Laid-Open No. 2003-108582 International Publication No. 2014/002776 Japanese Unexamined Patent Publication No. 2002-297605

Takeshi Yumura, Naoyuki Yoda, Yasuo Nozaki, Ken Mogi, Yuki Nishida, Sanyo Electric Co., Ltd. Information and Communication Systems Laboratories "Extraction of Consent Expressions from Text Databases", Proceedings of the National Convention, September 27, 1993 , 47th Artificial Intelligence and Cognitive Science, p. 79-80 Keiji Niisato, Kentaro Torizawa, "Automatic Acquisition of Upper and Lower Relationships between Words from HTML Documents", IPSJ Research Report Natural Language Processing (NL), November 6, 2003, p. 95-102

As mentioned above, the text data used for the search may be manually entered or modified. Therefore, there is a problem that the accuracy of the search varies because the text data used for the search has personality.

The present invention has been made in view of the above problems, and an object of the present invention is to provide an editing device or the like capable of suppressing the personality of text data and suppressing variations in search accuracy.

The editing device according to one aspect of the present invention is included in the acquired text data from an acquisition means for acquiring text data including one or more words and one or more documents to be searched using the text data. It is provided with an editing means for extracting related words for at least one word and editing the text data so that the search is executed including the extracted related words.

In the editing method according to one aspect of the present invention, text data including one or more words is acquired, and at least one included in the acquired text data from one or more documents to be searched using the text data. Related words are extracted for one word, and the text data is edited so that the search is executed including the extracted related words.

The editing program according to one aspect of the present invention is included in the acquired text data from a process of acquiring text data including one or more words and one or more documents to be searched using the text data. A computer is made to perform a process of extracting related words for at least one word and editing the text data so that the search is executed including the extracted related words.

According to the present invention, it is possible to suppress the personality of the text data and suppress the variation in the accuracy of the search.

It is a block diagram which shows the hardware composition of the computer apparatus which realizes an editing apparatus and an editing method in each embodiment. It is a block diagram which shows the functional structure of the editing apparatus which concerns on 1st Embodiment. It is a flowchart explaining the operation of the editing apparatus which concerns on 1st Embodiment. It is a figure which shows the edit screen in which a query is input. It is a figure which shows an example which the query was input to the edit screen. It is a block diagram which shows the functional structure of the query editing part of the editing apparatus which concerns on 1st Embodiment. It is a flowchart which shows the operation of the query editing part of the editing apparatus which concerns on 1st Embodiment. It is a figure which shows an example of the result of query analysis by the query analysis part of the editing apparatus which concerns on 1st Embodiment. It is a figure which shows an example of the document stored in the document storage part of the editing apparatus which concerns on 1st Embodiment. It is a figure which shows the example of the word obtained from the analysis result of the editing apparatus which concerns on 1st Embodiment, and the synonym extracted about the word. It is a figure which shows an example of the query edited by the query editing part of the editing apparatus which concerns on 1st Embodiment. It is a figure which shows the structure which connected the editing apparatus which concerns on 1st Embodiment, and the search apparatus which is the output destination of a query. It is a block diagram which shows the functional structure of the query acquisition part of the editing apparatus which concerns on 2nd Embodiment. It is a figure which shows an example of the editing screen which is output by the selection information input / output part of the editing apparatus which concerns on 2nd Embodiment. It is a flowchart which shows the operation of the query editing part of the editing apparatus which concerns on 2nd Embodiment. It is a block diagram which shows the functional structure of the query editing part of the editing apparatus which concerns on 3rd Embodiment. It is a flowchart which shows the operation of the query editing part of the editing apparatus which concerns on 3rd Embodiment. It is a figure which shows an example of the editing screen displayed by the selection information input / output part of the editing apparatus which concerns on 3rd Embodiment. It is a block diagram which shows the functional structure of the query editing part of the editing apparatus which concerns on 4th Embodiment. It is a flowchart which shows the operation of the query editing part of the editing apparatus which concerns on 4th Embodiment. It is a figure which shows an example of the editing screen displayed by the selection information input / output part of the editing apparatus which concerns on 5th Embodiment. It is a block diagram which shows the functional structure of the query editing part of the editing apparatus which concerns on 5th Embodiment. It is a figure which shows an example of the editing screen displayed by the selection information input / output part of the editing apparatus which concerns on 5th Embodiment. It is a figure which shows an example of the editing screen displayed by the selection information input / output part of the editing apparatus which concerns on 5th Embodiment. It is a figure which shows the example of the word obtained from the analysis result of the editing apparatus which concerns on 6th Embodiment, the synonym extracted about the word, and the number of appearances of a synonym. It is a figure which shows an example of the editing screen displayed by the selection information input / output part of the editing apparatus which concerns on 6th Embodiment. It is a block diagram which shows the functional structure of the editing apparatus which concerns on 7th Embodiment. It is a block diagram which shows the functional structure of the query editing unit and the search result evaluation unit in the editing apparatus which concerns on 7th Embodiment. It is a flowchart which shows the operation of the query editing part of the editing apparatus which concerns on 7th Embodiment. It is a flowchart which shows the operation of the search result evaluation part of the editing apparatus which concerns on 7th Embodiment. It is a figure which shows an example of a query and a search result. It is a figure which shows an example which plotted the rank and the score of each document included in a search result on XY coordinates. It is a figure which shows other examples of a query and a search result. It is a figure which shows the other example which plotted the rank and the score of each document included in a search result on XY coordinates. It is a block diagram which shows the functional structure of the query editing part of the editing apparatus which concerns on 8th Embodiment. It is a flowchart which shows the operation of the query evaluation part of the editing apparatus which concerns on 8th Embodiment. It is a block diagram which shows the functional structure of the query editing part of the editing apparatus which concerns on 9th Embodiment. It is a figure which shows an example of a syntax pattern. An example of the synonym information generated by the synonym information generation unit of the editing apparatus according to the ninth embodiment is shown. It is a block diagram which shows the functional structure of the editing apparatus which concerns on 10th Embodiment.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings.

First Embodiment The hardware constituting the editing apparatus according to the first embodiment and other embodiments will be described. FIG. 1 is a block diagram showing a hardware configuration of a computer device 10 that realizes an editing device and an editing method in each embodiment. In each embodiment, each component of the editing apparatus shown below indicates a block of functional units. Each component of the editing device can be realized, for example, by any combination of the computer device 10 and software as shown in FIG.

As shown in FIG. 1, the computer device 10 includes a processor 11, a RAM (Random Access Memory) 12, a ROM (Read Only Memory) 13, a storage device 14, an input / output interface 15, and a bus 16.

The storage device 14 stores the program 18. The processor 11 uses the RAM 12 to execute the editing program 18 related to the editing device. The program 18 may be stored in the ROM 13. Further, the program 18 may be recorded on the recording medium 20 and read out by the drive device 17, or may be transmitted from an external device via a network.

