JP2020149363A - Information processing apparatus, information processing method, and information processing program - Google Patents

Abstract

To efficiently utilize a plurality of information.SOLUTION: An information processing apparatus according to an embodiment of the present invention has an acquiring unit and a generating unit. The acquiring unit acquires a plurality of user identification information satisfying a predetermined condition and user information corresponding to the respective plurality of user identification information. The generating unit collects the plurality of user information acquired by the acquiring unit to thereby generate a learning data corresponding to a user indicated by the plurality of user identification information.SELECTED DRAWING: Figure 3

Description

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

Conventionally, a technique for authenticating a user using a plurality of pieces of information has been provided. For example, there is known a technique for authenticating a user based on a learning result learned based on a user's behavior history and input information on the user's behavior.

Japanese Unexamined Patent Publication No. 2017-134750

However, in the above-mentioned conventional technique, it is not always possible to efficiently use a plurality of pieces of information. For example, in the above-mentioned prior art, the user is only authenticated based on the learning result learned based on the user's behavior history and the input information regarding the user's behavior, and a plurality of information can be efficiently processed. It is not always available.

The present application has been made in view of the above, and an object of the present application is to provide an information processing device, an information processing method, and an information processing program capable of efficiently using a plurality of information.

The information processing device according to the present application includes a plurality of user identification information satisfying a predetermined condition, an acquisition unit that acquires user information corresponding to each of the plurality of user identification information, and a plurality of users acquired by the acquisition unit. It is characterized by including a generation unit that generates learning data corresponding to a user indicated by the plurality of user identification information by aggregating the information.

According to one aspect of the embodiment, there is an effect that a plurality of pieces of information can be efficiently used.

FIG. 1 is a diagram showing an example of a generation process executed by the information processing apparatus according to the embodiment. FIG. 2 is a diagram showing an example of a generation process for generating learning data executed by the information processing apparatus according to the embodiment. FIG. 3 is a diagram showing a configuration example of the information processing device according to the embodiment. FIG. 4 is a diagram showing an example of a credit score storage unit according to the embodiment. FIG. 5 is a flowchart showing an example of a flow of a learning model generation process executed by the information processing apparatus according to the embodiment. FIG. 6 is a flowchart showing an example of a flow of estimation processing executed by the information processing apparatus according to the embodiment. FIG. 7 is a diagram showing an example of a generation process for generating learning data executed by the information processing apparatus according to the modified example. FIG. 8 is a diagram showing an example of a generation process for generating learning data executed by the information processing apparatus according to the modified example. FIG. 9 is a diagram showing an example of a generation process for generating learning data executed by the information processing apparatus according to the modified example. FIG. 10 is a hardware configuration diagram showing an example of a computer that realizes the functions of the information processing device.

Hereinafter, the information processing apparatus, the information processing method, and the mode for carrying out the information processing program (hereinafter referred to as “the embodiment”) according to the present application will be described in detail with reference to the drawings. Note that this embodiment does not limit the information processing apparatus, information processing method, and information processing program according to the present application. In addition, each embodiment can be appropriately combined as long as the processing contents do not contradict each other. Further, in each of the following embodiments, the same parts are designated by the same reference numerals, and duplicate description is omitted.

[1. An example of the generation process shown by the information processing device]
An example of the generation process executed by the information processing apparatus 100 according to the embodiment will be described with reference to FIG. FIG. 1 is a diagram showing an example of a generation process executed by the information processing apparatus 100 according to the embodiment. Specifically, the information processing device 100 generates learning data corresponding to the user indicated by the plurality of user identification information by aggregating the user information corresponding to each of the plurality of user identification information satisfying a predetermined condition. ..

The predetermined condition referred to here is a condition that the user indicated by the plurality of user identification information is the same user, a user group to which a plurality of users having the same or similar attributes belong, a family member, a business company, or the like. Further, in the following, an example will be described as a case where a predetermined condition satisfies the condition of being the same user.

Further, the user identification information referred to here is user identification information that identifies a user who uses a service, a system, or the like. Further, the service and system referred to here may be any service or any system. For example, services and systems include financial services, portal sites such as search engines, game sites, travel reservation sites, Internet shopping sites, and the like.

Further, the plurality of user identification information referred to here includes a plurality of user identification information used by the same user in the same system. Further, the plurality of user identification information referred to here includes a plurality of user identification information used by the same user in different systems. For example, assume that the first system is system A. Further, it is assumed that the user identification information used in the first system is the first user identification information. Further, it is assumed that the second system is the system B. It is assumed that the user identification information used in the second system is the second user identification information. In this case, the user has a first user identification information corresponding to the system A and a second user identification information corresponding to the system B. Further, the plurality of user identification information referred to here includes a plurality of user identification information used by the same user in the same system or in a plurality of systems including different systems.

Further, for example, it is assumed that the first system is a financial service. Further, it is assumed that the user identification information used in the first system is the first user identification information. It is assumed that the second system is a portal site. Further, it is assumed that the user identification information used in the second system is the second user identification information. In this case, the user has a first user identification information that can be used in the financial service and a second user identification information that can be used in the portal site.

Further, in the following, an example will be described assuming that the first system is a financial service and the second system is a portal site. Further, an example will be described below assuming that the first user identification information used in the financial service is one and the second user identification information used in the portal site is plural.

The term "user" as used herein is a concept including one user (natural person), a family member, a group of users to which a plurality of users having the same or similar attributes belong, or a corporation such as a business company. Further, in the following, the user shall refer to one user.

As shown in FIG. 1, the information processing system 1 includes a financial server 20, a service server 30, and an information processing device 100. The financial server 20, the service server 30, and the information processing device 100 are communicably connected by wire or wirelessly via a predetermined communication network (not shown). The information processing system 1 shown in FIG. 1 may include a plurality of financial servers 20, a plurality of service servers 30, and a plurality of information processing devices 100.

The financial server 20 according to the embodiment is an information processing device that provides various financial services to users, and is realized by, for example, a server device or a cloud system. For example, the financial server 20 provides user registration information and information on user credit.

Further, for example, the financial server 20 is a server device used for network banking and payments made by various banks via a network. Further, for example, the financial server 20 manages the user's account, and according to the operation from the user via a predetermined network, the deposit amount, the deposit history, the withdrawal history, the debt balance notification, or various transfers. Etc. are realized.

The service server 30 according to the embodiment is a server device that provides various services to the user. For example, the service server 30 provides a portal service that provides a portal site customized for the user when a login is received from the user. Further, for example, the service server 30 includes a search service, a shopping service, an auction service, an information providing service (for example, a map information service, a navigation service, a news service, a weather forecast service), a task setting, and a schedule. You may provide various services such as a calendar service for registering.

The information processing device 100 according to the embodiment is realized by, for example, a server device or the like. Specifically, the information processing apparatus 100 acquires a plurality of user identification information satisfying a predetermined condition and user information corresponding to each of the plurality of user identification information. Then, the information processing device 100 aggregates a plurality of user information to generate learning data corresponding to the user indicated by the plurality of user identification information.

The user information here means, for example, the user's name, the user's age, the user's gender, the user's telephone number, the user's address, the user's email address, the user's annual income, the user's marriage history, the user's bankruptcy history, It is the number of times the user has browsed the content such as the weight of the user or the content such as the web page or the content for the application displayed on the browser such as the terminal device.

