JP7634499B2 - Information processing system, method and program - Google Patents

Description

This disclosure relates to technology for supporting evaluations such as calculating scores for users.

Conventionally, a judgment device has been proposed that includes a user information acquisition unit that acquires behavioral information indicating the user's behavior, and a creditworthiness judgment unit that judges the user's creditworthiness regarding his or her future ability to repay loans based on the behavioral information (see Patent Document 1). Also, a system has been proposed that determines whether or not to display a user score depending on the intimacy between users (see, for example, Patent Document 2).

JP 2021-174039 A JP 2020-129228 A

Technology has been proposed to calculate a user score that represents a user's trustworthiness, etc., based on the user's behavioral history. However, there are problems with this technology, such as the inability to calculate a user score or the accuracy of the calculated user score being insufficient when information about the target user is missing or the information is unreliable.

In view of the above problems, the present disclosure aims to realize evaluation such as calculation of a user score, or to improve the accuracy of the evaluation, even when information on the target user is missing or the information is unreliable.

An example of the present disclosure is an information processing system including a reference user identification means for identifying a reference user having a relationship with a target user, an attribute generation means for generating corresponding attribute data of the target user based on attribute data of the reference user identified for the target user, an attribute completion means for completing a corresponding attribute data group of the target user based on at least a portion of the generated corresponding attribute data of the target user, and a user score estimation means for estimating a user score to be set for the target user based on the completed corresponding attribute data group of the target user.

The present disclosure can be understood as an information processing device, a system, a method executed by a computer, or a program executed by a computer. The present disclosure can also be understood as such a program recorded on a recording medium readable by a computer or other device, machine, etc. Here, a recording medium readable by a computer, etc. refers to a recording medium that stores information such as data and programs through electrical, magnetic, optical, mechanical, or chemical action and can be read by a computer, etc.

According to the present disclosure, even when information on a target user is missing or the information is unreliable, it is possible to perform evaluation such as calculating a user score, or to improve the accuracy of the evaluation.

1 is a schematic diagram showing a configuration of an information processing system according to an embodiment. 1 is a diagram illustrating an outline of a functional configuration of an information processing device according to an embodiment. FIG. 10 is a diagram illustrating an example in which IP address data values are common in the embodiment. FIG. 4 is a diagram illustrating an example of graph data according to the embodiment. FIG. 10 is a diagram illustrating an example in which address data values are common in the embodiment. FIG. 4 is a diagram illustrating an example of graph data according to the embodiment. FIG. 10 is a diagram illustrating an example in which the value of credit card number data is common in the embodiment. FIG. 4 is a diagram illustrating an example of graph data according to the embodiment. FIG. 4 is a diagram illustrating an example of graph data according to the embodiment. FIG. 2 is a diagram illustrating an example of a cluster according to an embodiment. FIG. 13 is a diagram illustrating an example of visualization of classification according to the embodiment. FIG. 1 illustrates an example of determining relationship strength (closeness score) using a machine learning model according to an embodiment. FIG. 2 is a diagram illustrating a simplified concept of a decision tree of a machine learning model employed in an embodiment. 1 is a flowchart illustrating a flow of a machine learning process according to an embodiment. 11 is a flowchart showing a flow of a user score estimation process according to the embodiment.

Below, an embodiment of an information processing device, method, and program according to the present disclosure will be described with reference to the drawings. However, the embodiment described below is merely an example of an embodiment, and the information processing device, method, and program according to the present disclosure are not limited to the specific configuration described below. In implementing the present disclosure, a specific configuration according to the embodiment may be appropriately adopted, and various improvements and modifications may be made.

In this embodiment, an embodiment will be described in which the technology according to the present disclosure is implemented for a user score management system that manages a user score that indicates some measure related to a user (e.g., credit, etc.). However, the technology according to the present disclosure can be widely used as a technology for estimating a user score, and the application of the present disclosure is not limited to the examples shown in the embodiment.

<System Configuration>
1 is a schematic diagram showing the configuration of an information processing system according to this embodiment. In the system according to this embodiment, an information processing device 1 and one or more service providing systems 5 are connected to each other so as to be able to communicate with each other. A user is a user of a service provided by the service providing system 5, and receives the service by accessing the service providing system 5 from a user terminal.

The information processing device 1 is a computer including a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a storage device 14 such as an EEPROM (Electrically Erasable and Programmable Read Only Memory) or a HDD (Hard Disk Drive), a communication unit 15 such as a NIC (Network Interface Card), etc. However, the specific hardware configuration of the information processing device 1 can be omitted, replaced, or added as appropriate depending on the embodiment. Furthermore, the information processing device 1 is not limited to a device consisting of a single housing. The information processing device 1 may be realized by multiple devices using so-called cloud or distributed computing technology, etc.

The information processing device 1 manages a user score for each user and provides the user score to the service providing system 5. The service providing system 5 can customize services for the target user according to the user score provided by the information processing device 1.

The service providing system 5 is a computer equipped with a CPU, ROM, RAM, a storage device, a communication unit, an input device, an output device, etc. (not shown). Furthermore, these systems and terminals are not limited to devices consisting of a single housing. These systems and terminals may be realized by multiple devices using so-called cloud or distributed computing technology, etc.

In the system according to this embodiment, an electronic commerce system 40, a golf course reservation system 42, a travel reservation system 44, and a card management system 46 are connected to each other so that they can communicate with each other as the service providing system 5. However, the services provided by the service providing system 5 are not limited to the examples in this embodiment. The services provided by the service providing system 5 may be, for example, a map information service, a credit card/deferred payment service, an electronic money payment service, an online shopping service, an online reservation service, an operation center service, etc. Note that "deferred payment" is not limited to services such as Buy Now, Pay Later (BNPL), but includes the purchase of any product/service with deferred payment.

