Community system and community processing device

Abstract

The present invention comprises a summarization posting unit 105 that classifies a plurality of pieces of post information posted on a community site until a prescribed trigger condition is satisfied into one or more categories according to the content of each among the plurality of pieces of post information, generates summarized post information in which pieces of information relating to the one or more categories are summarized as a list, and posts the summarized post information on the community site. A user can recognize the specific category of a post being made on the community site and the various categories of posts that are actually being made, and as a result communications can be further expanded on the basis thereof and communication within the community can be expanded.

Description

Community system and community processing device

[0001] The present invention relates to a community system and a community processing device, and is particularly suitable for use in a system in which a computer and a plurality of users can post comments on a community site and communicate with each other.

[0002] In recent years, social networking services (SNS) have become increasingly popular as communication tools on the Internet, with many users posting various types of information on SNS. Groups (also called communities or circles) are formed within SNS, allowing users who participate in the groups to freely exchange information with each other. Multiple groups, each focused on a specific theme, are sometimes formed on SNS, allowing users to select and join groups that match their hobbies and interests.

[0003] Against this background, various systems have been proposed to support information exchange in communities (see, for example, Patent Documents 1 and 2). Patent Document 1 discloses a system that enables one community to automatically provide useful information to another, thereby efficiently sharing knowledge between communities. Patent Document 2 discloses a system that enables users to accurately participate in communities that deal with their desired topics, even if the topics in the communities change constantly.

[0004] The system described in Patent Document 1 manages multiple virtual communities that serve as forums for discussion, classifying and storing messages posted to each virtual community from each client terminal by topic. Furthermore, summary information (conclusions for conversations (threads)) that summarizes the messages stored in the virtual community by topic is managed for each virtual community. The summary information for a specific virtual community is then provided to other virtual communities that are determined to have discussions similar to that of the specific virtual community.

[0005] In the system described in Patent Document 2, a classification is continuously assigned to topics in a specified community based on multiple posting data continuously sent from a client device to the specified community, and information indicating the assigned classification is output to a client device accessing the specified community.

[0006] Japanese Patent Application Laid-Open No. 2002-342347 Japanese Patent Application Laid-Open No. 2006-227711

[0007] Although users who participate in communities and businesses that manage communities have various purposes for using social media, what is commonly desired in both cases is the activation of the community. However, neither of the systems described in Patent Documents 1 and 2 provides a mechanism for activating the community.

[0008] In other words, the system described in Patent Document 1 provides summary information of the results of discussions in a given community to other communities, but does not stimulate information exchange within a community. Furthermore, the system described in Patent Document 2 presents users participating in a community with a constantly changing topic classification, thereby encouraging users whose topics no longer suit their preferences to leave the community, but does not stimulate information exchange within the community.

[0009] The present invention has been made to solve such problems, and aims to provide a mechanism for revitalizing social media communities.

[0010] In order to solve the above-mentioned problems, the present invention classifies multiple pieces of posted information posted on a community site before a specified trigger condition is met into one or more categories according to their contents, generates summarized posted information that compiles information about the one or more categories into a list, and posts the summarized posted information on the community site.

[0011] According to the present invention configured as described above, users can recognize what categories of posts are being made on the community site, i.e., the various categories of posts that are actually being made, and can make new discoveries and gain new insights. If a category in which they are interested is found, they can expand their communication by replying to posts in that category, thereby making it possible to activate communication within the community.

[0012] FIG. 1 is a diagram illustrating an example of the overall configuration of a community system according to a first embodiment. FIG. 2 is a block diagram illustrating an example of the functional configuration of a server device according to the first embodiment. FIG. 3 is a diagram illustrating an example of summarized posted information posted by a summary posting unit in the first embodiment. FIG. 4 is a diagram illustrating an example of the operation of the server device according to the first embodiment. FIG. 5 is a flowchart illustrating an example of processing executed by the server device according to the first embodiment. FIG. 6 is a block diagram illustrating an example of the functional configuration of a server device according to a second embodiment. FIG. 7 is a diagram illustrating an example of the operation of a first mode of the server device according to the second embodiment. FIG. 8 is a block diagram illustrating an example of the functional configuration of a server device according to a third embodiment. FIG. 9 is a diagram schematically illustrating processing timing of the server device according to the third embodiment. FIG. 10 is a diagram illustrating an example of summarized posted information posted by a summary posting unit in the third embodiment.

[0013] (First Embodiment) A first embodiment of the present invention will now be described with reference to the drawings. Fig. 1 is a diagram showing an example of the overall configuration of a community system according to the first embodiment. As shown in Fig. 1, the community system of the first embodiment is configured to include a server device 10 (corresponding to a community processing device in the claims) and a plurality of user terminals 20. The server device 10 and the user terminals 20 are connected via a communication network 30 such as the Internet or a mobile phone network.

[0014] The community system of the first embodiment enables users of the computers of the server device 10 and the user terminals 20 to communicate by posting free comments within a community site formed in an online virtual space by the server device 10. Users can initiate messages themselves, and can also post replies to information posted by computers or other users. Computers can also initiate messages themselves, and can post replies to information posted by users.

[0015] 2 is a block diagram showing an example of the functional configuration of the server device 10 according to the first embodiment. As shown in Fig. 2, the server device 10 according to the first embodiment includes, as its functional configuration, an initial keyword registration unit 101, a derived keyword generation unit 102, a keyword-related posting unit 103, a posted information receiving unit 104, and a summarized posting unit 105. The server device 10 according to the first embodiment also includes, as storage media, a keyword storage unit 11 and a posted information storage unit 12.

[0016] The functional blocks 101 to 105 can be configured by either hardware or software. For example, the functional blocks 101 to 105 are realized by the operation of a program stored in a storage medium such as RAM, ROM, a hard disk, or a semiconductor memory under the control of a microcomputer configured with a CPU, RAM, ROM, etc. In addition to the CPU, a DSP (Digital Signal Processor) or the like may also be used.

[0017] The initial keyword registration unit 101 registers initial keywords (corresponding to "given keywords" in the claims) to be used in generating posted messages when a computer initiates a message posting. For example, an administrator of a community site using one of the user terminals 20 specifies an arbitrary keyword from the user terminal 20, and the initial keyword registration unit 101 accepts this and registers it in the server device 10 as an initial keyword.

