JP2003216563A - Method for analyzing community and community analysis server - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To appropriately mature community while using the advantage of the community. <P>SOLUTION: A server 16 analyzes the community formed by a BBS or chat room or the like. The server 16 is provided with a text mining part 28 which gives an attribute showing contents every data including texts accepted from a terminal equipment based on data including texts accumulated in DB 12, a community participation log DB 26 which stores a participant who is a caller of the attribute and the data and a date on which the data is sent or accepted by relating to the data, and an analysis part 30 which analyzes the situation of the community based on the attribute, participant and/or data memorized. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for analyzing a community of data communication via a network.

[0002]

2. Description of the Related Art With the spread of the Internet, electronic bulletin boards (BBS), chat rooms, mailing lists, and other tools used for communication between users have become popular, and many users use such tools. The users of these communication tools will form a certain community. The community is
There are closed ones that require a certain registration, such as email addresses, and open ones that allow you to freely participate only by the handle name. In these communities,
There is a possibility that a large amount of wisdom will be born from the accumulation of statements by the participants through free discussion. In addition, since the community operator does not manage each utterance in detail, it is possible to reduce the size of the system including the server software.

[0003]

On the other hand, on the other hand, since the remarks are relatively free, repetitive reproach or slander causes the community to fail to mature and the community itself to decline. There is also a risk of doing it.

On the other hand, a so-called Q & A site in which questions and answers are formalized and a predetermined answerer answers the question of the questioner has been proposed and is actually in operation. In such a Q & A site, participants' statements are formalized into questions or answers, and the contents thereof are also managed. For this reason, it is necessary to provide a mechanism for management, and there is a problem that a mechanism including software such as a server becomes large and complicated. In addition, since it is formalized and managed, it is difficult for a great deal of wisdom to be born from the accumulation of remarks. However, the site has the advantage that it is easy to mature and the risk of being ruined by slander or slander is small.

An object of the present invention is to provide a method capable of appropriately maturing a community while taking advantage of the advantage of the community.

[0006]

SUMMARY OF THE INVENTION An object of the present invention is to receive data including text from a terminal device via a network and make it available for browsing to another terminal device, or the other terminal device. A method of analyzing a community formed by transmitting to a device, the method comprising: assigning an attribute indicating the content to each data containing text received from a terminal device based on data containing stored text. Storing the attribute, the participant who is the sender of the data, and the date and time when the data was transmitted or received, in association with the data, and based on the stored attribute, participant and / or date and time. ,
And analyzing the situation of the community.

According to the present invention, an attribute is added to a text input by a terminal operator (participant) in a community formed by exchanging data with a terminal device via a network such as a BBS or a chat room. ,
The situation of the community can be analyzed based on the attributes. As a result, the rise and fall of the community can be known, and the operators and managers can take appropriate measures.

[0008] In a preferred embodiment, the analyzing step obtains an analysis result showing a temporal transition of the community.

In a more preferred embodiment, the attributes are topic offering, opinion statement, opinion reaction indicating affirmation and denial to the topic or opinion, and
The step of analyzing, including a rebellion that shows a negative or aggressive view of the topic of the community itself, includes the step of counting the number of data having any of the above-mentioned attributes in a predetermined period, and the growth rate of the number of data in adjacent periods. And a step of determining the situation of the community based on the growth rate. This makes it possible to know the decline of community people, the deterioration of conversation quality, and the mannerism.

Further, in a preferred embodiment, the analyzing step includes the step of counting the frequency indicating the access in a predetermined period, the step of calculating the growth rate of the frequency indicating the access in the adjacent period, and the growth rate. Based on the above, determining the load status of the server that executes the process for forming the community. This allows the administrator to take appropriate measures for load balancing.