The service providing system 5 notifies the information processing device 1 of a group of attribute data of the user acquired from the user when the service is provided. The information processing device 1 can also access the service providing system 5 to acquire user attribute data registered in the system for multiple users including the target user, and include the data in the group of attribute data. Here, the user attribute data includes account data, which is information about the user who uses the system, and service usage history data by the user. The content of the service usage history data varies depending on the content of the service, and may include, for example, historical data of the user's location information, payment history data of credit card usage amount/postpaid payment usage amount, electronic money usage history data, transaction history data, reservation history data, operation history data for the user from the operation center, and frequently visited places of stay identified based on historical data of location information. The account data includes, for example, a user ID, name data, address data, age data, gender data, telephone number data, mobile phone number data, credit card number data, IP address data, school data, and workplace data.

The user ID is, for example, the identification information of the user in the computer system. The name data is, for example, data indicating the name (first name and surname) of the user. The address data is, for example, data indicating the address of the user. If the computer system is an electronic commerce system 40, the address data may indicate the address to which the product purchased by the user is to be delivered. The age data is, for example, data indicating the age of the user. The gender data is, for example, data indicating the gender of the user. The telephone number data is, for example, data indicating the telephone number of the user. The mobile phone number data is, for example, data indicating the mobile phone number of the user. The credit card number data is, for example, data indicating the card number of the credit card used by the user for payment in the computer system. The IP address data is, for example, data indicating the IP address of the computer used by the user (for example, the IP address of the sender). The school data is, for example, data indicating the school (name of educational institution, address, etc.) of the user if the user is a student. The workplace data is data that indicates the user's workplace (company name, address, etc.) if the user is a working member of society, for example.

FIG. 2 is a diagram showing an outline of the functional configuration of the information processing device 1 according to this embodiment. The information processing device 1 functions as an information processing device including a graph data generating unit 21, a reference user identifying unit 22, a relationship identifying unit 23, a relationship strength determining unit 24, an attribute selecting unit 25, an attribute generating unit 26, an attribute complementing unit 27, a user score estimating unit 28, and a machine learning unit 29, by a program recorded in the storage device 14 being read into the RAM 13 and executed by the CPU 11, which controls each piece of hardware included in the information processing device 1. Note that in this embodiment and other embodiments described later, each function included in the information processing device 1 is executed by the CPU 11, which is a general-purpose processor, but some or all of these functions may be executed by one or more dedicated processors.

The graph data generating unit 21 generates graph data (social graph network) showing the relationships between users by identifying pairs of users who are related to each other based on the attribute data groups of each of the users. More specifically, the graph data generating unit 21 generates graph data including, for example, node data 50 corresponding to each of multiple users including the target user, and link data 52 corresponding to pairs of users who are related to each other (see Figures 4, 6, 8, and 9). The graph generating unit 21 predicts and creates implicit links between users by learning (representation learning, relationship learning, embedding learning, knowledge graph embedding) a user relationship graph consisting of nodes (users) connected by explicit links. At this time, the graph generating unit 21 may perform the learning based on a known embedding model or an extension thereof as appropriate.

For example, as shown in FIG. 3, suppose that a group of attribute data for user A is registered in electronic commerce system 40. Also, suppose that a group of attribute data for user B is registered in golf course reservation system 42. Also, suppose that attribute data for user C is registered in travel reservation system 44. And suppose that the value of the IP address data for user A registered in electronic commerce system 40, the value of the IP address data for user B registered in golf course reservation system 42, and the value of the IP address data for user C registered in travel reservation system 44 are the same.

In this case, as shown in FIG. 4, the graph data generation unit 21 generates graph data including node data 50a associated with user A, node data 50b associated with user B, node data 50c associated with user C, link data 52a indicating that user A is related to user B, link data 52b indicating that user A is related to user C, and link data 52c indicating that user B is related to user C. It is presumed that users with the same IP address use the same computer or share a global address in the same residence or workplace. For this reason, in this embodiment, such users are associated with each other.

For example, suppose that attribute data groups for user D, user E, and user F are registered in electronic commerce system 40, as shown in FIG. 5. Also, suppose that the address data values of user D, user E, and user F registered in electronic commerce system 40 are the same.

In this case, as shown in FIG. 6, the graph data generation unit 21 generates graph data including node data 50d associated with user D, node data 50e associated with user E, node data 50f associated with user F, link data 52d indicating that user D is related to user E, link data 52e indicating that user D is related to user F, and link data 52f indicating that user E is related to user F. It is presumed that users with the same address live together. Therefore, in this embodiment, such users are associated with each other.

For example, as shown in FIG. 7, suppose that a group of attribute data for user G is registered in electronic commerce system 40. Also, suppose that a group of attribute data for user H is registered in golf course reservation system 42. Also, suppose that a group of attribute data for user I is registered in travel reservation system 44. And suppose that the value of the credit card number data for user G registered in electronic commerce system 40, the value of the credit card number data for user H registered in golf course reservation system 42, and the value of the credit card number data for user I registered in travel reservation system 44 are the same.

In this case, as shown in FIG. 8, the graph data generation unit 21 generates graph data including node data 50g associated with user G, node data 50h associated with user H, node data 50i associated with user I, link data 52g indicating that user G is related to user H, link data 52h indicating that user G is related to user I, and link data 52i indicating that user H is related to user I. Users with the same credit card number are presumed to be family members, such as parents and children. For this reason, in this embodiment, such users are associated with each other.

The criteria for determining whether a pair of users is related to each other are not limited to those described above. A pair of users can be determined based on various criteria, such as location history and behavior history.

The link indicated by the link data 52 associating users identified as having a relationship with each other as described above is referred to as an explicit link. For example, a user connected to a first user via an explicit link and a user connected to a second user via an explicit link have a predetermined number of users in common (e.g., three or more). In this case, in this embodiment, for example, the graph data generation unit 21 generates link data 52 indicating that the first user is related to the second user. The link indicated by the link data 52 generated in this way is referred to as an implicit link.

For example, as shown in FIG. 9, it is assumed that node data 50j associated with user J and node data 50k associated with user K are connected by link data 52j indicating an explicit link. It is also assumed that node data 50j associated with user J and node data 50l associated with user L are connected by link data 52k indicating an explicit link. It is also assumed that node data 50j associated with user J and node data 50m associated with user M are connected by link data 52l indicating an explicit link.

