JP6560323B2 - Determination device, determination method, and determination program - Google Patents

Description

  The present invention relates to a determination device, a determination method, and a determination program.

  In general, when a financial institution such as a bank determines a loan allowance frame, business data provided by a data provider specializing in collection and analysis of business (company) information is referred to. Such business operator data is generated based on, for example, business operators' financial statements (for example, financial statements, profit and loss statements, etc.). That is, the financial institution uses the business data provided by the business in order to determine the credit information for determining whether or not the loan amount to each business is appropriate.

  Here, not only data with long update intervals (static data) such as financial data published in the fiscal year end, but also models with input of short update intervals (dynamic data) such as stock prices are created. And the system which can perform the newest company evaluation in time is proposed (for example, patent document 1). In addition to the rapid spread of the Internet in recent years, in addition to information based on business operators' financial statements as in the past, information users share information about businesses on the Internet, and the information they share is unified. A technique for enhancing the objectivity of business operator data by performing integrated management is known (for example, Patent Document 2).

JP 2000-348015 A JP 2010-26602 A

  However, in the above-described conventional technology, it is not always possible to determine the creditworthiness with respect to the operator from various viewpoints. Specifically, the above-mentioned conventional technology only shares information such as financial information of publicly disclosed business operators, business results of business operators that information users can know, and industry news. Even if they are integrated, it is difficult to appropriately evaluate the creditworthiness of the operator (for example, the growth potential and stability of the operator over the future). In addition, since data such as stock prices is not disclosed to unlisted companies, there is a shortage of data for corporate evaluation in some cases, making it difficult to determine the creditworthiness of the operator.

  The present application has been made in view of the above, and an object thereof is to provide a determination device, a determination method, and a determination program that can determine the creditworthiness of a business operator from various viewpoints.

  The determination apparatus according to the present application includes: an acquisition unit that acquires information about a user who uses a predetermined commercial transaction site on a network; and a first business that performs a commercial transaction at the commercial transaction site among information about the user acquired by the acquisition unit And a determination unit that determines the creditworthiness of the first operator based on an action history on the network of the first user who is a user with a transaction history.

  According to one aspect of the embodiment, there is an effect that it is possible to determine the creditworthiness with respect to the operator with high accuracy.

FIG. 1 is a diagram illustrating an example of a determination process according to the embodiment. FIG. 2 is a diagram illustrating a configuration example of the determination system according to the embodiment. FIG. 3 is a diagram illustrating a configuration example of the determination apparatus according to the embodiment. FIG. 4 is a diagram illustrating an example of the business operator information storage unit according to the embodiment. FIG. 5 is a diagram illustrating an example of a usage history table according to the embodiment. FIG. 6 is a diagram illustrating an example of a browsing history table according to the embodiment. FIG. 7 is a diagram illustrating an example of a payment information table according to the embodiment. FIG. 8 is a diagram illustrating an example of the attribute information table according to the embodiment. FIG. 9 is a diagram illustrating an example of a model storage unit according to the embodiment. FIG. 10 is a flowchart illustrating a generation processing procedure performed by the determination apparatus according to the embodiment. FIG. 11 is a flowchart illustrating a determination processing procedure performed by the determination apparatus according to the embodiment. FIG. 12 is a diagram illustrating an example of a determination process according to the modification. FIG. 13 is a hardware configuration diagram illustrating an example of a computer that realizes the function of the determination apparatus.

  Hereinafter, a mode for carrying out a determination device, a determination method, and a determination program according to the present application (hereinafter referred to as “embodiment”) will be described in detail with reference to the drawings. Note that the determination device, the determination method, and the determination program according to the present application are not limited by this embodiment. In the following embodiments, the same portions are denoted by the same reference numerals, and redundant description is omitted.

[1. Example of judgment process)
First, an example of the determination process according to the embodiment will be described with reference to FIG. FIG. 1 is a diagram illustrating an example of a determination process according to the embodiment. Specifically, FIG. 1 illustrates an example in which the determination device 100 according to the embodiment performs a process of determining the creditworthiness of a business opening a store on a predetermined commercial transaction site on the network.

  In the example illustrated in FIG. 1, the determination device 100, the user terminal 10, and the web server 30 are communicably connected via a communication network (for example, the Internet) (not shown).

  A determination apparatus 100 illustrated in FIG. 1 is a server apparatus that performs a process of determining a credit rating for a business operator. In the embodiment, the determination apparatus 100 determines the reliability of a business opening a store in a shopping mall that is a predetermined commercial transaction site on the network provided by the web server 30.

  In the embodiment, a business that opens a store in a shopping mall and whose business is subject to creditworthiness in the determination process according to the embodiment is referred to as a “first business”. Further, in the embodiment, among the businesses that open a store in the shopping mall, a business that is different from the first business and that can serve as learning data of a model for determining credit in the determination process according to the embodiment. Indicated as “second operator”. In addition, when there is no particular distinction between the first business operator and the second business operator, it is simply written as “business operator”.

The user terminal 10 ₁ and 10 ₂ shown in FIG. 1 is an information processing device used by a user. Specifically, the user terminals 10 ₁ and 10 ₂ allow the user to browse various web pages, purchase products on a website (for example, a shopping mall on a network), or evaluate product information (so-called review). Used to post. In the example of FIG. 1, the user terminal 10 ₁ is utilized by the user U01 which is an example of a user, the user terminal 10 ₂ is used by the user U02 is an example of a user. Also, unless otherwise identified and the user terminal 10 ₁ and 10 ₂ are collectively referred to as "user terminal 10". In the example of FIG. 1, the user terminal 10 is, for example, a smartphone or a tablet terminal.

  In the embodiment, among users, a user who has a transaction history with a first business operator (for example, a user who has a history of purchasing products at a store where the first business operator opened a store) is referred to as a “first user”. write. In addition, among users, a user who has a transaction history with the second operator is referred to as a “second user”. When the same user has a transaction history for both the first business operator and the second business operator, the user can be a “first user” or a “second user” depending on the processing status. Further, when there is no particular distinction between the first user and the second user, it is simply expressed as “user”.

  The web server 30 illustrated in FIG. 1 is a server device that provides various web pages when accessed from the user terminal 10. In the example illustrated in FIG. 1, the web server 30 provides a shopping mall site on the network. In a shopping mall provided by the web server 30, each business operator opens a store that handles products of various categories.

  Moreover, the web server 30 acquires the information regarding the user who uses a shopping mall through the website provided. Information about the user is stored in, for example, a storage unit included in the web server 30 or a predetermined external storage device. The information about the user means, for example, information transmitted from the user terminal 10 in accordance with a user operation when using a service provided from the web server 30. Specifically, the information about the user includes user attribute information (age, sex, etc.), shopping behavior in the shopping mall, payment information in the shopping mall, user rank in the shopping mall (for example, the user is a good member) Or not).

  In addition, the information regarding a user may be acquired via services other than a shopping mall. For example, the web server 30 provides various services such as a portal service, a news distribution service, a schedule management service, and a restaurant reservation service as services other than the shopping mall. And the web server 30 may acquire the information regarding a user through utilization of these services. Specifically, the web server 30 can acquire, as information about the user, information about news content browsed by the user, information about a schedule registered by the user with the service, and the like.

  By the way, a business that opens a business transaction site on a network such as a shopping mall provided by the web server 30 has a small business structure, a relatively new establishment, a private listing, or a personality. You may be an employer and do not have enough funds. For this reason, there is a demand for these businesses to obtain loans, and there is a demand for financial institutions and the like to obtain opportunities to finance these businesses.

