Graph data generation system, graph data generation method, and program

Abstract

To provide a credit determination system, a credit determination method, and a program capable of accurately determining credit of a person whose credit information is not known.SOLUTION: A relationship specifying unit (26) specifies a relationship between a target person and a reference person. A closeness score determining unit (28) determines a closeness score indicating closeness between the target person and the reference person in accordance with a determination criterion corresponding to the relationship between the target person and the reference person, based on an index indicating strength of the relationship between the target person and the reference person. A credit determining unit (32) determines credit of the target person based on information about the at least one reference person and the closeness score determined for the reference person.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a creditworthiness determination system, a creditworthiness determination method, and a program. [Background technology]

[0002] Patent Document 1 describes a technology for determining whether the outstanding loan balance of each loan contract included in the inquiry result file is equal to or less than the loan limit set by the total amount regulation of the Money Lending Business Act. Patent Document 1 also describes that when married persons each have a loan contract and the outstanding loan balances are added up, the loan limit set for the higher-income contract holder may be used, or the total of the loan limits for both parties may be used. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2009-301236 Summary of the Invention [Problem to be solved by the invention]

[0004] In the technology described in Patent Document 1, the above-mentioned judgment is made based on information on the loan contract registered with a credit information agency. Therefore, even if an attempt is made to judge a person's creditworthiness using the technology described in Patent Document 1, without the person's credit information, the person's creditworthiness cannot be accurately judged.

[0005] The present invention has been made in consideration of the above-mentioned problems, and one of its objectives is to provide a credit assessment system, a credit assessment method, and a program that can accurately assess the creditworthiness of a person for whom there is no credit information. [Means for solving the problem]

[0006] The creditworthiness determination system of the present invention is a creditworthiness determination system for determining the creditworthiness of a person of interest, and includes: a relationship identification means for identifying the relationship between the person of interest and a reference person; a proximity score determination means for determining a proximity score indicating the closeness between the person of interest and the reference person based on an indicator indicating the strength of the relationship between the person of interest and the reference person in accordance with a judgment criterion corresponding to the relationship between the person of interest and the reference person; and a creditworthiness determination means for determining the creditworthiness of at least one of the reference persons based on information about the reference person and the proximity score determined for the reference person.

[0007] In one aspect of the present invention, the closeness score determination means determines the closeness score indicating the closeness between the person of interest and the reference person based on the output when data representing the indicator is input into a trained machine learning model corresponding to the relationship between the person of interest and the reference person.

[0008] In addition, in one aspect of the present invention, the relationship identification means identifies the relationship between the person of interest and the reference person based on account data of the person of interest registered in a first computer system and account data of the reference person registered in a second computer system.

[0009] In one aspect of the present invention, the system further includes a pair identification means for identifying pairs of people who are related to each other based on the attributes of each of a plurality of people, and a reference person identification means for identifying as reference people a person identified as a person related to the person of interest, and a person who has a predetermined number or more of commonalities with the person of interest among the people identified as a related person.

[0010] In one aspect of the present invention, the relationship specification means specifies a family relationship between the person of interest and the reference person.

[0011] In one aspect of the present invention, the creditworthiness determining means determines the credit score of the person of interest.

[0012] In one aspect of the present invention, the relationship specification means specifies the relationship between the person of interest and the reference person based on a result of clustering based on values ​​associated with relationships between people.

[0013] In this aspect, the relationship identification means may identify the relationship between the person of interest and the reference person based on the results of clustering based on at least one of last name, IP address, address, credit card number, age difference, or gender.

[0014] In one aspect of the present invention, the credibility determination means performs a first determination to determine the credibility of the person of interest based on information about the person of interest, and if the credibility of the person of interest determined by the first determination is below a predetermined level, or if the credibility of the person of interest cannot be determined by the first determination, the credibility determination means performs a second determination to re-determine the credibility of the person of interest based on information about at least one of the reference persons.

[0015] In addition, the creditworthiness determination method of the present invention is a creditworthiness determination method for determining the creditworthiness of a person of interest, and includes the steps of: identifying the relationship between the person of interest and reference persons; determining a closeness score indicating the closeness between the person of interest and the reference persons based on an indicator indicating the strength of the relationship between the person of interest and the reference persons in accordance with a judgment criterion corresponding to the relationship between the person of interest and the reference persons; and determining the creditworthiness of at least one of the reference persons based on information about the reference person and the closeness score determined for the reference person.

