JP2008243033A - Retrieval system, retrieval method, program and storage medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a retrieval system capable of displaying a retrieval result so that related information for an input keyword can be noticed easily. <P>SOLUTION: In a DB server 111, the degree of conformance of each keyword with a result product is stored, and a related keyword and relevance with the related keyword are stored for each keyword. If retrieval is directed from a user terminal by specifying a keyword, a result product management server accesses the DB server and displays result products having high degree of conformance with the designated keyword. In addition, a related keyword having high relevance with the designated keyword is displayed. At this time, they are displayed in a form so that the relevance between the keywords may be shown explicitly. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a search system, a search method, a program, and a storage medium that can perform keyword search for a plurality of deliverables stored in a storage device.

  Currently, a keyword search for inputting a keyword as a search condition and extracting a product including the input keyword from a plurality of products to be searched such as a document file is generalized. However, in this keyword search, if the knowledge about the deliverable to be searched is scarce and the content to be searched is not clear, it is difficult to create an appropriate keyword, and there is a high possibility that a search omission will occur.

As a search method for preventing such a search omission in advance, a document search method has been proposed in which a search condition is expanded by adding a related keyword related to an input keyword to perform a search ( Non-patent document 1). Specifically, in this document search method, a related keyword related to the input keyword is extracted from the document to be searched, and the extracted related keyword is added to the input keyword to set a new search condition. Created. Then, a search is performed using the created new search condition, and the documents obtained by the search are ranked and displayed in a list in the order of suitability with the search condition.
Hiroko Mano, 2 others, “Ranking search technology for document search”, Ricoh Technical Report NO.29, December 2003, Ubiquitous Solutions Laboratory, Software Research and Development Headquarters <http://www.ricoh.co.jp/about /business_overview/report/29/pdf/A2902.pdf>

  The document search method described above makes it possible to reduce search omissions by extending search conditions. However, since the search results are displayed in order of the search conditions that are expanded by adding the related keywords, not the keywords previously input by the searcher, the degree of matching with the keywords input by the searcher is high. Finding high search results is difficult. In addition, it is difficult to grasp the degree of association between the input keyword and the extracted related keyword.

  An object of the present invention is to provide a search system, a search method, a program, and a storage medium that can display a search result so that information related to an input keyword can be easily noticed.

  In order to achieve the above object, the present invention relates to the degree of matching with a product associated with each of the keywords extracted from a plurality of products stored in the storage device, and to each keyword. With reference to at least one related keyword and the degree of relevance to the related keyword, a product that matches the keyword input as a search condition is extracted from a plurality of products stored in the storage device And a related keyword for each keyword and a degree of relation with the related keyword are extracted with reference to the related keyword for each keyword and the degree of relation with the related keyword. A plurality of deliverables stored in the storage device with reference to the degree of matching between the extraction means and the deliverables for each keyword Search means for searching for a first product that matches the keyword input as the search condition and a second product that matches the extracted related keyword, and the searched first product Display control means for displaying the product and the second product in a display form indicating the degree of association between the keyword for the first product and the related keyword for the second product. A search system is provided.

  In order to achieve the above object, the present invention is a search system for extracting a product that matches a keyword input as a search condition, and refers to the related keyword for each keyword and the degree of association with the related keyword. Then, an extraction process for extracting a related keyword for the keyword input as the search condition and a degree of relevance with the related keyword, and a matching degree with the deliverable for each keyword are stored in the storage device. A search step of searching for a first product that matches the keyword input as the search condition and a second product that matches the extracted related keyword from a plurality of deliverables, and the search The first product and the second product, the keyword for the first product and the related keyword for the second product. Providing a retrieval system characterized by comprising a display control step of displaying on the display a form indicating the relevance between.

  In order to achieve the above object, the present invention is a program for causing a computer to execute a search method for extracting a product that matches a keyword input as a search condition. Referring to the degree of association with the relevant keyword, an extraction step for extracting the relevant keyword for the keyword input as the search condition and the degree of association with the relevant keyword, and the degree of fitness of the product for each keyword Then, from among a plurality of deliverables stored in the storage device, a first deliverable that matches the keyword input as the search condition, and a second deliverable that matches the extracted related keyword, A search step for searching for the first deliverable and the second deliverable with respect to the first deliverable. And a display control step of displaying on the display a form indicating the relevance of the associated keywords for the second artifact and keyword provides a program for causing the computer to perform.

  In order to achieve the above object, the present invention provides a storage medium in which the above program is stored so as to be readable by a computer.

  According to the present invention, it is possible to display a search result so that information related to an input keyword can be easily noticed even when knowledge about a search target is scarce and the content to be searched is not clear.

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

  FIG. 1 is a block diagram showing a configuration of a product management system in which a search system according to an embodiment of the present invention is employed.

  As shown in FIG. 1, the product management system includes a product management server 101 and a DB server 111. At least one user terminal 102 is connected to the deliverable management server 101 via a network such as a LAN. The user terminal 102 issues an execution request for each process such as a product registration process, a search process, an acquisition process, and a reference process to the deliverable management server 101.

  The deliverable management server 101 includes a CPU, a memory, a Web server program 103, and a hard disk 123. The CPU of the deliverable management server 101 functions as a Web server by performing various processes according to the Web server program 103 read from the memory. By performing processing in accordance with the Web server program 103, it is possible to search for a product, publish it, provide it to the user, and register it from the user terminal 102. The hard disk 123 stores a plurality of deliverables 124 such as files and document files that can be searched for keywords.

  The Web server program 103 constitutes functional modules of a control unit 104, a screen display unit 105, a product file operation unit 106, a relevance calculation unit 107, a fitness calculation unit 108, and a search result creation unit 109. The control unit 104 controls acceptance of requests from the user terminal 102. The screen display unit 105 provides information for performing display control of a user interface displayed via the Web browser of the user terminal 102 as an HTML page. The product file operation unit 106 performs operations on the screen display unit 105 and a large number of product entities. The degree-of-association calculation unit 107 calculates the degree of association between a keyword and a related keyword related thereto. The goodness-of-fit calculation unit 108 calculates goodness that indicates how well the keyword and the product are suitable.

  The search result creation unit 109 creates a system diagram representing a product exceeding a predetermined fitness level and a list representing a list of remaining products not represented by the system diagram for the product extracted by the keyword search. To do. Here, the deliverables extracted by the keyword search include deliverables that match the keyword input as the search condition and deliverables that match the related keyword. Also, the above system diagram shows the results that exceed the specified fitness in the products that match the keyword entered as the search condition, and the results that exceed the specified fitness in the products that match the related keywords. It is a figure which expresses a thing so that the high degree of relevance between each keyword can be specified clearly.

  The DB server 111 includes a CPU, a memory, and a nonvolatile memory that stores each program. The CPU performs various functions by controlling each part of the DB server 111 according to a program.

  The DB server 111 is connected to the product management server 101 via a network. As this connection method, for example, a method such as JNDI (Java (registered trademark) Naming and Directory Interface) is used. The DB server 111 includes a DB management Facade 110, a product management DB 117, and a data access object.

  The DB management Fade 110 stores business logic that receives requests such as data acquisition, registration, and search from the deliverable management server 101 (Web server program 103) and performs processing according to the requests.

  The product management DB 117 includes a CPU and a memory, and manages data by the CPU performing various processes according to a program read from the memory. The product management DB 117 includes a product-keyword information management table 118, a related keyword management table 119, a product management table 120, a keyword management table 121, and a threshold management table 122. The deliverable-keyword information management table 118 is a table for managing information related to the deliverables 124 and keywords. The related keyword management table 119 is a table that manages the related keywords and the value of the related degree. The deliverable management table 120 is a table for managing deliverable information. The keyword management table 121 is a table for managing keyword information. The threshold management table 122 is a table for managing a plurality of thresholds such as a fit product determination threshold, a product display threshold, and a keyword display threshold described later.