In addition, it is assumed that the node data 50k associated with user K and the node data 50n associated with user N are connected by link data 52m indicating an explicit link. It is assumed that the node data 50l associated with user L and the node data 50n associated with user N are connected by link data 52n indicating an explicit link. It is assumed that the node data 50m associated with user M and the node data 50n associated with user N are connected by link data 52o indicating an explicit link.

In this case, the graph data generating unit 21 generates link data 52p (link data 52p indicating an implicit link) indicating that user J has a relationship with user N. In this way, user N is identified as a user who has a relationship with user J.

For example, suppose that a user connected to a first user via an explicit link or an implicit link has a predetermined number of users in common (e.g., three or more) with a second user connected to the first user via an explicit link or an implicit link. In this case, the graph data generating unit 21 may generate link data 52 indicating that the first user is related to the second user (link data 52 indicating an implicit link).

The reference user identification unit 22 refers to the graph data generated by the graph data generation unit 21, and identifies other users included in the graph data who are related to the target user as reference users for the target user. Here, the reference user identification unit 22 may identify, as reference users, users who are identified as users who are related to the target user, and users who have a predetermined number or more of users identified as related users in common with the target user. Furthermore, the reference user identification unit 22 may identify a reference user from among the multiple users based on the attributes of the target user and the attributes of the multiple users.

The reference user identification unit 22 may, for example, identify a user associated with node data 50 associated with a target user and node data 50 connected by link data 52 indicating an explicit link or an implicit link as a reference user for the target user.

The relationship identification unit 23 identifies the relationship between users. The relationship between users identified here may be, for example, (1) a parent-child relationship or a husband-wife relationship living in the same household, (2) a friendship relationship, (3) a relationship working in the same workplace, etc. However, the identified relationship is not limited to the examples in this disclosure. In this embodiment, the relationship identification unit 23 identifies the relationship between users based on the results of clustering based on values associated with the relationship between users. Here, the types of values that can be adopted as values associated with the relationship between users are not limited, but may include, for example, at least one of the user's name, IP address, address, credit card number, age, gender, place of school, place of work, and place of stay.

The relationship identification unit 23 identifies the relationship between the target user and the reference user. Here, the relationship identification unit 23 may identify the relationship between the target user and the reference user based on the attribute data group of the target user and the attribute data group of the reference user. Furthermore, the computer system in which the attribute data group of the target user is registered may be different from the computer system in which the attribute data group of the reference user is registered. For example, the relationship between the target user and the reference user may be identified based on the attribute data group of the target user registered in the electronic commerce system 40 and the attribute data group of the reference user registered in the golf course reservation system 42.

The relationship identification unit 23, for example, identifies pairs of node data 50 that are connected by link data 52. Then, the relationship identification unit 23 generates pair attribute data associated with the pair based on a group of user attribute data of the two users associated with the pair. Here, the pair attribute data includes, for example, a common IP flag, a common address flag, a common credit card number flag, a same last name flag, age difference data, pair gender data, a common school flag, a common workplace flag, a common place of stay flag, etc.

The IP common flag is, for example, a flag indicating whether the value of the IP address data included in one attribute data of the pair is the same as the value of the IP address data included in the other attribute data. For example, if the values of the IP address data are the same, the value of the IP common flag may be set to 1, and if the values of the IP address data are different, the value of the IP common flag may be set to 0.

The common address flag, common school flag, common workplace flag, and common place of stay flag are flags that indicate, for example, whether the values of the address data/school data/workplace data/place of stay data included in one attribute data group of the pair are the same as the values of the address data/school data/workplace data/place of stay data included in the other attribute data group. For example, if the values of the address data are the same, the value of the common address flag may be set to 1, and if the values of the address data are different, the value of the common address flag may be set to 0.

The credit card number common flag is, for example, a flag indicating whether the value of the credit card number data included in one attribute data group of the pair is the same as the value of the credit card number data included in the other attribute data group. For example, if the values of the credit card number data are the same, the value of the credit card number common flag may be set to 1, and if the values of the credit card number data are different, the value of the credit card number common flag may be set to 0.

The same surname flag is, for example, a flag indicating whether the surname indicated by the name data included in one attribute data group of the pair is the same as the surname indicated by the name data included in the other attribute data group. For example, if the surnames indicated by the name data are the same, the value of the same surname flag may be set to 1, and if the surnames indicated by the name data are different, the value of the same surname flag may be set to 0.

The age difference data is, for example, data indicating the difference between the value of the age data included in one attribute data group of the pair and the value of the age data included in the other attribute data group.

Paired gender data is, for example, data that indicates a combination of the gender data value contained in one attribute data group of the pair and the gender data value contained in the other attribute data group.

Then, the relationship identification unit 23 performs clustering using a general clustering method based on the values of the pair attribute data group associated with each of the multiple pairs, thereby classifying the multiple pairs into multiple clusters 54 as shown in FIG. 10.

Figure 10 is a diagram showing a schematic example of how a plurality of pairs are classified into five clusters 54 (54a, 54b, 54c, 54d, and 54e). The crosses shown in Figure 10 correspond to pairs. Each of the plurality of crosses is placed at a position that corresponds to the value of the pair attribute data of the pair corresponding to the cross. In the example of Figure 10, the plurality of pairs are classified into five clusters 54, but the number of clusters 54 into which the plurality of pairs are classified is not limited to five, and the plurality of pairs may be classified into four clusters 54, for example.

FIG. 11 is a diagram showing an example of visualization of classification when multiple pairs are classified into four clusters 54. As shown in FIG. 11, pairs with the same address, the same gender, an age difference of more than X years, and the same last name may be classified into a first cluster. Pairs with the same address, the same gender, an age difference of X years or less, and the same last name may be classified into a second cluster. Pairs with the same address, different gender, an age difference of more than Y years, and the same last name may be classified into a third cluster. Pairs with the same address, different gender, an age difference of Y years or less, and the same last name may be classified into a fourth cluster.

In this case, the first cluster is presumed to be cluster 54 associated with, for example, parents and children of the same sex. The second cluster is presumed to be cluster 54 associated with siblings of the same sex. The third cluster is presumed to be cluster 54 associated with parents and children of the opposite sex. The fourth cluster is presumed to be cluster 54 associated with a married couple.

