WO2021210210A1 - Dispositif de recherche documentaire, système de recherche documentaire et procédé de recherche documentaire - Google Patents

Dispositif de recherche documentaire, système de recherche documentaire et procédé de recherche documentaire Download PDF

Info

Publication number
WO2021210210A1
WO2021210210A1 PCT/JP2020/043312 JP2020043312W WO2021210210A1 WO 2021210210 A1 WO2021210210 A1 WO 2021210210A1 JP 2020043312 W JP2020043312 W JP 2020043312W WO 2021210210 A1 WO2021210210 A1 WO 2021210210A1
Authority
WO
WIPO (PCT)
Prior art keywords
document
search
terminal device
label information
database
Prior art date
Application number
PCT/JP2020/043312
Other languages
English (en)
Japanese (ja)
Inventor
佳那子 三田
健太郎 森本
勇樹 石川
Original Assignee
株式会社島津製作所
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 株式会社島津製作所 filed Critical 株式会社島津製作所
Priority to JP2022515199A priority Critical patent/JPWO2021210210A1/ja
Publication of WO2021210210A1 publication Critical patent/WO2021210210A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/332—Query formulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce

Definitions

  • the present disclosure relates to a document search device, a document search system, and a document search method, and more specifically, to a display technique of a document searched by the document search system.
  • Patent Document 1 discloses an example of a document retrieval system for providing the above-mentioned search service.
  • search service keywords or sentences input by the user are analyzed using natural language processing (NLP), and terms that match or are similar to the words (characteristic words) obtained by the analysis are analyzed. Extract the containing documents from the database and present them to the user.
  • NLP natural language processing
  • the number of documents to be extracted may be large, and in such a case, a document containing desired information from the documents displayed as a search result. It may take some time for the user to reach.
  • Patent Document 1 the extracted documents are displayed in a list, and the specified feature words are emphasized or displayed, or the features are displayed. By also displaying a graph showing the relevance of words, the user can easily analyze the search results.
  • Patent Document 1 When Japanese Patent Application Laid-Open No. 2008-293534 (Patent Document 1) is applied to the field of an analyzer as described above, it is basically a feature word obtained by analysis from an input keyword or document. A search is performed based on. However, for example, when the input keyword is only the sample name, the information such as the analyzer required by the user or the preprocessing may not be appropriately obtained.
  • the present disclosure has been made to solve such a problem, and an object thereof is to easily select a document that meets a user's request from the searched documents in a document search system. Is.
  • the document retrieval device is used in the document retrieval system.
  • the document retrieval system includes a terminal device and a database capable of communicating with the document retrieval device.
  • the database stores documents related to analysis and label information related to the documents.
  • the document retrieval device includes an analysis unit, a search unit, an acquisition unit, and a display control unit.
  • the analysis unit analyzes the search query entered by the user in the terminal device and interprets the character string included in the search query.
  • the search unit searches the database for documents related to the search query based on the analysis results of the analysis unit.
  • the acquisition unit acquires the label information associated with the searched document from the database.
  • the display control unit displays the searched document on the terminal device together with the label information related to the document.
  • the document retrieval system includes a terminal device, a database, and a server.
  • the terminal device accepts input from the user and displays the search result.
  • the database stores documents related to analysis and label information related to the documents.
  • the server is configured to be able to communicate with the terminal device and the database.
  • the server includes an analysis unit, a search unit, an acquisition unit, and a display control unit.
  • the analysis unit analyzes the search query entered by the user in the terminal device and interprets the character string included in the search query.
  • the search unit searches the database for documents related to the search query based on the analysis results of the analysis unit.
  • the acquisition unit acquires the label information associated with the searched document from the database.
  • the display control unit displays the searched document on the terminal device together with the label information related to the document.
  • the document search method is a step of i) receiving a search query entered by a user and ii) a step of analyzing the received search query and interpreting a character string included in the search query. And iii) the process of searching the database for documents related to the search query based on the analysis results, iv) the process of acquiring the label information associated with the searched documents from the database, and v) the searched documents. Is included on the display device together with the label information related to the document.
  • the label information associated with the document is displayed on the terminal device together with the document searched based on the search query input by the user. ..
  • the user can determine whether or not the document contains the desired content just by looking at the displayed label information. Can be done. As a result, the user can easily select a document that matches the user's request from the searched documents.
