JP2005235118A - Information extracting method and device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To extract a user who contributes the activation of communications on a network or a user who provides a trigger. <P>SOLUTION: In a communication providing device (10) on a network such as a message board, a mailing list, a chat or a blog, extracted user information (25), communication history information (27) and a parameter for extraction of information (29) are created from registered user information (16) and communication information (18) between users. From these pieces of information, there are extracted user properties including beginning users, high-rank users, important users, users who provide a trigger, follow users and similar users in the communication providing device (10) and the degree of association between users. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to communication provided over a network. More specifically, the present invention relates to an information extraction method and apparatus for extracting user characteristics in communication between users provided on a network.

  In recent years, with the development of the Internet, communication via a network has become widespread and has spread widely. In particular, in mailing lists, bulletin boards, chats, blogs, etc., regardless of whether users are known or unknown, a community is formed on the network, and active communication and information exchange are performed.

  Reflecting this situation, it is attracting attention that companies set up a community directly or indirectly on the network and use it to collect reputation and consumer needs for their products and services, and to promote their activities. Yes. In addition, users use the community on the network as a place for exchanging information, collecting information, and meeting new people.

  Such communities are essentially based on voluntary user participation. A stable and active community usually has a key person who contributes to the development of the community, and plays a role of providing topics, facilitating information exchange, and supporting novice users. Such a key person essentially communicates with many users, is trusted by many users, and thus has a great influence on the community, and also concentrates information. For this reason, negotiation activities focusing on such key persons are effective for companies in collecting and advertising information, and for users to maintain a comfortable and stable community. Therefore, it is important to be able to discover key persons and active environments that support active and sustainable communication.

  On the other hand, in a community that does not have an environment that activates communication between users, the members of the community are fixed, and even when the amount of communication is large, it is not always an active community. In addition, since there are many genres and topics in the community, it takes time to find an environment where users can feel easy to participate, and it is new to feel uneasy about communication with strangers. There is a problem that it is difficult for users to participate.

  In response to these problems, the community organizer now displays ranking results, evaluation results, and user profile information created from the number of participants per topic, number of accesses, number of writes, etc., from the communication history in the community. In this way, we provide information about the communication environment, and make efforts to make it easier to find communication topics and partners.

  The apparatus disclosed in Patent Document 1 does not display the user's characteristics only by profile information, but analyzes the profile information, and the contents (news, video, music, ( Background) images, advertisements, etc.) automatically. Then, by providing the user information composed of the collected contents to the communication partner, it is possible to provide information in consideration of personal interest and to provide a topic in communication with the partner.

  In addition, in order to revitalize a community, a method of ranking community activities is also known. For example, on the TeaCup bulletin board, the number of accesses and writes exceeds a certain standard, the number of accesses and the number of writes does not exceed the limit, it is easy to participate, and there are writings from a wide range of users The ranking is displayed (see Non-Patent Document 1). The review board of price.com ranks the display of the degree of excitement, threads with many replies, useful threads, and people with many writes (see Non-Patent Document 2). On the Slashdot Japan bulletin board, the moderator evaluates other users' comments (writes) and publishes the results. The purpose of this evaluation is to discover useful comments and make them visible to other people, so it is a principle that this evaluation is not performed by the personal hobbies and thoughts of the moderator (Non-Patent Document 3). reference).

JP 2003-223406 A (refer to page 4, means for solving the problem) Teacup Communication Co., Ltd., “About everyone's bulletin board ranking”, [on-line], [Search February 6, 2004], Internet <URL: http://bbs.teacup.com/help/rank_guide .html> Kakaku.com, Inc., “Write Ranking”, [on-line], [Search February 6, 2004], Internet <URL: http://www.kakaku.com/help/bbs_rank.htm> Open Source Development Network Japan, “Comments and Moderation”, [on-line], [Search February 6, 2004], Internet <URL: http://slashdot.jp/faq/com-mod.shtml>

  As described above, information extracted from the history of conventional communication and provided for the user to know about the community includes the number of participants (or the number of registrations), the number of accesses, the genre and thread with a large number of writes, and the number of writes. It is about user and high evaluation writing. However, with such information alone, the following problems remain.

  When a novice user who wants to participate in communication finds a communication partner, not only the information that the number of outgoing users is high, but also whether there are many outgoing calls with a specific user or between various users You may want to know if there are many outgoing calls. In addition, there are cases where it is desired to know whether a user makes a lot of calls with a specific user, a user who makes a lot of calls with various users, or a user who can expect a reply in a specific genre or thread. The same applies to a case where a key person who communicates with many users is identified when a company collects information or conducts advertising activities.