Further, in a preferred embodiment, the attribute is a rebellion in which a topic is provided in the community, an opinion statement, an opinion reaction indicating affirmation or denial to the topic or the opinion, and a denial or aggressive view of the topic of the community itself. And the step of analyzing includes the number of data with the attribute of opinion reaction in a predetermined period, and
Counting the number of data with the attribute of rebellion, calculating the rate of opinion reaction and the rate of rebellion from the number of data relating to the opinion reaction and the number of data relating to rebellion, respectively, based on the rate, Determining the quality. As a result, it is possible to know the deterioration or deterioration of the quality of the community, and it is possible to take measures against those who destroy the community (destroyer).

[0012] In another preferred embodiment, the analyzing step obtains analysis results for participants in the community. More specifically, the attributes are analyzed by including topic offering, opinion statement, opinion reaction indicating affirmation and denial to the topic and opinion in the community, and rebellion indicating negative or offensive views on the topic itself of the community. The step shows the step of counting the number of data items having the attributes of topic presentation and opinion statement of each participant, and the leader degree added after weighting the respective data amount of each participant item. Based on the step of calculating the frequency and the frequency indicating the leader level,
Determining the rank of the participants.

[0013] Further, the attribute analysis step includes and includes a topic offering in the community, an opinion statement, an opinion reaction indicating affirmation and denial of the topic or opinion, and a rebellion indicating a denial or an aggressive view of the topic of the community itself. However, the step of counting the number of data with each participant's opinion reaction indicating negative and the attribute of rebellion, and the destruction of the community weighted to the respective number of data of each participant The method may include a step of calculating a frequency indicating the degree of participation and a step of determining the rank of the participant based on the frequency indicating the degree of the destruction.

[0014] With these, it is possible to know the participants who can be leaders and the participants who may be annoying to the community. For these participants, necessary measures (for example, those who can be leaders can cooperate in the operation). , Etc.) can be taken. Further, in another preferred embodiment, the analyzing step is configured to perform a classification based on the text contained in the data.

Further, an object of the present invention is to receive data including text from a terminal device via a network,
A community analysis server that analyzes the community formed by making this available to other terminal devices or transmitting it to the other terminal devices, and the terminal device based on the data including the accumulated text. For each data including the text received from the text mining means for giving an attribute indicating the content, the attribute, the participant who is the sender of the data, and the date and time when the data was transmitted or received, The present invention is also achieved by a server comprising a participation log database stored in association with, and analysis means for analyzing the situation of a community based on the stored attributes, participants and / or date and time.

[0016]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the accompanying drawings. FIG. 1 is a block diagram showing an outline of the entire system according to the embodiment of the present invention. In the present embodiment, as shown in FIG.
Many personal computers (PC) 14-1, 14
-2, ... Are connected, and these PCs 14 form a community by accessing the server 16 that provides the community tool.

In the present embodiment, the server 16 exemplarily operates the BBS, but the present invention is not limited to this, and the user's utterance (input) such as chat and mailing list is once accepted by the server 16. , PC14
By the instruction from or by the processing of the server 16,
It may be configured to be transmitted to the PC 14 participating in the community. In the present specification, the user operating the PC 14 accessing the server 16 is also referred to as “participant” in some cases.

FIG. 2 shows a server 16 according to this embodiment.
3 is a block diagram showing the configuration of FIG. As shown in FIG. 2, the server 16 includes a communication management unit 18 that controls communication with the network 12, a BBS processing unit 20 that receives text input from participants and performs processing of adding the text to an electronic bulletin board, and participation. A bulletin board database (DB) 22 for accumulating text data input by a person, a log management unit 24 for grasping a participant's handle name, a date and time when an entry is made on a bulletin board, and a log for a participant in a community. Community participation log DB26 to accumulate and participant's B
Participants, etc. are referred to by referring to the data in the text mining unit 28 which executes text mining described later and the data in the community participation log DB 26 based on the input to the BS (hereinafter, also referred to as “speak” in some cases). The analysis unit 30 that performs an analysis regarding the analysis result, the analysis result DB 32 that stores the analysis result, the input from the operator (administrator) of the server 16 or the input from the user via the Internet, and the necessary analysis result is obtained. And a user interface (I / F) 34 for acquiring.

