JP2008210392A - Electronic bulletin board system, and recording medium recorded with program for constructing electronic bulletin board system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an electronic bulletin board system capable of making accurate retrieval precision of a description with an interest of a user, by comparing respectively the plurality of descriptions with a plurality of user profiles, and allowing automatic notification. <P>SOLUTION: The description transmitted to a server 1 is recorded in a description DB 7 in a description recording part 6. A content summarizing part 8 retrieves the description DB 7, records a summary of the description into a description content DB 9, and records a summary of a content of a thread containing the description into a subject content DB 10, when reception of a new description is transmitted from the description recording part 6. A content comparison part 11 compares the updated thread with a content of other thread, when the subject content DB 10 is updated, and transmits descriptions of introducing the threads each other as to the related threads, to a communication part 5, to be issued to each client. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

The present invention relates to an electronic bulletin board system, and more particularly to an electronic bulletin board system that automatically notifies an article of interest to a user and a recording medium that records a program for constructing the electronic bulletin board system.

Conventionally, in an information sharing type system such as an electronic bulletin board system in which many articles containing useful information have been posted from multiple people via a network, for example, in order to efficiently search for articles existing on the electronic bulletin board, posting The method of displaying the classification by the title of the posted article and the contents of the follow for a certain posted article are displayed in a branching manner so that the target article can be easily searched.

Another method for efficiently providing shared information in an information sharing type system is disclosed in Japanese Patent Laid-Open No. 9-26970, “Execution Method and Device by Computer for Searching Information” (hereinafter, Conventional Document 1). ) Is known. The method of this prior art document 1 creates a user-defined profile that specifies what kind of information the user is interested in, searches the database based on the contents of the user-defined profile, and uses this search result as a client. To present to the side.

As another technique, a technique described in Japanese Patent Laid-Open No. 3-70233 “Electronic Bulletin Board Device” (hereinafter, Conventional Document 2) is also known. In the technique of the conventional document 2, when a new article is posted on the electronic bulletin board system, the electronic bulletin board system notifies the user that a new article has been written by e-mail. A method of making it unnecessary is disclosed.
JP-A-9-26970 Japanese Patent Laid-Open No. 3-70233

By combining these technologies described in the conventional document 1 and the conventional document 2 with the conventional electronic bulletin board system, it seems that only the articles that the user is interested in can be notified. In the combined method, the user profile must be defined in advance. In addition, each posted article is compared with the user profile to determine whether the user is interested, so articles that the user is not interested in are notified by keywords included in the article. There is a problem that an article that the user is interested in is not notified because the keyword is not included.

SUMMARY OF THE INVENTION An object of the present invention is to provide an electronic bulletin board system and an electronic bulletin board that can automatically notify and automatically notify an article of interest to a user by comparing a plurality of articles with a plurality of user profiles. It is to provide a recording medium in which a program for constructing a system is recorded.

In the electronic bulletin board system according to the first aspect of the present invention, an article recording unit that records an article transmitted from a client in an article DB, and a content that summarizes the contents of the article based on the article recorded in the article DB are summarized. A content summary unit, an article content DB that records the article content summarized by the content summary unit, and a topic content DB that records the article content summarized by the content summary unit for each thread of the article recorded in the article DB; And a content comparison unit for comparing the content of each thread recorded in the topic content DB.

Explaining the operation of the first invention, the article sent from the client to the communication unit of the server is added with the ID used for the article in the article recording unit, the user name sent, and the information about which article to follow When the content summary unit is notified that a new article has been received from the article recording unit, it searches the article DB, records the summary of the article in the article content DB, and stores the article. A summary of the content of the included thread is recorded in the topic content DB. When the topic content DB is updated, the content comparison unit compares the updated thread with the content of other threads, and communicates and issues articles introducing each other thread to the related threads. .

Next, in the electronic bulletin board system according to the second invention of the present invention, in addition to the configuration of the first invention, the user model expressing the posted content of the user, the content recorded in the user model, and the user posted A user model update unit that updates the user model using the article content, a user relationship DB that manages whether the users know each other, and a user model comparison unit that compares the contents of the user models.