[0018] Alternatively, the initial keyword registration unit 101 may generate an arbitrary keyword and register it as an initial keyword. As an example of a method for generating a keyword, one keyword may be randomly extracted from a dictionary database in which a large number of keywords are recorded. Alternatively, a question sentence for which the keyword is to be answered may be input as a prompt to a large-scale language model (LLM), and the keyword output from the LLM as a result may be used as the initial keyword. In this case, the initial keyword registration unit 101 generates the question sentence using a predetermined template sentence. One or more template sentences may be prepared in advance. If there are multiple template sentences, one of them is selected and used.

[0019] The initial keyword registration unit 101 registers initial keywords periodically or irregularly. For example, it registers initial keywords once every day or once every two days. The number of initial keywords to be registered at one time is arbitrary, but it is assumed that the number is one, for example.

[0020] The derived keyword generation unit 102 generates one or more derived keywords based on an initial keyword every time the initial keyword is registered by the initial keyword registration unit 101. The derived keywords generated by the derived keyword generation unit 102 are used to generate posted text. The derived keywords are words related to the initial keyword, such as synonyms, associated words, and co-occurring words with respect to the initial keyword.

[0021] As an example of a method for generating derived keywords, synonyms or associative words for the initial keyword may be extracted by referring to a dictionary database that records the relationships between many synonyms and associative words. Alternatively, co-occurring words for the initial keyword may be detected by analyzing many sentences stored in a sentence database. In this case, the posted information storage unit 12 may be used as the sentence database.

[0022] Alternatively, a question sentence for asking for a derived keyword as an answer may be input to the LLM as a prompt, and the resulting keyword output from the LLM may be used as a derived keyword. In this case, the derived keyword generation unit 102 generates a question sentence using the initial keywords registered by the initial keyword registration unit 101 and a predetermined template sentence. One or more template sentences may be prepared in advance, and if there are multiple template sentences, one of them is selected and used.

[0023] The derived keyword generation unit 102 stores the generated derived keywords together with the initial keywords from which they were generated in the keyword storage unit 11. As described above, the registration of initial keywords is repeatedly executed, and one or more derived keywords are generated each time an initial keyword is registered, so that many keywords are successively stored in the keyword storage unit 11. At this time, keywords that are the same as keywords already stored in the keyword storage unit 11 do not need to be additionally stored. Alternatively, keywords may be classified according to the timing of generation and stored successively, regardless of whether they are the same as keywords already stored.

[0024] In the first embodiment, it is not essential to provide the derived keyword generation unit 102, but it is preferable to provide the derived keyword generation unit 102 in order to increase the variety of keywords used in generating posted messages. In particular, when a word specified by the administrator of the community site is to be registered as an initial keyword in the initial keyword registration unit 101, it is effective to generate derived keywords by the derived keyword generation unit 102.

[0025] The keyword-related posting unit 103 generates posting information for a message based on a plurality of keywords stored in the keyword storage unit 11 (including the initial keywords registered by the initial keyword registration unit 101 and the derived keywords generated by the derived keyword generation unit 102), and posts (communicates) the message on the community site. Hereinafter, this posting information will be referred to as keyword-related posting information. The keyword-related posting information posted by the keyword-related posting unit 103 is displayed on the community site and can be viewed by users on the user terminal 20.

[0026] If the derivative keyword generation unit 102 is omitted, the keyword storage unit 11 will store only the multiple initial keywords repeatedly registered by the initial keyword registration unit 101. In this case, the keyword-related posting unit 103 will generate posting information for a message based on the multiple initial keywords stored in the keyword storage unit 11 and post it on the community site.

[0027] The keyword-related posting unit 103 generates keyword-related posted information and posts it on the community site, for example, every time a keyword is stored in the keyword storage unit 11. For example, the keyword-related posting unit 103 may input a question to be answered by the posted message as a prompt to the LLM, and use the posted message output from the LLM as the keyword-related posted information. In this case, the keyword-related posting unit 103 generates the question message using a keyword extracted from the keyword storage unit 11 and a predetermined template message. One or more template messages may be prepared in advance, and if there are multiple template messages, one of them is selected and used.

[0028] Here, the keyword-related posting unit 103 selects one of the multiple keywords stored in the keyword storage unit 11 and generates keyword-related posted information using the selected keyword. The number of keywords selected each time is arbitrary, but for example, one. Also, the number of pieces of keyword-related posted information generated from one keyword each time is arbitrary, but for example, one. That is, as an example, the keyword-related posting unit 103 extracts one keyword from the keyword storage unit 11 each time a keyword is stored in the keyword storage unit 11, and generates one piece of keyword-related posted information using the extracted keyword.

[0029] The posted information receiving unit 104 receives a plurality of pieces of posted information posted on the community site by a plurality of users from their user terminals 20. Each time the posted information receiving unit 104 receives posted information from a user terminal 20, it sequentially stores the information in the posted information storage unit 12. Although not shown as a functional block in FIG. 2 , the posted information stored in the posted information storage unit 12 is displayed on the community site and made viewable by users on their user terminals 20.

[0030] The posted information received by the posted information receiving unit 104 from the user terminal 20 includes a first type of posted information originating from the user himself / herself, a second type of posted information that is a reply to keyword-related posted information posted by the keyword-related posting unit 103, and a third type of posted information that is a reply to posted information posted by another user. The posted information received by the posted information receiving unit 104 includes information that can identify the type of posted information.

[0031] The summary posting unit 105 classifies the plurality of pieces of posted information stored in the posted information storage unit 12 until a predetermined trigger condition is met into one or more categories according to the contents of each piece of posted information, generates summarized posted information that compiles information about the one or more categories into a list, and posts the summarized posted information on the community site. The generated categories are, for example, topics covered by the posted information. The summarized posted information posted by the summary posting unit 105 is displayed on the community site and made available for viewing by users on the user terminal 20.

[0032] The posted information processed by the summary posting unit 105 is the second type of posted information described above, i.e., posted information posted by multiple users as a reply to keyword-related posted information posted by the keyword-related posting unit 103 (hereinafter, sometimes referred to as computer-directed reply posted information).

