JP2009289068A - Characteristic estimating device, characteristic estimating system, characteristic estimating method, and characteristic estimating program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To precisely estimate a preference characteristic of a user in a communication network. <P>SOLUTION: In a teacher user characteristic specification unit 16b, an inquiry question transmitting and receiving control part 16a specifies a characteristic of a teacher user on the basis of inquiry question data from a teacher user terminal 30. An access history collection unit 16c collects access histories of a teacher user, an action data generation unit 16d generates action data of the teacher user from the history of the teacher user by using a history/action correspondence table of a history/action correspondence table storage unit 15d, a characteristic estimation model generation unit 16e generates a characteristic estimation model from the characteristic and the action data of the teacher user, an action data generation unit 16d generates action data of an estimated user from an access history of the estimated user by using the history/action correspondence table, and an estimated user characteristic estimation unit 16f estimates the characteristic of the estimated user from the action data of the estimated user and from the model. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a characteristic estimation device, a characteristic estimation system, a characteristic estimation method, and a characteristic estimation program for estimating user preference characteristics.

  Conventionally, the amount of information provided in a communication network such as the Internet has been on an explosive increase, and the variety of services available and products that can be purchased on the communication network such as the Internet has also been increasing. . For this reason, it has become difficult for a user who uses a communication network to actively search for information, services, and products intended for the user, and provides information, services, and products according to user preference characteristics. Technology needs to be realized.

  In order to provide information, services, and products according to the user's preference characteristics, the user's preference characteristics can be estimated numerically or expressed in text (for example, “SF favorite” etc.). It will be necessary. Currently, the following two methods are known as methods for objectively estimating user preference characteristics.

  The first method is a method that asks users about their preference characteristics directly or indirectly by conducting a questionnaire, etc., and the second method is the user's access to information, services, products, etc. This is a technique for estimating preference characteristics by analyzing a history.

  However, since the first method puts a burden on the user to input in a questionnaire or the like, the second method passively estimates the user's preference characteristics without imposing a load on the user. The importance is increasing.

  Here, in the second method, for example, a preference characteristic is estimated using, as parameters, an access history for the user's product information, a user's purchase history, or an access frequency for the user's information.

  In addition, as a classification example of preference characteristics, there is a classification based on innovator theory (see, for example, Non-Patent Document 1). Innovator theory is a theory about the spread of innovation (technological innovation) advocated by Rogers in 1962. In Innovator theory, Rogers is based on the user's characteristics (innovation) of novelty and topicality of products. There are 5 groups.

  That is, innovator theory is based on the user's “innovation” and changes the user's preference characteristics into innovators (Innovators), early adopters (Early Adopters), and Early Majority (Early Majority). ), Late Majority (Late Majority), and Laggards (Laggards).

  Innovators are adventurous users who are willing to adopt innovations, early adopters are users who are sensitive to trends and collect information on their own to decide whether to adopt innovations, Early Majority is a comparison Thought-minded, users who adopt innovation faster than average, Late Majority is relatively skeptical, users who make the same innovation choice after seeing the majority of the surroundings experiment, Lagard is the most conservative, uninterested in trends and trends and does not adopt until innovation is a tradition.

  Here, in the innovator theory, the innovator focuses on the novelty of innovation itself and focuses on the benefits of innovation while the early adopter focuses on the new benefits of innovation and gives it to other users. Is an important opinion leader for innovation to spread.

  However, in communication networks such as the Internet, the innovation cycle is fast, so we believe that innovators that access new product information as soon as possible have a function as an opinion leader, and the access history of users estimated as innovators Based on this, a method for recommending information to users other than innovators is known (see Non-Patent Document 1).

  In addition, a method is known in which user preference characteristics estimated from input information such as a questionnaire are updated using search and browsing based on purchase and purchase history of merchandise (see, for example, Patent Document 1). Furthermore, there is also known a method for estimating characteristics such as information acceptability and information transmission ability of a user from a user usage history specialized in a P2P (Peer to Peer) network (see, for example, Patent Document 2).

Masayuki Ishikawa, Takeshi Morita, Noriaki Izumi, Takahira Yamaguchi: Information Promotion Engine Based on Access History Analysis of Shared Content, 21st Annual Conference of Japanese Society for Artificial Intelligence, 3G8-2, 2007 Republished Patent W01 / 067319 JP 2004-362208 A

  By the way, the above-described conventional technique has a problem that it is difficult to accurately estimate the preference characteristic of the user in the communication network.

  In fact, if the correlation between the user group and the access time is calculated from the access history to the product of a user whose user group is known using the access history to the product on the Internet, no correlation is seen. Therefore, even when recommending information to users other than innovators based on the access history of users estimated as innovators because the access time to new product information is early, the estimation accuracy of users who are innovators is low, Low recommendation rate for users.

  In addition, when a user's preference characteristics are updated using search or browsing based on purchase and a purchase history of products, the updated preference characteristics are estimated from user input information such as a questionnaire. The characteristic cannot be estimated without imposing a load on the user.

  Also, when estimating the user's information acceptability and information transmission ability from the usage history of the P2P network, the type of usage history is limited to the history in the P2P network. Cannot be estimated.

  Accordingly, the present invention has been made to solve the above-described problems of the prior art, and a characteristic estimation device, a characteristic estimation system, and a characteristic estimation that can accurately estimate a user's preference characteristics in a communication network. It is an object to provide a method and a characteristic estimation program.

  In order to solve the above-described problems and achieve the object, this apparatus is a characteristic estimation apparatus that estimates characteristics of a user who uses a communication network, and is extracted from access information constituting an access history and the access information. History behavior correspondence information holding means for holding history behavior correspondence information in which user behavior information is associated, access history of characteristic specifying user who is a user whose characteristics are specified in advance, and history behavior correspondence information holding means Based on the history behavior correspondence information held by the user, the behavior information of the characteristic specific user is extracted, and the characteristic information is extracted from the extracted characteristic information of the characteristic specific user and the characteristic specified by the characteristic specific user in advance. Characteristic estimation model generation means for generating a characteristic estimation model for estimating the Based on the history history and the history behavior correspondence information held by the history behavior correspondence information holding means, the behavior information of the estimated user is extracted, the behavior information of the estimated user extracted, and the characteristic estimation model generation And a characteristic estimation unit that estimates the characteristic of the estimated user using the characteristic estimation model generated by the unit.

  The system includes a server that provides a service via a communication network, or a connection device connected to the server, and a user terminal that uses the service provided from the server, A characteristic estimation system for estimating characteristics when a user uses the communication network, wherein the server or the connection device includes access information constituting an access history and user behavior information extracted from the access information. The history behavior correspondence information holding means for holding the history behavior correspondence information in association with each other, the access history of the characteristic specifying user who is the user whose characteristics are specified in advance, and the history held by the history behavior correspondence information holding means Based on the action correspondence information, action information of the characteristic specifying user is extracted, and the extracted characteristic specifying user is extracted. Characteristic estimation model generating means for generating a characteristic estimation model for estimating the characteristic from the behavior information of the characteristic specific user and the characteristic specified in advance by the characteristic specifying user, and the terminal Generated by the historical behavior correspondence information held by the historical behavior correspondence information holding unit and the characteristic estimation model generation unit from the server or the connection device when the estimated user terminal is a terminal of an estimation user who is a user The behavior information of the estimated user is extracted based on the acquisition means for acquiring the characteristic estimation model, the access history of the estimated user, and the history behavior correspondence information acquired by the acquisition means. Using the behavior information of the estimated user and the characteristic estimation model acquired by the acquisition means. A characteristic estimating means for estimating a may be a requirement that with a.