  The data access object performs a process of perpetuating the value stored in the product management DB 117 and acquiring and registering the value in response to a request from the DB management Fade 110. The data access object includes an artifact-keyword class 112, an associated keyword class 113, an artifact class 114, a keyword class 115, and a threshold class 116. The product-keyword class 112 holds information for making the data of the product-keyword information management table 118 permanent. The related keyword class 113 holds information for making the data of the related keyword management table 119 permanent. The deliverable class 114 holds information for making the data of the deliverable management table 120 permanent. The keyword class 115 holds information for making the data of the keyword management table 121 permanent. The threshold class 116 holds information for making the data of the threshold management table 122 permanent.

  Next, the object of the product management system will be described with reference to FIG. FIG. 2 is a diagram showing objects of the product management system of FIG.

  In this deliverable management system, objects are divided into an application layer 200 and a data modeling layer 201. Here, the application layer 200 includes a control unit 104, a screen display unit 105, a product file operation unit 106, a relevance calculation unit 107, a fitness calculation unit 108, and a search result creation unit 109 configured by the Web server program 103. Including. The data modeling layer 201 includes a DB management Facade 110, a product-keyword class 112, a related keyword class 113, a product class 114, a keyword class 115, and a threshold class 116.

  Next, the product management table 120 for managing product information will be described with reference to FIG. FIG. 3 is a diagram showing a configuration example of the product management table 120 of FIG.

  In the deliverable management table 120, as shown in FIG. 3, a deliverable ID 301 that can uniquely identify a deliverable and a deliverable directory path 302 that indicates a storage destination directory path in the DB server 111 are described.

  Next, processing performed for the product class 114 for making the data of the product management table 120 permanent will be described with reference to FIGS. FIG. 4 is a sequence diagram showing a processing procedure for acquiring the total number of deliverables stored in the deliverable management server 101 of FIG. FIG. 5 is a sequence diagram illustrating a processing procedure for registering the product directory path in the product management table 120. FIG. 6 is a sequence diagram illustrating a procedure of processing for acquiring a product ID from the product management table 120. FIG. 7 is a sequence diagram illustrating a processing procedure for acquiring a product directory path. FIG. 8 is a sequence diagram illustrating a processing procedure for acquiring a combination list of a product ID and a product directory path.

  Hereinafter, it is assumed that the deliverable management server 101 functions as the Actor 501 by performing the function of each module in accordance with the Web server program 103 belonging to the application layer 200.

  First, processing for acquiring the total number of deliverables stored in the deliverable management server 101 will be described with reference to FIG.

  As shown in FIG. 4, the Actor 501 issues a total product number acquisition request to the DB management Fade 110 (M401). The DB management Facade 110 that has received the total product number acquisition request creates the product class 114, and acquires the product object list stored in the product management DB 117, which is returned from the created product class 114 (M402). Next, the DB management Facade 110 acquires the total number of deliverables with the deliverable object list as an argument (M403). Then, the DB management Facade 110 returns the acquired total number of deliverables to the Actor 501.

  Next, a process for registering the deliverable directory path in the deliverable management table 120 will be described with reference to FIG.

  As shown in FIG. 5, the Actor 501 issues a deliverable registration request with the deliverable directory path indicating the deliverable storage path as an argument to the DB management Facade 110 (M501). Upon receiving the product registration request, the DB management Facade 110 issues a product registration request with the product directory path as an argument to the product class 114 (M502). As a result, the product directory path is registered.

  Next, processing for acquiring a product ID from the product management table 120 will be described with reference to FIG.

  As shown in FIG. 6, the Actor 501 issues a product ID acquisition request with the product directory path as an argument to the DB management Facade 110 (M601). The DB management Facade 110 that has received the product ID acquisition request generates a product class 114 with the product directory path as an argument (M602). Next, the DB management Facade 110 extracts a product ID from the generated product class 114 (M603). Then, the DB management Facade 110 returns the extracted product ID to the Actor 501.

  Next, processing for acquiring a deliverable directory path will be described with reference to FIG.

  As shown in FIG. 7, the Actor 501 issues a deliverable directory path acquisition request with the deliverable ID as an argument to the DB management Facade 110 (M701). Receiving the deliverable directory path request, the DB management Facade 110 generates a deliverable class 114 with the deliverable ID as an argument (M702). Next, the DB management Facade 110 issues a deliverable directory path extraction request with the deliverable ID as an argument to the deliverable class 114 (M703). Then, the DB management Facade 110 acquires the product directory path extracted by the product class 114 in response to the request. The acquired deliverable directory path is returned from the DB management Facade 110 to the Actor 501.

  Next, processing for acquiring a combination list of product IDs and product directory paths will be described with reference to FIG.

  As shown in FIG. 8, the Actor 501 issues a request for acquiring a combination list of a product ID and a product directory path using the product ID list as an argument to the DB management Facade 110 (M801). Receiving the combination list acquisition request, the DB management Facade 110 generates a product class 114 with the product ID as an argument (M802). Next, the DB management Facade 110 issues a product directory path extraction request to the generated product class 114 (M803). Then, the DB management Facade 110 acquires the product directory path extracted in response to the request. Next, the DB management Facade 110 creates a combination list of the product ID and the acquired product directory path (M804).

  Here, the DB management Facade 110 repeats the above-described processing (M802 to M804) by the number of product IDs included in the product ID list. Then, the DB management Facade 110 returns the combination list of the product ID list and the product directory path to the Actor 501 after repeating the above processing.

  Next, the keyword management table 121 for managing keyword information will be described with reference to FIG. FIG. 9 is a diagram showing an example of the keyword management table 121 of FIG.

  As shown in FIG. 9, the keyword management table 121 includes a keyword ID 901 for uniquely identifying a keyword, and a keyword name 902 representing the name of the keyword.

  Next, processing performed for the keyword class 115 for making the data of the keyword management table 121 permanent will be described with reference to FIGS. FIG. 10 is a sequence diagram showing a procedure of processing for acquiring a keyword name. FIG. 11 is a sequence diagram showing a processing procedure for registering a keyword name in the keyword management table 121. FIG. 12 is a sequence diagram illustrating a processing procedure for acquiring a keyword ID.

  As shown in FIG. 10, the Actor 501 issues a keyword name acquisition request with the keyword ID as an argument to the DB management Facade 110 (M1001). Receiving the keyword name acquisition request, the DB management Facade 110 generates a keyword class 115 with the keyword ID as an argument (M1002). Next, the DB management Facade 110 issues a keyword name extraction request to the generated keyword class 115, and acquires the keyword name extracted in response to this request (M1003). Then, the DB management Facade 110 returns the acquired keyword name to the Actor 501.

  Next, processing for registering a keyword name in the keyword management table 121 will be described with reference to FIG.

  As shown in FIG. 11, the Actor 501 issues a keyword registration request with the keyword name as an argument to the DB management Facade 110 (M1101). Receiving the keyword registration request, the DB management Facade 110 issues a keyword name registration request with the keyword name as an argument to the keyword class 115 (M1102). As a result, the keyword name is registered in the keyword management table 121.

  Next, processing for acquiring a keyword ID will be described with reference to FIG.

  As shown in FIG. 12, the Actor 501 issues a keyword ID acquisition request with the keyword name as an argument to the DB management Facade 110 (M1201). Receiving the keyword ID acquisition request, the DB management Facade 110 generates a keyword class 115 with the keyword name as an argument (M1202). Next, the DB management Facade 110 issues a keyword ID extraction request to the generated keyword class 115, and acquires the keyword ID extracted in response to this request (M1203). Then, the DB management Facade 110 returns the acquired keyword ID to the Actor 501.

  Next, the product-keyword information management table 118 for managing product and keyword information will be described with reference to FIG. FIG. 13 is a diagram showing an example of the deliverable-keyword information management table 118.