An outline of the processing in the server 16 having the above configuration will be described with reference to FIG. Participants speak to the BBS on the server 16 (arrows 301 to 3
(See 06), the community 300 is formed.
With respect to various remarks of participants in such a community, the server 16 according to the present embodiment, respectively.
Assigns an attribute by text mining (step 311) and stores it in the community participation log DB 26 (step 312).

On the other hand, when the community administrator or the like operates the input device of the server 16 or gives a command from an external personal computer, the data analysis unit 30 causes the community participation log DB 26 to be displayed. The necessary data is acquired from the accumulated data (step 313), the necessary analysis is executed (step 314), the analysis result is stored in the analysis result DB 32, and this is presented to the administrator or the like ( Step 31
5).

Here, a brief description will be given of the remarks and text mining in the community 300. For example, as shown in FIG. 3, when the participant "A" provides some topic (see arrow 301), there are various reactions to this. For example, there is a case where the participant “A” responds to the topic “Is the Internet effective?” In the following order. -Participant "B": "Email is effective ..."-Participant "C": Negative reaction (denial) -Participant "D": "Communication within the company is smooth ..."- Participant "E": "It doesn't mean ..." ・ Participant "F": Positive reaction (affirmation)

In text mining, necessary keywords are found from the speech of each participant and attributes of the speech of the participant are added. In the present embodiment, any one of the attributes of “topic provision”, “opinion statement”, “opinion reaction (negative)”, “opinion reaction (affirmative)”, and “rebellion” is added to the participant's statement. Configured to give.

"Providing a topic" means creating a new thread in the BBS or presenting a new topic in the same thread. "Opinion statement" means
It means that he / she made an opinion, such as a proposal or a proposal, with respect to the provided topic. "Opinion reaction (denial)" and "opinion reaction (affirmation)" mean that a negative opinion and a positive opinion are expressed with respect to the statements belonging to "topic offer" and "opinion statement", respectively. is doing. Also,
“Rebellion” means something that is unrelated to the topic in progress, confuses other participants, or slanders the statements of other participants.

When the PC 14 accesses the server 16 and the participant operates the input device to speak, the BBS processing unit 2
With 0 or the like, the content of the statement is stored in the bulletin board DB 22 along with the date and time of the statement and the handle name of the participant. By accessing the server 16, the other participants can refer to the above-mentioned content of the statement and can speak by themselves.

The text mining unit 28 according to the present embodiment retrieves the contents of the message accumulated in the bulletin board DB 22 at a predetermined timing (for example, once a day at 5 am) and batch-processes each message. Execute the process to determine the attribute of. In the present embodiment, the attributes of the remarks are determined by batch processing, but the present invention is not limited to this, and each time there is a remark from the participant,
The text mining unit 28 may determine the attribute of the statement based on the content of the statement of the user.

FIG. 4 is a flowchart showing the processing executed by the text mining unit 28. When the text mining unit 28 is activated, the utterances of the participants accumulated from the previous process to the current process are acquired from the bulletin board DB 22 (step 401). Next, the content of the message is scanned (step 402), and it is determined whether or not a predetermined keyword is included (step 403). A plurality of this keyword is prepared for each of the above attributes. Therefore, when the keyword is found, the count (attribute count) value of the attribute to which the keyword belongs is incremented (step 404). Steps 402-40
The processing of 4 is repeated until the end of the speech of a participant is reached (see step 405). From the attribute count value, it is possible to know which attribute keyword is included and how many in the speech.

Therefore, the text mining unit 28
Compares the attribute count values for each attribute, determines the attribute with the largest count value as the attribute for the utterance, and stores the attribute in the community participation log DB 26 together with the date and time and the information that identifies the participant. (Step 407).

If there remains any message accumulated in the bulletin board DB 22 (Yes in step 408), the above steps 401 to 407 are repeated. In this way
An attribute can be given to each of the statements in a predetermined period. FIG. 5 is a diagram for explaining the data in the community participation log DB 26. Here, the date and time (see reference numeral 501), the participant who made the statement (see reference numeral 502), and the attribute to be given (see reference numeral 503) can be specified for each statement. For example, it can be seen that the participant "A" made a statement belonging to "topic offer" at "12:00: 00" on "December 1".