[0033] The trigger condition is a condition that must be satisfied to trigger the summary posting unit 105 to generate summarized posted information, which will be described later. For example, the trigger condition for generating summarized posted information can be that the number of computer-related reply posted information posted by multiple users in response to keyword-related posted information reaches a threshold value (e.g., a predetermined number such as five).

[0034] In this case, when the first five pieces of reply information to a computer are posted after the keyword-related posted information is posted, the summarize posting unit 105 performs classification on those five pieces of reply information to a computer. After that, when five more pieces of reply information to a computer are posted, the summarize posting unit 105 performs further classification on those five pieces of reply information to a computer. Thereafter, this classification process is continued sequentially for every five pieces of reply information to a computer.

[0035] The fact that a predetermined number or more of reply information to a computer is posted means that a predetermined number or more users are interested in, concerned about, or empathetic to the keyword-related posted information, and it is considered that there is a possibility that communication can be further developed based on the summarized posted information. Conversely, if the number of reply information to a computer posted is less than the predetermined number, it is considered that the number of users who are interested in, concerned about, or empathetic to the keyword-related posted information is small, and it is difficult to further develop communication. Therefore, summarized posted information is generated using the number of reply information to a computer posted as a trigger condition.

[0036] Here, the summary posting unit 105 generates a category for each of the multiple pieces of posted information according to its content. For example, the summary posting unit 105 inputs a question to the LLM as a prompt to request a category be assigned, and uses the resulting category output from the LLM as a category for the multiple pieces of posted information. At this time, the summary posting unit 105 generates the question using a predetermined template sentence and multiple pieces of posted information read from the posted information storage unit 12 each time a trigger condition is satisfied. One or more template sentences may be prepared in advance, and if there are multiple template sentences, one of them is selected and used.

[0037] For example, when classification is performed using five pieces of posted information as a unit, the following cases may occur: a first case in which the categories generated for the five pieces of posted information are all different and five different categories are assigned; a second case in which the categories generated for less than five pieces of posted information are the same and two, three, or four different categories are assigned; and a third case in which the categories generated for the five pieces of posted information are all the same and one type of category is assigned.

[0038] The summary posting unit 105 generates summarized posted information and posts it to the community site when multiple categories are assigned to multiple pieces of posted information, rather than all being the same (only one category is generated), i.e., in the first or second case. By generating and posting summarized posted information when multiple categories are generated, users can recognize that the computer-to-computer reply posted information, which is a response from multiple users to one piece of keyword-related posted information, relates to a variety of topics. In other words, users can be made aware of the various topics being discussed on the community site.

[0039] The summarized posted information posted to the community site includes at least one of the following: the name of the category, a description of the category, and examples of posted information belonging to the category. For example, the summarized posted information may be a list of the names of multiple categories.

[0040] Alternatively, the summarized posted information may include the names of multiple categories and examples of reply posted information to a computer belonging to each category. The example reply posted information to a computer may be the entire text or a portion thereof. For example, a maximum number of characters to be used as an example may be determined, and information copied from the beginning of the reply posted information to a computer within the number of characters may be included in the summarized posted information.

[0041] Alternatively, the summarized posted information may include the names of multiple categories, examples of computer-directed reply posted information belonging to each category, and descriptions of each category. In this example, the summarized posting unit 105 may input a question requesting the assignment of a category and the creation of a description to the LLM as a prompt, and generate the summarized posted information using the category names and descriptions and examples of computer-directed reply posted information output from the LLM as a result.

[0042] As another generation method, the summary posting unit 105 may generate categories and descriptions in two stages. For example, in a first stage, the summary posting unit 105 generates a question requesting the assignment of a category using a plurality of computer-directed reply posted information items read from the posted information storage unit 12 and a predetermined template sentence each time a trigger condition is satisfied, and inputs the question as a prompt to the LLM, thereby generating one or more categories. In a subsequent second stage, the summary posting unit 105 generates a question requesting the assignment of a description for each of the plurality of categories generated in the first stage, using one or more computer-directed reply posted information items belonging to the category, the category name, and a predetermined template sentence, and inputs the question as a prompt to the LLM, thereby generating a description for each category.

[0043] 3 is a diagram showing an example of summarized posted information posted by the summary posting unit 105. The summarized posted information shown in FIG. 3 includes category names 31, descriptions 32 of each category, and examples 33 of reply posted information to a computer belonging to each category.

[0044] The summarized posting unit 105 may generate summarized posted information and post it to the community site only when, in addition to the condition that multiple categories are assigned to multiple pieces of posted information, the condition that a category assigned in common to the multiple pieces of posted information is included, i.e., the second case, is satisfied. In this case, the number of pieces of posted information included in one category may be recorded in the summarized posted information. In this way, by posting summarized posted information only when a situation is included in which multiple users are posting about the same topic, the summarized posted information can allow users to recognize homogeneous categories in which multiple users are commonly interested.

[0045] Furthermore, in particular, in the second case, summarized posted information may be generated only when there are multiple sets of posted information assigned the same category, or when only a category commonly assigned to multiple pieces of posted information is included. The former means that summarized posted information is generated only when three categories are generated by a combination of two pieces of posted information, two pieces of posted information, and one piece of posted information, or when two categories are generated by a combination of two pieces of posted information and three pieces of posted information. The latter means that summarized posted information is generated only when two categories are generated by a combination of two pieces of posted information and three pieces of posted information. In the former case, summarized posted information may be generated only for the two categories generated from two pieces of posted information and two pieces of posted information. In this way, the summarized posted information can allow users to recognize only homogeneous categories in which multiple users are commonly interested.

[0046] The summary posting unit 105 may not execute the process of generating summarized posted information for keyword-related posted information if a predetermined number of pieces of reply posted information to a computer are not posted within a predetermined time after the keyword-related posted information is posted by the keyword-related posting unit 103. After summarized posted information has been posted once, if a predetermined number of pieces of reply posted information to a computer are not posted within a predetermined time after the most recent summarized posted information was posted, the process of generating summarized posted information for that keyword-related posted information may be stopped.