The input / output interface 15 exchanges data with peripheral devices (keyboard, mouse, display device, etc.) 19. The input / output interface 15 can function as a means for acquiring or outputting data. The bus 16 connects each component.

It should be noted that there are various modified examples of the method of realizing the editing device. For example, the editing device can be realized as a dedicated device. Further, the editing device can be realized by combining a plurality of devices.

A program for operating the configuration of each embodiment so as to realize the functions of the present embodiment and other embodiments (more specifically, a program for causing a computer to execute the process shown in FIG. 3 or the like) is recorded on a recording medium. A processing method of reading a program recorded on the recording medium as a code and executing the program on a computer is also included in the category of each embodiment. That is, a computer-readable recording medium is also included in the scope of each embodiment. Further, not only the recording medium on which the above-mentioned program is recorded but also the program itself is included in each embodiment.

As the recording medium, for example, a floppy (registered trademark) disk, a hard disk, an optical disk, a magneto-optical disk, a CD (Compact Disc) -ROM, a magnetic tape, a non-volatile memory card, or a ROM can be used. Further, the program recorded on the recording medium is not limited to the one that executes the processing by itself, but the one that operates on the OS (Operating System) and executes the processing in cooperation with other software and the function of the expansion board. Is also included in the category of each embodiment.

Next, the functional configuration of the editing apparatus according to the first embodiment of the present invention will be described. FIG. 2 is a block diagram showing a functional configuration of the editing device 100 according to the first embodiment of the present invention. The block shown in FIG. 2 may be mounted in a single device, or may be mounted in a plurality of devices separately. Data can be exchanged between blocks via any means such as a data bus, a network, and a portable storage medium.

The editing device 100 includes a query acquisition unit 110, a query editing unit 120, a query output unit 130, and a document storage unit 140.

The query acquisition unit 110 acquires the query used for the search. Here, the query is text data input by the user and used for the search. Therefore, the query acquisition unit 110 serves as an acquisition means for acquiring text data including one or more words.

The query editing unit 120 extracts related words from the query acquired by the query acquisition unit 110 for the words included in the document stored in the document storage unit 140, and edits the query based on the extracted related words. Therefore, the query editorial unit 120 extracts related words for at least one word included in the acquired text data from one or more documents to be searched using the text data, and searches including the extracted related words. Responsible for editing means of editing text data so that The query output unit 130 outputs the query edited by the query editing unit 120. The document storage unit 140 stores the document to be searched.

The query is, for example, a sentence, but is not limited thereto, and may be text data including one or more words. For example, the query may include words, phrases, clauses, sentences, or a list thereof. Here, the "word" included in the "word" or the "related words", "synonyms", "synonyms" and the like described below includes one or more morphemes. Related words include synonyms, synonyms and antonyms (opposites).

The document stored in the document storage unit 140 includes a published patent gazette, a published patent gazette, a patent gazette, a published document relating to a patent or a utility model such as a published utility model gazette, and other electronic materials and documents including some information. The document stored in the document storage unit 140 may be a document relating to a predetermined technical area. It may also be, for example, a formalized dissertation. In the following description, an example of searching for a public document relating to a patent will be mainly described. The document relating to a predetermined technical area may be, for example, a document relating to a designated technical area among various technical areas such as semiconductors, batteries, and wireless communication. Alternatively, the document relating to a predetermined technical area may be a document relating to a technical area designated by the International Patent Classification.

The editing device 100 is a device that edits the input query using the words included in the document stored in the document storage unit 140 and outputs the edited query. The edited query is output from the query output unit 130 to a search device that searches for documents related to the query. The search device may be a general search device having a function of searching for a document related to a query. The search device may be included in the editing device 100, or may be connected to the editing device 100 via wiring or a network. Alternatively, the editing device 100 may have a configuration included in the search device.

Here, the document storage unit 140 may be provided in the search device. When the search device is connected to the editing device 100 via wiring or a network, the query editing unit 120 accesses the document storage unit 140 included in the search device via wiring or a network.

FIG. 3 is a flowchart illustrating the operation of the editing device 100. The operation of the editing device 100 will be described with reference to FIG.

The query acquisition unit 110 of the editing device 100 acquires a query (step S110). FIG. 4 is a diagram showing an edit screen 111 in which a query is input. As shown in FIG. 4, the edit screen 111 includes an input field 112 and an execute button 113. In the edit screen 111, information for outputting the screen is generated by the query acquisition unit 110, and the information is output to the display device. The display device may be included in the editing device 100, or may be connected to the editing device 100 via wiring or a network.

Further, the editing screen 111 may be displayed by the search device. In this case, the edit screen 111 may be a general search screen displayed by the search device. That is, a search screen including a field for inputting a query and a button for instructing execution of the search may be used as the edit screen 111. In this case, the field for inputting the query may be used as the input field 112 of the edit screen 111, and the button instructing the search execution may be used as the execute button 113 of the edit screen 111.

FIG. 5 is a diagram showing an example in which a query is input to the edit screen 111. As shown in FIG. 5, the query may be, for example, text data containing keywords relating to the field of the document that the user wants to search.

The user inputs a query in the input field 112 and presses the execute button 113. When the execution button 113 is pressed, the query acquisition unit 110 acquires the query input in the input field 112. The query acquisition unit 110 is not limited to acquiring the text data input to the edit screen 111, and may read the text data from the file, for example.

Subsequently, the query editing unit 120 edits the query acquired by the query acquisition unit 110 (step S120). FIG. 6 is a block diagram showing a functional configuration of the query editing unit 120. As shown in FIG. 6, the query editing unit 120 includes a query analysis unit 121, a synonym extraction unit 122, and a synonym addition unit 123.

FIG. 7 is a flowchart showing the operation of the query editing unit 120. The operation of the query editing unit 120 will be described with reference to FIG. 7.

When the query editing unit 120 acquires a query from the query acquisition unit 110, the query analysis unit 121 analyzes the query (step S111). Specifically, the query analysis unit 121 has a function of performing natural language processing on the query and a function of extracting a word or a compound word. As natural language processing, for example, morphological analysis using known morphological analysis software may be performed.

FIG. 8 is a diagram showing an example of the result of the query being analyzed by the query analysis unit 121. FIG. 8 shows a query decomposed into a plurality of words by displaying a slash (/) between words. Words include words consisting of one or more morphemes, compound words, and the like. In addition, the word may include a word with a prefix or a suffix, or a compound word. It should be noted that, in order to show a query decomposed into a plurality of words, it is not limited to displaying a slash (/) between words as shown in FIG. 8, and for example, a comma is displayed between words. Or you may write each word in parentheses.

The query analysis unit 121 decomposes a query into morphemes including verbs, nomenclatures, etc. by morpheme analysis, extracts words from the decomposed query, and combines two or more words into a compound word, or a word and a suffix. Generate a compound word in combination with a prefix.

Subsequently, the synonym extraction unit 122 extracts synonyms from the document stored in the document storage unit 140 with respect to the words obtained from the analysis result.

