JP6697499B2 - Information processing apparatus, information processing method, and information processing program - Google Patents

Description

  The present invention relates to an information processing device, an information processing method, and an information processing program.

  In recent years, with the development of communication networks, various services have been provided by various service providers via the networks. Then, the user uses various devices via a network using a smart device such as a PC (Personal Computer), a smartphone, or a tablet terminal.

  In a service provided by a service provider, user information including user attributes and action history is acquired, and advertisements and other information according to the user are provided based on the user information. However, users may use different smart devices or create different accounts each time the service is used.

  In this case, the information of the same user is managed as a plurality of user information, and the attributes and action history of the same user are collected for each of the plurality of user information. Therefore, the service provider side may not be able to properly acquire the user information. Therefore, there has been proposed a technique of associating a plurality of user information as information of the same user based on the information acquired from the user's smart device (for example, see Patent Document 1).

JP, 2014-38420, A

  The degree of similarity of user information calculated when linking a plurality of user information is determined by a plurality of elements constituting the user information (for example, a user's name, address, preference, a plurality of elements included in an access log from a smart device). Etc.) between user information.

  However, among the plurality of elements that constitute the user information, there are elements that do not affect the similarity between the user information and elements that have a small effect. The processing load on the server becomes large. In addition, when a plurality of pieces of user information to be associated candidates are managed by different event persons, it may not be desirable to disclose all the elements that constitute the user information from the viewpoint of cost.

  The present application has been made in view of the above, and provides an information processing device, an information processing method, and an information processing program capable of efficiently performing a linking process of linking a plurality of user information. To aim.

  The information processing apparatus according to the present application compares a part of the first user information in the first user information group with a part of the second user information in the second user information group, and compares the first user information with the first user information. A determination unit that determines one or more first elements and one or more second elements of the second user information as comparison candidates, and stores information of the determined comparison candidates in a storage unit, and a storage unit that stores the information in the storage unit. A linking unit that links the first user information of the first user information group and the second user information of the second user information group based on the information of the compared candidates.

  According to one aspect of the embodiment, it is possible to provide an information processing device, an information processing method, and an information processing program that can efficiently perform a linking process of linking a plurality of user information.

FIG. 1 is an explanatory diagram of information processing according to the first embodiment. FIG. 2 is a diagram illustrating a configuration example of the information processing system according to the first embodiment. FIG. 3 is a diagram illustrating a configuration example of the information processing device according to the first embodiment. FIG. 4 is a diagram showing an example of a user information group table according to the first embodiment. FIG. 5 is a diagram showing an example of a user information group table according to the first embodiment. FIG. 6 is a diagram illustrating an example of the comparison candidate table according to the first embodiment. FIG. 7 is a diagram showing an example of the association information table according to the first embodiment. FIG. 8 is a flowchart showing an example of the flow of information processing in the information processing device according to the first embodiment. FIG. 9 is a flowchart showing the flow of the comparison candidate determination process shown in FIG. FIG. 10 is a flowchart showing the flow of the linking process shown in FIG. FIG. 11 is an explanatory diagram of information processing according to the second embodiment. FIG. 12 is a diagram illustrating a configuration example of the information processing device according to the second embodiment. FIG. 13 is a flowchart showing the flow of a comparison candidate determination process according to the second embodiment. FIG. 14 is a flowchart showing the flow of the linking process according to the second embodiment. FIG. 15 is a diagram illustrating an example of a hardware configuration of a computer that executes a program.

  Hereinafter, modes (hereinafter, referred to as “embodiments”) for implementing an information processing apparatus, an information processing method, and an information processing program according to the present application will be described in detail with reference to the drawings. Note that the information processing apparatus, the information processing method, and the information processing program according to the present application are not limited by this embodiment.

[1. First Embodiment]
[1.1. Information processing]
Information processing according to the first embodiment will be described with reference to FIG. 1. FIG. 1 is an explanatory diagram of information processing according to the embodiment, and in the first embodiment, information processing is executed by the information processing device 1.

  In the example illustrated in FIG. 1, in the information processing device 1, a plurality of operators C1 and C2 (hereinafter, sometimes collectively referred to as operators C) provide services as service providers, and users who use such services. It holds user information groups D1 and D2 (hereinafter, may be collectively referred to as user information group D) including user information that is information about the information.

  The service provided by the business operator C is an online service that provides services to users online. For example, shopping service, auction service, online banking service, video distribution service, music distribution service, news distribution service, home page provision. It is a service or a blog providing service. The service provided by the business operator C includes a service for providing a web page for selling and introducing the products provided by the business operator C. Further, the service provided by the business operator C may include an offline service.

  The user information group D is a data set including a plurality of user information and is stored in a storage unit such as a database managed by the business operator C. The information of the user who is the user of the service includes, for example, the demographic attribute of the user, the information of the smart device of the user, and the access history.

  In the example illustrated in FIG. 1, the business operator C1 includes a first user information group that includes user information (hereinafter, may be referred to as first user information) that is information regarding a user who is a service target of the business operator C1. Holds D1 information. The business operator C2 holds the information of the second user information group D2 including the user information (hereinafter sometimes referred to as the second user information) that is the information about the user to whom the business of the business operator C2 is provided. There is. Hereinafter, each of the information of the first user information group D1 and the information of the second user information group D2 may be simply referred to as the user information group D.

  The first user information group D1 and the second user information group D2 are information managed by different businesses C1 and C2, respectively. The first user information included in the first user information group D1 includes information of a plurality of elements, and similarly, the second user information included in the second user information group D2 includes information of a plurality of elements. It For example, the first user information and the second user information are, respectively, a user ID for identifying the user, information indicating the attribute of the user (hereinafter, referred to as attribute information), information indicating the action history of the user (hereinafter, action). (Described as history information) and the like.

  The first user information and the second user information are online services provided by the business operator C, such as information obtained from user input (for example, name, age, sex, address, etc.) or online services provided by the business operator C. It includes information indicating the user's preference obtained from the web page accessed by the user and the number of accesses. In addition, the first user information and the second user information include information indicating a user's preference obtained from a service or a product purchased by the online service provided by the business operator C. As described above, the first user information and the second user information include information obtained directly or indirectly by the online service provided by the business operator C.

  The first user information group D1 and the second user information group D2 include user information obtained from services provided by different businesses C1 and C2, respectively, and even the same user has different user information. Therefore, if the user information of the same user can be linked between the first user information group D1 and the second user information group D2, the user of the user who receives the service provision across the plurality of business operators C The amount of information can be increased. The larger the information amount of the user information, the more accurately the information based on the user information (for example, the advertisement information according to the user or the provision of the page according to the user) can be provided.

  Therefore, the information processing apparatus 1 according to the first embodiment executes information processing including a linking process of linking the first user information of the first user information group D1 and the second user information of the second user information group D2. To do. Such linking is also called name identification.

  First, the information processing device 1 acquires the information of the first user information group D1 from the device of the business operator C1 and the information of the second user information group D2 from the device of the business operator C2 (step S1). In the example shown in FIG. 1, the first user information group D1 includes the first user information of user IDs "U1000" to "U1999", and the second user information group D2 includes the user ID of "U2000". The second user information up to "U2999" is included.

  Note that in the example shown in FIG. 1, for convenience of explanation, it is assumed that there are 1000 pieces of first user information and 1000 pieces of second user information, but the number of first user information pieces and the number of second user information pieces are shown in FIG. It is not limited to the example shown. In addition, it is assumed that the plurality of elements forming the first user information are p elements (p is an integer of 2 or more) and the plurality of elements forming the second user information are p elements. The number of elements forming the user information is not limited to p.

  Next, the information processing device 1 is included in a part (for example, 200 pieces) of first user information among the plurality of first user information included in the first user information group D1 and the second user information group D2. A part of the plurality of pieces of second user information (for example, 500 pieces) of second user information is used as a judgment candidate, and a judgment process is performed to determine whether the pieces of user information are the same user (step S2).

  In the determination process, the information processing device 1 compares each piece of the first user information with the piece of the second user information. Specifically, the information processing apparatus 1 compares the p elements included in the first user information and the p elements included in the second user information between the elements, and the first user information and the second user information are compared. The degree of similarity with the user information is determined.

  Then, the information processing device 1 determines that the first user information and the second user information whose similarity is equal to or higher than the threshold Th1 are user information of the same user. The information processing device 1 counts the number of pairs of the first user information and the second user information (hereinafter, referred to as temporary binding pairs) that are determined to be the user information of the same user.