[0047] 4 is a diagram illustrating an example of the operation of the server device 10 according to the first embodiment configured as described above. In FIG. 4, the initial keyword registration unit 101 registers one initial keyword KW0. Once the initial keyword KW0 is registered, the derived keyword generation unit 102 generates one or more derived keywords KW1 to KWm (m≧1) based on the initial keyword KW0. The initial keyword registration unit 101 and the derived keyword generation unit 102 repeatedly execute this process, successively storing sets of keywords KW0 to KWm in the keyword storage unit 11.

[0048] The keyword-related posting unit 103 selects one keyword KWi from the plurality of keywords (plural sets of keywords KW0 to KWm) stored in the keyword storage unit 11 at each timing when a keyword is stored in the keyword storage unit 11, generates keyword-related posted information using the selected keyword KWi, and posts the generated keyword-related posted information on the community site.

[0049] The posted information receiving unit 104 sequentially receives reply posted information to a computer 1 to n (n≧1) posted by a plurality of users from the user terminal 20 as replies to the keyword-related posted information. The posted information receiving unit 104 receives posted information of the first to third types posted by a plurality of users on the community site, and identifies and recognizes reply posted information to a computer 1 to n of the second type posted in response to the keyword-related posted information.

[0050] Each time the summary posting unit 105 receives a predetermined number (e.g., five) of computer-to-computer reply posted information in response to keyword-related posted information, the summary posting unit 105 classifies the predetermined number of posted information into one or more categories, and if multiple categories are assigned to the predetermined number of posted information, generates summarized posted information and posts it on the community site.

[0051] 4 shows an example in which two categories C1-1 and C1-2 are generated from the first five pieces of computer-directed reply posted information 1 to 5, and summarized posted information S1 is generated by compiling information relating to the two categories C1-1 and C1-2 into a list. Also, FIG. 4 shows an example in which three categories C2-1 to C2-3 are generated from the next five pieces of computer-directed reply posted information 6 to 10, and summarized posted information S2 is generated by compiling information relating to the three categories C2-1 to C2-3 into a list.

[0052] 5 is a flowchart showing an example of processing executed by the server device 10 according to the first embodiment. The flowchart shown in FIG. 5 shows the flow of processing that is executed starting from the registration of one initial keyword.

[0053] First, the initial keyword registration unit 101 registers one initial keyword (step S1). Once the initial keyword is registered, the derived keyword generation unit 102 generates one or more derived keywords based on the registered initial keyword (step S2), and stores the generated derived keywords together with the initial keyword from which they were generated in the keyword storage unit 11 (step S3).

[0054] Next, the keyword-related posting unit 103 selects one keyword from the plurality of keywords currently stored in the keyword storage unit 11. Then, the selected keyword is used to generate keyword-related posted information, and the generated keyword-related posted information is posted on the community site (step S4).

[0055] The summary posting unit 105 then determines whether the number of reply computer-related posted information received by the posted information receiving unit 104 as replies to the keyword-related posted information has reached a predetermined number (step S5). If it determines that the number of reply computer-related posted information has reached the predetermined number, the summary posting unit 105 classifies the predetermined number of posted information into one or more categories, and if multiple categories are assigned to the predetermined number of posted information, generates summarized posted information and posts it on the community site (step S6). Then, the process returns to step S5.

[0056] After returning from step S6 to step S5, the summary posting unit 105 determines whether the number of reply computer-related posted information received by the posted information receiving unit 104 as replies to the keyword-related posted information has reached a predetermined number. If it determines that the number of reply computer-related posted information has reached the predetermined number, the summary posting unit 105 generates summarized posted information for the predetermined number of posted information and posts it on the community site (step S6). Similarly, the summary posting unit 105 generates and posts summarized posted information each time it receives a predetermined number of reply computer-related posted information.

[0057] If it is determined in step S5 that the number of reply posted information to the computer has not yet reached a predetermined number, the summary posting unit 105 determines whether a predetermined time has elapsed since the keyword-related posted information was posted in step S4 if no summarized posted information has been posted yet, or whether a predetermined time has elapsed since the summarized posted information was posted in step S6 if summarized posted information has already been posted once (step S7).

[0058] If the predetermined time has not yet elapsed, the process returns to step S5. On the other hand, if the predetermined time has elapsed, the summarized posting unit 105 stops the process of generating summarized posted information (step S8) and ends the process of the flowchart shown in Fig. 5. Note that in step S7, it may be determined whether a predetermined time has elapsed since the keyword-related posted information was posted, regardless of whether the summarized posted information has been posted once.

[0059] As explained in detail above, in the first embodiment, multiple pieces of computer-directed reply posting information posted on a community site before a specified trigger condition is met are classified into one or more categories according to their contents, and when multiple categories are assigned to the multiple pieces of computer-directed reply posting information, summarized posting information that lists information relating to the multiple categories is generated and posted on the community site.

[0060] According to the first embodiment configured as described above, users can recognize what categories are being posted in response to keyword-related posted information within the community site, i.e., the various categories of computer-directed reply posted information that are actually being posted, and can make new discoveries and discoveries. If a category among them interests the user, the user can broaden communication by replying to computer-directed reply posted information in that category. This makes it possible to activate communication within the community.

[0061] Second Embodiment Next, a second embodiment of the present invention will be described with reference to the drawings. The overall configuration of the community system according to the second embodiment is the same as that shown in Fig. 1. However, a server device 10A is used in place of the server device 10.

[0062] Fig. 6 is a block diagram showing an example of the functional configuration of a server device 10A according to the second embodiment. In Fig. 6, components having the same reference numerals as those shown in Fig. 2 have the same functions, and therefore, redundant explanations will be omitted here.

[0063] As shown in Fig. 6, the server device 10A according to the second embodiment includes, as its functional configuration, a keyword-related posting unit 103A and a summary posting unit 105A instead of the keyword-related posting unit 103 and the summary posting unit 105 shown in Fig. 2. The server device 10A according to the second embodiment also includes, as its functional configuration, a category keyword setting unit 111 and a second derived keyword generating unit 112. The server device 10A according to the second embodiment also includes, as a storage medium, a second keyword storage unit 13.