FIG. 9 is a diagram showing a part of a published patent gazette as an example of a document stored in the document storage unit 140. As shown in FIG. 9, the published patent gazette has a format including items such as "document name", "name of invention", "technical field", "background technology", and "form for carrying out the invention" in advance. It has been decided. After the line where each item name is described, the sentence related to that item is described.

The synonym extraction unit 122 extracts synonyms from the words included in each of the one or more documents stored in the document storage unit 140 based on the analysis result of the query analysis unit 121 (step S112).
The synonym extraction unit 122 sets the word extracted from the query or the compound word generated by the query analysis unit 121 as the target word for synonym extraction as described above. The synonym extraction unit 122 excludes particles from the words to be extracted. Further, the synonym extraction unit 122 may remove the adverb from the word to be extracted by the synonym.

For example, when the analysis result shown in FIG. 8 is used, the synonym extraction unit 122 stores synonyms such as “semiconductor laser”, “light source”, and “parallel light” in the document storage unit 140, respectively. Extract from the words contained in.

Known techniques may be used to extract synonyms. For example, the method disclosed in Non-Patent Document 1 may be used. That is, for example, when a word to be extracted as a synonym appears in a document stored in the document storage unit 140 in an alternating manner before and after the parenthesized expression, the synonym can be extracted. For example, suppose that the analysis yields the word "AAA". In the extraction of the synonym of "AAA", the synonym extraction unit 122 has a description of "AAA (BBB)" and a description of "BBB (AAA)" in any of the documents stored in the document storage unit 140. If included, "BBB" is extracted as a synonym for "AAA".

In the document exemplified in FIG. 9, the item "Background technology" includes the description "Regarding the semiconductor laser (laser diode) ...". When this document or other document stored in the document storage unit 140 includes the description "laser diode (semiconductor laser)", "laser diode" is a synonym for "semiconductor laser" obtained by query analysis. Is extracted.

FIG. 10 is a diagram showing an example of a word to be extracted as a synonym and a synonym extracted by the synonym extraction unit 122 for that word. In FIG. 10, among the words to be extracted as synonyms, the extracted synonyms are shown by taking "semiconductor laser", "parallel light", "reflected light" and "light detector" as examples, but other synonyms are shown. Synonyms are also extracted for words.

Subsequently, the synonym addition unit 123 edits the query using the synonyms extracted by the synonym extraction unit 122 (step S113). Editing means, for example, adding (including) the extracted synonyms to the query acquired by the query acquisition unit 110. Editing may also include replacing the words in the query with the extracted synonyms. Synonyms added to the query are used to search for related documents in the search device.

The synonym addition unit 123 adds the synonyms extracted by the synonym extraction unit 122 to the query acquired by the query acquisition unit 110. FIG. 11 is a diagram showing an example of an edited query. FIG. 11 shows an example in which the synonyms shown in FIG. 10 are added to the input query. FIG. 11 shows an example in which synonyms of the word are added in order, separated by a comma (,), following the word to be extracted, but the extraction is not limited to such order. The synonyms that have been added may be added together at the beginning or end of the sentence. That is, the extracted synonyms may be included in the edited query. Further, although the added synonyms are underlined, the notation is not limited to this, and the added synonyms may be displayed in an identifiable manner. For example, the added synonyms may be highlighted, displayed in different colors, displayed in highlighted characters, and the like. The added synonyms are used to search for related documents in the search device. This completes query editing.

The query output unit 130 outputs the query edited as described above (step S130). As an example, FIG. 12 is a diagram showing a configuration in which a search device 150, which is an output destination of a query, and an editing device 100 are connected. The query output unit 130 outputs the edited query to the search device 150. The search device 150 performs a search of the document storage unit 140 by using the words included in the query output by the query output unit 130.

As described above, the synonyms added to the query are extracted from the documents stored in the document storage unit 140 to be searched. Then, the search for the document storage unit 140 is executed using the query including the synonym.

As described above, according to the first embodiment, the query editing unit 120 extracts synonyms of words or compound words obtained by analyzing the input query from the document storage unit 140 to be searched. And add the word to the query. The search is performed using the query with the added synonyms. Therefore, according to the first embodiment, it is possible to suppress the personality of the query and suppress the variation in the accuracy of the search. Further, as described above, since the synonyms added to the query are extracted from the document storage unit 140 to be searched, the effect that the accuracy of the search can be improved can be obtained.

The synonym extraction unit 122 may consider the applicant in the synonym extraction when the document storage unit 140 stores a public document relating to a patent or a utility model. For example, the synonym extraction unit 122 may extract as a synonym a word that appears in the documents of different applicants by alternating before and after the parenthesized expression of the target word. That is, if the documents of different applicants include the description "AAA (BBB)" and the description "BBB (AAA)", "BBB" may be extracted as a synonym for "AAA". ..

Further, in a document of an applicant whose number of applications is larger than a predetermined standard, words appearing in front of and behind the parenthesized expression of the target word may be extracted as synonyms.

The second embodiment FIG. 13 is a block diagram showing a functional configuration of the query acquisition unit 160 of the editing device 100 according to the second embodiment. The editing device 100 according to the second embodiment includes a query acquisition unit 160 instead of the query acquisition unit 110 included in the editing device 100 according to the first embodiment shown in FIG. Except for the query acquisition unit 160, the editing device 100 according to the second embodiment has the same configuration as that of FIG. It is assumed that the document storage unit 140 stores documents related to a predetermined technical area.

The query acquisition unit 160 includes a selection information input / output unit 161.

The selection information input / output unit 161 generates information for selecting an item for extracting synonyms from the items included in each document stored in the document storage unit 140, and displays the information on the display device 162. The display device 162 may be connected to the editing device 100 via wiring or a network, or may be included in the editing device 100. The selection information input / output unit 161 serves as an output means for outputting information for selecting a predetermined item.

The selection information input / output unit 161 may display, for example, an item selection menu on the display device 162. Further, the selection information input / output unit 161 acquires information indicating the result of selection by the user with respect to the displayed information.

FIG. 14 is a diagram showing an example of the editing screen 163 output by the selection information input / output unit 161. As shown in FIG. 14, the selection information input / output unit 161 displays a pull-down menu 164 for accepting selection of synonym extraction items on the edit screen 163. As shown in FIG. 9, the published patent gazette, which is an example of a document stored in the document storage unit 140, includes a "document name", a "title of the invention", a "technical field", a "background technique", and an "invention". Items such as "form for carrying out" are predetermined. The selection information input / output unit 161 outputs a pull-down menu 164 that accepts selection of which of the statements the synonyms of the words included in the query should be extracted from the sentence. Also, in documents related to a given technical area, "Title", "Abstract", "Introduction (Introduction) and Background", "Technical Principles", "Experiments", "Discussion", "Results" It is predetermined by items such as.

