JP6075051B2 - Server apparatus, electronic conference system, and program - Google Patents

Description

  The present invention relates to a server device, an electronic conference system, and a program.

  There is known a technique for retrieving a related document related to a document designated by a user from a document group stored in a database or the like. For example, in Patent Document 1, another document related to the document is searched by calculating a distance representing a difference between a character string included in the log of the document and a character string included in the log of another document. Technology is disclosed.

  However, in the related art as described above, when the titles and contents of the documents in the document group tend to be shared, there is a possibility that related documents as expected by the user may not be searched. For example, if a document group includes meeting materials and minutes of multiple ongoing electronic meetings, such as meeting materials and minutes used in electronic meetings, these electronic documents Only meeting materials and minutes of meetings are searched as related documents.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a server device, an electronic conference system, and a program that can improve search accuracy.

In order to solve the above-described problems and achieve the object, a server device according to an aspect of the present invention includes, for each document related to an electronic conference, one or more information related to the document and one or more information related to the electronic conference. Included in the first similarity determination information for each first similarity determination information and a first generation unit that generates second similarity determination information including information different from the first similarity determination information and the first similarity determination information The degree of coincidence between each piece of information and each piece of information included in the other first similarity determination information is calculated, and the similarity between the document indicated by the first similarity determination information and the document indicated by the other first similarity determination information is calculated. Calculating the degree of coincidence between each piece of information included in the second similarity determination information and each piece of information included in the other second similarity determination information for each second similarity determination information. The document indicated by the information and the other second similarity determination A calculation unit for calculating a similarity between documents represented by the broadcast based on the similarity between documents showing each of the plurality of first similarity determination information, the first correlation information indicating a correlation between the plurality of documents A second generation unit that generates second correlation information indicating a correlation between the plurality of documents based on the similarity between the documents indicated by each of the plurality of second similarity determination information . Retrieval for searching for a related document related to a target document using a correlation information storage unit that stores the first correlation information and the second correlation information, and at least one of the first correlation information and the second correlation information A distribution unit that distributes the retrieved related document .

The electronic conferencing system according to another aspect of the present invention includes, for each document related to an electronic conference, first similarity determination information including one or more information related to the document and one or more information related to the electronic conference , and the first A first generation unit that generates second similarity determination information including information different from the similarity determination information, and each information included in the first similarity determination information and another first similarity determination for each of the first similarity determination information The degree of coincidence with each information included in the information is calculated to calculate the degree of similarity between the document indicated by the first similarity determination information and the document indicated by the other first similarity determination information , and the second similarity determination information Every time, the degree of coincidence between each piece of information included in the second similarity determination information and each piece of information included in the other second similarity determination information is calculated, and the document indicated by the second similarity determination information and the other second Calculate the similarity to the document indicated by the similarity judgment information A calculation unit that, based on the similarity between documents showing each of the plurality of first similarity determination information, and generates a first correlation information indicating a correlation between the plurality of documents, the plurality of second similar A second generation unit configured to generate second correlation information indicating a correlation between the plurality of documents based on a similarity between documents indicated by the determination information; and the generated first correlation information and the second correlation A correlation information storage unit for storing information; a search unit for searching for a related document related to a target document using at least one of the first correlation information and the second correlation information; and the searched related document A distribution unit for distribution .

The program according to another aspect of the present invention provides, for each document related to an electronic conference, first similarity determination information including one or more information related to the document and one or more information related to the electronic conference , and the first similarity determination. A first generation step of generating second similarity determination information including information different from the information, and each information included in the first similarity determination information and other first similarity determination information for each of the first similarity determination information The degree of coincidence with each included information is calculated to calculate the degree of similarity between the document indicated by the first similarity determination information and the document indicated by the other first similarity determination information, and for each second similarity determination information. The degree of coincidence between the information included in the second similarity determination information and the information included in the other second similarity determination information is calculated, and the document indicated by the second similarity determination information and the other second similarity determination Calculate similarity to document indicated by information A calculation step that, based on the similarity between documents showing each of the plurality of first similarity determination information, and generates a first correlation information indicating a correlation between the plurality of documents, the plurality of second similar A second generation step of generating second correlation information indicating a correlation between the plurality of documents based on the similarity between the documents indicated by the determination information; and the generated first correlation information and the second correlation A correlation information storing step for storing information; a search step for searching for a related document related to a target document using at least one of the first correlation information and the second correlation information; and the searched related document The distribution step of distributing is for causing a computer to execute.

  According to the present invention, it is possible to improve the search accuracy.