[0016] In addition, the program of the present invention causes a computer that judges the credibility of a person of interest to execute the following steps: identifying the relationship between the person of interest and reference persons; determining a closeness score indicating the closeness between the person of interest and the reference persons based on an indicator indicating the strength of the relationship between the person of interest and the reference persons, in accordance with judgment criteria corresponding to the relationship between the person of interest and the reference persons; and judging the credibility of at least one of the reference persons based on information about the reference person and the closeness score determined for the reference person. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a diagram illustrating an example of the overall configuration of a creditworthiness determination system according to an embodiment of the present invention. [Figure 2] 1 is a functional block diagram showing an example of the functions of a credit rating determination system according to an embodiment of the present invention. [Figure 3] FIG. 10 is a diagram schematically illustrating an example in which values ​​of IP address data are common. [Figure 4] FIG. 10 is a diagram illustrating an example of graph data. [Figure 5] FIG. 10 is a diagram schematically illustrating an example in which address data values ​​are common. [Figure 6] FIG. 10 is a diagram illustrating an example of graph data. [Figure 7] FIG. 10 is a diagram schematically illustrating an example in which the value of credit card number data is common. [Figure 8] FIG. 10 is a diagram illustrating an example of graph data. [Figure 9] FIG. 10 is a diagram illustrating an example of graph data. [Figure 10] FIG. 10 is a diagram illustrating an example of a cluster. [Figure 11] FIG. 10 is a diagram illustrating an example of visualization of classification. [Figure 12] FIG. 1 illustrates an example of determining a closeness score using a machine learning model. [Figure 13] FIG. 1 is a diagram illustrating an example of learning of a machine learning model. [Figure 14] FIG. 2 is a flowchart showing an example of the flow of processing performed in the credit rating determination system according to one embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0018] Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

[0019] Fig. 1 is a diagram showing an example of the overall configuration of a creditworthiness determination system 1 according to one embodiment of the present invention. As shown in Fig. 1, the creditworthiness determination system 1 according to this embodiment is a computer such as a server computer or a personal computer, and includes a processor 10, a memory unit 12, a communication unit 14, an operation unit 16, and an output unit 18. Note that the creditworthiness determination system 1 according to this embodiment may include multiple computers.

[0020] The processor 10 is, for example, a program-controlled device such as a microprocessor that operates according to a program installed in the credit rating assessment system 1. The memory unit 12 is, for example, a storage element such as a ROM or RAM, or a solid-state drive (SSD). The memory unit 12 stores programs and the like to be executed by the processor 10. The communication unit 14 is, for example, a communication interface for wired or wireless communication, and transmits and receives data to and from other computers or terminals via a computer network such as the Internet.

[0021] The operation unit 16 is an input device, and includes, for example, a touch panel, a pointing device such as a mouse, a keyboard, etc. The operation unit 16 transmits the operation content to the processor 10. The output unit 18 is, for example, an output device such as a display, such as a liquid crystal display unit or an organic EL display unit, or an audio output device, such as a speaker.

[0022] The programs and data described as being stored in the storage unit 12 may be supplied from another computer via a network. The hardware configuration of the creditworthiness assessment system 1 is not limited to the above example, and various hardware configurations are applicable. For example, the creditworthiness assessment system 1 may include a reading unit (e.g., an optical disk drive or a memory card slot) that reads a computer-readable information storage medium, or an input / output unit (e.g., a USB port) that inputs and outputs data to and from an external device. For example, the programs and data stored in an information storage medium may be supplied to the creditworthiness assessment system 1 via the reading unit or the input / output unit.

[0023] In the creditworthiness determination system 1 according to this embodiment, for example, the creditworthiness of a person whose credit information is not registered in a credit information agency (hereinafter referred to as a person of interest) is determined.

[0024] The functions of the creditworthiness determination system 1 according to this embodiment and the processes executed by the creditworthiness determination system 1 will be further described below.

[0025] Fig. 2 is a functional block diagram showing an example of functions implemented in the creditworthiness determination system 1 according to this embodiment. Note that the creditworthiness determination system 1 according to this embodiment does not need to implement all of the functions shown in Fig. 2, and functions other than the functions shown in Fig. 2 may also be implemented.