  In addition to the number of outgoing calls and writes, users other than novice users can increase their belonging to the community by grasping their degree of contribution to the community and their connection with other users. Motivate to join the community. In other words, it is not possible to know the degree of contribution of individual users to topics with a high number of accesses or writing, or the degree of contribution of individual users to the entire community only by the number of accesses to the topic or the degree of excitement due to the number of writing. . In addition, in the evaluation of the written content of the user, although it is possible to know about the evaluation of each written content, it is not possible to know the total contribution degree to a specific genre or the entire community.

  As described above, in the information provided in the past, characteristics of users participating in the community, i.e., beginner users who have joined the community and / or the number of outgoing calls is low, or with many users. Users who are communicating with each other, users who are communicating with many users about a specific genre or thread, giving the beginners the opportunity to speak, Information such as whether the user is following or the degree of communication between each user is not considered.

  The present invention has been made in view of such a problem, and the purpose of the present invention is to extract a user's characteristics for the entire community, or for each genre or thread, so that a person interested in the community can Easier to find a central person on a community or topic of interest, or to find someone to communicate with, and to help community users contribute to their community or communicate with others And the degree of connection with other users.

  According to the present invention, in order to achieve such an object, according to the first aspect of the present invention, in the communication providing apparatus on the network, the user information registered in the communication providing apparatus and the history of communication between users It is an information extraction method characterized by including a step of extracting a higher-level user group performing active communication from information based on the established degree of communication. According to this method, it is possible to know the upper user in the communication providing apparatus.

  According to the second aspect of the present invention, the communication providing apparatus on the network further includes a step of extracting an important user group that performs active communication for many genres and receives replies from many other users. It is the information extraction method characterized by providing. According to this method, it is possible to know important users in the communication providing apparatus.

  According to a third aspect of the present invention, the information providing method further comprises the step of extracting, by a communication providing apparatus on a network, an upper user group classified by genre that is actively communicating for a specific genre. It is. According to this method, it is possible to know the top users by genre in the communication providing apparatus.

  According to a fourth aspect of the present invention, the communication providing device on the network includes a step of extracting a beginner user group from user information registered in the communication providing device and history information of communication between users. Information extraction method. According to this method, it is possible to know a novice user in the communication providing apparatus.

  According to a fifth aspect of the present invention, there is provided the information extracting apparatus further comprising a step of extracting a trigger providing user group receiving replies from many beginner users in the communication providing apparatus on the network. According to this method, it is possible to know the opportunity providing user in the communication providing apparatus.

  The invention according to claim 6 further comprises the step of extracting a follow user group that is replying to replies from many beginner users in the communication providing apparatus on the network. It is. According to this method, it is possible to know the follow user in the communication providing apparatus.

  The invention according to claim 7 is the communication providing device on the network, based on the user information registered in the communication providing device and the communication history information between the users, based on the number of times communication between the users is established, It is an information extraction method characterized by including the step which extracts the connection degree between each user. According to this method, it is possible to know the degree of connection between users based on the establishment of communication in the communication providing apparatus.

  The invention according to claim 8 is a communication providing apparatus on a network, wherein each user set is classified into each genre, each thread, or each of the user information registered in the communication providing apparatus and the history information of communication between users. An information extraction method comprising a step of extracting a degree of connection between users based on the number of times the branch coexists. According to this method, it is possible to know the degree of connection between users based on coexistence in a common thread or branch in the communication providing apparatus.

  According to the ninth aspect of the present invention, in the communication providing apparatus on the network, each user's message is viewed by many users from the user information registered in the communication providing apparatus and the communication history information between users. It is the information extraction method characterized by including the step which extracts the to-be-viewed high-order user group. According to this method, it is possible to know about the browsing upper user in the communication providing apparatus.

  In the communication providing device on the network, the message of each user is highly evaluated by other users from the user information registered in the communication providing device and the history information of communication between users. It is the information extraction method characterized by including the step which extracts the high-ranking user group which has it. According to this method, it is possible to know the high-ranking user in the communication providing apparatus.

  The invention according to claim 11 is the information extraction method characterized in that, in the communication providing apparatus on the network, the established degree of communication is set as a ratio to the total number of users. According to this method, various characteristics of the user in the communication providing apparatus can be known based on the order ratio.

  According to a twelfth aspect of the present invention, in the communication providing apparatus on the network, the establishment of the communication is a direct bidirectional message exchange accompanied with a series of transmission / reception. According to this method, various characteristics of the user in the communication providing apparatus can be known based on direct bidirectional communication.