  • FIG. 1 is a diagram showing an overall configuration of a document retrieval system 10 according to an embodiment.
  • the document retrieval system 10 shown in FIG. 1 is a system that provides document information including desired information in response to a request from a user through a communication network 50.
  • the document retrieval system 10 provides information on analysis cases in response to a request from an information searcher (user), in which an apparatus manufacturer that manufactures and sells an analyzer serves as an information provider. The case where the system is used will be described as an example.
  • the document retrieval system 10 includes a server 100, a database 200, and at least one terminal device 300A to 300N.
  • the terminal devices 300A to 300N are also collectively referred to as "terminal device 300".
  • the server 100 and the database 200 are arranged in the device maker, and the terminal device 300 is a personal computer owned by each user or a mobile terminal such as a tablet or a smartphone.
  • the "server” of the embodiment corresponds to the "document retrieval device” in the present disclosure.
  • the terminal device 300 includes an input unit 301 and a display unit 302.
  • the input unit 301 includes, for example, a keyboard or a mouse.
  • the input unit 301 may be a microphone (and a converter that converts voice into digital data) that accepts user input through voice.
  • the display unit 302 is typically a display device such as a liquid crystal display. When the terminal device 300 is a tablet or a smartphone, the input unit 301 and the display unit 302 are realized as a touch panel in which these functions are integrally formed.
  • the server 100 is configured to be able to communicate with the terminal device 300 via a communication network 50 represented by the Internet. Further, the server 100 is configured to be able to communicate with the database 200 via wired communication or wireless communication.
  • the server 100 includes a CPU 101 which is a control device, a storage unit 102, and a communication unit 103.
  • the CPU 101, the storage unit 102, and the communication unit 103 are connected to a common communication bus 104, and are configured to enable data transmission to each other.
  • the storage unit 102 includes a storage memory such as a ROM and a RAM, and / or a large-capacity storage medium typified by an HDD and a DVD.
  • the communication unit 103 is a communication interface for communicating with the communication network 50, the database 200, and other devices (not shown).
  • the server 100 receives a user search query (search keyword) transmitted from the terminal device 300, extracts information suitable for the user's request from the database 200 based on the content of the search query, and extracts the information suitable for the user's request from the database 200. It is displayed on the terminal device 300.
  • search keyword search keyword
  • the database 200 is a storage device including a large-capacity storage medium.
  • the database 200 stores a large amount of document information regarding sample analysis and information related to the document.
  • the database 200 is provided outside the server 100, but the database 200 may form a part of the storage unit 102 provided inside the server 100. ..
  • part or all of the database may be connected to the server 100 via the communication network 50.
  • the database 200A may be included in the user's system.
  • FIG. 2 is a functional block diagram for explaining the details of the server 100 in the document retrieval system 10 of FIG.
  • the database 200 includes a document database (DB) 210 in which document information is stored, a sample DB 220 in which information about a sample (sample) is stored, and label management in which label information corresponding to each document included in the document DB 210 is stored. Includes DB230.
  • the information contained in each database is set / updated by the administrator of the database, for example, using the input device 250.
  • the input device 250 may be a device provided exclusively for the database 200, or a device attached to the server 100 may be used.
  • the information in the database 200 may be automatically updated at predetermined intervals.
  • the document DB 210 stores documents (analysis reports, technical papers, patent documents, etc.) related to a large number of analysis cases to be searched.
  • the document information stored in the document DB 210 may be classified into information that can be disclosed to the general public and information that is permitted to be disclosed only to a specific user.
  • the document information stored in the document DB 210 is stored as document data vectorized by "morphological analysis" used in natural language processing. Specifically, a document, a sentence in a document, and a word in the sentence are stored in a state of being vectorized (distributed expression) in space. By storing as such vectorized data, the similarity between words can be determined from the distance between the vectors.
  • the sample DB 220 stores information about other samples similar to the sample in association with each sample (sample). Both samples may be similar in chemical composition. For example, butyric acid and valeric acid belong to the same carboxylic acid (or fatty acid, or organic acid in the broader sense) and have similar carbon numbers, so they may be similar in chemical composition and related to each other. In addition, samples having similar chemical or physical properties may be associated with each other. Further, for example, for a nucleic acid-constituting base called adenine, the same nucleic acid-constituting bases such as cytosine, uracil, and thymine may be associated with each other.
  • label management DB 230 items (label information) related to each document stored in the document DB 210 are stored in association with each document.