The operation of the second invention will be described. When the user model update unit is notified that the article is received from the article recording unit, the user model of the user who transmitted the article is updated from the content of the received article. The user model comparison unit compares a user model of each user and searches for a plurality of users having similar user models. Furthermore, users who have not been written in the same thread from the user relation DB and have not received introductions from the system are extracted.

Since the users extracted in this way are likely not to know each other, the system introduces each user to each other. Users who have similar interests get to know each other and exchange information, thereby making it easier to search for only articles of interest.

Further, in the electronic bulletin board system of the third invention, in addition to the structure of the second invention, the article content DB
And a topic UM comparison unit that compares the contents of the user model.

Explaining the operation of the third invention, the topic UM comparison unit compares the user model of each user with the content of the thread recorded in the topic content DB, and for the client of the user who is likely to be interested, The ID of the latest article constituting the thread is transmitted.
In this way, the user can be automatically notified from the server of topics that he / she may be interested in by simply transmitting an article.

According to the present invention, by comparing a plurality of articles and a plurality of user profiles, respectively, it is possible to provide an electronic bulletin board system that makes the search accuracy of articles that the user is interested in more accurate and automatically notifies. it can.

Next, the electronic bulletin board system of the present invention configured as described above will be described in more detail with reference to the drawings.

(Embodiment in the First Invention) FIG. 1 is a block diagram showing an embodiment of the configuration of the electronic bulletin board system in the first invention of the present invention.

Referring to FIG. 1, the first invention of the present invention comprises one server 1 and a plurality of clients 2.

The server 1 may be configured by a recording medium 3 on which an electronic bulletin board server program is recorded.

The client 2 may be configured by the recording medium 4 on which the electronic bulletin board client program is recorded. In addition, the client 2 may be in any form that can refer to the content of the article on the electronic bulletin board recorded in the server 1 or the like or can post the article, and is not particularly defined in the present invention. Also, refer to the contents of the electronic bulletin board from the server 1 and other servers,
It is a form to receive a program that can be posted. For example, an electronic bulletin board is WWW (Wor
If it is constructed on (ld Wide Web), this recording medium is a browser program capable of browsing the WWW. The client program of the electronic bulletin board system may be recorded on the recording medium 4 or only partially recorded and may be received from a server or the like as necessary.

The server 1 has a communication unit 5 that communicates with the client 2, an article recording unit 6 that records an article transmitted from the client 2 in the article DB 7, and a content for each thread based on the article recorded in the article DB 7. The content summarizing section 8 to be summarized, the article content DB 9 for recording the article content summarized by the content summarizing section, the topic content DB 10 for recording the summary of the topic contents summarized by the content summarizing section, and the article content DB A content comparison unit 11 that compares the content of each thread is included.

Here, the present invention will be described in detail with specific operation examples. Assume that an article provided from the server 1 is displayed as shown in FIG. 2 in the client 2 used by the user E. In the example of Fig. 2, in response to the article "It seems that the new video card scheduled to be released from Company X seems to be fast", "There was an evaluation article in a magazine called Y."
It seems good to leave the company. Are followed, and they are displayed in a branching manner.

Then, a case where user E has posted a follow-up article “The price seems to be cheap according to that article” in response to an article “An evaluation article was published in a magazine called Y” will be described. The articles displayed on each client are part of the articles recorded in the server 1, and the server 1 stores some or all of the articles posted so far in the article DB.
7 is recorded. For example, the server 1 may set a period for storing articles, and articles that have expired may be deleted. In addition to the article content, for example, information such as the article poster and the article to be followed is added to the article received by the server 1. In addition to these, information such as the posting date and the title of the article may be added. Since these functions are functions realized in the prior art, detailed description will not be given. Further, the article DB 7 records the article content and the added information together with the article ID assigned to each received article.

