JP2013003763A - Information recommendation device, information recommendation method and information recommendation program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To recommend appropriate information to a user.SOLUTION: The information recommendation device according to an embodiment includes: a communication history acquisition unit for acquiring communication history among users; a group formation unit for forming a group of users on the basis of relation associated with communication between a predetermined user and other users, which can be acquired from the communication history; a browsing history acquisition unit for acquiring browsing history that is history of accesses to information on a network by other users; a recommendation information determination unit for, on the basis of the browsing history, determining recommendation information that is information accessed by other users belonging to a group formed by the group formation unit; and a recommendation information provision unit for providing the recommendation information determined by the recommendation information determination unit to the predetermined user.

Description

  Embodiments described herein relate generally to an information recommendation device, an information recommendation method, and an information recommendation program.

  In recent years, an enormous amount of information (electronic data) has been accumulated on a network with the speeding up and cost reduction of information processing technology and the spread of network technology such as Internet technology. This situation makes it difficult for a user to obtain desired information from an enormous amount of information.

  For this reason, an information recommendation technique for recommending various types of information represented by collaborative filtering has been proposed. The information recommendation technique is a technique for presenting information desired by a user by using a user's action history. In a typical information recommendation technique, information browsed by another user having a close taste is recommended based on the information browsing history of the user.

Badrul Sarwar et al, “Item-Based Collaborative Filtering Recommendation Algorithms”, GroupLens Research Group / Army HPC Research Center Department of Computer Science and Engineering University of Minnesota, Minneapolis, MN 55455, WWW10, May 2001 Miha Grcar et al, “Data Sparsity Issues in the Collaborative Filtering Framework”, Advances in Web Mining and Web Usage Analysis, Vol.4198, 2006 Aaron Clauset et al, “Finding Community Structure in Very Large Networks”, Physical Review E. Vol.70, No.6, 2004

  However, since the conventional information recommendation technique uses a browsing history of information by the user, it is difficult to recommend appropriate information unless the user has browsed a predetermined number of information.

  The present invention has been made in view of the above, and an object thereof is to provide an information recommendation device, an information recommendation method, and an information recommendation program that can recommend appropriate information to a user.

  The information recommendation device according to the embodiment includes a communication history acquisition unit that acquires a communication history between users, and a group of users based on a relationship related to communication between a predetermined user and another user obtained from the communication history. Formed by the group forming unit based on the browsing history, a grouping unit that forms a group, a browsing history acquisition unit that acquires a browsing history that is a history of accessing information on the network by other users A recommendation information determination unit that determines recommendation information that is information accessed by other users belonging to the group, and a recommendation information provision unit that provides the recommended information determined by the recommendation information determination unit to a predetermined user.

  The information recommendation device, the information recommendation method, and the information recommendation program according to the embodiment have an effect that it is possible to recommend appropriate information to the user.

FIG. 1 is a diagram illustrating a configuration example of a network including an information recommendation device according to the first embodiment. FIG. 2 is a block diagram illustrating a configuration example of the information recommendation device according to the first embodiment. FIG. 3 is a diagram illustrating an example of a communication history held by the communication history holding server according to the first embodiment. FIG. 4 is a diagram illustrating an example of group formation processing by the group formation unit in the first embodiment. FIG. 5 is a diagram illustrating an example of a browsing history held by the browsing history holding server according to the first embodiment. FIG. 6 is a diagram illustrating an example of recommendation information determination processing by the recommendation information determination unit according to the first embodiment. FIG. 7 is a flowchart illustrating the information recommendation processing procedure performed by the information recommendation apparatus according to the first embodiment. FIG. 8 is a diagram for explaining another embodiment of the information recommendation apparatus according to the first embodiment. FIG. 9 is a diagram for explaining another embodiment of the information recommendation apparatus according to the first embodiment. FIG. 10 is a diagram for explaining another embodiment of the information recommendation apparatus according to the first embodiment. FIG. 11 is a diagram for explaining another embodiment of the information recommendation device according to the first embodiment. FIG. 12 is a diagram for explaining another embodiment of the information recommendation apparatus according to the first embodiment. FIG. 13 is a diagram for explaining another embodiment of the information recommendation apparatus according to the first embodiment. FIG. 14 is a diagram illustrating a computer that executes an information recommendation program.

  Embodiments of an information recommendation device, an information recommendation method, and an information recommendation program according to the present invention will be described below in detail with reference to the drawings. In addition, this invention is not limited by this Example.

[Network configuration]
First, a configuration example of a network including an information recommendation apparatus according to the first embodiment will be described with reference to FIG. FIG. 1 is a diagram illustrating a configuration example of a network including an information recommendation device according to the first embodiment. The network 1 illustrated in FIG. 1 is an IP (Internet Protocol) network, for example, and includes user terminals 11 and 12, a communication history holding server 21, a browsing history holding server 22, a recommended information holding server 23, information And a recommendation device 100.

  The user terminals 11 and 12 are information processing apparatuses used by users who use the network 1 in the first embodiment, and are, for example, a PC (Personal Computer), a mobile phone, a PDA (Personal Digital Assistant), or the like.

  These user terminals 11 and 12 may access predetermined files or the like by Web contents such as Web pages and moving image data provided by various servers included in the network 1, FTP (File Transfer Protocol), or the like. . That is, the user terminal 11 or the user terminal 12 accesses a predetermined site (Web page or moving image data) provided by the Web server or a predetermined server provided by an FTP server or the like according to the user's operation content. Or access a file. Thereby, the user of the user terminal 11 or 12 can browse a Web page, moving image data, a file, and the like. In the example illustrated in FIG. 1, a recommended information holding server 23 described later is a Web server, an FTP server, or the like, and the user terminals 11 and 12 access Web content, files, and the like held in the recommended information holding server 23. Shall. In the following, electronic data such as Web pages, moving image data, and files browsed by the user may be simply referred to as “information”.

  The user terminals 11 and 12 may communicate with each other. For example, the user terminals 11 and 12 send / receive e-mails to / from each other via a mail server or base station (not shown), send / receive character data via a chat server, and realize a call between users. To send and receive voice data. In the first embodiment, for the communication performed between the user terminal 11 and the user terminal 12, the content of one user terminal published on a blog or SNS (Social Network Service) Includes a mode in which a user of the other user terminal browses, a mode in which a comment is written on such content, and the like.