FIG. 1 is a block diagram illustrating an example of the configuration of the electronic conference system according to the present embodiment. FIG. 2 is a block diagram illustrating an example of the configuration of the server apparatus according to the present embodiment. FIG. 3 is a diagram illustrating an example of a document table according to the present embodiment. FIG. 4 is a diagram showing an example of the full text search table of the present embodiment. FIG. 5 is a diagram illustrating an example of a conference table according to the present embodiment. FIG. 6 is a diagram illustrating an example of a device table according to the present embodiment. FIG. 7 is a diagram illustrating an example of a user table according to the present embodiment. FIG. 8 is a diagram illustrating an example of a department table according to the present embodiment. FIG. 9 is a diagram illustrating an example of the first similarity determination information according to the present embodiment. FIG. 10 is an explanatory diagram of a calculation example of points according to the degree of coincidence of values according to the present embodiment. FIG. 11 is an explanatory diagram of a calculation example of points according to the number of values of the present embodiment. FIG. 12 is a diagram illustrating an example of the first correlation information generation method according to the present embodiment. FIG. 13 is a diagram illustrating an example of the first correlation information generation method according to the present embodiment. FIG. 14 is a diagram illustrating an example of a related document search method using first correlation information according to the present embodiment. FIG. 15 is a diagram illustrating an example of a search history table according to the present embodiment. FIG. 16 is a diagram illustrating an example of the first similarity determination information after the change of the present embodiment. FIG. 17 is a diagram illustrating an example of a related document search method based on the first correlation information after the update according to the present embodiment. FIG. 18 is a diagram illustrating an example of a related document search method after a distance change using the first correlation information according to the present embodiment. FIG. 19 is a block diagram illustrating an example of the configuration of the terminal device according to the present embodiment. FIG. 20 is a flowchart illustrating an example of a flow of correlation information generation and update processing performed in the electronic conference system according to the present embodiment. FIG. 21 is a flowchart illustrating an example of a flow of related document distribution processing performed in the electronic conference system according to the present embodiment. FIG. 22 is a diagram illustrating an example of a keyword table according to a modification. FIG. 23 is a diagram illustrating an example of a hardware configuration of the server device and the terminal device according to the present embodiment and the modification.

  Hereinafter, embodiments of a server device, an electronic conference system, and a program according to the present invention will be described in detail with reference to the accompanying drawings.

  First, the configuration of the electronic conference system of this embodiment will be described.

  FIG. 1 is a block diagram showing an example of the configuration of the electronic conference system 1 of the present embodiment. As shown in FIG. 1, the electronic conference system 1 includes a server device 10, a PC 20-1, a notebook PC 20-2, a tablet 20-3, a smartphone 20-4, and a projector 30.

  The server device 10, the PC 20-1, the notebook PC 20-2, the tablet 20-3, the smartphone 20-4, and the projector 30 are connected via the network 2. The network 2 can be realized by, for example, a LAN (Local Area Network), a VPN (Virtual Private Network), or the Internet.

  The server device 10 manages an electronic conference, and can be realized by a computer installed with an electronic conference program for a server. Note that the server device 10 may be composed of one computer or a plurality of computers. Specifically, the server device 10 performs registration, editing, and deletion of electronic conferences, registration, update, deletion, and distribution of documents (meeting materials and minutes) and information used in electronic conferences. Hereinafter, document distribution will be mainly described.

  The PC 20-1, the notebook PC 20-2, the tablet 20-3, and the smartphone 20-4 are all terminal devices that can be used in an electronic conference, and can be realized by a computer in which an electronic conference program for clients is installed. . In the following description, when there is no need to distinguish between the PC 20-1, the notebook PC 20-2, the tablet 20-3, and the smartphone 20-4, the terminal device 20 may be referred to.

  Specifically, the terminal device 20 is used in an electronic conference registration request, edit request, and deletion request, or an electronic conference, with respect to the server device 10 by establishing communication with the server device 10. Requests for registration and update of documents and information, requests for deletion, and requests for distribution.

  When the projector 30 receives a screen displayed on the terminal device 20 and a projection request for the screen from the terminal device 20, the projector 30 projects the received screen onto a projection medium such as a screen (not shown).

  FIG. 2 is a block diagram illustrating an example of the configuration of the server device 10 according to the present embodiment. As illustrated in FIG. 2, the server device 10 includes a communication unit 110, an operation unit 120, a display unit 130, a storage unit 140, and a control unit 150.

  The communication unit 110 communicates with an external device such as the terminal device 20 via the network 2 and can be realized by a communication device such as a NIC (Network Interface Card).

  The operation unit 120 inputs various operations and can be realized by an input device such as a keyboard, a mouse, a touch pad, and a touch panel.

  The display unit 130 displays various screens and can be realized by a display device such as a liquid crystal display or a touch panel display.

  The storage unit 140 stores various programs such as a server electronic conference program executed by the server apparatus 10 and data used for various processes performed by the server apparatus 10. The storage unit 140 is, for example, a magnetic, optical, or electrical device such as a hard disk drive (HDD), a solid state drive (SSD), a memory card, an optical disk, a read only memory (ROM), and a random access memory (RAM). This can be realized by a storage device that can be stored.

  The storage unit 140 includes a reference information storage unit 141, a similarity determination information storage unit 143, and a correlation information storage unit 145.

  The reference information storage unit 141 includes a document table including document information including one or more information regarding documents used in the electronic conference, a conference table including conference information including one or more information regarding the electronic conference, and the electronic conference. Various tables such as a device table composed of device information including one or more pieces of information related to devices used in the Internet and a user table composed of user information including one or more pieces of information related to users participating in the electronic conference are stored. .

  FIG. 3 is a diagram illustrating an example of a document table according to the present embodiment. In the example illustrated in FIG. 3, the document table includes document information in which a document ID, a document name, a creator, a creation date, full-text search data, a frequent keyword, and a usage history are associated with each other. Yes.

  The document ID is a document identifier. The document name is the title of the document. The creator is the name or account of the user who created the document. The creation date is the date when the document was created. In the full-text search data, a full-text search ID of full-text search information constituting a full-text search table described later is set, and the full-text search information indicated by the full-text search ID is referred to. Frequent keywords are keywords that appear frequently in the full-text search data. In the usage history, a conference ID of conference information constituting a conference table described later and a device ID of device information configuring a device table described later are set, and the conference information indicated by the conference ID and the device information indicated by the device ID are referred to. Is done. The usage history may be a separate table. In each information, when “null” is set, it indicates that the information does not exist. However, the document information is not limited to such information.