[0026] As shown in Figure 2, the creditworthiness determination system 1 of this embodiment functionally includes, for example, a person attribute data acquisition unit 20, a graph data generation unit 22, a reference person identification unit 24, a relationship identification unit 26, a proximity score determination unit 28, a credit score acquisition unit 30, and a creditworthiness determination unit 32.

[0027] The person attribute data acquisition unit 20 and the credit score acquisition unit 30 are implemented mainly in the communication unit 14. The graph data generation unit 22, the reference person identification unit 24, the relationship identification unit 26, and the creditworthiness determination unit 32 are implemented mainly in the processor 10. The proximity score determination unit 28 is implemented mainly in the processor 10 and the memory unit 12.

[0028] The above functions may be implemented by executing a program including instructions corresponding to the above functions, which is installed in the credit rating assessment system 1, which is a computer, by the processor 10. The program may also be supplied to the credit rating assessment system 1 via a computer-readable information storage medium such as an optical disk, a magnetic disk, a magnetic tape, or a magneto-optical disk, or via the Internet, for example.

[0029] The creditworthiness assessment system 1 according to this embodiment is capable of communicating with multiple computer systems, such as an e-commerce system 40, a golf course reservation system 42, a travel reservation system 44, and a card management system 46 (see FIGS. 3, 5, and 7). Each of these computer systems stores account data, which is information about users who use the computer system. The creditworthiness assessment system 1 can access these computer systems and acquire the account data stored in the computer systems.

[0030] 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, and the like.

[0031] The user ID is, for example, identification information of the user in the computer system. The name data is, for example, data indicating the name (first and last name) of the user. The address data is, for example, data indicating the address of the user. If the computer system is an e-commerce system 40, the address data may indicate the address to which products purchased by the user are to be shipped. 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 on the computer system. The IP address data is, for example, data indicating the IP address of the computer used by the user (e.g., the IP address of the sender).

[0032] In this embodiment, for example, the person attribute data acquisition unit 20 acquires person attribute data indicating the attributes of multiple people including the person of interest. Here, an example of the person attribute data is the account data described above. The person attribute data acquisition unit 20 acquires the account data of the person from each of the multiple systems described above, for example.

[0033] In this embodiment, the graph data generation unit 22 identifies pairs of people who are related to each other based on the attributes of each of the multiple people. The graph data generation unit 22 may identify pairs of people who are related to each other based on the person attribute data of the multiple people. Note that the graph data generation unit 22 according to this embodiment corresponds to an example of a pair identification means, as recited in the claims, that identifies pairs of people who are related to each other based on the attributes of each of the multiple people.

[0034] The graph data generation unit 22 generates graph data including, for example, node data 50 each associated with a plurality of persons including a person of interest, and link data 52 associated with pairs of persons who are related to each other (see Figures 4, 6, 8, and 9).

[0035] For example, as shown in Figure 3, it is assumed that account data of user A is registered in electronic commerce system 40. It is also assumed that account data of user B is registered in golf course reservation system 42. It is also assumed that account data of user C is registered in travel reservation system 44.

[0036] Assume that the value of the IP address data of user A registered in the e-commerce system 40, the value of the IP address data of user B registered in the golf course reservation system 42, and the value of the IP address data of user C registered in the travel reservation system 44 are the same.

[0037] In this case, the graph data generation unit 22 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, as shown in FIG. 4.

[0038] It is assumed that users with the same IP address use the same computer, and therefore in this embodiment, such users are associated with each other.

[0039] Also, for example, it is assumed that account data of users D, E, and F are registered in the electronic commerce system 40 as shown in FIG.

[0040] It is also assumed that the address data values ​​of users D, E, and F registered in the electronic commerce system 40 are the same.

[0041] In this case, the graph data generation unit 22 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 has a relationship with user E, link data 52e indicating that user D has a relationship with user F, and link data 52f indicating that user E has a relationship with user F, as shown in FIG. 6.

[0042] Users who have the same address are assumed to live together, and therefore, in this embodiment, such users are associated with each other.

[0043] 7, for example, it is assumed that account data of user G is registered in electronic commerce system 40. It is also assumed that account data of user H is registered in golf course reservation system 42. It is also assumed that account data of user I is registered in travel reservation system 44.

