JP2005234689A - Community evaluation support method and apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide analytic data that can reveal the situation of a community on a network. <P>SOLUTION: A community evaluation support method has an input reception step of receiving classification standard information storing classification identifiers for classifying users' behavior from an administrator terminal 2, an extraction step of extracting an access history of user terminals 3 in a predetermined period from log information, a classification step of dividing the access history of the user terminals 3 into user units and classifying each user as any classification identifier of the classification standard information, an analysis step of calculating an evaluation value of clients belonging to each classification identifier for every classification identifier to analyze a community for every classification identifier, and an output step of outputting the analysis results by the analysis step to an output device. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a technique for supporting evaluation of a community existing on a network.

  Patent Document 1 discloses a technique for analyzing a community formed by users using a bulletin board service from information written in a bulletin board service that is an electronic bulletin board via a network. In other words, the information written on the bulletin board is text-mined, and the information written on the bulletin board is classified into "topics provided", "opinions", "aggressive opinions", etc., and the situation and quality of the community are analyzed . Text mining is a technique for analyzing a large amount of text data (sentences, words, etc.) and searching for correlations and patterns of text data lurking in the text data.

JP 2003-216563 A

  An administrator who manages and operates a website that provides various services such as bulletin board services, questionnaires, and product sales, for example, many users access the website, and more users have access to various services on the website. We will consider improvement measures to improve websites, such as improving services, advertising, and promotions. Therefore, the administrator analyzes various actions of users who access the Web site, and considers improvement measures such as measures for converting users into customers and means for increasing repeat users. In addition, after implementing these improvement measures, the administrator verifies how the situation of the community formed by the users accessing the website has changed by the improvement measures, and the future operation of the website. It is necessary to make use of it.

  However, in the above-mentioned Patent Document 1, no consideration is given to user behavior other than when writing text on a bulletin board service on a website. For example, actions such as “when reading contents published on a bulletin board” or “when browsing contents (Web pages) other than the bulletin board” are actions that cannot be ignored in evaluating the community.

  In Patent Document 1, information written on the bulletin board service is text mined. In general, text mining processing places a heavy load on the CPU. For this reason, it is necessary to introduce a high-performance computer system in order to ensure high analysis accuracy and high processing speed.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to provide analysis data that can grasp the status of a community on a network based on various actions of a user accessing a website. There is to do.

  In order to solve the above-described problems, the present invention supports the evaluation of a community by classifying user actions from log information and analyzing the community using a computer system.

  For example, the storage device stores log information that records the access status of a plurality of clients accessing the computer system.

  And an input receiving step for receiving, from an external system, an input of classification criteria information storing at least one classification identifier for classifying the behaviors of the plurality of clients on the network and community evaluation information to the arithmetic device; An extraction step for extracting the access histories of the plurality of clients in a predetermined period from the log information, and dividing the access histories of the plurality of clients for each client, and classifying each client based on the access history of each client A classification step for classifying into one of the classification identifiers of the reference information, and for each classification identifier, the evaluation value of the client belonging to the classification identifier is calculated based on the evaluation information and the access history of each client belonging to the classification identifier And for each classification identifier, An analysis step of analyzing the community, the results of analysis in the analyzing step, and an output step of outputting to the output device, to perform.

  In the present invention, it is possible to provide analysis data capable of grasping the status of a community on a network based on various actions of a user accessing a website.

  Hereinafter, embodiments of the present invention will be described.

  In the present embodiment, the community is a group of users who access any one of the Web pages provided by the service providing server 1 or a user who accesses a specific Web page (for example, a bulletin board service). An assembly consisting of The evaluation of the community is to determine whether or not the community is preferable for the administrator of the website based on the behavior of users constituting the community, the usage status of various services, communication between users, and the like. In the present embodiment, the community increases the number of accesses to the entire website by improving a bulletin board that promotes communication between users and a questionnaire service (web page) for collecting user information. The present invention will be described as an evaluation standard. In addition, this invention is good also as providing the reference | standard of evaluation of another community, without being limited to this.

FIG. 1 is a schematic diagram of a community evaluation support system to which an embodiment of the present invention is applied. As shown in the figure, the community evaluation support system of the present embodiment includes a service providing server 1, an administrator terminal 2, and at least one user terminal 3 _{1 to} 3 _n . These devices are connected via the Internet, an intranet, a LAN, or other wired or wireless network 9.

  The service providing server 1 is an information processing apparatus that provides various services such as bulletin boards, questionnaires, information provision, and product sales to the user terminal 3 via the network 9. The service providing server 1 is an information processing apparatus that provides the administrator terminal 2 with analysis information necessary for community evaluation.