As described above, the relationship identification unit 23 may identify the relationship between the target user and the reference user based on the results of clustering based on values associated with the relationships between users. A specific example of creating clusters of friendships or working in the same workplace by clustering based on the common school flag, common workplace flag, and common place of stay flag is roughly similar to the example described above, so an explanation is omitted. In addition, the relationship identification unit 23 may identify the relationship between the target user and the reference user based on the results of clustering based on at least one of the family name, IP address, address, credit card number, age difference, gender, school, place of work, and place of stay.

The relationship strength determination unit 24 determines the relationship strength (hereinafter also referred to as "closeness score") indicating the closeness between the target user and the reference user based on an index indicating the strength of the relationship between the target user and the reference user in accordance with a judgment criterion corresponding to the relationship between the target user and the reference user. In this embodiment, the relationship strength determination unit 24 determines the relationship strength (closeness score) indicating the closeness between the target user and the reference user based on the output when data indicating the index is input into a trained machine learning model corresponding to the relationship between the target user and the reference user.

Here, the relationship strength determination unit 24 may include a trained machine learning model that is associated with each of the above-mentioned clusters 54. For example, if multiple pairs are classified into five clusters 54, the relationship strength determination unit 24 may include five machine learning models. Then, the relationship strength determination unit 24 may determine a closeness score indicating the closeness between the target user and the reference user based on the output when data indicating an index indicating the strength of the relationship between the target user and the reference user is input to a trained machine learning model corresponding to the relationship between the target user and the reference user. In this case, the input/output relationship implemented in the trained machine learning model corresponds to the above-mentioned judgment criterion.

As shown in FIG. 12, the relationship strength determination unit 24 may input input data corresponding to a pair classified into a cluster 54 associated with the nth machine learning model to the nth machine learning model, which is the nth machine learning model. For example, if the relationship strength determination unit 24 includes five machine learning models, the above-mentioned value n is any integer between 1 and 5. Then, the relationship strength determination unit 24 may determine the value of the output data output from the nth machine learning model in response to the input of the input data as the value of the closeness score for the pair.

The input data associated with a pair may include, for example, some or all of the pair attribute data associated with the pair. The input data may also include data that is not included in the pair attribute data. For example, the input data may include data indicating the usage history of the electronic commerce system 40, data acquired by the relationship strength determination unit 24 from other information sources such as SNS, and the like. More specifically, for example, the input data may include data indicating the number of calls or message exchanges between the pair per unit period, the number of gifts sent by one to the other, the number of mutual friends in the pair, and the like.

In addition, the type of data included in the input data associated with a pair may be the same or different depending on the cluster 54 to which the pair belongs. For example, the type of data included in the input data input to the first machine learning model may be different from the type of data included in the input data input to the second machine learning model.

In this embodiment, for example, prior to the determination of the closeness score by the relationship strength determination unit 24, learning of the nth machine learning model is performed in advance using a given number of training data items associated with the nth machine learning model. This training data is, for example, prepared in advance so that the determination of the closeness score in the cluster 54 associated with the nth machine learning model is valid. Here, the closeness score set in the training data may be a rule-based set (annotated) closeness score. It may also be a closeness score that was previously output by the machine learning model and then modified by an administrator, etc.

Here, weakly supervised learning may be performed on the nth machine learning model. For example, the training data may include training input data that includes the same type of data as the input data input to the nth machine learning model, and training data that is compared with output data output from the nth machine learning model in response to the input of the training input data.

Here, for example, the closeness score described above takes a value of either 0 or 1. For example, if the pair is closely related, the closeness score value of the pair is determined to be 1, and if not, the closeness score value of the pair is determined to be 0. In this case, the teacher data may include a valid closeness score value for the corresponding learning input data and data indicating the probability that this value is valid. Then, for example, weakly supervised learning may be performed to update the parameter values of the nth machine learning model based on the value of the output data output from the nth machine learning model in response to the input of the learning input data included in the teacher data and the value of the teacher data included in the teacher data.

The above-mentioned closeness score does not have to be binary data that takes the value of either 0 or 1. For example, the above-mentioned closeness score may be a real number (e.g., a real number between 0 and 10 inclusive) that increases the closer the pair is to each other, or a multi-level integer value (e.g., an integer number between 1 and 10 inclusive).

Furthermore, the learning method of the machine learning model is not limited to weakly supervised learning. As a specific example, consider a pair that has a sibling relationship. In this case, input data associated with the pair is input to a trained machine learning model that corresponds to the sibling relationship. For example, if the address data values for the pair are the same, one of the pair has sent 50 gifts to the other, and the number of calls between the pair has been 1200, learning may be performed to output output data with a value of 1. For example, if the address data values for the pair are different, one of the pair has sent 2 gifts to the other, and the number of calls between the pair has been 30, learning may be performed to output output data with a value of 0. The criteria (e.g., threshold) for determining whether the value of the output data corresponding to the closeness score is 1 or 0 may differ depending on the machine learning model.

The attribute selection unit 25 selects the type of attribute data (type of attribute data to be complemented) generated by the attribute generation unit 26 according to the type of relationship between the target user and the reference user. Specific examples of the types of relationships between users and the types of attribute data selected according to the types of relationships include the relationships and attribute data exemplified below.

(1) Parent-child or marital relationship living in the same household When the relationship between users is a parent-child or marital relationship living in the same household, it can be assumed that mainly monetary variables and variables indicating behavior as a household are the same. Therefore, when such a relationship between users is identified, the attribute selection unit 25 selects, as the type of attribute data to be generated by the attribute generation unit 26, for example, household income, annual household income, place of residence, whether or not the household has insurance, amount of deposits, financial assets, whether or not the household subscribes to a newspaper, etc.

(2) Friendship Relationship When the relationship between users is a friendship relationship, it can be assumed that groups of people with the same sex, age, and hobbies are likely to become friends. Therefore, when the relationship between users is identified, the attribute selection unit 25 selects, for example, hobbies, frequently visited places/areas, age, sex, etc. as the types of attribute data to be generated by the attribute generation unit 26.