[0064] The functional blocks shown in Fig. 6 can be configured by either hardware or software, similar to the functional blocks shown in Fig. 2. For example, the functional blocks are realized by the operation of a program stored in a storage medium such as RAM, ROM, a hard disk, or a semiconductor memory under the control of a microcomputer configured with a CPU, RAM, ROM, etc. In addition to the CPU, a DSP or the like may also be used.

[0065] The keyword-related posting unit 103A has a first mode in which it generates keyword-related posted information using the initial keywords registered by the initial keyword registration unit 101 as given keywords, and a second mode in which it generates keyword-related posted information based on keywords indicating categories (hereinafter referred to as category keywords) generated by the summarize posting unit 105A, and executes a process of generating keyword-related posted information by switching from the first mode to the second mode. After switching from the first mode to the second mode, the keyword-related posting unit 103A repeatedly executes the process of the second mode one or more times as long as categories are generated by the summarize posting unit 105A. The process contents of the keyword-related posting unit 103A executed in the first mode and the second mode are the same as those in the first embodiment.

[0066] The summary posting unit 105A generates summarized posted information for computer-related reply posted information posted from the user terminal 20 as a reply to the keyword-related posted information posted by the keyword-related posting unit 103A in the first mode, and posts the summarized posted information on the community site. In addition, the summary posting unit 105A generates summarized posted information for computer-related reply posted information posted from the user terminal 20 as a reply to the keyword-related posted information posted by the keyword-related posting unit 103A in the second mode, and posts the summarized posted information on the community site. The processing content of the summary posting unit 105A executed in the first mode and the second mode is the same as that in the first embodiment.

[0067] The category keyword setting unit 111 sets a category keyword indicating a category generated by the summary posting unit 105A during processing in the first mode and the second mode. If there are multiple categories generated by the summary posting unit 105A, the category keyword setting unit 111 selects one of the categories and sets a category keyword indicating the selected category.

[0068] The category to be selected is arbitrary, but is limited to categories for which summarized posted information has been generated. For example, it is possible to randomly select one category from multiple categories for which summarized posted information has been generated. Alternatively, it is also possible to preferentially select a category that includes multiple pieces of posted information. A category that includes multiple pieces of posted information is a topic that is commonly discussed by multiple users, and is considered to indicate a high possibility of further development of communication.

[0069] The keyword indicating the category is, for example, the category name. If the summarized posting information generated by the summarized posting unit 105A includes a description of the category, a word extracted from the description may be set as the category keyword. Note that in the first mode processing starting with the registration of the initial keyword and the subsequent one or more second mode processings, the keyword set as the category keyword is preferably a word different from the initial keyword and a word different from any word that has already been set as a category keyword in the series of processing.

[0070] Each time a category keyword is set by the category keyword setting unit 111, the second derived keyword generation unit 112 generates one or more derived keywords based on the category keyword. The derived keywords generated by the second derived keyword generation unit 112 are words that are related to the category keyword. The second derived keyword generation unit 112 generates words that are subordinate concepts to the category keyword instead of or in addition to synonyms, associative words, and co-occurring words for the category keyword. The method of generating derived keywords may be the same as that of the derived keyword generation unit 102.

[0071] The second derived keyword generating unit 112 stores the generated derived keywords together with the category keywords from which they were generated in the second keyword storage unit 13. When the processing in the second mode is repeatedly executed, one or more derived keywords are generated each time a category keyword is set by the category keyword setting unit 111, and therefore many keywords are successively stored in the second keyword storage unit 13.

[0072] In the second embodiment, it is not essential to provide the derived keyword generation unit 102 and the second derived keyword generation unit 112, but it is preferable to provide the derived keyword generation unit 102 and the second derived keyword generation unit 112 in order to increase the variety of keywords used to generate posted messages. When the derived keyword generation unit 102 and the second derived keyword generation unit 112 are provided, the keyword-related posting unit 103A executes the following process.

[0073] That is, in the first mode, the keyword-related posting unit 103A generates keyword-related posted information based on the keywords stored in the keyword storage unit 11 by the processing of the initial keyword registration unit 101 and the derived keyword generation unit 102, and posts the generated information on the community site. In addition, in the second mode, the keyword-related posting unit 103A generates keyword-related posted information based on the keywords stored in the second keyword storage unit 13 by the processing of the category keyword setting unit 111 and the second derived keyword generation unit 112, and posts the generated information on the community site.

[0074] 7A and 7B are diagrams for explaining an example of operation of the server device 10A according to the second embodiment configured as described above. FIG. 7A shows an example of operation in the first mode, and FIG. 7B shows an example of operation in the second mode. Note that the [1] at the end of the reference numerals indicates the first mode, and [2] indicates the second mode. The example of operation in the first mode shown in FIG. 7A is the same as the example of operation in the first embodiment shown in FIG. 4.

[0075] 7A, the initial keyword registration unit 101 registers one initial keyword KW0[1]. Once the initial keyword KW0[1] is registered, the derived keyword generation unit 102 generates one or more derived keywords KW1[1] to KWm[1] (m≧1) based on the initial keyword KW0[1]. The initial keyword registration unit 101 and the derived keyword generation unit 102 repeatedly execute this process, and sequentially store sets of keywords KW0[1] to KWm[1] in the keyword storage unit 11.

[0076] The keyword-related posting unit 103A selects one keyword KWi[1] from among the multiple keywords (multiple sets of keywords KW0[1] to KWm[1]) stored in the keyword storage unit 11 at that time each time a keyword is stored in the keyword storage unit 11, and generates keyword-related posted information using the selected keyword KWi[1].Then, the generated keyword-related posted information is posted on the community site.

[0077] The posted information receiving unit 104 sequentially receives reply posted information to a computer 1 to n (n≧1) posted by a plurality of users from the user terminal 20 as replies to keyword-related posted information. The posted information receiving unit 104 receives posted information of the first to third types posted by a plurality of users on the community site, and identifies and recognizes reply posted information to a computer 1 to n of the second type that has been posted in response to keyword-related posted information posted based on the initial keyword KW0[1].