  In the deliverable-keyword information management table 118, as shown in FIG. 13, a keyword ID 1301, a deliverable ID 1302, a matching degree 1303, a keyword existence number 1304, and an acquisition result 1305 are described. The keyword ID 1301 is for uniquely identifying a keyword. The product ID 1302 is for uniquely identifying the product. The goodness of fit 1303 is a value indicating how well the keyword and the product are fit. The keyword existence number 1304 is the number of keywords corresponding to the keyword ID 1301 present in the deliverable. An acquisition result 1305 indicates whether or not a product obtained as a search result using a corresponding keyword has been acquired by the user terminal 102.

  Next, processing performed for the product-keyword class 112 for making the data of the product-keyword information management table 118 permanent will be described with reference to FIGS. FIG. 14 is a sequence diagram illustrating a procedure of processing for acquiring an acquired product ID list whose degree of conformity exceeds a threshold value. FIG. 15 is a sequence diagram showing a processing procedure for registering product information in the product-keyword information management table 118. FIG. 16 is a sequence diagram illustrating a procedure of processing for updating an acquisition result. FIG. 17 is a sequence diagram illustrating a processing procedure for acquiring a product ID list in which a specified keyword is included in a product. FIG. 18 is a sequence diagram illustrating a processing procedure for acquiring a keyword ID included in a product. FIG. 19 is a sequence diagram showing a procedure of processing for acquiring a product ID list to be displayed in the system diagram. FIG. 20 is a sequence diagram illustrating a processing procedure for acquiring a list of product IDs to be displayed in the search result list.

  First, the process of acquiring the acquired product ID list whose fitness level exceeds the threshold will be described with reference to FIG.

  As shown in FIG. 14, the Actor 501 issues an acquisition product ID list acquisition request with the matching degree exceeding the threshold value, with the keyword ID as an argument, to the DB management Facade 110 (M1401). Upon receiving this request, the DB management Facade 110 generates a product-keyword class 112 having the keyword ID as an argument, and receives a product-keyword class list as a return value (M1402).

  Next, the DB management Facade 110 issues a matching product determination threshold extraction request to the threshold class 116, and acquires the matching product determination threshold extracted in response to this request (M1403). Then, the DB management Facade 110 acquires, for the deliverables-keyword class 112, an obtained deliverable ID for obtaining a deliverable ID of a deliverable having a fitness degree exceeding the threshold, with the fit artifact determining threshold as an argument. An extraction request is issued (M1404).

  The product-keyword class 112 that has received the acquired product ID extraction request extracts the degree of matching between the designated keyword and the corresponding product (M1405). Subsequently, the deliverable-keyword class 112 extracts the acquisition result of the deliverable corresponding to the designated keyword (M1406). Then, the deliverable-keyword class 112 extracts the deliverable ID of the deliverable having a fitness level that exceeds the fit deliverable determination threshold from the acquired deliverable ID (M1407). The extracted product ID is returned to the DB management Facade 110.

  Next, the DB management Facade 110 generates a product ID list of product IDs received from the product-keyword class 112 (M1408). Then, the DB management Facade 110 repeats the above processing (M1404 to M1408) for the number of generated product-keyword classes, and then returns the product ID list to the Actor 501.

  Next, processing for registering product information in the product-keyword information management table 118 will be described with reference to FIG.

  As illustrated in FIG. 15, the Actor 501 issues a product information registration request with the keyword ID, product ID, fitness, keyword existence number, and acquisition result (= False) as arguments to the DB management Facade 110 (M1501). ). Upon receiving this request, the DB management Facade 110 issues a product information registration request with the keyword ID, product ID, degree of conformity, number of keywords present, and acquisition result (= False) as arguments to the product-keyword class 112. (M1502). Thus, each piece of information passed as an argument is registered in the keyword management table 121.

  Next, processing for updating the acquisition result will be described with reference to FIG.

  As shown in FIG. 16, the Actor 501 issues an acquisition result update request to the DB management Facade 110 with the keyword ID, product ID, and acquisition result (= True) as arguments (M1601). Receiving this request, the DB management Facade 110 generates a product-keyword class 112 having the keyword ID and product ID as arguments (M1602). Next, the DB management Facade 110 issues an acquisition result update request with the acquisition result (= True) as an argument to the generated artifact-keyword class 112 (M1603). Thereby, the acquisition result is updated.

  Next, processing for acquiring a product ID list including a specified keyword will be described with reference to FIG.

  As shown in FIG. 17, the Actor 501 issues a product ID list acquisition request with the keyword ID as an argument to the DB management Facade 110 (M1701). Receiving this request, the DB management Facade 110 generates a product-keyword class 112 having the keyword ID as an argument, and acquires a product-keyword class list including the generated product-keyword class (M1702). Next, the DB management Facade 110 issues a product ID extraction request to the generated product-keyword class 112, and acquires the product ID extracted in response to this request (M1703). Then, the DB management Facade 110 adds the acquired product ID to the product ID list (M1704).

  The DB management Facade 110 repeats the above processes (M1703 and M1704) for the number of the extracted product-keyword objects. Thereafter, the DB management Facade 110 returns the created product ID list to the Actor 501.

  Next, a process for acquiring a keyword ID included in a product will be described with reference to FIG.

  As illustrated in FIG. 18, the Actor 501 issues a keyword ID acquisition request for acquiring a keyword ID included in the product, with the product ID as an argument, to the DB management Facade 110 (M1801). Receiving this request, the DB management Facade 110 generates a product-keyword class 112 having the product ID as an argument, and acquires a product-keyword class list including the product-keyword class 112 (M1802).

  Next, the DB management Facade 110 issues a keyword ID extraction request to the generated product-keyword class 112, and acquires the keyword ID extracted in response to the request (M1803). Subsequently, the DB management Facade 110 adds the acquired keyword ID to the keyword ID list (M1804). Then, the DB management Facade 110 repeats the above processing (M1803 and M1804) for the number of deliverables-keyword classes. After repeating this process, the DB management Facade 110 returns the keyword ID list to the Actor 501.

  Next, a process for acquiring the ID list of deliverables displayed on the system diagram will be described with reference to FIG.

  As shown in FIG. 19, the Actor 501 issues a system diagram display product ID list acquisition request with the keyword ID as an argument to the DB management Facade 110 (M1901). Upon receiving this request, the DB management Facade 110 generates a deliverable-keyword class of an object having a keyword ID as an argument and holding information on the deliverable including the keyword, and acquires a deliverable-keyword class list (M1902). ).

  Next, the DB management Facade 110 issues a request for extracting a product display threshold to the threshold class 116, and acquires the product display threshold extracted in response to the request (M1903). Then, the DB management Facade 110 uses the obtained product display threshold as an argument for the product-keyword class 112, and the degree of matching between the specified keyword and the product exceeds the product display threshold. A product product ID extraction request is issued (M1904).

  The product-keyword class 112 that has received the product ID extraction request extracts the degree of matching between the designated keyword and the product (M1905). The product-keyword class 112 extracts only product IDs whose extracted fitness exceeds the display threshold of the product and returns the extracted product ID to the DB management Fade 110 (M1906).

  The DB management Facade 110 adds the product ID returned from the product-keyword class 112 to the product ID list (M1907). Then, the DB management Facade 110 repeats the above processing (M1904 and M1907) as many times as the number of acquired product-keyword class lists. After repeating this process, the DB management Facade 110 returns the product ID list to the Actor 501.

  Next, processing for acquiring a list of product IDs to be displayed in the search result list will be described with reference to FIG.

  As shown in FIG. 20, the Actor 501 issues a search result list display product ID list acquisition request with the keyword ID as an argument to the DB management Facade 110 (M2001). This is a request for acquiring the ID of a deliverable that is not displayed in the system diagram. Receiving this request, the DB management Facade 110 generates a product-keyword class 112 having the keyword ID as an argument, and acquires a product-keyword class list including this class (M2002).