  • the items included in the label information include, for example, the name (sample name) of the sample to be analyzed in the document, the analysis method (analysis type), the presence / absence and type of pretreatment, and the analysis conditions. Includes at least one.
  • the server 100 includes a receiving unit 110, an analysis unit 120, a search unit 130, a ranking generation unit 140, a label acquisition unit 150, and a display control unit 160 as functional blocks.
  • Each of the above functional blocks is mainly realized by executing the program stored in the storage unit 102 on the CPU 101.
  • a part of the functional block may be realized by a dedicated hardware circuit.
  • the receiving unit 110 receives the information of the search query QRY input by the user in the terminal device 300 of FIG. 1 via the communication network 50, and outputs the received search query QRY to the analysis unit 120.
  • the search query QRY is a character string expressed in natural language, and is a set of words or a series of sentences.
  • the analysis unit 120 performs "morphological analysis" on the search query QRY from the reception unit 110 based on a predefined search dictionary, and divides the search query QRY into the smallest unit of words. Then, the analysis unit 120 calculates the word vector WVC (or feature vector) that characterizes the divided words. The analysis unit 120 outputs the calculated word vector WVC information to the search unit 130.
  • the search unit 130 extracts the characteristics of the search query QRY from the word vector WVC calculated by the analysis unit 120, and based on the extracted characteristics, matches the characteristics from the document information stored in the document DB 210. Search for document information DOC that contains similar content.
  • the search in the search unit 130 is not a keyword search that returns document information that matches the character string (keyword) input as the search query QRY, but a semantic search that interprets the content of the analysis case to be searched from the keyword. Is done.
  • the search unit 130 outputs the document information DOC extracted from the document DB 210 by the semantic search to the ranking generation unit 140 and the display control unit 160. Further, the search unit 130 outputs the word vector WVC from the analysis unit 120 to the ranking generation unit 140.
  • the ranking generation unit 140 first calculates the degree of similarity with the keyword included in the search query QRY for each extracted document information DOC.
  • the similarity is calculated by the distance between the word vector WVC and the vector of each word included in the document information DOC, and it is determined that the closer the distance between the two vectors is, the higher the similarity is. If the similarity is higher than the predetermined reference value, it is determined that the keyword included in the search query QRY and the word in the document information DOC match.
  • the ranking generation unit 140 calculates an index value (related score) based on the similarity calculated above. Specifically, the association score is expressed as the sum of similarities. It is determined that the higher the association score, the higher the association between the search query QRY and the document information DOC.
  • the ranking generation unit 140 ranks the extracted document information DOC based on the matching with the keyword included in the search query QRY and the related score, and the ranking information RNK indicating the ranking is used as the label acquisition unit 150. And output to the display control unit 160.
  • the ranking information RNK also includes the above-mentioned related score information.
  • the label acquisition unit 150 acquires the label information LBL individually set for each of the extracted document information DOCs from the label management DB 230 based on the ranking information RNK from the ranking generation unit 140. Then, the label acquisition unit 150 acquires information (sample information) SPL about another sample similar to the sample from the sample DB 220 based on the sample name included in the acquired label information LBL.
  • the label acquisition unit 150 counts the breakdown of the elements included in each label information item for a predetermined number of documents in descending order of ranking (that is, in descending order of relevance). For example, the number of samples included in the sample name item is counted for the label information of the top 50 documents in the ranking.
  • the label acquisition unit 150 outputs the label information LBL including the above count number information and the sample information SPL regarding a similar sample to the display control unit 160.
  • the display control unit 160 receives the document information DOC from the search unit 130, the ranking information RNK from the ranking generation unit 140, and the label information LBL and sample information SPL from the label acquisition unit 150. Based on this information, the display control unit 160 edits the search result screen information DSP to be displayed on the display unit 302 of the terminal device 300, and the terminal for which the search query QRY is requested via the communication network 50. The search result screen information DSP is output to the device.
  • the display control unit 160 displays the extracted document information DOC according to the ranking information RNK. Specifically, the display control unit 160 displays a document containing a word matching the keyword included in the search query QRY at a higher level. Then, for documents that do not include words that match the keywords, documents with a higher association score are displayed so as to be ranked higher, following the documents that match the keywords.