Here, for the sake of explanation, the article DB 7 records items such as the article ID, the name of the user who posted the article, the article to be followed, and the article content in a table format as shown in FIG. To do. In FIG. 3, the articles shown in FIG. 2 are recorded as article IDs 100, 101, and 102. In addition, for example, an article with article ID 10 is recorded.

If user E posts a follow-up article with the article ID 101, “Your evaluation article was in a magazine called Y”, “The price seems to be cheaper according to that article.” E inputs an article from the client 2 and transmits it to the server 1, and the server 1 receives the article transmitted from the client 2 in the communication unit 5 (502). This communication unit 5
The article received at is delivered to the article recording unit 6 (506).

The received article is assigned an article ID in the article recording unit 6, and the user name and follow-up ID
Are recorded in the article DB 7 together with information such as the article content (607). In this example, when the article DB 7 is in the state as shown in FIG. 3, if the user E posts an article “The price seems to be cheap according to that article”, the article recording unit 6 displays the article DB 7 in FIG. 4. It is updated as follows.

When the received article is recorded in the article DB 7 (607), the article recording unit 6 sends the recorded content to the communication unit 5.
(605) and transmitted to all the clients connected to the server 1 (205, etc.) to update the display state of the clients. However, instead of a method of sending a newly posted article from the server 1 to the client, a method of sending back the content recorded in the article DB 7 only when a newly registered article is requested from the client may be used. However, the present invention is not limited to this form.

The article recording unit 6 records the received article in the article DB 7 (607) and simultaneously transmits the article ID of the received article to the content summary unit 8 (608). The content summary unit 8 retrieves the received article from the article DB 7 (807), and records the summary of the article content in the article content DB 9 (809).

The summary of the article content is expressed, for example, by a set of several keywords. As a means for generating a summary from the content of an article, for example, a method of extracting a keyword by performing morphological analysis can be used as disclosed in JP-A-8-221440.

For example, a description will be given assuming that the content summary unit 8 extracts nouns as keywords from the article as a summary of the article content. When the article DB 7 is as shown in FIG. 3, the article content DB 9 records the article ID and the extracted keyword for each article in a table format as shown in FIG.

When the article recording unit 6 informs the content summary unit 8 that a new article with the article ID 103 has been recorded in the article DB 7 (608), the content summary unit 8 receives an article ID of 10 from the article DB 7.
3 is retrieved (809), nouns are extracted as keywords, and the article content DB 9 is updated as shown in FIG. 6 (809).

When the summary of the received article is recorded in the article content DB 9 (809), the content summary unit 8 searches the article DB 7 for a thread having the article as a terminal (807). Here, the article at the end refers to the article that is not followed, the thread that ends at an article is the article that the article at the end follows, and the article that the article follows It refers to a series of articles that date back.

For example, the article with the article ID 103 newly added to the article DB 7 is the last article, and the article that has followed the follow-up destination of this article, that is, the article that is the follow-up article with the article ID 103 in the example of FIG. An ID 101 and an article ID 100 that is a follow-up article of the article ID 101 are threads that end at the article ID 103.

When the content summary unit 8 searches for a thread that ends in the article added to the article DB 7 (8
07), the keyword of the article included in the thread is searched from the article content DB 9 (809).
), The topic content DB 10 is updated as a keyword representing the content of the thread (8010).
).

For example, as shown in FIG. 7, the topic content DB 10 records the article IDs at the tips and ends of threads and keywords representing the contents of the threads in a table format. The keyword representing the content of the thread is determined by taking the logical sum of the keywords of the articles included in the thread.

As a means for determining a keyword representing the content of a thread from keywords of articles included in the thread, in addition to a method of taking a logical sum, keywords included in articles that are older than a certain degree (for example, articles that have been followed more than 10 times) There are a method of ignoring, a method of determining the total number of keywords representing the contents of the thread, and prioritizing a keyword included in a new article or a keyword appearing multiple times.