  Next, the information processing apparatus 1 determines a comparison candidate, which is an element used for comparison in a later-described linking process, out of p elements included in each of the first user information and the second user information (step S3). .. Specifically, the information processing apparatus 1 determines an element that has a small influence on the above-described similarity, which is an index indicating whether or not the user information is the same user, and the p-elements have a small influence on the similarity 1. A plurality of elements obtained by excluding the above elements are determined as comparison candidates. The comparison candidate is a linking condition in the linking process, and is an element used in the linking process.

  The information processing apparatus 1 sets the plurality of first user information items included in the first user information group D1 and the plurality of second user information items included in the second user information group D2 as pegging elements among p elements. Linking processing is performed using one or more elements of the comparison candidates (step S4).

  For example, it is assumed that the elements 1, 2, and 3 are comparison candidates. In this case, the information processing device 1 includes elements 1 to 3 forming the first user information and elements 1 to 3 forming the second user information among the elements (for example, the elements 1 with each other, the elements 2 with each other, and the element 3 with each other. For each first user information, and the degree of similarity with the second user information is determined.

  Then, the information processing device 1 associates the second user information having the similarity with the first user information with a threshold value or more with the first user information. When there are a plurality of pieces of second user information whose similarity to the first user information is greater than or equal to a threshold value, the information processing apparatus 1 associates the first user information and the second user information having the highest similarity with each other.

  As described above, since the information processing device 1 uses only a part of the plurality of elements forming the user information for the binding process, the processing load on the information processing device 1 can be reduced.

  Further, when the information processing device 1 is managed by the business operator C1, the business operator C2 only needs to disclose all the elements that constitute the second user information, only for a part of the second user information. After the elements of the comparison candidates are determined, only the elements of the comparison candidates can be disclosed. Therefore, the business operator C2 does not have to disclose all the elements configuring the second user information to the business operator C1 after the elements of the comparison candidates are determined, and the security is improved. It is possible to suppress the cost required for user information transmission processing and the like.

[1.2. Information processing system]
FIG. 2 is a diagram illustrating a configuration example of the information processing system according to the first embodiment. As illustrated in FIG. 2, the information processing system 100 according to the first embodiment includes an information processing device 1, a plurality of server devices 2 ₁ to 2 _m (m is an integer of 3 or more), and a plurality of terminal devices 3. _{1 to} 3 _n (n is an integer of 2 or more). The information processing device 1 is managed by the business operator C1 as described above. Further, the server devices 2 _{1 to} 2 _m-1 are managed by the business operators C2 to Cm, respectively.

The information processing device 1, the plurality of server devices 2 ₁ to 2 _m-1 (hereinafter, sometimes collectively referred to as the server device 2) and the plurality of terminal devices 3 _{1 to} 3 _n (hereinafter, collectively referred to as the terminal device 3) Are connected to each other via a network 4 in a wired or wireless manner so that they can communicate with each other. The network 4 is, for example, a LAN (Local Area Network) or a WAN (Wide Area Network) such as the Internet. The terminal apparatus ₃ 1 to 3 _n, the user _U 1 ~U _n (hereinafter, may be collectively referred to as user U) operated by.

  The terminal device 3 is, for example, a smart device (communication terminal) such as a smartphone, a tablet terminal, a PDA (Personal Digital Assistant), and a personal computer, and can execute an application such as a browser. The terminal device 3 can access the information processing device 1 and the server device 2 via the network 4 from a browser or a dedicated application. The user U of the terminal device 3 can use various online services provided by the information processing device 1 and the server device 2.

  The information processing device 1 and the server device 2 can each acquire and store user information from the terminal device 3 of each user U who provides an online service. The server device 2 stores a user information group D including user information acquired from the terminal device 3 of each user U in a database (hereinafter, referred to as DB).

  The information processing device 1 and the server device 2 can provide the information corresponding to the user information to the terminal device 3 of the user U. For example, the information processing device 1 and the server device 2 provide the information of the page dedicated to the user U to the terminal device 3 of the user U based on the held user information, or the user U has a high degree of interest. Advertisement information can be provided to the terminal device 3 of the user U. Accordingly, it is possible to provide the user U with appropriate information for each user U.

  The information processing device 1 acquires the user information group D held by each server device 2 from each server device 2 via the network 4 and stores it. Then, the information processing device 1 can associate the user information between the user information groups D. Hereinafter, the configuration and operation of the information processing device 1 will be specifically described.

[1.3. Information processing device 1]
FIG. 3 is a diagram illustrating a configuration example of the information processing device 1 according to the first embodiment. As illustrated in FIG. 3, the information processing device 1 includes a communication unit 10, a storage unit 11, and a control unit 12 (controller), and provides an online service to the user U of the terminal device 3. Hereinafter, the communication unit 10, the storage unit 11, and the control unit 12 will be specifically described in this order.

[1.3.1. Communication unit 10]
The communication unit 10 is a communication interface that transmits and receives information to and from the network 4, and connects to the network 4 by wire or wirelessly. The control unit 12 can send and receive various information to and from the plurality of server devices 2 and the terminal devices 3 via the communication unit 10 and the network 4.

[1.3.2. Storage unit 11]
The storage unit 11 includes a service information DB 20, user information group tables 21 and 22, a comparison candidate table 23, and a linking information table 24. The service information DB 20, the user information group tables 21 and 22, the comparison candidate table 23, and the association information table 24 are, for example, semiconductor memory devices such as flash memory, or storage devices such as HDD (Hard Disk Drive) and optical disks. ..

  The service information DB 20 is a storage unit that stores information used for services provided by the information processing device 1 to the user U of the terminal device 3. For example, the service information DB 20 stores information such as products used for shopping services and auction services, and video content used for video distribution services.

  The user information group table 21 is set with user information that is information of the user U who uses the service provided by the information processing device 1. The user information stored in the user information group table 21 is the above-mentioned first user information. FIG. 4 is a diagram showing an example of the user information group D stored in the user information group table 21.

  As shown in FIG. 4, the user information group table 21 is information associated with information such as “user ID”, “psychographic attribute”, “demographic attribute”, and “access history”.

The “user ID” is identification information of the user U, and in the example shown in FIG. 4, the user IDs of the users U ₁ , U ₂ , U ₃ , and U ₄ are “U1001”, “U1002”, “U1003”, and “U1003”. "U1004" is set. The user ID is a user's account for a service that uses the information processing device 1. The user ID may be any information as long as it identifies the user U, and the identification information of the terminal device 3 used by the user U may be used as the user ID. For example, the user ID may be identification information specified by an HTTP (Hypertext Transfer Protocol) cookie transmitted and received between the terminal device 3 and the information processing device 1.

  The “psychographic attribute” is information indicating the user U's sense of values, lifestyle, personality, taste, and the like. The "psychographic attribute" is classified into attribute elements such as "car" and "travel". In the example shown in FIG. 4, “1” is stored for each attribute element when the preference of the user U is relatively high, and “0” is stored otherwise. The “psychographic attribute” is not limited to the example of the attribute element shown in FIG. 4, but includes various attribute elements such as economy, politics, baseball, soccer, other sports, sweets, personal computers, white goods, and furniture. May be. The attribute element information is not limited to two levels of 0 and 1, and may be a value of three levels or more.

  “Demographic attribute” indicates demographic attribute information of the user U. The "demographic attribute" is classified into attribute elements such as "gender" and "age" of the user U, for example. “Gender” stores “1” when the user U is a female, and stores “2” when the user U is a male. Further, the age of the user U is stored in “age”. The “demographic attribute” is not limited to the example of the attribute element shown in FIG. 4, and may include various attribute elements such as occupation, family structure, annual income, address, birthplace, and educational background of the user U.

  The “access history” includes access information and action history information that the information processing device 1 acquires from the terminal device 3 when the user of the terminal device 3 uses the service provided by the information processing device 1. The access information includes information such as "date and time" and "browser". The “date and time” indicates the date and time of access from the terminal device 3. In other words, the “date and time” indicates the date and time when the service provided by the information processing device 1 was used by the terminal device 3. The “browser” is information on the browser used by the terminal device 3, and includes “BR01” which is the name of the browser such as “BR01 (8.1)” and “8.1” which is the version information. It is represented by a combination.

  The access information also includes “time difference information”, “route information”, “OS”, “plug-in”, “IP address”, and the like. The “time difference information” indicates information regarding the time difference between the information transmitted from the terminal device 3 and the information received by the information processing device 1. The “route information” indicates information about a route until the information transmitted from the terminal device 3 reaches the information processing device 1. The “route information” includes information for identifying a router or a proxy server used for communication between the information processing device 1 and the terminal device 3.

  The “OS” is information on the OS (Operating System) used by the terminal device 3, and includes the OS name and version name. “Plug-in” indicates information on the plug-in incorporated in the browser (for example, identification information such as the name of the plug-in). The “IP address” indicates the IP (Internet Protocol) address of the terminal device 3.