  Although FIG. 1 shows an example in which the network 1 includes two user terminals 11 and 12, the number of user terminals included in the network 1 is not limited to this. For example, the network 1 may include three or more user terminals.

  The communication history holding server 21 holds a history of communication between the user terminal 11 and the user terminal 12 (hereinafter sometimes referred to as “communication history”). For example, the communication history holding server 21 acquires the communication history of the user terminal 11 and the like from various servers that relay communication between user terminals. For example, the communication history holding server 21 acquires a communication history from the user terminal 11 or the like. The communication history held by the communication history holding server 21 will be described later.

  The browsing history holding server 22 holds a browsing history of various information by the user of the user terminal 11 or the user terminal 12 (hereinafter may be referred to as “browsing history”). For example, the browsing history holding server 22 acquires an access history of the user terminal 11 or the like as a browsing history from various servers accessed by the user terminal 11 or the like. For example, the browsing history holding server 22 acquires the access history to various servers as the browsing history from the user terminal 11 or the like. The browsing history held by the browsing history holding server 22 will be described later.

  The recommendation information holding server 23 is, for example, a Web server or an FTP server, and holds Web content, files, and the like. That is, the recommendation information holding server 23 holds various electronic data accessed by the user terminal 11 and the user terminal 12. The browsing history holding server 22 collects browsing history from the recommended information holding server 23.

  In the example illustrated in FIG. 1, an example in which one recommended information holding server 23 is included in the network 1 is illustrated, but the number of recommended information holding servers included in the network 1 is not limited thereto. For example, the network 1 may include a plurality of recommended information holding servers that are Web servers and a plurality of recommended information holding servers that are FTP servers.

  Further, the recommended information holding server 23 illustrated in FIG. 1 may not be a Web server or an FTP server. For example, the recommendation information holding server 23 may be a server that holds information accessed by the user terminal 11 or the user terminal 12 from Web content or a file.

  The information recommendation apparatus 100 provides the user terminal 11 and the user terminal 12 with information (electronic data such as Web content) that recommends browsing. For example, the information recommendation apparatus 100 illustrated in FIG. 1 acquires a communication history from the communication history holding server 21 when receiving an information recommendation request for requesting information recommendation from the user terminal 11. And the information recommendation apparatus 100 forms the group of the user terminal group which has communicated in the past based on the communication history acquired from the communication history holding server 21. For example, when the user terminal 11 and the user terminal 12 have communicated, the information recommendation apparatus 100 forms a group including the user terminal 11 and the user terminal 12.

  Subsequently, the information recommendation apparatus 100 acquires a browsing history from the browsing history holding server 22. And the information recommendation apparatus 100 is the information accessed in the past by the user terminal 12 belonging to the same group as the user terminal 11 that is the transmission source of the information recommendation request based on the browsing history acquired from the browsing history holding server 22. (Electronic data such as Web contents) is determined as recommendation information recommended to the user terminal 11. At this time, for example, the information recommendation apparatus 100 determines to preferentially use information that is frequently accessed by the user terminal 12 as recommendation information. Then, the information recommendation device 100 acquires the recommended information determined in this way from the recommended information holding server 23, and provides the acquired recommended information to the user terminal 11.

  Thus, the information recommendation apparatus 100 according to the first embodiment forms a group including the user terminals 12 that have communicated with the user terminal 11 when receiving the information recommendation request from the user terminal 11. Then, information such as Web contents accessed in the past by the user terminal 12 belonging to the group is provided to the user terminal 11 as recommendation information.

  Here, when the user terminal 11 and the user terminal 12 are in the same group, both users have communicated such as sending and receiving e-mails and chatting and calling, so their hobbies and preferences are similar. There is a possibility. That is, since the information recommendation apparatus 100 uses the information that has been browsed by the user of the user terminal 12 with similar hobbies and preferences as the recommendation information, there is a possibility that the user of the user terminal 11 may desire. High information can be provided to the user terminal 11.

  Moreover, since the information recommendation apparatus 100 determines recommendation information based on the information browsed in the past by the user of the user terminal 12, the information has never been browsed by the user of the user terminal 11 in the past. Even in this case, recommendation information can be provided to the user terminal 11.

  Furthermore, since the information recommendation device 100 forms a group based on the communication history and uses information that has been accessed by user terminals belonging to the same group as recommendation information, an enormous amount of information is stored on the network 1. Even in the case where the information is stored, information that is highly likely to be desired by the user of the user terminal 11 can be provided to the user terminal 11 from such an enormous amount of information.

  In addition, the problem that it is not possible to recommend appropriate information to users who have a small browsing history, and the problem that it is not possible to recommend appropriate information because a huge amount of information has been accumulated on the network is “Data Sparsity”. The information recommendation device 100 according to the first embodiment can solve the “Data Sparsity” problem as described above.

  Hereinafter, the information recommendation apparatus 100 will be described in detail. Hereinafter, as in the example illustrated in FIG. 1, the case where the information recommendation apparatus 100 receives an information recommendation request from the user terminal 11 will be described as an example. However, the information recommendation device 100 may accept an information recommendation request from a user terminal other than the user terminal 11 (for example, the user terminal 12).

[Configuration of information recommendation device]
Next, the information recommendation apparatus 100 shown in FIG. 1 will be described with reference to FIG. FIG. 2 is a block diagram illustrating a configuration example of the information recommendation apparatus 100 according to the first embodiment. As illustrated in FIG. 2, the information recommendation device 100 according to the first embodiment includes a communication history acquisition unit 111, a group formation unit 112, a browsing history acquisition unit 113, a recommendation information determination unit 114, and a recommendation information provision unit. 115.

  The communication history acquisition unit 111 acquires a communication history from the communication history holding server 21 when an information recommendation request is received from the user terminal 11. Note that the user terminal 11 transmits an information recommendation request including a user ID for identifying the user of the user terminal 11 to the information recommendation device 100.