  Next, the DB management Facade 110 issues a request for extracting the product display threshold to the threshold class 116, and extracts the product display threshold from the value returned in response to this request (M2003). Then, the DB management Facade 110 issues a deliverable product ID list extraction request to the deliverables-keyword class 112 using the deliverable display threshold as an argument and having a fitness level equal to or lower than the deliverable display threshold ( M2004).

  The product-keyword class 112 that has received the product ID list extraction request extracts the degree of matching between the designated keyword and the product (M2005). Then, the deliverable-keyword class 112 compares the extracted matching level with the display threshold of the deliverable, and extracts the deliverable ID of the deliverable that is equal to or lower than the display threshold of the deliverable (M2006). The extracted product ID is returned to the DB management Facade 110.

  The DB management Facade 110 adds the product ID returned from the product-keyword class 112 to the product ID list (M2007). Then, the DB management Facade 110 repeats the above processing (M2004 and M2007) as many times as the number of acquired product-keyword classes. After repeating this process, the DB management Facade 110 returns the product ID list to the Actor 501.

  Next, a related keyword management table 119 for managing related keyword information will be described with reference to FIG. FIG. 21 is a diagram illustrating an example of the related keyword management table 119.

  In the related keyword management table 119, as shown in FIG. 21, a keyword ID 2101, a related keyword 2102 and a degree of relevance 2103 are described. The keyword ID 2101 uniquely identifies a keyword. The related keyword ID 2102 uniquely identifies a related keyword related to the keyword. The degree of relevance 2103 is a numerical value indicating how related the keyword and the related keyword are, and is set to a larger numerical value as the degree of association is higher.

  Next, processing performed on the related keyword class 113 for making the data of the related keyword management table 119 permanent will be described with reference to FIGS. FIG. 22 is a sequence diagram showing a procedure of processing for acquiring a combination of a related keyword whose relevance exceeds a threshold and a relevance. FIG. 23 is a sequence diagram illustrating a procedure of processing for registering the related keyword and the degree of association in the related keyword management table 119. FIG. 24 is a sequence diagram showing a procedure of processing for acquiring a system diagram display related keyword.

  First, a process for acquiring a combination of a related keyword whose relevance exceeds a threshold and a relevance will be described with reference to FIG.

  As shown in FIG. 22, the Actor 501 issues an acquisition request to the DB management Facade 110 for acquiring a combination of a related keyword having a keyword ID as an argument and a related degree exceeding the keyword display threshold and the related degree. (M2201). Receiving this acquisition request, the DB management Facade 110 generates the related keyword class 113 using the keyword ID as an argument, and acquires the related keyword class list (M2202).

  Next, the DB management Facade 110 issues a request for extracting a keyword display threshold to the threshold class 116, and extracts a keyword display threshold returned as a response to this request (M2203). Then, the DB management Facade 110 issues a relevance degree extraction request to the related keyword class 113 and acquires the relevance degree extracted in response to this request (M2204).

  Next, the DB management Facade 110 compares the acquired degree of association with the keyword display threshold, and extracts a keyword ID whose degree of association exceeds the keyword display threshold based on the comparison result (M2205). Then, the DB management Facade 110 adds the extracted related keyword and degree of association to the combination list of the related keyword and degree of association (M2206).

  The DB management Facade 110 repeats the above processing (M2204 to M2206) for the number of generated related keyword classes. After this process is repeated, the DB management Facade 110 returns a combination list of related keywords and degrees of association to the Actor 501.

  Next, processing for registering the related keyword and the related degree in the related keyword management table 119 will be described with reference to FIG.

  As shown in FIG. 23, the Actor 501 issues a related keyword and related degree registration request to the DB management Facade 110 using the keyword ID, the related keyword, and the related degree as arguments (M2301). Receiving this request, the DB management Facade 110 issues a related keyword and related degree registration request to the related keyword class 113 with the keyword ID, the related keyword ID, and the related degree as arguments (M2302). With this request, the related keyword and the degree of association are registered in the related keyword management table 119.

  Next, processing for acquiring related keywords to be displayed in the system diagram will be described with reference to FIG.

  As shown in FIG. 24, the Actor 501 issues a system diagram display related keyword acquisition request with the keyword ID as an argument to the DB management Facade 110 (M2401). Receiving this request, the DB management Facade 110 generates a related keyword class with the keyword ID as an argument, and acquires a related keyword class list (M2402).

  Next, the DB management Facade 110 issues a request for extracting the keyword display threshold to the threshold class 116, and acquires the keyword display threshold extracted in response to this request (M2403). Then, the DB management Facade 110 issues a request to the related keyword class 113 to extract a related keyword whose relevance degree exceeds the keyword display threshold, with the keyword display threshold as an argument (M2404).

  The related keyword class 113 that has received the request extracts the degree of relevance of the keyword included in the related keyword class list (M2405). Then, the related keyword class 113 compares the extracted relevance level with the keyword display threshold value, and extracts the keyword ID of the keyword whose relevance level exceeds the keyword display threshold value based on the comparison result (M2406). . The extracted keyword ID is returned to the DB management Facade 110.

  Upon receiving the keyword ID from the related keyword class 113, the DB management Facade 110 adds the keyword ID to the keyword ID list (M2407). Next, the DB management Facade 110 repeats the above processing (M2404 and M2407) for the number of the related keyword classes 113. Then, after repeating the above processing, the DB management Facade 110 returns the keyword ID list to the Actor 501.

  Next, the threshold management table 122 for managing each threshold value will be described with reference to FIG. FIG. 25 is a diagram illustrating an example of the threshold management table 122.

  As shown in FIG. 25, the threshold management table 122 includes a column 2501 representing the name of each threshold and a column 2502 holding a threshold value corresponding to each threshold name.

  Next, processing performed in the application layer 200 in the Web server program 103 of the deliverable management server 101 according to the operation of the user of the user terminal 102 will be described.

  First, a user interface used when registering a product from the user terminal 102 into the product management server 101 will be described with reference to FIGS. 26 and 27. FIG. FIG. 26 is a diagram illustrating an example of a deliverable registration dialog that the screen display unit 105 provides as a user interface to the user terminal 102. FIG. 27 is a diagram illustrating an example of a registration completion dialog provided by the screen display unit 105 to the user terminal 102.

  When registering a product from the user terminal 102 into the product management server 101, HTML data indicating a product registration dialog is transmitted to the user terminal 102 by the screen display unit 105 of the product management server 101. . The dialog is displayed as an HTML page via the Web browser of the user terminal 102. This deliverable registration dialog includes a field 2601, a registration button 2602, and a cancel button 2603 as shown in FIG. A field 2601 is a field for designating a product to be registered in the product management server 101 and a directory path of the registration destination. A registration button 2602 is a button for instructing an operation for registering a deliverable to the registration destination specified in the field 2601. A cancel button 2603 is a button for canceling the product registration operation.

  When a product registration operation is performed on the product registration dialog, the specified product is registered in the specified registration destination according to the registration operation. When the registration is normally performed, a registration completion dialog for informing the user that the product registration process has been normally performed is displayed on the user terminal 102 as shown in FIG. This registration completion dialog is provided as an HTML page by the screen display unit 105 and is displayed by the Web browser of the user terminal 102.

  The deliverable management server 101 registers the deliverable specified by the registration operation on the user terminal 102, and calculates the degree of conformity indicating how much the keyword and the deliverable are suitable. A process from registration of a product to calculation of a fitness level in the product management server 101 will be described with reference to FIG. FIG. 28 is a sequence diagram illustrating a processing procedure from registration of a product to calculation of a fitness level in the product management server 101.

  As shown in FIG. 28, the user terminal 102 issues a product registration request with the product directory path and the product entity as arguments to the control unit 104 of the product management server 101 (M2801). Upon receiving this request, the control unit 104 issues a product registration request with the product directory path and the product entity as arguments to the product file operation unit 106 (M2802).