[0044] Assume that the value of the credit card number data of user G registered in the e-commerce system 40, the value of the credit card number data of user H registered in the golf course reservation system 42, and the value of the credit card number data of user I registered in the travel reservation system 44 are the same.

[0045] In this case, the graph data generation unit 22 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, as shown in Figure 8.

[0046] Users with the same credit card number are presumed to be family members such as parents and children, etc. Therefore, in this embodiment, such users are associated with each other.

[0047] The criteria for determining whether or not a pair of people are related to each other are not limited to those described above.

[0048] Furthermore, the link indicated by the link data 52 that associates people who have been identified as having a relationship with each other as described above will be referred to as an explicit link.

[0049] For example, suppose that a person who is connected to a first person via an explicit link and a person who is connected to a second person via an explicit link have a predetermined number or more of them in common (e.g., three or more people). In this case, in this embodiment, for example, the graph data generation unit 22 generates link data 52 indicating that the first person has a relationship with the second person. The link indicated by the link data 52 generated in this way is referred to as an implicit link.

[0050] 9, it is assumed that link data 52j indicating an explicit link connects node data 50j associated with user J and node data 50k associated with user K. It is also assumed that link data 52k indicating an explicit link connects node data 50j associated with user J and node data 50l associated with user L. It is also assumed that link data 52l indicating an explicit link connects node data 50j associated with user J and node data 50m associated with user M.

[0051] Also, 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. Also, 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. Also, 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.

[0052] In this case, the graph data generation unit 22 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 person who has a relationship with user J.

[0053] Furthermore, for example, if a person who is connected to a first person via an explicit link or an implicit link and a person who is connected to a second person via an explicit link or an implicit link have a predetermined number or more of common people (e.g., three or more), the graph data generating unit 22 may generate link data 52 indicating that the first person has a relationship with the second person (link data 52 indicating an implicit link).

[0054] The graph data generating unit 22 may generate graph data based on person attribute data that is different from account data.

[0055] In this embodiment, for example, the reference person identification unit 24 identifies a reference person who is a person related to the person of interest. Here, the reference person identification unit 24 may identify, as reference persons, a person identified as a person related to the person of interest and a person who has a predetermined number or more of common characteristics with the person of interest among the people identified as related. Furthermore, the reference person identification unit 24 may identify a reference person from among a plurality of people based on the attributes of the person of interest and the attributes of the plurality of people.

[0056] The reference person identification unit 24 may, for example, identify a person associated with node data 50 associated with a person of interest and node data 50 connected by link data 52 indicating an explicit link or an implicit link as a reference person for the person of interest.

[0057] In this embodiment, the relationship identification unit 26, for example, identifies the relationship between the person of interest and the reference person. Here, the relationship identification unit 26 may identify the relationship between the person of interest and the reference person based on the account data of the person of interest and the account data of the reference person. Here, the computer system in which the account data of the person of interest is registered may be different from the computer system in which the account data of the reference person is registered. For example, the relationship between the person of interest and the reference person may be identified based on the account data of the person of interest registered in the e-commerce system 40 and the account data of the reference person registered in the golf course reservation system 42.

[0058] Furthermore, the relationship specification unit 26 may specify the relationship between the person of interest and the reference person as a family.

[0059] The relationship identification unit 26, for example, identifies a pair of node data 50 connected by link data 52. Then, the relationship identification unit 26 generates pair attribute data associated with the pair based on the person attribute data of the two people associated with the pair.

[0060] 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, and pair gender data.

[0061] The IP common flag is a flag that indicates whether the value of the IP address data included in one account data of the pair is the same as the value of the IP address data included in the other account data. For example, if the values ​​of the IP address data are the same on a given day, 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.

[0062] The address commonality flag is, for example, a flag indicating whether the value of the address data included in one account data of the pair is the same as the value of the address data included in the other account data. For example, if the address data values ​​are the same, the address commonality flag may be set to a value of 1, and if the address data values ​​are different, the address commonality flag may be set to a value of 0.

[0063] The credit card number common flag is, for example, a flag indicating whether the value of the credit card number data included in one account data of the pair is the same as the value of the credit card number data included in the other account data. 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.