  Further, the action history information includes a history of actions performed by the user U of the terminal device 3 on the information processing device 1. For example, the action history information includes the URL of the page accessed from the terminal device 3 and the history of products and services purchased by the user U.

  Information of the plurality of user information groups D acquired from the plurality of server devices 2 is set in the user information group table 22. The information of the plurality of user information groups D is set in the user information group table 22, for example. FIG. 5 is a diagram showing an example of the user information group table 22 according to the first embodiment. As shown in FIG. 5, the user information group table 22 is information in which a plurality of access elements such as “company ID”, “user ID”, “attribute”, and “access history” are associated.

  The “company ID” is information for identifying the company C, and in the example shown in FIG. 5, the company ID “C102” is the company ID of the company C2, and the company ID “C103”. Is the company ID of the company C3. The company ID of the company C1 is "C101".

  The “attribute” is information indicating the attribute of the user U, and includes, for example, a psychographic attribute, a demographic attribute, and the like, as in the example shown in FIG. In addition, the “access history” is the access information and behavior that the server device 2 acquires from the terminal device 3 when the user U of the terminal device 3 uses the service provided by the server device 2 as in the example shown in FIG. Contains historical information.

  The “user ID”, “attribute”, and “access history” are information managed and generated by the server device 2, and the user information group D of each business operator C differs among business operators C. For example, the “user ID” may be assigned different identification information among the business operators C, and the attribute elements included in the “attribute” may be the same for the business operators C.

  The user information group table 22 may be provided for each user information of the business operator C. Further, the user information group D included in the user information group table 22 may be deleted from the user information group table 22 by the control unit 12 after being used in the associating process described below by the control unit 12.

  The comparison candidate table 23 includes information indicating the comparison candidates determined by the control unit 12. FIG. 6 is a diagram showing an example of the comparison candidate table 23 according to the first embodiment. As shown in FIG. 6, the comparison candidate table 23 includes information associated with information such as “company ID”, “business type”, and “comparison candidate”.

  The “company ID” is the company ID of the company C, the “business type” is information indicating the business type of the company C, and the “comparison candidate” is the linking process among the plurality of second elements. It is information indicating a comparison candidate that is the second element used in.

  The association information table 24 includes association information indicating the association result of the user information between the business operators C by the control unit 12. FIG. 7 is a diagram showing an example of the association information table 24 according to the first embodiment. As shown in FIG. 7, the association information table 24 includes “first business operator ID”, “second business operator ID”, “user ID of first business operator”, “user ID of second business operator”, and the like. Information is associated information.

  The “first company ID” and the “second company ID” are the company IDs of the two companies C with which the user information is associated. The "user ID of the first operator" is a user ID included in the user information of one operator C of the two operators C to which the user information is linked, and the "user ID of the second operator" is , The user ID included in the user information of the other operator C of the two operators C associated with the user information.

  The association information table 24 illustrated in FIG. 7 includes the association result of the user information of the business operator C1 having the business operator ID “C101” and the business operator C2 having the business operator ID “C102”. The linking result is linked with the user ID. That is, the user ID included in the user information of the business operator C1 and the user ID included in the user information of the business operator C2 are associated and linked.

  For example, in the example illustrated in FIG. 7, the user ID “U1001” and the user ID “U2002” are associated with each other, the user ID “U1002” and the user ID “U2032” are associated with each other, and the user ID “U1003” and the user are associated with each other. The ID “U2028” is linked. Therefore, in the example illustrated in FIG. 7, the user information of the user ID “U1001” is associated with the user information of the user ID “U2002”, and the user information of the user ID “U1002” is the user information of the user ID “U2032”. Is tied to. Also, the user information of the user ID “U1003” is linked to the user information of the user ID “U2028”.

[1.3.3. Controller 12]
The control unit 12 includes, for example, a microcomputer having a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), an input / output port, and various circuits. The control unit 12 includes, for example, a microcomputer having a CPU, a ROM, a RAM, an input / output port, and various circuits. The control unit 12 includes a service providing unit 30, an information acquisition unit 31, a determination unit 32, a tying unit 33, and an information output unit 34. The functions of the service providing unit 30, the information acquisition unit 31, the determination unit 32, the tying unit 33, and the information output unit 34 are, for example, that the CPU of the control unit 12 reads a program stored in the ROM of the control unit 12. It is realized by executing.

  The service providing unit 30, the information acquiring unit 31, the determining unit 32, the associating unit 33, and the information output unit 34 are each partly or wholly an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array). It may be configured by hardware.

[1.3.3.1. Service provision department 30]
The service providing unit 30 provides a service to the user U of the terminal device 3. The service providing unit 30 can provide the service to the user U of the terminal device 3 by acquiring the information from the service information DB 20 in response to the request from the terminal device 3 and transmitting the information to the terminal device 3.

  For example, the service providing unit 30 obtains information about products used for the service from the service information DB 20 and sends it to the terminal device 3, or obtains moving image content and music content from the service information DB 20 and sends it to the terminal device 3. You can

  Further, the service providing unit 30 stores the above-mentioned user information acquired from the terminal device 3 in providing the service to the terminal device 3 in the user information group table 21. For example, the service providing unit 30 can store the access history in the user information group table 21 based on the information transmitted from the terminal device 3.

  Further, the service providing unit 30 may determine the psychographic attribute based on the result of the questionnaire input for the user U to use the service provided by the information processing device 1, and set the psychographic attribute in the user information group table 21. it can. The service providing unit 30 can also determine the psychographic attribute based on the action history information described above and set it in the user information group table 21. Further, the service providing unit 30 acquires, from the terminal device 3, the user information (for example, name, age, address) input for the user U to use the service provided by the information processing device 1, and the user information group table. It can be set to 21.

[1.3.3.2. Information acquisition unit 31]
The information acquisition unit 31 acquires user information from the server device 2 and sets it in the user information group table 22. For example, the information acquisition unit 31 may be a server device 2 ₁ acquires information of the second user information group D2, sets information of the second user information group D2 in the user information group table 22. The information acquisition unit 31 can be set from the server apparatus 2 ₂ acquires information of the user information group D3, the information of the user information group D3 in the user information group table 22.

Server 2 _1, when updating the second user information in the user information group D2 by the provision of the service, and transmits the update information indicating the updated contents of the user information included in the second user information group D2 to the information processing apparatus 1 .. Information acquisition unit 31, based on the acquired update information from the server device 2 _1, it is possible to update the second user information group D2.

Similarly, the server apparatus 2 _2, when updating the user information of the user information group D3 by the provision of the service, and transmits the update information indicating the updated contents of the user information included in the user information group D3 to the information processing apparatus 1. Information acquisition unit 31, based on the acquired update information from the server apparatus 2 _2, it is possible to update the user information group D3.

The information acquisition unit 31 acquires, from the server device 2, information indicating a linking request that requests linking of user information. For example, the server device 2 ₁ performs pegging request for association between user information and the user information of the operators C2 operators C1, the server apparatus 2 _2, the user information and company C3 operators C1 A linking request for requesting linking with user information can be performed.

Server 2 _1, among a plurality of user information included in the second user information group D2, the number required for the determination of the comparison element by determining unit 32 (hereinafter, the initial number of users) information processing user information content Send to device 1. Thereafter, the server device 2 _1, after obtaining the information indicating the comparison element from the information processing apparatus 1, excluding elements other than Comparative Element of the plurality of elements constituting the user information included in the second user information group D2 User information is transmitted to the information processing device 1.

Similarly, the server apparatus 2 _2, among the plurality of user information included in the user information group D3, and transmits the user information of the initial number of users to the information processing apparatus 1. Then, the server apparatus 2 _2, after obtaining the information indicating the comparison element from the information processing apparatus 1, user information, excluding the elements other than Comparative Element of the plurality of elements constituting the user information included in the user information group D3 Is transmitted to the information processing device 1.

  Note that the server device 2 sends the user information including all the elements constituting the user information included in the user information group D3 to the information processing device 1 regardless of whether the comparison element is determined by the information processing device 1. You can also send it.

[1.3.3.3. Determining unit 32]
The determination unit 32 includes some or all of the plurality of first user information items included in the first user information group D1 of the business operator C1 and the user information items included in the user information group D of the business operator C other than the business operator C1 ( Hereinafter, based on a part of the second user information), the linking unit 33 determines a comparison candidate to be used for comparison.

  Specifically, the determining unit 32 includes a plurality of elements that make up the first user information (hereinafter sometimes referred to as the first element) and a plurality of elements that make up the second user information (hereinafter, the second element). Element and sometimes described) is performed.