  Here, FIG. 3 shows an example of a communication history held by the communication history holding server 21 in the first embodiment. As illustrated in FIG. 3, the communication history holding server 21 holds a communication history including “communication source user ID” and “destination user ID”. “Communication source user ID” indicates an identifier for identifying the user of the user terminal that has started communication. “Destination user ID” indicates an identifier for identifying a user who is a communication partner of the user indicated by the communication source user ID. For example, when the user with the user ID “A” sends an e-mail to the user with the user ID “B”, the communication history holding server 21 uses the communication source user ID “A” and the communication destination use. The communication history with the user ID “B” is held. Hereinafter, the user ID “A” is the user ID of the user of the user terminal 11, and the user ID “B” is the user ID of the user of the user terminal 12. To do.

  That is, in the example shown in FIG. 3, the user terminal 11 held by the user with the user ID “A” communicates with the user terminal 12 held by the user with the user ID “B”. It shows that there is. Similarly, the user terminal 11 held by the user with the user ID “A” communicated with the user terminal (not shown in FIG. 1) held by the user with the user ID “D”. It shows that there is. In the example illustrated in FIG. 3, the user terminal 12 held by the user with the user ID “B” includes the user terminal held by the user with the user ID “D” and the user ID “C”. ”Indicates that the user has communicated with the user terminal held by the user.

  The communication history acquisition unit 111 acquires the communication history as illustrated in FIG. 3 from the communication history holding server 21. Note that the communication history acquisition unit 111 holds only the communication history in which the user ID included in the information recommendation request is stored in at least one of “communication source user ID” and “destination user ID”. You may acquire from the server 21. In the case of the above example, the communication history acquisition unit 111 receives the communication history of the communication source user ID “A” and the communication destination user ID “B”, the communication source user ID “A”, and the communication history holding server 21. Only the communication history of the communication destination user ID “D” may be acquired.

  Returning to the description of FIG. 2, the group forming unit 112 forms a group of user terminal groups based on the relationship regarding communication between user terminals obtained from the communication history acquired by the communication history acquiring unit 111. The group forming unit 112 according to the first embodiment forms a group of user terminal groups that have communicated in the past based on the communication history acquired by the communication history acquiring unit 111. Specifically, the group forming unit 112 in the first embodiment forms a group of user IDs of other user terminals that have communicated with the user terminal 11 that is the transmission source of the information recommendation request. .

  An example of group formation processing by the group formation unit 112 in the first embodiment will be described with reference to FIG. FIG. 4 is a diagram illustrating an example of group formation processing by the group formation unit 112 in the first embodiment. Here, it is assumed that the communication history holding server 21 holds the communication history illustrated in FIG. In addition, the user ID included in the information recommendation request received from the user terminal 11 is “A”.

  As shown in FIG. 3, “B” and “D” exist in the communication destination user ID corresponding to the communication source user ID “A”. That is, the user terminal 11 held by the user with the user ID “A” includes the user terminal held by the user with the user ID “B” and the use held by the user with the user ID “D”. Have communicated with a user terminal. Therefore, the group forming unit 112 acquires the user ID “A” included in the information recommendation request and the communication destination user IDs “B” and “D” corresponding to the communication source user ID “A”. . Then, as illustrated in FIG. 4, the group forming unit 112 forms a group G1 including the user IDs “A”, “B”, and “D”.

  The group forming unit 112 also acquires a communication source user ID corresponding to the communication destination user ID “A”, and includes the acquired user ID in the group G1. However, in the example shown in FIG. 3, since “A” does not exist in the communication destination user ID, the group forming unit 112 also acquires the communication source user ID corresponding to the communication destination user ID “A”. Do not do.

  4 shows an example in which the group forming unit 112 includes the user ID “A” of the user who is the transmission source of the information recommendation request in the group G1. However, the group forming unit 112 may not include the user ID “A” in the group G1. This is because the recommendation information determination unit 114, which will be described later, uses the information browsed by users other than the user with the user ID “A” as recommendation information.

  FIG. 4 shows an example in which the information recommendation apparatus 100 receives an information recommendation request including the user ID “A”, so that the group forming unit 112 has the user IDs “A”, “B”, and “D”. ”Was formed. However, for example, when the information recommendation request including the user ID “B” is received, the group forming unit 112 forms a group including the user IDs “B”, “C”, and “D”.

  Returning to the description of FIG. 2, the browsing history acquisition unit 113 acquires the browsing history from the browsing history holding server 22. Note that the browsing history acquisition unit 113 may acquire the browsing history when the information recommendation request is received from the user terminal 11 or when the group is formed by the group forming unit 112. Good. When a browsing history is transmitted from the browsing history holding server 22, the browsing history acquisition unit 113 acquires the browsing history every time it is transmitted by the browsing history holding server 22.

  Here, FIG. 5 shows an example of a browsing history held by the browsing history holding server 22 in the first embodiment. As illustrated in FIG. 5, the browsing history holding server 22 holds a browsing history including “user ID” and “information ID”. “User ID” indicates an identifier for identifying a user such as the user terminal 11 and corresponds to the “source user ID” and the “destination user ID” illustrated in FIG. “Information ID” indicates an identifier for identifying information such as Web content.

  That is, in the example shown in FIG. 5, the user terminal 12 held by the user with the user ID “B” is divided into information indicated by the information ID “I11” and information indicated by the information ID “I12”. This indicates that the user has accessed. Similarly, in the example shown in FIG. 5, the user terminal held by the user with the user ID “C” is divided into information indicated by the information ID “I11” and information indicated by the information ID “I12”. This indicates that the user has accessed. In the example illustrated in FIG. 5, the user terminal held by the user with the user ID “D” accesses the information indicated by the information ID “I11” and the information indicated by the information ID “I13”. It has shown that it has done.

  The browsing history acquisition unit 113 may acquire only the browsing history corresponding to the user ID included in the group formed by the group forming unit 112 from the browsing history holding server 22. In the case of the above example, the browsing history acquisition unit 113 receives the communication history of the user ID “B” and the information ID “I11” and the communication of the user ID “B” and the information ID “I12” from the browsing history holding server 22. The history, the communication history of the user ID “D” and the information ID “I11”, and the communication history of the user ID “D” and the information ID “I13” may be acquired.