In FIG. 5, each attribute column is provided to indicate the attribute and a flag can be set in a predetermined column, but the present invention is not limited to this.
It goes without saying that the text or the number indicating the attribute may be arranged in a single column. Further, for items whose attributes cannot be specified, columns such as "unknown" and "other" may be provided. In this way, when the data to which the date and time, the participant, and the attribute are added is accumulated for each utterance, the analysis unit 30 generates data with various items as keys based on the accumulated data. In the present embodiment, the analysis unit 30 generates the above-mentioned data at a predetermined timing (for example, once a week, at 5 am on Sunday), but the present invention is not limited to this, and the management is not limited to this. It may be configured to generate the necessary data in response to the request of the person.

As shown in FIG. 1, in the present embodiment, data for observing the overall tendency such as the number of participants and the number of messages belonging to each attribute for a predetermined period (for example, one hour) are stored. In response to a request from the administrator, an overall analysis unit 36 that generates an analysis result indicating the overall tendency from the above data, and data for viewing the tendency of the speech for each participant are generated. A participant analysis unit 38 that executes a predetermined analysis and an analysis support unit 40 that analyzes the content of the statement (text) are included as necessary.

For example, the overall analysis unit 36 is based on the data in the community participation log DB 26, and is in the form of tabular data (total analysis table 42) showing the number of participants and the total number of statements for each predetermined period, and the number of each attribute of the statement. ) Is generated. Reference numeral 600 in FIG. 6 is a diagram showing an example of the overall analysis table 42. As shown in FIG. 6, in the table 600, the number of participants, the number of statements, each attribute (topic offer, opinion statement, opinion reaction (negative), opinion reaction (positive) and hourly (see reference numeral 601) and The number of rebellions is shown.
6 is, for example, in batch processing at a predetermined timing,
The overall analysis table 42 may be generated and stored in the analysis result DB 32. Alternatively, it may be configured to generate this in response to an instruction from the administrator.

The overall analysis table 42 is used to know the overall tendency of the community, as will be described later. In the present embodiment, the community year / month transition (see reference numeral 611) for analyzing the transition of the situation of the community on a year-by-month basis (for example, on a daily basis) and the community time analysis for analyzing the transition of the situation of the community on an hourly basis. (See reference numeral 612) and community quality analysis (see reference numeral 613) for analyzing the qualitative situation of the community can be executed. 9A to 9C are flowcharts showing the respective processes.

As shown in FIG. 9A, in the transition of the community year / month, first, the tabular data (entire analysis table 42) is acquired from the analysis result DB 32 (step 901). The overall analysis unit 36 calculates the growth rate of the total number of participants, the number of participants, the number of remarks, the number of topics, etc. for each day (between all days and the current day) (step 902). Next, the overall analysis unit 36 refers to the growth rate of each item to determine the situation of the community (step 903).

More specifically, for example, when the growth rate of the number of participants and / or the total number of participants is continuously negative, it is determined that the community is declining. If the growth rate of the number of utterances is continuously negative, it is recognized that the quality of the conversation is deteriorated. Alternatively, if the growth rate of the number of topics is continuously negative, it is determined that the topics are rut. It should be noted that the term continuously minus here is not limited to staying negative without becoming positive for a predetermined period of time, and the negative continues for a predetermined period of time. Cases are also included.

The overall analysis unit 36 generates an image showing necessary measures based on the obtained judgment result such as the situation of the community (for example, decline, deterioration of quality, rut, etc.) and presents it to the manager. (Step 904). The community time analysis and community quality analysis by the overall analysis unit 36 will be briefly described below.