  FIG. 4 is a diagram showing an example of the full text search table of the present embodiment. In the example illustrated in FIG. 4, the full-text search table includes full-text search information in which full-text search IDs, document IDs, full-text search data, and keywords are associated with each other.

  The full text search ID is an identifier of full text search data. As the document ID, the document ID of the document information constituting the document table is set. The full-text search data is all text data of the document indicated by the document ID, and is, for example, a character string that has been subjected to OCR (Optical Character Reader) processing, but is not illustrated in the example illustrated in FIG. The keyword is a character string that appears more than a predetermined number of times in the full-text search data. However, the full text search information is not limited to such information.

  Full-text search information is referenced from document information. For example, in the full text search data of document information, [full text search data. 1] is set, the full-text search information whose full-text search ID is “1” is referred to.

  FIG. 5 is a diagram illustrating an example of a conference table according to the present embodiment. In the example illustrated in FIG. 5, the conference table includes a conference ID, a conference name, the number of times, an organizer, a participant, a document, a related link, a date, a conference location, a conference content, an agenda, and the like. The conference information is associated with the frequent keywords.

  The conference ID is a conference identifier. The meeting name is the title of the meeting. The number of times is the number of meetings. The organizer is the name or account of the user who hosts the conference. A participant is the name or account of a meeting attendee other than the organizer. The document is a document related to the conference, and the document ID of the document information constituting the document table is set. The related link is link information such as a URL (Uniform Resource Locator). The date and time is the date and time when the conference is held. The meeting place is a place where a meeting such as a meeting room is held. The content of the conference is the content of the conference, for example, audio data of the conference. The agenda is the conference agenda. A frequent keyword is a character string that appears more than a predetermined number of times in the content of a meeting. In each information, when “null” is set, it indicates that the information does not exist. However, the conference information is not limited to such information. For example, information on devices used in the conference may be included. In this case, the device ID of the device information constituting the device table described later is set for the device used, and the device information indicated by the device ID is referred to.

  The conference information is referenced from the document information. For example, in the usage history of document information, [Conference. 1] is set, the conference information whose conference ID is “1” is referred to.

  FIG. 6 is a diagram illustrating an example of a device table according to the present embodiment. In the example illustrated in FIG. 6, the device table includes device information in which a device ID, a device name, a location, and an IP address are associated with each other.

  The device ID is a device identifier. The device name is the name of the device. The place is the place where the device is installed. The IP address is the IP address of the device. In each information, when “null” is set, it indicates that the information does not exist. However, the device information is not limited to such information.

  The device information is referenced from the document information. For example, in the usage history of document information, [Device. 1] is set, the device information with the device ID “1” is referred to. Similarly, the device information may be referred to from the conference information. For example, in a device using conference information, [device. 1] is set, the device information with the device ID “1” is referred to.

  FIG. 7 is a diagram illustrating an example of a user table according to the present embodiment. In the example illustrated in FIG. 7, the user ID, the user, and the department to which the user belongs are associated with user information. However, the user information is not limited to such information.

  The user ID is a user identifier. A user is a user name or account. The department ID of the department information constituting the department table described later is set for the department to which the department belongs, and the department information indicated by the department ID is referred to.

  FIG. 8 is a diagram illustrating an example of a department table according to the present embodiment. In the example illustrated in FIG. 8, the department table includes department information in which a department ID and a department name are associated with each other. However, the department information is not limited to such information. The department ID is an identifier of the department, and the department name is the name of the department.

  The department information is referred to from the user information. For example, in the department to which the user information belongs, [Department. 2] is set, the department information with the department ID “2” is referred to.

  Details of the similarity determination information storage unit 143 and the correlation information storage unit 145 will be described later.

  The control unit 150 controls each unit of the server device 10 and can be realized by a control device such as a CPU (Central Processing Unit). The control unit 150 includes a registration unit 151, a first generation unit 153, a calculation unit 155, a second generation unit 157, a search unit 159, and a distribution unit 161. Here, the control unit 150 activates (executes) the electronic conference program for the server stored in the storage unit 140, whereby the registration unit 151, the first generation unit 153, the calculation unit 155, and the second generation unit 157. The search unit 159 and the distribution unit 161 are realized as software.

  The registration unit 151 registers information of each table stored in the reference information storage unit 141. Specifically, the registration unit 151 requests registration, editing, and deletion of electronic conferences from the terminal device 20, registration requests, update requests, deletion requests, and distribution of documents and information used in electronic conferences. Registration, update, and deletion of information in each table stored in the reference information storage unit 141 are performed in accordance with the request.

  For example, since a registration request, an edit request, and a deletion request for an electronic conference are made from the terminal device 20 before the conference, the registration unit 151 registers, updates, and deletes the conference information that configures the conference table. Similarly, since registration requests, edit requests, and delete requests for documents and information used in the electronic conference are made from the terminal device 20 before the conference, the registration unit 151 includes the document information and device table constituting the document table. Registration, update, and deletion of the device information constituting the user information and the user information constituting the user table.

  In addition, for example, when a conference is started, device information is transmitted from the terminal device 20 or the projector 30 used in the conference. Therefore, the registration unit 151 selects the used device included in the conference information constituting the conference table. Register or update.