Further, the user information referred to here is information on the user's hobbies and tastes, information on the degree of credit of the user to financial services, the amount of income of the user, and the like. In the following, an example in which the information processing apparatus 100 estimates the credit score, which is information on the credit degree of the user regarding the financial service, will be described as the user information.

In the information processing device 100, whether or not the first user identification information used in the first system and the second user identification information used in the second system are used by the same user. Is determined to be the same user by the address, telephone number, e-mail address, etc. registered in association with each user identification information.

Further, in the information processing device 100, whether or not the first user identification information used in the first system and the second user identification information used in the second system are used by the same user. Is determined to be the same user based on the cookie information when the service is used, the terminal ID unique to the terminal, and the like.

Further, in the following, the first user identification information used in the financial service and the plurality of second user identification information used in the portal site will be described by giving an example as being used by the same user. To do.

Hereinafter, an example of the generation process by the information processing apparatus 100 will be described along the flow with reference to FIG.

First, as shown in FIG. 1, the information processing apparatus 100 acquires the user identification information and the user information from the financial server 20 (step S1). For example, in the example of FIG. 1, the information processing apparatus 100 acquires the user's gender, the user's marriage history, the bankruptcy history, and the user's credit score from the financial server 20 as user information.

Subsequently, the information processing apparatus 100 acquires a plurality of user identification information and user information corresponding to each of the plurality of user identification information (step S2). For example, in the example of FIG. 1, the information processing apparatus 100 acquires the user's gender, the user's marriage history, and the number of views from the service server 30 as user information.

Then, the information processing apparatus 100 aggregates a plurality of user information to generate learning data corresponding to the user indicated by the plurality of user identification information (step S3). Specifically, the information processing device 100 identifies the first user information corresponding to the first user identification information that identifies the user used in the first system and the user used in the second system. The second user identification information is a combination with the aggregated user information obtained by aggregating the second user information corresponding to each of the plurality of second user identification information satisfying a predetermined condition. It is generated as learning data corresponding to the user indicated by the information.

Here, a generation process for generating learning data will be described with reference to FIG. FIG. 2 is a diagram showing an example of a generation process for generating learning data executed by the information processing apparatus 100 according to the embodiment. Table TB1 is a table showing the relationship between the first user identification information and the plurality of second user identification information. For example, as shown in FIG. 2, the first user identification information is the user specified by "U1". Specifically, the user information corresponding to the first user identification information U1 is gender, marriage history, bankruptcy history, and credit score.

Further, for example, in the gender corresponding to the first user identification information U1, the score 1 indicates a male and the score 0 indicates a female. For example, since it has a gender score of "1" corresponding to the first user identification information U1, it indicates that it is a male.

Further, for example, in the marriage history corresponding to the first user identification information U1, a score of 1 indicates a marriage history, and a score of 0 indicates no marriage history. For example, since the score of the marriage history corresponding to the first user identification information U1 is "1", it indicates that the person has a marriage history.

Further, for example, in the bankruptcy history corresponding to the first user identification information U1, a score 1 indicates that there is a bankruptcy history, and a score 0 indicates that there is no bankruptcy history. For example, since the bankruptcy history score "0" corresponding to the first user identification information U1, it indicates that there is no bankruptcy history. Further, for example, the credit score corresponding to the first user identification information U1 is "0.9".

Further, for example, as shown in FIG. 2, the second user identification information is the user specified by "B1". Further, the user information corresponding to the second user identification information B1 is the gender, the marriage history, and the number of times the content is viewed.

Further, for example, in the gender corresponding to the second user identification information B1, the score 1 indicates a male and the score 0 indicates a female. For example, since it has a gender score of "1" corresponding to the second user identification information B1, it indicates that it is a male.

Further, for example, in the marriage history corresponding to the second user identification information B1, a score of 1 indicates a marriage history, and a score of 0 indicates no marriage history. For example, since the score of the marriage history corresponding to the second user identification information B1 is "1", it indicates that the person has a marriage history. Further, for example, the number of views corresponding to the second user identification information B1 is "20".

Further, as shown in FIG. 2, the second user identification information is the user specified by "B2". Specifically, the user information corresponding to the second user identification information B2 is the gender, the marriage history, and the number of times the content is viewed.

Further, for example, in the gender corresponding to the second user identification information B2, a score of 1 indicates a male and a score of 0 indicates a female. For example, since the gender score corresponding to the second user identification information B2 is "0", it indicates that the woman is a woman.

Further, for example, in the marriage history corresponding to the second user identification information B2, a score of 1 indicates a marriage history, and a score of 0 indicates no marriage history. For example, since the score of the marriage history corresponding to the second user identification information B2 is "1", it indicates that the person has a marriage history. Further, for example, the number of views corresponding to the second user identification information B2 is "10".

Further, as shown in FIG. 2, the second user identification information is the user specified by "B3". Specifically, the user information corresponding to the second user identification information B3 is the gender, the marriage history, and the number of times the content is viewed.

Further, for example, in the gender corresponding to the second user identification information B3, a score of 1 indicates a male and a score of 0 indicates a female. For example, since it has a gender score of "1" corresponding to the second user identification information B3, it indicates that it is a male.

Further, for example, in the marriage history corresponding to the second user identification information B3, a score of 1 indicates a marriage history, and a score of 0 indicates no marriage history. For example, since the score of the marriage history corresponding to the second user identification information B3 is "1", it indicates that the person has a marriage history. Further, for example, the number of views corresponding to the second user identification information B3 is "20".

In the example of FIG. 2, the information processing apparatus 100 aggregates a plurality of user information to generate learning data corresponding to the user indicated by the plurality of user identification information. More specifically, the information processing apparatus 100 generates learning data from aggregated user information aggregated based on the number of the same user information among a plurality of user information.

For example, as for gender, marriage history, and bankruptcy history, a large number of them shall be used as user information. Further, the number of views shall be the sum of the number of views for each user identification information as the user information. In this case, in the example of FIG. 2, since the information processing apparatus 100 has many male users in the user information corresponding to the second user identification information, it is assumed that the gender is male.

Further, it is assumed that the information processing apparatus 100 has a marriage history because the user information corresponding to the second user identification information has a lot of marriage history. Further, in the user information corresponding to the second user identification information, the information processing device 100 uses "50", which is the total number of views, as the number of views. As described above, the information processing device 100 is a combination of the first user information corresponding to the first user identification information and the aggregated user information in which the second user information corresponding to the second user identification information is aggregated. A table TB2 is generated as training data.

Returning to FIG. 1, an example of the generation process according to the embodiment will be described. The information processing device 100 generates a learning model M1 based on a credit score of a predetermined user and learning data of a predetermined user (step S4). For example, the information processing device 100 learns based on a predetermined user's credit score and a predetermined user's learning data by conventional techniques of machine learning methods such as logistic regression, SVM (Support Vector Machine), and deep learning. Generate model M1.

Further, the information processing device 100 is not limited to the above-mentioned processing example, and specifically, the information processing apparatus 100 corresponds to a positive example based on the number of positive examples associated with the user information corresponding to each of the plurality of user identification information. Generate training data to be processed. Further, specifically, the information processing apparatus 100 generates learning data corresponding to the negative examples based on the number of negative examples associated with the user information corresponding to each of the plurality of user identification information.