  In general, when a financial institution makes a loan, it determines whether the loan is possible and how much the loan is based on the creditworthiness (credit) of the operator. The creditworthiness of such a business is measured by, for example, the financial institution itself or a specialized organization that evaluates the business. However, small businesses often do not publish quarterly or semi-annual financial information, capital relationships, etc., and there are cases where specialized institutions and financial institutions cannot measure the creditworthiness of businesses. In this case, the business operator cannot receive the loan, and the financial institution misses the opportunity for the loan.

  Therefore, the determination apparatus 100 according to the embodiment determines the creditworthiness of a business operator who opens a store based on information about a user who has a transaction history in a store that has opened a store in a shopping mall.

  This is a user whose first user who uses the store opened by the first operator is solid and reliable (for example, a user who has not been paid for settlement, a user who views a relatively large amount of economic news, etc.). The reason is that the first operator is also assumed to be a reliable operator. That is, the determination apparatus 100 determines a store having a relatively large number of reliable customers as a store with high creditworthiness, and further determines the first business operator that operates the store as a business with high creditworthiness.

  In such determination processing, the determination apparatus 100 is information related to the second business operator who is a business operator for which creditworthiness is set in advance (for example, some creditworthiness information is given by a financial institution, a specialized institution, or a shopping mall side). May be used. For example, the determination apparatus 100 learns the correlation between the second business operator and information on the second user who has a transaction history with the second business operator. Thereby, the determination apparatus 100 learns what information a user who has a transaction with a store with high (or low) creditworthiness has. Specifically, the determination apparatus 100 generates a model in which the correlation between the information relating to the second operator and the second user is learned. And the determination apparatus 100 determines the reliability of a 1st provider using the produced | generated model. As a result, the determination apparatus 100 can accurately determine the degree of credit even for a relatively small first business operator that has not released information such as financial status and stock price. Hereinafter, the determination process according to the embodiment will be described along the flow with reference to FIG. 1.

  In the example shown in FIG. 1, the shopping mall provided by the web server 30 is operated by a business operator A01 (an example of a first business operator) that is a business operator to which credit information is not given from a financial institution or a specialized institution. There is a store page 61 which is a web page of the store AAA. Further, in the shopping mall, there is a store page 71 that is a web page of a store BBB operated by a business operator A02 (an example of a second business operator) that is a business operator whose credit rating has already been set (having credit information). To do. In the example of FIG. 1, the reliability is assumed to be indicated by a score (index value) having a maximum of 100 and a minimum of 0. In the example of FIG. 1, a credit rating indicating a relatively high score of “85” is set for the business operator A02, and such a numerical value indicates that the business operator A02 is evaluated as a business enterprise having a relatively high credit quality. Show. In addition, although illustration is abbreviate | omitted in FIG. 1, in the shopping mall which the web server 30 provides, a considerable number of 2nd business so that sufficient information for performing the determination process which concerns on embodiment is obtained. There shall be a person.

With respect to the store of such a shopping mall, the user takes an action such as browsing a product or purchasing a product. For example, the user terminal 10 ₁ in accordance with operation by the user U01, accesses the shopping mall web server 30 is provided, utilizing a shopping mall (step S11). For example, the user U01 purchases a product from the store AAA. In the example of FIG. 1, the user U01 corresponds to the first user because the user U01 is a user having a transaction history with the operator A01, which is the first operator.

Similarly, the user terminal 10 ₂ in accordance with operation by the user U02, accesses the shopping mall web server 30 is provided, utilizing a shopping mall (step S12). For example, the user U02 purchases a product from the store BBB. In the example of FIG. 1, the user U02 corresponds to the second user because the user U02 is a user who has a transaction history with the operator A02, which is the second operator. Although not shown in the example of FIG. 1, there are a plurality of users U01 and users U02, and a considerable number of accesses sufficient to obtain a sufficient amount of information for performing the determination processing according to the embodiment are provided. Suppose you are going.

  In addition, it is assumed that the user U01 and the user U02 appropriately use services other than the shopping mall provided by the web server 30. For example, it is assumed that the user U01 and the user U02 use a news service distributed via a network, a schedule service provided by the web server 30, and the like. These user behavior histories and the like are all included in the information related to the user according to the embodiment.

  Here, the determination apparatus 100 acquires from the web server 30 the preset creditworthiness of the second operator and information regarding the user U02 (second user) who has made a transaction with the second operator (step S13). ).

  And the determination apparatus 100 produces | generates the model which learned the correlation with the reliability regarding a 2nd provider, and the information regarding a 2nd user based on the acquired information (step S14). The model is a model for indicating the correlation between the creditworthiness of the business operator and the information on the user. For example, when information on the first user is input to the model, the model corresponding to the first user is displayed. The credit rating of one operator is output. Although details of the model generation processing will be described later, the determination apparatus 100 uses, for example, the reliability of the operator A02 as correct data (teacher data, objective variables), and features (explanatory variables) various information about the second user corresponding to the operator A02. A regression learning model such as a support vector machine is generated.

  Thereafter, the determination apparatus 100 acquires information on the user U01 (first user) who has made a transaction with the business operator A01, which is the first business operator to be processed (step S15).

  And the determination apparatus 100 outputs the score (index value) which shows the reliability of provider A01 by inputting into the model which produced | generated the information regarding provider U01 to process and user U01 with a transaction history. The determination apparatus 100 determines the credit rating of the first operator based on the output score (step S16).

  That is, the model generated by the determination apparatus 100 outputs a score indicating the creditworthiness of the business operator A01 based on the information about the first user when the information about the first user who has business with the business operator A01 is input. . For example, if the information related to the user U01 is very similar to the information related to the user U02, the model outputs a score similar to “85”, which is the credit rating of the business operator A02, as the credit quality of the business operator A01. The determination apparatus 100 does not acquire information such as the financial status of the business operator A01 by this determination processing, but based on information related to the user U01 with a transaction (for example, various behavior histories of the user U01), the business operator A01. Can be determined.

  As described above, the determination apparatus 100 according to the embodiment acquires information about a user who uses a predetermined commercial transaction site on the network. And the determination apparatus 100 is a 1st provider based on the action history on the network of the 1st user who is a user with a transaction history with the 1st provider who performs a commercial transaction in the commercial transaction site among the information regarding the acquired user. Judgment of creditworthiness.

  That is, the determination apparatus 100 according to the embodiment includes not only financial information generally used for measuring the creditworthiness of a business operator, but also information on users who have transactions with a store opened in a commercial transaction site such as a shopping mall. Use to determine the creditworthiness of the operator. Thereby, the determination apparatus 100 can determine the creditworthiness with respect to the operator from various viewpoints regardless of the viewpoint of the financial information, the sales amount, etc. of the operator. Hereinafter, the configuration of the determination apparatus 100 that performs the above-described processing and the determination system 1 including the determination apparatus 100 will be described in detail.

[2. (Configuration of judgment system)
The configuration of the determination system 1 including the determination apparatus 100 according to the embodiment will be described with reference to FIG. FIG. 2 is a diagram illustrating a configuration example of the determination system 1 according to the embodiment. As illustrated in FIG. 2, the determination system 1 according to the embodiment includes a user terminal 10, an operator terminal 20, a web server 30, and a determination device 100. These various apparatuses are communicably connected via a network N (for example, the Internet) in a wired or wireless manner. Note that the determination system 1 illustrated in FIG. 2 may include a plurality of user terminals 10, a plurality of business operator terminals 20, and a plurality of web servers 30.