  Returning to the description of FIG. 2, the recommendation information determination unit 114 is a transmission source of the information recommendation request based on the group formed by the group formation unit 112 and the browsing history acquired by the browsing history acquisition unit 113. Recommendation information which is information recommended for the user terminal 11 is determined. Specifically, the recommendation information determination unit 114 according to the first embodiment extracts a user ID other than the user ID of the user terminal 11 from the user IDs included in the group formed by the group formation unit 112. Subsequently, the recommended information determination unit 114 acquires the information ID of the information browsed by the user indicated by the extracted user ID from the browsing history, and counts the number of the acquired information ID as the number of browsing for each information ID. . Then, the recommended information determination unit 114 determines to preferentially use information with a large number of browsing times as recommended information.

  An example of recommendation information determination processing by the recommendation information determination unit 114 according to the first embodiment will be described with reference to FIG. FIG. 6 is a diagram illustrating an example of recommendation information determination processing by the recommendation information determination unit 114 according to the first embodiment. Here, it is assumed that the group illustrated in FIG. 4 is formed by the group forming unit 112. Moreover, the browsing history acquisition part 113 shall acquire the browsing history illustrated in FIG. Further, it is assumed that the information recommendation device 100 has received an information recommendation request including the user ID “A” from the user terminal 11.

  As shown in FIG. 4, the group G1 formed by the group forming unit 112 includes user IDs “A”, “B”, and “D”. Therefore, the recommendation information determination unit 114 extracts the user IDs “B” and “D” other than the user ID “A” of the user terminal 11 from the group G1. Subsequently, the recommended information determination unit 114 determines the information IDs “I11” and “I12” corresponding to the user ID “B” and the information ID corresponding to the user ID “D” from the browsing history illustrated in FIG. “I11” and “I13” are acquired. That is, the recommended information determination unit 114 acquires the information ID “I11” twice, acquires the information ID “I12” once, and acquires the information ID “I13” once. For this reason, as shown in FIG. 6, the recommended information determination unit 114 counts the number of browsing times “2” of the information ID “I11”, counts the number of browsing times “1” of the information ID “I12”, and sets the information ID “ The number of browsing times “1” of “I13” is counted. Then, the recommended information determination unit 114 determines to preferentially use the information indicated by the information ID “I11” having a large number of browsing among the information IDs “I11” to “I13” as the recommended information.

  Returning to the description of FIG. 2, the recommendation information provision unit 115 provides the recommendation information determined by the recommendation information determination unit 114 to the user terminal 11 that is the transmission source of the information recommendation request. Specifically, the recommendation information providing unit 115 acquires the recommendation information from the recommendation information holding server 23 that holds the recommendation information determined by the recommendation information determination unit 114. Then, the recommendation information providing unit 115 transmits the recommendation information acquired from the recommendation information holding server 23 to the user terminal 11.

  The recommended information providing unit 115 uses a list of “recommended information (for example, information title and information storage location information)” and “viewing frequency” sorted by the number of browsing times counted by the recommended information determining unit 114. The user terminal 11 may be provided. The recommendation information providing unit 115 also lists “recommendation information (for example, information title and information storage location information)” and “view count” sorted by the time when each information is registered in the recommendation information holding server 23. May be provided to the user terminal 11.

[Processing procedure by information recommendation device]
Next, a procedure of information recommendation processing by the information recommendation device 100 according to the first embodiment will be described with reference to FIG. FIG. 7 is a flowchart illustrating the information recommendation processing procedure performed by the information recommendation apparatus 100 according to the first embodiment.

  As illustrated in FIG. 7, the communication history acquisition unit 111 of the information recommendation device 100 determines whether an information recommendation request is received from the user terminal (step S101). At this time, when the communication history acquisition unit 111 does not receive the information recommendation request (No at Step S101), the communication history acquisition unit 111 waits until the information recommendation request is received.

  On the other hand, when receiving an information recommendation request (Yes at Step S101), the communication history acquisition unit 111 acquires a communication history from the communication history holding server 21 (Step S102). Then, the group forming unit 112 forms a group including the user terminal that is the transmission source of the information recommendation request and other user terminals that have communicated with the user terminal (step S103). The browsing history acquisition unit 113 acquires a browsing history from the browsing history holding server 22 (step S104).

  Subsequently, an information ID indicating information browsed by the user indicated by the user ID included in the group formed by the group forming unit 112 is acquired from the browsing history, and the number of acquired information IDs is counted as the number of browsing times. (Step S105). Then, the recommended information determination unit 114 determines that the information indicated by the information ID having the largest number of browsing times is the recommended information (step S106). And the recommendation information provision part 115 acquires the recommendation information determined by the recommendation information determination part 114 from the recommendation information holding | maintenance server 23, and provides the acquired recommendation information to the user terminal which is the transmission source of an information recommendation request ( Step S107).

[Effect of Example 1]
As described above, the information recommendation apparatus 100 according to the first embodiment recommends information such as Web content accessed by a user terminal that has communicated with the user terminal that is the transmission source of the information recommendation request. This recommendation information is provided as information to the user terminal that is the transmission source of the information recommendation request. As a result, the information recommendation device 100 according to the first embodiment can recommend the appropriate recommendation information to the user terminal even when various types of information have not been accessed by the user terminal that is the transmission source of the information recommendation request. Even when a huge amount of information is accumulated on the network 1, it is possible to provide appropriate recommendation information to the user terminal. In other words, the information recommendation device 100 according to the first embodiment can solve the problem of “Data Sparsity”.

  By the way, the information recommendation apparatus, the information recommendation method, and the information recommendation program according to the present invention may be implemented in various different forms other than the above-described embodiments. Therefore, in the second embodiment, another embodiment of the information recommendation device, the information recommendation method, and the information recommendation program according to the present invention will be described.

[Group formation 1]
In the first embodiment, the example in which the group forming unit 112 groups the user ID groups of users who have communicated with the user indicated by the user ID included in the information recommendation request as a group. In other words, when the user ID is a node and the user IDs of users who have communicated with each other are represented by an undirected graph that connects with edges (also called “links”), The group forming unit 112 forms a user ID included in the information recommendation request and another user ID group connected by an edge as the same group. However, the group formation processing by the group formation unit 112 is not limited to this example.