[0078] Each time the summary posting unit 105A receives a predetermined number (e.g., five) of computer-to-computer reply posted information in response to keyword-related posted information, the summary posting unit 105A classifies the predetermined number of posted information into one or more categories, and if multiple categories are assigned to the predetermined number of posted information, it generates summarized posted information and posts it on the community site.

[0079] 7A shows an example in which two categories C1-1[1] and C1-2[1] are generated from the first five pieces of computer-related reply posted information 1 to 5, and summarized posted information S1[1] is generated by compiling information related to the two categories C1-1[1] and C1-2[1] into a list. Also, FIG. 7A shows an example in which three categories C2-1[1] to C2-3[1] are generated from the next five pieces of computer-related reply posted information 6 to 10, and summarized posted information S2[1] is generated by compiling information related to the three categories C2-1[1] to C2-3[1] into a list.

[0080] 7B, the category keyword setting unit 111 arbitrarily selects one category from among the multiple categories C1-1[1], C1-2[1], C2-1[1] to C2-3[1], ... generated by the summary posting unit 105A in the first mode, and sets a keyword indicating the selected category as the category keyword KW0[2]. The example of FIG. 7B shows a state in which category C1-2[1] is selected and category keyword KW0[2] is set.

[0081] When the category keyword KW0[2] is set, the second derived keyword generation unit 112 generates one or more derived keywords KW1[2] to KWp[2] (p≧1) based on the category keyword KW0[2]. The category keyword setting unit 111 and the second derived keyword generation unit 112 repeatedly execute this process each time a category is generated by the summarize posting unit 105A, and sequentially store sets of keywords KW0[2] to KWp[2] in the second keyword storage unit 13.

[0082] The keyword-related posting unit 103A selects one keyword KWj[2] from among the multiple keywords (multiple sets of keywords KW0[2] to KWp[2]) stored in the second keyword storage unit 13 at each timing when a keyword is stored in the second keyword storage unit 13, and generates keyword-related posted information using the selected keyword KWj[2]. Then, the generated keyword-related posted information is posted on the community site.

[0083] The posted information receiving unit 104 sequentially receives reply posted information to a computer 1 to q (q≧1) posted by a plurality of users from the user terminal 20 as replies to keyword-related posted information. The posted information receiving unit 104 receives posted information of the first to third types posted by a plurality of users on the community site, and identifies and recognizes reply posted information to a computer 1 to q of the second type that has been posted in response to keyword-related posted information posted based on the category keyword KW0[2].

[0084] Each time the summary posting unit 105A receives a predetermined number (e.g., five) of computer-to-computer reply posted information in response to keyword-related posted information, the summary posting unit 105A classifies the predetermined number of posted information into one or more categories, and if multiple categories are assigned to the predetermined number of posted information, it generates summarized posted information and posts it on the community site.

[0085] 7B shows an example in which three categories C1-1[2] to C1-3[2] are generated from the first five pieces of computer-directed reply posted information 1 to 5, and summarized posted information S1[2] is generated by compiling information related to the three categories C1-1[2] to C1-3[2] as a list. Also, FIG. 7B shows an example in which two categories C2-1[2] and C2-2[2] are generated from the next five pieces of computer-directed reply posted information 6 to 10, and summarized posted information S2[2] is generated by compiling information related to the two categories C2-1[2] and C2-2[2] as a list.

[0086] Although not shown in the figure, the category keyword setting unit 111 arbitrarily selects one category from among the multiple categories C1-1[2] to C1-3[2], C2-1[2], C2-2[2], ... generated by the summary posting unit 105A during the first processing in the second mode, and sets a keyword indicating the selected category as a new category keyword KW0[2]. Then, the same processing as in FIG. 7B is executed again. Thereafter, the processing in the second mode shown in FIG. 7B is repeatedly executed as long as the condition is met, for example, that a predetermined number of computer-to-computer reply postings are posted within a predetermined time from the time when the most recent summarized posting was posted.

[0087] As described in detail above, in the second embodiment, the keyword-related posting unit 103A has two processing modes: a first mode in which keyword-related posting information is generated and posted based on an initial keyword, and a second mode in which keyword-related posting information is generated and posted based on a category keyword indicating a category generated by the summary posting unit 105A, and the first mode is switched to the second mode, and the processing of the second mode is repeatedly executed as long as predetermined conditions are met.

[0088] This not only generates and posts summarized posting information once based on computer-to-computer reply posting information sent by multiple users who have responded expressing interest, concern, or sympathy for keyword-related posting information, but also posts keyword-related posting information again based on keywords indicating the category assigned when the summarized posting information was generated, thereby expanding the scope of conversation based on keyword-related posting information related to categories in which multiple users have shown interest, concern, or sympathy, and further developing communication.

[0089] When generating derivative keywords from category keywords, derivative keywords that are subordinate to the category keywords are also generated, making it possible to post keyword-related information that encourages communication that delves deeper into the topic indicated by the category keyword. This not only broadens the topic, but also leads to communication that allows for in-depth discussion of a single topic.

[0090] Furthermore, summarized posting information is re-posted based on computer-to-computer reply posting information that is replied to by multiple users in response to newly posted keyword-related posting information based on category keywords, which increases the opportunities for users to make new discoveries and realize new things, and further develops communication from here.

[0091] (Third Embodiment) Next, a third embodiment of the present invention will be described with reference to the drawings. The overall configuration of the community system according to the third embodiment is the same as that shown in FIG. 1. However, a server device 10B is used instead of the server device 10. The third embodiment is a further application of the second embodiment.

[0092] Fig. 8 is a block diagram showing an example of the functional configuration of a server device 10B according to the third embodiment. In Fig. 8, components having the same reference numerals as those shown in Fig. 6 have the same functions, and therefore, redundant explanations will be omitted here.

[0093] As shown in FIG. 8, the server device 10B according to the third embodiment has, as its functional configuration, a keyword-related posting unit 103B and a summary posting unit 105B instead of the keyword-related posting unit 103A and the summary posting unit 105A shown in FIG. 6.

[0094] The functional blocks shown in Fig. 8 can also be configured by either hardware or software. For example, the functional blocks are realized by the operation of a program stored in a storage medium such as RAM, ROM, a hard disk, or a semiconductor memory under the control of a microcomputer configured with a CPU, RAM, ROM, etc. In addition to the CPU, a DSP or the like may also be used.