  In the communication providing apparatus on the network according to the thirteenth aspect of the present invention, the establishment of communication is an indirect transmission / reception message exchange accompanying transmission / reception in each genre, each thread, or each branch. This is a characteristic information extraction method. According to this method, various characteristics of the user in the communication providing apparatus can be known based on indirect bidirectional communication.

  The invention according to claim 14 is an information extraction method characterized in that the communication providing apparatus on the network is based on user information or communication history information for a predetermined period. According to this method, various characteristics of the user in the communication providing apparatus can be known for a predetermined period.

  The invention according to claim 15 is a communication providing apparatus for providing communication between users via a network, a history information collection unit, an extracted user information database, a communication history information database, an information extraction parameter database, An information extraction apparatus including an information extraction unit. ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to extract the information regarding a user's various characteristics using with a communication provision apparatus.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a functional block diagram showing the overall configuration of an embodiment of the present invention. The communication providing device 10, the information extracting device 20, and the user terminals 40 a, 40 b, and 40 n are connected via a network 30 so that they can communicate with each other. The communication providing apparatus 10 provides an environment for communication between users of the user terminals 40a, 40b, and 40n on the network 30. Usually, communication via the communication providing apparatus 10 is performed as follows.

  First, at least one content serving as the core of communication is provided to the communication providing apparatus 10. Such content is provided in the form of multimedia including text, audio and video, and can be composed of specific topics, projects, online games and the like. People who are interested in such content access the communication providing apparatus 10 via the network 30 from the user terminals 40a, 40b, and 40n. At this time, depending on the service provided by the communication providing apparatus 10, there may be a request for user registration, or a request not required. When user registration is requested, the user registers user information described in detail below in the communication providing apparatus 10 via the network 30.

  The communication providing apparatus 10 includes a mechanism in which a user who accesses accesses a message (including content) to other users. For example, in a bulletin board, mailing list, chat, blog, etc., a message such as text or HTML format is transmitted from the user terminals 40a, 40b, 40n to the communication providing apparatus 10. In the case of bulletin boards, chats, and blogs, the transmitted message is released on the communication providing device 10 and can be viewed by other users. In the case of a mailing list, a message is delivered in the form of an e-mail to a user who has registered an e-mail address as a destination. All users participating in the service can publish a new message, publish a message to a previously published message, or view a published message. Such communication activities and their contents can be recorded in the communication providing apparatus including the registered user information (when user registration is required). In this way, communication is established when a plurality of users exchange messages via the communication providing device 10 on the network 30.

  In one embodiment of the present invention, the communication providing apparatus 10 includes basic user information (user management ID, address information, password, date of subscription, etc.) and user attribute information (age, occupation, etc.) for providing communication. Etc.) and information (title, content) and related information (genre / thread ID, communication ID, source address, communication time, communication type, reply destination communication ID, etc.) of communication between users. Including the communication information database 18, the communication control unit 12 that performs access from the user terminals 40 a, 40 b, and 40 n and controls the communication providing device 10, and the information extraction device 20. And a communication information notification unit 14 responsible for Esu. The communication providing apparatus 10 is usually provided as a server in a client server environment, but is not limited thereto.

  In one embodiment of the present invention, the information extraction device 20 includes a history information collection unit 22, an information extraction unit 24, an extraction user information database 25, a communication history information database 27, and an information extraction parameter database 29. Prepare. As will be described in more detail below, the history information collection unit 22 inquires and / or registers information identifying the user in the extracted user information database 25 based on the user information and communication information from the communication information notification unit 14. Then, the communication information is processed and registered in the communication history information database 27, and information related to the information extraction parameter database 29 is recorded from the user information and the communication information. Based on instructions from the communication information notification unit 14, the information extraction unit 24 uses the information stored in the extracted user information database 25, the communication history information database 27, and the information extraction parameter database 29, such as beginner users and important users. Characteristics, users with similar user characteristics, and the degree of connection depending on the frequency of communication between users are extracted.

  The functions of the communication providing device 10 and the information extracting device 20 can be implemented by software, hardware, or a combination thereof. Also, these functions can be installed in one place, or can be installed in a plurality of places via the network 30 or another network, for example. The network 30 may be, for example, a bidirectional open or closed network such as the Internet, an intranet, a VPN, a LAN, and a WAN that are used on a public telephone network, a mobile phone network, a dedicated line, and the like. Examples of user terminals include personal computers, digital information terminals, mobile phones, and set-top boxes that are physically connected to the network 30 regardless of whether they are wired or wireless and can logically communicate with the communication providing device 10. , Online game terminals, and other terminals.

  The method and apparatus according to the embodiment of the present invention will be mainly described with respect to bulletin boards, mailing lists, chats, and blogs provided on the network 30 by way of example, but those skilled in the art will appreciate many others provided on the network. It will be understood that the present invention can also be applied to other forms of communication.