(3) Relationship of working in the same workplace When the relationship between users is a relationship of working in the same workplace, it can be assumed that a group with the same educational level and specialty field often works in the same workplace. Therefore, when the relationship between users is identified, the attribute selection unit 25 selects, for example, the genre of specialized books to be purchased, educational level, etc. as the type of attribute data to be generated by the attribute generation unit 26.

In this embodiment, a method has been described in which the attribute selection unit 25 selects the type of attribute data to be complemented (generated) on a rule basis, but the method of selecting the type of attribute data to be complemented is not limited to the example in this embodiment. For example, a method may be adopted in which the type of attribute data to be complemented is selected using a machine learning model that has learned the presence or absence of correlation between the type of relationship between users and the type of similar attribute data, and the degree of correlation.

The attribute generation unit 26 generates attribute data for complementing missing attribute data or unreliable attribute data in the attribute data group of the target user based on information about at least one reference user identified for the target user. Here, the attribute generation unit 26 refers to the type of attribute data selected by the attribute selection unit 25 from the attribute data group of the reference user as the information about the reference user, and generates attribute data of the target user corresponding to the referred attribute data.

Specifically, when the relationship between the target user and the reference user is "(1) parent-child relationship or marital relationship residing in the same household," the attribute generation unit 26 refers to the reference user's attribute data for attribute data such as household income, annual household income, place of residence, whether or not the household has insurance, deposit amount, financial assets, whether or not the household subscribes to a newspaper, etc., and generates corresponding attribute data for the target user based on the attribute data. When the relationship between the target user and the reference user is "(2) friendship," the attribute generation unit 26 refers to the reference user's attribute data for attribute data such as hobbies, frequently visited places and areas, age, sex, etc., and generates corresponding attribute data for the target user based on the attribute data. When the relationship between the target user and the reference user is "(3) working in the same workplace," the attribute generation unit 26 refers to the reference user's attribute data for attribute data such as genre of specialized books purchased, educational level, etc., and generates corresponding attribute data for the target user based on the attribute data.

The attribute generation unit 26 may generate the attribute data of the target user by directly copying the parameters of the attribute data of the reference user to the corresponding attribute data of the target user. However, the attribute generation unit 26 may generate the corresponding attribute data of the target user by performing some processing on the parameters of the attribute data of the reference user. For example, when generating the attribute data of the target user, the attribute generation unit 26 may refer to the closeness score determined for the reference user, and generate the attribute data of the target user based on the parameters of the attribute data of the reference user and the closeness score.

For example, the attribute generation unit 26 may generate attribute data for the target user by weighting the parameters of the attribute data of the reference user based on the closeness score. In this case, the attribute generation unit 26 sets a larger weighting coefficient the higher the closeness score between the target user and the reference user indicates the strength of the relationship between the users. Then, by performing processing using the weighting coefficient on the parameters of the attribute data of the reference user (for example, simply multiplying the parameters by the weighting coefficient), the parameters of the attribute data supplemented for the target user can be made to be closer to the parameters of the attribute data of the referenced reference user.

In addition, here, when multiple reference users are identified, attribute data of the target user may be generated based on the multiple reference users. For example, the attribute generation unit 26 may acquire a closeness score and a parameter of the attribute data to be complemented for each of the multiple reference users, weight the parameters acquired from each reference user based on the closeness score, and use the average (not limited to the average, but other statistics such as the median may also be used) of the multiple weighted parameters acquired for each reference user as the parameter of the corresponding attribute data of the target user.

For example, the attribute generation unit 26 may generate attribute data of the target user using an attribute generation model that uses as input values at least some parameters of the attribute data group of the target user before complementation, at least some parameters of the attribute data group of the reference user, and a closeness score between the target user and the reference user, and that uses as output values the attribute data of the target user to be complemented. As in the case of employing weighting, when employing an attribute generation model, the attribute generation model is generated and/or updated so that the parameters of the attribute data to be complemented for the target user are closer to the parameters of the attribute data of the referenced reference user as the closeness score between the target user and the reference user is higher. As in the case of employing weighting, the attribute generation model may be input with closeness scores and attribute data related to multiple reference users, and parameters of the attribute data to be complemented of the target user may be output.

The attribute complementing unit 27 complements the attribute data group for the user based on at least a portion of the generated attribute data. The attribute data group for the user includes attribute data including account data and usage history data acquired from the service providing system 5, and in this case, the attribute complementing unit 27 determines at least a portion of the attribute data generated by the attribute generating unit 26 as at least a portion of the attribute data group for the target user, and complements the attribute data group for the user.

Here, the attribute data complemented by the attribute complement unit 27 may include demographic attributes, behavioral attributes, or psychographic attributes. Demographic attributes are, for example, the user's gender, family structure, age, etc., behavioral attributes are, for example, whether or not a cash advance has been used, whether or not a revolving payment has been used, deposit and withdrawal history for a specific account, commercial transaction history for some product including gambling or lotteries (which may include online transaction history in an online marketplace, etc.), etc., and psychographic attributes are, for example, preferences for gambling or lotteries. However, the available user attributes are not limited to the examples in this embodiment. For example, "time required for operation (calls, etc.)" and "credit card usage amount/deferred payment usage amount" from an operation center service, etc. may also be used as attribute data.

The user score estimation unit 28 estimates a user score to be set for the user based on the complemented attribute data group. In this embodiment, the user score estimation unit 28 estimates a user score to be set for the user by inputting the attribute data group of the user into a user score estimation model. Here, the output value of the user score estimation model is a user score normalized/standardized with 0 as the minimum value and 1 as the maximum value. Here, the attribute data group of the target user input into the user score estimation model includes attribute data generated by the attribute generation unit 26. As described above, the attribute data generated by the attribute generation unit 26 may include, for example, household income, annual household income, place of residence, whether or not the household has insurance (as a household), amount of deposits, financial assets, whether or not the household subscribes to a newspaper, hobbies, frequently visited places/areas, age, sex, genre of specialized books purchased, educational level, etc.