In the time analysis, the overall analysis unit 36
After obtaining the tabular data (overall analysis table 42) (step 911), the growth rate of the total number of participants, the number of participants, the number of statements, the number of topics, etc. by time (between adjacent time zones) is calculated. (Step 912) With reference to the growth rate of each item, the situation regarding the number of accesses such as the participation time and the number of participations by the participants is grasped (step 913). Thereby, the server load can be calculated, and a suggestion of server distribution according to the server load can be presented (step 914).

In the quality analysis, the overall analysis unit 36
After obtaining tabular data (overall analysis table 42) (step 921), the rebellion rate (the ratio of the anti-random number to the number of messages and / or the ratio of the anti-random number to the number of topics) and the reaction rate (the number of messages) Reaction to (positive and negative)
The ratio of the number and / or the ratio of the number of reactions (positive and negative) to the number of topics is calculated (step 92).
2). For example, when it is determined that the reaction rate is negative in time series, the overall analysis unit 36 determines that the quality of the community is degraded, while the repulsion rate is time series. If it is determined to be positive, it is determined that the quality of the community is degraded (step 923). In this way, when the quality of the community is deteriorated, a statement that has an attribute of rebellion, that is, a measure against the community destroyer is presented (step 924).

Next, the processing by the participant analysis unit 38 will be described. The participant analysis unit 38 uses tabular data (participant analysis table 4) for checking the tendency of remarks for each participant based on the data in the community participation log DB 26.
4) is generated. Reference numeral 700 in FIG. 7 is a diagram showing an example of the participant analysis table 44. As shown in FIG.
In 00, the number of participants for each participant (the number of statements: code 7
01), attributes (see reference numeral 702), and evaluation of participants (see reference numeral 703). Participant Analysis Department 3
Similarly to the overall analysis unit 36, 8 generates the participant analysis table 44 by batch processing, and the analysis result DB 32
Remember. Alternatively, it may be generated in response to an instruction from the administrator. The participant may be evaluated by giving a score to each attribute and assigning it to the range to which the sum total belongs. For example, “+ a” is provided for topic offering, “+ b (where a> b)” is provided for opinion, and “−” is provided for rebellion.
You just have to decide the score such as "c".

The participant analysis table 44 is used to know the tendency of each participant participating in the community.
In the present embodiment, a community leader analysis (see reference numeral 711) for finding a participant who can be a community leader and a community disrupter analysis (see reference numeral 712) for finding a participant who becomes a destroyer are executed. It has become so. 10 (a) and 10 (b)
FIG. 7 is a flowchart showing a process in a community leader analysis and a community vandal analysis.

As shown in FIG. 10 (a), in the community leader analysis, the participant analysis section 38 acquires tabular data (participant analysis table 44) (step 1001), and the period (for example, week). The number of topics provided and the number of opinions made by each participant are grasped (step 10).
02). Next, the leader frequency is calculated based on the grasped number of topics provided and the number of opinions and comments (step 100).
3). This is done by assigning a score to each of the number of topics provided and the number of opinions and comments, and obtaining the sum of the scores assigned to each of the participants' remarks within a predetermined period, and using the sum of the scores to give an index indicating the degree of leadership. Give it. Next, the participant analysis unit 38 creates a ranking of the index indicating the leader degree and presents it to the manager (step 100).
4). By referring to this ranking, the administrator can request, for example, a participant in the top ranking to cooperate in the management of the community as a leader.

On the other hand, as shown in FIG. 10B, in the community vandal analysis, after obtaining the tabular data (participant analysis table 44) (step 1011), the anti-random number and the opinion reaction number (negative ) Is grasped (step 1012), and the vandal frequency of each participant is calculated based on these (step 1013). This is done by assigning a score to each of the anti-random number and opinion reaction number (negative), obtaining the sum of the scores assigned to each utterance of the participants within a predetermined period, and determining the destroyer frequency according to the sum of the scores. It is sufficient to give the index shown. Next, the participant analysis unit 38
A ranking of an index indicating the frequency of vandals is created and presented to the manager (step 1004). The administrator refers to this ranking to restrict participation in the community to participants whose vandal count exceeds a prescribed value (for example, denying access by the participant's handle name), etc. It is possible to take the necessary measures.