Here, a case will be described in which the article ID 103 is added to the article DB 7 when the topic content DB 10 is recorded as shown in FIG. The content summary unit 8 records the article with the article ID 103 in the article content DB 9 (908), and searches the article DB 7 for a thread having the article with the article ID 103 as a terminal (807). Articles having article IDs 100, 101, and 103 are searched as threads to be obtained.

Since the article with the article ID 103 is a follow article with the article ID 101 when the article DB 7 is examined (807), the content summary unit 8 has the terminal article ID 101 of the thread of the topic content DB 10 as 101.
Is updated to 103 (8010). Article IDs 100 and 101
, 103 is searched from the article content DB 9 (809), and logical OR is taken and updated as a keyword representing the content of the thread whose end article ID is 103 in the topic content DB 10 (8010). When the article with the article ID 103 is added in this way, the topic content DB 10 is updated from the state shown in FIG. 7 to the state shown in FIG.

When the topic content DB 10 is updated, the content comparison unit 11 compares a keyword representing the topic content of the updated thread with a keyword having a different article ID at the tip in another thread, and searches for a similar one. . The determination of whether the contents are close is, for example, when there are several or more (for example, two or more) common keywords included in both, or when the total number of keywords representing the thread contents is determined Can be based on a case where a certain percentage of keywords is included (for example, 10% or more). Further, even if the keywords are not exactly the same, it is possible to use a means for determining that the keywords are the same using a synonym dictionary or a thesaurus dictionary.

Whether or not the contents in the present embodiment are similar will be described as being determined based on whether or not there are two or more common keywords included in both.

When the topic content DB 10 is updated (8010), the content comparison unit 11 reads the topic content DB 10.
The keyword representing the content of the updated thread is retrieved from (11010). Article ID1
When the article 03 is recorded in the article DB 7 and the article content DB 9 and the topic content DB 10 are updated, the keyword “X company, sale schedule, video card, Y indicating the content of the thread whose end article ID is 103” is shown. , Magazine, evaluation article "is searched.

Since the thread whose end article ID is 103 has the article ID at the tip of the thread as 100, the content comparison unit 11 is a thread whose tip article ID is not 100, and “X company, planned to be released, video card, A thread including two or more keywords such as “Y, magazine, evaluation article” is searched from the topic content DB 10 (11010). As a thread that satisfies the condition, a thread having an article ID of 10 at the end of the thread is searched, and the content comparison unit 11 is a thread having a content close to the thread having the article ID of 103 and the thread having the article ID of 10 at the end. It is judged that.

This example describes the case where only one thread is found as a close thread,
The following description is the same when a plurality of threads are searched.

When a thread with similar content is searched, the content comparison unit 11 posts an article introducing each other to each thread. However, if an article introducing each other has already been posted, it will not be posted again. For example, 1 thread for posting articles that introduce each other
You may make it contribute again only when there is a certain interval, such as when following 0 or more.

For example, the posting user name is “System”, the article content is “following the article with article ID 10”, and there is an article such as “There is an article with content such as a video card and price in a thread including an article with ID 103”, and the article ID 103. The article content is "
There is an article with contents such as video card and price in a thread containing an article with ID 10.
Is generated and passed to the communication unit 5 (1105).

The communication unit 5 handles the article generated by the content comparison unit 11 in the same manner as the article transmitted from the client. That is, it is passed to the article recording unit 6 (506) and recorded in the article DB 7 (607)
The recorded content is returned to the communication unit (605) and transmitted to all the clients connected to the server 1 (502, etc.) to update the display state of the client.

However, the article generated by the content comparison unit 11 is registered as having no keyword representing the article content when recorded in the article content DB 9 (809).

Therefore, an article such as “There is an article with a content such as a video card and price in a thread containing an article with ID 103” and an article content “following the article with article ID 103 is a video card in a thread containing an article with ID 10”. When an article such as “There is an article with a price” is generated by the contents comparison unit 11, the article DB 7 is updated as shown in FIG. 9 (607), and the article contents DB 9 is updated as shown in FIG. 10 (809). ). Article content D
By updating B9, the content summary unit 8 updates the article ID at the end of the thread in the topic content DB 10 as shown in FIG. 11 (8010).