  In addition, this method is a characteristic estimation method for estimating characteristics of a user who uses a communication network, and is a history in which access information constituting an access history is associated with user behavior information extracted from the access information Based on the historical action correspondence information holding step for holding action correspondence information, the access history of the characteristic specifying user who is the user whose characteristic is specified in advance, and the historical action correspondence information held in the historical action correspondence information holding step. Then, the behavior information of the characteristic specific user is extracted, and a characteristic estimation model for estimating the characteristic is generated from the extracted behavior information of the characteristic specific user and the characteristic specified in advance of the characteristic specific user. A characteristic estimation model generation step, an access history of an estimated user who is a target of the characteristic estimation, and the history behavior Based on the history behavior correspondence information held in the response information holding step, the behavior information of the estimated user is extracted, the behavior information of the estimated user extracted, and the characteristic generated by the characteristic estimation model generation step And a characteristic estimation step of estimating the characteristic of the estimated user using an estimation model.

  Further, this program is a characteristic estimation program for causing a computer to execute a characteristic estimation method for estimating characteristics of a user who uses a communication network. The program includes access information constituting an access history, and user's information extracted from the access information. The historical behavior correspondence information holding procedure for holding historical behavior correspondence information associated with behavior information, the access history of a characteristic specifying user who is a user whose characteristics are specified in advance, and the historical behavior correspondence information holding procedure Based on the history behavior correspondence information, the behavior information of the characteristic identification user is extracted, and the characteristic is estimated from the extracted behavior information of the characteristic identification user and the characteristic specified in advance of the characteristic identification user. A characteristic estimation model generation procedure for generating a characteristic estimation model for the user, and a user who is a target of the characteristic estimation Based on the access history of the user and the history behavior correspondence information held by the history behavior correspondence information holding procedure, the behavior information of the estimated user is extracted, the behavior information of the estimated user extracted, and the characteristic estimation It is a requirement that the computer execute a characteristic estimation procedure for estimating the characteristic of the estimated user using the characteristic estimation model generated by the model generation procedure.

  According to the disclosed apparatus, system, method, and program, it is possible to accurately estimate a preference characteristic of a user in a communication network. In addition, it is possible to estimate the preference characteristics without placing an input burden on the user's questionnaire, it is possible to estimate the preference characteristics corresponding to the case where the user's preference characteristics change dynamically, Corresponding to a large-scale environment, it is possible to estimate the preference characteristics of many users.

  Exemplary embodiments of a characteristic estimation device, a characteristic estimation system, a characteristic estimation method, and a characteristic estimation program according to the present invention will be described below in detail with reference to the accompanying drawings.

  In the first embodiment described below, a case where the function of the characteristic estimation apparatus according to the present invention is installed in a “server” that provides a service via a communication network will be described.

  In the following, a user who uses a communication network such as the Internet via a “server” and whose characteristics (preference characteristics in the Internet) are to be estimated is described as an “estimated user”. Further, a terminal device owned by the “estimated user” and requesting a service via the network such as the Internet from the “server” based on the input information of the “estimated user” is described as “estimated user terminal”.

  First, the overall configuration of the characteristic estimation system including the server in the first embodiment will be described. FIG. 1 is a diagram for explaining the overall configuration of a characteristic estimation system including a server according to the first embodiment.

  As shown in FIG. 1, the server 10 according to the first embodiment is connected to the estimated user terminal 20 and is summarized to estimate the characteristics of the estimated user who owns the estimated user terminal 20. Here, the server 10 in the first embodiment is further connected to a teacher user terminal 30 that is a terminal of a “teacher user”.

  For example, as illustrated in FIG. 1, the server 10 according to the first embodiment is connected to a plurality of teacher user terminals 30 such as a teacher user terminal 30 of the teacher user 1 and a teacher user terminal 30 of the teacher user 2. Here, the “teacher user” is a user whose characteristics in the communication network are specified in advance based on information acquired from the teacher user terminal 30 before estimating the characteristics of the estimated user. The “teacher user” corresponds to the “characteristic specifying user” described in the claims.

  Further, as the estimated user terminal 20 and the teacher user terminal 30, for example, a personal computer, a PDA, or a device having a communication function such as a mobile phone or a PHS can be cited, and a web browser or the like dedicated for using the network. Provide an application.

  The server 10 according to the first embodiment is characterized mainly in that it is possible to accurately estimate the preference characteristics of the estimated user in a communication network such as the Internet by using the characteristics specified in advance by the teacher user. .

  This main feature will be described with reference to FIGS. FIG. 2 is a block diagram showing the configuration of the teacher user terminal, FIG. 3 is a block diagram showing the configuration of the server in the first embodiment, and FIG. 4 is a diagram for explaining the teacher user characteristic specifying unit. 5 is a diagram for explaining the access history management storage unit, FIG. 6 is a diagram for explaining the history / behavior correspondence table storage unit, and FIG. 7 is a teacher by the behavior data generation unit. FIG. 8 is a diagram for explaining generation of user behavior data, FIG. 8 is a diagram for explaining generation of behavior data of the estimated user by the behavior data generation unit, and FIG. 9 is a diagram for explaining the characteristic estimation model generation unit. FIG.

  First, the teacher user terminal 30 will be described. As shown in FIG. 2, the teacher user terminal 30 connected to the server 10 includes an input unit 31, an output unit 32, a communication unit 33, an input / output control I / F unit 34, a storage unit 35, and processing. Part 36. In addition to the functional blocks shown in FIG. 2, the storage unit 35 and the processing unit 36 include various functional blocks for executing functions as a personal computer, for example.

  The input unit 31 inputs various information and is configured to include a keyboard, a mouse, and the like, and particularly as closely related to the present invention, an answer of the teacher user to the questionnaire question data received from the server 10 is, for example, , Input from the keyboard.

  The output unit 32 outputs various types of information, and is configured to include a monitor and a speaker. In particular, as closely related to the present invention, the questionnaire question data received from the server 10 is displayed on the monitor screen. To do.

  The communication unit 33 performs communication of data transmitted and received through a network. In particular, as closely related to the present invention, the communication unit 33 receives questionnaire question data from the server 10 and stores the received questionnaire question data, which will be described later. It transfers to 35a, or the questionnaire questionnaire data already input stored in the questionnaire memory | storage part 35a mentioned later are transmitted to the server 10. FIG.

  The input / output control I / F unit 34 controls data transfer between the input unit 31, the output unit 32, the communication unit 33, the storage unit 35, and the processing unit 36.

  The storage unit 35 stores data used for various types of processing by the processing unit 36 and various types of processing results by the processing unit 36, and particularly as closely related to the present invention, as shown in FIG. 2, a questionnaire storage unit 35a Is provided.

  The questionnaire storage unit 35a stores the questionnaire question data received by the communication unit 33 from the server 10 and transferred from the input / output control I / F unit 34, and has been input by the questionnaire input processing unit 36a described later. Memorize questionnaire question data.