For example, the information processing apparatus 100 learns the learning data corresponding to the positive example with a credit score of "1.0". Further, the information processing apparatus 100 learns the learning data corresponding to the negative example with a credit score of "0.0".

Then, the information processing device 100 outputs the score of the user who is the prediction target when the learning data of the user who is the prediction target is input based on the learning model M1. Then, the information processing apparatus 100 calculates the credit score of the user by converting the score output from the learning model M1 into a probability value by using Sigmoid fitting or the like. In this way, the information processing device 100 estimates the credit score of the user to be predicted based on the learning model M1.

Then, the information processing device 100 refers to the user specified by the first user identification information "U100" (hereinafter, the user specified by the first user identification information "UN" is "user UN (N is an arbitrary numerical value)". ) ”) Is received from the financial server 20 (step S5).

Subsequently, the information processing apparatus 100 acquires the user identification information and the user information of the user U100 from the financial server 20 (step S6). For example, in the example of FIG. 1, the information processing apparatus 100 acquires the user's gender, the user's marriage history, and the bankruptcy history from the financial server 20 as user information.

Then, the information processing device 100 acquires the plurality of user identification information of the user U100 and the user information corresponding to each of the plurality of user identification information (step S7). For example, in the example of FIG. 1, the information processing apparatus 100 acquires the user's gender, the user's marriage history, and the number of views from the service server 30 as user information.

Subsequently, the information processing device 100 generates the learning data RC1 corresponding to the user U100 (step S8). For example, the information processing device 100 generates the learning data RC1 based on the learning data generation process shown in FIG.

Subsequently, the information processing apparatus 100 estimates the credit score of the user U100 by inputting the learning data RC1 corresponding to the user U100 into the learning model M1 (step S9). For example, in the example of FIG. 1, the information processing apparatus 100 inputs the learning data RC1 corresponding to the user U100 into the learning model M1, and the score is output from the learning model M1. Then, the information processing device 100 estimates the credit score of the user U100 by converting the score output by the learning model M1 into a probability value using sigmoid fitting or the like. Then, the information processing device 100 transmits the credit score of the user U100 to the financial server 20 (step S10).

As described above, the information processing apparatus 100 according to the embodiment acquires a plurality of user identification information satisfying a predetermined condition and user information corresponding to each of the plurality of user identification information. Then, the information processing device 100 aggregates a plurality of user information to generate learning data corresponding to the user indicated by the plurality of user identification information. As a result, the information processing device 100 according to the embodiment efficiently uses the plurality of information because it aggregates the plurality of user information to generate learning data corresponding to the user indicated by the plurality of user identification information. be able to.

This point will be described. Explaining with reference to the example of FIG. 1, the information processing apparatus 100 includes user information corresponding to the first user identification information and aggregated user information in which user information corresponding to a plurality of second user identification information is aggregated. Generate training data based on the combination. As a result, the information processing apparatus 100 can process the user information corresponding to the first user identification information used by the same user in association with the user information corresponding to the second user identification information. In this way, the information processing apparatus 100 does not need to learn the duplicate learning data, and can efficiently generate the learning model.

Further, since the information processing device 100 can generate a learning model by appropriately associating user information with each user, it is possible to generate a learning model that can output a highly accurate credit score. Further, since the information processing device 100 can learn a large amount of user information corresponding to a plurality of user identification information, it is possible to generate a learning model capable of outputting a highly accurate credit score. Further, since the information processing device 100 does not learn duplicate learning data, it is possible to generate a learning model in which the influence of data bias is reduced. Therefore, since the information processing device 100 can give an appropriate score to the user, the credit score of the user can be estimated with high accuracy.

[2. Information processing device configuration]
Next, the configuration of the information processing apparatus 100 according to the embodiment will be described with reference to FIG. FIG. 3 is a diagram showing a configuration example of the information processing device 100 according to the embodiment. As shown in FIG. 3, the information processing device 100 includes a communication unit 110, a storage unit 120, and a control unit 130.

(About communication unit 110)
The communication unit 110 is realized by, for example, a NIC (Network Interface Card) or the like. Then, the communication unit 110 is connected to the network by wire or wirelessly, and transmits / receives information between the financial server 20 and the service server 30.

(About storage unit 120)
The storage unit 120 is realized by, for example, a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory (Flash Memory), or a storage device such as a hard disk or an optical disk. The storage unit 120 has a learning model 121 and a credit score storage unit 122.

(About credit score storage unit 122)
The credit score storage unit 122 according to the embodiment stores information regarding the user's credit score. Here, FIG. 4 shows an example of the credit score storage unit 122 according to the embodiment. In the example shown in FIG. 4, the credit score storage unit 122 has items such as "second user identification information" and "credit score".

The "second user identification information" is an identifier that identifies a user. The "credit score" is a credit score associated with the "first user identification information", and is information regarding the credit score estimated based on the learning model. For example, in FIG. 4, the credit score of "BC1" identified by the second user identification information is "0.7".

(About control unit 130)
The control unit 130 is a controller, and for example, various programs (information processing programs) stored in a storage device inside the information processing device 100 by a CPU (Central Processing Unit), an MPU (Micro Processing Unit), or the like. (Corresponding to one example) is realized by executing RAM as a work area. Further, the control unit 130 is a controller, and is realized by, for example, an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

As shown in FIG. 3, the control unit 130 includes an acquisition unit 131, a reception unit 132, a generation unit 133, a learning unit 134, an estimation unit 135, and a transmission unit 136, and the information described below. Realize or execute the function or action of processing. The internal configuration of the control unit 130 is not limited to the configuration shown in FIG. 3, and may be another configuration as long as it is a configuration for performing information processing described later. Further, the connection relationship of each processing unit included in the control unit 130 is not limited to the connection relationship shown in FIG. 3, and may be another connection relationship.

(About acquisition unit 131)
The acquisition unit 131 acquires various types of information. Specifically, the acquisition unit 131 acquires a plurality of user identification information satisfying a predetermined condition and user information corresponding to each of the plurality of user identification information. For example, in the example of FIG. 1, the acquisition unit 131 acquires the gender of the user, the marriage history of the user, the bankruptcy history, and the credit score of the user from the financial server 20 as user information. Further, for example, in the example of FIG. 1, the acquisition unit 131 acquires the user's gender, the user's marriage history, and the number of views from the service server 30 as user information.

(About reception desk 132)
The reception unit 131 receives various requests. Specifically, the request for acquiring the credit score of the user is received from the financial server 20. For example, in the example of FIG. 1, the reception unit 131 receives the credit score of the user U100, which is the user to be predicted, from the financial server 20.

(About generator 133)
The generation unit 133 aggregates the plurality of user information acquired by the acquisition unit 131 to generate learning data corresponding to the user indicated by the plurality of user identification information. Specifically, the generation unit 133 identifies the first user information corresponding to the first user identification information that identifies the user used in the first system and the user used in the second system. The second user identification information is a combination of the second user identification information and the aggregated user information obtained by aggregating the second user information corresponding to each of the plurality of second user identification information satisfying a predetermined condition. Is generated as learning data corresponding to the user indicated by.