  The service providing server 1 includes an input receiving unit 11 that receives registration of classification criterion information 300 and the like described later from the administrator terminal 2, a log extracting unit 12 that extracts an access history 400 of each user terminal 3 described later, To the level setting unit 13 that sets the level for each classification code of the reference information 300, the analysis unit 14 that analyzes the access history 400, the display processing unit 15 that outputs the analysis result to the administrator terminal 2, and the user terminal 3 A service providing unit 16 that provides various website services such as a bulletin board service, a communication processing unit 17 that transmits / receives data to / from other devices via a network, classification criteria information 300, an access history 400, and an analysis result table to be described later And a storage unit 18 for storing various information such as 1200.

  The administrator terminal 2 is an information processing device that receives a community analysis result analyzed by the service providing server 1 via a network and displays it on an output device. Then, the administrator terminal 2 includes an input receiving unit 21 that receives input of the classification standard information 300 and evaluation items described in FIG. 3, a display unit 22 that displays the analysis result of the community analyzed by the service providing server 1, And a communication processing unit 23 that transmits and receives data to and from other devices via the network. The evaluation item is an item that serves as an index for the administrator to perform community evaluation. In this embodiment, the criterion for community evaluation is to increase the number of accesses of the entire website provided by the service providing server 1 by improving the bulletin board service and the questionnaire. Therefore, the evaluation item is the number of accesses of the entire website.

  Note that the administrator who uses the administrator terminal 2 is a person who manages the website provided by the service providing server 1 and evaluates the community using the community analysis result analyzed by the service providing server 1. . Specifically, the administrator accesses (visits) many users from the service providing server 1 and examines and implements various improvement measures such as improvement of services and implementation of promotions so that the services can be used. In addition, the administrator will grasp the effects of implementing the improvement measures, the degree of community growth, and so forth, and will make further improvements.

  The user terminal 3 is an information processing device that uses various services of a Web site provided by the service providing server 1. The user terminal 3 acquires various information provided by the service providing server 1 using a web browser or the like based on a user instruction, and displays the information on the output device of the user terminal 3. Further, the user terminal 3 receives information input from the input device based on a user instruction and transmits the information to the service providing server 1. That is, the user can use various services on the website using the user terminal 3 such as browsing or writing information written on a bulletin board, answering a questionnaire, searching for information, and purchasing a product. The web browser is application software for browsing information published on the Internet.

  In the present embodiment, the service providing server 1 and the administrator terminal 2 are each one. However, the service providing server 1 and the administrator terminals 2 may be plural.

  The service providing server 1, the administrator terminal 2, and the user terminal 3 described above each include a CPU 901 as shown in FIG. 2, a memory 902 such as a RAM, an external storage device 903 such as an HDD, a keyboard, and the like. And a general-purpose computer system including an input device 904 such as a mouse, an output device 905 such as a display or a printer, a communication control device 906 for connecting to a network, and a bus 907 for connecting these devices. Can be used. In this computer system, the CPU 901 executes a predetermined program loaded on the memory 902, thereby realizing each function of each device.

  For example, each function of the service providing server 1, the administrator terminal 2, and the user terminal 3 is a program for the service providing server 1, and the CPU 901 of the service providing server 1 is a program for the administrator terminal 2. Is implemented by the CPU 901 of the administrator terminal 2 and the CPU 3 of the user terminal 3 in the case of a program for the user terminal 3, respectively. In this case, the memory 902 or the external storage device 903 of the service providing server 1 is used for the storage unit 18 of the service providing server 1.

  Next, the classification reference information 300 will be described. The classification reference information 300 is obtained by associating and classifying user actions that serve as indices when an administrator evaluates a community with address information. The classification standard information 300 is input to the input receiving unit 21 of the administrator terminal 2 by the administrator using the input device 904. The classification standard information 300 input to the administrator terminal 2 is transmitted to the service providing server 1 via the network 9 and stored in the storage unit 18 of the service providing server 1.

  FIG. 3 shows an example of the table configuration of the classification reference information 300. The classification reference information 300 includes a classification code 301, a classification name 302, and address information 303, as illustrated. The classification code 301 is set with a code (number, alphabet, etc.) for identifying the user's behavior that is subject to community evaluation. In addition, the action name of the user who is the target of community evaluation is set in the classification name 302.

  Further, in the address information 303, identification information indicating destinations of various services (Web pages) corresponding to the user's behavior that is subject to community evaluation is set. For example, a URL (Uniform Resource Locator) or a URI (Uniform Resource Identifier) is used as the address information 303. In the present embodiment, a URL is used.

Here, in the present embodiment, the criterion for community evaluation is to increase the number of accesses of the entire Web site provided by the service providing server 1 by improving the bulletin board service and the questionnaire. Therefore, the user's actions for the bulletin board and the questionnaire service are classified and registered in the classification reference information 300.
In the example shown in FIG. 3, five actions with classification codes 301 “C001” to “C005” are registered. A classification name 302 with a classification code 301 of “C001” is “access to home page”, which means access to a top page (home page) that is the entrance of a Web site. In the corresponding address information 303, the URL of the home page is set.