[0064] The same last name flag is a flag that indicates, for example, whether the last name indicated by the name data included in one of the account data of the pair is the same as the last name indicated by the name data included in the other account data. For example, if the last names indicated by the name data are the same, the value of the same last name flag may be set to 1, and if the last names indicated by the name data are different, the value of the same last name flag may be set to 0.

[0065] The age difference data is, for example, data indicating the difference between the value of age data included in one of the account data of the pair and the value of age data included in the other account data.

[0066] The paired gender data is, for example, data indicating a combination of the value of gender data included in one of the account data of the pair and the value of gender data included in the other account data.

[0067] Then, the relationship identification unit 26 performs clustering using a general clustering method based on the values ​​of the pair attribute data associated with each of the multiple pairs, thereby classifying the multiple pairs into multiple clusters 54 as shown in Figure 10.

[0068] 10 is a diagram schematically illustrating an example of how a plurality of pairs are classified into five clusters 54 (54a, 54b, 54c, 54d, and 54e). The crosses shown in FIG. 10 correspond to pairs. Each of the crosses is located at a position corresponding to the value of the pair attribute data of the pair corresponding to the cross.

[0069] In the example of Figure 10, multiple pairs are classified into five clusters 54, but the number of clusters 54 into which multiple pairs are classified is not limited to five, and for example, multiple pairs may be classified into four clusters 54.

[0070] FIG. 11 is a diagram showing an example of visualization of classification when a plurality of pairs are classified into four clusters 54.

[0071] As shown in FIG. 11, pairs with the same address, the same sex, an age difference of more than X years, and the same last name may be classified into a first cluster. Also, pairs with the same address, the same sex, an age difference of X years or less, and the same last name may be classified into a second cluster. Also, pairs with the same address, different sex, an age difference of more than Y years, and the same last name may be classified into a third cluster. Also, pairs with the same address, different sex, an age difference of Y years or less, and the same last name may be classified into a fourth cluster.

[0072] 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, for example, siblings of the same sex. The third cluster is presumed to be cluster 54 associated with, for example, parents and children of the opposite sex. The fourth cluster is presumed to be cluster 54 associated with, for example, a married couple or siblings of the opposite sex.

[0073] As described above, the relationship identification unit 26 may identify the relationship between the person of interest and the reference person based on the result of clustering based on values ​​associated with relationships between people. Alternatively, the relationship identification unit 26 may identify the relationship between the person of interest and the reference person based on the result of clustering based on at least one of last name, IP address, address, credit card number, age difference, and gender.

[0074] In this embodiment, for example, the closeness score determination unit 28 determines a closeness score indicating the closeness between the person of interest and the reference person based on an index indicating the strength of the relationship between the person of interest and the reference person, in accordance with a judgment criterion corresponding to the relationship between the person of interest and the reference person.

[0075] Here, the proximity score determiner 28 may include trained machine learning models each associated with the above-mentioned clusters 54. For example, if a plurality of pairs are classified into five clusters 54, the proximity score determiner 28 may include five machine learning models.

[0076] The proximity score determination unit 28 may then determine a proximity score indicating the proximity between the person of interest and the reference person based on the output of a trained machine learning model corresponding to the relationship between the person of interest and the reference person when data indicating an index indicating the strength of the relationship between the person of interest and the reference person is input. In this case, the input-output relationship implemented in the trained machine learning model corresponds to the above-mentioned judgment criterion.

[0077] 12, the closeness score determination unit 28 may input, to the n-th machine learning model, input data corresponding to a pair classified into cluster 54 associated with the n-th machine learning model. For example, if the closeness score determination unit 28 includes five machine learning models, the value n described above is any integer between 1 and 5. Then, the closeness score determination unit 28 may determine, as the closeness score value for the pair, the value of the output data output from the n-th machine learning model in response to the input of the input data.

[0078] The input data associated with a pair may include, for example, some or all of the pair attribute data associated with the pair. Furthermore, the input data may include data not included in the pair attribute data. For example, the input data may include data indicating the usage history of the e-commerce system 40, data acquired by the proximity score determination unit 28 from other information sources such as social networking sites, etc. More specifically, for example, the input data may include data indicating the number of calls or messages exchanged between the pair per unit period, the number of gifts sent by one party to the other, the number of mutual friends the pair has, etc.