[0095] In the third embodiment, the keyword-related posting unit 103B alternately switches between the first mode process and the first second mode process. For example, the first mode process and the first second mode process are alternately switched every day. The second and subsequent second mode processes are executed sequentially.

[0096] In each of the first and second modes, the processing up to the posted information receiving unit 104 and the processing of the summarized posting unit 105B are alternately switched and executed on a daily basis. In this case, the timing of the processing of the summarized posting unit 105B is the same as the timing of the processing up to the posted information receiving unit 104, which is executed for the posting of keyword-related posted information related to the category generated by the processing. The meaning of "the same timing" will be described later.

[0097] 9 is a diagram showing the timing of the above processing. As shown in FIG. 9, on the first day, the initial keyword registration unit 101, the derivative keyword generation unit 102, the keyword related posting unit 103B, and the posted information receiving unit 104 perform processing in the first mode.

[0098] On the second day, the summary posting unit 105B processes the multiple computer-to-computer reply posting information stored in the posting information storage unit 12 during the processing on the first day as a continuation of the first mode, and the category keyword setting unit 111, second derived keyword generation unit 112, keyword-related posting unit 103B, and posting information receiving unit 104 process the second mode.

[0099] On the third day, the processing of the summarize posting unit 105B is performed as a continuation of the second mode for the plurality of computer-to-computer reply posted information stored in the second posted information storage unit 13 during the processing on the second day, and the processing of the category keyword setting unit 111, the second derived keyword generating unit 112, the keyword-related posting unit 103B, and the posted information receiving unit 104 is performed as second-mode processing. In addition, the processing of the initial keyword registration unit 101, the derived keyword generating unit 102, the keyword-related posting unit 103B, and the posted information receiving unit 104 is also performed as new first-mode processing.

[0100] The same process is repeated from the fourth day onwards.

[0101] In the third embodiment, the summarized posting unit 105B includes, in the summarized posting information, introduction information for the keyword-related posted information when one of a plurality of categories included in the generated summarized posting information matches a category corresponding to a keyword used by the keyword-related posting unit 103B when generating the keyword-related posted information at the same time as the generation of the summarized posted information. The same timing means “on the same day” when processing is performed every other day as shown in FIG. 9 .

[0102] The processing of this summary posting unit 105B will be described with reference to Figure 9. During processing on the second day, summary posting unit 105B determines whether any of a plurality of categories included in the summarized posted information generated on the second day as processing in the first mode matches a category corresponding to a keyword used by keyword-related posting unit 103B when generating keyword-related posted information on the second day as processing in the first second mode, and if it determines that there is a match, generates summarized posted information including information introducing the keyword-related posted information and posts it to the community site.

[0103] Furthermore, during processing on the third day, the summary posting unit 105B determines whether any of the multiple categories included in the summarized posting information generated on the third day as the first processing in the second mode matches a category corresponding to the keyword used by the keyword-related posting unit 103B when generating the keyword-related posting information on the third day as the second processing in the second mode, and if it determines that there is a match, it generates summarized posting information including introduction information for the keyword-related posting information and posts it on the community site.

[0104] As described in detail in the first embodiment, when the summary posting unit 105B generates summarized posted information, it may generate a category but not generate summarized posted information that includes that category. Furthermore, when the keyword-related posting unit 103B generates keyword-related posted information, it generates the keyword-related posted information by selecting one of the multiple keywords that have been stored up to that point in the keyword storage unit 11 or the second keyword storage unit 13. Therefore, any one of the multiple categories included in the summarized posted information generated by the summary posting unit 105B may not necessarily match the category corresponding to the keyword selected from the keyword storage units 11 and 13 by the keyword-related posting unit 103B.

[0105] Therefore, the summary posting unit 105B determines whether the categories match, and if the category of the keyword-related posted information matches the category included in the summarized posted information, the summarized posted information includes introductory information about the keyword-related posted information related to the same category, thereby creating a situation in which the user can easily reply to the keyword-related posted information.

[0106] The introduction information can be, for example, link information to the keyword-related posted information. In this way, users can easily view the keyword-related posted information by following the link information and can easily reply to it. Note that instead of or in addition to the link information, the posted text of the keyword-related posted information or a summary thereof may be used as the introduction information.

[0107] Fig. 10 is a diagram showing an example of summarized posted information incorporating introduction information for keyword-related posted information. In Fig. 10, the same elements as those included in the summarized posted information shown in Fig. 3 are denoted by the same reference numerals. The summarized posted information shown in Fig. 10 includes introduction information 34 for keyword-related posted information in addition to a category name 31, a description 32 for the category, and examples 33 of computer-directed reply posted information belonging to the category.

[0108] According to the third embodiment configured as described above, keyword-related posted information relating to a category introduced by the summarized posted information is posted in a form that is incorporated into the summarized posted information as introduction information, thereby providing a user with an environment in which they can easily access keyword-related posted information of a category in which they are interested by viewing the summarized posted information.

[0109] Furthermore, according to the third embodiment, the processing in the first mode and the processing in the second mode are executed in parallel at the same time (on the same day), so that the keyword-related posted information posted by the processing in the first mode can broaden the range of topics and widen the scope of user participation, while the keyword-related posted information posted by the processing in the second mode can provide more in-depth topics and encourage deeper comments.

[0110] In the above first to third embodiments, an example of generating summarized posting information when multiple categories are assigned to multiple pieces of computer-to-computer reply posting information (in the first or second case) has been described. However, summarized posting information may also be generated in the third case where the same category is generated for multiple pieces of posting information and only one category is assigned.

[0111] In this case, any comments a user makes in response to keyword-related posted information will always be included in the summary posting. This allows users to safely post computer-directed reply information without worrying that their comments will be excluded from the summary posting. This provides an environment where users can post without hesitation, contributing to invigorating communication on the community site.