  The processing unit 36 executes various processes based on the data transferred from the input / output control I / F unit 34, and particularly as closely related to the present invention, as shown in FIG. 2, a questionnaire input processing unit 36a Is provided.

  The questionnaire input processing unit 36a controls to display the questionnaire question data stored in the questionnaire storage unit 35a on a monitor provided in the output unit 32. Further, the questionnaire input processing unit 36a accepts the teacher user's answer with reference to the questionnaire question data displayed on the monitor provided in the output unit 32 via the input unit 31, and the input questionnaire question data into which the answer has been input. Is stored in the questionnaire storage unit 35a.

  In addition, the questionnaire input processing unit 36a, when receiving an input completion notification of questionnaire question data from the teacher user via the input unit 31, inputs the input questionnaire question data stored in the questionnaire storage unit 35a. The data is transferred to the communication unit 33 via the / F unit 34, and the communication unit 33 transmits the transferred questionnaire question data that has been transferred to the server 10.

  Next, the server 10 will be described. As shown in FIG. 3, the server 10 connected to the teacher user terminal 30 and the estimated user terminal 20 includes an input unit 11, an output unit 12, a communication unit 13, an input / output control I / F unit 14, and a storage. Unit 15 and processing unit 16. The storage unit 15 and the processing unit 16 include various functional blocks for executing functions as a server in addition to the functional blocks shown in FIG.

  The input unit 11 inputs various types of information and is configured to include a keyboard, a mouse, and the like. Particularly, as closely related to the present invention, the input unit 11 is stored in a questionnaire management storage unit 15a described later and transmitted to the teacher user terminal 30. Survey questionnaire data, a history / behavior correspondence table stored in the history / behavior correspondence table storage unit 15d described later, a time for estimating the characteristics of the estimated user (hereinafter referred to as a characteristic estimation time), etc. Accept and input from 10 managers.

  The output unit 12 outputs various types of information and is configured to include a monitor and a speaker. Particularly, as closely related to the present invention, the input questionnaire question data received from the teacher user terminal 30 and a teacher described later The characteristics of the teacher user specified by the user characteristic specifying unit 16b, the estimated user characteristic estimated by the estimated user characteristic estimation unit 16f described later, and the like are displayed on the monitor screen.

  The communication unit 13 performs communication of data transmitted and received through the network, and particularly as closely related to the present invention, transmits the questionnaire question data stored in the questionnaire management storage unit 15a described later to the teacher user terminal 30. The received questionnaire question data is received from the teacher user terminal 30 and the received questionnaire question data is transferred to the questionnaire management storage unit 15a described later, or based on the control of the access history collection unit 16c described later. The access history of the teacher user terminal 30 and the estimated user terminal 20 is received, and the received access history is transferred to the access history management storage unit 15c described later.

  The input / output control I / F unit 14 controls data transfer between the input unit 11, the output unit 12, the communication unit 13, the storage unit 15, and the processing unit 16.

  The storage unit 15 stores data used for various types of processing by the processing unit 16 and various types of processing results by the processing unit 16, and particularly as closely related to the present invention, as shown in FIG. 15a, teacher user characteristic storage unit 15b, access history management storage unit 15c, history / behavior correspondence table storage unit 15d, behavior data storage unit 15e, characteristic estimation model storage unit 15f, and estimated user characteristic storage unit 15g With.

  The questionnaire management storage unit 15a stores questionnaire question data transmitted to the teacher user terminal 30 based on the control of the questionnaire transmission / reception control unit 16a described later, and the input questionnaire question data received from the teacher user terminal 30 is described later. Based on the control of the questionnaire transmission / reception control unit 16a, the identifier (user ID) assigned to the teacher user is stored in association with it.

  The teacher user characteristic storage unit 15b stores the characteristics of the teacher user specified by the teacher user characteristic specifying unit 16b described later. The access history management storage unit 15c stores the teacher user and the estimation collected by the access history collection unit 16c described later. The user access history is stored, the behavior data storage unit 15e stores the behavior data of the teacher user and the estimation user generated by the behavior data generation unit 16d described later, and the characteristic estimation model storage unit 15f stores the characteristic estimation described later. The characteristic estimation model generated by the model generation unit 16e is stored, and the estimated user characteristic storage unit 15g stores the characteristics of the estimated user estimated by the estimated user characteristic estimation unit 16f described later.

  The history / behavior correspondence table storage unit 15d stores a history / behavior correspondence table in which access information constituting the access history is associated with user behavior information extracted from the access information. The history / action correspondence table corresponds to “history action correspondence information” recited in the claims, and the history / action correspondence table storage unit 15d also corresponds to “history action correspondence information holding means”.

  The contents stored in each functional block provided in the storage unit 15 will be described in detail later.

  The processing unit 16 executes various processes based on the data transferred from the input / output control I / F unit 14, and particularly as closely related to the present invention, as shown in FIG. 3, a questionnaire transmission / reception control unit 16a. A teacher user characteristic specifying unit 16b, an access history collecting unit 16c, a behavior data generating unit 16d, a characteristic estimation model generating unit 16e, and an estimated user characteristic estimating unit 16f. Here, the behavior data generation unit 16d and the characteristic estimation model generation unit 16e correspond to the “characteristic estimation model generation unit” described in the claims, and the behavior data generation unit 16d and the estimated user characteristic estimation unit 16f are the same. This corresponds to “characteristic estimation means”.

  The questionnaire transmission / reception control unit 16 a controls the questionnaire question data stored in the questionnaire management storage unit 15 a to be transmitted to the teacher user terminal 30 via the communication unit 13 based on an instruction from the administrator of the server 10. The questionnaire transmission / reception control unit 16a receives the input questionnaire question data transmitted from the teacher user terminal 30 via the communication unit 13, and the received questionnaire question data is allocated to each teacher user. Control is performed so as to be associated with the user ID and stored in the questionnaire management storage unit 15a.

  The teacher user characteristic specifying unit 16b specifies a characteristic for each teacher user based on the already input questionnaire question data stored in the questionnaire management storage unit 15a.

  For example, the teacher user characteristic specifying unit 16b analyzes the already input questionnaire question data, and specifies characteristics based on “innovation” or the like for each teacher user as shown in FIG.

  That is, the teacher user characteristic specifying unit 16b analyzes the questionnaire data that has already been input, and as shown in FIG. 4, each of the teacher users is innovator, early adapter, early majority, late majority, or lagard. Identify and classify into characteristic groups based on “innovation”. Further, the teacher user characteristic specifying unit 16b assigns each teacher user a characteristic group based on other than “innovation”, for example, a characteristic group based on demographic information such as sex, age, occupation, etc., as shown in FIG. Categorizes and classifies them into characteristic groups based on psychographic information other than innovation, such as luxury-oriented. As a result, as shown in FIG. 4, for each teacher user (user IDs: ku0001, ku0002,...), Identification based on innovation, demographic information, psychographic information other than innovation, and the like is specified. .

  And the teacher user characteristic specific | specification part 16b matches the characteristic for every specified teacher user with the user ID allocated for every teacher user, and stores it in the teacher user characteristic memory | storage part 15b.