  Further, for example, when a document distribution request is made from the terminal device 20 during a meeting, the distribution unit 161 described later acquires the document from the reference information storage unit 141 and distributes the document to the terminal device 20, so the registration unit 151. Registers or updates the usage history included in the document information constituting the document table, and registers or updates the document included in the conference information constituting the conference table. The document distribution request may be a request for distribution of a document included in the conference information constituting the conference table, or a request for distribution of a document not included in the conference information configuring the conference table. May be.

  For each document related to the electronic conference, the first generation unit 153 generates first similarity determination information including one or more information related to the document and one or more information related to the electronic conference, and stores the first similarity determination information in the similarity determination information storage unit 143. . Specifically, the first generation unit 153 generates first similarity determination information with reference to the information of each table stored in the reference information storage unit 141.

  Here, in the present embodiment, the information about the document is each information of the document information and each information of the full text search information, and the information about the meeting is each information of the meeting information and each information of the device information. However, the present invention is not limited to this. For example, if user information is referred to in document information or meeting information, each information on user information and each information on department information can be included in information on documents and information on meetings.

  FIG. 9 is a diagram illustrating an example of the first similarity determination information according to the present embodiment. In the example shown in FIG. 9, the similarity determination information corresponds to the document ID of the document information, the document name of the document information, the meeting name of the meeting information, the creator of the document information, the participants of the meeting information, and the devices using the meeting information. It is information attached.

  The first generation unit 153 also generates second similarity determination information including information different from the first similarity determination information for each document related to the electronic conference, and stores the second similarity determination information in the similarity determination information storage unit 143. In other words, the first generation unit 153 also generates second similarity determination information including information different from the similarity determination information illustrated in FIG.

  In addition, although the 1st production | generation part 153 shall produce | generate 1st similarity judgment information and 2nd similarity judgment information before a meeting, a production | generation timing is not limited to this. Moreover, if the information of each table memorize | stored in the reference information memory | storage part 141 is updated with the progress of a meeting, the 1st production | generation part 153 will also update 1st similarity judgment information and 2nd similarity judgment information.

  For each first similarity determination information generated by the first generation unit 153, the calculation unit 155 determines the degree of coincidence between each piece of information included in the first similarity determination information and each piece of information included in the other first similarity determination information. And the degree of similarity between the document indicated by the first similarity determination information and the document indicated by the other first similarity determination information is calculated.

  For example, the calculation unit 155 acquires two pieces of first similarity determination information from the similarity determination information storage unit 143. Here, the calculation unit 155 may acquire the two pieces of first similarity determination information in the order of document IDs, may be acquired randomly, or may be acquired according to a predetermined rule. Then, the calculation unit 155 calculates points according to the degree of coincidence of values and the number of values between each piece of information included in each of the acquired two first similarity determination information, and totals the points between the calculated pieces of information. The similarity between the documents indicated by each of the two first similarity determination information is calculated. The calculation unit 155 repeats this process until the similarities of all pairs of the first similarity determination information are calculated.

  FIG. 10 is an explanatory diagram of an example of calculating points according to the degree of coincidence of values according to the present embodiment, and FIG. 11 is an explanatory diagram of an example of calculating points according to the number of values according to the present embodiment. In the example illustrated in FIG. 10, when the information is a keyword, the calculation unit 155 calculates 1 point if the information completely matches, 0.5 points if the information partially matches, and 0 points if the information does not match. Similarly, when the information is ID, the calculation unit 155 calculates 1 point if they match, and 0 points if they do not match. In addition, when the information is a date, the calculation unit 155 calculates 1 point if they match, 0.3 points if they approximate, and 0 points if they do not match. In the example illustrated in FIG. 11, if the number of values is 1, the calculation unit 155 calculates 1 point if they match, and if there are a plurality of values, the calculation unit 155 calculates points (0 to 0) according to the ratio of the number of matched values. 1) is calculated, and if the number of values is 0, 0 points are calculated.

  Although detailed description is omitted, the calculation unit 155 also performs the same process as the first similarity determination information for the second similarity determination information, and the second similarity determination information indicates each second similarity determination information. The degree of similarity between the document and the document indicated by the other second similarity determination information is calculated.

  The second generation unit 157 generates first correlation information indicating the correlation between the plurality of documents based on the similarity between the documents indicated by each of the plurality of first similarity determination information calculated by the calculation unit 155, Stored in the correlation information storage unit 145.

  12 and 13 are diagrams illustrating an example of the first correlation information generation method according to the present embodiment. In the present embodiment, as shown in FIGS. 12 and 13, a diagram like a correlation map is assumed as the first correlation information, but the present invention is not limited to this.

  The second generation unit 157 arranges each document on the correlation map based on the similarity between documents indicated by each of the plurality of first similarity determination information calculated by the calculation unit 155. Specifically, as illustrated in FIG. 12, the second generation unit 157 determines the distance between the documents in the correlation from the similarity between the documents calculated by the calculation unit 155, and responds to the determined distance. These documents are arranged at positions on the correlation map. In this way, the strength of the correlation can be expressed by the similarity.

  For example, when the similarity between the document 1 and the document 2 is 5 pt, the second generation unit 157 arranges the document 1 and the document 2 as illustrated in FIG. Next, since the similarity between the document 1 and the document 3 is 3 pt, the second generation unit 157 determines that the distance between the document 1 and the document 3 is greater than the distance between the document 1 and the document 2 as illustrated in FIG. Document 3 is arranged so as to be far away. At this time, the second generation unit 157 arranges the document 3 in consideration of the similarity between the document 2 and the document 3.