  Upon receipt of this request, the deliverable file operation unit 106 stores the substance of the deliverable in the storage location in the hard disk 123 specified by the deliverable directory path (M2803). Next, the deliverable file operation unit 106 cuts out a keyword included in the stored deliverable entity as a morpheme (minimum unit of meaningful words) by morphological analysis, converts it into a standard form, and extracts it (M2804). Then, the deliverable file operation unit 106 calculates the existence number of each extracted keyword in the deliverable entity, and creates a combination list of keywords and the existence quantity (M2805). The combination list of keywords and the number of keywords present is returned to the control unit 104. Here, the deliverable file operation unit 106 repeats the above processing (M2805) for the number of keywords included in the keyword list.

  Next, the control unit 104 issues a registration completion screen display request to the screen display unit 105 (M2806). Upon receiving this request, the screen display unit 105 sends an HTML page constituting a registration completion dialog for notifying the user that the product registration process has been normally performed to the user terminal 102 (M2807).

  Further, the control unit 104 issues a fitness level registration request using the product directory path and the product file operation unit 106 as an argument to the fitness level calculation unit 108. (M2808).

  Upon receiving this request, the fitness level calculation unit 108 issues a request for obtaining the total number of deliverables to the DB management Facade 110, and obtains the total number of deliverables stored in the deliverable management server 101 (M2809). . Next, the fitness level calculation unit 108 issues a product registration request with the product directory path as an argument to the DB management Facade 110 (M2810). Then, the fitness level calculation unit 108 issues a product ID acquisition request with the product directory path as an argument to the DB management Facade 110, and acquires the product ID returned in response to this request (M2811). .

  Next, the matching level calculation unit 108 issues a keyword registration request with the keyword name as an argument to the DB management Facade 110 (M2812). Subsequently, the fitness level calculation unit 108 issues a keyword ID acquisition request with the keyword name as an argument to the DB management Facade 110, and acquires the keyword ID returned in response to this request (M2813). Then, the fitness level calculation unit 108 issues an acquisition request for a product ID list including a keyword having the acquired keyword ID as an argument to the DB management Facade 110, and the product ID returned in response to the request. A list is acquired (M2814). This deliverable ID list is a list showing a list of deliverables including a specified keyword.

  Next, the fitness level calculation unit 108 calculates the number of deliverables included in the obtained deliverable ID list (M2815). Based on the number of existing keywords, the number of deliverables, and the total number of deliverables, the goodness-of-fit calculation unit 108 calculates the suitability representing how much the keywords and deliverables are compatible (M2816). As a method for calculating the fitness, a method used for a general search engine, for example, a method described in Non-Patent Document 1 described above is used.

  Next, the fitness level calculation unit 108 issues a product information registration request with the product ID, keyword ID, fitness level, number of keywords present, and acquisition result (= False) as arguments to the DB management Facade 110 (M2817). . In response to this request, the DB management Facade 110 registers the product information in the product-keyword information management table 118.

  Here, the fitness level calculation unit 108 repeats the above processing (M2812-M2817) for the number of keywords included in the combination list of keywords and the number of keywords present.

  Next, keyword search using keywords input at the user terminal 102 will be described with reference to FIGS. 29 and 30. FIG. FIG. 29 is a diagram showing an example of a keyword search dialog provided by the screen display unit 105 as a user interface to the user terminal. FIG. 30 is a diagram illustrating an example of a search result screen displayed on the user terminal 102.

  When the user performs a keyword search via the user terminal 102, a search dialog as shown in FIG. 29 is displayed on the user terminal 102, for example. This search dialog is transmitted as HTML data by the screen display unit 105, and is displayed as an HTML page by the Web browser of the user terminal 102.

  The search dialog includes a field 2901 and a search button 2902 as shown in FIG. A field 2901 is a field for designating a keyword used for performing a keyword search for the deliverable stored in the deliverable management server 101. A search button 2902 is a button for instructing execution of a keyword search using the keyword specified in the field 2901.

  When performing a keyword search, first, in response to the user operating the user terminal 102 on the search dialog, a keyword is input in the field 2901 and the search button 2902 is pressed. As a result, the keyword is transmitted from the user terminal 102 to the deliverable management server 102. The deliverable management server 101 that has received the keyword starts a keyword search. The search result by this keyword search is sent to the user terminal 102 and displayed as a search result screen as shown in FIG.

  The search result screen is provided as an HTML page by the screen display unit 105 and is displayed by the Web browser of the user terminal 102. This search result screen includes a plurality of fields 3001, 3003, 3004, 3005, 3007, a search button 3002, and a display button 3006, as shown in FIG.

  Here, the field 3001 is a text field for displaying the input keyword. The field 3001 is also used as an input field for a keyword to be redesignated when the user wants to perform a search again.

  A search button 3002 is a button for instructing execution of a keyword search using the keyword input in the field 3001. A field 3003 is a field for displaying a related keyword for the keyword displayed in the field 3001. A field 3004 includes a text field for displaying a keyword display threshold. A field 3005 includes a text field for displaying a display threshold value of the deliverable. A display button 3006 is a button for instructing to recreate and display the system diagram based on the threshold values displayed in the fields 3004 and 3005.

  The threshold values displayed in the fields 3004 and 3005 are previously stored in the memory, and the system diagram is displayed based on the threshold values. This threshold is a value indicating how much a search result having a high degree of relevance with a keyword is displayed. In the present embodiment, as the threshold value is larger, only search results with higher relevance are displayed. However, regarding the threshold value fixed in the system, if the value is too small, the number of related keywords and deliverables displayed in the system diagram becomes enormous, and it may be difficult to see them. If the threshold is too large, an appropriate number of related keywords and deliverables may not be displayed. For this reason, the present embodiment is configured such that each field 3004, 3005 can be used to input a threshold value by a user operation. Thus, when the user inputs an appropriate value and presses the display button 3006, the display in the system diagram is controlled.

  A field 3007 is a field for displaying a system diagram created from an input keyword and a related keyword extracted based on the keyword. A field 3008 is a field for displaying a list of deliverables that are not displayed in the system diagram among the retrieved deliverables.

  Here, in detail, the system diagram displayed in the field 3007 displays a search result obtained by performing a keyword search under a search condition including an input keyword and a related keyword related thereto. Here, the related keyword is extracted as a keyword having high relevance to the input keyword. That is, the related keyword displayed in the system diagram has a degree of relevance exceeding the display threshold of the keyword.

  The system diagram includes a plurality of search results 3009, 3013-3015. This search result 3009 shows a list 3011 of names that can identify deliverables that match the input keyword 3010. The search result 3009 is displayed at the center position of the system diagram. Here, the product that matches the input keyword 3010 has a matching value with the keyword that exceeds the display threshold of the product.

  Each of the search results 3013 to 3015 is a search result for each related keyword with respect to the input keyword, and shows a list of names that can specify the related keyword and the corresponding product. The search results 3013 to 3015 are displayed so as to be arranged around the search result 3009 as a center.

  The name of each deliverable in the list of search results 3009, 3013-3015 is linked to the deliverable. In response to the user of the user terminal 102 clicking a corresponding name with a pointer such as a mouse, the deliverable management server 101 transmits the deliverable to the user terminal 102. As a result, the user can browse and acquire the deliverables. Further, the search result 3009 based on the input keyword is displayed so as to be distinguishable from other search results 3013 to 3015. For example, a display form is used in which the display frame of the search result 3009 is thick and the color of the display frame is changed so that the search result 3009 is based on the keyword input at a glance.