  For example, the group forming unit 112 has communicated with the user ID of the user Y who has communicated with the user X indicated by the user ID included in the information recommendation request, and the user Y. A group including a user ID of a certain user Z may be formed. That is, when the relationship between the user IDs is represented by the undirected graph, the group forming unit 112 not only includes the user IDs included in the information recommendation request and other user ID groups connected by the edge. The other user ID group connected to the other user ID group by the edge may be formed as the same group.

  This point will be specifically described with reference to FIG. FIG. 8 is a diagram for explaining another embodiment of the information recommendation device 100 according to the first embodiment. 8 shows an example of the communication history acquired from the communication history holding server 21 by the communication history acquisition unit 111, and the right diagram of FIG. 8 shows an example of the group formed by the group forming unit 112. Show.

  In the example illustrated in FIG. 8, it is assumed that the information recommendation apparatus 100 has received an information recommendation request including the user ID “A”. In such a case, the group forming unit 112 acquires communication destination user IDs “B” and “D” corresponding to the communication source user ID “A”. Thereby, the group formation unit 112 acquires the user IDs “B” and “D” of the users who have communicated with the user indicated by the user ID “A”. Further, the group forming unit 112 acquires communication destination user IDs “C” and “D” corresponding to the communication source user ID “B”, and receives communication destination users corresponding to the communication source user ID “D”. The ID “C” is acquired, and the communication source user ID “E” corresponding to the communication destination user ID “D” is acquired. Thereby, the group forming unit 112 acquires the user IDs “C” and “D” of the users who have communicated with the user indicated by the user ID “B”, and the user ID “D”. The user IDs “C” and “E” of the users who have communicated with the user indicated by “” are acquired. The group forming unit 112 has communicated with the user ID “A” included in the information recommendation request and the user with the user ID “A” as shown in the right diagram of FIG. The user IDs “B” and “D” of the user and the user ID “of the user who has communicated with the user with the user ID“ B ”or the user with the user ID“ D ”. A group G1 including “C” and “E” is formed.

  As described above, the group forming unit 112 may also include users who have communicated indirectly with the user who is the transmission source of the information recommendation request in the same group. In the example illustrated in FIG. 8, the group forming unit 112 uses the number of nodes (user IDs) that mediate connection with the user ID included in the information recommendation request as “1” or “2”. The person ID group was formed as the same group. However, the group forming unit 112 may form a group of user IDs in which the number of nodes that mediate connection with the user ID included in the information recommendation request is “3” or more as the same group.

[Group formation 2]
In addition, when the group forming unit 112 is represented by an undirected graph in which user IDs of users who have communicated with each other are connected by an edge, the group of user IDs having a dense connection relation are set as the same group. It may be formed. For example, the group forming unit 112 can specify a user ID group having a close connection relationship using a community extraction algorithm (see Non-Patent Document 3 and the like).

  This point will be specifically described with reference to FIG. FIG. 9 is a diagram for explaining another embodiment of the information recommendation apparatus 100 according to the first embodiment. 9 shows an example of a communication history acquired from the communication history holding server 21 by the communication history acquisition unit 111, and the central diagram of FIG. 9 shows an example of an undirected graph having a user ID as a node. 9 shows an example of a group formed by the group forming unit 112.

  In the example illustrated in FIG. 9, the group forming unit 112 generates an undirected graph having the user ID illustrated in the central diagram of FIG. 9 as a node from the communication history illustrated in the left diagram of FIG. And the group formation part 112 specifies user ID group with which a connection relation is dense by applying said community extraction algorithm with respect to this undirected graph. In the example illustrated in FIG. 9, the group forming unit 112 identifies the user IDs “A”, “B”, “C”, and “D” as a group of user IDs that are closely connected, and the user IDs “E”, “F”, and “G” are specified as user ID groups that are closely connected. In such a case, the group forming unit 112 forms a group G1 including user IDs “A”, “B”, “C”, and “D” as shown in the right diagram of FIG. A group G2 including “E”, “F”, and “G” is formed.

  As described above, the information recommendation apparatus 100 may form a group of user IDs having a close connection relationship as the same group, so that the hobby / preference may be similar to the user who is the transmission source of the information recommendation request. Information viewed by high users can be recommended.

[Group formation 3]
Moreover, the group formation part 112 may form a group using the frequency | count (communication frequency) of communication performed between users, and communication amount. Specifically, the group formation unit 112 weights the connection relation between user IDs according to the communication frequency or communication amount, and forms the user ID groups connected with a weight greater than or equal to a predetermined value as the same group. May be. In other words, the group forming unit 112 may form a group of user IDs having a dense connection relationship as the same group, assuming that the connection relationship between the user IDs having a high communication frequency or communication amount is close.

  This point will be specifically described with reference to FIG. FIG. 10 is a diagram for explaining another embodiment of the information recommendation device 100 according to the first embodiment. 10 shows an example of the communication history acquired from the communication history holding server 21 by the communication history acquisition unit 111, and the right diagram of FIG. 10 shows an example of an undirected graph having the user ID as a node. Indicates. FIG. 10 shows an example in which the group forming unit 112 forms a group according to the traffic.

  In the example illustrated in FIG. 10, the group forming unit 112 generates an undirected graph having the user ID illustrated in the right diagram of FIG. 10 as a node from the communication history illustrated in the left diagram of FIG. At this time, the group formation part 112 weights the edge between each user ID using the communication amount contained in the communication history. In the example illustrated in FIG. 10, the group forming unit 112 uses the ratio of the communication amount between users as a weight for each edge. And the group formation part 112 forms the user ID group connected by the weight more than a predetermined value as the same group based on the weight provided to each edge. For example, in the example shown in FIG. 10, when the group forming unit 112 forms user ID groups having a weight of “2” or more as the same group, as in the example shown in FIG. 9 described above, A group G1 including user IDs “A”, “B”, “C”, and “D” is formed, and a group G2 including user IDs “E”, “F”, and “G” is formed.