  FIG. 2 shows an embodiment of the database structure of the user information database 16. FIG. 2A shows an exemplary database structure 200a for a bulletin board, chat, blog, etc., and FIG. 2B shows an exemplary database structure 200b for a mailing list or the like. The user information database 16 includes user management ID fields 202a and 202b, address information fields 204a and 204b, password fields 206a and 206b, subscription date fields 208a and 208b, and user attribute information fields 210a and 210b. It is.

  The user management ID fields 202a and 202b are identifiers that uniquely identify the user in the communication providing apparatus 10, and are set in the communication providing apparatus at the time of user registration or first access, for example. In the case of a bulletin board, a chat, a blog, etc., the address information field 204a is the user's address in the communication providing device 10, and may be set by the user or the communication providing device. In the case of a mailing list or the like, the address information field 204b is an address of a user on the network 30, and usually an e-mail address is set. Password fields 206a and 206b store passwords for authentication set by the user or the communication providing device in correspondence with the address information 204a and 204b. However, depending on the service provided by the communication providing apparatus 10, password authentication may not be required.

  In the subscription date fields 208a and 208b, the date when the user first participates or registers in the communication is recorded and set. The user attribute information fields 210a and 210b store, for example, user attribute information provided by the user when the user first participates or registers in the communication. The user attribute information fields 210a and 210b include, for example, age fields 212a and 212b, occupation fields 214a and 214b, but are not limited thereto. Hereinafter, the data stored in the user information database 16 is referred to as “user information”.

  FIG. 3 shows an embodiment of the database structure of the communication information database 18. 3A shows an exemplary database structure 300a in a bulletin board, chat, blog, etc., and FIG. 3B shows an exemplary database structure 300b in a mailing list. The communication information database 18 includes genre / thread ID fields 302a and 302b, communication ID fields 304a and 304b, source address fields 306a and 306b, communication time fields 308a and 308b, title fields 310a and 310b, and contents. Fields 312a and 312b, communication type fields 314a and 314b, and reply destination communication ID fields 316a and 316b are included.

  The genre / thread ID fields 302a and 302b are identifiers that identify the genre and / or thread of communication, and are set based on the genre and / or thread targeted by the user's communication. The genre is a classification of communication contents set by the user or an organizer who provides a service in the communication providing apparatus 10, and the user can narrow down an object of interest for communication. The genre ID (“J003” in FIG. 3) can be set by the communication providing apparatus 10 when a new genre is set. A thread is a set of a plurality of communications derived from one message and connected. Therefore, in many cases, a topic caused by the original message is a target of communication within one thread, and the user can select a thread as a target of more detailed communication interest. The thread ID (“T008” in FIG. 3) can be set by the communication providing apparatus 10 when a new message is set.

  Furthermore, as shown in FIG. 9, one thread 100 that originates from the communication ID “0005” may have branch points 102 and 104. In such a tree-like thread, a branch can be defined as a smaller unit than the thread. In the example of FIG. 9, there are three branches “0005”-“0006”-“0009”, “0005”-“0007”-“0008”-“0011”, and “0005”-“0007”-“0010”. It is shown. In this way, communication can be classified into branches that are more detailed units than threads. This branch can be derived from the data stored in the communication history information database 27 described below in response to a request.

  The communication ID fields 304a and 304b are identifiers for identifying individual communications (including new transmission, reply, and browsing), and are uniquely set by the communication providing apparatus 10 for each communication. In the sender address fields 306a and 306b, address information 204a and 204b of the user who communicated with the communication providing apparatus 10 is stored. The communication time fields 308a and 308b record and set the time when communication is performed by the communication providing apparatus 10. In the title fields 310a and 310b, the communication title is extracted from the specific communication by the communication providing apparatus 10 and stored (in the case of browsing, “null”). In the content fields 312a and 312b, the communication content is extracted from the specific communication by the communication providing apparatus 10 and stored (in the case of browsing, “null”).

  The communication type fields 314a and 314b indicate whether the communication performed is a new call (“new” in FIG. 3) or a reply to another message (“reply” in FIG. 3). Whether or not there is (not shown) is determined and set by the communication providing apparatus 10. The reply destination communication ID fields 316a and 316b are set to “null” when the communication that has been performed is a new call, and when the reply is a reply, the communication ID that is the object of the reply is acquired, set, and viewed Acquires and sets a communication ID to be browsed. Hereinafter, data stored in the communication information database 18 is referred to as “communication information”.