That is, when the user clicks the triangle mark on the right side of the pull-down menu 164, a list of selectable items is displayed directly under the pull-down menu 164 according to the control of the selection information input / output unit 161. This list includes items such as those mentioned above. The user selects an item from this list.

When a query is input to the input field 112 and an item is selected via the pull-down menu 164 and the execute button 113 is pressed, the query acquisition unit 110 acquires the query and information indicating the selected item. do. The selection information input / output unit 161 is not limited to accepting the selection of items by the pull-down menu. In the input field of the synonym extraction item, the name of the item such as "invention name", "technical field", "background technology", "form for carrying out the invention" may be directly input. Further, the item to be selected is not limited to one, and a plurality of items may be selected by a pull-down menu or direct input.

When the query is acquired by the query acquisition unit 110, the query editing unit 120 edits the query.

FIG. 15 is a flowchart showing the operation of the query editing unit 120 of the editing device 100 according to the second embodiment. The query editing unit 120 performs the same operations as the operations S111 and S113 shown by the same reference numerals as those in FIG. 15 for the query analysis unit 121 and the synonym addition unit 123, respectively.

The synonym extraction unit 122 of the query editing unit 120 extracts synonyms from the sentence related to the selected item of the document stored in the document storage unit 140 with respect to the word to be extracted by the synonym (S210). For example, when "semiconductor laser" is included in the target word for synonym extraction and "background technique" is selected by the pull-down menu 164, the synonym extraction unit 122 stores each document stored in the document storage unit 140. Among them, the synonyms of "semiconductor laser" are extracted from the sentence related to the item of "background technology".

When synonyms are extracted from all the descriptions of the document stored in the document storage unit 140, unrelated words (noise) may be extracted as synonyms. Therefore, in the present embodiment, noise can be reduced by extracting synonyms from the sentences related to the items selected by the user.

As described above, according to the second embodiment, the query acquisition unit 160 accepts the selection of the item for extracting the synonyms of the words included in the query, and the synonym extraction unit 122 relates to the selected item. Extract synonyms from sentences. Therefore, it is possible to reduce the extraction of words that are not related to the desired search range as synonyms, and thus it is possible to improve the accuracy of the search.

Third Embodiment FIG. 16 is a block diagram showing a functional configuration of the query editing unit 170 of the editing apparatus 100 according to the third embodiment. The editing device 100 according to the third embodiment includes a query editing unit 170 instead of the query editing unit 120 included in the editing device 100 according to the first embodiment shown in FIG. Except for the query editing unit 170, the editing device 100 according to the third embodiment has the same configuration as that of FIG.

The query editing unit 170 includes a selection information input / output unit 124a in addition to the configuration of the query editing unit 120 shown in FIG.

The selection information input / output unit 124a has a function of accepting the selection of the target word for which synonyms are extracted. That is, the selection information input / output unit 124a outputs information for the user to select whether or not to extract synonyms for the word or compound word obtained by the query analysis unit 121. The selection information input / output unit 124a also acquires information indicating the result selected by the user. Therefore, the selection information input / output unit 124a serves as an output means for outputting information for selecting whether or not to extract related words for at least one word included in the acquired text data.

FIG. 17 is a flowchart showing the operation of the query editing unit 170. The query editing unit 170 performs the same operations as the operations S111 and S113 shown by the same reference numerals as those in FIG. 17 for the query analysis unit 121 and the synonym addition unit 123, respectively, and therefore detailed description thereof will be omitted. do.

The selection information input / output unit 124a selectively displays the word or compound word obtained by the query analysis unit 121 on the display device 162 as a candidate for extracting synonyms (S220). FIG. 18 is a diagram showing an example of the editing screen 171 displayed by the selection information input / output unit 124a. As shown in FIG. 18, the selection information input / output unit 124a displays the synonym extraction candidate 172 on the edit screen 171. Specifically, the selection information input / output unit 124a displays the word or compound word 173 obtained by the query analysis unit 121 and the check box 174 associated with each word.

The word or compound word 173 to be displayed is a word excluding particles among the words or compound words obtained by the query analysis unit 121. Check box 174 allows selection of associated words. When the check box 174 is clicked by the user, the associated word is selected as the target word for synonym extraction. Although it is exemplified that the check box 174 is used for selection, the present invention is not limited to this.

The user selects a target word from the candidates for synonym extraction, checks the check box 174 associated with the target word, and presses the execute button 113 (Yes in S230). The user can select the target word for synonym extraction, for example, so as to reduce the noise contained in the search result.

The selection information input / output unit 124a acquires the selected word and notifies the synonym extraction unit 122. The synonym extraction unit 122 acquires the selected word and extracts the synonym from the document stored in the document storage unit 140 with respect to the selected word (step S240).

As described above, according to the third embodiment, the selection information input / output unit 124a accepts the selection of the target word for extracting the synonyms, so that the words including the synonyms that may cause noise can be excluded. , The effect that the desired search result can be obtained for the user can be obtained.

Fourth Embodiment FIG. 19 is a block diagram showing a functional configuration of the query editing unit 180 of the editing apparatus 100 according to the fourth embodiment. The editing device 100 according to the fourth embodiment includes a query editing unit 180 instead of the query editing unit 120 included in the editing device 100 according to the first embodiment shown in FIG. Except for the query editing unit 180, the editing device 100 according to the fourth embodiment has the same configuration as that of FIG.

The query editing unit 180 includes a selection information input / output unit 124b in addition to the configuration of the query editing unit 120 shown in FIG.

The selection information input / output unit 124b has a function of accepting selection of a word used for editing a query among synonyms extracted by the synonym extraction unit 122. That is, the selection information input / output unit 124b outputs information for the user to select a word to be included in the query among the synonyms extracted by the synonym extraction unit 122. Therefore, the selection information input / output unit 124b serves as an output means for outputting information for selecting a related word to be included in the text data from the extracted related words.

FIG. 20 is a flowchart showing the operation of the query editing unit 180. The query editing unit 180 performs the same operations as the operations S111, S112, and S113 shown by the same reference numerals as those in FIG. 20 for the query analysis unit 121, the synonym extraction unit 122, and the synonym addition unit 123, respectively. A detailed description of them will be omitted.

The selection information input / output unit 124b displays the extracted synonyms in a selectable manner (S250). FIG. 21 is a diagram showing an example of the editing screen 181 displayed by the selection information input / output unit 124b. As shown in FIG. 21, the selection information input / output unit 124b displays the synonym-extracted word 182 and the synonym-extracted result 183 on the edit screen 181.

The result of synonym extraction 183 includes the extracted synonyms 184 and the check boxes 185 associated thereto. When the check box 185 is clicked by the user like the check box 174, the associated word is selected as a synonym.

The user selects a word to be used for editing the query from the result of synonym extraction, checks the check box 185 associated with the check box 185, and presses the execute button 113 (Yes in S260). The user can select synonyms, for example, to reduce noise in the search results.

The selection information input / output unit 124b acquires the selected word and notifies the synonym addition unit 123. The synonym addition unit 123 edits the query using the notified synonyms (S113).