  Then, the determining unit 32 determines, based on the result of the comparison process, one or more first elements of the comparison candidates among the plurality of first elements and one or more second elements of the comparison candidates among the plurality of second elements. To determine each. The determination unit 32 sets information on the determined comparison candidates in the comparison candidate table 23.

  Here, it is assumed that the plurality of first elements forming the user information of the first user information group D1 and the plurality of second elements forming the user information of the second user information group D2 have a one-to-one correspondence. , And the number of first elements and the number of second elements are both p. That is, it is assumed that there are p pairs of the first element and the second element (hereinafter, referred to as element pairs). The determination unit 32 compares the first element and the second element included in each element pair.

  The correspondence between the first element and the second element means that the first element and the second element can be compared with each other, for example, that the first element and the second element are the same or similar. The determination unit 32 can handle the plurality of second elements as one second element in order to associate the first element with the second element, and treat the plurality of first elements as one first element. You can

  The number of the first elements and the number of the second elements are not limited to p, and the number of the first elements and the number of the second elements may be different from each other. When the number of the first element and the number of the second element are different from each other, the determining unit 32 may use only the pair of the first element and the second element corresponding to each other and not use the remaining elements. it can.

  The determining unit 32 can make the plurality of first elements correspond to the second element as one first element, and can make the plurality of second elements correspond to the first element as one second element. For example, the determining unit 32 can treat the two second elements “overseas travel” and “domestic travel” as one second element “travel”.

The determination unit 32 can determine the similarity Rd between the user information by comparing the first element and the second element included in each element pair. For example, the determination unit 32 can determine the degree of similarity Rd between the user information by the calculation of the following formula (1). In the following formula (1), x1 _i is the information of the first element of the i-th (for example, 1 ≦ i ≦ p) element pair, and x2 _i is the information of the second element of the i-th element pair. is there. In the following formula (1), (x1 _i −x2 _i ) ² can also be called a dissimilarity score.

For example, the user information managed by the business operator C1 and the user information managed by the business operator C2 have the same attribute element "travel", both of which are set to "1", and the attribute element "travel" compares 1 Suppose it is the th attribute element. In this case, x1 ₁ = x2 ₁ = 1 and (x1 ₁ −x2 ₁ ) ² = 0. Further, it is assumed that the attribute element “travel” of the user information managed by the business operator C1 is “1”, and the attribute element “travel” of the user information managed by the business operator C2 is “0”. In this case, x1 ₁ = 1 and x2 ₁ = 0, and (x1 ₁ −x2 ₁ ) ² = 1.

Further, the user information managed by the business operator C1 includes the attribute element “travel”, the user information managed by the business operator C2 includes attribute elements “domestic travel” and “overseas travel”, and the attribute element “travel” is “1”. , And at least one of the attribute elements “domestic travel” and “overseas travel” is “1”. In this case, the determination unit 32 can set x1 ₁ = x2 ₁ = 1.

When the attribute element information is not a two-step value (1 and 0) but a three-step value or more, the similar calculation can be performed to obtain the dissimilarity score. Information on attribute elements such as address, age, and name is compared based on preset criteria, and a value corresponding to (x1 _i −x2 _i ) ² is calculated by the determining unit 32. For example, when the attribute element is an address, the determination unit 32 calculates that the dissimilarity score approaches 0 as the address becomes closer.

  The method of determining the degree of similarity Rd is not limited to the above example. For example, the determination unit 32 can also calculate the similarity Rd by dividing the right side of the above formula (1) by the number of elements to be compared (hereinafter, sometimes referred to as a comparison element). Further, the determining unit 32 can weight the dissimilarity score for each element and calculate the square root of the sum, and use the reciprocal of the square root as the similarity Rd.

  The determining unit 32 can also calculate the total value of the similarity scores for each comparison element and calculate the similarity Rd based on the total value. The determination unit 32 increases the similarity score of each comparison element as the first element and the second element of the element pair are similar to each other. The determining unit 32 may calculate the similarity Rd between the user information by, for example, calculating the total value of the similarity scores for each comparison element and dividing the square root of the total value by the number of comparison elements. it can.

  Further, the determining unit 32 calculates a total value by weighting and accumulating the similarity scores of the comparison elements, and using the square root of the total value as a coefficient according to the number of comparison elements as the number of comparison elements. It is also possible to calculate the degree of similarity between user information by dividing by the multiplied value.

  The determination unit 32 determines whether or not the similarity Rd between the user information is equal to or more than a threshold Th1 and counts the number of temporary linking pairs which are pairs of user information with the similarity Rd equal to or more than the threshold Th1. I do. Such a temporary tied pair is a pair of user information tied when all the comparison elements are used in the tied portion 33. Hereinafter, the number of temporary tied pairs obtained by using all of the first element and the second element corresponding to each other will be referred to as the reference number of temporary tied pairs.

  Next, the determination unit 32 reduces the number of comparison elements by one, performs the above-described comparison processing, determination processing, and counting processing to obtain the number of temporary tied pairs. Hereinafter, the number of provisional tied pairs obtained by reducing the number of comparison elements by one or more is referred to as the number of comparison pairs. Then, the determining unit 32 calculates the ratio Rp of the number of comparison pairs to the number of reference temporary tied pairs. For example, when the number of reference temporary tied pairs is q (q ≦ p) and the number of comparison pairs is q−2, the determination unit 32 is Rc = q / (q−2).

  The number of reference provisional pegging pairs is the number of provisional pegging pairs when the number of first elements and all of the second elements corresponding to each other are used, and the ratio Rc of the number of comparison pairs to the number of reference provisional pegging pairs is the reference provisional pegging pair. It can be said that the success probability Rs with respect to the number of linked pairs.

  The determining unit 32 performs the above-described comparison processing, determination processing, and counting processing while changing the elements that are not used in the comparison processing among the p elements of the comparison elements, and determines the combination of comparison elements or each comparison element. The success probability Rs is calculated. Then, the determination unit 32 determines an element having a small success probability Rs as an element having a small influence on the similarity Rd, and obtains a plurality of elements obtained by excluding one or more elements having a small influence on the similarity Rd from the p elements. Elements can be determined as comparison candidates.

  Further, the determining unit 32 increases the number of elements used in the comparison process until the success probability Rs reaches a predetermined range Rth1 (for example, a range of 85% to 90%), thereby comparing a plurality of elements from p elements. Can be determined as a candidate. For example, the determining unit 32 can determine the success probability Rs each time an element is added, and can determine one or more elements used for comparison when the success probability Rs is within the predetermined range Rth1 as a comparison candidate. In this case, the determining unit 32 can add elements in descending order of success probability Rs as elements to be used in the comparison process.

  Further, the determining unit 32 can also calculate the reduction rate Rr of the number of comparison pairs with respect to the number of reference provisional tied pairs. For example, when the number of reference temporary tied pairs is q and the number of comparison pairs is q-2, the determination unit 32 is Rr = {1-q / (q-2)}.

  The determination unit 32 can determine the reduction rate Rr for each element pair while changing the element pairs that are not used in the comparison process among the q element pairs. When the reduction rate Rr is less than or equal to the predetermined range Rth2 (for example, a range of 10% to 15%), the determining unit 32 determines an element not used in the comparison process as an element pair that has a small influence on the success probability Rs. can do.

  Further, the determination unit 32 performs the comparison process and the determination process while sequentially excluding elements from the smallest reduction ratio Rr from the comparison target until the reduction ratio Rr of the temporary tied pair falls within the predetermined range Rth2. The information processing apparatus 1 can determine one or more elements used for comparison as a comparison candidate when the reduction rate Rr of the temporary tied pair falls within the predetermined range Rth2. Note that the determination unit 32 can also determine, as a comparison candidate, elements other than one or more elements whose reduction rate Rr of the temporary tied pair is equal to or less than the threshold Th2.

  In this way, the determination unit 32 can determine the comparison candidate based on the result of the repeated comparison while increasing or decreasing the first element and the second element, respectively.

  Further, the determination unit 32 determines the comparison candidate based on the comparison candidate table 23 and based on the information indicating the comparison candidate set for the other company C without performing the above-described comparison processing and determination processing. You can also For example, the determination unit 32 can set the same comparison candidate for the user information group D of the business operator C having the same or similar business type based on the comparison candidate table 23.

  For example, as illustrated in FIG. 6, it is assumed that the comparison candidate table 23 is set, and in the comparison candidate table 23, the comparison candidate is set to the business operator C2 having the business operator ID “C103”. In this case, the business type of the business operator C2 having the business ID “C102” and the business type of the business operator C3 having the business ID “C103” are both automobile manufacturing. Therefore, the determination unit 32 can determine the comparison candidate associated with the company ID “C103” as the comparison candidate for the company C2 having the company ID “C102”.