  Returning to FIG. 3, the access history collection unit 16 c collects the access history of the teacher user from the teacher user terminal 30 based on the characteristic estimation timing set by the administrator of the server 10, and the estimated user terminal 20 receives the estimated user's access history. Collect access history. Then, the access history collection unit 16c stores the collected access history in the access history management storage unit 15c in association with the user ID assigned for each teacher user and estimated user.

  For example, as shown in FIG. 5, the access history management storage unit 15 c stores “URL: http://www.example.com/search?key=xxx” accessed by the teacher user 1 and “date / time: 2008 / The access history such as “3/28 12:30” is stored in association with “ku0001” which is the user ID assigned to the teacher user 1. In addition, although the present Example demonstrated the case where the object which collects the access history of a teacher user and an estimation user is the teacher user terminal 30 and the estimation user terminal 20, this invention is not limited to this, The server 10 that provides services to the teacher user and the estimated user, or a server other than the server 10 may be a target for collecting access history of the teacher user and the estimated user.

  Returning to FIG. 3, the behavior data generation unit 16 d determines that the teacher user and the access history management storage unit 15 c store the access history stored in the access history management storage unit 15 c and the history and behavior correspondence table stored in the history and behavior correspondence table storage unit 15 d. Estimate user behavior data is generated. The action data corresponds to “behavior information” described in the claims.

  Here, the history / action correspondence table stored in the history / action correspondence table storage unit 15d will be described again in detail with reference to FIG. As described above, the history / behavior correspondence table is information in which access information constituting the access history is associated with the user behavior information extracted from the access information, and specifically, the access history is constituted. This is information in which an action ID is associated with each type of URL as access information, and furthermore, an action group in which each action ID is roughly classified by type is associated.

  For example, as shown in FIG. 6, the history / action correspondence table storage unit 15d indicates that the type of URL accessed by the user is that the user accesses the portal site “http://www.example.com/” If it is “http://www.example.com/search?key= <keyword>” indicating that a keyword has been entered on the site and the search has been executed, it corresponds to “action ID: A01”. A history / behavior correspondence table is stored in which “action ID: A01” is an action ID corresponding to “ID: A” roughly classified into “behavior group: search”.

  In addition, as shown in FIG. 6, the history / behavior correspondence table storage unit 15d displays “behavior ID: B01” corresponding to “ID: B” roughly classified as “behavior group: browsing” as a specific portal. It stores in association with each type of URL indicating that a page on the site has been browsed, and “action ID: C01” corresponding to “ID: C” broadly classified as “action group: purchase” is specified. A history / action correspondence table stored in association with each type of URL indicating that a product has been purchased using a portal site is stored. The history / behavior correspondence table storage unit 15d has “ID: D” “aggressiveness for information collection” as “ID: D”, “ID: E” “acceptability for recommendations from others”, as shown in FIG. “ID: F” as “sensitivity to topicality (popularity)”, “ID: G” as “sensitivity to new products”, “ID: H” as “sensitivity to evaluation”, “ID: I” History / Behavior Correspondence of “Action ID” for each URL type for “Action Group”, which is roughly divided into “Aggression for Information Transmission”, “ID: J” and “Price Sensitivity” Remember the table.

  Returning to FIG. 3, the behavior data generation unit 16 d firstly includes the access history of the teacher user collected by the access history collection unit 16 c and stored in the access history management storage unit 15 c and the history / behavior correspondence table storage unit in the characteristic estimation timing. Based on the history / action correspondence table stored in 15d, action data of the teacher user is generated. Note that the behavior data generation unit 16d generates behavior data using an amount that is determined in advance by an administrator or an access history for a total period.

  Specifically, the behavior data generation unit 16d reads “URL: http://www.example.com/search?key=xxx” in the access history of the teacher user terminal 30 (user ID: ku0001) of the teacher user 1. 5 ”is specified as“ action ID: A01 ”from the history / action correspondence table (see FIG. 6) stored in the history / action correspondence table storage unit 15d.

  Then, the behavior data generation unit 16d calculates a frequency for each behavior ID by identifying a behavior ID for each piece of “URL” information constituting the access history of the teacher user. Furthermore, the behavior data generation unit 16d sums the frequencies of all the behavior IDs that constitute the same behavior group, and calculates “total frequency of“ ID: A ”, total frequency of“ ID: B ”, and“ ID: C ”. Action data for each teacher user consisting of the total frequency... Is generated, and the result is stored in the action data storage unit 15e.

  For example, as illustrated in FIG. 7, the behavior data generation unit 16 d analyzes the “URL” of the access history of the teacher user whose user ID is “ku0001”, and is roughly classified into “activity group: search”. : “A” total frequency is “1”, “action group: browsing” is broadly divided into “ID: B” total frequency is “0”, and “action group: purchasing” is broadly divided into “ID” : The action data of the teacher user 1 with the total frequency of “C” being “0” is generated. Although not shown in FIG. 7, the behavior data generation unit 16 d analyzes the “URL” of the access history of the teacher user and calculates the total frequency for the above “ID: D” to “ID: J”. Then, action data for each teacher user regarding “ID: D” to “ID: J” is also generated.

  Note that the behavior data generation process of the estimated user by the behavior data generation unit 16d will be described later.

  Returning to FIG. 3, the characteristic estimation model generation unit 16e obtains characteristics from the behavior data for each teacher user stored in the behavior data storage unit 15e and the characteristic groups for each teacher user stored in the teacher user characteristic storage unit 15b. A characteristic estimation model for estimation is generated, and the result is stored in the characteristic estimation model storage unit 15f.

  Specifically, the characteristic estimation model generation unit 16e uses a characteristic group based on “innovation” as an objective variable, and uses a total frequency for each ID of the action group as an explanatory variable, to calculate a characteristic estimation model for estimating the characteristic, It is generated using a pattern recognition technique such as linear discriminant analysis.

  For example, the characteristic estimation model generation unit 16e sets the characteristic group based on “innovation” as the objective variable “Y”, sets the total frequency of “ID: A” as the explanatory variable “X1,” and sets the total frequency of “ID: B”. Is the explanatory variable “X2”, the total frequency of “ID: C” is the explanatory variable “X3”, the total frequency of “ID: D” is the explanatory variable “X4”, and the total frequency of “ID: E” is described The variable “X5”, the total frequency of “ID: F” is the explanatory variable “X6”, the total frequency of “ID: G” is the explanatory variable “X7”, and the total frequency of “ID: H” is the explanatory variable “X”. X8 ”, the total frequency of“ ID: I ”is the explanatory variable“ X9 ”, the total frequency of“ ID: J ”is the explanatory variable“ X10 ”, and the discriminant function“ Y = aX1 + bX2 + cX3 + dX4 + eX5 + fX6 + gX7 + hX8 + iX9 + jX10 ”is set. The optimal parameters “a, b, c, d, e, f, g, h, i, j” by linear discriminant analysis from the “characteristics based on“ innovation ”and the total frequency for each action group” for each user. Is calculated to generate a characteristic estimation model.

  In addition, the characteristic estimation model generation unit 16e also performs the following for each characteristic group based on demographic information for each teacher user specified by the teacher user characteristic specifying unit 16b and for each characteristic group based on psychographic information other than innovation. Similarly, a characteristic estimation model with the objective variable is generated.