More specifically, the generation unit 133 generates learning data from the aggregated user information aggregated based on the number of the same user information among the plurality of user information acquired by the acquisition unit 131. For example, as for gender, marriage history, and bankruptcy history, a large number of them shall be used as user information. Further, the number of views shall be the sum of the number of views for each user identification information as the user information. In this case, in the example of FIG. 2, the generation unit 133 assumes that the gender is male because there are many males in the user information corresponding to the second user identification information.

Further, the generation unit 133 assumes that the user information corresponding to the second user identification information has a marriage history because there are many marriage histories. Further, in the user information corresponding to the second user identification information, the generation unit 133 sets “50”, which is the total number of views, as the number of views. As described above, the generation unit 133 is a combination of the first user information corresponding to the first user identification information and the aggregated user information in which the second user information corresponding to the second user identification information is aggregated. Table TB2 is generated as training data.

(About learning unit 134)
The learning unit 134 generates the learning model M1 based on the information on the credit of the predetermined user and the learning data of the predetermined user. For example, in the example of FIG. 1, the learning unit 134 learns based on a predetermined user's credit score and a predetermined user's learning data by a conventional technique of a machine learning method such as logistic regression, SVM, and deep learning. Generate model M1.

Further, for example, the learning unit 134 learns the learning data corresponding to the positive example with a credit score of "1.0". Further, the learning unit 134 learns the learning data corresponding to the negative example with a credit score of "0.0".

(About estimation unit 135)
The estimation unit 135 is based on a plurality of user identification information acquired by the acquisition unit 131 that satisfies a predetermined condition and aggregated user information that aggregates user information corresponding to each of the plurality of user identification information, and is used for a predetermined user. Estimate user information.

For example, in the example of FIG. 1, the estimation unit 135 estimates the credit score of the user U100 by inputting the learning data RC1 corresponding to the user U100 into the learning model M1. For example, in the example of FIG. 1, the estimation unit 135 inputs the learning data RC1 corresponding to the user U100 into the learning model M1, and the score is output from the learning model M1. Then, the estimation unit 135 estimates the credit score of the user U100 by converting the score output by the learning model M1 into a probability value using sigmoid fitting or the like. Then, the estimation unit 135 stores the credit score of the user U100 in the credit score storage unit 122.

(About transmitter 136)
The transmission unit 136 transmits various information. Specifically, the transmission unit 136 transmits the user's credit score stored in the credit score storage unit 122 to the user. For example, in the example of FIG. 1, the transmission unit 136 transmits the credit score of the user U100 to the financial server 20.

[3. Processing procedure (1) Generate learning model]
Next, the procedure of the learning model generation process executed by the information processing apparatus 100 according to the embodiment will be described with reference to FIG. FIG. 5 is a flowchart showing an example of a flow of a learning model generation process executed by the information processing apparatus 100 according to the embodiment.

As shown in FIG. 5, the acquisition unit 131 acquires the user identification information and the user information from the financial server 20 (step S101). Then, when the acquisition unit 131 has not acquired the user identification information and the user information (step S101; No), the acquisition unit 131 waits until the user identification information and the user information are acquired from the financial server 20.

On the other hand, when the user identification information and the user information are acquired from the financial server 20 (step S101; Yes), the acquisition unit 131 obtains the plurality of user identification information and the user information corresponding to each of the plurality of user identification information. Obtained from the service server 30 (step S102). Then, when the acquisition unit 131 does not acquire the plurality of user identification information and the user information corresponding to each of the plurality of user identification information (step S102; No), the generation unit 133 together with the plurality of user identification information. Wait until the user information corresponding to each of the plurality of user identification information is acquired from the service server 30.

On the other hand, when the acquisition unit 131 acquires the plurality of user identification information and the user information corresponding to each of the plurality of user identification information from the service server 30, the generation unit 133 has a plurality of users (step S102; Yes). By aggregating the information, learning data corresponding to a predetermined user indicated by a plurality of user identification information is generated (step S103).

For example, as for gender, marriage history, and bankruptcy history, a large number of them shall be used as user information. Further, the number of views shall be the sum of the number of views for each user identification information as the user information. In this case, in the example of FIG. 2, the generation unit 133 assumes that the gender is male because there are many males in the user information corresponding to the second user identification information.

Further, the generation unit 133 assumes that the user information corresponding to the second user identification information has a marriage history because there are many marriage histories. Further, in the user information corresponding to the second user identification information, the generation unit 133 sets “50”, which is the total number of views, as the number of views. As described above, the generation unit 133 is a combination of the first user information corresponding to the first user identification information and the aggregated user information in which the second user information corresponding to the second user identification information is aggregated. Table TB2 is generated as training data.

Then, the learning unit 134 generates a learning model based on the combination of the learning data corresponding to the predetermined user and the credit score of the predetermined user (step S104). For example, in the example of FIG. 1, the learning unit 134 learns based on a predetermined user's credit score and a predetermined user's learning data by a conventional technique of a machine learning method such as logistic regression, SVM, and deep learning. Generate model M1.

Further, for example, the learning unit 134 learns the learning data corresponding to the positive example with a credit score of "1.0". Further, the learning unit 134 learns the learning data corresponding to the negative example with a credit score of "0.0".

[4. Processing procedure (2) Estimate credit score]
Next, the procedure of the estimation process executed by the information processing apparatus 100 according to the embodiment will be described with reference to FIG. FIG. 5 is a flowchart showing an example of a flow of estimation processing executed by the information processing apparatus 100 according to the embodiment.

As shown in FIG. 5, the reception unit 132 receives a request for acquiring the credit score of the user U100 (step S201). Then, when the reception unit 131 has not received the user's credit score acquisition request (step S201; No), the acquisition unit 131 waits until the user's credit score acquisition request is received.

On the other hand, when the reception unit 131 receives the user's credit score acquisition request (step S201; Yes), the acquisition unit 131 acquires the user identification information of the user U100 and the user information from the financial server 20 (step S202). ). Then, when the acquisition unit 131 does not acquire the user identification information of the user U100 and the user information from the financial server 20 (step S202; No), the user identification information of the user U100 and the user information are acquired by the financial server 20. Wait until you get from.

On the other hand, when the acquisition unit 131 acquires the user identification information of the user U100 and the user information from the financial server 20 (step S202; Yes), each of the plurality of user identification information of the user U100 and the plurality of user identification information. The user information corresponding to is acquired from the service server 30 (step S203). Then, the generation unit 133 does not acquire the plurality of user identification information of the user U100 and the user information corresponding to each of the plurality of user identification information from the service server 30 (step S203; No). , Waits until the plurality of user identification information of the user U100 and the user information corresponding to each of the plurality of user identification information are acquired from the service server 30.

On the other hand, when the acquisition unit 131 acquires the plurality of user identification information of the user U100 and the user information corresponding to each of the plurality of user identification information from the service server 30, the generation unit 133 has a plurality of cases (step S203; Yes). By aggregating the user information of, the learning data corresponding to the user U100 indicated by the plurality of user identification information is generated (step S204).

Then, the estimation unit 135 estimates the user's credit score by inputting the learning data generated by the generation unit 133 into the learning model (step S205). For example, in the example of FIG. 1, the estimation unit 135 estimates the credit score of the user U100 by inputting the learning data RC1 corresponding to the user U100 into the learning model M1.