  Further, instead of setting the same comparison candidate for the business operator C having the same or similar business type, the determination unit 32 is the same for the user information group D of the service of the same or similar type. A comparison candidate can be set. That is, the determination unit 32 can set the same comparison candidate for each service type in which the user information group D is independently managed, instead of the business type of the business operator C.

  If the same comparison candidate cannot be set between the user information groups D of the business operator C having the same or similar business type, the determination unit 32 selects one of the plurality of comparison candidates set in one user information group D. It is also possible to use only common elements as comparison candidates for the other user information group. The same applies to services of the same or similar type.

  In addition, the determination unit 32 determines one or more elements used for comparison when the success probability Rs is in the predetermined range Rth1 while increasing the comparison candidates determined in the past with priority, as the comparison candidates. You can Further, the determination unit 32 preferentially increases the comparison candidates determined in the past, and increases one or more elements used for comparison when the decrease rate Rr of the temporary tied pair falls within the predetermined range Rth2. Can be determined as

  The comparison process and the determination process may be performed by the associating unit 33, and in this case, the determining unit 32 can determine the comparison candidate based on the similarity Rd determined by the associating unit 33.

[1.3.3.4. Stringing part 33]
The linking unit 33 acquires the plurality of first user information items of the first user information group D1 from the user information group table 21, and acquires the plurality of second user information items of the second user information group D2 from the user information group table 22. .. Further, the linking unit 33 acquires information on the comparison candidates associated with the business operators C1 and C2 from the comparison candidate table 23.

  Then, the linking unit 33 determines the degree of similarity between the plurality of first user information items in the first user information group D1 of the business operator C1 and the second user information items included in the user information group D of the business operator C other than the business operator C1. Rd is determined using the comparison candidates set in the comparison candidate table 23, and based on the determination result, a linking process for linking the first user information and the second user information is performed.

  Specifically, the associating unit 33 selects, from among the plurality of first elements constituting the first user information, information on one or more first elements of the comparison candidates and among the plurality of second elements constituting the second user information. Based on the comparison with the information of one or more second elements of the comparison candidates, the linking process for linking the first user information and the second user information is performed. Hereinafter, an element pair of the first element comparison candidate and the second element comparison candidate that correspond to each other will be referred to as a comparison candidate pair.

  The tying unit 33 compares, for each comparison candidate pair, the first element and the second element included in the comparison candidate pair, similarly to the comparison process in the determination unit 32, and determines the similarity score for each comparison candidate pair. To do. Then, the associating unit 33 determines the similarity Rd between the first user information and the second user information based on the total value of the similarity scores of the comparison candidate pairs, as in the determination process of the determining unit 32. To do.

  The tying unit 33 can calculate the similarity Rd between the user information by the calculation of the above-described formula (1). The tying unit 33 can also calculate the similarity Rd between the user information by the above-described calculation other than the calculation of the expression (1).

  The tying unit 33 determines whether or not the degree of similarity Rd between the first user information and the second user information is a threshold Th1 or more, and the first user information and the second user information whose similarity Rd is the threshold Th1 or more. Link with. The linking is performed by generating linking information that associates the user IDs included in the user information. The association unit 33 sets the generated association information in the association information table 24. The association information includes the user ID of the first user information and the user ID of the second user information that are associated with each other.

  Further, the associating unit 33 performs the above-described comparison processing using one or more comparison candidates having a relatively high success probability Rs among the plurality of comparison candidates determined in the determination processing, and the first user information and the second user information It is possible to determine whether or not the degree of similarity Rd with the user information is greater than or equal to the threshold value Th3 (<Th1). The associating unit 33 can perform an exclusion process of excluding the first user information and the second user information whose similarity Rd is less than the threshold Th3 from the user information of the association candidate.

  Next, the associating unit 33 performs the associating process of the first user information and the second user information after performing the above-described exclusion process, using all the comparison candidates determined in the determining process. Accordingly, since it is possible to perform the linking process after narrowing down the user information of the linking candidates by the comparison candidate having the relatively high success probability Rs, it is possible to reduce the processing load of the linking process.

  The associating unit 33 can also repeatedly add one or more comparison candidates in order from the one with the highest success probability Rs, and gradually narrow down the user information of the associative candidates. The associating unit 33 increases the threshold Th3 each time the user information of the associating candidates is narrowed down. The tying unit 33 can perform the tying processing after performing the narrowing down in two steps or three or more steps.

  For example, when performing the narrowing down in the second stage, the tying unit 33 uses thresholds Th31 and Th32 (Th31 <Th32 <Th3) as the threshold Th3. The associating unit 33 performs the comparison process using one or more comparison candidates with the highest success probability Rs, and excludes the first user information and the second user information whose similarity Rd is less than the threshold Th31 from the user information of the association candidate. To do.

  Next, the associating unit 33 adds one or more comparison candidates having the next highest similarity Rd, and uses the one or more comparison candidates having the highest success probability Rs and the one or more comparison candidates having the next highest probability Rs to perform the association. A comparison process is performed between the first user information and the second user information that are not excluded from the candidates. Then, the associating unit 33 excludes the first user information and the second user information whose similarity Rd is less than the threshold Th32 from the user information of the association candidate. The associating unit 33 can perform the associating process on the first user information and the second user information of the associating candidates thus narrowed down in two stages.

  In this way, the associating unit 33 can reduce the load of the associating process because only a part of the plurality of elements constituting the user information is used for the associating process. Further, the business operator C2 only needs to disclose all the elements that constitute the second user information, as long as it is a part of the second user information, and once the elements of the comparison candidate are once determined, Only the elements can be disclosed. Therefore, since the user information transmitted to the information processing device 1 by the business operators C other than the business operator C1 does not include a part, the security is improved and the cost for the transmission processing of the second user information is suppressed. be able to.

[1.3.3.5. Information output unit 34]
The information output unit 34 can acquire the association information and the information of the user information group D corresponding to the association request from the server device 2 from the storage unit 11 and transmit the information to the server device 2 that is the request source.

For example, it assumed to be the pegging request pegging request from the server apparatus 2 ₁ requests the association between the user information and the user information of the operators C2 operators C1. In this case, the information output unit 34 can transmit the correlating information attaching cord and user information of the user information and company C2 operators C1, and information of the user information group D1 to server 2 _1.

The information output unit 34 generates the unified user information is user information that integrates the user information of the user information and operator C2 of the cord with the carrier C1, and transmits the integrated user information to the server device 2 ₁ be able to. The integrated user information may include only the attribute element, or may include both the attribute element and the access element, for example.

  Further, the information output unit 34 can transmit the user ID and the user information when the information acquisition unit 31 receives the user information transmission request including the user ID from the server device 2 that has made the association request. The information output unit 34 can acquire the user information corresponding to the user information transmission request from the storage unit 11 and can transmit the acquired user information to the server device 2.

For example, the received by the information transmission request information acquisition unit 31 requests transmission of user information operators C1 carrying user information corresponding from the server device 2 ₁ with the user ID "U2002." In this case, the information output unit 34 determines that the user ID of the user information of the business operator C1 who holds the user information corresponding to the user ID “U2002” is “U1001” based on the association information table 24.

The information output unit 34 acquires the user information of the user ID "U1001" from the user information group table 22, and transmits the acquired user information of the user ID "U1001" and the server device _{2 1.} The information output unit 34 can also transmit the user information and integrate user information that integrates the user information of the user ID "U2002" of the user ID "U1001" to the server device _{2 1.}

  Further, the information output unit 34 can also transmit the user information including the user ID included in the association information to the server device 2 every time the association information is added. Further, the information output unit 34 can also send the updated user information to the server device 2 each time the user information including the user ID included in the association information is updated.

  It should be noted that the information output unit 34 can transmit only the elements of the comparison candidates in the first user information of the business operator C1 to the server device 2. As a result, the business operator C1 can disclose only the information of the comparison candidate elements, and therefore, the business operator C1 does not disclose all the elements constituting the first user information to the business operator C2. However, it is possible to improve the security and suppress the cost required for the transmission process of the first user information.

[1.4. Processing flow of information processing apparatus 1]
An example of an information processing procedure in the information processing device 1 will be described with reference to FIG. FIG. 8 is a flowchart showing an example of the flow of information processing in the information processing device 1. Such information processing is a process that is repeatedly executed.

  As shown in FIG. 8, the control unit 12 determines whether or not there is a service request from the terminal device 3 (step S10). When determining that there is a service request (step S10: Yes), the control unit 12 provides information and processing according to the service request to provide the service (step S11).