  Here, when the characteristic estimation model generated by the characteristic estimation model generation unit 16e is stored in the characteristic estimation model storage unit 15f, the behavior data generation unit 16d collects the access history by the access history collection unit 16c at the characteristic estimation timing. Based on the access history for each estimated user stored in the management storage unit 15c and the history / behavior correspondence table stored in the history / behavior correspondence table storage unit 15d, action data for each estimated user is generated. Then, the behavior data generation unit 16d stores the generated behavior data for each estimated user in the behavior data storage unit 15e. Note that the behavior data generation unit 16d generates behavior data using an amount that is determined in advance by an administrator or an access history for a total period.

  For example, as illustrated in FIG. 8, the behavior data generation unit 16 d stores the history / behavior correspondence stored in the history / behavior correspondence table storage unit 15 d of the access history “URL” of the estimated user 1 whose user ID is “su0001”. By analyzing using the table (see FIG. 6), the total frequency of “ID: A”, which is roughly classified as “behavior group: search”, is “1”, and is broadly classified as “behavior group: browsing”. The behavior data of the estimated user 1 is generated by setting the total frequency of “ID: B” to “1” and the total frequency of “ID: C” roughly classified as “behavior group: purchase” as “0”. Although not shown in FIG. 8, the behavior data generation unit 16d analyzes the “URL” of the access history of the estimated user and calculates the total frequency for the above “ID: D” to “ID: J”. And the action data for every presumed user regarding “ID: D” to “ID: J” are also generated.

  Returning to FIG. 3, the estimated user characteristic estimation unit 16 f uses the characteristic estimation model stored in the characteristic estimation model storage unit 15 f and the behavior data for each estimated user stored in the behavior data storage unit 15 e. The characteristic is estimated, and the result is stored in the estimated user characteristic storage unit 15g.

  Specifically, the estimated user characteristic estimation unit 16f calculates the estimated user based on the value calculated by substituting the total frequency for each action group in the estimated user's action data into the discriminant function generated as the characteristic estimation model. Estimate the characteristics of

  For example, as shown in FIG. 9, the estimated user characteristic estimation unit 16 f uses a characteristic estimation model in which the preference characteristic based on “innovation” of the estimation user 1 whose user ID is “su0001” is the objective variable “innovation”. As an “innovator”. In addition, the estimated user characteristic estimation unit 16f estimates the characteristic of the estimated user 1 by using a characteristic estimation model that uses a characteristic group based on demographic information or a characteristic group based on psychographic information other than innovation as an objective variable. Similarly, the estimated user characteristic estimation unit 16f also estimates characteristics of the estimated user 2 or the like whose user ID is “su0002”. Note that the characteristics of the identified teacher user and the estimated estimated user are not limited to the characteristics described above, and the present invention can be applied to various types of preference characteristics. It is.

  Various processes are automatically or manually performed based on the characteristics estimated by the estimated user characteristic estimation unit 16f. For example, among the services provided on the communication network such as the Internet, automatically by the server 10 or manually by the administrator of the server 10, the preferences of the estimated users are matched based on the estimated characteristics. Services are extracted and the extracted services are recommended to each estimated user. Alternatively, a marketing survey based on the estimated characteristics is performed by a person in charge of product development.

  In addition, when the characteristic estimation timing described above is set based on a predetermined condition by the administrator of the server 10 and becomes the characteristic estimation timing, “collection of access history of teacher user and estimated user by the access history collection unit 16c” “Generation of teacher user's behavior data by behavior data generation unit 16d and generation of characteristic estimation model by property estimation model generation unit 16e” and “Generation of behavior data of estimated user by behavior data generation unit 16d and estimated user characteristics” A series of characteristic estimation processing including “estimation user characteristic estimation by the estimation unit 16 f” is performed.

  Here, the characteristic estimation process may be executed in one characteristic estimation timing set by the administrator of the server 10, but the present invention is not limited to this. That is, in order to cope with dynamic user characteristic changes, multiple characteristic estimation timings are set, and by repeatedly performing characteristic estimation processing for each of the multiple characteristic estimation timings, the estimated user characteristics are updated as needed It may be the case.

  For example, when a plurality of times designated by the administrator of the server 10 are set as “a plurality of characteristic estimation timings”, or at predetermined time intervals designated by the administrator of the server 10 (for example, every month) The time may be set as “a plurality of characteristic estimation timings”. Alternatively, the access history collection unit 16c monitors the access history of the estimated user terminal 20, and each time when the amount of increase in the accumulated amount in the access history becomes a predetermined value is referred to as “a plurality of characteristic estimation times”. It may be set.

  Next, the procedure of processing by the teacher user terminal 30 will be described with reference to FIG. FIG. 10 is a flowchart for explaining a processing procedure by the teacher user terminal.

  As shown in FIG. 10, when the teacher user terminal 30 receives the questionnaire question data from the server 10 (Yes in step S1001), the teacher user terminal 30 stores the received questionnaire question data in the questionnaire storage unit 35a, and the questionnaire input processing unit 36a Control is performed so that the questionnaire question data stored in the storage unit 35a is displayed on a monitor provided in the output unit 32 (step S1002).

  When the questionnaire input processing unit 36a determines that the input of the questionnaire question data has been completed by receiving an input completion notification of the questionnaire question data from the teacher user via the input unit 31 (Yes in step S1003), the input is completed. The questionnaire question data is transmitted to the server 10 (step S1004), and the process is terminated.

  Next, a procedure of teacher user characteristic specifying processing by the server 10 will be described with reference to FIG. FIG. 11 is a flowchart for explaining the procedure of the teacher user characteristic specifying process by the server.

  As shown in FIG. 11, when the server 10 receives an instruction to send questionnaire question data from the administrator (Yes in step S1101), the questionnaire transmission / reception control unit 16a transmits the questionnaire question data stored in the questionnaire management storage unit 15a to the teacher. Control is performed to transmit to the user terminal 30 via the communication unit 13 (step S1102).

  Then, when the questionnaire transmission / reception control unit 16a receives the inputted questionnaire question data from the teacher user terminal 30 (step S1103), the received questionnaire question data is stored in the questionnaire management storage unit 15a, and the teacher user characteristic specifying unit 16b. Specifies the characteristics for each teacher user based on the already input questionnaire question data stored in the questionnaire management storage unit 15a (step S1104), and ends the process.

  Next, the procedure of the estimation user characteristic estimation process by the server 10 will be described with reference to FIG. FIG. 12 is a flowchart for explaining the procedure of the estimation user characteristic estimation process by the server.

  As shown in FIG. 12, when the server 10 comes to the characteristic estimation time set by the administrator (Yes at Step S1201), the access history collection unit 16c collects the access history of the teacher user and the estimated user (Step S1202). The behavior data generation unit 16d uses the history / behavior correspondence table stored in the history / behavior correspondence table storage unit 15d from the access history of the teacher user collected by the access history collection unit 16c to obtain the behavior data of the teacher user. Generate (step S1203).

  Then, the characteristic estimation model generation unit 16e generates a characteristic estimation model from the characteristics of the teacher user specified by the teacher user characteristic specification unit 16b and the behavior data of the teacher user generated by the behavior data generation unit 16d (step S1204).