  The user terminal 10 is, for example, an information processing apparatus such as a smartphone, a desktop PC (Personal Computer), a notebook PC, a tablet terminal, a mobile phone, a PDA (Personal Digital Assistant), or a wearable device. It is. The user terminal 10 accesses the web server 30 in accordance with an operation by the user, so that various contents (for example, the store page 61 of the store AAA and the store BBB shown in FIG. Page 71). Then, the user terminal 10 displays the acquired content on a display device (for example, a liquid crystal display). In the present specification, the user and the user terminal 10 may be identified with each other. For example, “providing content to the user” may actually mean “providing content to the user terminal 10 used by the user”.

  The business entity terminal 20 is an information processing device used by the business operator. Specifically, the business entity terminal 20 performs a procedure for opening a store in a shopping mall, a procedure for listing products in a store, and the like in accordance with the operation of the business operator. In the embodiment, the business operator is a concept that includes a wide range of business operators, and is not limited to a stock company or the like, and may be an individual business owner or a legal entity.

  The web server 30 is a server device that provides various contents such as a web page when accessed from the user terminal 10. In the example of FIG. 1, an example is shown in which the web server 30 provides a website corresponding to a shopping mall, and provides web pages of the business operators A01 and business operators A02 that are open in the shopping mall. However, the service provided by the web server 30 is not limited to this, and the web server 30 is, for example, a news site, a weather forecast site, a shopping site, a finance (stock price) site, a route search site, a map providing site, a travel site, a food and drink Various web pages related to store introduction sites, web blogs, etc. may be provided.

  In addition, the web server 30 may acquire information on each website and information regarding users that can be acquired on the network. The information on the user is information transmitted from the user terminal 10 according to the user's operation when using services provided from various websites, for example. For example, the information about the user includes attribute information (gender, age, address, occupation, annual income, etc.) registered by the user when using the service. Moreover, the information regarding a user contains a user's action information. For example, user behavior information includes browsing information for browsing news content by users, information on search queries transmitted on search sites, information on purchase behavior on shopping sites and reviews from users, and SNS ( Information on messages on the Social Networking Service site.

  In the embodiment, a plurality of web servers 30 may exist. In this case, the web server 30 that provides the user terminal 10 with the shopping mall may be different from the web server 30 that acquires information about the user.

  The determination apparatus 100 acquires information on a user who uses a predetermined commercial transaction site (such as the shopping mall shown in FIG. 1) on the network, and among the acquired information on the user, the first business operator who performs a commercial transaction on the commercial transaction site It is a server apparatus which determines the creditworthiness of a 1st provider based on the action history on the network of the 1st user who is a user with a transaction history. Note that the determination device 100 may also serve as the configuration of the web server 30 described above. That is, the determination apparatus 100 itself may provide a shopping mall, and the determination apparatus 100 itself may acquire information from the user terminal 10 that has accessed the shopping mall.

  In addition, the determination device 100 performs credit rating of the operator by investigating, acquiring, and managing information processing terminals installed in financial institutions that actually provide loans to the operator and financial information related to the operator. You may communicate mutually with the information processing terminal etc. which were installed in the specialized organization. For example, the determination apparatus 100 may acquire rating information (corresponding to “credit score” in the embodiment) of the second operator from an information processing terminal installed in a financial institution or a specialized institution. In addition, when the financial information of the business operator A01 or the business operator A02 is published, the determination apparatus 100 may appropriately acquire financial information from an information processing terminal installed in a financial management company or the like. In addition, when the financial information related to the business operator A01 or the business operator A02 can be acquired from a public institution or the like, the determination apparatus 100 may appropriately access the public institution or acquire information. These processes may be performed using, for example, a predetermined program that crawls on a network and acquires information.

[3. (Configuration of judgment device)
Next, the configuration of the determination apparatus 100 according to the embodiment will be described with reference to FIG. FIG. 3 is a diagram illustrating a configuration example of the determination apparatus 100 according to the embodiment. As illustrated in FIG. 3, the determination apparatus 100 includes a communication unit 110, a storage unit 120, and a control unit 130. The determination device 100 includes an input unit (for example, a keyboard and a mouse) that receives various operations from an administrator who uses the determination device 100, and a display unit (for example, a liquid crystal display) that displays various types of information. You may have.

(About the communication unit 110)
The communication unit 110 is realized by, for example, a NIC (Network Interface Card). The communication unit 110 is connected to a communication network by wire or wireless, and transmits / receives information to / from the user terminal 10 or the like via the communication network.

(About the storage unit 120)
The storage unit 120 is realized by, for example, a semiconductor memory device such as a RAM (Random Access Memory) or a flash memory, or a storage device such as a hard disk or an optical disk. The storage unit 120 according to the embodiment includes a provider information storage unit 121, a feature information storage unit 122, and a model storage unit 127. Hereinafter, each storage unit will be described in order.

(About the company information storage unit 121)
The business operator information storage unit 121 stores information related to a business operator who opens a store on a predetermined commercial transaction site. Information stored in the business operator information storage unit 121 is acquired from the web server 30 via, for example, a network. Here, FIG. 4 illustrates an example of the provider information storage unit 121 according to the embodiment. As illustrated in FIG. 4, the business operator information storage unit 121 includes items such as “business operator ID”, “store”, “corporate personality”, “credit”, and “customer information”.

  “Business ID” indicates identification information for identifying a business that has opened a store on a commercial transaction site. In the present specification, identification information may be used as a reference symbol used for explanation. For example, a business operator whose business ID is “A01” may be described as “business business A01”.

  “Store” indicates the name of the store where the business operator opens a store on the commercial transaction site. In addition, a business operator may open a plurality of stores instead of opening only one store at one commercial transaction site. “Corporate status” indicates the legal status of the operator.

  “Credit level” indicates the credit level of the business. The credit rating is stored in the business operator information storage unit 121 after the credit rating is determined by the determination unit 134 described later. For example, for an operator whose credit rating has not been determined at present (first operator in the embodiment, which corresponds to the operator A01 in FIG. 4), the credit rating of the operator A01 who received the loan is an item of credit rating. Is blank (in the example of FIG. 4, indicated as “none”). Also, for example, a business operator whose credit rating has already been determined or whose credit rating has been set by a specialized organization (second business operator in the embodiment. In FIG. 4, business operator A02, business operator A03, and business operator A04 are applicable. Yes), a numerical value indicated by 0 to 100 is stored in the item of reliability.

  In the “customer information”, information related to users who are customers of a store where a business operator opens a store is stored. In the example shown in FIG. 4, information stored in “customer information” is represented by a concept such as “B01”. Actually, the item of customer information includes information for identifying a user with a transaction history. In addition, various pieces of information such as the number of users who are customers, specific transaction details, and transaction amounts are stored.

  That is, in FIG. 4, as an example of data stored in the business operator information storage unit 121, the business operator A01 identified by the business operator ID “A01” has opened a store “AAA”, and its legal entity is “ This indicates that the credit rating is “None” and the customer information is “B01”.

  Although illustration in FIG. 4 is omitted, the business operator may open a plurality of stores instead of opening a single store at one commercial transaction site.

(About the feature information storage unit 122)
The feature information storage unit 122 stores information about features used in the model generation process. Specifically, the feature information storage unit 122 stores various information regarding users on the network as feature information. As shown in FIG. 3, the feature information storage unit 122 includes data tables such as a usage history table 123, a browsing history table 124, a payment information table 125, and an attribute information table 126. Hereinafter, each data table will be described in order.

(About the usage history table 123)
The usage history table 123 stores the usage history of the store by the user as an example of information related to the user. Here, FIG. 5 shows an example of the usage history table 123 according to the embodiment. As illustrated in FIG. 5, the usage history table 123 includes items such as “user ID”, “use store”, and “number of transactions”.

  “User ID” indicates identification information for identifying the user or the user terminal 10 used by the user. “Used store” indicates the name of the store used by the user. “Number of transactions” indicates the number of times the user has performed transactions with individual stores (for example, purchase of goods, use of services, etc.).