  FIG. 4 shows an embodiment of the database structure of the extracted user information database 25. FIG. 4A shows an exemplary database structure 400a for a bulletin board, chat, blog, etc., and FIG. 4B shows an exemplary database structure 400b for a mailing list or the like. The extracted user information database 25 stores the extracted user management ID fields 402a and 402b that store data extracted from the user management ID fields 202a and 202b, and the data that is extracted from the address information fields 204a and 204b. Address information fields 404a and 404b, extraction date fields 406a and 406b in which data extracted from the subscription date fields 208a and 208b are stored, and data extracted from the user attribute information fields 210a and 210b are stored. Extracted user attribute information 408a and 408b. In extracted user attribute information fields 408a and 408b, extracted age fields 410a and 410b in which data extracted from age fields 212a and 212b are stored, and extracted occupation fields in which data extracted from occupation fields 214a and 214b are stored. 412a and 412b are included, but not limited thereto. In one embodiment of the present invention, data stored in these fields is extracted and stored by the history information collection unit 22 based on user information from the communication information notification unit 14. The user information may be transmitted from the communication information notification unit 14 to the history information collection unit 22 every time a communication occurs or may be transmitted periodically for updating. In another embodiment, user information may be transmitted on demand from the communication information notification unit 14 in response to a request from the history information collection unit.

  FIG. 5 shows an embodiment of the database structure of the communication history information database 27. FIG. 5A shows an exemplary database structure 500a in a bulletin board, chat, blog, etc., and FIG. 5B shows an exemplary database structure 500b in a mailing list or the like. The communication history information database 27 includes source user management ID fields 502a and 502b, reply destination user management ID fields 504a and 504b, genre / thread ID fields 506a and 506b, communication time fields 508a and 508b, and a communication ID. Fields 510a and 510b, reply destination communication ID fields 512a and 512b, and communication type fields 514a and 514b are included.

  Data stored in these fields is extracted from the communication information and user information by the history information collection unit 22 and stored. Specifically, the user management IDs 202a and 202b of the user who performed communication from the user information are extracted and stored in the transmission source user management ID fields 502a and 502b. In the reply-destination user management ID fields 504a and 504b, user management IDs 202a and 202b of users who are communication targets are extracted from the user information and stored. Genre / thread ID fields 506a and 506b, communication time fields 508a and 508b, communication ID fields 510a and 510b, return destination communication ID fields 512a and 512b, and communication type fields 514a and 514b include a genre / thread of communication from communication information. IDs 300a and 300b, communication times 308a and 308b, communication IDs 304a and 304b, reply destination communication IDs 316a and 316b, and communication types 314a and 314b are extracted and stored, respectively. Hereinafter, the data stored in the communication history information database 27 is referred to as “communication history information”.

  FIG. 6 shows an embodiment of the database structure of the information extraction parameter database 29. 6A shows an exemplary database structure 600a in a bulletin board, chat, blog, etc., and FIG. 6B shows an exemplary database structure 600b in a mailing list or the like. The information extraction parameter database 29 includes user management ID fields 602a and 602b, genre / thread ID fields 604a and 604b, update date fields 606a and 606b, and communication type fields 608a and 608b. The communication type fields 608a and 608b include the number of new calls 610a and 610b, the number of replies 612a and 612b, and the number of views 614a. These items increase or decrease according to the classification of the communication types 514a and 514b However, it is not limited to these. Data stored in these fields is extracted and stored by the history information collection unit 22 as follows. User management ID fields 602a and 602b are acquired from user information, and genre / thread ID fields 604a and 604b are acquired from communication information. Next, the update dates 606a and 606b are updated based on the communication times 308a and 308b of the communication information, and the new outgoing numbers 610a and 610b, the reply numbers 612a and 612b, and the browse numbers 614a and 614b Aggregated and recorded. Hereinafter, the data stored in the information extraction parameter database 29 will be referred to as “information extraction parameters”.

  Next, with reference to FIG. 7, the flow of information extraction processing in a bulletin board, chat, blog, etc. will be described as one embodiment according to the present invention. It should be noted that the following description applies similarly to the processing in the mailing list with reference to FIGS. 2 (b) to 6 (b). It will also be apparent to those skilled in the art that the present invention is applicable to other communications provided over a network.

  FIG. 7 is a flow of processing when a communication event occurs from the user terminal 40a to the communication providing apparatus 10. As shown in the figure, a communication event is received by the communication control unit 12 by a reply from the user terminal 40a (processing 1). The communication control unit 12 compares the address information and password of the user of the user terminal 40a included in this communication event with the address information 204a and password 206a stored in the user information database 16 and authenticates them (Process 2). Furthermore, the communication information included in the communication event and the communication information obtained from the related information are registered in the communication information database 18 (process 3). The communication control unit 12 displays the contents of communication and related information disclosed as a result of communication on the user terminal 40a (process 4).