As described above, according to the fourth embodiment, the selection information input / output unit 124b accepts the selection of synonyms to be included in the query among the synonyms extracted by the synonym extraction unit 122, which causes noise. It is possible to exclude synonyms, and it is possible to obtain a search result desirable for the user.

Fifth Embodiment FIG. 22 is a block diagram showing a functional configuration of the query editing unit 190 of the editing apparatus 100 according to the fifth embodiment. The editing device 100 according to the fifth embodiment includes a query editing unit 190 instead of the query editing unit 120 included in the editing device 100 according to the first embodiment shown in FIG. Except for the query editing unit 190, the editing device 100 according to the fifth embodiment has the same configuration as that of FIG.

The query editing unit 190 includes a selection information input / output unit 124a in the third embodiment and a selection information input / output unit 124b in the fourth embodiment. That is, the query editing unit 190 of the editing device 100 according to the fifth embodiment has a function of combining the query editing unit 170 in the third embodiment and the query editing unit 180 in the fourth embodiment.

FIG. 23A is a diagram showing an example of the editing screen 191a displayed by the selection information input / output unit 124a. The edit screen 191a shows the same contents as the edit screen 171 shown in FIG. The user selects a target word from the synonym extraction candidates 172, checks the check box 174, and presses the execute button 113. The selection information input / output unit 124a acquires the selected word and notifies the synonym extraction unit 122. The synonym extraction unit 122 extracts synonyms with respect to the selected word, as described in the third embodiment.

FIG. 23B is a diagram showing an example of the editing screen 191b displayed by the selection information input / output unit 124b. As shown in FIG. 23B, the edit screen 191b has the result of synonym extraction 183 in addition to the information of the edit screen 191a shown in FIG. 23A. The result of synonym extraction 183 is a synonym extracted by the synonym extraction unit 122. The user selects a word to be used for editing the query from the result of extracting synonyms 183, and presses the execute button 113. The selection information input / output unit 124b acquires the selected word and notifies the synonym addition unit 123. The synonym addition unit 123 edits the query using the notified synonyms.

As described above, according to the fifth embodiment, the selection information input / output unit 124a accepts the selection of the target word for which the synonym is to be extracted, and the selection information input / output unit 124b is the extracted synonym. Of these, accept the selection of words to be included in the query. Therefore, it is possible to exclude synonyms that may cause noise and words that include the synonyms, and it is possible to obtain an effect that more desirable search results can be obtained for the user.

Sixth Embodiment The editing apparatus according to the sixth embodiment has the same configuration as the editing apparatus 100 according to the fifth embodiment. That is, the editing device according to the sixth embodiment is the editing device 100 provided with the query editing unit 190 shown in FIG. 22. However, the synonym extraction unit 122 of the query editing unit 190 according to the sixth embodiment counts the number of times the synonym appears in the document stored in the document storage unit 140 (hereinafter referred to as "appearance number"). Has the function of

FIG. 24 is a diagram showing an example of a word to be extracted as a synonym, a synonym extracted by the synonym extraction unit 122 for the word, and the number of appearances of the synonym. The synonym extraction unit 122 counts up the number of appearances each time a certain synonym is extracted, and holds the total number of appearances for each synonym.

FIG. 25 is a diagram showing an example of the editing screen 192 displayed by the selection information input / output unit 124b. As shown in FIG. 25, the edit screen 192 has 186 results of the number of occurrences of synonym extraction, in addition to the information on the edit screen 191b shown in FIG. 23B.

When selecting a word to be used for editing a query from the result 183 of the synonym extraction, the user can refer to and select the result 186 of the number of occurrences. The user may select a synonym having a large number of occurrences or a synonym having a small number of occurrences, depending on the importance of the word in the query. The user may select synonyms based on selection rules according to the importance of the word.

For example, when a public document relating to a patent is registered in the document storage unit 140 and a word from which a synonym is extracted is a word indicating a characteristic of a technical idea specified by a query, the user appears relatively with respect to the word. Select synonyms that are less frequent. On the other hand, by selecting a synonym with a small number of appearances, a desired document may be omitted from the search results. Therefore, in some cases, a synonym with a relatively large number of appearances may be selected. Alternatively, in addition to the synonyms that appear less frequently, synonyms that appear relatively frequently may also be selected.

On the other hand, when the word from which the synonym is extracted is not a word indicating the characteristic of the technical idea specified by the query, the user selects the synonym having a relatively large number of occurrences. Alternatively, a synonym with a relatively small number of occurrences may be selected.

In this way, the user may select a synonym according to the number of occurrences of the synonym, based on the relationship between the technical idea and the word specified by the query.

As described above, the user selects a word to be used for editing the query, checks the check box 185 associated with the word, and presses the execute button 113.

The selection information input / output unit 124b acquires the selected word and notifies the synonym addition unit 123. The synonym addition unit 123 edits the query using the notified synonyms.

As described above, according to the sixth embodiment, the selection information input / output unit 124b outputs the number of occurrences of the extracted synonyms to the editing screen 192. Therefore, the user can select a synonym according to the relationship between the content of the query and the word, and the effect that a more desirable search result for the user can be obtained can be obtained.

Seventh Embodiment FIG. 26 is a block diagram showing a functional configuration of the editing apparatus 200 according to the seventh embodiment. As shown in FIG. 26, the editing device 200 according to the seventh embodiment includes a query acquisition unit 110, a query output unit 130, and a document storage unit 140 similar to the editing device 100 described in the first embodiment. The editing device 200 includes a query editing unit 210 instead of the query editing unit 120 of the editing device 100. The editing device 200 further includes a search result evaluation unit 220.

FIG. 27 is a block diagram showing a functional configuration of the query editing unit 210 and the search result evaluation unit 220 in the editing device 200. As shown in FIG. 27, the query editing unit 210 includes a query analysis unit 121, a synonym extraction unit 122, a synonym selection unit 212, a synonym addition unit 123, and a feedback acquisition unit 213.

Further, the search result evaluation unit 220 includes a search result acquisition unit 221, a determination unit 222, and a feedback unit 223. In FIG. 27, the same elements as those shown in FIG. 6 are designated by the same reference numerals, and detailed description thereof will be omitted.

The editing device 200 according to the seventh embodiment has a function of feeding back the search result using the query edited by the query editing unit 210 to the query.

The synonym selection unit 212 selects one or more of the synonyms extracted by the synonym extraction unit 122. The synonym selection unit 212 may, for example, randomly select one or a plurality of synonym sets for synonyms. Therefore, the synonym selection unit 212 serves as a related word selection means for selecting one or a plurality of related words among the related words extracted by the synonym extraction unit 122 included in the query editing unit 210.

The feedback acquisition unit 213 acquires feedback of the search result from the feedback unit 223 of the search result evaluation unit 220.

The search result acquisition unit 221 acquires the result of the search executed by the search device 150. The determination unit 222 determines whether or not the acquired search result satisfies the criteria. If the search result does not meet the criteria, the feedback unit 223 instructs the query editing unit 210 to edit the query again.