  Further, as illustrated in FIG. 13, the second generation unit 157 determines the direction between the documents in the correlation from the ratio of the degree of coincidence between the information in the similarity between the documents calculated by the calculation unit 155. Then, these documents are arranged at positions on the correlation map corresponding to the determined direction. In this way, the correlation classification can be changed depending on which information has a high degree of coincidence. In this case, the second generation unit 157 performs correlation grouping based on information having a high degree of coincidence, and arranges the documents so as to be grouped by branching from the starting document.

  For example, the number of information is N, the similarity between the document A and the document Z having a high rate of coincidence of the information A, the similarity between the document B and the document Z having a high rate of coincidence of the information B, and the information C Assume that the degree of similarity between the document C and the document Z having a high percentage of coincidence is the same. In this case, the second generation unit 157 arranges the document B at a position at an angle of 360 / N degrees from the document Z with respect to the arrangement position of the document A, and arranges the document Z with respect to the arrangement position of the document B. Document C is placed at a position with a 360 / N degree angle at the starting point.

  If there is a document D having a high degree of coincidence of information B but a low similarity to document Z, the second generation unit 157 arranges document D at a position farther in the arrangement direction of document B. . Furthermore, if there is a document E or F with a high degree of coincidence of information B but a low degree of similarity with document Z and a degree of similarity or coincidence with document B different from document D, the second generation unit Reference numeral 157 denotes the arrangement direction (branch) of the document B, but the documents E and F are arranged at different positions (distances) and distances from the document B.

  Although detailed description is omitted, the second generation unit 157 correlates the plurality of documents based on the similarity between documents indicated by the plurality of second similarity determination information calculated by the calculation unit 155. Is generated and stored in the correlation information storage unit 145. The method for generating the second correlation information is the same as the method for generating the first correlation information.

  When a request for distribution of the related document of the target document is made together with the document ID of the target document currently being browsed from the terminal device 20 during the meeting, the search unit 159 stores the first stored in the correlation information storage unit 145. A related document related to the target document is searched using at least one of the correlation information and the second correlation information.

  Specifically, the search unit 159 identifies the position of the target document on the correlation information of at least one of the first correlation information and the second correlation information, and is located at a certain distance around the identified position (constant correlation). Documents having a relationship) are acquired as related documents from the reference information storage unit 141 or the Web.

  When a plurality of pieces of correlation information are stored in the correlation information storage unit 145 such as the first correlation information and the second correlation information, the search unit 159 may use the latest correlation information for the initial search. .

  FIG. 14 is a diagram illustrating an example of a related document search method using first correlation information according to the present embodiment. In the example shown in FIG. 14, document 1 is the target document, and document 2, document 3, and document 8 located at a fixed distance from the position of document 1 are related documents.

  Note that the related document searched by the search unit 159 is not what the user expects, and the terminal device 20 may request a redistribution of the related document. In this case, the search unit 159 may re-search related documents using the correlation information specified in the re-distribution request, or re-search related documents using correlation information in a predetermined order. May be.

  The distribution unit 161 distributes the related document searched by the search unit 159 and the related document re-searched to the terminal device 20. Specifically, the distribution unit 161 distributes the search result of the search unit 159, the re-search result of the search unit 159, and the searched related document to the terminal device 20.

  Here, correction of correlation information will be described. When the related document searched using the correlation information by the search unit 159 is not what the user expects, the reliability of the correlation is low. For this reason, when the re-search ratio, which is the number of searches until the related document is expected by the user, satisfies the predetermined condition, the correlation information is corrected.

  Therefore, the registration unit 151 registers the search history information in the search history table stored in the correlation information storage unit 145 when the search by the search unit 159 is performed.

  FIG. 15 is a diagram illustrating an example of a search history table according to the present embodiment. In the example shown in FIG. 15, the search history ID, the document, the number of re-searches, the search route, and the re-search ratio are associated with search history information. However, the search history information is not limited to such information.

  The search history ID is an identifier of the search history. The document indicates a combination of the target document and the target related document. The number of re-searches is the number of re-searches until the target related document is searched. The search route is the order of correlation information used for the search. The re-search ratio is a re-search ratio until a target related document is searched.

  For example, in the example shown in FIG. 14, when the document 1 is the target document, the document 4 and the document 20 are not related documents, but the document 4 and the document 20 are related documents expected by the user. . In this case, in the search history information shown in FIG. 15, the re-search ratio is 90% in order to make the document 4 or the document 20 a related document of the document 1, so that the correlation information is corrected. Note that any method described below may be used to correct the correlation information.

  For example, the first generation unit 153 changes the information included in the similarity determination information of the correlation information generation source used for the initial search, the calculation unit 155 calculates the similarity from the changed similarity determination information, The 2 generation unit 157 updates the correlation information based on the similarity between documents indicated by the changed similarity determination information.

  FIG. 16 is a diagram illustrating an example of the first similarity determination information after the change of the present embodiment. In the example shown in FIG. 16, the first similarity determination information includes the document ID of the document information, the document name of the document information, the meeting name of the meeting information, the creator of the document information, the participants of the meeting information, and the device using the meeting information. In addition, it is information in which frequent keywords of document information are associated.

  Here, the reason why the frequent keywords of the document information are added to the first similarity determination information is that the frequent keywords of the document 1, the document 4, and the document 20 are completely matched, and the correlation is expected to be high. Because.