[0112] In the first to third embodiments, an example has been described in which one piece of keyword-related posted information is generated by selecting one of the multiple keywords stored in the keyword storage units 11 and 13. However, two or more keywords may be selected to generate two or more pieces of keyword-related posted information. For example, in the third embodiment, first keyword-related posted information may be generated using a category keyword set by the category keyword setting unit 111 based on one of the categories generated by the summarize posting unit 105, and second keyword-related posted information may be generated by selecting one of the multiple keywords stored in the second keyword storage unit 13 up to that point.

[0113] In this configuration, one of the categories included in the summarized posted information generated by the summarized posting unit 105 and the category corresponding to the first keyword-related posted information will always match at the same time, so it is possible to generate summarized posted information by incorporating introduction information of the first keyword-related posted information. In addition, in parallel with this, it is also possible to post second keyword-related posted information using keywords related to any category stored in the second keyword storage unit 13.

[0114] Furthermore, the above-described embodiments are merely examples of specific embodiments for carrying out the present invention, and the technical scope of the present invention should not be construed as being limited thereby. In other words, the present invention can be carried out in various forms without departing from the gist or main characteristics thereof.

[0115] 10, 10A, 10B Server device 11 Keyword storage unit 12 Posted information storage unit 13 Second keyword storage unit 20 User terminal 101 Initial keyword registration unit 102 Derived keyword generation unit 103, 103A, 103B Keyword-related posting unit 104 Posted information reception unit 105, 105A, 105B Summarized posting unit 111 Category keyword setting unit 112 Second derived keyword generation unit

Claims

1. A community system comprising: a posted information storage unit that stores a plurality of pieces of posted information posted by a plurality of users on a community site; and a summarized posting unit that classifies the plurality of pieces of posted information stored in the posted information storage unit into one or more categories according to the content of each piece of posted information until a predetermined trigger condition is met, generates summarized posted information that compiles information relating to the one or more categories into a list, and posts the summarized posted information on the community site.

2. A community system as described in claim 1, characterized in that the summary posting unit generates a category for each of the multiple posted information according to its content, and when multiple categories are assigned to the multiple posted information, generates the summarized posted information and posts it on the community site.

3. The community system according to claim 2, wherein the summary posting unit generates the summarized posted information when there are multiple groups of posted information assigned to the same category.

4. The community system according to claim 1, wherein the summarized posted information includes at least one of the name of the category, a description of the category, and examples of posted information belonging to the category.

5. The community system described in claim 1 further comprises a keyword-related posting unit that generates posted information related to a given keyword based on the keyword and posts the generated keyword-related posted information on the community site, and the summary posting unit generates the summarized posted information based on posted information posted by the multiple users as replies to the keyword-related posted information.

6. The community system described in claim 5, further comprising a derived keyword generation unit that generates a derived keyword based on the given keyword, and wherein the keyword-related posting unit generates the keyword-related posting information based on at least one keyword out of a plurality of keywords including the given keyword and the derived keyword generated by the derived keyword generation unit, and posts the keyword-related posting information on the community site.

7. The community system described in claim 5, characterized in that the predetermined trigger condition is that the number of posted information posted by the plurality of users as replies to the keyword-related posted information reaches a threshold value.

8. The community system described in claim 5 or 6, further comprising a category keyword setting unit that sets keywords indicating the category generated by the summary posting unit, and wherein the keyword-related posting unit further generates the keyword-related posting information based on the category keywords set by the category keyword setting unit.

9. The community system described in claim 8, further comprising a second derived keyword generation unit that generates derived keywords based on the category keywords set by the category keyword setting unit, and wherein the keyword-related posting unit generates the keyword-related posting information based on at least one keyword out of a plurality of keywords including the category keywords set by the category keyword setting unit and the derived keywords generated by the second derived keyword generation unit, and posts the keyword-related posting information on the community site.

10. A community system as described in claim 8, wherein the category keyword setting unit selects one of the multiple categories generated by the summary posting unit and sets a category keyword indicating the selected category.

11. The community system described in claim 8, characterized in that the keyword-related posting unit has a first mode in which the keyword-related posting information is generated using an arbitrarily set initial keyword as the given keyword, and a second mode in which the keyword-related posting information is generated based on the category keyword set by the category keyword setting unit, and is executed by switching from the first mode to the second mode.

12. The community system described in claim 11, characterized in that the summary posting unit includes introductory information about the keyword-related posting information that relates to the same category as one of the multiple categories included in the summarized posting information, in the keyword-related posting information that is generated by the keyword-related posting unit at the same time as the generation of the summarized posting information.

13. The community system described in claim 12, characterized in that the summary posting unit determines whether any of the multiple categories included in the summarized posting information matches a category corresponding to a keyword used by the keyword-related posting unit when generating the keyword-related posting information at the same time as the summarized posting information is generated, and if there is a match, includes introduction information for the keyword-related posting information in the summarized posting information.

14. A community processing device comprising: a posting information storage unit that stores multiple pieces of posting information posted by multiple users on a community site; and a summary posting unit that classifies the multiple pieces of posting information stored in the posting information storage unit into one or more categories according to their contents until a predetermined trigger condition is met, generates summarized posting information that compiles information related to the one or more categories into a list, and posts the summarized posting information on the community site.

15. The community processing device according to claim 14, further comprising a keyword-related posting unit that generates posted information related to a given keyword based on the keyword and posts the generated keyword-related posted information on the community site, wherein the summary posting unit generates the summarized posted information based on posted information posted by the plurality of users as replies to the keyword-related posted information.

16. The community processing device described in claim 15, further comprising a category keyword setting unit that sets keywords indicating the category generated by the summary posting unit, and wherein the keyword-related posting unit further generates the keyword-related posting information based on the category keywords set by the category keyword setting unit.

17. The community processing device described in claim 16, characterized in that the keyword-related posting unit has a first mode in which the keyword-related posting information is generated using an arbitrarily set initial keyword as the given keyword, and a second mode in which the keyword-related posting information is generated based on the category keyword set by the category keyword setting unit, and is executed by switching from the first mode to the second mode.

18. The community processing device described in claim 17, characterized in that the summary posting unit includes introductory information about the keyword-related posting information that is generated by the keyword-related posting unit at the same time as the generation of the summarized posting information, and that relates to the same category as one of the multiple categories included in the summarized posting information.

Images