FIG. 28 is a flowchart showing the operation of the query editing unit 210. The query analysis unit 121, the synonym extraction unit 122, and the synonym addition unit 123 of the query editing unit 210 perform the same operations as the operations S111, S112, and S113 shown by the same reference numerals as those in FIG. 28, respectively. The detailed explanation of is omitted.

The synonym selection unit 212 selects one or more of the synonyms extracted by the synonym extraction unit 122 (S213). The synonym selection unit 212 may randomly select synonyms. The synonym addition unit 123 edits the query using the synonyms selected by the synonym selection unit 212 (S113). Here, the query obtained by this editing is referred to as "query 1".

The "query 1" is output by the query output unit 130 to the search device 150 connected via wiring or a network as illustrated in FIG. The search device 150 executes a search using "query 1".

FIG. 29 is a flowchart showing the operation of the search result evaluation unit 220. The operation of the search result evaluation unit 220 will be described with reference to FIG. 29. The search result evaluation unit 220 acquires search results from the search device 150 in the search result acquisition unit 221 (S221).

FIG. 30 is a diagram showing an example of “query 1” and “search result 1”. "Query 1" shows an example in which synonyms "laser diode" and "colimated light" are added to the query input as shown in FIG. The "search result 1" includes information about a document extracted as a related document as a result of a search using "query 1". The results shown in FIG. 30 show information about N documents ranked from 1st to N (N is an integer). Here, it is shown that the higher the "score" is, the higher the ranking is. The "score" is an index regarding the similarity between query 1 and N documents. The higher the score, the higher the similarity between Query 1 and the document. As a method for calculating the score, a method can be used in which the importance of the words contained in the document is evaluated by a method such as TF (Term Frequency) -IDF (Inverse Assessment Frequency).

Information about the document may include, for example, publication number, registration number, application number, applicant, title of invention, filing date, publication date.

FIG. 31 is a diagram in which the rank and score of each document included in the search result 1 are plotted on the XY coordinates with the X-axis as the rank attached to the documents included in the search result and the Y-axis as the score. be.

When the search result is acquired as described above, the determination unit 222 determines whether or not the result satisfies the criterion (S222). Here, the editing device 100 holds a reference in advance. The criteria may be arbitrary, for example, a criterion regarding the highest score or a criterion regarding the number of documents having a certain score or higher. Specifically, the criteria may be "the score of the document with the highest ranking is 80 or more" or "there are 10 or more documents with the score of 70 or more".

For example, when the criterion is "the score of the document with the first rank is 80 or more", in the search result 1 shown in FIG. 31, the score of the document with the first rank is "65.3", so the criterion is set. Not satisfied. In this case, the determination unit 222 determines that the search result does not satisfy the criteria (No in S222). Then, the feedback unit 223 instructs the query editing unit 210 to reselect synonyms (S223).

The query editing unit 210 acquires the above instruction in the feedback acquisition unit 213. When the feedback acquisition unit 213 acquires the above instruction, the feedback acquisition unit 213 instructs the synonym selection unit 212 to reselect the synonym. The synonym selection unit 212 reselects synonyms in response to the above instructions. At this time, the synonym selection unit 212 selects a synonym different from the already selected synonym. The synonym selection unit 212 may select a set of synonyms that are all different from the already selected synonyms, may select a set of synonyms that are partially different, or may select a plurality of already selected synonyms. You may delete a part of. The synonym addition unit 123 edits the query using the synonyms selected by the synonym selection unit 212. The query obtained by this editing is referred to as "query 2".

FIG. 32 is a diagram showing an example of “query 2” and “search result 2”. "Query 2" shows an example in which synonyms "diode laser" and "return light" are added to the query input as shown in FIG. "Search result 2" indicates the result of a search using "query 2".

FIG. 33 is a diagram showing the results of plotting the search result 1 and the search result 2 on the XY coordinates, as in the case of FIG. 31. Similarly to the above, when the criterion is "the score of the document having the first rank is 80 or more", the determination unit 222 determines that the search result 2 satisfies the criterion (Yes in S222).

As described above, in the present embodiment, the selection of synonyms and the editing of the query are repeated until the search result meets the criteria, and when the search result meets the criteria, the editing of the query ends.

As described above, according to the seventh embodiment, the search result evaluation unit 220 determines whether or not the search result using the query to which the synonym is added satisfies the criteria, and if not, the search result evaluation unit 220 determines whether or not the search result satisfies the criteria. , Instruct the query editorial unit 210 to edit the query. The query editing unit 210 edits the query by adding synonyms different from the added synonyms until the search result meets the criteria. Therefore, according to the present embodiment, there is an effect that the search result satisfying the criteria can be obtained without the user selecting a synonym.

Eighth Embodiment FIG. 34 is a block diagram showing a functional configuration of the query editing unit 310 of the editing apparatus 100 according to the eighth embodiment. The editing device 100 according to the eighth embodiment includes a query editing unit 310 instead of the query editing unit 120 included in the editing device 100 according to the first embodiment shown in FIG. Except for the query editing unit 310, the editing device 100 according to the eighth embodiment has the same configuration as that of FIG.

The query editing unit 310 includes a query evaluation unit 311 in addition to the query analysis unit 121, the synonym extraction unit 122, and the synonym addition unit 123 described in the first embodiment. The query evaluation unit 311 includes a score calculation unit 312 and a synonym selection unit 212. In this embodiment, the query evaluation unit 311 will be mainly described.

The score calculation unit 312 of the query evaluation unit 311 calculates the query score based on the result of the analysis by the query analysis unit 121. The synonym selection unit 212 has the same function as the synonym selection unit 212 described in the seventh embodiment.

Here, in general, a higher-lower relationship may occur between words depending on the range of meanings (concepts) included in the words. A hypernym (a word indicating a superordinate concept) has a wider concept to be included than a subordinate word (a word indicating a subordinate concept).

For example, in the word "recognition" and the word "image recognition", "recognition" is associated with a hypernym and "image recognition" is associated with a hyponym.

Here, when the query used for the search contains a relatively large number of hypernyms, the search result may include documents that are not related to the document that the user wants to search by expanding the search range. Therefore, in the present embodiment, it will be described that the query is edited when the ratio of the hypernyms included in the query is larger than the standard.

In this embodiment, it is assumed that a word is a hypernym and a compound word is a hyponym. In the above example, "recognition" is a word and a hypernym, and "image recognition" is a compound word and a hyponym. Here, a word is a grammatically minimum linguistic unit that has meaning, and a compound word is a word formed by combining two or more words. A compound word is a word in grammar, but here, "word" and "compound word" are treated separately.

The smaller the ratio of "the number of compound words" to "the number of words + the number of compound words" included in the query, the more hypernyms are included. In this embodiment, in such a case, the query is edited.