  A classification name 302 having a classification code 301 of “C002” is “browsing bulletin board”, which means that an existing bulletin board is accessed and contents written on the bulletin board are browsed. In the corresponding address information 303, the URL of the bulletin board is set.

  A classification name 302 with a classification code 301 of “C003” is “write to bulletin board”, which means that some information (opinions / sentences) is written to an existing bulletin board. In the corresponding address information 303, a URL for writing to the bulletin board is set.

  A classification name 302 whose classification code 301 is “C004” is “installation of bulletin board”, which means that a new bulletin board is installed. That is, when a user wants to communicate with another user about a topic that the user is interested in, this means that a new bulletin board related to the topic is installed. In the corresponding address information 303, a URL for newly installing a bulletin board is set.

  A classification name 302 with a classification code 301 of “C005” is “questionnaire answer”, which means that a user answers a questionnaire published on a website. In the corresponding address information 303, a URL for answering the questionnaire is set.

The address information 303 is a destination when the user terminal 3 accesses the service providing server 1, and is acquired in the log file of the service providing server 1 every time the user terminal 3 accesses the service providing server 1, That is, it is information included in an access history 400 described later. Thus, by creating the classification reference information 300 in which the address information 303 included in the access history 400 is associated with the user's behavior, the user's behavior can be analyzed using the access history 400.
Next, the access history 400 will be described. The access history 400 is a storage unit that extracts information necessary for community evaluation from a log file acquired (accumulated) by the log extraction unit 12 of the service providing server 1 by a function such as an OS (Operating System) (not shown). 18 is stored. Note that the log file (not shown) is an access record from the user terminal 3 stored in the file, and access data (user information such as the host name and IP address of the user terminal, user ID, access information) for each access. Date, access destination such as URL, route information, etc.) are added.
FIG. 4 shows an example of the table structure of the access history 400. Each record of the access history 400 has a user ID 401 and address information 402 that are data items of access data stored in the log file. In the access history 400, records for a predetermined period are stored in time series.

  Next, the flow of level setting processing of the service providing server 1 in this embodiment will be described. The level setting process is a process performed before executing an analysis process (see FIG. 6) to be described later. A level is set for each classification code (user action) set in the above-described classification standard information 300 (see FIG. 3). Is a process of adding. Levels are set in the order from the user who is likely to take an action (small conscious resistance and low threshold) to the person who does not easily take action (high conscious resistance and high threshold). To do.

  For example, in a certain three behaviors A, B, and C, if there are many users who first take the behavior of A and then act in the order of B and C, then the behavior pattern of A → B → C is common Conceivable. In this case, A can be considered to have less conscious resistance (lower threshold) when taking action than B is B and B is C. In this case, the level is set to increase in the order of A, B, and C. Thus, by setting the level in the user's behavior, it is possible to grasp the degree of community growth.

  FIG. 5 is a process flow diagram of the level setting process of the service providing server 1.

  First, the input receiving unit 11 receives the classification reference information 300 and the evaluation items (the number of accesses to the entire website) transmitted from the administrator terminal 2 via the communication processing unit 17 and stores them in the storage unit 18 ( S501). Next, the log extraction unit 12 extracts an access history 400 (see FIG. 4) for a predetermined period from the log file and stores it in the storage unit 18 (S502). The predetermined period is, for example, a predetermined period such as one month or six months.

  Next, the level setting unit 13 reads the access history 400 stored in the storage unit 18, converts each address information 402 of the access history 400 to the classification code 301 set in the classification reference information 300, and stores it in the storage unit 18. (S503). That is, the level setting unit 13 converts each address information 403 of the access history 400 into a corresponding classification code 301 with reference to the classification standard information 300 stored in the storage unit 18. The level setting unit 13 deletes the record having the address information 403 that is not registered in the classification reference information 300.

  6 shows a code conversion of the access history 400 shown in FIG. As illustrated, the access history 600 after code conversion has a user ID 601 and a classification code 602. For example, in the case of the record in the first line of the access history 400 shown in FIG. 4, the address information 402a is “http://aaa.bbb.co.jp”. Therefore, the level setting unit 13 refers to the classification standard information 300 and converts the address information 402a into the corresponding classification code “C001” (see FIG. 6: 602a). Further, in the case of the address information 402 b of the records in the second and third lines, the address information 402 b does not exist in the classification reference information 300. Therefore, the level setting unit 13 deletes these records. Further, in the case of the address information 402 c of the records in the fourth and fifth lines, the respective address information 402 c exists in the classification reference information 300. Therefore, the level setting unit 13 converts the address information 403 into corresponding classification codes (“C002” and “C003”) (see FIG. 6: 602c).