[0079] Furthermore, the types 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.

[0080] In this embodiment, for example, learning of the n-th machine learning model is performed in advance using a plurality of given training data associated with the n-th machine learning model before the determination of the closeness score by the closeness score determination unit 28. This training data is, for example, prepared in advance so that the determination of the closeness score in the cluster 54 associated with the n-th machine learning model is valid.

[0081] Here, the n-th machine learning model may be subjected to weakly supervised learning. For example, the training data may include learning input data containing the same type of data as the input data input to the n-th machine learning model, and teacher data to be compared with output data output from the n-th machine learning model in response to the input of the learning input data, as shown in FIG.

[0082] Here, for example, it is assumed that the above-mentioned closeness score takes a value of either 0 or 1. For example, if a 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.

[0083] In this case, the training data may include data indicating a valid closeness score value for the corresponding training input data and the probability that this value is valid.

[0084] Then, for example, weakly supervised learning may be performed to update the parameter values ​​of the nth machine learning model based on the values ​​of output data output from the nth machine learning model in response to the input of learning input data included in the training data and the values ​​of the teacher data included in the training data.

[0085] It should be noted that the above-mentioned closeness score does not have to be binary data taking 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) that increases as the relationship between the pair becomes closer, or a multi-level integer value (e.g., an integer between 1 and 10).

[0086] Furthermore, the learning method for the machine learning model is not limited to weakly supervised learning.

[0087] As a specific example, consider a pair of siblings. 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 pair has made 1,200 phone calls, learning may be performed to output output data with a value of 1. Alternatively, if the address data values ​​for the pair are different, one of the pair has sent 2 gifts to the other, and the pair has made 30 phone calls, learning may be performed to output output data with a value of 0.

[0088] The criteria (for example, a 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.

[0089] In this embodiment, the credit score acquisition unit 30 acquires, for example, credit score data indicating the credit score of the reference person.

[0090] The credit score acquisition unit 30 may access the system of a credit information agency and acquire credit score data indicating the credit score of the reference person from the system.

[0091] The credit score obtaining unit 30 may access the card management system 46 and obtain credit score data indicating the credit score of the reference person from the card management system 46.

[0092] In this embodiment, for example, the credibility determination unit 32 determines the credibility of at least one reference person based on information about the reference person and a closeness score between the reference person and the person of interest determined for the reference person.

[0093] Here, an example of the information about the reference person is credit score data acquired by the credit score acquisition unit 30 and indicating the credit score of the reference person.

[0094] Furthermore, the creditworthiness determination unit 32 may determine the credit score of the person of interest as an example of the creditworthiness of the person of interest. For example, the creditworthiness determination unit 32 may determine the weighted average of the credit scores based on the proximity scores as the credit score of the person of interest.

[0095] For example, suppose there are m reference persons for a certain person of interest. In this case, a value p may be calculated by summing the products of the credit score data values ​​for the m reference persons and the closeness scores associated with the pair of the reference person and the person of interest. Then, a value q may be calculated by summing the closeness scores associated with the pair of the reference person and the person of interest for the m reference persons. Then, the value p divided by the value q may be determined as the credit score of the person of interest.

[0096] It should be noted that the creditworthiness determination unit 32 does not need to determine the creditworthiness of the person of interest based on the credit score of the reference person. For example, the creditworthiness determination unit 32 may determine the creditworthiness of the person of interest based on, for example, the income of the reference person, the popularity of the reference person, demographics, etc.

[0097] The creditworthiness of a person of interest determined by the creditworthiness determination unit 32 is not limited to the credit score of the person of interest. For example, the creditworthiness determination unit 32 may determine the credit limit of the person of interest. Furthermore, for example, the creditworthiness determination unit 32 may determine whether or not to permit new user registration for any membership service, or whether or not to permit deferred payment for any online shopping or flea market service.

[0098] Here, an example of the flow of processing for determining the creditworthiness of a certain person of interest, which is performed by the creditworthiness determination system 1 according to this embodiment, will be described with reference to the flow diagram illustrated in FIG. 14. In the following processing example, it is assumed that graph data for multiple people including the person of interest has already been generated, and that clusters 54 associated with multiple pairs have been identified. It is also assumed that a machine learning model associated with each cluster 54 has already been trained. It is also assumed that the initial values ​​of the values ​​p and q are set to 0.