  The communication control unit 12 notifies the communication information notification unit 14 of the occurrence of communication together with the communication information and the corresponding user information (Process 5). Thereafter, the communication information notification unit 102 notifies the history information collection unit 22 of the received communication information and corresponding user information (processing 6). The history information collection unit 22 collates with the extracted user information database 25 based on the user management ID 202a of the corresponding user information, and registers the extracted user information 400a for a new user (processing 7). The history information collection unit 22 further processes user information and communication information (process 8), and registers data in the communication history information database 27 (process 9). Further, the history information collection unit 22 registers and updates data in the information extraction parameter database 29 from the user information and communication history information (processing 10). In this process 10, the item (for example, the number of new calls) corresponding to each genre / thread ID 604a targeted by the communication is incremented with respect to the corresponding user management ID 602a of the user who performed the communication. The update date 606a is updated.

  FIG. 8 shows a flow of processing for extracting and providing information in the information extracting apparatus 20 as an embodiment according to the present invention. As shown in the figure, according to the information extraction event performed from the communication information notification unit 14 to the information extraction unit 24 (Process 11), the information extraction unit 24 uses the extracted user information database 25 and the information extraction parameter database 29 from A beginner user is extracted (process 12).

  In this process 12, for example, when the difference in the number of days between the extracted subscription date 406a and the current date is smaller than a predetermined value, it is determined that the user is a novice user. Alternatively, the difference in the number of days between the extracted subscription date 406a and the current date is greater than or equal to a predetermined value, but the communication type 608a is a predetermined standard over all genres / thread IDs 604a or for a specific genre / thread ID. It can also be determined that the user is a novice user when the value (for example, the number of new outgoing calls or the number of replies is less than 3 times and the number of browsing times is less than 10). In addition, items such as scores from other users and certification from the administrator of the communication providing device 10 can be added to the information extraction parameter database 29, and a novice user can be determined using these criteria. . Furthermore, a novice user can also be determined by combining a plurality of the above criteria.

Further, the information extraction unit 24 extracts users and user groups based on user characteristics such as a higher-order user, important user, and specific important user from the extracted user information 400a and communication history information 500a (process 13). For example, using the following equation, the highest user with the highest Cd (i) and the user with the highest Cd (i) are regarded as the upper user group based on the highest value.

Cd (i) = d (i) / (g-1) (Formula 1)

Here, Cd (i) is a value obtained by standardizing the centrality based on the order of the user i, and the order in this expression is the number of different users who have established communication. d (i) is the order of user i, and g is the total number of users (that is, g-1 is the maximum order that one user can have). Therefore, Cd (i) according to this equation is a value between the minimum value 0 and the maximum value 1 for each user. Here, establishment of communication means the following cases.

  As one-way communication, new transmission, reply, reception (when a reply is received), browsing (when viewed), and / or viewed (when viewed) can be regarded as establishment of communication. Also, as a direct two-way communication, when a reply is received from a user to his / her message and a reply is made to the reply message, or a reply is made to a message and the reply message is When a reply is received, it can also be regarded as establishment of communication. In addition, as indirect two-way communication, one or a plurality of users exist between a plurality of messages (new transmission, reply, or reception) within each genre, within each thread, or within each branch. If communication is established with one or a plurality of users, or if a message is present among a plurality of messages of a certain user, communication with the user may be regarded as established. it can. Further, as indirect communication, when messages of a plurality of users coexist in each genre, each thread, or each branch, it can be considered that communication is established between the plurality of users. In addition, in all the cases described above, a threshold can be provided for the number of established communications with other users, and the case where the threshold is exceeded can be counted as established communications.

  Furthermore, in the upper user group specified by Expression 1, from the communication history information 500a, among the users whose number of genre / thread IDs 604a that have made and / or received 612b a number of times greater than or equal to a predetermined value, A user having the largest number of users who have received a reply is designated as the most important user, and a user within a predetermined number or a predetermined ratio from the top in the descending order of the number of users who have received a reply, or a user having a predetermined value or more as an important user group. In the upper user group, the user having the largest number of users who have transmitted and / or received for a specific genre / thread ID from the communication history information 500a is identified as the most important user, and the transmission and / or Alternatively, a user within a predetermined number or within a predetermined ratio from the top, or a user having a predetermined value or more is set as a specific important user group in descending order of the number of received users.