  That is, in FIG. 5, as an example of data stored in the usage history table 123, the user U01 identified by the user ID “U01” uses “Store AAA”, “Store EEE”, “Store FFF”, and the like. It has a history, the number of transactions with the store AAA is “16” times, the number of transactions with the store EEE is “7” times, and the number of transactions with the store FFF is “5” times. Show.

(About the browsing history table 124)
The browsing history table 124 stores a history of browsing content by the user. In the embodiment, the content includes a wide range of web pages, application content, and the like on which various information is provided provided via a network. In the example of FIG. 6, news content is taken as an example among the content provided to the user. The news content includes, for example, news content on a news site provided from the web server 30, news content provided via a news dedicated application, and the like. Here, FIG. 6 shows an example of the browsing history table 124 according to the embodiment. As illustrated in FIG. 6, the browsing history table 124 includes items such as “user ID”, “total period”, and “browsing history”. The “browsing history” has a middle item “news content”. Further, the “news content” has small items such as “category”, “number of times of browsing”, and “viewing time”.

  “User ID” corresponds to the same item shown in FIG. “Total period” indicates a period during which the user's browsing history is totaled. “Browsing history” indicates the browsing history of the user. “News content” indicates information related to browsing of news content in the browsing history.

  “Category” indicates a category of individual news content. Categories are tagged to individual news content, for example, by those who provide news distribution services. “Number of browsing” indicates the number of times the news content has been browsed by category. “Browsing time” indicates the time (in the example of FIG. 6, the unit is “minute”) for browsing the news content by category.

  That is, in FIG. 6, as an example of data stored in the browsing history table 124, the browsing history of news content in the counting period from “October 1, 2017 to October 31, 2017” of the user U01 is “ The number of browsing times of the “Economy” category is “21” and the browsing time is “110” minutes, the number of browsing times of the “Sports” category is “5” and the browsing time is “40” minutes, This indicates that the number of browsing of the “entertainment” category is “3” and the browsing time is “20” minutes.

  6 shows an example in which the browsing history is stored with the aggregation period set to “1 month”. However, the aggregation period is not limited to this example, and may be changed as appropriate according to a model generation process and a credit determination process described later. May be.

(About payment information table 125)
The payment information table 125 stores information related to payment at a commercial transaction site among information related to users. Here, FIG. 7 shows an example of the payment information table 125 according to the embodiment. As shown in FIG. 7, the settlement information table 125 includes items such as “user ID”, “total period”, and “transaction history”. The “transaction history” has small items such as “number of payments”, “number of unpaid times”, and “number of installments paid”.

  “User ID” and “Total period” correspond to the same items shown in FIG. “Transaction history” indicates a history related to a user's transaction. “Number of payments” indicates the number of times the user has made a payment on the commercial transaction site. “Number of unpaid times” indicates the number of transactions that have been made on the commercial transaction site, but have been unpaid for the reason that payment is not properly performed, for example. “Number of installment payment usage” indicates the number of times the installment payment method is adopted in the settlement.

  That is, in FIG. 7, as an example of data stored in the settlement information table 125, the transaction count “30” is shown as a transaction history in the total period from “October 1, 2017 to October 31, 2017” of the user U01. "" And unpaid times "2", and the number of installments used is "5".

(About the attribute information table 126)
The attribute information table 126 stores user attribute information among information about users. Here, FIG. 8 shows an example of the attribute information table 126 according to the embodiment. As shown in FIG. 8, the attribute information table 126 has items such as “user ID” and “attribute information”. The “attribute information” has small items such as “age”, “sex”, “address”, “occupation”, “language”, and “user rank”.

  “User ID” corresponds to the same item shown in FIG. “Attribute information” indicates user attribute information. “Age” indicates the age of the user. “Gender” indicates the gender of the user. “Address” indicates the user's address. “Occupation” indicates the occupation of the user. “Use language” indicates a language that can be used by the user. In the example of FIG. 8, information stored in the items of “occupation” and “language” is represented by concepts such as “E01” and “F01”. The language name (Japanese, English, etc.) is stored.

  “User rank” indicates the rank of the user in the commercial transaction site. The rank of the user on the commerce site is the rating information of the user based on the original criteria of the commerce site. For example, the user pays a fixed amount every month (paid member) or the transaction amount exceeding a predetermined amount It is a ranking made by various information such as being an excellent member who pays. In the example of FIG. 8, the information stored in the item “user rank” is represented by a concept such as “G01”. However, in practice, a specific rank name (rank corresponding to a good member, a normal member, or the like) is used. Is stored.

  That is, in FIG. 8, as an example of data stored in the attribute information table 126, the attribute information of the user U01 is age “30's”, sex “female”, and address “A prefecture. , Occupation “E01”, language “F01”, and user rank “G01”.

(About the model storage unit 127)
The model storage unit 127 stores information related to the model generated by the determination apparatus 100. Here, FIG. 9 shows an example of the model storage unit 127 according to the embodiment. As illustrated in FIG. 9, the model storage unit 127 includes items such as “model ID”, “information update date”, and “learning data”. Further, the “learning data” has small items such as “the business ID of the second business” and “second user information”.

  “Model ID” indicates identification information for identifying a model. “Information update date” indicates the date on which the model was updated. “Learning data” indicates data used for model generation (learning). “The business operator ID of the second business operator” indicates the business operator ID of the second business operator used for generating the model. Note that the business ID shown in FIG. 9 is the same as the business ID shown in FIG. “Second user information” indicates information of a second user corresponding to each second operator. In the example of FIG. 9, the information stored in the “second user information” is indicated by a concept such as “H01”, but actually, information for identifying the second business operator and the user who has a transaction history. In addition, various pieces of information such as the number of second users as customers of the respective second business operators, specific transaction contents, and transaction amounts are stored.

  That is, in FIG. 9, as an example of data stored in the model storage unit 127, the model M01 identified by the model ID “M01” is updated to “November 1, 2017”, and the model M01 For example, the second business operator that has become the learning data is the business operator identified by the business operator ID “A02” or “A03”, and the second user information corresponding to the business operator A02 is “H01”. The second user information corresponding to the user A03 is “H02”.

(About the control unit 130)
The control unit 130 is a controller, for example, various programs (an example of a determination program) stored in a storage device inside the determination apparatus 100 by a CPU (Central Processing Unit), an MPU (Micro Processing Unit), or the like. This is realized by executing the RAM as a work area. The control unit 130 is a controller, and is realized by an integrated circuit such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA).

  As shown in FIG. 3, the control unit 130 according to the embodiment includes an acquisition unit 131, a generation unit 132, a reception unit 133, and a determination unit 134, and has information processing functions and operations described below. Realize or execute. Note that the internal configuration of the control unit 130 is not limited to the configuration illustrated in FIG. 3, and may be another configuration as long as the information processing described below is performed. Further, the connection relationship between the processing units included in the control unit 130 is not limited to the connection relationship illustrated in FIG. 3, and may be another connection relationship.

(About the acquisition unit 131)
The acquisition unit 131 acquires various types of information. For example, the acquisition unit 131 acquires information about a user who uses a predetermined commercial transaction site on the network. Specifically, the acquisition unit 131 acquires an action history on the network of a first user who is a user having a transaction history with a first business operator who performs a commercial transaction at a commercial transaction site.

  For example, the acquisition unit 131 acquires a browsing history of browsing content by the first user as an action history on the network of the first user. Specifically, the acquisition unit 131 acquires information on web pages browsed by the user and information on application content.

  More specifically, the acquisition unit 131 acquires the number of times or time when the first user browses the news content in a predetermined period (for example, the aggregation period shown in FIG. 5) and the category of the browsed news content.