Next, the level setting unit 13 divides the access history 600 after code conversion for each user (S504).
FIG. 7 shows an example of the access history 700 divided for each user, in which actions only for the user with the user ID “U001” (that is, classification codes) are arranged in time series. The access history 700 for each user includes a No 701 in which numbers are set in order from “1”, a user ID 702, and a classification code 703. Note that No. 701 is given to facilitate the description described below. Therefore, the access history 700 does not have to have No701.

  Next, the level setting unit 13 selects one of the access histories 700 (see FIG. 7) divided for each user (S505). The level setting unit 13 may select any of the access histories 174 of the users who have not performed the processing from S506 described below, for example, the user IDs may be selected in order from the smallest.

  Then, the level setting unit 13 uses the transition order flow to determine whether or not the classification code 703 of the record first appears in order from the top (No: “1”) record of the access history 700 of the selected user. (S506). The transition order flow represents the transition of the user's behavior (growth order), and specifically represents the classification code 703 stored in the access history 700 in the order of appearance. Each time a new classification code appears, the level setting unit 13 adds the classification code and creates a transition order flow. Therefore, the level setting unit 13 determines whether or not the classification code 703 first appears depending on whether or not the classification code 703 to be processed is already set in the transition order flow.

  FIG. 8 is a diagram schematically illustrating a transition order flow for each user. The level setting unit 13 sets the classification code 703 in order from the first area to the fifth area each time a new classification code appears. In the present embodiment, since the classification reference information has five classification codes, the transition order flow has a maximum of five areas. If there are fewer than five classification codes, only the classification codes that have appeared are set in order from the first area.

  When the classification code 703 of the record to be processed is not set in the transition order flow, that is, when the classification code appears first (S506: YES), the level setting unit 13 sets the classification code in the transition order. The flow is set in an unset area (a blank area where no classification code is set) (S507). Then, the level setting unit 13 determines whether or not all records in the access history 700 have been processed (whether or not the record to be processed is the last record) (S508). If processing has not been performed for all records (S508: NO), the level setting unit returns to S506 to process the next record.

  On the other hand, when the classification code 703 of the record to be processed has already been set in the transition order flow, that is, when the classification code does not appear first (S506: NO), the level setting unit displays the classification code 703 in the transition order. The processing of S508 described above is performed without setting the flow.

  If all records in the access history 700 have been processed (S508: YES), the transition flow of the processing target user is completed at this point. Next, the level processing unit 13 determines whether or not the processing of the access history 700 for all users has been performed (S509). If there is an access history 700 of a user who has not been processed yet (S509: NO), the process returns to S505, and the processing from S506 to S509 is performed on the access history 700 of the unprocessed user. And the level setting part 13 produces the transition order flow of all the users.

  Here, the processing from S506 to S509 will be specifically described by taking the user access history 700 shown in FIG. 7 as an example. First, the level setting unit 13 reads the record of No “1” in FIG. 7 and determines whether or not the classification code “C001” of the record appears for the first time (S506). At this time, since no classification code is set in the transition order flow, the level setting unit 13 determines that the classification code “C001” first appears (S506: YES). Then, the level setting unit 13 sets the classification code “C001” in the first area of the transition flow (S507). Since the record of No “1” is not the record of the last line (S508: NO), the level setting unit subsequently processes the record of No “2”.

  The classification code of No “2” is “C001”. The level setting unit 13 refers to the transition order flow and determines that the classification code “C001” does not appear first because “C001” is already stored in the first area of the transition order flow. (S506: NO). Accordingly, the record of No “3” is processed without adding the classification code “C001” to the transition order flow.

  The classification code of the record with No “3” is “C002”. The level setting unit refers to the transition order flow, and since “C002” is not stored in the transition order flow at this time, it is determined that the classification code “C002” first appears (S506: YES). Therefore, the level setting unit 13 sets the classification code “C002” in the second area of the transition order flow.

  As described above, when the level setting unit 13 performs the processing from S506 to S508 up to the last line (No “11”) of the access history 700, the flow of user transition shown in FIG. 7 is “C001 → C002 → C005 → C003 → C004 ”(see FIG. 8).

  As described above, after creating a transition order flow for all users (S509: YES), the level setting unit 13 totals the transition order flows for all users (S510). That is, the level setting unit 13 integrates (superimposes) the transition order flows of all users, and creates an overall transition order flow.

  FIG. 9 schematically shows the transition sequence flow of the entire user. In the example shown in FIG. 9, first, there are 764 users who have performed the action with the classification code “C001”, of which 358 users have performed the action of “C002”, and the remaining 406 users. Indicates that only the action of “C001” was performed. Further, there are 358 users who have performed the action of “C002”, of which 185 perform the action of “C005” next, another 49 perform the action of “C003” next, and the remaining 124 This indicates that the person has only performed the actions “C001” and “C002”.