In this way, it is not just a comparison of articles, but the relation of threads is judged from the contents of multiple articles that are in a follow relationship, and related threads for multiple users who are reading different threads Can be notified.

(Second Embodiment) Next, a second embodiment of the present invention will be described.

In the electronic bulletin board system according to the embodiment of the second invention, as shown in FIG. 12, a user model 13 expressing the user's interest in addition to the components of the electronic bulletin board system according to the embodiment of the first invention. A user model update unit 12 that updates the user model 13 using the content recorded in the user model 13 and the content of an article posted by the user, a user relationship DB 14 that manages whether the users know each other, and a user And a user model comparison unit 15 for comparing contents recorded in the model 13.

The user model 13 is created by the user model updating unit 12 and manages, for example, as shown in FIG. 13, a user name and a keyword representing the posted article content of the user in a tabular format. The keyword representing the content of the posted article of the user is determined by taking the logical sum of all the keywords included in the articles that the user has posted so far. Also,
The user model 13 can also employ a method that allows each user to register keywords of interest to the user.

The user model update unit 12 determines a keyword representing the content of the posted article of the user. In addition to the method of calculating the logical sum of all the keywords included in the articles that the user has posted so far, for example, the method of calculating the logical sum of the keywords included in only the latest 10 articles posted by the user, There is a method of determining the total number of keywords to be recorded, and deleting old keywords every time a new keyword is added.

For example, as shown in FIG. 14, the user relationship DB 14 manages whether or not each user knows each other in a table format. Assume that users who have posted articles in the same thread know each other, and FIG. 14 shows that user B and user C, user B and user D, and user C and user D know each other. ing.

Here, the user model 13 is recorded as shown in FIG. 13, the article DB 7 is recorded as shown in FIG. 3, the article content DB 9 is recorded as shown in FIG. 5, and the topic content DB 10 is recorded as shown in FIG. The operation of the embodiment of the second invention will be described for the case where an article with ID 103 is posted.

In the electronic bulletin board system according to the second embodiment, the operations described below are added in addition to the operations described in the first embodiment.

The article with the article ID 103 received by the communication unit 5 is transferred to the article recording unit 6, and the article DB 7, the article content DB 9, and the topic content DB 10 are respectively shown in FIG. It is updated as shown in FIGS.

In the embodiment of the second invention, it is further notified that the article ID of the article newly posted from the article recording unit 6 to the user model updating unit 12 is 103 (6012).

The user model update unit 12 checks whether the user E is recorded in the user model 13 (12013). If it has already been recorded, the keyword included in the article posted from the article content DB 9 is checked (1209), and logically added with the already registered posted content keyword and updated (12013). If not recorded, user E is newly registered (1
2013), the keyword included in the article posted from the article content DB 9 is checked (1209).
), And newly registered in the posted content keyword (12013). The user model 13 is updated as shown in FIG. 15 from the state of FIG.

Further, the article recording unit 6 searches the article DB 7 that the users who have posted to the thread including the article with the article ID 103 posted by the user E are B, C, and D (607), and the user E is stored in the user relation DB 14 with the user E. (15014) and updated as shown in FIG.

When the user model 13 is updated, the user model comparison unit 15 searches the post content keyword of the user E updated from the user model 13 (15013) and compares it with the post content keyword of another user.

Judgment of whether or not the posted content is similar is, for example, when there are several or more (for example, two or more) common keywords included in either, or by determining the total number of posted content keywords, It can be based on the case where a certain percentage or more (for example, 10% or more) is included. Further, even if the keywords are not exactly the same, it is possible to use a means for determining that the keywords are the same using a synonym dictionary or a thesaurus dictionary.

Whether the posted contents in this example are close or not is based on 2 common keywords included in both
A description will be given assuming that the determination is based on whether or not the number is more than one.