  FIG. 10 shows an example in which the group forming unit 112 forms a group according to the amount of communication. However, the group forming unit 112 assigns a higher weight between user IDs having a higher communication frequency, and the communication frequency. A lower weight is assigned to a lower user ID. For example, in the example shown in FIG. 10, the communication with the communication amount “1” is performed between the user with the user ID “D” and the user with the user ID “E”. Assigns a weight “1” to the edge connecting the user ID “D” and the user ID “E”. However, when the communication with the communication amount “1” is performed a plurality of times (for example, N times) between the user with the user ID “D” and the user with the user ID “E”, the group For example, the forming unit 112 assigns a weight “N” to an edge connecting the user ID “D” and the user ID “E”.

  In this way, the information recommendation device 100 may have similar hobbies and preferences to the user who is the source of the information recommendation request by forming a group of user ID groups based on the communication frequency or communication volume. It is possible to recommend information viewed by high-level users. For example, when the communication frequency between predetermined user terminals is high, it indicates that communication is frequently performed between the user terminals. When there is a large amount of communication between user terminals, This indicates that the other user has written a long comment on the blog published by the user, a long-time call between the users, and the like. Both users of such user terminals have high intimacy and are likely to have similar hobbies and preferences. For this reason, the information recommendation apparatus 100 can use the information browsed by users who are likely to have similar hobbies / preferences as recommendation information by forming a group using communication frequency or communication volume. become.

  Note that the group forming unit 112 may selectively use a process of forming a group based on the communication frequency and a process of forming a group based on the communication amount according to the communication content. For example, the group forming unit 112 may perform a process of forming a group based on the communication frequency when communication such as e-mail and the like in which there is no significant difference in communication volume is performed between users. When communication with a large difference in communication volume such as a call is performed between users, a process of forming a group based on the communication volume may be performed.

[Group formation 4]
Further, the group forming unit 112 may form user ID groups of users having similar communication times as a group instead of forming user ID groups of users who have communicated as a group. Good. Specifically, the group formation unit 112 forms a group of user IDs whose communication times are included in a predetermined time range among communication source user IDs held in the communication history holding server 21 as the same group. May be.

  This point will be specifically described with reference to FIG. FIG. 11 is a diagram for explaining another embodiment of the information recommendation device 100 according to the first embodiment. 11 shows an example of a communication history acquired from the communication history holding server 21 by the communication history acquisition unit 111, and the right diagram of FIG. 11 shows an example of an undirected graph having a user ID as a node. Indicates.

  In the example illustrated in FIG. 11, the group forming unit 112 generates an undirected graph having the user ID illustrated in the right diagram of FIG. 11 as a node from the communication history illustrated in the left diagram of FIG. At this time, the group forming unit 112 connects the communication source user IDs whose communication times included in the communication history are included in a predetermined time range by an edge. For example, it is assumed that the predetermined time is “one hour before and after”. In this case, the group forming unit 112 transmits the communication source user ID “A” that communicated at the communication time “10:00” and the communication time “1 hour after the communication time“ 10:00 ”. The communication source user ID “F” that communicated at “11:00” is connected. Similarly, the group forming unit 112 communicates with the communication source user ID “F” communicating at the communication time “11:00” and the communication source user ID “F” communicating at the communication time “12:00”. E ”. In this manner, the group forming unit 112 forms a group including the user IDs “A” to “G” as shown in the right diagram of FIG. Further, the user ID included in the group may be determined based on the communication time of the user ID that transmitted the information recommendation request. For example, in the example illustrated in FIG. 11, the information recommendation apparatus 100 receives the information recommendation request including the user ID “D”, and a predetermined time “one hour before and after the time when the user ID“ D ”communicates” Suppose that In this case, the group forming unit 112 forms a group including the user IDs “C”, “D”, and “E”.

  In this way, the information recommendation device 100 forms a group of user IDs of communication source users with similar communication times at the start of communication as the same group, so that the user who is the source of the information recommendation request and the hobby / Information viewed by users who are likely to have similar preferences can be recommended. For example, users who have similar times for sending emails, browsing other people's blogs, making phone calls, etc. have similar lifestyles even if they have never communicated with each other. May have similar interests and preferences. For this reason, the information recommendation apparatus 100 can use the information browsed by users who are likely to have similar hobbies / preferences as recommendation information by forming a group using the similarity of communication times. become.

[Group formation 5]
In addition, the group forming unit 112 does not form a group including user ID groups of users who have communicated with each other, but forms a group including user ID groups of users with similar traffic. May be. Specifically, the group forming unit 112 forms a group of user IDs whose communication amount is included in a predetermined range among communication source user IDs held in the communication history holding server 21 as the same group. Also good.

  This point will be specifically described with reference to FIG. FIG. 12 is a diagram for explaining another embodiment of the information recommendation device 100 according to the first embodiment. 12 shows an example of the communication history acquired from the communication history holding server 21 by the communication history acquisition unit 111, and the right diagram of FIG. 12 shows an example of an undirected graph having the user ID as a node. Indicates.

  In the example shown in FIG. 12, the group forming unit 112 generates an undirected graph having the user ID shown in the right diagram of FIG. 12 as a node from the communication history shown in the left diagram of FIG. At this time, the group forming unit 112 connects the communication source user IDs having similar communication amounts included in the communication history with edges. For example, it is assumed that a group is formed by dividing into a user ID group having a traffic volume of “10” or more and a user ID group having a traffic volume of less than “10”. In such a case, as shown in the right diagram of FIG. 12, the group forming unit 112 has a communication source user ID “A” that has performed communication with a communication amount “10” and a communication amount “20”. A group is formed in which communication source user IDs “D” and “F” that have performed communication are connected to each other. Further, the group forming unit 112 performs communication with the communication source user IDs “B”, “C”, and “E” that performed communication with the communication amount “1” and with the communication amount “2”. A group in which the communication source user ID “G” is connected to each other is formed.

  In this way, the information recommendation apparatus 100 has similar hobbies and preferences to the user who is the transmission source of the information recommendation request by forming the user ID groups of communication source users with similar traffic as the same group. It is possible to recommend information viewed by users who are likely to be. For example, users who have similar e-mail sizes, call durations with others, and the like may have similar hobbies and preferences even if they have never communicated with each other. For this reason, the information recommendation apparatus 100 can use information viewed by a user who is likely to have similar hobbies and preferences as a recommendation information by forming a group using the similarity of the traffic. become.