  In the upper user group, except for the important user group (including the most important user) and the specific important user group (including the specified most important user), transmission and / or reception of the number of times greater than or equal to a predetermined value from the communication history information 500a. The number of genres / thread IDs that have been transmitted is equal to or greater than a predetermined value, and within a predetermined number of people from the top in descending order of the number of transmissions and / or receptions in the genre / thread ID that has performed transmission and / or reception greater than the predetermined value, or Users within a predetermined ratio or more than a predetermined value are extracted as quasi-important user groups.

  In addition, a user who has received a reply from the communication history information 500a up to a predetermined number of times and who has received the largest number of users who have received the reply has received the reply. Users with a predetermined number of users or within a predetermined ratio from the top in the descending order of the number of users, or users with a predetermined value or more are extracted as a provision user group. Furthermore, for a novice user who has replied to two or more genres, threads or branches from the communication history information 500a, within a predetermined number or within a predetermined ratio from the top in descending order of the number of replies to the initial reply by the novice user Or, a user having a predetermined value or more is set as a follow user group.

  Moreover, in users other than the beginner user group and the upper user group, users having similar user characteristics that match the conditions of the important user group, the specific important user group, the semi-important user group, the opportunity providing user group, and the follow user group are selected. It extracts as each similar user group (process 14). Further, the communication history information 500a is used to count the number of established communications between the users and extract the degree of connection between the users (process 15). The degree of connection between the users can also be totaled based on the number of times of coexistence with a specific user in genre units, thread units, or branch units. The information extraction unit notifies the communication information notification unit 14 of the above-described user characteristics and connection degree (process 16). Through such processing, user characteristics, similarity of user characteristics, and degree of connection can be extracted.

  In addition, according to an embodiment of the present invention, it is possible to set a certain period and extract the user characteristics, the similarity of user characteristics, and the degree of connection within the period. For example, it is possible to extract various characteristics of the user in a certain past period, or to set periods such as the past week, the past January, the past half year, and the past year based on the present. Thereby, the temporal change of the user characteristics can be examined.

  Furthermore, according to an embodiment of the present invention, the communication history information 500a includes a user with the largest number of browses within a predetermined difference with reference to the highest number of browsed users and the value with the highest number of browses. The users who are to be viewed can be set as the upper-viewed user group. Further, according to an embodiment of the present invention, as a communication type, a vote is added, a user who has the highest evaluation value by voting is the highest voting user, and a value having the highest evaluation value by voting is used as a reference. Users included in the difference can be set as a voting upper user group.

  In view of the many possible embodiments to which the principles of the present invention may be applied, the embodiments and examples described herein in connection with the drawings are merely illustrative and are intended to limit the scope of the invention. It should be recognized that it should not be interpreted. For example, as will be appreciated by those skilled in the art, the embodiments illustrated herein may be modified in configuration and detail without departing from the spirit of the invention. Moreover, the illustrative steps may be changed, supplemented, and / or reordered without departing from the invention. Accordingly, the invention described herein is contemplated by all such embodiments that fall within the scope of the appended claims and their equivalents.

It is a functional block diagram which shows the whole structure in one Example of the information extraction method by this invention. It is a figure which shows the data structure of the exemplary user information by this invention, (a) is a figure in the case of a bulletin board, a chat, or a blog, (b) is a figure in the case of a mailing list. 2A and 2B are diagrams illustrating a data structure of exemplary communication information according to the present invention, in which FIG. 1A is a diagram for a bulletin board, chat, or blog, and FIG. 2B is a diagram for a mailing list. It is a figure which shows the data structure of the example extraction user information by this invention, (a) is a figure in the case of a bulletin board, a chat, or a blog, (b) is a figure in the case of a mailing list. It is a figure which shows the data structure of the exemplary communication history information by this invention, (a) is a figure in the case of a bulletin board, a chat, or a blog, (b) is a figure in the case of a mailing list. It is a figure which shows the data structure of the parameter for an exemplary information extraction by this invention, (a) is a figure in the case of a bulletin board, a chat, or a blog, (b) is a figure in the case of a mailing list. It is a figure which shows the flow of a process from generation | occurrence | production of a communication event to storing various information in one Example by this invention. It is a figure which shows the flow of a process until it extracts predetermined information from the various information of FIG. 7 stored in one Example by this invention. It is a figure which shows the example of a tree-like thread | sled which has three branches in a communication provision apparatus.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Communication provision apparatus 12 Communication control part 14 Communication information notification part 16 User information database 18 Communication information database 20 Information extraction apparatus 22 History information collection part 24 Information extraction part 25 Extraction user information database 27 Communication history information database 29 Parameter database for information extraction 30 Network 40a, 40b, 40n User terminal 42a, 42b, 42n Transceiver 100 Thread 102, 104 Branch point

Claims (15)