Next, the processing in the analysis support unit 40 will be briefly described. The analysis support unit 40, for example,
A classification-based data table (see reference numeral 46 in FIG. 1) in which the comments are classified is generated using the attributes of the comments of the participants as keys.
FIG. 8 is a diagram showing an example of a classification-based data table. As shown in FIG. 8, in the tabular data 800,
In association with the item of classification (see reference numeral 801),
A text indicating the content of the user's statement (see reference numeral 802),
The participant who made the statement (see reference numeral 803) and the date and time (see reference numeral 804) are displayed.

The analysis support unit 40 generates the above-mentioned classification-specific data table and the classification-specific data table using other items as keys, if necessary, and, in response to a request from the administrator, information required by the administrator. Can be presented. The analysis support unit 40 may extract the text for each attribute or calculate the total number of messages for each attribute based on the classification-based data table 46. In this way, the analysis support unit 40 can execute the trend analysis and the keyword analysis at the content level based on the content of the extracted text. In addition, the volume of the topic can be grasped based on the calculated total number. The analysis support unit 40 can present these analysis results to the manager, and can implement product development and problem analysis according to the instructions from the manager.
For example, if you have a community about product evaluation,
If the number of topics is large, it is determined that a problem may occur in the product, and it is understood that urgent action is required. In addition, in a community related to product evaluation, it is possible to extract product problems by performing text keyword analysis.

According to the present embodiment, the comments of the participants to the community are accumulated, and the community and the participants can be analyzed by the accumulated comments of the participants.
This makes it possible to take appropriate measures to manage the community smoothly. The above measures are not only for the participants who ask the participants to participate in the operation as leaders, or to prohibit the participants' access, but also for equipment such as load balancing by distributing the servers. It also includes those related to.

The above-described embodiment is a community that gives an opinion about a specific product in order to evaluate and develop the product.
It can be applied to various communities, such as a community that gives opinions on the optimization and efficiency of work within a company. Here, unlike the so-called Q & A site, participants can freely speak without being put in the form of questions and answers. On the other hand, depending on the evaluation of the participants, it is possible to provide some protection against the destruction of the community. The present invention is not limited to the above embodiments, and various modifications can be made within the scope of the invention described in the claims, and those are also included in the scope of the present invention. Needless to say.

For example, in the above-described embodiment, the overall analysis unit 36 and the participant analysis unit 38 respectively use the overall analysis table 42 and the participant analysis table 4 respectively.
4 has been created, and various analyzes such as year-month changes and time changes have been executed. However, the present invention is not limited to the analysis shown in the above-described embodiment, and it may be configured to execute a part of these or to be able to execute other analysis. Is clear.

Further, in the above-described embodiment, the quality of the community is judged based on the rebellion rate and the reaction rate, and the countermeasures against the destroyers are presented to the manager (FIG. 9 (c)). )), In addition to this, when there is a remark that includes slander or slander of another person, notify the administrator of the remark and notify the administrator to take necessary measures. Is also desirable.

Further, although a single community is operated in the above-mentioned embodiment, the present invention can be used to operate a plurality of communities. In this case, the method described in the above embodiment is adopted for the evaluation of each community, and in addition to this,
Necessary data etc. can be exchanged between communities. For example, in each community, based on the judgment of the situation, the attributes of the necessary participants are specified, and the participants of other communities who have the necessary attributes are identified as their own community. You may ask for participation.

For example, in a community where the number of opinions and reactions (affirmation) is large compared to the number of topics provided and opinions and comments, it is desirable that participants who provide topics or make comments and opinions participate. In this case, the desired attribute is transmitted to the other community, and the information regarding the participant of the attribute is notified in the other community. If the e-mail address of the participant is known, the administrator of the community may send an e-mail asking the participant to participate in the community. On the other hand, if the participant's e-mail address is unknown and personal notification to the participant is impossible, a message from the administrator will be posted on the bulletin board etc. in the community in which the participant is participating. , Leave a message to encourage participation in other communities.