  Moreover, the acquisition part 131 acquires the information regarding a user's transaction and payment in a commercial transaction site. Specifically, the acquisition unit 131 acquires information on at least one of the number of payments at the first user's commercial transaction site, the number of unpaid payments, and the number of use of installments.

  The acquisition unit 131 acquires the attribute information of the first user. For example, the acquisition unit 131 acquires information related to the rank of the first user's commercial transaction site. In addition, the acquisition unit 131 acquires at least one of the first user's age, sex, address, occupation, and language used.

  The acquisition unit 131 acquires not only information related to the first user but also information related to the second user when model generation processing by the generation unit 132 described later is performed.

  Moreover, the acquisition part 131 may acquire various information about the information acquirable via a network besides the action log | history and attribute information which were mentioned above.

  For example, the acquisition unit 131 may acquire information related to the evaluation from the user with respect to the business operator at the commercial transaction site. For example, the commerce site has a system that accepts evaluations from users (for example, scores indicated by numerical values in five stages) for each business (for each store), and receives scores and reviews for each product. In this case, the acquisition unit 131 may acquire information such as the contents of evaluation performed by each user, the number of evaluations performed, and the details of specific reviews.

  In addition, the acquisition unit 131 acquires information on the transition of the number of reactions from the user to the store or product opened by the business operator, or the evaluation value transition of the user score on the store or product opened by the business operator. May be. That is, the acquisition unit 131 acquires an increase / decrease in the number of reviews posted in a predetermined period and a transition in an increase / decrease in scores.

  Moreover, the acquisition part 131 may acquire the word contained in the review from the user with respect to the store or goods which the company opened as the information regarding the evaluation from the user with respect to the company. For example, the acquisition unit 131 performs morphological analysis on reviews from the user, extracts words included in the reviews, and acquires information about the extracted words. Further, the acquisition unit 131 may acquire only words having high weights by using an algorithm for scoring each word appearing in the review, such as tf-idf (Term Frequency-Inverse Document Frequency). In addition, the acquisition unit 131 may acquire a comment (that is, a word included in the comment) returned by the operator in response to a review from the user for a store or a product opened by the operator.

  Moreover, the acquisition part 131 may acquire the information regarding the reliability set beforehand by the 2nd provider. For example, the acquisition unit 131 acquires the creditworthiness of a business set by a specialized period such as a rating company. Or the acquisition part 131 may acquire the creditworthiness uniquely set in the commercial transaction site (shopping mall etc.). In this case, the credit rating may be set by various methods such as a store evaluation at the commerce site (the above five-stage numerical value) or a numerical value set corresponding to the store ranking in the shopping mall. It may be what was done. In addition, the credit rating may be represented not only by a numerical value but also by a binary value such as “whether or not the operator has trust”.

  The acquisition unit 131 stores the acquired information in the storage unit 120 as appropriate. For example, the acquisition unit 131 stores information related to the transaction at the commercial transaction site in the usage history table 123 and the payment information table 125. In addition, the acquisition unit 131 stores information regarding the browsing behavior of the user on the network in the browsing history table 124. The acquisition unit 131 stores user attribute information in the attribute information table 126. Further, the acquisition unit 131 stores information related to the creditworthiness of the business operator in the business operator information storage unit 121. Further, the acquisition unit 131 may appropriately acquire information necessary for processing from the storage unit 120.

  In addition, the acquisition part 131 may acquire the information of various users and a provider at random, without specifying the user and provider who acquire information. For example, the acquisition unit 131 uses a program such as a search robot used for a search engine or the like to crawl on the Internet, thereby acquiring information about the user or the provider as needed, or updating the acquired information. May be.

  Further, the acquisition unit 131 may acquire such information when the financial information of the business operator can be acquired from, for example, a financial institution or a financial management company. When financial information is acquired by the acquisition unit 131, the determination unit 134 described later may determine the credit rating for the first operator in consideration of the financial information.

(About the generator 132)
The production | generation part 132 produces | generates the model for determining the reliability of a 1st provider. Specifically, the generation unit 132 is a provider different from the first provider among the companies that perform the commercial transaction on the commercial transaction site, and the creditworthiness of the second provider that is the provider whose creditworthiness is determined in advance. And a model for determining the creditworthiness of the operator based on the correlation between the second user and information related to the second user who is a user with the second operator (hereinafter, this model is referred to as “first model”). Generate).

  For example, the generating unit 132 determines whether the second user's commerce site has at least one of the number of payments, the number of unpaid payments, or the number of use of installments, and the correlation between the creditworthiness of the second operator. One model is generated.

  Further, the generation unit 132 determines the first model based on the number of times or time when the second user browses the news content in the predetermined period, and the correlation between the browsed news content category and the creditworthiness of the second operator. May be generated. In addition, the production | generation part 132 may produce | generate a 1st model not only with above-described information but with the various information regarding the 2nd user acquired by the acquisition part 131 as a feature.

  For example, the generation unit 132 generates, as the first model, a learning model in which each piece of information related to the second user acquired by the acquisition unit 131 is used as a feature and the preset creditworthiness of the second operator is correct data. Specifically, the generation unit 132 generates a first model that recursively learns what kind of weight is given to what kind of information among the pieces of information about the second user to be a feature. In other words, in the determination of the creditworthiness of the second operator acquired as the correct answer data among the information related to the second user, the generation unit 132 has a greater influence on the determination of the creditworthiness. A first model indicating is generated.

  Below, an example of the 1st model which the generating part 132 generates is shown. In addition, the 1st model which the production | generation part 132 produces | generates is not restricted to the following, When it is a model produced | generated based on the preset creditworthiness of the 2nd provider, and the information regarding a 1st user is input In addition, any model may be used as long as it outputs a score indicating the creditworthiness of the first operator.

  For example, the generation unit 132 sets the second operator's credit rating as an objective variable in machine learning. And the production | generation part 132 makes each information regarding the 2nd user acquired by the acquisition part 131 an explanatory variable (feature). And the production | generation part 132 produces | generates the 1st model for determining a provider's reliability using an objective variable and an explanatory variable. That is, when the information of the first user is input to the first model, a score indicating the credit rating of the first operator who uses the first user as a customer is output. For example, the generation unit 132 can use any information stored in the feature information storage unit 122 as an explanatory variable.

  First, the production | generation part 132 produces | generates the formula which shows the relationship between the reliability of a 2nd provider, and the information regarding the 2nd user who is the customer of the said 2nd provider. Such formulas are generated as many as the number of second operators serving as learning data. Furthermore, the generation unit 132 learns and determines what weight each piece of information about the second user has with respect to the event of the creditworthiness of the second operator. Thereby, the production | generation part 132 can obtain the information of how much each feature contributes in determination of the creditworthiness of a 2nd provider. For example, the generation unit 132 creates the following formula (1).

y = ω ₁ · x ₁ + ω ₂ · x ₂ + ω ₃ · x ₃ ... + ω _N · x _N (1) (N is an arbitrary number)

  The above formula (1) is created for each second operator, for example. In the above formula (1), “y” indicates “predetermined reliability of the second operator”. In the learning, if the operator A02 shown in FIG. 4 is correct data, “y” takes a value of “85”. Also, if the operator A03 shown in FIG. 4 is correct data, “y” takes a value of “45”.

In the above formula (1), “x” corresponds to each explanatory variable that is information (feature) about the second user. In the above formula (1), “ω” is a coefficient of “x” and indicates a predetermined weight value. Specifically, “ω ₁ ” is a weight value of “x ₁ ”, “ω ₂ ” is a weight value of “x ₂ ”, and “ω ₃ ” is a weight value of “x ₃ ”. It is. As described above, the above equation (1) is a variable including the explanatory variable “x” corresponding to the feature information extracted from the information related to the second user and the predetermined weight value “ω” (for example, “ω _1. x ₁ ").