  When the process of step S11 ends or when it is determined that there is no service request (step S10: No), the control unit 12 determines whether or not the user information is acquired from the terminal device 3 (step S12). When it is determined that the user information is acquired (step S12: Yes), the control unit 12 updates the user information group table 21 stored in the storage unit 11 (step S13).

  When the process of step S13 ends or when it is determined that the user information has not been acquired (step S12: No), the control unit 12 determines whether or not there is a binding request from the server device 2 (step S14). ). When it is determined that there is a linking request (step S14: Yes), the control unit 12 performs a comparison candidate determination process (step S15). The process of step S15 is the process of steps S20 to S23 shown in FIG. 9, and will be described later.

  The control unit 12 performs a linking process for linking the first user information and the second user information using the one or more comparison candidates determined in step S15 (step S16). The process of step S16 is the process of steps S30 and S31 shown in FIG. 10, and will be described later.

  When it is determined that there is no association request (step S14: No) or when the process of step S16 ends, the control unit 12 determines whether there is an information transmission request from the server device 2 (step S17). When determining that there is an information transmission request (step S17: Yes), the control unit 12 transmits information according to the information transmission request to the server device 2 (step S18). When the process of step S18 ends or when it is determined that there is no information transmission request (step S17: No), the control unit 12 ends the process illustrated in FIG.

  FIG. 9 is a flowchart showing the flow of the comparison candidate determination process of step S15 shown in FIG. As illustrated in FIG. 9, the control unit 12 displays the information of the comparison candidates associated with the business operator C of the business type that is the same as or similar to the business type of the business operator C that manages the user information group D that determines the comparison candidates. Then, it is determined whether or not it is included in the comparison candidate table 23 (step S20). When the control unit 12 determines that the information of the comparison candidates of the same or similar business type is not included in the comparison candidate table 23 (step S20: No), the success probability Rs of each element is determined (step S21). ..

  Next, the control unit 12 determines comparison candidates (step S22). Specifically, the control unit 12 determines a comparison candidate based on the success probability Rs determined in step S21. When the process of step S22 ends, the control unit 12 sets the information of the comparison candidate in the comparison candidate table 23 and updates the comparison candidate table 23 (step S23), and ends the process shown in FIG.

  For example, the control unit 12 sets the information of the comparison candidates determined in step S22 in the comparison candidate table 23. In addition, when the control unit 12 determines that the information of the comparison candidates of the same or similar business type is included in the comparison candidate table 23 (step S20: Yes), the same or the same information set in the comparison candidate table 23 is set. Comparison candidates of similar business types are set in the comparison candidate table 23 as comparison candidates.

  FIG. 10 is a flowchart showing the flow of the linking process of step S16 shown in FIG. As shown in FIG. 10, the control unit 12 narrows down the linking candidates while sequentially increasing the comparison candidates (step S30). Then, the control unit 12 associates the user information narrowed down in step S30 with all the comparison candidates (step S31), and ends the processing illustrated in FIG.

[2. Second Embodiment]
[2.1. Information processing]
In the first embodiment, the narrowed-down second element is determined as the associating condition, and the associating process is performed by the narrowed-down second element, whereas in the second embodiment, the narrowed-down second user information is associated. The difference is that the linking process is performed with the second user information that has been determined and narrowed down. In the following, components having the same functions as those in the first embodiment will be designated by the same reference numerals and description thereof will be omitted, and the description will focus on the points different from the first embodiment.

  Information processing according to the second embodiment will be described with reference to FIG. 11. FIG. 11 is an explanatory diagram of information processing according to the embodiment, and in the second embodiment, information processing is executed by the information processing device 1A.

  First, the information processing apparatus 1A acquires the information of the first user information group D1 from the device of the business operator C1 and the information of the second user information group D2 from the device of the business operator C2 (step S1).

  Next, the information processing device 1A is included in a part (for example, 200 pieces) of first user information among the plurality of first user information included in the first user information group D1 and the second user information group D2. A part of the plurality of pieces of second user information (for example, 500 pieces) of second user information is used as a judgment candidate, and a judgment process is performed to determine whether the pieces of user information are the same user (step S2).

  In the determination processing, the information processing apparatus 1A compares each piece of the first user information with the piece of the second user information, similarly to the processing of the information processing apparatus 1. Specifically, the information processing apparatus 1A compares a plurality of first elements included in the first user information with a plurality of second elements included in the second user information, and compares the elements with the first user information. The degree of similarity Rd with the second user information is determined. Then, the information processing apparatus 1A determines that the first user information and the second user information whose similarity Rd is the threshold Th1 or more are the user information of the same user.

  Next, the information processing apparatus 1A generates a linking possibility model having a plurality of first elements constituting the first user information as features, using the similarity Rd as a correct answer label. Then, the information processing apparatus 1A determines a candidate for the linking process (hereinafter, also referred to as a linking candidate) that is the first user information used for comparison in the linking process described later based on the linking possibility model ( Step S3A). The linking possibility model is, for example, a machine learning model such as a linear regression model or a support vector machine, and the degree of linking possibility can be calculated.

  The information processing apparatus 1A determines, as the linking candidate, the first user information having a predetermined ratio (for example, 20%) having a higher degree of linking possibility among the plurality of first user information included in the first user information group D1. be able to. The pegging candidate is a pegging condition that defines the first user information used in the pegging process.

  The information processing apparatus 1A performs a process of associating the first user information, which is the candidate for the process of association determined in step S3A, and the plurality of second user information included in the second user information group D2 (step S4).

  As described above, since the information processing apparatus 1A uses only a part of the first user information having a high degree of possibility of association for the association processing, the processing load on the information processing apparatus 1A can be reduced. In addition, the first user information having a large amount of information of the first element effective for the association among the plurality of first elements can be preferentially selected by the association possibility model and can be selected as the association candidate. It is not necessary to disclose the first user information whose amount of information is biased to the first element, which is unnecessary for linking, out of the plurality of first elements constituting the information, to the business operator C2, which improves security and improves the first user. It is possible to suppress the cost required for information transmission processing.

[2.2. Information processing apparatus 1A]
FIG. 12 is a diagram illustrating a configuration example of the information processing device 1A according to the second embodiment. As illustrated in FIG. 12, the information processing device 1A includes a communication unit 10, a storage unit 11, and a control unit 12A (controller), and provides an online service to the user U of the terminal device 3. Since the communication unit 10 and the storage unit 11 are the same as those in the first embodiment, the control unit 12A will be specifically described below.

[2.2.1. Controller 12A]
The control unit 12A includes, for example, a microcomputer having a CPU, a ROM, a RAM, an input / output port, and various circuits. The control unit 12A includes, for example, a microcomputer having a CPU, a ROM, a RAM, an input / output port, and various circuits. The control unit 12A includes a service providing unit 30, an information acquisition unit 31, a determination unit 32A, a binding unit 33A, and an information output unit 34. The functions of the service providing unit 30, the information acquisition unit 31, the determination unit 32A, the tying unit 33A, and the information output unit 34 are, for example, that the CPU of the control unit 12A reads a program stored in the ROM of the control unit 12A. It is realized by executing.

  The service providing unit 30, the information acquisition unit 31, the determination unit 32A, the linking unit 33A, and the information output unit 34 may be partially or wholly configured by hardware such as ASIC and FPGA. Since the service providing unit 30, the information acquisition unit 31, and the information output unit 34 are the same as those in the first embodiment, the determination unit 32A and the tying unit 33A will be specifically described below.

[2.2.1.1. Determining unit 32A]
The determination unit 32A includes a part of the plurality of first user information included in the first user information group D1 of the business operator C1 and the user information included in the user information group D of the business operators C other than the business operator C1 (hereinafter, Based on a part or all of the second user information), the linking unit 33A determines a linking candidate to be used for comparison.

  Specifically, the determination unit 32A compares the plurality of first elements forming the first user information with the plurality of second elements forming the second user information by the same process as the determination unit 32. Perform processing. Then, the determining unit 32A determines the similarity Rd between the user information by the same process as the determining unit 32 based on the result of the comparison process.

  32 A of determination parts perform learning by machine learning models, such as a linear regression model, using the similarity Rd as a correct label and each 1st element which comprises 1st user information as a feature, for example. A learning model such as the linear regression model can also be called a success probability model. The determination unit 32A performs a calculation using a machine learning model for each first user information included in the user information group D1 of the business operator C1 and ranks higher than the calculation result (for example, a predetermined value with respect to the total number of the first user information). It is possible to determine the first user information of (number of ratios) as a linking candidate.