An information extracting method in an information extracting device for extracting user characteristics from user information registered in the communication providing device and history information of communication between the users in a communication providing device for providing communication between users via a network And the information extraction device comprises:
Obtaining the number of different users with whom communication has been established as an order based on at least one of the user information and the history information of the communication;
Extracting the users whose orders satisfy a predetermined criterion as a higher-level user group.   The information extracting device has a number of genres in which the order of the communication satisfies a predetermined criterion based on at least one of the user information and the communication history information, and the number of users who have received a reply is The information extracting method according to claim 1, further comprising: extracting a user satisfying a predetermined criterion as an important user group.   The information extraction device extracts, as a higher-level user group by genre, users who satisfy the predetermined standard for the order of the communication in a specific genre based on at least one of the user information and the communication history information. The information extraction method according to claim 1, further comprising: An information extracting method in an information extracting device for extracting user characteristics from user information registered in the communication providing device and history information of communication between the users in a communication providing device for providing communication between users via a network And the information extraction device comprises:
The information extracting method according to claim 1, further comprising: extracting a user who does not satisfy a predetermined criterion as a beginner user group based on at least one of the user information and the communication history information.   The information extraction device extracts, as a trigger providing user group, a user whose number of users who have received a reply from the user of the beginner user group satisfies a predetermined criterion based on at least one of the user information and the communication history information The information extracting apparatus according to claim 4, further comprising a step of:   The information extraction device is a user who follows a user whose number of users who have replied to a reply from a user in the beginner user group satisfies a predetermined criterion based on at least one of the user information and the history information of the communication The information extracting method according to claim 4, further comprising a step of extracting as a group. An information extracting method in an information extracting device for extracting user characteristics from user information registered in the communication providing device and history information of communication between the users in a communication providing device for providing communication between users via a network And the information extraction device comprises:
Extracting the degree of connection between users by counting the number of times communication has been established for each user based on at least one of the user information and the communication history information. Information extraction method. An information extracting method in an information extracting device for extracting user characteristics from user information registered in the communication providing device and history information of communication between the users in a communication providing device for providing communication between users via a network And the information extraction device comprises:
Based on at least one of the user information and the communication history information, for each user, by counting the number of times each set of users coexisted in each genre, each thread, or each branch, between each user An information extraction method comprising a step of extracting a degree of connection. An information extracting method in an information extracting device for extracting user characteristics from user information registered in the communication providing device and history information of communication between the users in a communication providing device for providing communication between users via a network And the information extraction device comprises:
A step of extracting, as a higher-viewed-number user group, a user whose number of users viewed from other users satisfies a predetermined criterion for each user's message based on at least one of the user information and the communication history information An information extraction method comprising: An information extracting method in an information extracting device for extracting user characteristics from user information registered in the communication providing device and history information of communication between the users in a communication providing device for providing communication between users via a network And the information extraction device comprises:
Based on at least one of the user information and the history information of the communication, a step of extracting, as a higher-ranking user group, a user whose score voted by another user satisfies a predetermined criterion for each user's message An information extraction method characterized by that.   4. The information extraction apparatus according to claim 1, wherein the order is based on a ratio between a number of different users who have established the communication and a total number of users to be extracted in the history information of the communication. Information extraction method described in 1.   In the information extraction device, the establishment of the communication is performed by a second user returning a second message to the first message of the first user, and the first user responding to the second message. The information extraction method according to claim 1, wherein the information extraction method is achieved between the first and second users when a third message is returned to the user.   In the information extracting apparatus, the communication is established when the one or more users exist between the messages of the first user in each genre, each thread, or each branch. The information extraction method according to claim 1, wherein the information extraction method is performed between the user and the one or more users.   14. The information extracting method according to claim 1, wherein the extracting device is based on the user information or the communication history information for a predetermined period. A communication providing apparatus for providing communication between users via a network, wherein the information extracting apparatus extracts user characteristics from user information registered in the communication providing apparatus and communication contents provided by the communication providing apparatus. There,
Extracting predetermined user information from the user information;
Extracting predetermined history information from the content of the communication,
A history information collection unit that aggregates predetermined parameter information from the content of the communication;
An extracted user information database in which the predetermined user information extracted by the history information collection unit is stored;
A communication history information database in which the predetermined history information extracted by the history information collection unit is stored;
An information extraction parameter database in which the predetermined parameter information tabulated by the history information collection unit is stored;
Extracting a user whose predetermined information in the extracted user information database and the information extraction parameter database does not satisfy a predetermined standard as a beginner user group,
An information extraction unit comprising: an information extraction unit that extracts, as upper user groups, users whose number of users who have communicated so far from information in the communication history information database satisfies a predetermined standard apparatus.

Images