FIG. 35 is a flowchart showing the operation of the query evaluation unit 311. The operation of the query evaluation unit 311 will be described with reference to FIG. 35.

The query evaluation unit 311 acquires the analysis result of the query from the query analysis unit 121 (S310). It is assumed that the query evaluation unit 311 has acquired the result of the query analysis as shown in FIG. 8 in the first embodiment.

The score calculation unit 312 of the query evaluation unit 311 calculates the ratio of the "composite word number" to the "word number + compound word number" included in the acquired result as the query score α (S320).

In the example shown in FIG. 8, for example, "semiconductor laser", "parallel light", "reflected light", "light path", and "light detector" are compound words, and "light source", "emission", and "laser" are used. The words "light", "lens", "disc", "top", "collection", "light", "do", "original", "return", and "detection".

In this case, the total number of words and compound words is "18", and the number of compound words is "5". Therefore, the score calculation unit 312 calculates the score α = 5/18. The score is calculated excluding particles.

Here, the score calculation unit 312 holds in advance a reference regarding the score α. The score calculation unit 312 determines whether or not the score α satisfies the standard (S330). For example, when setting a standard that the "number of compound words" in "the number of words + the number of compound words" is more than half, the standard of "score α> 0.5" can be set.

If the score α does not meet the criteria, the score calculation unit 312 instructs the synonym selection unit 212 to select a synonym. The synonym selection unit 212 arbitrarily selects a synonym from the synonyms extracted by the synonym extraction unit 122 (S340).

The synonym selection unit 212 notifies the synonym addition unit 123 of the selected synonym. The synonym addition unit 123 adds the notified synonym to the query.

The selection of synonyms is not limited to selection by the synonym selection unit 212 as described above, and as shown in the third embodiment, the fourth embodiment and the fifth embodiment, the user You may accept your choice. For example, if the score of the query does not meet the criteria, the score calculation unit 312 may accept the user's selection as shown in the third embodiment, the fourth embodiment, and the fifth embodiment.

As described above, according to the eighth embodiment, the query evaluation unit 311 adds synonyms to the query when the ratio of the number of compound words included in the query is smaller than the standard, thereby improving the accuracy of the search. The effect of being able to do is obtained.

Ninth Embodiment FIG. 36 is a block diagram showing a functional configuration of the query editing unit 320 of the editing apparatus 100 according to the ninth embodiment. The query editing unit 320 of the editing device 100 according to the ninth embodiment includes a query evaluation unit 330 instead of the query evaluation unit 311 included in the query editing unit 310 shown in FIG. 34. The query editing unit 320 has the same configuration as that of FIG. 34 except for the query evaluation unit 311.

The query evaluation unit 330 includes a synonym information generation unit 331, a score calculation unit 332, and a synonym selection unit 212. The synonym information generation unit 331 extracts synonyms of words or compound words obtained by the query analysis unit 121 from the document storage unit 140. Known techniques may be used to extract synonyms. For example, the method described in Non-Patent Document 2 can be used. Particles are not subject to synonym extraction.

FIG. 37 is a diagram showing an example of a syntax pattern. In FIG. 37, the notations "CCC" and "DDD" are used as examples of words. In the syntactic pattern as illustrated in FIG. 37, for the words "CCC" and "DDD", "CCC" can be associated with a hyponym and "DDD" can be associated with a hypernym.

It is assumed that the word "CCC" is extracted by the analysis of the query analysis unit 121. When the document storage unit 140 has a synonym pattern as shown in FIG. 37, for example, a synonym pattern such as "DDD called CCC", the synonym information generation unit 331 extracts "DDD" as a synonym for "CCC" and "" "CCC" is associated with a subordinate word, and "DDD" is associated with a hypernym. It should be noted that the relationship is not limited to the upper-lower relationship using the syntax pattern as shown in FIG. 37, but is a higher-level, middle-level, lower-level relationship, or a higher relationship (hereinafter, based on the breadth of the concept to be included). , Also referred to as "hierarchical relationship").

The hypernym has a higher degree of abstraction than the hyponym. Hyponyms can be said to be more embodied words than hypernyms.

The synonym information generation unit 331 executes the above-mentioned extraction of synonyms and the extraction of mutual hierarchical relationships for the words or compound words obtained by the query sentence analysis unit 121. Information including the result of execution is referred to as synonym information.

FIG. 38 shows an example of synonym information generated by the synonym information generation unit 331. The synonym information generation unit 331 notifies the score calculation unit 332 of the generated synonym information.

The score calculation unit 332 calculates the query score based on the acquired synonym information. First, the score calculation unit 332 assigns a score according to the depth of the hierarchy of the extracted synonyms, that is, the degree of abstraction, to each word excluding the particles obtained by the query sentence analysis unit 121. Specifically, a score is given to each word so that the deeper the hierarchy of the extracted synonyms, that is, the lower the degree of abstraction (the higher the degree of concreteness), the higher the score. The depth of the hierarchy indicates the degree of abstraction when the word is layered according to the abstraction level. Words with a deeper hierarchy (lower) indicate relatively lower words, and are scored so that the score is higher.

For example, suppose that when the query includes the word "recognition", the synonyms "pattern recognition", "image recognition", and "speech recognition" are extracted in a hierarchical relationship as shown in FIG. 38. That is, it is assumed that a hierarchical relationship is extracted in which "recognition" is a hypernym, "pattern recognition" is a middle word, and "image recognition" and "speech recognition" are hyponyms. In this case, the score calculation unit 332 determines that the depth of the hierarchy of these words is "3", that is, three layers. Then, the score calculation unit 332 uses, for example, "10" for the hypernym "recognition", "20" for the middle word "pattern recognition", and "image recognition" and "speech recognition" for the hyponyms. A score of "30" is given.

Of the words or compound words included in the query, words for which synonyms are not extracted may be given "10". Further, for words from which synonyms are extracted in the second layer, "10" may be given to the hypernym and "20" may be given to the hyponym.

As described above, the synonym information generation unit 331 creates a thesaurus database based on the words included in the document stored in the document storage unit 140, and the score calculation unit 332 assigns points based on the thesaurus database. In this case, relative points can be given to the words included in the query based on the document storage unit 140. Alternatively, the score calculation unit 332 may give a score based on the existing thesaurus database. In this case, a more absolute score can be given.

Then, the score calculation unit 332 calculates the score of the query by totaling the scores given to the words as described above. In addition, based on the total score of the query when the synonyms of all the words are not extracted from the document storage unit 140, the higher the hierarchy is, the higher the score is given to the words from which the synonyms are extracted as described above. You may calculate the percentage of total points in the query. Similar to the score calculation unit 312 of the seventh embodiment, the score calculation unit 332 holds the standard in advance, and when the above score or ratio does not meet the standard, the synonym selection unit 212 selects a synonym. Instruct.