For example, in the above formula (1), suppose that “x ₁ ” is “the number of settlements of the second user”. Further, “x ₂ ” is assumed to be “the number of unpaid times for the second user”. Further, “x ₃ ” is assumed to be “the number of times the second user has browsed the economic category news content”. In this case, the above formula (1) corresponding to the operator A01 can be expressed as the following formula (2).

y _{(operator A02)} (= 1) = ω ₁ · (number of settlements of second user) + ω ₂ · (number of unpaid times of second user) + ω ₃ · (second user browses news content in economic category) Number of times) ... + ω _N · x _N (2) (N is an arbitrary number)

The production | generation part 132 produces | generates the said Formula (2) for every 2nd provider, and for every 2nd user of a 2nd provider. Then, the generation unit 132 sets the generated expression as a machine learning sample. Then, the generation unit 132 derives a value corresponding to the predetermined weight value “ω” by performing calculation processing (regression learning) of a sample expression. That is, the generation unit 132 determines a predetermined weight value “ω” that satisfies the above equation (2). In other words, the generation unit 132 can determine the weight value “ω” indicating the influence of the predetermined explanatory variable on the objective variable “y”. For example, if “second user's unpaid number” greatly contributes to “second operator's creditworthiness” compared to other variables, it corresponds to “second user's unpaid number”. The value of the weight value “ω ₂ ” is calculated to be larger than that of other variables.

  Further, for example, if “the number of times the second user browses news content in the economic category” contributes to the numerical value of the correct answer data as compared with other variables, for example, “news content in the economic category” There is a possibility that a business operator who has many users who have a large number of “viewing times” has a relatively high credit rating.

  In the above example, the number of payments, the number of unpaid payments, and the number of times of browsing news content in the economic category are shown as features. However, these numerical values may be converted to be used as explanatory variables by a predetermined process. . For example, the generation unit 132 may normalize and use these numerical values by a known method. Alternatively, for example, the generation unit 132 divides the number of payments of the second user into three stages, “high”, “medium”, and “low”, compared with the average value of all users who use the commercial transaction site, and each of them is an explanatory variable. It may be used as In this case, for example, the generation unit 132 can determine the weight value given to the objective variable (= credibility) by the explanatory variable that the unpaid number is “high”. As a result, the generation unit 132 can generate a first model that outputs a result such as “a higher business operator having a second user who has a larger number of unpaid times is judged to have a lower reliability”.

  As described above, the generation unit 132 obtains the relevance between the tendency of the score indicating “the creditworthiness of the second operator” and the information (feature) regarding the second user who is the customer of the second operator. The first model is generated.

  In the above formula (2), the left side is not set as a numerical value of creditworthiness (such as “85”), but an operator whose creditworthiness numerical value exceeds a predetermined number (for example, “70”) is expressed as “trusted (= 1) ) ”, And a regression equation in which a business with a reliability value of a predetermined number or less is“ no trust (= 0) ”may be used. In addition, the generation unit 132 assumes a predetermined error in learning, and uses a method such as a least square method that approximates a value obtained by squaring the difference from the error to a minimum value, and optimizes “ω”. A solution may be calculated.

  In addition, although description is abbreviate | omitted in said example, the production | generation part 132 may produce | generate a 1st model using all the characteristics regarding a 2nd user. Moreover, the production | generation part 132 may add the information regarding a 2nd provider as a feature. For example, the generation unit 132 may use financial information in a store operated by the second business operator or sales information of a product handled by the store as a feature. Further, the generation unit 132 may use the legal character of the second business operator as a feature. Through this learning, for example, the generation unit 132 determines that the legality such as “individual business owner” or “limited company” is determined to have a low credit rating, and “corporation” is determined to have a relatively high credit rating. One model may be generated. In this case, “attributes” such as a stock company and an individual business owner have a weight with respect to an event of creditworthiness of the business operator. Depending on the learning result, there may be a result that a certain feature has little influence on the event of creditworthiness. In this case, it is assumed that the weight (ω) corresponding to the feature in the equation (2) approaches “0”.

  The illustrated first model is an example, and the first model generated by the generation unit 132 is not limited to the above example. That is, the generation unit 132 is a model that is generated based on the preset creditworthiness of the second operator, and when information related to the first user is input, a score indicating the creditworthiness of the first operator is generated. If it is a model to output, the 1st model different from the said example may be produced | generated suitably by combining a known method. For example, in the above example, the first model is generated by regression analysis. However, the generation unit 132 may generate the first model by other statistical processing or use a neural network. The first model may be generated. In the learning process (machine learning), various methods such as linear regression, logistic regression, support vector machine, decision tree, and hazard analysis may be used.

  As described above, the generation unit 132 determines the creditworthiness of the business operator based on the correlation between the creditworthiness of the second business operator and information on the second user who is a user who has a transaction with the second business operator. A first model for generating is generated. In other words, the generation unit 132 includes information related to the customer (second user) of the second operator whose credit rating has already been set, and information related to the customer (first user) of the first operator whose credit rating has not been determined. From the correlation, a first model for determining the creditworthiness of the first operator is generated. In other words, the generation unit 132 can generate a model for determining the creditworthiness of a business operator from the viewpoint of information about the customer held by the business operator, unlike the viewpoints of financial status, sales, and the like. The generation unit 132 stores the generated model in the model storage unit 127.

(About the reception unit 133)
The accepting unit 133 accepts a request regarding determination of the creditworthiness of the first operator. Specifically, the accepting unit 133 according to the embodiment accepts a request for determining the credit rating of the first operator from, for example, a financial institution or the like. The accepting unit 133 may identify the first business operator to be processed in response to accepting the request, and cause the obtaining unit 131 to obtain information on the first user corresponding to the first business operator.

(About determination unit 134)
Of the information about the user acquired by the acquisition unit 131, the determination unit 134 is based on the behavior history on the network of the first user who is a user having a transaction history with the first business operator who performs a commercial transaction on the commercial transaction site. Judge the creditworthiness of one operator.

  For example, the determination unit 134 determines the credit rating of the first operator based on the tendency of the browsing history of the first user. Specifically, the determination unit 134 determines the reliability of the first operator based on the number of times or time when the first user browses the news content in a predetermined period and the tendency of the category of the browsed news content.

  For example, the determination unit 134 has a relatively large number of users and the number and time of browsing the news content categorized as “economy” and “finance” among the categories of news content with respect to all users of the commerce site. Is a user with high creditworthiness. And the determination part 134 has a high ratio with which the user with a high credit rating is a customer (for example, the ratio which the said 1st provider occupies for all the customers exceeds 50%). Judge relatively high.

  Alternatively, the determination unit 134 may determine the creditworthiness of the first operator based on at least one of information on the number of payments at the first user's commercial transaction site, the number of unpaid payments, and the number of use of installments.

  For example, the determination unit 134 estimates that a user having a lower number of unsettled payments with respect to the number of payments at the first user's commercial transaction site is a user with higher reliability. And the determination part 134 determines a credit degree comparatively high about the 1st provider with a high ratio that the user with a high credit degree is a customer. In addition, the determination unit 134 estimates that a user having a smaller number of installment payments in the settlement with respect to the number of settlements in the first user's commercial transaction site is a user with higher reliability. And the determination part 134 determines a credit degree comparatively high about the 1st provider with a high ratio that the user with a high credit degree is a customer.