When the user model 13 is updated as shown in FIG. 15, the user model comparison unit 15
Since keywords such as “article, price” are common, it is determined that the user A is close to the content posted by the user E. In this example, a case where only one person is searched for as a user whose posting content is close is described. However, the following description is the same when a plurality of users are searched.

When a user whose posting content is close is searched, the user model comparison unit 15 searches for the user relationship DB1.
4 is searched (15014), and the users whose posting contents are close and do not know each other are searched for. When the user relationship DB 14 is as shown in FIG. 16, the user A and the user E are regarded as not knowing each other. Therefore, in the user model comparison unit 15, the user A and the user E are close to each other. Discovered as users who do not know each other.

Next, the user model comparison unit 15 transmits a message for introducing each other to the clients of the user A and the user E via the communication unit 5 (1505) (502 and the like). Since the information is not valuable to users other than A and E, it is not posted as an article, but is transmitted as a message using means that can be viewed only by a specific user. As means for sending a message, e-mail or the like can be used, for example.

In this example, for user A, “A person named E makes a posting that includes keywords such as articles and prices.” "Users A and E
To the client.

Users who are close to each other and who do not know each other know each other and exchange information, so that they can teach articles that only one of them knows. In addition, since the user model is automatically updated by the posted article, the user does not need to update the user model himself / herself, and the burden on the user can be reduced.

  (Third Embodiment) Next, a third embodiment of the present invention will be described.

In the electronic bulletin board system according to the embodiment of the third invention, as shown in FIG. 18, each thread recorded in the topic content DB 10 in addition to the components of the electronic bulletin board system according to the embodiment of the second invention. And a topic UM comparison unit 16 that compares a keyword representing the content of the posting for each user recorded in the user model 13.

The operation of the embodiment of the third invention will be described by taking as an example the case where the article having the article ID 103 described above is received.

In the electronic bulletin board system according to the third embodiment, the following operations are added in addition to the operations described in the first and second embodiments.

By receiving the article with the article ID 103, the topic content DB 10 is as shown in FIG.
Further, it is assumed that the user model 13 is updated as shown in FIG. The topic UM comparison unit 16 includes a keyword representing the content of each thread recorded in the topic content DB 10 and the user model 13.
Are compared with the keywords representing the posted contents for each user recorded in the list (16010, 1601).
3) It is determined whether the topic content and the posted content of the user are close.

Judgment whether the topic content and the posted content of the user are similar, for example, when there are several or more (for example, two or more) common keywords included in both, recorded in the topic content keyword and the posted content keyword It is possible to determine the total number of keywords to be determined and use a case in which more than a certain proportion of keywords (for example, 10% or more) is included. Further, even if the keywords are not exactly the same, it is possible to use a means for determining that the keywords are the same using a synonym dictionary or a thesaurus dictionary.

In the following description, it is assumed that whether the topic content and the posted content are close is determined based on whether there are two or more common keywords included in both.

When the topic content DB 10 is as shown in FIG. 11 and the user model 13 is as shown in FIG. 15, the topic UM comparison unit 16 determines the posting content keyword of the user A in FIG. 15 and the thread end in FIG. Since the keywords such as “video card, price” are common among the content keywords with the article ID 105, it is determined that the posted content of the user A is close to the topic content of the thread with the end article ID 105. In addition, user B's posted content and the last article I
The topic content of the thread whose D is 105, the topic content of the thread whose post ID of the user B and the end article ID is 102, the topic content of the thread whose post content of the user C and the end article ID is 105, the user E And the topic content of the thread whose end article ID is 105 are determined to be close.

With respect to a thread and a user who are determined to be close to the topic content and the posted content, it is checked from the article DB 7 whether or not the user has posted an article to the thread (1607).
If the message has not been posted, the topic UM comparison unit 16 sends a message introducing the thread to the user's client via the communication unit 5 (1605) (502, etc.). In this example, for the article ID 105 at the end of the thread, since the user A has not posted an article to the thread, a message introducing the thread having the article ID 105 at the end of the thread is transmitted to the user A. . For threads and users that are judged to be close to other topic content and posted content, the thread is introduced because it can be understood that the user has posted to the thread by examining the article DB 7 (1607). Do not send messages.