[0099] First, the reference person identification unit 24 identifies, as a reference person, a person corresponding to node data 50 that is connected to the node data 50 corresponding to the person of interest by an explicit link or an implicit link (S101). Here, for example, it is assumed that at least one reference person is identified.

[0100] Then, the relationship identification unit 26 selects one reference person for whom the processes shown in S103 to S109 have not yet been executed from among the reference persons identified in the process shown in S101 (S102).

[0101] Then, the relationship specifying unit 26 specifies the cluster 54 corresponding to the pair of the person of interest and the reference person selected in the process shown in S102 (S103).

[0102] Then, the proximity score determination unit 28 generates input data corresponding to the pair of the person of interest and the reference person selected in the process shown in S102 (S104).

[0103] Then, the proximity score determination unit 28 inputs the input data generated in the process shown in S104 to the trained machine learning model associated with the cluster 54 identified in the process shown in S103 (S105).

[0104] Then, the proximity score determination unit 28 determines the value of the proximity score to be associated with the pair of the person of interest and the reference person based on the output data output from the machine learning model in response to the input executed in the process shown in S105 (S106).

[0105] Then, the credit score acquisition unit 30 acquires the credit score data of the reference person (S107).

[0106] Then, the creditworthiness determination unit 32 adds the product p1 of the value of the credit score data acquired in the process shown in S107 and the value of the closeness score determined in the process shown in S106 to the value p (S108).

[0107] Then, the credibility determination unit 32 adds the value q1 of the closeness score determined in the process shown in S106 to the value q (S109).

[0108] Then, the relationship identification unit 26 checks whether the processes shown in S103 to S109 have been executed for all of the reference persons identified in the process shown in S101 (S110).

[0109] If the processes shown in S103 to S109 have not been executed for all of the reference persons identified in the process shown in S101 (S110: N), the process returns to the process shown in S102.

[0110] If the processes shown in S103 to S109 have been executed for all of the reference persons identified in the process shown in S101 (S110: Y), the creditworthiness determination unit 32 identifies the value p divided by the value q as the credit score of the person of interest (S111), and the process shown in this processing example is terminated.

[0111] In this way, according to this embodiment, it is possible to accurately determine the creditworthiness of a person for whom there is no credit information.

[0112] This embodiment can also be used to determine the credibility of a new person of interest whose corresponding node data 50 is not included in the graph data. For example, based on the person attribute data of the new person of interest, node data 50 corresponding to the person of interest and at least one link data 52 connected to the node data 50 may be generated. Then, a person connected to the node data 50 corresponding to the person of interest by the link data 52 may be identified as a reference person of the person of interest.

[0113] Then, for each of the at least one identified reference person, paired attribute data corresponding to the pair between the reference person and the person of interest may be generated. Then, a cluster 54 corresponding to the relationship between the reference person and the person of interest may be identified based on the paired attribute data and the result of the previously performed clustering.

[0114] Then, for each of the identified at least one reference person, a trained machine learning model corresponding to the identified cluster 54 may be identified. Then, a closeness score for the pair of the reference person and the person of interest may be determined based on an output from the machine learning model when input data corresponding to the pair of the reference person and the person of interest is input to the machine learning model.

[0115] The creditworthiness of the new person of interest may then be determined based on the proximity score and credit score for each reference person, as described above.

[0116] The present invention is not limited to the above-described embodiment.

[0117] For example, the credibility determination unit 32 may perform a first determination to determine the credibility of a person of interest based on information about the person of interest. In the first determination, the credibility of the person of interest may be determined based only on the information about the person of interest, without using information about people other than the person of interest (for example, information about the above-mentioned reference person).

[0118] Then, if the creditworthiness of the person of interest determined by the first determination described above is below a predetermined level (for example, the value of the person's credit score is below a predetermined value), the creditworthiness determination unit 32 may perform a second determination to re-determine the creditworthiness of the person of interest based on information about at least one reference person, as described above.

[0119] In addition, if the creditworthiness of the person of interest cannot be determined in the first determination described above, the creditworthiness determination unit 32 may perform a second determination to re-determine the creditworthiness of the person of interest based on information about at least one reference person, as described above.