  After that, the behavior data generation unit 16d uses the history / behavior correspondence table stored in the history / behavior correspondence table storage unit 15d from the access history of the estimated user collected by the access history collection unit 16c. Data is generated (step S1205).

  Subsequently, the estimated user property estimation unit 16f estimates the estimated user property from the estimated user behavior data generated by the behavior data generation unit 16d and the property estimation model generated by the property estimation model generation unit 16e ( In step S1206), the process ends. It should be noted that the characteristic estimation process shown in FIG. 12 can be performed in response to dynamic user characteristic changes by repeatedly executing the characteristic estimation process every time the characteristic estimation timing comes.

  As described above, in the first embodiment, the questionnaire transmission / reception control unit 16a receives the input questionnaire question data from the teacher user terminal 30, and the teacher user characteristic specifying unit 16b is based on the input questionnaire question data. Identify the characteristics for each teacher user. Then, when the characteristic estimation timing comes, the access history collection unit 16c collects the access history of the teacher user and the estimated user from the teacher user terminal 30 and the estimated user terminal 20, and the behavior data generation unit 16d determines from the access history of the teacher user. The history / behavior correspondence table storage unit 15d stores the teacher / user behavior data using the history / behavior correspondence table, and the characteristic estimation model generation unit 16e includes the teacher user characteristic identification unit 16b. Is used as an objective variable, and a characteristic estimation model is generated using the teacher user's action data generated by the action data generation unit 16d as an explanatory variable.

  Then, the behavior data generation unit 16d uses the history / behavior correspondence table stored in the history / behavior correspondence table storage unit 15d from the access history of the estimated user collected by the access history collection unit 16c. The estimated user characteristic estimation unit 16f estimates the estimated user characteristic from the estimated user behavior data generated by the behavior data generation unit 16d and the characteristic estimation model generated by the characteristic estimation model generation unit 16e. .

  From such a thing, since the characteristic of the user to be estimated can be estimated based on the model generated from the access history of the user in which the preference characteristic in the communication network such as the Internet is actually specified, as described above, It is possible to accurately estimate the preference characteristics of the estimated user in a communication network such as the Internet.

  Further, since only the access history of the estimated user is required when estimating the estimated user characteristics, it is possible to easily estimate the preference characteristics of the estimated user in a communication network such as the Internet.

  In addition, it is possible to estimate the preference characteristics without imposing the burden of inputting questionnaire question data on the estimated user. Thereby, the estimation user can passively use a service optimized based on his / her preference characteristics.

  Further, since it is possible to update and estimate the characteristics of the estimated user simply by repeatedly collecting the access history of the estimated user, it is possible to follow the dynamically changing preference characteristics.

  In addition, although the present Example demonstrated the case where the estimation user terminal 20 was connected to the server 10, this invention is not limited to this, If an access history of an estimation user can be collected The estimated user terminal 20 may not be connected to the server 10.

  In the present embodiment, the case where the access history is collected and the estimated user's characteristic is estimated at the characteristic estimation timing has been described. However, the present invention is not limited to this, and the access history is collected in advance. It may be a case where the characteristics of the estimated user are estimated at the characteristic estimation timing. In addition, the time when a new estimated user is set may be used as the characteristic estimation time, and the characteristic of the new estimated user may be estimated using the already generated characteristic estimation model.

  In the above-described first embodiment, the case where the server 10 as the characteristic estimation apparatus executes the characteristic estimation process collectively has been described. However, in the second embodiment, the server 10 and the estimated user terminal 20 perform the characteristic estimation process. The case of sharing and executing will be described with reference to FIGS. 13 and 14. FIG. 13 is a block diagram illustrating a configuration of the server in the second embodiment, and FIG. 14 is a block diagram illustrating a configuration of the estimation user terminal in the second embodiment.

  In addition, the characteristic estimation system comprised including the server 10 in Example 2 becomes the structure similar to the characteristic estimation system comprised including the server 10 in Example 1 shown in FIG.

  First, the server 10 according to the second embodiment will be described. As illustrated in FIG. 13, the estimated user characteristic estimation unit 16 f is deleted from the server 10 according to the second embodiment as compared with the server 10 according to the first embodiment illustrated in FIG. 3.

  Here, the server 10 in the second embodiment performs the questionnaire question data transmission / reception process by the questionnaire transmission / reception control unit 16a and the teacher user characteristic specifying process by the teacher user characteristic specifying unit 16b in the same manner as the server 10 in the first example. Execute.

  However, in the access history collection process by the access history collection unit 16c, the server 10 according to the second embodiment collects only the access history of the teacher user at the time of characteristic estimation, and the behavior data generation process by the behavior data generation unit 16d Only the behavior data of the teacher user is generated using the history / behavior correspondence table stored in the behavior correspondence table storage unit 15d.

  And the server 10 in Example 2 performs the generation process of the characteristic estimation model by the characteristic estimation model production | generation part 16e similarly to the server 10 in Example 1. FIG.

  The contents stored in each functional block of the storage unit 15 according to the second embodiment are that the access history management storage unit 15c stores only the access history of the teacher user collected by the access history collection unit 16c, and the behavior data storage unit 15e. Is the same as that of the first embodiment except that only the behavior data of the teacher user generated by the behavior data generation unit 16d is stored.

  Then, the estimation user terminal 20 in Example 2 is demonstrated. The estimated user terminal 20 in the first embodiment is a general device (for example, a personal computer) as a device having a communication function, but the estimated user terminal 20 in the second embodiment is a user (estimated user) that owns the estimated user terminal 20. ) To estimate the characteristics of the apparatus.

  That is, as shown in FIG. 14, the estimated user terminal 20 in the second embodiment is connected to the server 10 and includes an input unit 21, an output unit 22, a communication unit 23, an input / output control I / F unit 24, A storage unit 25 and a processing unit 26 are provided. Note that the storage unit 25 and the processing unit 26 include various functional blocks for executing functions as a personal computer, for example, in addition to the functional blocks shown in FIG.

  The input unit 21 inputs various types of information and includes a keyboard and a mouse. The output unit 22 outputs various types of information and includes a monitor and a speaker.

  The communication unit 23 performs communication of data transmitted and received through the network. In particular, as closely related to the present invention, the communication unit 23 receives the history / action correspondence table stored in the history / action correspondence table storage unit 15d from the server 10. Then, it is transferred to the history / behavior correspondence table storage unit 25b described later, or the characteristic estimation model generated by the characteristic estimation model generation unit 16e is received from the server 10 and transferred to the characteristic estimation model storage unit 25d described later. To do. The communication unit 23 corresponds to “acquiring means” recited in the claims.

  Here, the communication unit 23 receives the history / action correspondence table from the server 10 in advance before executing the characteristic estimation process. In addition, the communication unit 23 receives the characteristic estimation model from the server 10 every time the characteristic estimation model is generated by the characteristic estimation model generation unit 16e.

  The input / output control I / F unit 24 controls data transfer between the input unit 21, the output unit 22, the communication unit 23, the storage unit 25, and the processing unit 26.

  The storage unit 25 stores data used for various types of processing by the processing unit 26 and various types of processing results by the processing unit 26, and particularly as closely related to the present invention, as shown in FIG. Unit 25a, history / behavior correspondence table storage unit 25b, behavior data storage unit 25c, characteristic estimation model storage unit 25d, and estimated user characteristic storage unit 25e.