  • the display control unit 160 displays the label information and the related score together with the excerpt of the title and contents of each document to be displayed. Further, the display control unit 160 displays a breakdown of the number of elements of each label information item included in a predetermined number of documents at the top of the ranking.
  • FIG. 3 is a flowchart for explaining the document retrieval control executed on the server 100.
  • the document search control described in FIG. 3 is executed in response to the server 100 receiving the search query QRY from the user.
  • the server 100 acquires the search query QRY transmitted from the terminal device 300 in step 10 (hereinafter, step is abbreviated as S).
  • step is abbreviated as S.
  • the server 100 analyzes the search query QRY by natural language processing and extracts the keywords included in the search query QRY. More specifically, in S20, the server 100 divides the search query QRY into the smallest unit of words (keywords) using "morphological analysis". In morphological analysis, each keyword is vectorized.
  • Semantic search is a technique for identifying a search target that matches a search character string based on the semantic expression of the character string of the input search query, the intention of requesting the search, the purpose, and the like. That is, it is a technique that does not capture the search character string as a mere character string, but as a character string having a meaning including a search request.
  • the server 100 calculates the similarity and the association score between the word included in the extracted document and the keyword extracted from the search query QRY (S40).
  • the similarity between a keyword and a word in the searched document is defined by the distance between each word vector. For example, similarity is represented by a number between 0 and 1. The similarity decreases from 1 as the distance between the vectors increases. Word fluctuations and synonyms are registered in the dictionary in advance and treated as the same word. For fluctuations, synonyms, and synonyms of unregistered expressions, the similarity is set to a value close to 1 when vectorized. Further, the server 100 calculates a "relationship score" indicating the degree of relevance of the document by using the similarity of each word for the document that does not include the word that matches the keyword.
  • the server 100 generates a ranking (ranking) of the extracted documents using the similarity and the related score calculated in S40.
  • a document containing a word that matches a keyword is set to have a higher ranking than a document that does not contain a matching word, and a document having a large number of matching words is set to have a higher ranking.
  • the higher the related score the higher the ranking is set.
  • the server 100 acquires the label information related to the extracted document from the label management DB 230 in S60. Then, in S70, the server 100 acquires the name of a similar sample corresponding to the label (sample label) of the sample name included in the label information from the sample DB 220.
  • the server 100 counts the number of elements for each item of the label information included in the document of the higher ranking according to the ranking generated in S50. For example, with respect to the item of "analyzer" of the label information, the number of the analyzer AA, the analyzer BB, and the analyzer CC is counted in the label information of the top 50 documents.
  • the server 100 edits the acquired document information, label information, similar sample information, and ranking information as search result screen information in S90, and the terminal device 300 to which the search query QRY is transmitted via the communication network 50. Output to.
  • the searched documents are displayed in the order of ranking together with various information related to the documents.
  • FIG. 4 is a diagram showing a first example of the display screen 400 displayed on the display unit 302 of the terminal device 300.
  • the display screen 400 has an input area 410, a search button 420 for transmitting a search request, a first display area 430 (430A to 430C) for displaying search results by the server 100, and a first display area 430 (430A to 430C). Includes 2 display areas 440.
  • the input area 410 is an area for the user to input a search query using the input unit 301.
  • the search button 420 When the user selects the search button 420 while the search query is input, the information of the search query is transmitted from the terminal device 300 to the server 100.
  • the first display area 430 is an area in which the details of the document searched by the server 100 are displayed.
  • the first display area 430 includes three display areas 430A to 430C, and information on the top three documents among the searched documents is shown.
  • the number of documents that can be displayed in the first display area 430 is not limited to three, and three or more documents may be displayed depending on the screen size and screen arrangement of the display unit 302. Further, the first display area 430 can be scrolled by a user touching the mouse, keyboard, or screen, and the scrolling can display information of a lower ranking.
  • each display area in the first display area in addition to the document name, an area 341 for displaying the related score, an area 342 for displaying the match / similar keyword, and an area 433 for displaying the outline of the document or a part of the document. 435 to 437 for displaying label information, and 438 for displaying similar sample names.
  • the document A including the word matching the keyword included in the search query is displayed in the display area 430A.
  • "no related score” is displayed in the area 431
  • the matching keyword is displayed in the area 432.