For example, in the example of FIG. 1, the estimation unit 135 inputs the learning data RC1 corresponding to the user U100 into the learning model M1, and the score is output from the learning model M1. Then, the estimation unit 135 estimates the credit score of the user U100 by converting the score output by the learning model M1 into a probability value using sigmoid fitting or the like.

Then, the transmission unit 136 transmits the credit score estimated by the estimation unit 135 to the financial server 20 (step S206). For example, in the example of FIG. 1, the transmission unit 136 transmits the credit score of the user U100 to the financial server 20.

[5. Modification example]
The information processing device 100 described above may be implemented in various different forms other than the above-described embodiment. Therefore, another embodiment of the information processing apparatus 100 will be described below.

[5-1. Different services and systems]
In the above embodiment, the plurality of user identification information possessed by the user has been described by way of example assuming that the first system is a financial service and the second system is a portal site, but the present invention is not limited to the above embodiment.

For example, the plurality of user identification information held by the user may be the first user identification information used on the game site and the second user identification information used on the portal site. Further, for example, the plurality of user identification information possessed by the user may be the first user identification information used on the Internet shopping site and the second user identification information used on the portal site. Further, for example, the plurality of user identification information possessed by the user may be the first user identification information used on the Internet shopping site and the second user identification information used on the travel reservation site.

[5-2. User identification information]
In the above embodiment, it is assumed that the plurality of user identification information possessed by the user is one first user identification information used in the financial service and the second plurality of user identification information used in the portal site. Although the description has been given with an example, the present invention is not limited to the above embodiment.

For example, the plurality of user identification information possessed by the user may include a plurality of first user identification information used in the financial service and one second user identification information used in the portal site. Further, for example, the plurality of user identification information possessed by the user may include a plurality of first user identification information used in the financial service and a plurality of second user identification information used in the portal site. ..

[5-3. User information (1)]
In the above embodiment, there is an example in which the generation unit 133 of the information processing device 100 uses a large number of genders and marital history as user information, and the generation unit 133 uses the total number of views for each user identification information as user information. However, the present invention is not limited to the above embodiment.

For example, suppose that the marriage history is predicted with high accuracy. In this case, the generation unit 133 may determine that there is no marriage history when there is no marriage history. Further, for example, the generation unit 133 may use a statistical value (for example, an average value or the like) of each viewing number, or a maximum value or a minimum value, as the number of viewings. Further, for example, the generation unit 133 may use a statistic (for example, an average value or the like) of each number of views, a maximum value or a minimum value, or the like as the number of views in increments of 10.

Further, specifically, the acquisition unit 131 acquires the user information including the information corresponding to the plurality of items as the user information corresponding to each of the plurality of user identification information, and the generation unit 133 acquires the user information including the information corresponding to the plurality of items for each item. Learning data is generated by aggregating the information corresponding to the items. More specifically, the generation unit 133 may generate learning data by aggregating the information corresponding to each item for each item. For example, in the case of gender, when the generation unit 133 generates learning data from aggregated user information aggregated based on the number of genders, it may not be the original gender. In this case, the generation unit 133 may generate learning data for men and learning data for women, respectively.

[5-4. User information (2)]
In the above embodiment, an example of the acquisition process in which the acquisition unit 131 of the information processing device 100 acquires the gender, marriage history, bankruptcy history, or number of views as user information has been described, but the acquisition process is not limited to the above. Specifically, the acquisition unit 131 may acquire the age of the user. Then, the generation unit 133 may generate the learning data corresponding to the user indicated by the plurality of user identification information by aggregating the plurality of user information. Further, in the following, an example of generating learning data by aggregating a plurality of second user information will be described.

For example, it is assumed that the age category of the user is less than 20 years old, 20 to 30 years old, 40 to 50 years old, and 60 years old or older. In addition, it is assumed that the user information has a large number of age categories of the user. In this case, in the example of FIG. 7, since the number of generation units 133 is large in the 20s to 30s, it is assumed that the age of the user is in the 20s to 30s.

For example, it is assumed that the age category of the user is less than 20 years old, 20 to 30 years old, 40 to 50 years old, and 60 years old or older. In addition, it is assumed that the intermediate value is the user information. In this case, in the example of FIG. 7, since the generation unit 133 has data in the 20s to 30s, the 40s to 50s, and the 60s and over, the user is in the 40s to 50s, which is an intermediate value. Suppose it is age.

For example, it is assumed that the age category of the user is less than 20 years old, 20 to 30 years old, 40 to 50 years old, and 60 years old or older. In addition, the average value shall be the user information. Further, it is assumed that the score of under 20 years old is 1, the score of 20 to 30 years old is 2, the score of 40 to 50 years old is 3, and the score of 60 years old or older is 4. In this case, in the example of FIG. 7, since the generation unit 133 has data in the 20s to 30s, the 40s to 50s, and the 60s and over, the average value is (2 × 2 + 3 × 1 + 4 × 1). Calculate as / 4 = 2.75. At this time, the generation unit 133 calculates the average value as 3 by rounding off the average value to an integer. From this, it is assumed that the generation unit 133 is the age of the user in his 40s to 50s.

For example, it is assumed that the age category of the user is less than 20 years old, 20 to 30 years old, 40 to 50 years old, and 60 years old or older. Further, it is assumed that the average value considering the weight is used as the user information. Further, it is assumed that the score of under 20 years old is 1, the score of 20 to 30 years old is 2, the score of 40 to 50 years old is 3, and the score of 60 years old or older is 4. Further, it is assumed that the weight between the 40s to 50s and the 60s and over is 0.5. In this case, in the example of FIG. 7, since the generation unit 133 has data in the 20s to 30s, the 40s to 50s, and the 60s and over, the weighted average value is set to (2 × 2 + 3 × 1 ×). Calculate as 0.5 + 4 × 1 × 0.5) / 4 = 2.5. At this time, the generation unit 133 calculates the average value as 3 by rounding off the average value to an integer. From this, it is assumed that the generation unit 133 is the age of the user in his 40s to 50s.

[5-5. When considering the update date and time of user information]
In the above embodiment, an example of the generation process in which the generation unit 133 of the information processing apparatus 100 generates learning data based on the user information in which the user information corresponding to each of the plurality of user identification information is aggregated has been described. Not limited to processing. Specifically, the generation unit 133 may generate learning data from aggregated user information that aggregates a plurality of user information based on the update date and time corresponding to each of the plurality of user information. Further, in the following, an example of generating learning data by aggregating a plurality of second user information will be described.

More specifically, the generation unit 133 learns from the user information aggregated based on the score corresponding to each of the plurality of user information and the weight calculated from the update date and time corresponding to each of the plurality of user information. It may be generated as data.

For example, in the example of FIG. 8, it is assumed that the date and time updated together with the score are stored in Table TB4. Further, it is assumed that the update date and time of January, February, March ... Are weights "1", "2", "3" ... In this case, in the example of FIG. 8, the generation unit 133 calculates the weighted average value of the user information 1 as (0 × 6 + 1 × 12 + 1 × 10) / (6 + 12 + 10) = 0.58. At this time, the generation unit 133 calculates 1 by rounding off the score of the user information 1 to an integer. Further, the generation unit 133 calculates the weighted average value of the user information 2 as (0 × 6 + 0 × 5 + 0 × 12) / (6 + 5 + 12) = 0.