[Recommended information]
In the first embodiment, the recommendation information determination unit 114 recommends information based on the number of times (viewing frequency) of browsing information by other users who belong to the same group as the user who has transmitted the information recommendation request. An example of determining is shown. However, the recommendation information determination process by the recommendation information determination unit 114 is not limited to this example.

  For example, the recommendation information determination unit 114 determines the other user based on the number of times of communication (communication frequency) and the amount of communication performed between the user who is the transmission source of the information recommendation request and another user. It is also possible to weight the information browsed by and to change the probability of determining the recommended information based on the assigned weight. Specifically, the recommendation information determination unit 114 determines between the user who is the transmission source of the information recommendation request, the other user, and the other user based on the communication history acquired by the communication history acquisition unit 111. The communication frequency or communication amount of communication is acquired. And the recommendation information determination part 114 gives priority to each information by weighting the information browsed by the other user according to this communication frequency or communication amount. Then, the recommended information determination unit 114 determines that the information with higher priority is to be recommended information with a higher probability, and the information with lower priority is determined to be recommended information with a lower probability.

  This point will be specifically described with reference to FIG. FIG. 13 is a diagram for explaining another embodiment of the information recommendation device 100 according to the first embodiment. 13 shows an example of the browsing history acquired from the browsing history holding server 22 by the browsing history acquisition unit 113, and the lower left diagram of FIG. 13 shows the communication history holding server 21 from the communication history holding server 21. An example of the acquired communication history is shown, and the right figure of FIG. 13 shows an example of the priority given to each information by the recommended information determination unit 114. FIG. 13 illustrates an example in which the browsing history acquisition unit 113 weights various types of information according to the amount of communication. Here, the information recommendation apparatus 100 receives an information recommendation request including the user ID “A”, and a group including the user IDs “A”, “B”, and “D” is formed by the group forming unit 112. Shall be.

  As shown in the upper left diagram of FIG. 13, each piece of information indicated by the information ID “I11”, “I12”, or “I13” is browsed once. For this reason, in this example, the browsing frequency of each information is equivalent. Here, as shown in the lower left diagram of FIG. 13, the user with the user ID “A” and the user with the user ID “B” perform communication with the communication amount “1”, and the user ID The user of “A” and the user of user ID “D” are performing communication with a communication volume of “3”. At this time, the recommended information determination unit 114 gives the weight “1” to the information of the information ID “I11” or “I12” browsed by the user with the user ID “B”, and also uses the user ID. A weight “3” is assigned to the information of the information ID “I13” viewed by the user “D”. In the example illustrated in FIG. 13, the recommendation information determination unit 114 sets such weight as a priority. Here, the higher the numerical value, the higher the priority.

  And the recommendation information determination part 114 gives priority to the information with a large numerical value of a priority, and provides it to the user terminal which is the transmission source of an information recommendation request. Specifically, the recommendation information determination unit 114 receives information ID “I13” having a priority “3” rather than information ID “I11” or “I12” having a priority “1”. Provide priority. For example, the recommended information determination unit 114 provides the information ID “I13” with a probability of 3 times out of 5 times, while providing the information with the information ID “I11” with a probability of 1 time out of 5 times, Similarly, information of information ID “I12” is provided with a probability of 1 out of 5 times.

  In FIG. 13, the recommendation information determination unit 114 gives an example of giving priority to each information according to the communication amount. However, the recommendation information determination unit 114 gives priority to each information according to the communication frequency. The process of giving is also performed. For example, in the example shown in FIG. 13, the information indicated by the information ID “I11” is viewed once by the user with the user ID “B”, and the user and the user with the user ID “A” are viewed. Since the amount of communication performed with the user with ID “B” is “1”, the recommended information determination unit 114 assigns “1” to the priority of the information ID “I11”. . However, when the information indicated by the information ID “I11” is viewed multiple times (for example, N times) by the user with the user ID “B” or the user ID “D”, the recommended information determination unit 114 assigns “N” to the priority of the information ID “I11”.

  As described above, the information recommendation device 100 assigns a priority to each piece of information based on the communication frequency and communication amount of communication performed between users, and thus the user who is the transmission source of the information recommendation request. Information viewed by users who are likely to have similar hobbies and preferences can be recommended. For example, between users with high communication frequency and between users with a large amount of communication, there is a high possibility of closeness and similar hobbies and preferences. For this reason, the information recommendation apparatus 100 determines recommendation information according to the priority based on the communication frequency and the communication amount, and uses information browsed by users who are likely to have similar hobbies and preferences as recommendation information. It becomes possible.

[User ID]
Moreover, in the said Example, the information recommendation apparatus 100 showed the example which performs various processes using user ID. For example, the information recommendation device 100 in the above embodiment receives an information recommendation request including a user ID, and forms a group of user ID groups. However, the information recommendation apparatus 100 may perform various processes using a terminal ID for identifying the user terminal instead of the user ID. For example, the information recommendation apparatus 100 may receive an information recommendation request including a terminal ID and form a group of terminal ID groups.

[Communication history, browsing history]
In the above-described embodiment, the information recommendation apparatus 100 acquires the communication history from the communication history holding server 21 and acquires the browsing history from the browsing history holding server 22. However, the information recommendation device 100 may have a function equivalent to that of the communication history holding server 21 and the browsing history holding server 22. For example, the information recommendation apparatus 100 may have a function of collecting communication history and browsing history from various servers such as a Web server. In such a case, the information recommendation device 100 may not perform processing for acquiring a communication history or a browsing history from the communication history holding server 21 or the browsing history holding server 22.

[System configuration]
In addition, among the processes described in the above embodiment, all or a part of the processes described as being automatically performed can be manually performed, or the processes described as being performed manually can be performed. All or a part can be automatically performed by a known method. In addition, the processing procedures, specific names, and information including various data and parameters shown in the document and drawings can be arbitrarily changed unless otherwise specified. For example, the various types of information illustrated in FIGS. 3 to 5 are merely examples, and can be changed to arbitrary information.