  Note that the determination unit 32A performs one or more first elements of the comparison candidates among the plurality of first elements and one or more second elements of the comparison candidates among the plurality of second elements by the same processing as the determination unit 32. Can also be determined respectively. In this case, the determination unit 32A can also perform the calculation using the machine learning model by using only the comparison candidates among the plurality of first elements that form the first user information. Thereby, the calculation load of the machine learning model can be reduced.

  In addition, the determination unit 32A can determine one or more comparison candidates based on the information amount of one or more comparison candidates. For example, the determination unit 32A may determine, as the linking candidate, the first user information in which the information amount of each comparison candidate is equal to or more than the threshold value among the plurality of first user information included in the user information group D1 of the business operator C1. it can. Further, the determining unit 32A determines, as the linking candidate, the first user information in which the total value of the information amounts of the one or more comparison candidates is equal to or more than the threshold value among the plurality of first user information included in the user information group D1. You can

  In addition, the determining unit 32A performs the process of determining the first user information whose total value of the information amount of the comparison candidates is equal to or more than the threshold value as the association candidate while increasing the comparison candidates, and the first information included in the user information group D1. It can be performed until the number reaches a predetermined ratio with respect to the total number of user information.

[2.2.1.2. Pegging portion 33A]
The linking unit 33A acquires a plurality of first user information items of the first user information group D1 from the user information group table 21. In addition, the associating unit 33A acquires the plurality of associating candidates determined by the determining unit 32A from the determining unit 32A.

  Then, the associating unit 33A performs the same process as the associating unit 33, and the similarity between the plurality of first user information items of the first user information group D1 of the business operator C1 and the plurality of associating candidates determined by the determining unit 32A. Rd is determined, and based on the determination result, a linking process for linking the first user information and the second user information is performed.

[2.3. Processing flow of information processing apparatus 1A]
An example of an information processing procedure in the information processing apparatus 1A will be described with reference to FIGS. 13 and 14. Note that steps S15 and S16 will be specifically described below because the processing is the same as that of the first embodiment except the processing of step S15 shown in FIG. 8 and the stringing processing of step S16 shown in FIG.

  FIG. 13 is a flowchart showing the flow of a comparison candidate determination process according to the second embodiment. As illustrated in FIG. 13, the control unit 12A creates a linking possibility model for calculating the degree of linking possibility between the first user information and the second user information based on a part of the first user information. Generate (step S41). Then, the control unit 12A determines, by using the associating possibility model, the first user information of a predetermined ratio (for example, 20%) having a higher degree of associating possibility among the plurality of first user information as the associating candidate. (Step S42).

  Note that the control unit 12A associates the first user information and the second user information whose similarity Rd is the threshold Th1 or more as user information of the same user when generating the association possibility model in step S41. You can In this case, the control unit 12A can determine the degree of linking possibility using the linking possibility model for the remaining first user information for which the degree of similarity Rd has not been calculated in step S41.

  FIG. 14 is a flowchart showing the flow of the linking process according to the second embodiment. As illustrated in FIG. 14, the control unit 12A compares the linking candidate determined in step S41 with each piece of the second user information to obtain the similarity Rd (step S50). Then, the control unit 12A associates the first user information and the second user information whose similarity Rd is the threshold Th1 or more as the user information of the same user (step S51), and ends the process illustrated in FIG. ..

[3. Modification example)
In the above-described embodiment, the control units 12 and 12A of the information processing devices 1 and 1A perform the linking process when the linking request is received from the server device 2, but the business operator who does not request the linking from the server device 2. It is also possible to perform a process of associating the user information between Cs.

  Further, the associating units 33 and 33A can execute the above-described associating process when the processing load of the information processing devices 1 and 1A is equal to or less than the predetermined value. This can prevent the processing load of the information processing device 1 from becoming too high.

  Further, in the above-described embodiment, an example in which one user information group D is used for a plurality of services provided by the information processing devices 1 and 1A has been shown, but a plurality of services can be managed separately. In this case, for example, the user information group D is managed for each service. The control units 12 and 12A of the information processing devices 1 and 1A can also perform the above-described linking process between the user information groups D for each service in the information processing device 1.

  Further, the business operator C other than the business operator C1 may determine the elements to be included in the user information transmitted to the information processing apparatus 1, and transmit the user information including only the determined elements from the server apparatus 2 to the information processing apparatus 1. it can. For example, the business operator C2 can transmit the second user information excluding the element of the demographic attribute to the information processing device 1. As a result, it is possible to prevent elements that the business operator C2 does not want to disclose to the business operator C1 from being transmitted to the information processing apparatus 1.

  The control unit 12 can narrow down the second user information by the same processing as the control unit 12A. That is, the control unit 12 can narrow down the second element to determine it as a comparison element and narrow down the grandmother second user information to determine it as a comparison candidate. The control unit 12 can compare the comparison candidate with the second user information for the comparison element.

  In addition, the determination unit 32 of the control unit 12 can determine the first user information and the second user information that the similarity Rd is determined to be equal to or greater than the threshold Th1 as the user information of the same user. In this case, the associating unit 33 can also handle the remaining second user information that is not handled by the determining unit 32 as a candidate for associating.

[4. Hardware configuration]
Each of the information processing apparatuses 1 and 1A in the above-described embodiment is realized by the computer 200 having the configuration illustrated in FIG. 15, for example, executing a program.

  FIG. 15 is a diagram illustrating an example of a hardware configuration of a computer that executes a program. The computer 200 includes a CPU 201, a RAM 202, a ROM 203, a HDD (Hard Disk Drive) 204, a communication interface (I / F) 205, an input / output interface (I / F) 206, and a media interface (I / F) 207.

  The CPU 201 operates based on a program stored in the ROM 203 or the HDD 204 to control each unit. The ROM 203 stores a boot program executed by the CPU 201 when the computer 200 starts up, a program dependent on the hardware of the computer 200, and the like.

  The HDD 204 stores data used by the programs executed by the CPU 201. The communication interface 205 corresponds to the communication unit 10, receives data from another device via the network 4, sends the data to the CPU 201, and transmits the data generated by the CPU 201 to the other device via the network 4.

  The CPU 201 controls an output device such as a display and a printer and an input device such as a keyboard and a mouse via the input / output interface 206. The CPU 201 acquires data from the input device via the input / output interface 206. The CPU 201 also outputs the generated data to the output device via the input / output interface 206.

  The media interface 207 reads a program or data stored in the recording medium 208 and provides the program or data to the CPU 201 via the RAM 202. The CPU 201 loads the program from the recording medium 208 onto the RAM 202 via the media interface 207 and executes the loaded program. The recording medium 208 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or a PD (Phase change rewritable Disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory. Etc.

  The CPU 201 of the computer 200 realizes each function of the control units 12 and 12A described above by executing the program loaded on the RAM 202. Further, each of the control units 12 and 12A may be partially or entirely configured by only hardware.

  The CPU 201 of the computer 200 reads the program from the recording medium 208 and executes the program. However, as another example, the program may be acquired from another device via the network 4.

  The HDD 204 corresponds to the storage unit 11 and stores the same data as the storage unit 11. Further, instead of the HDD 204, a semiconductor memory device such as a RAM or a flash memory, or a storage device such as an optical disk may be used.

[5. effect〕
The information processing devices 1 and 1A according to the embodiment include a storage unit 11 and control units 12 and 12A. The storage unit 11 stores at least a part of information of the first user information group D1 including a plurality of first user information and at least a part of information of the second user information group D2 including a plurality of second user information. To do. The control units 12 and 12A determine the first user information and the second user information based on the result of the comparison between the plurality of first elements forming the first user information and the plurality of second elements forming the second user information. A linking condition for linking and is determined, and a linking process for linking the first user information and the second user information is performed based on the linking condition. Thereby, the information processing device 1 can efficiently perform the linking process, for example.

  The control unit 12 also includes a determination unit 32 and a tying unit 33. Based on the result of the comparison between the plurality of first elements forming the first user information and the plurality of second elements forming the second user information, the determining unit 32 selects one of the plurality of first elements as a comparison candidate. The above first element and one or more second elements of the comparison candidates among the plurality of second elements are respectively determined. The associating unit 33 determines the first based on the comparison between the information of one or more first elements of the comparison candidates among the plurality of first elements and the information of the one or more second elements of the comparison candidates among the plurality of second elements. Linking processing for linking the user information and the second user information is performed. As described above, the information processing apparatus 1 can efficiently perform the linking process because only some of the plurality of elements forming the user information are used for the linking process. For example, the processing load on the information processing device 1 can be reduced. Further, it is possible to suppress the cost required for the transmission processing of the second user information and the like.