  FIG. 17 is a diagram illustrating an example of a related document search method based on the first correlation information after the update according to the present embodiment. In the example shown in FIG. 17, the document 2, the document 3, the document 4, the document 8, and the document 20 are related documents with respect to the document 1 that is the target document.

  Further, for example, the search unit 159 may change the distance to be a related document in the correlation information used for the initial search. FIG. 18 is a diagram illustrating an example of a related document search method after a distance change using the first correlation information according to the present embodiment. In the example illustrated in FIG. 18, the distance to be the related document is twice that in the example illustrated in FIG. 14, and the document 4 and the document 20 are the related documents with respect to the document 1 that is the target document.

  Further, for example, the calculation unit 155 may perform weighting when calculating the similarity from the similarity determination information of the correlation information generation source used for the initial search. For example, when generating correlation information for which it is determined that the degree of association is higher when the document names match, the calculation unit 155 calculates the degree of coincidence assuming that the document name has twice the similarity to other information. That's fine.

  For example, the search unit 159 may change the reference order of the correlation information. For example, the search unit 159 may re-search related documents preferentially for correlation information having a high re-search frequency.

  The correlation information may be corrected based on an instruction from the terminal device 20.

  FIG. 19 is a block diagram illustrating an example of the configuration of the terminal device 20 of the present embodiment. As illustrated in FIG. 19, the terminal device 20 includes a communication unit 210, an operation unit 220, a display unit 230, a storage unit 240, and a control unit 250.

  The communication unit 210 communicates with an external device such as the server device 10 via the network 2 and can be realized by a communication device such as a NIC.

  The operation unit 220 inputs various operations and can be realized by an input device such as a keyboard, a mouse, a touch pad, and a touch panel.

  The display unit 230 displays various screens and can be realized by a display device such as a liquid crystal display or a touch panel display.

  The storage unit 240 stores various programs such as a terminal electronic conference program executed by the terminal device 20 and data used for various processes performed by the terminal device 20. The storage unit 240 can be realized by a storage device that can store magnetically, optically, or electrically, such as an HDD, an SSD, a memory card, an optical disk, a ROM, and a RAM.

  The control unit 250 controls each unit of the terminal device 20 and can be realized by a control device such as a CPU. The control unit 250 makes a registration request, an edit request, and a deletion request for an electronic conference, a registration request, an update request, a deletion request, and a distribution request for documents and information used in the electronic conference to the server device 10. .

  The control unit 250 makes a distribution request or redistribution request for the related document of the target document currently being browsed, and receives the search result or research result of the related document from the server device 10. The related document search result and re-search result need not be a correlation map as shown in FIG. 14, but may be a list or the like.

  The control unit 250 displays the search result and the re-search result of the related document on the display unit 230. Alternatively, the control unit 250 may generate a view from the search result and the re-search result and display the view on the display unit 230. The search result or the re-search result generated as a view in (1) may be displayed on the display unit 230 as it is.

  Further, when requesting the server device 10 to redistribute the related document, the control unit 250 may specify correlation information used for re-searching the related document, or may request only re-search of the related document, The search method may be left to the server device 10, a change in the search range of related documents may be designated, or a combination of these may be requested to redistribute related documents.

  Next, the operation of the electronic conference system of this embodiment will be described.

  FIG. 20 is a flowchart illustrating an example of a flow of correlation information generation and update processing performed in the electronic conference system 1 according to the present embodiment.

  First, when information is registered in each table stored in the reference information storage unit 141, the first generation unit 153 generates similarity determination information for each document, and the calculation unit 155 performs the first similarity determination information for each document. The degree of coincidence between the information included in the similarity determination information and the information included in the other similarity determination information is calculated, and the similarity between the document indicated by the similarity determination information and the document indicated by the other similarity determination information The second generation unit 157 generates correlation information indicating the correlation between the plurality of documents based on the similarity between the documents indicated by the plurality of similarity determination information (step S100).

  Thereafter, the first generation unit 153, the calculation unit 155, and the second generation unit 157 update the correlation information when the correlation information update condition is satisfied (Yes in step S102) (step S104).

  Here, the correlation information update condition is satisfied, for example, when information is newly registered in each table stored in the reference information storage unit 141 or existing information is updated. In this case, the 1st production | generation part 153, the calculation part 155, and the 2nd production | generation part 157 update correlation information in the procedure similar to step S100.

  The correlation information update condition is also satisfied when the re-search ratio, which is the number of searches until the related document is expected by the user, satisfies a predetermined condition.

  In this case, the first generation unit 153 changes the information included in the similarity determination information of the correlation information generation source used for the initial search, the calculation unit 155 calculates the similarity from the changed similarity determination information, The second generation unit 157 may update the correlation information based on the similarity between documents indicated by the changed similarity determination information.

  In addition, the search unit 159 may update the correlation information by changing the distance to be a related document in the correlation information used for the initial search.

  Further, the calculation unit 155 performs weighting when calculating the similarity from the similarity determination information of the correlation information generation source used for the initial search, and the second generation unit 157 is based on the weighted similarity between documents. Then, the correlation information may be updated.

  FIG. 21 is a flowchart illustrating an example of a flow of related document distribution processing performed in the electronic conference system 1 according to the present embodiment.

  First, in the search unit 159 of the server device 10, the terminal device 20 receives a distribution request for the related document of the target document together with the document ID of the target document (step S200).

  Subsequently, the search unit 159 searches for related documents of the target document using the correlation information stored in the correlation information storage unit 145 (step S202). If there is no related document (No in step S204), the process ends.