The terminal device accessing the server is
The present invention is not limited to a personal computer, and may be a mobile terminal such as a mobile phone or a PDA. As a network, in addition to the Internet, a LAN in a company.
Or WAN may be used. Furthermore, in the above-described embodiment, the server 16 that executes the processing associated with the operation of the BBS, the chat room, etc. is provided with various analysis functions, but the present invention is not limited to this, and the accumulated participants It is needless to say that a separate server (analysis server) may be configured to be able to perform overall analysis, participant analysis, etc., based on the above statement.

[0051]

According to the present invention, it is possible to provide a method capable of appropriately maturing a community while taking advantage of the community.

[Brief description of drawings]

FIG. 1 is a block diagram showing an outline of an entire system according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of a server according to the present embodiment.

FIG. 3 is a diagram showing an outline of processing by a server according to the present embodiment.

FIG. 4 is a flowchart showing a process executed by the text mining unit according to the present embodiment.

FIG. 5 is a diagram for explaining data in a community participation log DB according to the present embodiment.

FIG. 6 is a diagram for explaining data in the community participation log database according to the present embodiment.

FIG. 7 is a diagram for explaining data in the community participation log database according to the present embodiment.

FIG. 8 is a diagram for explaining data in a community participation log database according to the present embodiment.

FIG. 9 is a flowchart showing processing executed by the overall analysis unit according to the embodiment of the present invention.

FIG. 10 is a flowchart showing processing executed by a participant analysis unit according to the embodiment of the present invention.

[Explanation of symbols]

12 networks 14 personal computer 16 servers 18 Communication Management Department 20 BBS processing unit 22 Bulletin board DB 24 Log Management Department 26 community participation log DB 28 Text Mining Department 30 Analysis Department 32 Analysis result DB 34 User I / F 36 Overall Analysis Department 38 Participant Analysis Department 40 Analysis Support Department

Claims (18)