  The access history management storage unit 25a stores the access history collected by the access history collection unit 26a described later.

  The history / behavior correspondence table storage unit 25b stores the same information as the history / behavior correspondence table received by the communication unit 23, that is, the history / behavior correspondence table stored in the history / behavior correspondence table storage unit 15d of the server 10. .

  The behavior data storage unit 25c stores behavior data generated by a behavior data generation unit 26b described later.

  The characteristic estimation model storage unit 25d stores the characteristic estimation model received by the communication unit 23, that is, the characteristic estimation model generated by the characteristic estimation model generation unit 16e of the server 10 and stored in the characteristic estimation model storage unit 15f.

  The estimated user characteristic storage unit 25e stores the estimated user characteristic estimated by the estimated user characteristic estimation unit 26c described later.

  The processing unit 26 executes various processes based on the data transferred from the input / output control I / F unit 24, and particularly as closely related to the present invention, as shown in FIG. 3, the access history collecting unit 26a And an action data generation unit 26b and an estimated user characteristic estimation unit 26c.

  The access history collection unit 26a collects the access history of the estimated user who owns the estimated user terminal 20 provided on the basis of the characteristic estimation timing set by the administrator of the server 10, and the result is the access history management storage unit. 25a.

  Based on the access history stored in the access history management storage unit 25a and the history / behavior correspondence table stored in the history / behavior correspondence table storage unit 25b, the behavior data generation unit 26b determines the estimated user terminal 20 provided therein. The behavior data of the estimated user who owns it is generated, and the result is stored in the behavior data storage unit 25c.

  The estimated user characteristic estimation unit 26c uses the characteristic estimation model stored in the characteristic estimation model storage unit 25d and the behavior data stored in the behavior data storage unit 25c to estimate the user who owns the estimation user terminal 20 provided therein. The characteristic is estimated, and the result is stored in the estimated user characteristic storage unit 25e.

  That is, in the first embodiment, the server 10 collectively estimates the characteristics of each estimated user. However, in the second embodiment, the estimated user terminal 20 of each estimated user owns the characteristics of the estimated user who owns the estimated user. Is estimated.

  Various processes are performed automatically or manually based on the characteristics estimated by the estimated user characteristic estimation unit 26c. For example, automatically by the estimated user terminal 20 or the server 10 that has acquired the estimated characteristics from the estimated user terminal 20 or manually by the administrator of the server 10 that has acquired the estimated characteristics from the estimated user terminal 20 From the services provided on a communication network such as the Internet, a service that matches the preference for the estimated user is extracted, and the extracted service is recommended to the estimated user. Alternatively, a marketing survey based on the estimated characteristics is performed by a person in charge of product development.

  In addition, as described in the first embodiment, the characteristic estimation timing is set with a plurality of characteristic estimation timings in order to cope with dynamic user characteristic changes, and the characteristic estimation process is repeated for each of the plurality of characteristic estimation timings. By doing so, the characteristics of the estimated user may be updated and estimated at any time.

  In the present embodiment, the history / behavior correspondence table is received from the server 10 in advance and stored in the history / behavior correspondence table storage unit 25b, and then the behavior data is generated. It is not limited, and the behavior data may be generated from the access history of the estimated user while inquiring the information of the history / behavior correspondence table to the server 10.

  Next, the procedure of the characteristic model generation process by the server 10 in the second embodiment will be described with reference to FIG. FIG. 15 is a flowchart for explaining the procedure of the characteristic model generation process by the server in the second embodiment. The procedure of the process of the teacher user terminal 30 in the second embodiment and the procedure of the characteristic specification process of the teacher user of the server 10 in the second embodiment are the same as those described in the first embodiment with reference to FIGS. 10 and 11. Since it is the same, description is abbreviate | omitted.

  As shown in FIG. 15, when the server 10 reaches the characteristic estimation timing set by the administrator (Yes at step S1501), the access history collection unit 16c collects the access history of the teacher user, and the behavior data generation unit 16d Generates behavior data of the teacher user from the access history using the history / behavior correspondence table stored in the history / behavior correspondence table storage unit 15d (step S1502).

  Then, the characteristic estimation model generation unit 16e generates a characteristic estimation model from the characteristics of the teacher user specified by the teacher user characteristic specification unit 16b and the behavior data of the teacher user generated by the behavior data generation unit 16d (step S1503). Here, the characteristic estimation model generation unit 16e stores the generated characteristic estimation model in the characteristic estimation model storage unit 15f.

  After that, the communication unit 13 transmits the characteristic estimation model stored in the characteristic estimation model storage unit 15f to the estimation user terminal 20 (step S1504), and ends the process.

  Then, the procedure of the process by the presumed user terminal 20 in Example 2 is demonstrated using FIG. FIG. 16 is a flowchart for explaining a process performed by the estimation user terminal according to the second embodiment.

  As illustrated in FIG. 16, when the estimation user terminal 20 according to the second embodiment receives the characteristic estimation model from the server 10 (Yes in step S1601), the access history collection unit 26a owns the estimation user terminal 20 with which the estimation user terminal 20 is provided. By collecting the access history of the estimated user, the behavior data generation unit 26b generates behavior data from the access history using the history / behavior correspondence table stored in the history / behavior correspondence table storage unit 25b (step S1602). .

  Then, the estimated user characteristic estimation unit 26c estimates the characteristics of the estimated user from the behavior data generated by the behavior data generation unit 26b and the characteristic estimation model stored in the characteristic estimation model storage unit 25d (step S1603). The process ends. In this embodiment, the case where the access history is collected and the estimated user's characteristics are estimated at the characteristic estimation timing has been described. However, the present invention is not limited to this, and the access history is collected in advance. It may be a case where the characteristics of the estimated user are estimated at the characteristic estimation timing. In addition, the time when the user who owns the user is set as a new estimated user is also used as the property estimation time, and the estimation user terminal 20 acquires the already generated property estimation model and estimates the property of the user who owns the user. It may be the case.

  As described above, in the second embodiment, since the estimated user terminal 20 estimates the estimated user characteristics, it is possible to perform distributed access history analysis and characteristic estimation on the terminal side, and also supports a large-scale environment. It becomes possible to do.

  In the first embodiment, the server 10 collectively executes the characteristic estimation process. In the second embodiment, the server 10 executes the characteristic estimation model generation process. However, the present invention is not limited to this. It is not limited, and the connection device connected to the server 10 may execute the processing of the server 10 described in the first and second embodiments.

  In addition, among the processes described in this embodiment, all or part of the processes described as being performed automatically can be performed manually, or the processes described as being performed manually can be performed. All or a part can be automatically performed by a known method. For example, the characteristic estimation timing may always be manually set by the administrator of the server 10. In addition, the processing procedure, control procedure, specific name, and information including various data and parameters shown in the above-described document and drawings can be arbitrarily changed unless otherwise specified.

  Further, each component of each illustrated apparatus is functionally conceptual, and does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution / integration of each device is not limited to that shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various loads or usage conditions. Can be integrated and configured. For example, the behavior data generation unit 16d may be distributed to a processing unit for generating behavior data for a teacher user and a processing unit for generating behavior data for an estimated user. Further, all or any part of each processing function performed in each device may be realized by a CPU and a program analyzed and executed by the CPU, or may be realized as hardware by wired logic.