  Further, the determination unit 32 determines the success probability of linking the first user information and the second user information, and determines the comparison candidate based on the determined success probability. Thereby, the linking process can be performed more efficiently.

  Further, the associating unit 33 performs the associating process while gradually increasing the comparison candidates in descending order of success probability. Thereby, for example, since the linking process can be performed after narrowing down the user information of the linking target, the load on the linking process can be reduced.

  Further, the determination unit 32 determines a comparison candidate based on the result of repeated comparison while increasing or decreasing the first element and the second element, respectively. This makes it possible to appropriately determine the comparison candidates.

  In addition, the determination unit 32, based on the result of the association of the association unit 33 with respect to a part of the second user information of the plurality of second user information, one or more first elements and one or more first elements used for comparison in the association process. Determine two elements. The associating unit 33 performs the associating process by comparing the one or more first elements determined by the determining unit 32 with the one or more second elements. Thereby, the accuracy of the tying process can be improved.

  The determining unit 32 also determines one or more second elements of the comparison candidates based on the comparison candidates determined in the past. This can reduce the load on the binding process.

  In addition, the control unit 12A includes a determination unit 32A and a tying unit 33A. The determining unit 32A sets a linking condition for selecting the first user information that is a candidate for the linking process from the plurality of first user information based on the result of the comparison between the plurality of first elements and the plurality of second elements. decide. The associating unit 33A performs the associating process based on the comparison between the first user information determined as a candidate for the associating process among the plurality of first user information and the plurality of second user information. Since only a part of the first user information is used for the linking process, the processing load on the information processing device 1A can be reduced.

  Further, the above-described information processing device 1 may be realized by a plurality of server computers, or may be realized by calling an external platform or the like with an API (Application Programming Interface) or network computing depending on the function. The configuration can be changed flexibly.

  Also, the above-mentioned "section (module, unit)" can be read as "means" or "circuit". For example, the information acquisition unit can be read as an information acquisition unit or an information acquisition circuit.

  Although the present invention has been described using the above embodiment, the technical scope of the present invention is not limited to the scope described in the above embodiment. It is apparent to those skilled in the art that various changes or improvements can be added to the above-described embodiment. It is also apparent from the scope of the claims that the embodiments added with such changes or improvements can be included in the technical scope of the present invention.

1, 1A Information processing device 2, 2 _{1 to} 2 _m-1 Server device 3, 3 _{1 to} 3 _n Terminal device 4 Network 10 Communication unit 11 Storage unit 12, 12A Control unit 20 Service information DB
21, 22 User information group table 23 Comparison candidate table 23 Linking information table 30 Service providing section 31 Information acquiring section 32, 32A determining section 33, 33A Linking section 34 Information output section

Claims (9)

Comparing a plurality of first elements constituting a part of the first user information of the first user information group and a plurality of second elements constituting a part of the second user information of the second user information group By determining the similarity between the first user information and the second user information, the first element that is determined to have little influence on the similarity among the plurality of first elements forming the first user information And one or more first elements excluding the second element that excludes the second element that is determined to have a small influence on the similarity among the plurality of second elements that constitute the second user information. A determination unit that determines the comparison candidate and stores information of the determined comparison candidate in a storage unit;
The degree of similarity between the first user information and the second user information is determined using the one or more first elements and the one or more second elements stored as the comparison candidates in the storage unit, and the determination is made. A linking unit for linking the first user information of the first user information group and the second user information of the second user information group based on the similarity. Processing equipment. Comparing a plurality of first elements constituting a part of the first user information of the first user information group and a plurality of second elements constituting a part of the second user information of the second user information group And a reference pair number that is the number of combinations of the first user information and the second user information that are determined to be information of the same user, and some first elements of the plurality of first elements The number of comparison pairs, which is the number of combinations of the first user information and the second user information, which are determined to be information of the same user by comparing a part of the second elements of the plurality of second elements calculated respectively bets, the ratio of the number the comparison pairs for the number of the reference pair and the success probability of linkage, based on the probability of success, excluding the effect is small element relative to the success probability of the plurality of first element One or more first elements and one or more second elements excluding elements of the plurality of second elements that have little influence on the success probability are determined as comparison candidates, and information of the determined comparison candidates is stored. A decision section to be stored in the section,
The degree of similarity between the first user information and the second user information is determined using the one or more first elements and the one or more second elements stored as the comparison candidates in the storage unit, and the determination is made. A linking unit for linking the first user information of the first user information group and the second user information of the second user information group based on the similarity. Processing equipment. The tying part is
The information processing apparatus according to claim 2, wherein the associating process is performed while gradually increasing the comparison candidates in order of increasing influence on the success probability. The determination unit is
The number of comparison pairs is obtained based on a result of repeated comparison while increasing or decreasing the first element and the second element as the part of the first element and the part of the second element , respectively. The information processing device according to claim 2 or 3 . The determination unit is
Based on the comparison candidates determined in the past, information processing apparatus according to any one of claims 1-4, characterized in that determining the one or more second element of the comparison candidate. An information processing method performed by a computer,
Comparing a plurality of first elements constituting a part of the first user information of the first user information group and a plurality of second elements constituting a part of the second user information of the second user information group By determining the similarity between the first user information and the second user information, the first element that is determined to have little influence on the similarity among the plurality of first elements forming the first user information And one or more first elements excluding the second element that excludes the second element that is determined to have a small influence on the similarity among the plurality of second elements that constitute the second user information. A determination step of determining as a comparison candidate and storing information of the determined comparison candidate in a storage unit;
The degree of similarity between the first user information and the second user information is determined using the one or more first elements and the one or more second elements stored as the comparison candidates in the storage unit, and the determination is made. A linking step of linking the first user information of the first user information group and the second user information of the second user information group based on the determined degree of similarity. Processing method. An information processing method performed by a computer,
Comparing a plurality of first elements constituting a part of the first user information of the first user information group and a plurality of second elements constituting a part of the second user information of the second user information group And a reference pair number that is the number of combinations of the first user information and the second user information that are determined to be information of the same user, and some first elements of the plurality of first elements The number of comparison pairs, which is the number of combinations of the first user information and the second user information, which are determined to be information of the same user by comparing a part of the second elements of the plurality of second elements calculated respectively bets, the ratio of the number the comparison pairs for the number of the reference pair and the success probability of linkage, based on the probability of success, excluding the effect is small element relative to the success probability of the plurality of first element One or more first elements and one or more second elements excluding elements of the plurality of second elements that have little influence on the success probability are determined as comparison candidates, and information of the determined comparison candidates is stored. A decision process to be stored in the department,
The degree of similarity between the first user information and the second user information is determined using the one or more first elements and the one or more second elements stored as the comparison candidates in the storage unit, and the determination is made. A linking step of linking the first user information of the first user information group and the second user information of the second user information group based on the determined degree of similarity. Processing method. Comparing a plurality of first elements constituting a part of the first user information of the first user information group and a plurality of second elements constituting a part of the second user information of the second user information group By determining the similarity between the first user information and the second user information, the first element that is determined to have little influence on the similarity among the plurality of first elements forming the first user information And one or more first elements excluding the second element that excludes the second element that is determined to have a small influence on the similarity among the plurality of second elements that constitute the second user information. A determination procedure of determining as a comparison candidate and storing information of the determined comparison candidate in a storage unit,
The degree of similarity between the first user information and the second user information is determined using the one or more first elements and the one or more second elements stored as the comparison candidates in the storage unit, and the determination is made. A linking procedure for linking the first user information of the first user information group and the second user information of the second user information group based on the similarity .
An information processing program that causes a computer to execute. Comparing a plurality of first elements constituting a part of the first user information of the first user information group and a plurality of second elements constituting a part of the second user information of the second user information group And a reference pair number that is the number of combinations of the first user information and the second user information that are determined to be information of the same user, and some first elements of the plurality of first elements The number of comparison pairs, which is the number of combinations of the first user information and the second user information, which are determined to be information of the same user by comparing a part of the second elements of the plurality of second elements calculated respectively bets, the ratio of the number the comparison pairs for the number of the reference pair and the success probability of linkage, based on the probability of success, excluding the effect is small element relative to the success probability of the plurality of first element One or more first elements and one or more second elements excluding elements of the plurality of second elements that have little influence on the success probability are determined as comparison candidates, and information of the determined comparison candidates is stored. Decision procedure to be stored in the department,
The degree of similarity between the first user information and the second user information is determined using the one or more first elements and the one or more second elements stored as the comparison candidates in the storage unit, and the determination is made. A linking procedure for linking the first user information of the first user information group and the second user information of the second user information group based on the similarity .
An information processing program that causes a computer to execute.

Images