[Claims] 1. A community formed by receiving data including text from a terminal device via a network and making the data available to other terminal devices or transmitting the data to the other terminal devices. A method of analyzing the above, wherein, based on the data including the stored text, a step of assigning an attribute indicating the content of each data including the text received from the terminal device, the attribute, and the sender of the data. Storing a certain participant and the date and time when the data was transmitted or received in association with the data, and analyzing the situation of the community based on the stored attribute, the participant and / or the date and time. A method characterized by having. 2. The method according to claim 1, wherein the analyzing step obtains an analysis result indicating a temporal transition of a community. 3. The analysis includes the topic providing, opinion statement, opinion reaction indicating affirmation and denial to the topic or opinion, and rebellion indicating a negative or offensive view of the topic of the community, wherein the attribute includes: The step of counting the number of data having any of the attributes in a predetermined period,
The method according to claim 2, further comprising: a step of calculating a growth rate of the number of data in adjacent periods, and a step of judging a situation of the community based on the growth rate. 4. The analyzing step includes the step of counting the frequency indicating the access in a predetermined period, the step of calculating the growth rate of the frequency indicating the access in an adjacent period, and the community based on the growth rate. Determining the load status of the server that executes the processing for forming
Or the method described in 3. 5. The analysis includes the topic providing, opinion statement, opinion reaction indicating affirmation and denial to a topic or an opinion, and rebellion indicating a negative or offensive view of the topic itself of the community, wherein the attribute includes: The step of counting the number of data having the attribute of opinion reaction and the number of data having the attribute of rebellion in a predetermined period, and the number of data relating to the opinion reaction and the number of data relating to rebellion, respectively. 5. The method according to any one of claims 2 to 4, comprising the steps of calculating the rate of religion and the rate of rebellion and determining the quality of the community based on the rate. 6. The method according to claim 1, wherein the analyzing step obtains an analysis result regarding participants of the community. 7. The analysis, wherein the attribute includes topic offering in the community, opinion statement, opinion reaction indicating affirmation and denial to the topic or opinion, and rebellion indicating denial or offensive opinion to the topic itself of the community, The step of counting the number of data items having the attributes of topic presentation and opinion statement of each participant, and the leader degree added after weighting the respective data items of each participant. 7. The method of claim 6, comprising the steps of calculating an indicated frequency and determining a rank of a participant based on the reader indicative frequency. 8. The analysis includes the topic providing, opinion statement, opinion reaction indicating affirmation and denial to a topic or an opinion, and rebellion indicating a negative or offensive view of the topic of the community, wherein the attribute includes The step of counting the number of data with each participant's opinion reaction indicating negative and the attribute of rebellion, and adding the data after weighting the respective number of data of each participant 8. The method according to claim 6, further comprising: a step of calculating a frequency indicating the degree of vandalism, and a step of determining a rank of the participant based on the frequency indicating the degree of vandalism. 9. A method according to any one of the preceding claims, characterized in that the analyzing step is arranged to perform a classification based on the text contained in the data. 10. A community formed by receiving data including text from a terminal device via a network, making the data available to other terminal devices, or transmitting the data to the other terminal devices. Is a community analysis server that analyzes, based on the data including the stored text, a text mining unit that assigns an attribute indicating the content to each data including the text received from the terminal device, the attribute, and the data. The participation log database that stores the participant who is the sender and the date and time when the data was transmitted or received, in association with the data, and the situation of the community based on the stored attribute, participant and / or date and time. And a analyzing means for analyzing the server. 11. The server according to claim 10, wherein the analysis unit is configured to obtain an analysis result indicating a temporal transition of a community. 12. The analysis, wherein the attribute includes topic offering, opinion statement, opinion reaction indicating affirmation and denial of a topic or opinion in the community, and rebellion indicating a denial or offensive view of the topic itself of the community, The means is configured to count the number of data having any of the attributes in a predetermined period, calculate a growth rate of the number of data in an adjacent period, and determine the situation of the community based on the growth rate. The server according to claim 11, which has been executed. 13. The analysis means counts the frequency indicating the access in a predetermined period, calculates the growth rate of the frequency indicating the access in an adjacent period, and based on the growth rate, the load condition of the server. 13. The server according to claim 11, wherein the server is configured to determine 14. The analysis includes the topic providing, opinion statement, opinion reaction indicating affirmation and denial to a topic or an opinion, and rebellion indicating a negative or offensive view on the topic itself of the community, wherein the attribute includes: The means counts the number of data having the attribute of opinion reaction and the number of data having the attribute of rebellion in a predetermined period, and the rate of opinion reaction and the number of data relating to rebellion are respectively calculated from the number of data relating to the opinion reaction and the number of data relating to rebellion. The server according to any one of claims 11 to 13, wherein the server is configured to calculate a rate of rebellion and determine the quality of the community based on the rate. 15. The server according to claim 10, wherein the analysis unit is configured to obtain an analysis result regarding participants of the community. 16. The analysis includes the topic providing, opinion statement, opinion reaction indicating affirmation and denial to a topic or an opinion, and rebellion indicating a negative or offensive view of the topic itself of the community, wherein the attribute includes: The means counts the number of data items having the attributes of topic presentation and opinion statement of each participant,
It is configured to calculate the frequency indicating the leader degree, which is obtained by weighting the respective data numbers of the respective participants, and to determine the rank of the participant based on the frequency indicating the leader degree. The server according to claim 15, wherein: 17. The analysis includes the topic providing, opinion statement, opinion reaction indicating affirmation and denial to a topic or an opinion, and rebellion indicating a negative or offensive view of the topic of the community, wherein the attribute includes The means for destroying the community, in which the means counts the number of data having the opinion reaction indicating negative and the attribute of rebellion of each participant, weights the respective number of data of each participant, and adds them. The server according to claim 15 or 16, wherein the server is configured to calculate a frequency indicating a degree and determine the rank of the participant based on the frequency indicating the degree of the destroyer. 18. The server according to claim 10, wherein the analysis unit is configured to perform classification based on the text included in the data.