  Moreover, the determination part 134 may determine the reliability of a 1st provider based on the information regarding the rank in a 1st user's commercial transaction site. For example, the determination unit 134 estimates that a user who is a paid member of a commercial transaction site has a higher degree of trustworthiness than other users. Alternatively, the determination unit 134 estimates that a user who is determined to be a good member according to the sales amount of the commercial transaction site or the like has higher reliability than other users. And the determination part 134 determines a credit degree comparatively high about the 1st provider with a high ratio that the user with a high credit degree is a customer.

  Moreover, the determination part 134 may determine the reliability of a 1st provider based on at least any one information of a 1st user's age, sex, an address, occupation, or a language to be used. For example, the determination unit 134 sets a numerical value that serves as a criterion for determining creditworthiness for each piece of user attribute information. Specifically, the determination unit 134 sets the level of trust for each age, gender, address, occupation, or language used, and based on the attribute information of all the first users corresponding to the first operator, Judge the creditworthiness of one operator.

  As described above, the determination unit 134 sets a predetermined criterion regarding the creditworthiness for the user information, statistically compares the information about the first user corresponding to the first operator, and based on the compared information. The credit rating of the first operator may be calculated. Furthermore, the determination unit 134 may perform determination using the model generated by the generation unit 132. Specifically, the determination unit 134 determines the credit rating of the first operator based on the result output from the first model generated by the generation unit 132.

  As described above, the first model generated by the generation unit 132 indicates the correlation between the existing credit level set by the specialized organization and the information related to the second user corresponding to the second operator. For this reason, the determination part 134 can output the reliability corresponding to the said 1st provider by inputting the information regarding the 1st user made into a process target into a 1st model. That is, the determination unit 134 determines the reliability of the first operator based on the output score.

  Note that the determination unit 134 may use the output score itself as the credit rating of the first operator, or perform processing such as normalizing the output score, and calculate the value calculated after the processing by the first operator. It may be creditworthiness.

[4. Processing procedure)
Next, the procedure of the generation process performed by the determination apparatus 100 according to the embodiment will be described with reference to FIG. FIG. 10 is a flowchart illustrating a generation processing procedure performed by the determination apparatus 100 according to the embodiment.

  As illustrated in FIG. 10, the determination apparatus 100 obtains a credit level set in advance for the second operator from the web server 30 or a specialized institution for generating a model (step S101). Subsequently, the determination apparatus 100 acquires information about a user (second user) who has a transaction with the second business operator (step S102).

  And the determination apparatus 100 produces | generates a 1st model based on the reliability regarding a 2nd provider, and the information regarding a 2nd user (step S103). Subsequently, the determination apparatus 100 stores the generated first model in the model storage unit 127 (step S104).

  Next, a determination process procedure performed by the determination apparatus 100 according to the embodiment will be described with reference to FIG. FIG. 11 is a flowchart illustrating a determination processing procedure performed by the determination apparatus 100 according to the embodiment. Note that the determination apparatus 100 starts the determination process according to the embodiment, for example, when a request for determining the credit rating is transmitted from a financial institution or the like.

  First, the determination apparatus 100 determines whether or not information about a user (first user) who has a transaction with a first business operator to be processed has been acquired (step S201). If the information has not been acquired, the determination apparatus 100 stands by until it is acquired (step S201; No).

  On the other hand, when the information regarding the first user is acquired (step S201; Yes), the determination apparatus 100 inputs the acquired information regarding the first user into the first model (step S202).

  And the determination apparatus 100 outputs the score which shows the reliability of a 1st provider using a 1st model. Furthermore, the determination unit 134 determines the credit rating of the first operator based on the output score (step S204).

[5. (Modification)
The determination apparatus 100 according to the above-described embodiment may be implemented in various different forms other than the above-described embodiment. Therefore, in the following, another embodiment of the determination apparatus 100 will be described.

[5-1. (User credit rating model)
In the said embodiment, the determination apparatus 100 produces | generates a 1st model based on the correlation with the creditworthiness of a 2nd provider, and the information regarding a 2nd user, and determines the trustworthiness of a 1st provider using a 1st model. Explained the process. Here, in a case where the “user's credit rating” is acquired in advance, the determination device 100 may determine the credit rating of a business operator who has a transaction with the user based on the user's credit rating. In this case, the determination apparatus 100 is different from the first model based on the correlation between the information related to the second user and the credit set for the second user (hereinafter referred to as “user credit”). A model, which is a model for determining the user's trustworthiness (hereinafter referred to as “second model”), is generated. And the determination apparatus 100 determines the reliability of the 1st provider used as a process target using a 2nd model. This point will be described with reference to FIG.

  FIG. 12 is a diagram illustrating an example of a determination process according to the modification. FIG. 12 shows an example in which the user U01 and the user U02 use the store AAA and the store BBB opened in the shopping mall as in FIG.

Similar to FIG. 1, the user terminal 10 ₁ in accordance with operation by the user U01, accesses the shopping mall web server 30 is provided, utilizing a shopping mall (step S21). For example, the user U01 purchases a product from the store AAA. Similarly, the user terminal 10 ₂ in accordance with operation by the user U02, accesses the shopping mall web server 30 is provided, utilizing a shopping mall (step S22).

  Here, in the example of FIG. 12, it is assumed that a predetermined user credit is set for the user U02 who is the second user. For example, the user credit level set according to the shopping mall transaction or the attribute of the user U02 is set for the user U02. Specifically, the user U02 has a user credit rating (0 to 100) indicated by a numerical value from 0 to 100 in accordance with a rating given to the user from each store with which the user U02 is dealing, a shopping mall usage amount, and the like. In the example of 12, “70”) is set.

  In the example of FIG. 12, the determination apparatus 100 acquires the user trustworthiness of the user U02 together with information such as the action history of the user U02 (step S23). And the determination apparatus 100 produces | generates the model (2nd model) for determining a user's reliability (step S24).

  For example, the determination apparatus 100 places the user reliability set in advance for the user U02 on the right side, and puts the features described in the embodiment (the number of settlements, news content browsing history, user rank, and other attribute information) on the left side. Generate a formula. It is assumed that the determination apparatus 100 has acquired information from a sufficient number of users U02 required for learning. And the determination apparatus 100 produces | generates the 2nd model which shows the correlation with each feature and user credit by learning the expression produced | generated for every user U02 recursively similarly to the production | generation process of a 1st model. .

  Thereafter, the determination apparatus 100 acquires information regarding the user U01 (first user) who is a user who has made a transaction with the business operator A01, which is the first business operator to be processed (step S25). The determination apparatus 100 inputs the acquired information regarding the first user to the second model, and determines the user credit of each first user (step S26).

  And the determination apparatus 100 determines the creditworthiness of a 1st provider based on the creditworthiness of a 1st user (step S27). For example, the determination apparatus 100 determines the average value of the reliability determined for the user U01 as the reliability of the first business operator who has a transaction with the user U01.

  As described above, the determination apparatus 100 according to the modification includes the second model for determining the user's trustworthiness based on the correlation between the information related to the second user and the user trustworthiness set for the second user. Is generated. And the determination apparatus 100 determines the reliability of a 1st provider based on the output result at the time of inputting the information regarding a 1st user into a 2nd model.

  That is, the determination apparatus 100 according to the modified example determines the creditworthiness of the user itself, and further determines the creditworthiness of a business operator who has a transaction with the user. Thereby, the determination apparatus 100 can determine the creditworthiness of the business operator from various viewpoints such as, for example, a business operator that has many users with high user creditworthiness as customers. Moreover, according to this method, the determination apparatus 100 can determine the creditworthiness of the first operator based on the creditworthiness of the user itself even if the creditworthiness of the second operator is not set in advance.