  • the documents B and C shown in the display areas 430B and 430C do not include words matching the keywords included in the search query, the related score is displayed in the area 431. Further, in the area 432, a word similar to the keyword is shown.
  • the analyzer is shown in the area 435, the presence / absence and type of pretreatment are shown in the area 436, and the name (sample name) of the sample to be analyzed is shown in the area 437. It is shown. Then, in the area 438, the name of the sample similar to the sample displayed in the area 437 is displayed.
  • the second display area 440 is an area for displaying a breakdown of label information for a document having a high ranking among the extracted documents. For example, as shown in FIG. 4, in the second display area 440, the breakdown of the elements included in the label information in the top 50 documents in the ranking is displayed.
  • the items of the analyzer include 30 documents having the label of the analyzer AA, 15 documents having the label of the analyzer BB, and 5 documents having the label of the analyzer CC.
  • the items of preprocessing include 15 documents that perform preprocessing EE, 10 documents that perform preprocessing FF, and 25 documents that do not perform preprocessing.
  • the display area of each searched document is limited, the content of the document shown in the area 433 can be partial. Therefore, it may not be possible to determine whether or not the actual content described in the document contains the information desired by the user only from the similarity with the document name and the keyword extracted from the search query. ..
  • the present embodiment by preliminarily assigning information that is generally considered to be requested by the user to each document as label information, it is not necessary to read the contents of the document in detail.
  • Basic information such as the sample targeted by the document and the analyzer used can be recognized from the label information.
  • the sample name similar to the sample extracted from the search query together, it is possible to understand the applicability of the analysis method described in the document to other samples. Further, even when the search query does not clearly describe the sample to be analyzed (or when the specific sample name cannot be specified), the sample name and / or the area displayed in the area 437. From the similar sample name displayed on 438, it is possible to determine whether or not the content of the document is a document that meets the user's request.
  • the label information related to the document is displayed, it is possible to easily select a document that matches the user's request from the searched documents. Become.
  • the contents of the label information displayed in the areas 345 to 437 and the similar sample names displayed in the area 438 are all contents because the size of the display area shown in FIG. 4 is limited. May not be displayed. Therefore, when the user selects the item by clicking on these areas or by hovering the mouse cursor over the area, as shown in FIG. 5, the selected item
  • the detailed information may be displayed in the pop-up window 450 or another screen.
  • the document retrieval device is used in a document retrieval system.
  • the document retrieval system includes a terminal device and a database capable of communicating with the document retrieval device.
  • the database stores documents related to analysis and label information related to the documents.
  • the document retrieval device includes an analysis unit, a search unit, an acquisition unit, and a display control unit.
  • the analysis unit analyzes the search query entered by the user in the terminal device and interprets the character string included in the search query.
  • the search unit searches the database for documents related to the search query based on the analysis results of the analysis unit.
  • the acquisition unit acquires the label information associated with the searched document from the database.
  • the display control unit displays the searched document on the terminal device together with the label information related to the document.
  • the label information associated with the document is displayed on the terminal device together with the document searched based on the search query input by the user.
  • the user can determine whether or not the document contains the desired content just by looking at the displayed label information. Can be done. As a result, the user can easily select a document that matches the user's request from the searched documents.
  • the label information includes an item of the name of the sample targeted by the document.
  • the document search device acquires a sample name similar to the sample name included in the label information of the searched document from the database and displays it on the terminal device.
  • a sample name similar to the sample name included in the label information of the searched document is displayed together with the document. This makes it possible to understand the applicability of the analytical techniques described in the retrieved document to other samples. Therefore, even if the sample that the user wants to analyze is not described in the searched document, it is possible to easily determine whether or not the analysis method described in the document is applicable.
  • the document search device calculates the similarity with the character string included in the search query for the searched document, and searches according to the index value based on the similarity. It further includes a ranking generator that ranks the documents that have been created. The document search device displays the searched document on the terminal device according to a set ranking.