  When the related document exists (Yes in step S204), the distribution unit 161 distributes the related document and the search result of the related document to the terminal device 20 (step S206).

  Subsequently, when making a redistribution request for a related document, the control unit 250 of the terminal device 20 sends a redistribution request for the related document of the target document to the server device 10 together with the document ID of the target document (Yes in step S208). . The related document re-distribution request is made when, for example, the user of the terminal device 20 confirms the distributed related document and the search result of the related document, and the related document expected by the user is not distributed. Is called.

  On the other hand, the control unit 250 of the terminal device 20 displays the related document that is the user's purpose on the display unit 230 when the related document re-distribution request is not made (No in Step S208) (Step S210).

  As described above, in the present embodiment, similarity determination information including information regarding documents and information regarding meetings is generated for each document, and the similarity determination information is based on the degree of coincidence between pieces of information between the similarity determination information. Similarity is calculated, and correlation information indicating a correlation between documents is generated based on the calculated similarity. Therefore, according to the present embodiment, since the information regarding the correlation information is taken into account, the search accuracy can be further improved by using the correlation information for the search.

  In the present embodiment, correlation information is generated in advance before searching for related documents. Therefore, according to the present embodiment, the search speed can be improved by searching related documents using the correlation information. In particular, the search speed can be improved as the number of documents to be searched increases. However, this embodiment does not exclude the generation of correlation information when searching for related documents. If sufficient performance can be expected from the number of documents to be searched and the performance of the server device 10, correlation information may be generated when searching for related documents.

  In the present embodiment, when the related document searched using the correlation information is not what the user expects, the correlation information is corrected when a predetermined condition is satisfied. For this reason, according to the present embodiment, it is possible to prevent the search accuracy of the search using the correlation information from deteriorating and maintain high search accuracy.

(Modification)
In addition, this invention is not limited to the said embodiment, A various deformation | transformation is possible. In the above embodiment, the direction between documents is determined from the ratio of the degree of coincidence between pieces of information in the similarity between documents, and these documents are arranged at positions on the correlation map according to the determined direction. An example of generating correlation information has been described.

  In this case, if the number of documents stored in the reference information storage unit 141 is large, it is assumed that it takes a long time to generate correlations for the first time, and to update the correlations when adding and correcting documents. There may be noise in the correlation. For this reason, the second generation unit 157 generates correlation information using a method in which some information is used as a key to perform gentle grouping in advance and integrate the results of generating correlations within the group. Also good. In this way, it can be expected to increase the speed of correlation information generation, decrease the processing amount, and increase the accuracy of correlation similarity.

  For example, the second generation unit 157 uses the keyword table shown in FIG. 22 to rank the appearance frequency of keywords in the entire document, and selects keywords from the top to the rank in which all documents are classified. Grouping. The keyword table shown in FIG. 22 can be generated by the registration unit 151 analyzing the full text data of each document and extracting frequent keywords for each document.

(Hardware configuration)
FIG. 23 is a diagram illustrating an example of a hardware configuration of the server device 10 and the terminal device 20 according to the present embodiment and the modification. The server device 10 and the terminal device 20 of the present embodiment and the modification include a control device 810 such as a CPU, a storage device 820 such as a ROM and a RAM, an external storage device 830 such as an HDD, and a display device 840 such as a display. And an input device 850 such as a keyboard and a mouse, and a communication device 860 such as a communication interface, and has a hardware configuration using a normal computer.

  The programs executed in the server device 10 and the terminal device 20 according to the present embodiment and the modification are files in an installable format or an executable format, and are CD-ROM, CD-R, memory card, DVD (Digital Versatile Disk). The program is stored in a computer-readable storage medium such as a flexible disk (FD).

  Further, the program executed by the server device 10 and the terminal device 20 of the present embodiment and the modification may be provided by being stored on a computer connected to a network such as the Internet and downloaded via the network. Good. In addition, the program executed by the server device 10 and the terminal device 20 according to the present embodiment and the modification may be provided or distributed via a network such as the Internet. In addition, the program executed by the server device 10 and the terminal device 20 according to the present embodiment and the modification may be provided by being incorporated in advance in a ROM or the like.

  The program executed by the server device 10 and the terminal device 20 according to the present embodiment and the modified example has a module configuration for realizing the above-described units on a computer. As actual hardware, the CPU reads out a program from the HDD to the RAM and executes the program, whereby the above-described units are realized on the computer.

1 Electronic Conference System 2 Network 10 Server Device 20 Terminal Device 20-1 PC
20-2 Notebook PC
20-3 tablet 20-4 smartphone 30 projector 110 communication unit 120 operation unit 130 display unit 140 storage unit 141 reference information storage unit 143 similarity determination information storage unit 145 correlation information storage unit 150 control unit 151 registration unit 153 first generation unit 155 Calculation unit 157 Second generation unit 159 Search unit 161 Distribution unit 210 Communication unit 220 Operation unit 230 Display unit 240 Storage unit 250 Control unit 810 Control device 820 Storage device 830 External storage device 840 Display device 850 Input device 860 Communication device

JP2011-59920A

Claims (12)