  Then, the level setting unit 13 specifies the transition order that is most likely to occur from the transition order flow of the entire user (that is, the largest number of users), and sets the level (S511). In the transition sequence flow of the entire user shown in FIG. 9, the level setting unit 13 specifies “C001 → C002 → C005 → C003 → C004”. Then, the level setting unit 13 sets a level for each classification code in the identified transition order. That is, the level setting unit 13 sets C001 to level 1, C002 to level 2, C005 to level 3, C003 to level 4, and C004 to level 5. Then, the level setting unit 13 registers the level set for each classification code in the classification reference information 300 stored in the storage unit 18 (S512).

  Note that FIG. 10 is obtained by setting the level 304 to the classification reference information shown in FIG. In this way, by setting the level for each classification code, for example, a user who has only taken actions of the classification codes of “Level 1” and “Level 2” has the upper level “Level 3”. Suppose you take action with a classification code. In this case, the administrator of the website can determine that the user has grown from “level 2” to “level 3”.

  Next, the flow of analysis processing of the service providing server 1 in this embodiment will be described. The analysis process is executed at a predetermined timing (cycle) after the execution of the level setting process (see FIG. 5). For example, it is possible to analyze a log file for one month from the first day of every month and perform it once a month periodically. Alternatively, a log file for one week from Monday to Sunday may be analyzed and periodically executed once a week.

  FIG. 11 is a process flow diagram of the analysis process of the service providing server 1.

  First, the log extraction unit 12 extracts an access history 400 (see FIG. 4) for a predetermined period from the log file at a predetermined timing, and stores it in the storage unit 18 (S1101). Then, the analysis unit 14 converts the access history 400 stored in the storage unit 18 into address codes 402 of the access history 400 into classification codes (see FIG. 6) in the same manner as S503 in the level setting process (see FIG. 5) (see FIG. 6). S1102). In this case, the analysis unit 14 leaves the record of the address information 402 that is not registered in the classification reference information 300 as it is without being converted into a classification code, unlike the process of S503. And the analysis part 14 divides | segments the access log | history 600 after converting into a classification code for every user similarly to S504 of FIG. 5 (S1103).

  Next, the analysis unit 14 reads the classification criteria information 300 (see FIG. 10) after level setting stored in the storage unit 18 (S1104). And the analysis part 14 sets a classification code to all the users using classification standard information 300 and access history 700 for every user, and classifies a user (S1105). That is, the analysis unit 14 reads out each classification code of the access history 700 for each user, and specifies the highest level classification code among them. Then, the analysis unit 14 sets the highest level classification code as the classification code of the user.

  For example, when the user access history 700 for a predetermined period is FIG. 7 and the classification criterion information 300 is FIG. 10, the classification code of the highest level (level 5) is “C004”. Therefore, the analysis unit 14 searches for “C004” in the access history 700. Since the classification code of the record of No “11” is “C004”, the analysis unit 14 sets the classification code of the user to “C004” of level 5. The analysis unit 14 sets the classification codes for all users in this way.

  Next, the analysis unit 14 calculates the evaluation value of the user belonging to each classification code (S1106). The evaluation value is a value indicating how much the user belonging to each classification code is taking an action that meets the expectations of the administrator, and the evaluation items received from the administrator terminal 2 in the level setting process (FIG. 5: S501) Is calculated based on The evaluation item in this embodiment is to increase the number of accesses of the entire website. For this reason, the evaluation value is a value with which the number of accesses of the entire website can be grasped. Therefore, the evaluation value of the present embodiment is obtained by averaging the number of accesses to the entire Web site of each user belonging to each classification code. That is, for each classification code, the analysis unit 14 calculates the number of records of each access history 700 of the user belonging to the classification code, and uses the average value as the evaluation value. When the administrator evaluates to increase only the number of accesses to a specific Web page, a value that can grasp the number of accesses to the specific Web page is set as an evaluation value.

  Next, the analysis unit 14 calculates an evaluation value for the entire user (S1107). In the present embodiment, the evaluation value for the entire user is the number of accesses of the entire website of all users. Therefore, the analysis unit 14 calculates the sum total of the product of the number of users belonging to each classification code and the evaluation value. Then, the analysis unit 14 stores the result of the processing from S1105 to S1107 in the storage unit 18.

  FIG. 12 is an example of an analysis result table showing the results of the processing from S1105 to S1107. As shown in the figure, this analysis result table has a classification code 1201, a classification name 1202, a level 1203, the number of users 1204, and an evaluation value 1205. The evaluation value 1205 includes an evaluation value 1205a for each classification code 1201 and an evaluation value 1205b for the entire user. The evaluation value 1205b for the entire user is the total sum of the number of users 1204 for each classification court 1201 multiplied by the evaluation value 1205. Specifically, 725 × 10 + 562 × 68 + 81 × 136 + 15 × 287 + 164 × 95 = 76367. The number of users 1204 represents the number of users classified into any of the classification codes in S1105.