The machine learning unit 29 generates and/or updates a user score estimation model used for user score estimation by the user score estimation unit 28. The user score estimation model may be a machine learning model that outputs a user score indicating some measure related to the user (e.g., trust, etc.) when one or more attribute data (group of attribute data) related to the target user is input, or may be some function or statistical model capable of outputting a user score.

When generating and/or updating the user score estimation model, the machine learning unit 29 creates teacher data for each user based on the data acquired from the service providing system 5, in which a group of attribute data including the demographic attributes of the user is defined as an input value and the user score related to the user is defined as an output value. Then, the machine learning unit 29 creates a user score estimation model based on the teacher data. As described above, the group of attribute data input to the user score estimation model includes attribute data generated by the attribute generation unit 26, and is combined with the user score of the corresponding user and input to the machine learning unit 29 as teacher data. The user score set in the teacher data may be a user score set (annotated) based on a rule base. It may also be a user score that was previously output by the user score estimation model and then corrected by an administrator or the like.

A machine learning model generation framework that can be adopted when implementing the technology according to the present disclosure is based on an ensemble learning algorithm, for example. For example, a machine learning framework (for example, LightGBM) based on a gradient boosting decision tree (GBDT) may be adopted as the framework. In other words, the framework may be a machine learning framework based on a decision tree model that transfers the error between the correct answer and the predicted value between the previous and next weak learners (weak classifiers). The predicted value here refers to the predicted value of the user score, for example. In addition to LightGBM, the framework may adopt boosting methods such as XGBoost and CatBoost. According to a framework using a decision tree, a machine learning model with relatively high performance can be generated with less effort in parameter adjustment compared to a framework using a neural network. However, the machine learning model generation framework that can be adopted when implementing the technology according to the present disclosure is not limited to the examples in this embodiment. For example, instead of a gradient boosting decision tree, another learning device such as a random forest may be used as the learning device, or a learning device that is not a so-called weak learning device such as a neural network may be used. In particular, when a learning device that is not a so-called weak learning device such as a neural network is used, ensemble learning does not need to be used.

FIG. 13 is a simplified diagram showing the concept of a decision tree in a machine learning model employed in this embodiment. When a gradient boosting machine learning framework based on a decision tree algorithm is employed, the branching conditions of each node of the decision tree are optimized. Specifically, in the gradient boosting machine learning framework based on a decision tree algorithm, a user score is calculated for each user group having attributes indicated by each of two child nodes branched from one parent node, and the branching conditions of the parent node are optimized so that the difference between these user scores is large (for example, so that the difference is maximized or is equal to or greater than a predetermined threshold), that is, so that the two child nodes branch neatly. For example, if the attribute indicated as the branching condition of the node is age, the age set as the branching threshold may be changed, or the branching condition may be changed to an attribute other than age. In this way, the branching conditions of all nodes of the decision tree are recursively optimized, thereby improving the estimation accuracy of the user score based on the attribute data group.

In addition, when the attribute generation unit 26 generates attribute data to be complemented using an attribute generation model, the machine learning unit 29 further generates and/or updates the attribute generation model used by the attribute generation unit 26 to generate the attribute data to be complemented for the target user. The attribute generation model is a machine learning model that outputs attribute data to be complemented for the target user when one or more attribute data and a closeness score related to one or more reference users are input.

When generating and/or updating the attribute generation model, the machine learning unit 29 creates training data in which the attribute data and closeness scores of one or more reference users among the data acquired from the service providing system 5 are defined as input values and one attribute data (attribute data to be complemented related to the target user) is defined as an output value. Here, the output value (parameter of the attribute data to be complemented of the target user) set in the training data used to generate and/or update the attribute generation model may be an output value set (annotated) based on a rule base (for example, the above-mentioned weighting calculation method). It may also be an output value that was previously output by the attribute generation model and then modified by an administrator, etc.

Then, the machine learning unit 29 generates or updates an attribute generation model based on the teacher data. One or more attribute data and the closeness score are combined with the corresponding attribute data and input to the machine learning unit 29 as teacher data. In addition, in generating or updating the attribute generation model, the machine learning model generation framework that can be adopted is not limited, but a machine learning framework of gradient boosting based on a decision tree algorithm may be adopted, as in the user score estimation model described above.

<Processing flow>
Next, a process flow executed by the information processing system according to the present embodiment will be described. Note that the specific contents and processing order of the processes described below are an example for implementing the present disclosure. The specific contents and processing order may be appropriately selected according to the embodiment of the present disclosure.

Figure 14 is a flowchart showing the flow of the machine learning process according to this embodiment. The process shown in this flowchart is executed at a timing specified by the administrator.

In this embodiment, in the machine learning process, a user score estimation model is generated and/or updated. The machine learning unit 29 creates training data including a combination of a group of attribute data for each user previously accumulated in the service providing system 5 and a user score previously determined for the corresponding user (step S101). Then, the machine learning unit 29 inputs the created training data into the user score estimation model, and creates or updates a user score estimation model used for user score estimation by the user score estimation unit 28 (step S102). Thereafter, the process shown in this flowchart ends. Note that when the attribute generation unit 26 uses an attribute generation model for attribute complementation, the attribute generation model may also be generated and/or updated in a similar process flow.

Figure 15 is a flowchart showing the flow of the user score estimation process according to this embodiment. The process shown in this flowchart is executed for each target user at a timing specified by the administrator. Here, the target users are users whose attribute data is missing or has low reliability. Examples of unreliable attribute data include attribute data generated based on insufficiently accumulated history data and attribute data that clearly contradicts the contents of other attribute data. Note that here, it is assumed that graph data has already been generated for multiple users including the target user, and that each machine learning model has already been trained.

In steps S201 and S203, the reference user is identified, and the relationship between the target user and the reference user is identified. The reference user identification unit 22 refers to the graph data, and identifies one or more other users corresponding to the node data 50 that is connected to the node data 50 corresponding to the target user by an explicit link or an implicit link as reference users (step S201). Then, the relationship identification unit 23 identifies the type of relationship between the users (specifically, parent-child relationship living in the same household/married couple relationship/friendship relationship/working in the same workplace, etc.) for each pair of the target user and one or more reference users identified in step S201 (step S202). After that, the process proceeds to step S203.