The synonym selection unit 212 may select any synonym extracted by the synonym extraction unit 122. Alternatively, a synonym of a word having a relatively low score as described above may be selected. Alternatively, as described above, the synonyms extracted by the synonym information generation unit 331 may be selected. At this time, for example, a synonym (concrete related word) that is a subordinate word may be selected from the word included in the query. The synonym selection unit 212 notifies the synonym addition unit 123 of the selected synonym or synonym. The synonym addition unit 123 adds the notified synonym or synonym to the query. The synonym addition unit 123 may replace the word included in the query with the notified synonym or synonym.

As described above, according to the ninth embodiment, the query evaluation unit 330 determines the degree of abstraction of each word based on the word included in the query and the synonyms of the word, and each of the determined words. If the degree of abstraction of the query obtained from the degree of abstraction of the word is larger than the standard, synonyms or synonyms are added to the query, which has the effect of improving the accuracy of the search.

The tenth embodiment FIG. 39 is a block diagram showing a functional configuration of the editing device 400 according to the tenth embodiment. As shown in FIG. 37, the editing device 400 includes an acquisition unit 410 and an editing unit 420. The acquisition unit 410 acquires text data including one or more words. The editorial unit 420 extracts related words for at least one word included in the acquired text data from one or more documents to be searched using the text data, and the search is executed including the extracted related words. Edit the text data so that it is done.

By adopting the above configuration, according to the ninth embodiment, it is possible to obtain the effect that the personality of the text data can be suppressed and the variation in the accuracy of the search can be suppressed.

The present invention has been described above with reference to the above-described embodiments. However, the invention is not limited to the embodiments described above. That is, the present invention can be applied to various aspects that can be understood by those skilled in the art, such as various combinations or selections of various disclosed elements within the scope of the present invention.

Some or all of the above embodiments may also be described, but not limited to:

(Appendix 1) Acquisition means for acquiring text data containing one or more words, and
Related words are extracted from one or more documents to be searched using the text data for at least one word included in the acquired text data, and the search is executed including the extracted related words. An editing device provided with an editing means for editing the text data.

(Appendix 2)
The editing device according to Appendix 1, wherein the document is a document relating to a predetermined technical area.

(Appendix 3)
The editing means according to Appendix 1 or Appendix 2 for extracting the related word from a sentence related to a predetermined item of the document.

(Appendix 4)
Further provided with an output means for outputting information for selecting the predetermined item,
The editing device according to Appendix 3, wherein the editing means extracts the related word from a sentence related to a selected predetermined item in the document.

(Appendix 5)
Further, an output means for outputting information for selecting whether or not to extract the related word is provided for at least one word included in the acquired text data.
The editing device according to any one of Supplementary note 1 to Supplementary note 4, wherein the editing means extracts the related word from the word selected to extract the related word.

(Appendix 6)
Further provided with an output means for outputting information for selecting a related word to be included in the text data from the extracted related words.
The editing device according to any one of Supplementary note 1 to Supplementary note 5, wherein the editing means edits the text data so that the search is executed including the selected related word.

(Appendix 7)
Further provided with a related word selection means for selecting one or more related words from the related words extracted by the editing means.
The editing means acquires the result of a search executed using text data including the related word selected by the related word selection means, and obtains a related word different from the related word selected by the related word selection means. The editing device according to any one of Supplementary note 1 to Supplementary note 6, which is obtained from the related word selection means and edits the text data so that the search is executed including the different related words.

(Appendix 8)
When the score of the result of the acquired search does not meet the criteria, the editing means acquires a related word different from the related word selected by the related word selection means from the related word selection means, and obtains the different association. The editing device according to Appendix 7, which edits the text data so that the search including words is executed.

(Appendix 9)
The editing means edits the text data so that the search is executed including the extracted related words when the ratio of the number of compound words to the number of words included in the text data does not meet the standard. The editing device according to any one of Supplementary note 1 to Supplementary note 8.

(Appendix 10)
The editing means applies to the text data among the related words based on the degree of abstraction of each of the words determined based on the words included in the text data and the related words extracted for the words. The editing device according to any one of Supplementary note 1 to Supplementary note 9, which edits the text data so that the search is executed including related words embodied from the included words.

(Appendix 11)
Get text data containing one or more words,
Related words are extracted from one or more documents to be searched using the text data for at least one word included in the acquired text data, and the search is executed including the extracted related words. An editing method for editing the text data.

(Appendix 12)
The process of acquiring text data containing one or more words,
Related words are extracted from one or more documents to be searched using the text data for at least one word included in the acquired text data, and the search is executed including the extracted related words. An editing program that causes a computer to perform the process of editing the text data.

10 Computer device 11 Processor 14 Storage device 15 Input / output interface 16 Bus 17 Drive device 18 Program 19 Peripheral device 20 Recording medium 100 Editing device 110 Query acquisition unit 120 Query editing unit 121 Query analysis unit 122 Synonymous word extraction unit 123 Synonymous word addition unit 124a Selection information input / output unit 124b Selection information input / output unit 130 Query output unit 140 Document storage unit 150 Search device

Claims (9)

An acquisition method for acquiring text data containing one or more words,
Related words are extracted from one or more documents to be searched using the text data for at least one word included in the acquired text data, and the search is executed including the extracted related words. Is equipped with an editing means for editing the text data .
The editing means extracts the related word from a sentence related to a predetermined item of the document.
Editing device. The editing device according to claim 1, wherein the document is a document relating to a predetermined technical area. Further provided with an output means for outputting information for selecting the predetermined item,
The editing device according to claim 1 or 2 , wherein the editing means extracts the related word from a sentence related to a selected predetermined item in the document. Further, an output means for outputting information for selecting whether or not to extract the related word is provided for at least one word included in the acquired text data.
The editing device according to any one of claims 1 to 3 , wherein the editing means extracts the related word from the word selected to extract the related word. Further provided with an output means for outputting information for selecting a related word to be included in the text data from the extracted related words.
The editing device according to any one of claims 1 to 4 , wherein the editing means edits the text data so that the search is executed including the selected related word. Further provided with a related word selection means for selecting one or more related words from the related words extracted by the editing means.
The editing means acquires the result of a search executed using text data including the related word selected by the related word selection means, and obtains a related word different from the related word selected by the related word selection means. The editing device according to any one of claims 1 to 5 , which is obtained from the related word selection means and edits the text data so that the search is executed including the different related words. The editing means edits the text data so that the search is executed including the extracted related words when the ratio of the number of compound words to the number of words included in the text data does not meet the standard. The editing device according to any one of claims 1 to 6 . The computer
Get text data containing one or more words,
For at least one word contained in the acquired text data, related words were extracted from the sentences related to a predetermined item of the document from one or more documents to be searched using the text data, and the extracted words were extracted. An editing method for editing the text data so that the search is executed including related words. The process of acquiring text data containing one or more words,
For at least one word contained in the acquired text data, related words were extracted from the sentences related to a predetermined item of the document from one or more documents to be searched using the text data, and the extracted words were extracted. An editing program that causes a computer to perform a process of editing the text data so that a search including related words is executed.

Images