  Next, the display processing unit 15 outputs the analysis result to the administrator terminal 2 using the communication processing unit 17 (S1108). Then, the output unit 22 of the administrator terminal 2 outputs the analysis result to the output device 905 using the communication processing unit.

  FIGS. 13 and 14 show examples of output screens output to the output device 905 of the administrator terminal 2. First, the output screen shown in FIG. 13 shows the evaluation value for each level [Table 1], the change of the evaluation value for the entire user as a graph [Table 2], and the change in the number of constituent members for each level. [Table 3].

  In [Table 1], a part of the analysis result table shown in FIG. 12 is displayed. [Table 1] has a level 1301, a classification name 1302, and an evaluation value 1303 as illustrated. The administrator of the Web site can grasp the transition (growth order) of each action of the user at the time when the analysis process is performed, based on [Table 1] displayed on the output device 905 of the administrator terminal 2. . In the example shown, the user “accesses the home page”, then “views the bulletin board”, then “answers the questionnaire”, then “writes on the bulletin board”, and then “sets the bulletin board”, The administrator can grasp that the behavior pattern is performed. As a result, regarding the next action of the user group at level 2 (browsing the bulletin board), it is understood that the “questionnaire answer” (level 3) is more likely to change than the “writing on the bulletin board” (level 4). Therefore, the administrator considers increasing the number of accesses to the entire Web site by taking some measure so that the user group classified as level 2 can answer a questionnaire first.

  In addition, by referring to the evaluation value for each level, the number of accesses to the website of the user group of each level can be evaluated. For example, in the illustrated example, the number of accesses to the website of the user group who performs level 4 “writing on the bulletin board” is the highest. Thereby, the administrator examines a measure for growing the users of level 1 to 3 to level 4 in order to increase the number of accesses of the entire website.

  [Graph 1] represents the transition of the evaluation value of the entire user. The evaluation value for the entire user is calculated in S1107 of FIG. 11 and stored in the storage unit 18 in a format such as the analysis result table 1200 shown in FIG. The analysis result table 1200 is created every time analysis processing is performed at a predetermined timing, and is accumulated in the storage unit 18. The display processing unit 15 extracts the evaluation value of the entire user of the analysis result table 1200 stored in the storage unit 18 for a predetermined period from the time of the current analysis processing, and displays it on a graph in time series. Thereby, the administrator can grasp | ascertain how the evaluation value of the whole user is changing. In the example shown in [Graph 1], it can be seen that the number of accesses to the entire Web site is gradually increasing.

  [Graph 2] represents the transition of the number of users constituting each level. The display processing unit 15 extracts the analysis result table 1200 for a predetermined period stored in the storage unit 18 as in [Graph 1], and graphs the transition of the number of users belonging to each level (each classification code) in time series. To display. The display processing unit 15 may display the transition of the composition ratio of each level in chronological order instead of the number of users configuring each level.

  As a result, the manager can grasp what level of the person is performing what level of action. For example, at the present time or in the past, a graph is output that has a large number of users who perform level 1 (access to the homepage) but a small number of users of level 3 (questionnaire responses) and level 4 (write to the bulletin board). Suppose. In this case, the administrator considers a measure for improving the bulletin board and the Web page of the questionnaire so as to increase the answer to the questionnaire and the writing on the bulletin board. In addition, after taking any improvement measures, the administrator can evaluate and confirm whether or not the effect of the implemented improvement measures is as expected by looking at this graph.

  The output screen shown in FIG. 14 is a graph showing the level transition (action history) of each user. The vertical axis of this graph displays the classification name corresponding to the level, and the horizontal axis displays the time when the analysis processing was performed. For example, the display processing unit 15 can output an action history for each user by storing not only the number of users 1204 but also a user ID belonging to the classification code in the analysis result table 1200 shown in FIG. Note that the administrator causes the output device 905 to display an action history of a user ID desired to be displayed by inputting a user ID from a user selection screen (not shown).

  This graph allows the administrator to grasp what level each individual user is currently in, and what level the past has been in the past, thereby making it possible to make a detailed strategy for each user.

  The embodiment of the present invention has been described above.

In this embodiment, the service providing server 1 classifies user actions using the access history, and sets a level based on a general growth order (transition order) of each action. Then, at a predetermined timing, the behavior of the user is analyzed based on the level, and analysis data necessary for community evaluation of the user accessing the website is output. Further, the degree of contribution (for example, the number of accesses) of users belonging to each category is calculated as an evaluation value. From these, grasp the situation of the community, such as the user composition at each current level and past level, the evaluation value (contribution) of users belonging to each level, and the general growth order (transition order) between actions of each classification Can do. Therefore, the administrator can evaluate an appropriate community and can plan measures for improving the Web site. In addition, after taking some improvement measures, the administrator can evaluate and confirm whether or not the effect of the implemented improvement measures has resulted in an expected result. In addition, the administrator can grasp what classification should be promoted so that the user belonging to which classification will be led to which classification in the future Note that the present invention is not limited to the above embodiment, Many modifications are possible within the scope of the gist.