Further, the topic UM comparison unit 16 records which thread is introduced to which user so that the same thread is not introduced to the same user. Since the information is not valuable to other users other than the user, the information is not posted as an article, and is transmitted as a message using a means that only the user can see. As means for sending a message, e-mail or the like can be used, for example.

In this example, a case where only one set of users and threads is searched is described, but the following description is the same when a plurality of sets of users and threads are searched.

The topic UM comparison unit 16 sends a message to the user A, for example, “A thread that includes an article whose article ID is 105 includes a keyword such as a video card and a price”. 5 (1605) and transmit (502, etc.).

In this way, the user can automatically notify the server of a topic that he / she may be interested in according to the posted content simply by posting the article.

It is a block diagram which shows the function structure of embodiment of 1st invention of this invention. It is a figure which shows an example of the article displayed with the client by this invention. It is a figure which shows one Example of article DB by this invention. It is a figure which shows one Example of article DB by this invention. It is a figure which shows one Example of article content DB by this invention. It is a figure which shows one Example of article content DB by this invention. It is a figure which shows one Example of topic content DB by this invention. It is a figure which shows one Example of topic content DB by this invention. It is a figure which shows one Example of article DB by this invention. It is a figure which shows one Example of article content DB by this invention. It is a figure which shows one Example of topic content DB by this invention. It is a block diagram which shows the function structure of embodiment of 2nd invention of this invention. It is a figure which shows one Example of article content DB by this invention. It is a figure which shows one Example of user relationship DB by this invention. It is a figure which shows one Example of article content DB by this invention. It is a figure which shows one Example of user relationship DB by this invention. It is a figure which shows one Example of user relationship DB by this invention. It is a block diagram which shows the function structure of embodiment of 3rd invention of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Server 2 Client 3 Program recording medium 4 Program recording medium 5 Communication part 6 Article recording part 7 Article DB
8 Content summary section 9 Article content DB
10 Topic DB
11 Content Comparison Unit 12 User Model Update Unit 13 User Model 14 User Relationship DB
15 User model comparison unit 16 Article UM comparison unit

Claims (2)

An electronic bulletin board system that records articles posted from a client in an article DB and provides the contents of the articles recorded in the article DB as an electronic bulletin board,
Content summarizing means for creating a topic content summary representing a summary of the content of a thread which is a set of articles using the article recorded in the article DB;
A topic content DB that records the topic content summary created by the content summarization means;
Refer to the topic content DB, compare the topic content summary between the threads, and if there is a thread with the topic content summary close between the threads, generate an article introducing the thread, Content comparison means for recording the generated article in the article DB as an article of a thread whose topic content summary is close;
An electronic bulletin board system comprising at least. An electronic bulletin board system that records articles posted from a client in an article DB and provides the contents of the articles recorded in the article DB as an electronic bulletin board,
Article recording means for recording an article posted from a client via the communication unit in the article DB;
An article content DB for storing an article content summary representing a summary of the contents of the articles stored in the article DB;
Among the articles stored in the article DB, a topic content DB that stores a topic content summary representing a summary of the contents of a thread that is a set of a series of articles, for each thread;
When a new article is recorded in the article DB by the article recording means, the article content summary of the article is created and stored in the article content DB, and the created article content summary is used to create the article content summary. Content summarizing means for updating the content of the topic content DB for the thread related to an article;
When the topic content DB is updated, the topic content DB is referred to, the topic content summaries are compared between the threads, and if there is a thread with a close topic content summary between the threads, A content comparison unit that generates an article introducing a thread and records the generated article in the article DB as an article of a thread whose topic content summary is close;
An electronic bulletin board system comprising at least.

Images