Further, the generation unit 133 calculates the weighted average value of the user information 3 as (1 × 6 + 0 × 6 + 0 × 7) / (6 + 6 + 7) = 0.32. At this time, the generation unit 133 calculates 0 by rounding off the score of the user information 3 to an integer. Further, the generation unit 133 calculates the weighted average value of the user information 4 as (1 × 6 + 1 × 7 + 0 × 12) / (6 + 7 + 12) = 0.52. At this time, the generation unit 133 calculates 1 by rounding off the score of the user information 4 to an integer. In this way, the generation unit 133 may generate learning data by aggregating the scores.

As a result, the generation unit 133 of the information processing apparatus 100 according to the embodiment generates learning data from the user information that aggregates the plurality of user information based on the update date and time of the user information of any one of the plurality of user information. , Multiple information can be used efficiently.

In the example of FIG. 8, the update date and time is set to month, but it may be year, day, hour, minute, second, or a combination thereof. Further, in the above example, the generation unit 133 generates the learning data based on the table TB4, but the example is not limited to the above example. For example, the generation unit 133 may generate learning data by adopting the score of the latest update date and time.

[5-6. When considering the estimation accuracy of user information]
In the above embodiment, an example of the generation process in which the generation unit 133 of the information processing apparatus 100 generates learning data based on the user information in which the user information corresponding to each of the plurality of user identification information is aggregated has been described. Not limited to processing. Specifically, the generation unit 133 may generate learning data from the aggregated user information that aggregates the plurality of user information based on the estimation accuracy corresponding to each of the plurality of user information acquired by the acquisition unit 131. Good. Further, in the following, an example of generating learning data by aggregating a plurality of second user information will be described.

More specifically, the generation unit 133 aggregates user information based on the score corresponding to each of the plurality of user information and the weight calculated from the estimation accuracy of the score corresponding to each of the plurality of user information. It may be generated as training data from.
You may.

For example, in the example of FIG. 9, it is assumed that the estimation accuracy of the score is stored in Table TB5 together with the score indicating the user information. In this case, in the example of FIG. 9, the generation unit 133 sets the weighted average value of the user information 1 to (0 × 0.7 + 1 × 0.8 + 1 × 0.6) / (0.7 + 0.8 + 0.6) = 0. Calculated as .66. At this time, the generation unit 133 calculates 1 by rounding off the score of the user information 1 to an integer. Further, the generation unit 133 calculates the weighted average value of the user information 2 as (0 × 0.5 + 0 × 0.5 + 0 × 0.6) / (0.5 + 0.5 + 0.6) = 0.

Further, the generation unit 133 calculates the weighted average value of the user information 3 as (1 × 1 + 0 × 0.9 + 0 × 0.7) / (1 + 0.9 + 0.7) = 0.38. At this time, the generation unit 133 calculates 0 by rounding off the score of the user information 3 to an integer. Further, the generation unit 133 calculates the weighted average value of the user information 4 as (1 × 0.5 + 1 × 0.8 + 0 × 0.7) / (0.5 + 0.8 + 0.7) = 0.65. At this time, the generation unit 133 calculates 1 by rounding off the score of the user information 4 to an integer. In this way, the generation unit 133 may generate learning data by aggregating the scores.

As a result, the generation unit 133 of the information processing apparatus 100 according to the embodiment generates learning data from the user information that aggregates the plurality of user information based on the estimation accuracy of the user information of any one of the plurality of user information. , Multiple information can be used efficiently.

In the above example, the generation unit 133 generates the learning data based on the table TB5, but the example is not limited to the above example. For example, the generation unit 133 may generate the learning data by adopting the score having the highest estimation accuracy.

[5-7. Other learning data]
In the above embodiment, the generation unit 133 of the information processing apparatus 100 shows an example in which the correct answer data aggregates the learning data in which the correct answer data is classified into two values of the positive example and the negative example, but the correct answer data is a plurality of three values or more. Learning data that can take values may be aggregated. Specifically, the generation unit 133 may generate learning data from aggregated user information that is user information that is correct answer data and is aggregated based on the number of the same user information. For example, the generation unit 133 generates learning data from aggregated user information that includes information on the income amount as correct answer data and is aggregated based on the same number of income amounts among a plurality of user information. You may.

For example, it is assumed that the user information includes the user gender, marriage history, bankruptcy history, and income amount. Further, it is assumed that the user information corresponding to the user A is the user information A. Further, it is assumed that the user information corresponding to the user B is the user information B. Further, it is assumed that the user information corresponding to the user C is the user information C. Further, it is assumed that the income amount included in the user information A is "250,000 yen". Further, it is assumed that the income amount included in the user information B is "200,000 yen". Further, it is assumed that the amount of income included in the user information C is "250,000 yen". In this case, since the generation unit 133 has a large number of income amounts of "250,000 yen", the learning data may be generated assuming that the income amount included in the aggregated user information is "250,000 yen".

Further, in the above example, the generation unit 133 has given an example of generating learning data from aggregated user information aggregated based on the number of income amounts that are the same, but the present invention is not limited to the above generation process. For example, the generation unit 133 may generate learning data from the aggregated user information aggregated based on the average value of the income amount included in each of the plurality of user information. Further, for example, the generation unit 133 may generate learning data from the aggregated user information aggregated based on the maximum value among the income amounts included in each of the plurality of user information. Further, for example, the generation unit 133 may generate learning data from the aggregated user information aggregated based on the minimum value among the income amounts included in each of the plurality of user information.

As a result, the learning unit 134 of the information processing apparatus 100 according to the embodiment generates learning data from the aggregated user information aggregated based on the number of the same user information among the plurality of user information, so that the plurality of information Can be used efficiently.

In the above processing example, the generation unit 133 shows an example in which the learning data is generated from the aggregated user information that is the user information that is the correct answer data and is aggregated based on the number of the same user information. It does not have to be limited to. Specifically, the generation unit 133 may generate learning data corresponding to each class by aggregating the classes associated with each of the plurality of user information. More specifically, the generation unit 133 generates learning data corresponding to each class based on the number of the same classes among the classes associated with the user information corresponding to each of the plurality of user identification information. It may be generated.

For example, the user information A to D are assumed to be user information corresponding to a family composed of three people, a father, a mother, and a child. Further, it is assumed that the class corresponding to the user information A is "0". Here, class "0" indicates a father. Further, it is assumed that the class corresponding to the user information B is "0". Further, it is assumed that the class corresponding to the user information C is "1". Here, class "1" indicates a mother. Further, it is assumed that the class corresponding to the user information D is "2". Here, the class "2" indicates a child. In addition, it is assumed that a class having a large number of the same classes is a class corresponding to aggregated user information. In this case, since the generation unit 133 has a large number of classes "0" among the user information A to D, the learning data is assumed that the class corresponding to the aggregated user information in which the user information A to D is aggregated is "0". To generate. Further, the association between the user information and the class in the above example is realized by, for example, multi-class classification by a conventional technique such as Naive Bayes.

[5-8. Positive or negative example]
In the above embodiment, when the learning unit 134 of the information processing apparatus 100 has a large number of positive or negative examples corresponding to the combination of the first user identification information and the second user identification information, the first user Although an example of the learning process for learning the learning data corresponding to the combination of the identification information and the second user identification information as a positive example or a negative example has been described, the present invention is not limited to the above learning process.