  For example, in the present embodiment, increasing the number of accesses to the entire website is a criterion for community evaluation. However, the present invention is not limited to this. For example, the number of accesses to a product sales service provided by a website may be used as a criterion for community evaluation. In the product sales service, the purchase price when the user actually purchases the product may be used as a standard for community evaluation. Further, in the present embodiment, input from the administrator terminal 2 is accepted in the evaluation item level setting process (FIG. 5: S501). However, the input process from the administrator terminal 2 may not be performed by setting the evaluation items in advance.

  In the present embodiment, the level setting unit 13 of the service providing server 1 sets a level for each classification code. However, the input of the level of each classification code may be received from the input receiving unit 21 of the administrator terminal 3.

  Moreover, in this embodiment, the analysis process (refer FIG. 11) of the service provision server 1 is implemented at the predetermined timing. However, for example, it may be performed at an arbitrary timing such as an instruction from the administrator terminal 2. In this case, the log extraction unit 12 extracts an access history for a predetermined period such as the past 10 days from the present from the log file. Alternatively, the analysis process may be performed in real time every time the user accesses the website.

  In addition, the classification standard information (see FIG. 3) of the present embodiment has one address information 303 corresponding to one classification code name 301. However, a plurality of address information 303 corresponding to one classification code 301 may exist. If it is complicated to set a plurality of address information one by one, for example, by setting a URL directory, all Web pages located below the directory may be set as corresponding address information.

  In this embodiment, an access history including a log file or address information acquired by a general function of the Web server is extracted, and each user's behavior is analyzed. However, the present invention is not limited to this, and various actions of users are accumulated in a log file by devising screen transitions or using technologies such as cgi, thereby analyzing user actions. It is good.

  In addition, by setting a classification code or level in the user's behavior, the administrator measures and analyzes how much the service contributes to the entire website when a new service is added on the website. it can. In addition, the administrator can use the measured / analyzed data in order to examine measures for improving the contribution degree of the entire website.

FIG. 1 is a schematic diagram of a community evaluation support system to which an embodiment of the present invention is applied. FIG. 2 is a diagram illustrating a hardware configuration example of the service providing server, the administrator terminal, and the user terminal. FIG. 3 is a diagram illustrating a table configuration example of the classification criterion information. FIG. 4 is a diagram showing an example of a table structure of access history. FIG. 5 is a process flow diagram of the level setting process of the service providing server. FIG. 6 is a diagram illustrating a table configuration example of an access history after code conversion. FIG. 7 is a diagram illustrating a table configuration example of an access history (after code conversion) for each user. FIG. 8 is a schematic diagram of a transition order flow for each user. FIG. 9 is a schematic diagram of a transition sequence flow for the entire user. FIG. 10 is a diagram illustrating a table configuration example of classification reference information after level setting. FIG. 11 is a processing flowchart of analysis processing of the service providing server. FIG. 12 is a diagram illustrating a table configuration example of the analysis result table. FIG. 13 is an example of an output screen (overall analysis result) of the administrator terminal. FIG. 14 is an example of an output screen (analysis result for each user) of the administrator terminal.

Explanation of symbols

1: service providing server, 2: administrator terminal, 3 _{1 to} 3 _n : user terminal, 9: network, 11: input reception unit, 12: log extraction unit, 13: level setting unit, 14: analysis unit, 15: Display processing unit, 16: service providing unit, 17: communication processing unit, 18: storage unit, 21: input receiving unit, 22: output unit, 23: communication processing unit, 300 classification criteria information, 400: access history, 1200: Analysis result table

Claims (19)