  In addition, the search results 3013 to 3015 based on the related keywords are displayed so as to be connected to the search results 3009 based on the input keywords by a line 3012. By displaying the search results in association with each other in this way, it is clearly indicated that the search results 3013 to 3015 based on the related keywords are related to the search results 3009. In addition, on a line 3012 connecting the search results 3013 to 3015 based on the related keywords and the search results 3009 based on the input keywords, the level of relevance is shown so that the level of relevance between the related keywords and the input keywords can be understood. The value is displayed. The calculation of the degree of association between the related keyword and the input keyword will be described later. Instead, the thickness of each line 3012 that connects the search results 3013 to 3015 by the related keyword and the search result 3009 by the input keyword may be changed according to the degree of relevance, or each line 3012 may be changed. You may make it change the color of.

  Further, as shown in FIG. 30, the display positions of the search results 3013 to 3015 based on the related keywords may be changed with respect to the search results 3009 based on the input keywords according to the degree of relevance. . That is, among the search results 3013 to 3015 by the related keyword, the higher the degree of relevance is displayed at a position closer to the search result 3009 by the input keyword. In this way, the degree of relevance can be indicated using the arrangement interval between the deliverables.

  Also, among the deliverables with the names displayed in the system diagram, for the already obtained deliverables, display the names in different colors to indicate that they have already been obtained. Also good.

  According to the above system diagram, the product whose fitness with the input keyword exceeds the display threshold of the product and the product whose fitness with the related keyword exceeds the display threshold of the product are classified. Is displayed. In addition, the degree of association between keywords for each deliverable is specified. Therefore, it is possible to provide a search result display screen that allows a user to easily perceive related information for an input keyword even when a user performs a search in which knowledge about a search target is scarce and content to be searched is not clear. it can.

  In addition, a product having a matching degree with the input keyword equal to or less than the display threshold of the product is also displayed in 3008, which is an area different from the display area 3007 of the system diagram. As a result, more products can be referred to, and the system diagram is not difficult to see with many products.

  Next, keyword search processing in the present embodiment will be described with reference to FIG. FIG. 31 is a sequence diagram showing the procedure of keyword search processing. Here, in accordance with a user operation, a keyword is input from the search dialog (FIG. 29), a keyword search is performed based on the keyword, and the search result (FIG. 30) is displayed. Will be described.

  When a keyword is input from the search dialog (FIG. 29) by the user and the search button is pressed, the user terminal 102 uses the input keyword as an argument to the control unit 104 as shown in FIG. A search request is issued (M3101). Receiving this request, the control unit 104 issues a search result creation request with the input keyword as an argument to the search result creation unit 109 (M3102).

  Upon receiving the request, the search result creation unit 109 issues a keyword ID acquisition request with the input keyword as an argument to the DB management Facade 110, and acquires the keyword ID returned in response to this request (M3103). ). Next, the search result creation unit 109 makes a request to the DB management Facade 110 to acquire a list of combinations of related keywords and related degrees whose related degrees exceed the keyword determination threshold in the system diagram display, using the acquired keyword ID as an argument. (M3014). In response to this request, the DB management Facade 110 returns a combination list of related keywords and degrees of association to the search result creation unit 109, and the search result creation unit 109 acquires the combination list.

  Next, the search result creation unit 109 issues a system diagram display related keyword ID acquisition request with the keyword ID as an argument to the DB management Facade 110 (M3105). Then, the search result creation unit 109 acquires the keyword ID list returned from the DB management Facade 110 in response to the request. Here, the keyword ID list is a combination of the acquired related keyword ID and the input keyword ID. The search result creation unit 109 issues a system diagram display product acquisition request with the keyword ID as an argument to the DB management Facade 110 (M3106). Then, the search result creation unit 109 returns the product to be displayed in the system diagram returned from the DB management Facade 110 in response to the request (those whose fitness with keywords exceeds the display threshold of the product). Acquire a product ID list.

  Next, the search result creation unit 109 issues a request for acquiring a combination list of a product ID and a product directory path, using the product ID list as an argument, to the DB management Facade 110 (M3107). Then, the search result creation unit 109 acquires the combination list of the product ID and the product directory path to be displayed in the system diagram returned from the DB management Facade 110 in response to the request.

  Next, the search result creation unit 109 issues a keyword name acquisition request with the keyword ID as an argument to the DB management Facade 110 (M3108). Then, the search result creation unit 109 acquires the keyword name returned from the DB management Facade 110 in response to the request.

  Next, the search result creation unit 109 creates system diagram display information for displaying the system diagram based on the combination list of the keyword name, product ID, and product directory path (M3109). Subsequently, the search result creation unit 109 issues a search result list display product acquisition request with the keyword ID as an argument to the DB management Facade 110 (M3110). Then, the search result creation unit 109 acquires a deliverable ID list indicating a list of deliverables displayed as a search result returned in response to the request.

  Next, the search result creation unit 109 issues a request for acquiring a combination list of a product ID and a product directory path using the acquired product list as an argument to the DB management Facade 110 (M3111). Then, the search result creation unit 109 acquires a combination list of the product ID and the product directory path returned from the DB management Facade 110 in response to the request.

  Next, the search result creation unit 109 creates search result display information based on the product ID and product directory path combination list (M3112).

  The search result creation unit 109 repeats the above processing (M3106 to M3112) for the number of keyword IDs of the keywords displayed in the system diagram. This creates information displayed as a system diagram (field 3007) and other search results (field 3008) as shown in FIG.

  After the above processing is repeated by the search result creation unit 109, the control unit 104 issues a search screen display request to the screen display unit 105 (M3113). Upon receiving this request, the screen display unit 105 issues a combination keyword / relationship degree acquisition request to the search result creation unit 109 (M3114). Then, the screen display unit 105 acquires a combination of the related keyword and the degree of relevance returned in response to the request.

  Next, the screen display unit 105 issues a system diagram display information acquisition request to the search result creation unit 109 (M3115). Then, the screen display unit 105 responds to the above request, the keyword displayed in the system diagram returned from the search result creation unit 109, and the degree of matching with the input keyword exceeds the display threshold of the product. Acquire information on the deliverables. This is a product displayed in the area 3007 indicating the system diagram.

  Next, the screen display unit 105 issues a search result display information acquisition request to the search result creation unit 109 (M3116). Then, the screen display unit 105 acquires information on the deliverables to be displayed in the other search results (field 3008) returned from the search result creation unit 109 in response to the request. This product has a keyword displayed in the system diagram and a matching degree with the input keyword is less than or equal to the display threshold of the product. This is a deliverable displayed in the area 3008 indicating other search results.

  Next, the screen display unit 105 creates a search result screen (FIG. 30) based on the acquired product information (information acquired in response to a request from M3115 and M3116) (M3117). Then, the screen display unit 105 issues a search result display request to the user terminal 102 (M3118). Along with this request, the screen display unit 105 sends HTML page information forming a search result screen to the user terminal 102. Receiving this HTML page information, the user terminal 102 displays a search result screen via the Web browser.

  Next, processing in the case of performing a re-search by changing a threshold value or changing or adding a keyword on the search result screen in FIG. 30 will be described with reference to FIG. FIG. 32 is a sequence diagram showing a processing procedure when re-searching is performed in accordance with re-input processing such as threshold value change or keyword change or addition on the search result screen of FIG.

  In the search result screen (FIG. 30), when there is an instruction to change the threshold value and press the display button 3006 by the user's operation, the keyword search is performed again based on the input information. Also, when there is an instruction to change or add a keyword and press the display button 3006, the keyword search is performed again based on the input information. As a result, the search result screen is recreated and displayed. When a related keyword is selected as a keyword as a search condition, a related keyword for the selected related keyword is extracted, and a search for a product that matches these keywords is performed. Then, the search result screen of the deliverable extracted by the search is created again and displayed.

  Thus, when a search condition such as a keyword and / or threshold (keyword display threshold or deliverable display threshold) is changed on the search result screen, the search condition including the changed keyword and / or threshold is changed. A keyword search is performed based on this. This keyword search is executed according to the sequence shown in FIG. The sequence shown in FIG. 32 is the same as the sequence shown in FIG. However, the sequence shown in FIG. 32 differs from the sequence shown in FIG. 31 in that the changed keyword and / or threshold value are used as arguments in the following requests.