  Further, each component of each illustrated apparatus is functionally conceptual, and does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution / integration of each device is not limited to that shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various loads or usage conditions. Can be integrated and configured.

[program]
It is also possible to create a program in which the processing executed by the information recommendation device 100 described in the first and second embodiments is described in a language that can be executed by a computer. For example, an information recommendation program in which processing executed by the information recommendation apparatus 100 according to the first embodiment is described in a language that can be executed by a computer can be created. In this case, when the computer executes the information recommendation program, the same effect as in the first embodiment can be obtained. Further, the information recommendation program is recorded on a computer-readable recording medium, and the information recommendation program recorded on the recording medium is read by the computer and executed, thereby realizing the same processing as in the first embodiment. Also good. Hereinafter, an example of a computer that executes an information recommendation program that realizes the same function as the information recommendation apparatus 100 illustrated in FIG. 2 will be described.

  FIG. 14 is a diagram illustrating a computer 1000 that executes an information recommendation program. As illustrated in FIG. 14, the computer 1000 includes, for example, a memory 1010, a CPU 1020, a hard disk drive interface 1030, a disk drive interface 1040, a serial port interface 1050, a video adapter 1060, and a network interface 1070. These units are connected by a bus 1080.

  The memory 1010 includes a ROM (Read Only Memory) 1011 and a RAM (Random Access Memory) 1012 as illustrated in FIG. The ROM 1011 stores a boot program such as BIOS (Basic Input Output System). The hard disk drive interface 1030 is connected to the hard disk drive 1031 as illustrated in FIG. The disk drive interface 1040 is connected to the disk drive 1041 as illustrated in FIG. For example, a removable storage medium such as a magnetic disk or an optical disk is inserted into the disk drive. The serial port interface 1050 is connected to, for example, a mouse 1051 and a keyboard 1052 as illustrated in FIG. The video adapter 1060 is connected to a display 1061, for example, as illustrated in FIG.

  Here, as illustrated in FIG. 14, the hard disk drive 1031 stores, for example, an OS 1091, an application program 1092, a program module 1093, and program data 1094. That is, the information recommendation program is stored in, for example, the hard disk drive 1031 as a program module in which a command to be executed by the computer 1000 is described. For example, the communication history acquisition procedure for executing the same information processing as the communication history acquisition unit 111 illustrated in FIG. 2, the group formation procedure for executing the same information processing as the group formation unit 112, and the browsing history acquisition unit 113 The browsing history acquisition procedure for executing the information processing, the recommended information determining procedure for executing the same information processing as the recommendation information determining unit 114, and the recommended information providing procedure for executing the same information processing as the recommended information providing unit 115 The described program module 1093 is stored in the hard disk drive 1031.

  Various data held by the information recommendation device 100 described in the above embodiment is stored as program data in, for example, the memory 1010 or the hard disk drive 1031. Then, the CPU 1020 reads the program module 1093 and the program data 1094 stored in the memory 1010 and the hard disk drive 1031 to the RAM 1012 as necessary, and acquires a communication history acquisition procedure, a group formation procedure, a browsing history acquisition procedure, a recommended information determination procedure, The recommended information provision procedure is executed.

  Note that the program module 1093 and the program data 1094 related to the information recommendation program are not limited to being stored in the hard disk drive 1031, but are stored in, for example, a removable storage medium and read out by the CPU 1020 via the disk drive or the like. Also good. Alternatively, the program module 1093 and the program data 1094 related to the information recommendation program are stored in another computer connected via a network (LAN (Local Area Network), WAN (Wide Area Network), etc.), and the network interface 1070 is stored. Via the CPU 1020.

DESCRIPTION OF SYMBOLS 1 Network 11 User terminal 12 User terminal 21 Communication history holding server 22 Browsing history holding server 23 Recommended information holding server 100 Information recommendation apparatus 111 Communication history acquisition part 112 Group formation part 113 Browsing history acquisition part 114 Recommended information determination part 115 Recommendation Information provision department

Claims (6)

A communication history acquisition unit for acquiring communication history between users;
A group forming unit for forming a group of users based on a relationship regarding communication between a predetermined user and other users obtained from the communication history acquired by the communication history acquiring unit;
A browsing history acquisition unit that acquires a browsing history that is a history of accessing information on the network by the other users;
Based on the browsing history acquired by the browsing history acquisition unit, a recommendation information determination unit that determines recommended information that is information accessed by the other user belonging to the group formed by the group formation unit;
An information recommendation device comprising: a recommendation information providing unit that provides recommendation information determined by the recommendation information determination unit to the predetermined user. The recommendation information determination unit
Of the information accessed by the other user belonging to the group formed by the group forming unit, it is determined to preferentially use information that is frequently accessed by the other user as recommended information. The information recommendation device according to claim 1. The communication history acquisition unit
Get communication history including communication frequency or communication volume in communication between users,
The recommendation information determination unit
Based on the communication history acquired by the communication history acquisition unit, the communication frequency or the communication amount in communication between the predetermined user and the other user with respect to the information accessed by the other user. The information recommendation apparatus according to claim 1, wherein priority is given according to the information, and it is determined that information having a high priority is preferentially used as recommendation information. The communication history acquisition unit
Get communication history including communication volume or communication time or communication frequency in communication between users,
The group forming unit
Based on the communication history acquired by the communication history acquisition unit, forming a group of users whose communication volume difference or communication time difference is within a predetermined range or whose communication frequency is higher than a predetermined level as the group. The information recommendation device according to any one of claims 1 to 3. An information recommendation method executed by an information recommendation device,
A communication history acquisition step for acquiring communication history between users;
A group formation step of forming a group of users based on a relationship related to communication between a predetermined user and other users obtained from the communication history acquired by the communication history acquisition step;
A browsing history acquisition step of acquiring a browsing history that is a history of accessing information on the network by the other user;
Based on the browsing history acquired by the browsing history acquisition step, a recommendation information determination step of determining recommendation information that is information accessed by the other user belonging to the group formed by the group formation step;
A recommendation information providing step of providing recommendation information determined in the recommendation information determination step to the predetermined user.   The information recommendation program for functioning a computer as an information recommendation apparatus as described in any one of Claims 1-4.

Images