A community evaluation support method for supporting evaluation of a community existing on a network using a computer system,
The storage device of the computer system stores log information that records the access status of a plurality of clients that access the computer system,
The computing device of the computer system includes:
An input receiving step for receiving input of classification criteria information storing community classification information and at least one classification identifier for classifying actions on the network of the plurality of clients from an external system;
An extraction step of extracting access histories of the plurality of clients in a predetermined period from the log information;
A classification step of dividing the access histories of the plurality of clients for each client, and classifying each client into one of the classification identifiers of the classification criterion information based on the access history of each client;
An analysis step of calculating an evaluation value of a client belonging to the classification identifier for each classification identifier based on the evaluation information and an access history of each client belonging to the classification identifier, and analyzing the community for each classification identifier; ,
An output step of outputting the result analyzed in the analysis step to an output device. A community evaluation support method, comprising: A community evaluation support method according to claim 1,
The extraction step, the classification step, the analysis step, and the output step are performed at a predetermined timing. A community evaluation support method according to claim 1,
The extraction step, the classification step, the analysis step, and the output step are performed based on an instruction received from the external system. A community evaluation support method according to claim 1,
The extraction step, the classification step, the analysis step, and the output step are performed each time any one of the plurality of clients accesses the computer system. A community evaluation support method according to claim 1,
The computing device of the computer system includes:
A community evaluation support method further comprising a level setting step of setting a level for each of the classification identifiers based on an access history of each of the clients. A community evaluation support method according to claim 5,
In the community setting support method, the level setting step comprises analyzing the order of behavior of each client from the access history of each of the clients, and setting the level based on the order of behavior transition analyzed. A community evaluation support method according to claim 6,
The classification step reads the access history of each of the clients, converts each of the client actions stored in the access history into one of the classification identifiers, and classifies the highest level among the converted classification identifiers A community evaluation support method characterized by setting an identifier as a classification identifier of the client. A community evaluation support method according to claim 1,
The evaluation value of a client belonging to the classification identifier is an average access count of the client belonging to the classification identifier to the computer system. A community evaluation support method according to claim 1,
The evaluation value of the client belonging to the classification identifier is an average access count of the client belonging to the classification identifier to a predetermined Web page of the computer system. A community evaluation support method according to claim 1,
The analysis step calculates, for each classification identifier, a value obtained by multiplying the number of clients belonging to the identifier by the evaluation value of the identifier, and further, an overall evaluation that is a sum of values calculated for each classification identifier A community evaluation support method characterized by calculating a value and analyzing the community. A community evaluation support method according to claim 10,
The analysis step stores the overall evaluation value in the storage device;
In the output step, the overall evaluation value stored in the storage device is output to the output device in time series. A community evaluation support method according to claim 1,
The output step calculates the composition ratio of the classification identifier from the number of clients of each of the classification identifiers classified in the classification step, and outputs it to the output device. A community evaluation support method according to claim 12,
In the classification step, each client is classified into one of the classification identifiers of the classification criterion information, and the number of clients belonging to each of the classification identifiers is stored in the storage device.
The output step outputs the composition ratio of the classification identifier to the output device in time series based on the number of clients belonging to each of the classification identifiers stored in the storage device. Method. A community evaluation support method according to claim 1,
The classification step classifies each client into one of the classification identifiers of the classification reference information, stores the classification identifier for each client in the storage device,
The output step outputs the classification identifier set for each of the clients stored in the storage means to the output device in time series for each client. A community evaluation support device that supports evaluation of a community existing on a network,
Storage means for storing log information recording the access status of each of a plurality of clients accessing the community evaluation support device;
Input acceptance means for accepting input of classification criteria information having at least one classification identifier for classifying actions on the network of the plurality of clients and community evaluation information from an external system;
Extraction means for extracting access histories of the plurality of clients for a predetermined period from the log information;
Classifying means for dividing the access history of the plurality of clients for each client, and classifying each client into one of the classification identifiers of the classification criteria information based on the access history of each client;
An analysis means for calculating an evaluation value of a client belonging to the classification identifier for each classification identifier based on the evaluation information and an access history of each client belonging to the classification identifier, and analyzing the community for each classification identifier; ,
A community evaluation support apparatus comprising: output means for outputting the result of analysis by the analysis means. The community evaluation support device according to claim 15,
A community evaluation support device further comprising level setting means for setting a level for each of the classification identifications based on an access history of each of the clients. The community evaluation support device according to claim 15,
The analysis means calculates, for each classification identifier, a value obtained by multiplying the number of clients belonging to the identifier and the evaluation value of the identifier, and further, an overall evaluation that is a sum of values calculated for each classification identifier Community evaluation support device characterized by calculating values and analyzing communities. The community evaluation support device according to claim 15,
The analysis unit stores the analysis result of the community in the storage unit,
The output means outputs the community analysis results stored in the storage means to the output means in time series. A computer-readable community evaluation support program for supporting evaluation of communities existing on a network,
The storage device of the computer system stores log information that records the access status of a plurality of clients accessing the computer system,
The program is stored in a computing device of the computer system.
An input receiving step for receiving input of classification criteria information storing community classification information and at least one classification identifier for classifying actions on the network of the plurality of clients from an external system;
An extraction step of extracting access histories of the plurality of clients in a predetermined period from the log information;
A classification step of dividing the access histories of the plurality of clients for each client, and classifying each client into one of the classification identifiers of the classification criterion information based on the access history of each client;
An analysis step of calculating an evaluation value of a client belonging to the classification identifier for each classification identifier based on the evaluation information and an access history of each client belonging to the classification identifier, and analyzing the community for each classification identifier; ,
An output step of outputting the result of analysis in the analysis step to an output device.

Images