  Specifically, the user terminal 102 uses the keyword and / or threshold value (keyword display threshold value or product display threshold value) changed in the search result screen (FIG. 30) as an argument to the control unit 104. A search request is issued (M3201). Upon receiving the request, the control unit 104 issues a search result creation request with the keyword and / or threshold as arguments to the search result creation unit 109 (M3202).

  Each request (M3203 to M3306, M3208, M3210) issued from the search result creation unit 109 to the DB management Facade 110 includes the keyword or threshold value as an argument.

  Further, when the user selects a related keyword, the same sequence as the sequence shown in FIG. 31 is performed. However, the selected keyword is used as an argument of a search request issued from the user terminal 102 to the control unit 104.

  Next, a process for displaying or saving a product obtained by keyword search will be described with reference to FIG. FIG. 33 is a diagram showing an example of a product acquisition dialog displayed on the user terminal 102 of FIG.

  When the name of a desired product is selected from the list of product names displayed on the search result screen (FIG. 30) by an operation by the user of the user terminal 102, the result as shown in FIG. An object acquisition dialog is displayed. This deliverable acquisition dialog is provided as an HTML page by the screen display unit 105 and is displayed by the Web browser of the user terminal 102.

  The deliverable acquisition dialog includes an open button 3301, a save button 3302, and a cancel button 3303. The open button 3301 is a button for referring to the contents of the product having the name selected on the search result screen. When the open button 3301 is pressed, the content management server 101 sends the content of the selected product to the user terminal 102, and the content of the product is displayed by the Web browser.

  The save button 3302 is a button for saving the selected deliverable in the user terminal. When the save button 3302 is pressed, the selected product is downloaded from the product management server 101 to the user terminal 102, and the product is stored in the memory of the user terminal 102. A cancel button 3303 is a button for canceling a process performed by pressing each of the buttons 3301 and 3302.

  Next, the relevance calculation process performed in response to the pressing operation of the save button 3302 on the product acquisition dialog by the user will be described with reference to FIG. FIG. 34 is a sequence diagram showing the relevance calculation processing procedure of the product management server 101 of FIG.

  When the user presses the save button 3302 on the product acquisition dialog, the user terminal 102 provides the keyword, product ID, and product directory path to the control unit 104 as shown in FIG. A product acquisition request is issued (M3401). Here, the keyword passed as an argument is the keyword input in the field 3001 on the search result screen (FIG. 30) displayed on the user terminal 102. The product ID is a product ID of the product selected on the search result screen. The product directory path is the directory path of the selected product.

  Upon receiving the request, the control unit 104 issues a product acquisition request with the product directory path as an argument to the product file operation unit 106 (M3502). Then, the control unit 104 acquires the product entity stored on the product directory path returned from the product file operation unit 106 in response to the request. The substance of the acquired product is transmitted from the control unit 104 to the user terminal 102 (M3403).

  In addition, the control unit 104 issues a relevance level registration request with the keyword name and the product ID as arguments to the relevance level calculation unit 107 (M3404). Receiving this request, the degree-of-association calculation unit 107 issues a keyword ID acquisition request with the keyword name as an argument to the DB management Facade 110 (M3405). Then, the relevance calculation unit 107 acquires the keyword ID returned from the DB management Facade 110 in response to the request (keyword extraction unit).

  Next, the degree-of-association calculation unit 107 issues an acquisition result update request with the acquired keyword ID, product ID, and acquisition result (= true) as arguments to the DB management Facade 110 (M3406). Receiving this request, the DB management Facade 110 associates the keyword ID with the deliverable acquired based on the keyword ID and registers it in the deliverable-keyword information management table 118.

  Next, the degree-of-association calculation unit 107 issues a request to the DB management Facade 110 to acquire a product ID list of acquired products whose matching level exceeds a threshold value using the keyword ID as an argument (M3407). Then, the relevance calculating unit 107 returns the deliverable of the deliverable that has been returned from the DB management Facade 110 in response to the above request and whose degree of relevance with the keyword exceeds the applicable deliverable determination threshold and has already been acquired by the user. Get an ID list.

  Next, the degree-of-association calculation unit 107 issues a keyword ID list acquisition request with the acquired product ID list as an argument to the DB management Facade 110 (M3408). Then, the relevance calculation unit 107 acquires a keyword ID list of keywords included in the product acquired by the user, returned from the DB management Facade 110 in response to the request.

  Next, the degree-of-association calculation unit 107 issues an acquisition product ID list acquisition request with the keyword ID included in the acquired keyword ID list as an argument to the DB management Facade 110 and the degree of matching with the keyword exceeds a threshold value. (M3409). Then, the relevance calculation unit 107 returns the product that has been returned from the DB management Facade 110 in response to the above request and whose degree of matching with the keyword included in the product ID list exceeds the threshold and has already been acquired by the user. The deliverable ID list is acquired.

  Next, the relevance calculation unit 107 calculates the relevance between the keywords based on the first keyword product ID list and the second keyword product ID list (M3410). Specifically, the duplication number of the acquired product having a high matching degree with each keyword in both lists is calculated, and this is set as the degree of association between the keywords. Here, the first keyword product ID list is a product ID list of acquired products having a high degree of matching with the input keyword. The second keyword product ID list is a product ID list of acquired products having a high degree of matching with the keywords included in the acquired products.

  Next, the degree-of-relevance calculation unit 107 issues a related keyword registration request that arguments the keyword ID of the keyword used for the search, the keyword ID acquired in the process M3408, and the degree of association to the DB management Facade 110. (M3411). Receiving this request, the DB management Facade 110 registers or updates the degree of association between keywords in the related keyword management table 119.

  As a method for calculating the relevance, there is a method using Robertson's calculation formula based on an established model, for example, in addition to the method according to the sequence of FIG. Details of this method are described in the document “Robertson, S.E. and Walker, S.“ On relevance weights with little relevance information, ”SIGIR 97, ACM Press, pp. 16-24.

  In the method using Robertson's calculation formula based on this established model, there is no need to limit the timing of registration and processing of deliverables. For example, in the sequence of FIG. 31, each time a keyword search is performed using the input keyword, the input keyword is related to the keyword present in the product obtained by the keyword search using the keyword. It is possible to calculate the degree.

  The object of the present invention is achieved by executing the following processing. That is, a storage medium that records a program code of software that realizes the functions of the above-described embodiments is supplied to a system or apparatus, and a computer (or CPU, MPU, etc.) of the system or apparatus is stored in the storage medium. This is the process of reading the code.

  In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiments, and the program code and the storage medium storing the program code constitute the present invention.

  Moreover, the following can be used as a storage medium for supplying the program code. For example, floppy (registered trademark) disk, hard disk, magneto-optical disk, CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD + RW, magnetic tape, nonvolatile memory card, ROM or the like. Alternatively, the program code may be downloaded via a network.

  Further, the present invention includes a case where the function of the above-described embodiment is realized by executing the program code read by the computer. In addition, an OS (operating system) running on the computer performs part or all of the actual processing based on an instruction of the program code, and the functions of the above-described embodiments are realized by the processing. Is also included.

  Furthermore, a case where the functions of the above-described embodiment are realized by the following processing is also included in the present invention. That is, the program code read from the storage medium is written in a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer. Thereafter, based on the instruction of the program code, the CPU or the like provided in the function expansion board or function expansion unit performs part or all of the actual processing.