For example, the learning unit 134 is a characteristic combination, and the first user identification information is based on a positive example or a negative example corresponding to the combination of the first user identification information and the second user identification information. The learning data corresponding to the combination of the second user identification information and the second user identification information may be learned as a positive example or a negative example.

Further, for example, the learning unit 134 is a positive example or a negative example in consideration of the update information of the user information or the estimation accuracy of the user information, and corresponds to the combination of the first user identification information and the second user identification information. Learning data corresponding to the combination of the first user identification information and the second user identification information may be learned as a positive example or a negative example based on the attached positive example or negative example.

[6. Hardware configuration]
Further, the financial server 20, the service server 30, and the information processing apparatus 100 according to the above-described embodiment are realized by, for example, a computer 1000 having a configuration as shown in FIG. Hereinafter, the information processing device 100 will be described as an example. FIG. 10 is a hardware configuration diagram showing an example of a computer 1000 that realizes the functions of the information processing device 100. The computer 1000 includes a CPU 1100, a RAM 1200, a ROM (Read Only Memory) 1300, an HDD (Hard Disk Drive) 1400, a communication interface (I / F) 1500, an input / output interface (I / F) 1600, and a media interface (I / F). ) Has 1700.

The CPU 1100 operates based on a program stored in the ROM 1300 or the HDD 1400, and controls each part. The ROM 1300 stores a boot program executed by the CPU 1100 when the computer 1000 is started, a program that depends on the hardware of the computer 1000, and the like.

The HDD 1400 stores a program executed by the CPU 1100, data used by such a program, and the like. The communication interface 1500 receives data from another device via the network N and sends it to the CPU 1100, and the CPU 1100 transmits the data generated by the CPU 1100 via the network N to the other device.

The CPU 1100 controls an output device such as a display or a printer, and an input device such as a keyboard or a mouse via the input / output interface 1600. The CPU 1100 acquires data from the input device via the input / output interface 1600. Further, the CPU 1100 outputs the data generated via the input / output interface 1600 to the output device.

The media interface 1700 reads a program or data stored in the recording medium 1800 and provides the program or data to the CPU 1100 via the RAM 1200. The CPU 1100 loads the program from the recording medium 1800 onto the RAM 1200 via the media interface 1700, and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or PD (Phase change rewritable Disk), a magneto-optical recording medium such as MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory. And so on.

For example, when the computer 1000 functions as the information processing device 100 according to the embodiment, the CPU 1100 of the computer 1000 realizes the function of the control unit 130 by executing the program loaded on the RAM 1200. Further, the data in the storage unit 120 is stored in the HDD 1400. The CPU 1100 of the computer 1000 reads and executes these programs from the recording medium 1800, but as another example, these programs may be acquired from another device via the network N.

[7. Others]
Further, among the processes described in the above-described embodiments and modifications, all or part of the processes described as being automatically performed can be manually performed, or the processes described as being manually performed. It is also possible to automatically perform all or part of the processed processing by a known method. In addition, the processing procedure, specific name, and information including various data and parameters shown in the above document and drawings can be arbitrarily changed unless otherwise specified. For example, the various information shown in each figure is not limited to the illustrated information.

Further, each component of each of the illustrated devices is a functional concept, and does not necessarily have to be physically configured as shown in the figure. That is, the specific form of distribution / integration of each device is not limited to the one shown in the figure, and all or part of the device is functionally or physically distributed / physically in arbitrary units according to various loads and usage conditions. It can be integrated and configured.

In addition, the above-described embodiments and modifications can be appropriately combined as long as the processing contents do not contradict each other.

Further, the above-mentioned "section, module, unit" can be read as "means" or "circuit". For example, the generation unit can be read as a generation means or a generation circuit.

[8. effect〕
As described above, the information processing apparatus 100 according to the embodiment includes an acquisition unit 131 and a generation unit 133. The acquisition unit 131 acquires a plurality of user identification information satisfying a predetermined condition and user information corresponding to each of the plurality of user identification information. The generation unit 133 aggregates the plurality of user information acquired by the acquisition unit 131 to generate learning data corresponding to the user indicated by the plurality of user identification information.

As a result, the information processing device 100 according to the embodiment efficiently uses the plurality of information because it aggregates the plurality of user information to generate learning data corresponding to the user indicated by the plurality of user identification information. be able to.

Further, in the information processing apparatus 100 according to the embodiment, the acquisition unit 131 is the first user information corresponding to the first user identification information for identifying the user used in the first system, and the second system. The second user identification information that identifies the user to be used, and the second user information corresponding to each of the plurality of second user identification information satisfying a predetermined condition, is acquired, and the generation unit 133 A combination of the first user information acquired by the acquisition unit 131 and the aggregated user information obtained by aggregating a plurality of second user information is generated as learning data corresponding to the user indicated by the second user identification information.

As a result, the information processing apparatus 100 according to the embodiment corresponds to the user indicated by the second user identification information by combining the first user information and the aggregated user information obtained by aggregating the plurality of second user information. Since it is generated as learning data, a plurality of pieces of information can be used efficiently.

Further, in the information processing apparatus 100 according to the embodiment, the generation unit 133 is an aggregated user that the generation unit 133 aggregates based on the number of the same user information among the plurality of user information acquired by the acquisition unit 131. Generate learning data from the information.

As a result, the information processing device 100 according to the embodiment generates learning data from the aggregated user information aggregated based on the number of the same user information among the plurality of user information, so that the plurality of information can be efficiently generated. It can be used.

Further, in the information processing apparatus 100 according to the embodiment, the generation unit 133 uses the aggregated user information that aggregates the plurality of user information based on the update date and time corresponding to each of the plurality of user information acquired by the acquisition unit 131. Generate training data.

As a result, the information processing device 100 according to the embodiment generates learning data from the aggregated user information that aggregates the plurality of user information based on the update date and time corresponding to each of the plurality of user information. It can be used efficiently.

Further, in the information processing apparatus 100 according to the embodiment, the generation unit 133 uses the aggregated user information obtained by aggregating the plurality of user information based on the estimation accuracy corresponding to each of the plurality of user information acquired by the acquisition unit 131. Generate training data.

As a result, the information processing device 100 according to the embodiment generates learning data from the aggregated user information that aggregates the plurality of user information based on the estimation accuracy corresponding to each of the plurality of user information. It can be used efficiently.

Further, in the information processing apparatus 100 according to the embodiment, the generation unit 133 is positive based on the number of positive examples associated with the user information corresponding to each of the plurality of user identification information acquired by the acquisition unit 131. Generate training data corresponding to the example.

As a result, the information processing device 100 according to the embodiment generates learning data corresponding to the positive examples based on the number of positive examples associated with the user information corresponding to each of the plurality of user identification information. Multiple pieces of information can be used efficiently.

Further, in the information processing apparatus 100 according to the embodiment, the generation unit 133 is negative based on the number of negative examples associated with the user information corresponding to each of the plurality of user identification information acquired by the acquisition unit 131. Generate training data corresponding to the example.

As a result, the information processing device 100 according to the embodiment generates learning data corresponding to the negative examples based on the number of negative examples associated with the user information corresponding to each of the plurality of user identification information. Multiple pieces of information can be used efficiently.