In steps S203 and S204, the type of attribute data to be complemented is selected, and a closeness score between users is determined. The attribute selection unit 25 selects the type of attribute data to be complemented for the target user according to the type of relationship identified in step S202 (step S203). In addition, the relationship strength determination unit 24 determines the closeness score value associated with each pair of the target user and each reference user (S204). After that, the process proceeds to step S205.

In step S205, attribute data to be complemented for the target user is generated. The attribute generation unit 26 generates attribute data to be complemented for the target user based on the parameters of the attribute data of the reference user that correspond to the attribute data to be complemented and the closeness score determined in step S204 for the reference user. Then, the process proceeds to step S206.

In steps S206 and S207, a user score is estimated and output. The attribute complementing unit 27 adds the complemented attribute data generated in step S205 to the attribute data group previously stored for the target user, such as by being acquired from the service providing system 5, to generate an attribute data group for the user (step S206). Then, the user score estimating unit 28 inputs the attribute data group including the attribute data complemented for the target user in step S206 into a user score estimation model, and acquires the output value as the user score to be set for the user (step S207). However, the method of estimating the user score is not limited to the example in this embodiment. For example, the user score may include a value calculated by inputting the attribute data group into a predetermined function that is not a machine learning model. After that, the processing shown in this flowchart ends.

The user score set for each user is provided to other systems, such as the service providing system 5, and is used to customize the services provided to the target user by the other systems, such as the service providing system 5.

This embodiment can also be used to estimate a user score for a new target user whose corresponding node data 50 is not included in the graph node. For example, based on the user attribute data of the new target user, node data 50 corresponding to the target user and at least one link data 52 connected to the node data 50 may be generated. Then, a user connected to the node data 50 corresponding to the target user by the link data 52 may be identified as a reference user of the target user.

＜Effects＞
According to this embodiment, missing attributes of a user are complemented from a social graph network that covers the relationships between users, and a user score is estimated/determined using the complemented attribute group, so that even when information about a target user is missing or the reliability of the information is low, it is possible to calculate a user score or improve the accuracy of the calculated user score. In addition, by using various user attribute data, it is possible to calculate a highly accurate user score even when a certain range of attribute data (for example, credit card department) cannot be used due to regulations, laws, etc., or when some attribute data does not exist for the target user.

<Variations>
In the embodiment described above, an example of an information processing device including a graph data generation unit 21, a reference user identification unit 22, a relationship identification unit 23, a relationship strength determination unit 24, an attribute selection unit 25, an attribute generation unit 26, an attribute completion unit 27, a user score estimation unit 28, and a machine learning unit 29 was described, but some of these functional units may be omitted to the extent that the invention disclosed herein can be implemented.

For example, in the embodiment described above, the relationship strength (closeness score) between the target user and the reference user was generated and referenced when generating the attribute data to be complemented, but the generation and reference of the closeness score may be omitted when generating the attribute data to be complemented. In this case, of the functional units of the information processing device 1 described with reference to FIG. 2, the relationship strength determination unit 24 may be omitted. Furthermore, when generating the attribute data, the attribute generation unit 26 may generate the attribute data to be complemented for the target user based on the attribute data of the reference user without performing weighting with reference to the closeness score.

For example, the attribute generation unit 26 may generate attribute data of the target user using an attribute generation model in which at least some parameters of the attribute data group of the reference user and the closeness score between the target user and the reference user are used as input values, and the complemented attribute data of the target user is used as output values. At this time, the attribute generation model is appropriately pre-trained according to the state of the input values and output values.

For example, the attribute generation unit 26 may generate attribute data of the target user using an attribute generation model in which at least some parameters of the attribute data group of the target user and/or at least some parameters of the attribute data group of the reference user are used as input values, and the attribute data of the target user to be complemented is used as output values. At this time, the attribute generation model is appropriately pre-trained according to the state of the input value and the output value. At this time, the attribute generation unit 26 may determine a predetermined attribute generation model according to the type of relationship and/or the closeness score between the target user to be processed and the reference user, among a plurality of attribute generation models that differ for each relationship and/or closeness score between the target user and the reference user, and generate attribute data of the target user to be complemented. Here, each of the plurality of attribute generation models may be pre-trained based on teacher data in which the type of relationship and/or the closeness score is common or similar (within a predetermined range).

For example, the attribute generation unit 26 may generate attribute data of a target user using an attribute generation model in which the embedded representation (vector representation, feature representation) of the user on the graph data is used as an input value as at least a portion of the parameters of the attribute data group of the user (target user, reference user) and the complemented attribute data of the target user is used as an output value. The attribute generation model may include as an input value the distance or inner product between the target user and the reference user on the graph data (the distance or inner product in a vector space based on the graph data). In this case, the attribute generation model is appropriately pre-trained according to the state of the input value and the output value.

Furthermore, for example, if the attribute data output by the attribute generation model is a missing value (missing attribute data) or an invalid value (low-reliability attribute data) in the attribute data group of the target user before the completion, the attribute completion unit 27 may determine that the output attribute data is part of the attribute data group of the target user.

Furthermore, for example, the attribute selection unit 25 or the attribute completion unit 27 may treat attribute data with a high weight in an ensemble learning model such as a gradient boosting decision tree employed as a user score estimation model, as attribute data to be completed. Here, the attribute data with a high weight may be, for example, attribute data corresponding to a tree with a weight exceeding a predetermined weight in the user score estimation model, or attribute data corresponding to a tree showing a high weight (above a predetermined rank) in the user score estimation model.