  • the searched document is preferentially displayed on the terminal device as a document having a higher relevance to the search query. Therefore, the user can easily select a document closer to the user's request from the searched documents.
  • Paragraph 4 The document search device described in Paragraph 3 displays the index value of the searched document on the terminal device together with the searched document.
  • the index value of the searched document is displayed on the terminal device.
  • the relevance to the search query entered by the user can be recognized as a numerical value. Therefore, it is possible to easily select a document closer to the user's request from the searched documents.
  • a predetermined number of documents can be displayed in descending order of relevance. If there are many related documents and all the searched documents are displayed, the communication load between the server and the terminal device may increase or the display may take a long time. By displaying a predetermined number of documents, it is possible to improve the response of displaying the search results.
  • the breakdown of the label information included in the searched document is displayed on the terminal device.
  • Candidates for analyzers, preprocessing, etc. suitable for the content of search queries from users can be easily understood.
  • the label information attached to each document and the breakdown of the label information it is possible to easily select a document that matches the user's request from the search results.
  • the label information further includes an analysis method, an analyzer, the presence / absence and type of preprocessing, and at least one of analysis conditions. ..
  • the user can obtain information on the analysis method, analyzer, presence / absence and type of preprocessing, and analysis conditions for the searched document. Can be obtained.
  • the document retrieval device displays detailed information of the selected item on the terminal device when the area of the label information displayed on the terminal device is selected by the user. do.
  • the document retrieval device described in item 8 when the label information area in the terminal device is selected, the detailed information of the item is displayed on the terminal device. Since there is a limit to the number of characters in the search result display area and there are many chemical substances with long names, it may not be possible to display all the contents in the label information display area. By displaying the detailed information when the corresponding item of the label information is selected, it is possible to further facilitate the understanding of the relationship between the user's request and the searched document.
  • the document search system is a terminal device that accepts input from a user and displays search results, a database that stores a document related to analysis and label information related to the document, and a terminal. It includes a device and a server capable of communicating with the database.
  • the server includes an analysis unit, a search unit, an acquisition unit, and a display control unit.
  • the analysis unit analyzes the search query entered by the user in the terminal device and interprets the character string included in the search query.
  • the search unit searches the database for documents related to the search query based on the analysis results of the analysis unit.
  • the acquisition unit acquires the label information associated with the searched document from the database.
  • the display control unit displays the searched document on the terminal device together with the label information related to the document.
  • the label information associated with the document can be displayed on the terminal device together with the document searched based on the search query input by the user.
  • the user can determine whether or not the document contains the desired content just by looking at the displayed label information. Can be done. As a result, the user can easily select a document that matches the user's request from the searched documents.
  • the document search method includes a step of receiving a search query input by a user, a step of analyzing the received search query, and a step of interpreting a character string included in the search query. Based on the results, the process of searching the database for documents related to the search query, the process of acquiring the label information associated with the searched document from the database, and the process of retrieving the searched document with the label information related to the document. It also includes a step of displaying on a display device.
  • the label information associated with the document is displayed on the display device together with the document searched based on the search query input by the user.
  • the user can determine whether or not the document contains the desired content just by looking at the displayed label information. Can be done. As a result, the user can easily select a document that matches the user's request from the searched documents.
  • 10 document search system 50 communication network, 100 server, 102 storage unit, 103 communication unit, 104 communication bus, 110 reception unit, 120 analysis unit, 130 search unit, 140 ranking generation unit, 150 label acquisition unit, 160 display control unit , 200, 200A database, 210 document DB, 220 sample DB, 230 label management DB, 250 input device, 300, 300A to 300N terminal device, 301 input unit, 302 display unit, 400 display screen, 410 input area, 420 search button 430, 430A, 430B, 430C, 440 display area, 450 pop-up window.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • General Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Marketing (AREA)
  • Finance (AREA)
  • Economics (AREA)
  • Development Economics (AREA)
  • Accounting & Taxation (AREA)
  • Mathematical Physics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