It is a block diagram which shows the structure of the product management system by which the search system which concerns on one embodiment of this invention is employ | adopted. It is a figure which shows the object of the product management system of FIG. It is a figure which shows the structural example of the product management table 120 of FIG. FIG. 2 is a sequence diagram illustrating a processing procedure for acquiring the total number of deliverables stored in the deliverable management server 101 in FIG. 1. FIG. 10 is a sequence diagram illustrating a processing procedure for registering a product directory path in a product management table 120. FIG. 10 is a sequence diagram illustrating a procedure of processing for acquiring a product ID from the product management table 120. FIG. 10 is a sequence diagram illustrating a processing procedure for acquiring a deliverable directory path. FIG. 10 is a sequence diagram illustrating a processing procedure for acquiring a combination list of a product ID and a product directory path. It is a figure which shows an example of the keyword management table 121 of FIG. It is a sequence diagram which shows the procedure of the process which acquires a keyword name. FIG. 11 is a sequence diagram illustrating a procedure of processing for registering a keyword name in a keyword management table 121. It is a sequence diagram which shows the procedure of the process which acquires keyword ID. FIG. 10 is a diagram illustrating an example of a deliverable-keyword information management table 118. It is a sequence diagram which shows the procedure of the process which acquires the acquisition product ID list | wrist in which the fitness exceeded the threshold value. FIG. 11 is a sequence diagram illustrating a processing procedure for registering product information in a product-keyword information management table 118. It is a sequence diagram which shows the procedure of the process which updates an acquisition result. It is a sequence diagram which shows the procedure of the process which acquires the product ID list | wrist in which the designated keyword is contained in a product. It is a sequence diagram which shows the procedure of the process which acquires keyword ID contained in a deliverable. It is a sequence diagram which shows the procedure of the process which acquires ID list of the deliverable displayed on a systematic diagram. It is a sequence diagram which shows the procedure of the process which acquires the list | wrist of product ID displayed on a search result list. It is a figure which shows an example of the related keyword management table 119. It is a sequence diagram which shows the procedure of the process which acquires the combination of the related keyword and relevance degree in which the relevance degree exceeded the threshold value. It is a sequence diagram which shows the procedure of the process which registers a related keyword and a related degree in the related keyword management table 119. It is a sequence diagram which shows the procedure of the process which acquires a systematic diagram display related keyword. It is a figure which shows an example of the threshold value management table. It is a figure which shows an example of the deliverable registration dialog which the screen display part 105 provides with respect to the user terminal 102 as a user interface. 5 is a diagram illustrating an example of a registration completion dialog provided by the screen display unit 105 to the user terminal 102. FIG. FIG. 10 is a sequence diagram illustrating a processing procedure from registration of a product to calculation of a fitness level in the product management server 101. It is a figure which shows an example of the keyword search dialog which the screen display part 105 provides as a user interface with respect to a user terminal. It is a figure which shows an example of the search result screen displayed on the user terminal. It is a sequence diagram which shows the procedure of the process of keyword search. FIG. 31 is a sequence diagram illustrating a processing procedure when re-searching is performed by changing a threshold value or changing or adding a keyword on the search result screen of FIG. 30. It is a figure which shows an example of the product acquisition dialog displayed on the user terminal of FIG. It is a sequence diagram which shows the procedure of the relevance degree calculation process of the product management server 101 of FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 101 Product management server 102 User terminal 103 Web server program 104 Control part 105 Screen display part 106 Product file operation part 107 Relevance degree calculation part 108 Conformity degree calculation part 109 Search result creation part 110 DB management Facade
111 DB server 117 deliverable management DB
118 Product-Keyword Information Management Table 119 Related Keyword Management Table 120 Product Management Table 121 Keyword Management Table 122 Threshold Management Table

Claims (13)

A degree of matching with a product associated with each of the keywords extracted from a plurality of products stored in the storage device, at least one related keyword associated with each keyword, and the related keyword; A search system for extracting a product that matches a keyword input as a search condition from a plurality of products stored in the storage device,
Extracting means for extracting a related keyword for the keyword inputted as the search condition and a related degree with the related keyword with reference to the related keyword for each keyword and the related degree with the related keyword;
With reference to the degree of matching with the product for each keyword, the first product and the extraction that match the keyword input as the search condition from the plurality of products stored in the storage device Search means for searching for a second deliverable that matches the related keyword that has been set,
The display which displays the searched said 1st product and said 2nd product with the display form which shows the relevance degree of the keyword with respect to said 1st product, and the related keyword with respect to said 2nd product. And a control means.   The display control means displays the first product and the second product at different positions on the screen, respectively, and a keyword for the first product and a related keyword for the second product. 2. The search system according to claim 1, wherein a display showing a degree of association with the first and second deliverables is displayed in association with the display of the first and second deliverables.   The display control means displays the second product around the first product, and the degree of relevance between the keyword for the first product and the related keyword for the second product is high. 3. The search system according to claim 2, wherein an interval between display positions of the first product and the second product is changed according to the display.   The display control means displays the second product around the first product, and further displays a line connecting the first product and the second product as the first product. The search system according to claim 2, wherein a display form according to a degree of association between a keyword for an object and a related keyword for the second product is displayed.   5. The display control unit according to claim 1, wherein the display control means displays a product having a fitness level greater than a preset threshold value among the first product and the second product. The search system according to one.   The search system according to claim 5, wherein the display control unit displays a product having a fitness level equal to or lower than the threshold value at a position different from a position where the product having a fitness level greater than the threshold value is displayed. .   7. The extraction unit according to claim 1, wherein the extraction unit extracts a related keyword having a higher relevance level than a pre-set relevance level as a related keyword related to the keyword input as the search condition. The search system according to one. Means for holding the degree of matching between the keyword and the deliverable for each keyword;
The search system according to any one of claims 1 to 7, further comprising: a related keyword related to the keyword, and means for holding a degree of association between the keyword and the related keyword. In accordance with a user operation, an input means for re-inputting a related keyword for the second product as a keyword of the search condition,
When the related keyword for the second product is re-input as the keyword of the search condition, the extraction means extracts the related keyword corresponding to the re-input keyword and the degree of association with the related keyword, A third product that matches the re-input keyword and a fourth product that matches the extracted related keyword are extracted by the search means, and the third result is extracted by the display control means. The product and the fourth product are displayed in a display form indicating a degree of association between a keyword for the third product and a related keyword for the fourth product. 9. The search system according to any one of 8. Registration means for registering a product in the storage device;
Extraction means for extracting keywords from the deliverables;
A degree-of-fit calculation means for calculating a degree of match between the keyword extracted from the product and the product;
Relevance calculating means for calculating relevance between the keyword extracted from the deliverable and the related keyword;
When a new product is registered in the storage device by the registration unit, a keyword extracted from the new product, a degree of matching between the keyword extracted from the new product and the product, Update for updating the degree of matching with the product for each keyword and the related keyword for each keyword and the degree of association with the related keyword based on the keyword related to the keyword extracted from the new product and its degree of association The search system according to claim 1, further comprising: means. A search system for extracting deliverables that match a keyword entered as a search condition,
An extracting step of referring to the related keyword for each keyword and the degree of association with the related keyword, and extracting the related keyword for the keyword input as the search condition and the degree of association with the related keyword;
With reference to the degree of matching with the product for each keyword, the first product that matches the keyword input as the search condition is extracted from the plurality of products stored in the storage device. A search process for searching for a second product that matches the related keyword,
The display which displays the searched said 1st product and said 2nd product with the display form which shows the relevance degree of the keyword with respect to said 1st product, and the related keyword with respect to said 2nd product. A search system comprising a control step. A program for causing a computer to execute a search method for extracting a product that matches a keyword input as a search condition,
An extraction step of referring to the related keyword for each keyword and the related degree with the related keyword to extract the related keyword for the keyword input as the search condition and the related degree with the related keyword;
With reference to the degree of matching with the product for each keyword, the first product that matches the keyword input as the search condition is extracted from the plurality of products stored in the storage device. A search step for searching for a second product that matches the related keyword;
The display which displays the searched said 1st product and said 2nd product with the display form which shows the relevance degree of the keyword with respect to said 1st product, and the related keyword with respect to said 2nd product. A program for causing a computer to execute a control step.   A storage medium storing the program according to claim 12 so as to be readable by a computer.

Images