  The characteristic estimation method described in this embodiment can be realized by executing a program prepared in advance on a computer such as a personal computer or a workstation. This program can be distributed via a network such as the Internet. The program can also be executed by being recorded on a computer-readable recording medium such as a hard disk, a flexible disk (FD), a CD-ROM, an MO, and a DVD and being read from the recording medium by the computer.

  As described above, the characteristic estimation device, the characteristic estimation system, the characteristic estimation method, and the characteristic estimation program according to the present invention are useful when estimating the characteristics of a user who uses a communication network. Suitable for accurately estimating preference characteristics.

It is a figure for demonstrating the whole structure of the characteristic estimation system containing the server in Example 1. FIG. It is a block diagram which shows the structure of a teacher user terminal. 1 is a block diagram illustrating a configuration of a server in Embodiment 1. FIG. It is a figure for demonstrating a teacher user characteristic specific | specification part. It is a figure for demonstrating an access history management memory | storage part. It is a figure for demonstrating a log | history and action corresponding | compatible table memory | storage part. It is a figure for demonstrating the teacher user's action data generation | occurrence | production by an action data generation part. It is a figure for demonstrating the estimation user's action data generation by an action data generation part. It is a figure for demonstrating a characteristic estimation model production | generation part. It is a flowchart for demonstrating the procedure of the process by a teacher user terminal. It is a flowchart for demonstrating the procedure of the characteristic specification process of the teacher user by a server. It is a flowchart for demonstrating the procedure of the estimation user's characteristic estimation process by a server. It is a block diagram which shows the structure of the server in Example 2. FIG. It is a block diagram which shows the structure of the presumed user terminal in Example 2. FIG. 12 is a flowchart for explaining a procedure of a characteristic model generation process by a server in the second embodiment. It is a flowchart for demonstrating the process by the estimation user terminal in Example 2. FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Server 11 Input part 12 Output part 13 Communication part 14 Input / output control I / F part 15 Storage part 15a Questionnaire management storage part 15b Teacher user characteristic storage part 15c Access history management storage part 15d History / action correspondence table storage part 15e Action data Storage unit 15f characteristic estimation model storage unit 15g estimated user characteristic storage unit 16 processing unit 16a questionnaire transmission / reception control unit 16b teacher user characteristic identification unit 16c access history collection unit 16d behavior data generation unit 16e characteristic estimation model generation unit 16f estimation user characteristic estimation unit 20 Estimated user terminal 30 Teacher user terminal 31 Input unit 32 Output unit 33 Communication unit 34 Input / output control I / F unit 35 Storage unit 35a Questionnaire storage unit 36 Processing unit 36a Questionnaire input processing unit

Claims (6)

A characteristic estimation device for estimating characteristics of a user who uses a communication network,
History behavior correspondence information holding means for holding history behavior correspondence information in which access information constituting an access history is associated with user behavior information extracted from the access information;
Based on the access history of the characteristic specifying user who is the user whose characteristic is specified in advance and the history action correspondence information held by the history action correspondence information holding means, the action information of the characteristic specification user is extracted and extracted. Characteristic estimation model generation means for generating a characteristic estimation model for estimating the characteristic from the behavior information of the characteristic specific user and the characteristic specified in advance of the characteristic specific user;
Based on the access history of the estimated user who is the user whose characteristics are to be estimated and the history behavior correspondence information held by the history behavior correspondence information holding means, the behavior information of the estimation user is extracted and extracted Characteristic estimation means for estimating the characteristics of the estimated user using the behavior information of the estimated user and the characteristic estimation model generated by the characteristic estimation model generation means;
A characteristic estimation apparatus comprising:   The characteristic estimation apparatus according to claim 1, wherein the characteristic estimation unit updates and estimates the characteristic of the estimated user for each time based on a predetermined condition. A server that provides a service via a communication network, or a connection device connected to the server, and a user terminal that uses the service provided by the server. A characteristic estimation system for estimating characteristics when used,
The server or the connection device is
History behavior correspondence information holding means for holding history behavior correspondence information in which access information constituting an access history is associated with user behavior information extracted from the access information;
Based on the access history of the characteristic specifying user who is the user whose characteristic is specified in advance and the history action correspondence information held by the history action correspondence information holding means, the action information of the characteristic specification user is extracted and extracted. Characteristic estimation model generation means for generating a characteristic estimation model for estimating the characteristic from the behavior information of the characteristic specific user and the characteristic specified in advance of the characteristic specific user;
With
The terminal
In the case of an estimated user terminal that is a terminal of an estimated user who is a user whose characteristics are to be estimated,
An acquisition means for acquiring the historical behavior correspondence information held by the historical behavior correspondence information holding means and the characteristic estimation model generated by the characteristic estimation model generation means from the server or the connection device;
Based on the access history of the estimated user and the history behavior correspondence information acquired by the acquisition means, the behavior information of the estimated user is extracted, and the extracted behavior information of the estimated user is extracted by the acquisition means. Characteristic estimation means for estimating the characteristics of the estimated user using the acquired characteristic estimation model;
A characteristic estimation system comprising:   The characteristic estimation system according to claim 3, wherein the characteristic estimation unit updates and estimates the characteristic of the estimated user for each time based on a predetermined condition. A characteristic estimation method for estimating characteristics of a user who uses a communication network,
A history behavior correspondence information holding step for holding history behavior correspondence information in which access information constituting the access history is associated with user behavior information extracted from the access information;
Based on the access history of the characteristic identification user who is the user whose characteristic is specified in advance and the history behavior correspondence information held by the history behavior correspondence information holding step, the behavior information of the characteristic identification user is extracted and extracted. A characteristic estimation model generating step for generating a characteristic estimation model for estimating the characteristic from the behavior information of the characteristic specific user and the characteristic specified in advance of the characteristic specific user;
Based on the access history of the estimated user who is the user whose characteristic is to be estimated and the history behavior correspondence information held by the history behavior correspondence information holding step, the behavior information of the estimation user is extracted and extracted. A characteristic estimation step of estimating the characteristics of the estimated user using the behavior information of the estimated user and the characteristic estimation model generated by the characteristic estimation model generation step;
The characteristic estimation method characterized by including. A characteristic estimation program for causing a computer to execute a characteristic estimation method for estimating characteristics of a user who uses a communication network,
A history action correspondence information holding procedure for holding history action correspondence information in which access information constituting an access history is associated with user action information extracted from the access information;
Based on the access history of the characteristic identification user who is the user whose characteristic is specified in advance and the history behavior correspondence information held by the history behavior correspondence information holding procedure, the behavior information of the characteristic identification user is extracted and extracted A characteristic estimation model generating procedure for generating a characteristic estimation model for estimating the characteristic from the behavior information of the characteristic specific user and the characteristic specified in advance of the characteristic specific user;
Based on the access history of the estimated user who is the user whose characteristics are to be estimated and the history behavior correspondence information held by the history behavior correspondence information holding procedure, the behavior information of the estimation user is extracted and extracted. A characteristic estimation procedure for estimating the characteristics of the estimated user using the behavior information of the estimated user and the characteristic estimation model generated by the characteristic estimation model generation procedure;
A characteristic estimation program for causing a computer to execute.

Images