[0120] Furthermore, the specific character strings and numerical values ​​described above and the specific character strings and numerical values ​​in the drawings are examples, and the present invention is not limited to these character strings and numerical values. [Explanation of symbols]

[0121] 1 Creditworthiness Determination System, 10 Processor, 12 Memory Unit, 14 Communication Unit, 16 Operation Unit, 18 Output Unit, 20 Person Attribute Data Acquisition Unit, 22 Graph Data Generation Unit, 24 Reference Person Identification Unit, 26 Relationship Identification Unit, 28 Proximity Score Determination Unit, 30 Credit Score Acquisition Unit, 32 Creditworthiness Determination Unit, 40 E-Commerce System, 42 Golf Course Reservation System, 44 Travel Reservation System, 46 Card Management System, 50, 50a, 50b, 50c, 50d, 50e, 50f, 50g, 50h, 50i, 50j, 50k, 50l, 50m, 50n Node data, 52, 52a, 52b, 52c, 52d, 52e, 52f, 52g, 52h, 52i, 52j, 52k, 52l, 52m, 52n, 52o, 52p Link data, 54, 54a, 54b, 54c, 54d, 54e Cluster.

Claims

1. a node data generating means for generating a plurality of node data items each associated with a different person; a pair identification means for identifying a pair of a first person and a second person as a pair of persons related to each other when a value of a certain attribute included in account data of a first person registered in a first computer system is the same as a value of the attribute included in account data of a second person different from the first person registered in a second computer system different from the first computer system; an explicit link data generating means for generating explicit link data that connects the node data associated with each of the two people included in the identified pair; an implicit link data generating means for generating implicit link data connecting first node data and second node data when a predetermined number of node data are common to a first node data among the plurality of node data and connected by the explicit link data to a second node data different from the first node data among the plurality of node data; A graph data generation system including:

2. the implicit link data generation means generates implicit link data connecting the third node data and the fourth node data when a predetermined number or more of the node data are in common between the node data that is connected to third node data among the plurality of node data by the explicit link data or the implicit link data and the node data that is connected to fourth node data that is different from the third node data among the plurality of node data by the explicit link data or the implicit link data; The graph data generation system according to claim 1 .

3. The attribute is an IP address or a credit card number. The graph data generation system according to claim 1 or 2.

4. A plurality of trained machine learning models, each of which is associated with a cluster that classifies pairs of people; a reference person specifying means for specifying, as a reference person, a person associated with the node data connected to the node data associated with the person of interest by the explicit link data or the implicit link data; a relationship specification means for specifying the cluster into which the pair of the person of interest and the reference person is classified, based on a result of clustering based on a value associated with the relationship between the person of interest and the reference person; and a proximity score determination means for determining a proximity score indicating the proximity between the person of interest and the reference person based on an output when data indicating an index indicating the strength of the relationship between the person of interest and the reference person is input to the machine learning model corresponding to the identified cluster. The graph data generation system according to claim 1 .

5. A step in which node data generation means generates a plurality of node data each associated with a different person; a step in which, when a value of a certain attribute included in account data of a first person registered in a first computer system is the same as a value of the attribute included in account data of a second person different from the first person registered in a second computer system different from the first computer system, the pair of the first person and the second person is identified as a pair of persons who are related to each other; a step in which explicit link data generating means generates explicit link data that connects the node data associated with each of the two people included in the identified pair; an implicit link data generating means for generating implicit link data connecting first node data and second node data when the node data connected to first node data of the plurality of node data by the explicit link data and the node data connected to second node data different from the first node data of the plurality of node data by the explicit link data have a predetermined number of common elements; A graph data generation method including:

6. generating a plurality of node data items each associated with a different person; a step of identifying a pair of a first person and a second person as a pair of persons related to each other when a value of an attribute included in account data of a first person registered in a first computer system is the same as a value of the attribute included in account data of a second person different from the first person registered in a second computer system different from the first computer system; generating explicit link data connecting the node data associated with each of the two people included in the identified pair; generating implicit link data connecting the first node data and the second node data when a predetermined number of node data or more are in common between the node data connected to first node data of the plurality of node data by the explicit link data and the node data connected to second node data different from the first node data of the plurality of node data by the explicit link data; A program that causes a computer to execute the following.

Images