1. Information processing device

Claims (16)

A reference user specification means for specifying a reference user having a relationship with the target user based on graph data indicating the relationship between the users;
an attribute generating means for generating corresponding attribute data of the target user based on attribute data of the reference user identified for the target user;
an attribute complementing means for complementing a group of attribute data corresponding to the target user based on at least a part of the generated attribute data corresponding to the target user;
a user score estimation means for estimating a user score to be set for the target user based on the complemented attribute data group corresponding to the target user;
An information processing system comprising: a graph data generating means for generating the graph data by identifying pairs of users who are related to each other based on a group of attribute data of each of the users;
The information processing system according to claim 1 . A relationship specification means for specifying a relationship between users based on a result of clustering based on at least one of the user's name, IP address, address, credit card number, age, sex, school, place of employment, and place of stay, as a value associated with the relationship between users;
A reference user identification means for identifying a reference user having a relationship with the target user;
an attribute generating means for generating corresponding attribute data of the target user based on attribute data of the reference user identified for the target user;
an attribute complementing means for complementing a group of attribute data corresponding to the target user based on at least a part of the generated attribute data corresponding to the target user;
a user score estimation means for estimating a user score to be set for the target user based on the complemented attribute data group corresponding to the target user;
An information processing system comprising: and a relationship strength determination means for determining a relationship strength indicating a closeness between the target user and the reference user based on an index indicating a strength of the relationship between the target user and the reference user in accordance with a judgment criterion corresponding to the relationship between the target user and the reference user,
the attribute generating means generates, for at least one of the reference users, corresponding attribute data of the target user based on information about the reference user and the relationship strength determined for the reference user;
The information processing system according to claim 3 . The relationship strength determination means determines the relationship strength indicating the closeness between the target user and the reference user based on an output when data representing the index is input to a trained machine learning model corresponding to the relationship between the target user and the reference user.
5. The information processing system according to claim 4. A reference user identification means for identifying a reference user having a relationship with the target user;
an attribute selection means for selecting a type of attribute data to be complemented for the target user according to a type of relationship between the target user and the reference user ;
an attribute generating means for generating attribute data corresponding to the target user based on the type of attribute data selected by the attribute selecting means from the attribute data group of the reference users identified for the target user;
an attribute complementing means for complementing a group of attribute data corresponding to the target user based on at least a part of the generated attribute data corresponding to the target user;
a user score estimation means for estimating a user score to be set for the target user based on the complemented attribute data group corresponding to the target user;
An information processing system comprising: The user score estimation means estimates a user score to be set for the target user by inputting a group of attribute data of the target user into a machine learning model.
The information processing system according to claim 1 . The user score estimation means estimates the user score using a machine learning model generated using a machine learning framework based on a gradient boosting decision tree.
The information processing system according to claim 7. The user score estimation means estimates a user score to be set for the target user by using the machine learning model generated using teacher data in which a group of attribute data including demographic attributes of the user is an input value and the user score related to the user is an output value.
9. The information processing system according to claim 7 or 8. The attribute complementing means generates attribute data for complementing missing attribute data or low reliability attribute data among the attribute data group of the target user based on the attribute data of the reference user.
The information processing system according to any one of claims 1 to 9. The computer
A reference user identifying step of identifying a reference user having a relationship with the target user based on graph data indicating the relationship between the users;
an attribute generating step of generating corresponding attribute data of the target user based on attribute data of the reference user identified for the target user;
an attribute complementation step of complementing a group of attribute data corresponding to the target user based on at least a part of the generated attribute data corresponding to the target user;
a user score estimating step of estimating a user score to be set for the target user based on the complemented attribute data group corresponding to the target user;
How to do it. The computer
a relationship identification step of identifying a relationship between the users based on a result of clustering based on at least one of the users' names, IP addresses, addresses, credit card numbers, ages, sexes, schools, places of employment, and places of stay as values associated with the relationships between the users;
A reference user identification step of identifying a reference user who is related to the target user;
an attribute generating step of generating corresponding attribute data of the target user based on attribute data of the reference user identified for the target user;
an attribute complementation step of complementing a group of attribute data corresponding to the target user based on at least a part of the generated attribute data corresponding to the target user;
a user score estimating step of estimating a user score to be set for the target user based on the complemented attribute data group corresponding to the target user;
How to do it. The computer
A reference user identification step of identifying a reference user who is related to the target user;
an attribute selection step of selecting a type of attribute data to be complemented for the target user according to a type of relationship between the target user and the reference user ;
an attribute generating step of generating corresponding attribute data of the target user based on the attribute data of the type selected in the attribute selecting step from among the attribute data group of the reference users identified for the target user;
an attribute complementation step of complementing a group of attribute data corresponding to the target user based on at least a part of the generated attribute data corresponding to the target user;
a user score estimating step of estimating a user score to be set for the target user based on the complemented attribute data group corresponding to the target user;
How to do it. Computer,
A reference user specification means for specifying a reference user having a relationship with the target user based on graph data indicating the relationship between the users;
an attribute generating means for generating corresponding attribute data of the target user based on attribute data of the reference user identified for the target user;
an attribute complementing means for complementing a group of attribute data corresponding to the target user based on at least a part of the generated attribute data corresponding to the target user;
a user score estimation means for estimating a user score to be set for the target user based on the complemented attribute data group corresponding to the target user;
A program that functions as a Computer,
A relationship specification means for specifying a relationship between users based on a result of clustering based on at least one of the user's name, IP address, address, credit card number, age, sex, school, place of employment, and place of stay, as a value associated with the relationship between users;
A reference user identification means for identifying a reference user having a relationship with the target user;
an attribute generating means for generating corresponding attribute data of the target user based on attribute data of the reference user identified for the target user;
an attribute complementing means for complementing a group of attribute data corresponding to the target user based on at least a part of the generated attribute data corresponding to the target user;
a user score estimation means for estimating a user score to be set for the target user based on the complemented attribute data group corresponding to the target user;
A program that functions as a Computer,
A reference user identification means for identifying a reference user having a relationship with the target user;
an attribute selection means for selecting a type of attribute data to be complemented for the target user according to a type of relationship between the target user and the reference user ;
an attribute generating means for generating attribute data corresponding to the target user based on the type of attribute data selected by the attribute selecting means from the attribute data group of the reference users identified for the target user;
an attribute complementing means for complementing a group of attribute data corresponding to the target user based on at least a part of the generated attribute data corresponding to the target user;
a user score estimation means for estimating a user score to be set for the target user based on the complemented attribute data group corresponding to the target user;
A program that functions as a

Images