Un système de recherche documentaire (10) comprend un serveur (100), une base de données (200) et un dispositif terminal (300). Le serveur peut communiquer avec le dispositif terminal et la base de données. Le serveur comprend une unité d'analyse (120), une unité d'extraction (130), une unité d'acquisition (150) et une unité de commande d'affichage (160). L'unité d'analyse analyse une interrogation d'extraction entrée dans le dispositif terminal par un utilisateur et analyse une chaîne de caractères incluse dans l'interrogation d'extraction. L'unité d'extraction extrait des documents, associés à l'interrogation d'extraction, à partir de la base de données. L'unité d'acquisition acquiert des informations d'étiquette, associées aux documents extraits, à partir de la base de données. L'unité de commande d'affichage affiche les documents extraits, conjointement avec les informations d'étiquette, sur le dispositif terminal.
PCT/JP2020/043312 2020-04-15 2020-11-20 Dispositif de recherche documentaire, système de recherche documentaire et procédé de recherche documentaire WO2021210210A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2022515199A JPWO2021210210A1 (fr) 2020-04-15 2020-11-20

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2020-072863 2020-04-15
JP2020072863 2020-04-15

Publications (1)

Publication Number Publication Date
WO2021210210A1 true WO2021210210A1 (fr) 2021-10-21

Family

ID=78083580

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2020/043312 WO2021210210A1 (fr) 2020-04-15 2020-11-20 Dispositif de recherche documentaire, système de recherche documentaire et procédé de recherche documentaire

Country Status (2)

Country Link
JP (1) JPWO2021210210A1 (fr)
WO (1) WO2021210210A1 (fr)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003173340A (ja) * 2001-12-05 2003-06-20 Shimadzu Corp 機器分析用データ管理装置
JP2008014895A (ja) * 2006-07-10 2008-01-24 Shimadzu Corp 同一性判定プログラム
JP2010055518A (ja) * 2008-08-29 2010-03-11 Nippon Telegraph & Telephone West Corp 検索装置及び方法、ならびに、コンピュータプログラム

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003173340A (ja) * 2001-12-05 2003-06-20 Shimadzu Corp 機器分析用データ管理装置
JP2008014895A (ja) * 2006-07-10 2008-01-24 Shimadzu Corp 同一性判定プログラム
JP2010055518A (ja) * 2008-08-29 2010-03-11 Nippon Telegraph & Telephone West Corp 検索装置及び方法、ならびに、コンピュータプログラム

Also Published As

Publication number Publication date
JPWO2021210210A1 (fr) 2021-10-21

Similar Documents

Publication Publication Date Title
US9384245B2 (en) Method and system for assessing relevant properties of work contexts for use by information services
US9864808B2 (en) Knowledge-based entity detection and disambiguation
US7895595B2 (en) Automatic method and system for formulating and transforming representations of context used by information services
US8209321B2 (en) Emphasizing search results according to conceptual meaning
CA2702651C (fr) Systeme et procede de recherche de documents
US8001135B2 (en) Search support apparatus, computer program product, and search support system
US10108720B2 (en) Automatically providing relevant search results based on user behavior
US20050160080A1 (en) System and method of context-specific searching in an electronic database
US20030220913A1 (en) Techniques for personalized and adaptive search services
US20180004838A1 (en) System and method for language sensitive contextual searching
US20120078979A1 (en) Method for advanced patent search and analysis
Ru et al. Indexing the invisible web: a survey
JP2000200281A (ja) 情報検索装置および情報検索方法ならびに情報検索プログラムを記録した記録媒体
KR100455439B1 (ko) 인터넷 자원의 도메인화와 확장 사이트맵을 이용한정보검색 및 브라우징방법
WO2021210210A1 (fr) Dispositif de recherche documentaire, système de recherche documentaire et procédé de recherche documentaire
KR101238927B1 (ko) 전자도서컨텐츠 검색 서비스 시스템 및 전자도서컨텐츠 검색 서비스 방법
Iyad et al. Towards supporting exploratory search over the Arabic web content: The case of ArabXplore
JP5068304B2 (ja) 抽出装置、方法及びプログラム
EP2185999A2 (fr) Mise en évidence de résultats de recherche en fonction d'une signification conceptuelle
JP2010108427A (ja) 検索サーバ及び方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20930673

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2022515199

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20930673

Country of ref document: EP

Kind code of ref document: A1