Further, in the information processing apparatus 100 according to the embodiment, the acquisition unit 131 acquires user information including information corresponding to a plurality of items as user information corresponding to each of the plurality of user identification information, and the generation unit 133 , Learning data is generated by aggregating the information corresponding to the item for each item.

As a result, the information processing apparatus 100 according to the embodiment can generate learning data by aggregating the information corresponding to the items for each item, so that a plurality of information can be efficiently used.

Further, in the information processing apparatus 100 according to the embodiment, the generation unit 133 aggregates user information for each user information in which the information corresponding to a specific item is the same among the plurality of items acquired by the acquisition unit 131. Generate training data from.

As a result, the information processing device 100 according to the embodiment generates learning data from aggregated user information aggregated for each user information in which the information corresponding to a specific item is the same among the plurality of items, so that the plurality of information Can be used efficiently.

Further, in the information processing device 100 according to the embodiment, the acquisition unit 131 includes, as a predetermined condition, the same user as the user indicated by the plurality of user identification information, a user group to which a plurality of users having the same or similar attributes belong, and a family member. Alternatively, when the condition of being a business company is satisfied, a plurality of user identification information and user information corresponding to each of the plurality of user identification information are acquired.

As a result, the information processing device 100 according to the embodiment is a user group, a family member, or a business company to which the users indicated by the plurality of user identification information belong to the same user, or a plurality of users having the same or similar attributes, as predetermined conditions. When a certain condition is satisfied, a plurality of user identification information and user information corresponding to each of the plurality of user identification information are acquired, so that the plurality of information can be efficiently used.

Further, as described above, the information processing apparatus 100 according to the embodiment includes an acquisition unit 131 and an estimation unit 135. The acquisition unit 131 acquires a plurality of user identification information satisfying a predetermined condition and user information corresponding to each of the plurality of user identification information. The estimation unit 135 estimates the user information of a predetermined user based on the aggregated user information obtained by aggregating a plurality of user information acquired by the acquisition unit 131.

As a result, the information processing device 100 according to the embodiment estimates the user information of a predetermined user based on the aggregated user information that aggregates the plurality of user information, so that the plurality of information can be efficiently used. ..

Although some of the embodiments of the present application have been described in detail with reference to the drawings, these are examples, and various modifications are made based on the knowledge of those skilled in the art, including the embodiments described in the disclosure column of the invention. It is possible to carry out the present invention in other improved forms.

1 Information processing system 20 Financial server 30 Service server 100 Information processing device 110 Communication unit 120 Storage unit 121 Learning model 122 Credit score storage unit 130 Control unit 131 Acquisition unit 132 Reception unit 133 Generation unit 134 Learning unit 135 Estimating unit 136 Transmission unit

Claims (15)

A plurality of user identification information satisfying a predetermined condition, an acquisition unit for acquiring user information corresponding to each of the plurality of user identification information, and an acquisition unit.
A generation unit that generates learning data corresponding to a user indicated by the plurality of user identification information by aggregating a plurality of user information acquired by the acquisition unit.
An information processing device characterized by being equipped with. The acquisition unit
The first user information corresponding to the first user identification information for identifying the user used in the first system and the second user identification information for identifying the user used in the second system. Acquire the second user information corresponding to each of the plurality of second user identification information satisfying a predetermined condition, and obtain the second user information.
The generator
A combination of the first user information acquired by the acquisition unit and the aggregated user information obtained by aggregating the plurality of second user information is generated as learning data corresponding to the user indicated by the second user identification information. To do,
The information processing apparatus according to claim 1. The generator
The learning data is generated from the aggregated user information aggregated based on the number of the same user information among the plurality of user information acquired by the acquisition unit.
The information processing apparatus according to claim 1 or 2. The generator
The learning data is generated from the aggregated user information that aggregates the plurality of user information based on the update date and time corresponding to each of the plurality of user information acquired by the acquisition unit.
The information processing apparatus according to any one of claims 1 to 3. The generator
The learning data is generated from the aggregated user information that aggregates the plurality of user information based on the estimation accuracy corresponding to each of the plurality of user information acquired by the acquisition unit.
The information processing device according to any one of claims 1 to 4, wherein the information processing device is characterized. The generator
The learning data corresponding to the positive example is generated based on the number of positive examples associated with the user information corresponding to each of the plurality of user identification information acquired by the acquisition unit.
The information processing device according to any one of claims 1 to 5, wherein the information processing device is characterized. The generator
The learning data corresponding to the negative example is generated based on the number of negative examples associated with the user information corresponding to each of the plurality of user identification information acquired by the acquisition unit.
The information processing device according to any one of claims 1 to 6, wherein the information processing device is characterized. The acquisition unit
As user information corresponding to each of the plurality of user identification information, user information including information corresponding to a plurality of items is acquired.
The generator
The learning data is generated by aggregating the information corresponding to the item for each item.
The information processing device according to any one of claims 1 to 7, wherein the information processing device is characterized. The generator
Among the plurality of items acquired by the acquisition unit, the learning data is generated from the aggregated user information aggregated for each user information in which the information corresponding to the specific item is the same.
The information processing apparatus according to claim 8. The acquisition unit
When the predetermined condition is that the user indicated by the plurality of user identification information is the same user, a user group to which a plurality of users having the same or similar attributes belong, a family member, or a business company, the plurality of users. Acquire the identification information and the user information corresponding to each of the plurality of user identification information.
The information processing device according to any one of claims 1 to 9, wherein the information processing device is characterized. A plurality of user identification information satisfying a predetermined condition, an acquisition unit for acquiring user information corresponding to each of the plurality of user identification information, and an acquisition unit.
An estimation unit that estimates user information of a predetermined user based on aggregated user information that aggregates a plurality of user information acquired by the acquisition unit.
An information processing device characterized by being equipped with. Information processing method executed by a computer
A plurality of user identification information satisfying a predetermined condition, an acquisition process for acquiring user information corresponding to each of the plurality of user identification information, and an acquisition process.
A generation step of generating learning data corresponding to a user indicated by the plurality of user identification information by aggregating a plurality of user information acquired by the acquisition step, and a generation step.
An information processing method characterized by including. A plurality of user identification information satisfying a predetermined condition, an acquisition procedure for acquiring user information corresponding to each of the plurality of user identification information, and an acquisition procedure.
A generation procedure for generating learning data corresponding to a user indicated by the plurality of user identification information by aggregating a plurality of user information acquired by the acquisition procedure, and a generation procedure.
An information processing program characterized by having a computer execute. Information processing method executed by a computer
A plurality of user identification information satisfying a predetermined condition, an acquisition process for acquiring user information corresponding to each of the plurality of user identification information, and an acquisition process.
An estimation process that estimates user information of a predetermined user based on aggregated user information that aggregates a plurality of user information acquired by the acquisition process, and an estimation process.
An information processing method characterized by including. A plurality of user identification information satisfying a predetermined condition, an acquisition procedure for acquiring user information corresponding to each of the plurality of user identification information, and an acquisition procedure.
An estimation procedure for estimating user information of a predetermined user based on aggregated user information obtained by aggregating a plurality of user information acquired by the acquisition procedure, and an estimation procedure.
An information processing program characterized by having a computer execute.

Images