For each document related to the electronic conference, one or more information about the document, first similarity determination information including one or more information about the electronic conference , and second similarity determination information including information different from the first similarity determination information A first generation unit for generating;
For each of the first similarity determination information, a document indicated by the first similarity determination information by calculating the degree of coincidence between each information included in the first similarity determination information and each information included in the other first similarity determination information And the degree of similarity between the document indicated by the other first similarity determination information and each information included in the second similarity determination information and other second similarity determination information for each second similarity determination information. A calculation unit that calculates the degree of coincidence with each piece of information included and calculates the degree of similarity between the document indicated by the second similarity determination information and the document indicated by the other second similarity determination information ;
Based on the similarity between documents indicated by each of the plurality of first similarity determination information, first correlation information indicating a correlation between the plurality of documents is generated , and each of the plurality of second similarity determination information indicates A second generation unit that generates second correlation information indicating a correlation between the plurality of documents based on a similarity between documents ;
A correlation information storage unit for storing the generated first correlation information and the second correlation information;
A search unit that searches for a related document related to a target document using at least one of the first correlation information and the second correlation information;
A distribution unit that distributes the retrieved related documents;
A server device comprising:   2. The server device according to claim 1, wherein the second generation unit further generates the first correlation information based on a rate of coincidence between the pieces of information in the similarity.   The second generation unit determines a distance between the documents in the correlation based on the similarity, and a direction between the documents in the correlation based on a rate of coincidence between the pieces of information in the similarity. The server device according to claim 2, which determines The said search part uses the newest correlation information for an initial search, when searching the said related document using the some correlation information memorize | stored in the said correlation information storage part . 4. The server device according to any one of 3 . The search unit according to any one of claims 1 to 4, wherein when the related document is searched using a plurality of correlation information stored in the correlation information storage unit, the correlation information specified by the user is used for the re-search. The server apparatus as described in any one . The first generation unit changes information included in a plurality of similarity determination information that is a generation source of correlation information used for an initial search,
The calculation unit calculates, for each similarity determination information, the degree of coincidence between each piece of information included in the similarity determination information and each piece of information included in other similarity determination information, and the document indicated by the similarity determination information and the other Further calculate the similarity with the document indicated by the similarity determination information,
The server according to claim 1 , wherein the second generation unit updates correlation information used for the initial search based on a similarity between documents indicated by each of the plurality of similarity determination information. apparatus. The calculation unit weights the degree of coincidence between each piece of information included in the similarity determination information and each piece of information included in other similarity determination information for each piece of similarity determination information that is a source of correlation information used for the initial search. To calculate the similarity between the document indicated by the similarity determination information and the document indicated by the other similarity determination information,
The server according to claim 1 , wherein the second generation unit updates correlation information used for the initial search based on a similarity between documents indicated by each of the plurality of similarity determination information. apparatus. The server device according to any one of claims 1 to 7 , wherein the search unit changes a range to be a related document in the correlation information used for the initial search, and searches for the related document. The server device according to any one of claims 1 to 8 , wherein the search unit changes the correlation information used for the initial search to correlation information having a high re-search frequency, and searches for related documents. A reference information storage unit that stores a document table including document information including one or more information about the document and a meeting table including meeting information including one or more information about the electronic conference;
The first generation unit, based on the plurality of document information and the plurality of conference information, the server apparatus according to any one of claims 1-9 for generating said plurality of similarity determination information. For each document related to the electronic conference, one or more information about the document, first similarity determination information including one or more information about the electronic conference , and second similarity determination information including information different from the first similarity determination information A first generation unit for generating;
For each of the first similarity determination information, a document indicated by the first similarity determination information by calculating the degree of coincidence between each information included in the first similarity determination information and each information included in the other first similarity determination information And the degree of similarity between the document indicated by the other first similarity determination information and each information included in the second similarity determination information and other second similarity determination information for each second similarity determination information. A calculation unit that calculates the degree of coincidence with each piece of information included and calculates the degree of similarity between the document indicated by the second similarity determination information and the document indicated by the other second similarity determination information ;
Based on the similarity between documents indicated by each of the plurality of first similarity determination information, first correlation information indicating a correlation between the plurality of documents is generated , and each of the plurality of second similarity determination information indicates A second generation unit that generates second correlation information indicating a correlation between the plurality of documents based on a similarity between documents ;
A correlation information storage unit for storing the generated first correlation information and the second correlation information;
A search unit that searches for a related document related to a target document using at least one of the first correlation information and the second correlation information;
A distribution unit that distributes the retrieved related documents;
An electronic conference system comprising: For each document related to the electronic conference, one or more information about the document, first similarity determination information including one or more information about the electronic conference , and second similarity determination information including information different from the first similarity determination information A first generating step for generating;
For each of the first similarity determination information, a document indicated by the first similarity determination information by calculating the degree of coincidence between each information included in the first similarity determination information and each information included in the other first similarity determination information And the degree of similarity between the document indicated by the other first similarity determination information and each information included in the second similarity determination information and other second similarity determination information for each second similarity determination information. A calculation step of calculating the degree of coincidence with each piece of information included to calculate the degree of similarity between the document indicated by the second similarity determination information and the document indicated by the other second similarity determination information ;
Based on the similarity between documents indicated by each of the plurality of first similarity determination information, first correlation information indicating a correlation between the plurality of documents is generated , and each of the plurality of second similarity determination information indicates A second generation step of generating second correlation information indicating a correlation between the plurality of documents based on the similarity between the documents ;
A correlation information storing step for storing the generated first correlation information and the second correlation information;
A search step of searching for a related document related to a target document using at least one of the first correlation information and the second correlation information;
A distribution step of distributing the retrieved related document;
A program that causes a computer to execute.

Images