  Note that the determination apparatus 100 can use the same feature as that of the first model as the feature of the second model. For example, the determination apparatus 100 determines whether the second user's user credit rating is based on the correlation between at least one of the number of payments at the second user's commercial transaction site, the number of unpaid payments or the number of installments, and the user credit of the second user. Two models are generated. Alternatively, the determination apparatus 100 determines the second model based on the correlation between the number of times or time when the second user browses the news content in the predetermined period and the category of the browsed news content and the credit rating of the second user. It may be generated.

[5-2. Information to use)
In the above-described embodiment, an example has been described in which the determination device 100 determines the creditworthiness of the operator using information on a user who has a transaction with the operator. Here, the determination apparatus 100 may calculate the creditworthiness of the operator using a known model in addition to the information related to the user. As an example, the determination apparatus 100 may use any known company evaluation model. For example, the Alt value (Edward Altman) Z value and the like are widely known as a company evaluation model for obtaining a score indicating the probability of bankruptcy of an operator based on financial information. The determination apparatus 100 may employ a score indicating the bankruptcy probability of the business operator as one of the elements for determining the creditworthiness using such an existing model.

  In addition, the determination device 100 may store, for example, the number of accesses (page view) of a store where a business operator opens a store, the number of reviews, the user unit price of a user (for example, the purchase amount per user in a predetermined period), Alternatively, the creditworthiness of the business operator may be determined using information such as the number of times the business operator name or product has been searched and the increase rate of the search frequency.

[5-3. Amount of information)
In the above embodiment, the determination apparatus 100 has shown an example in which a model is generated based on various information that can be acquired on the network. Here, the determination apparatus 100 may perform processing using only information acquired from general users who use various websites exceeding a certain threshold.

  For example, reviews regarding products in a certain store, user evaluations, and the like may show a biased tendency if they are not based on a certain number or more of data. In this case, since the influence of the biased data tends to affect the regression equation, the determination apparatus 100 may generate a model that cannot accurately determine the creditworthiness of the operator. For this reason, the determination apparatus 100 can use only data for which the total number of scores and reviews transmitted from the user exceeds a certain number for a predetermined store as data handled in the model determination process. Thereby, the determination apparatus 100 can generate a model that outputs a highly reliable score.

[5-4. correction〕
The determination apparatus 100 may generate a model that can correct the output score according to the actual economic state. For example, the determination apparatus 100 classifies a business operator whose business condition is improved when the yen is strong, a business operator who is not affected, and a business person whose business condition deteriorates. Then, when generating the model, the determination apparatus 100 adopts the trend of the value of the yen in a predetermined period as a feature, thereby outputting a model that outputs a corrected score in consideration of the trend of the value of the yen. Generate. The determination apparatus 100 reflects such correction information in the model by, for example, acquiring information on the business operator in the long term and accumulating information on the linkage with the trend of the value of the yen.

  The correction as described above may be performed, for example, by limiting the type of business of the second operator input to the model to the import business or the export business.

[6. Others]
In addition, among the processes described in the above embodiment, all or part of the processes described as being automatically performed can be performed manually, or the processes described as being performed manually can be performed. All or a part can be automatically performed by a known method. In addition, the processing procedures, specific names, and information including various data and parameters shown in the document and drawings can be arbitrarily changed unless otherwise specified.

  Further, each component of each illustrated apparatus is functionally conceptual, and does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution / integration of each device is not limited to that shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various loads or usage conditions. Can be integrated and configured.

  For example, the information in the storage unit 120 illustrated in FIG. 3 may be held in an external storage server or the like without being held by the determination apparatus 100. In this case, the determination apparatus 100 acquires various pieces of stored information by accessing the storage server.

  In addition, for example, the determination device 100 described above includes a front-end server side that mainly executes exchanges with external devices such as acquiring various types of information and accepting determination requests for creditworthiness of operators, You may distribute to the back end server side which performs a determination process etc. In this case, for example, the front-end server includes at least the acquisition unit 131. In addition, the back-end server has at least a determination unit 134.

[7. Hardware configuration)
Moreover, the determination apparatus 100, the user terminal 10, the provider terminal 20, and the web server 30 according to the embodiment described above are realized by a computer 1000 configured as shown in FIG. 13, for example. Hereinafter, the determination apparatus 100 will be described as an example. FIG. 13 is a hardware configuration diagram illustrating an example of a computer 1000 that implements the functions of the determination apparatus 100. The computer 1000 includes a CPU 1100, RAM 1200, ROM 1300, HDD 1400, communication interface (I / F) 1500, input / output interface (I / F) 1600, and media interface (I / F) 1700.

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

  The HDD 1400 stores a program executed by the CPU 1100, data used by the program, and the like. The communication interface 1500 receives data from other devices via the communication network 500 (corresponding to the network N shown in FIG. 2) and sends the data to the CPU 1100, and other data generated by the CPU 1100 via the communication network 500 To the device.

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

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

  For example, when the computer 1000 functions as the determination apparatus 100, the CPU 1100 of the computer 1000 implements the function of the control unit 130 by executing a program loaded on the RAM 1200. In addition, each data in the storage unit 120 is stored in the HDD 1400. The CPU 1100 of the computer 1000 reads these programs from the recording medium 1800 and executes them, but as another example, these programs may be acquired from other devices via the communication network 500.

[8. effect〕
As described above, the determination apparatus 100 according to the embodiment includes the acquisition unit 131 and the determination unit 134. The acquisition unit 131 acquires information about a user who uses a predetermined commercial transaction site on the network. Of the information about the user acquired by the acquisition unit 131, the determination unit 134 is based on the behavior history on the network of the first user who is a user having a transaction history with the first business operator who performs a commercial transaction on the commercial transaction site. Judge the creditworthiness of one operator.

  As described above, the determination apparatus 100 according to the embodiment is not limited to financial information or the like that is generally used to measure the creditworthiness of a business operator, but is also used by a user who has a transaction with a store that is opened in a commercial transaction site such as a shopping mall. Use the information to determine the creditworthiness of the operator. Thereby, the determination apparatus 100 can determine the creditworthiness with respect to the operator from various viewpoints regardless of the viewpoint of the financial information, the sales amount, etc. of the operator.

  Moreover, the acquisition part 131 acquires the browsing history in which the said 1st user browsed the content as an action history on the network of a 1st user. The determination unit 134 determines the creditworthiness of the first operator based on the tendency of the first user's browsing history.

  As described above, the determination apparatus 100 according to the embodiment can determine the creditworthiness of the business operator from various viewpoints by using the browsing history of the content that is one of the daily actions of the user.

  In addition, the acquisition unit 131 acquires the number of times or time when the first user browses the news content in the predetermined period and the category of the browsed news content. The determination unit 134 determines the reliability of the first operator based on the number of times or time when the first user browses the news content in the predetermined period and the tendency of the category of the browsed news content.

  As described above, the determination apparatus 100 according to the embodiment uses the browsing history of news content, and for example, a store (operator) that has many users who frequently browse economic news among customers has high creditworthiness. The operator can be evaluated using an index that is difficult to assume as a general operator evaluation index.

  In addition, the acquisition unit 131 acquires at least one of information on the number of payments at the first user's commercial transaction site, the number of unpaid payments, and the number of use of installments. The determination unit 134 determines the creditworthiness of the first operator based on at least one of information on the number of settlements at the first user's commercial transaction site, the number of unpaid payments, and the number of installments.

  As described above, the determination apparatus 100 according to the embodiment uses the behavior related to the user's payment for the processing, and for example, determines a high degree of trustworthiness for a firm with a firm customer who makes a payment steadily. It is possible to evaluate operators from various viewpoints.

  Moreover, the acquisition part 131 acquires the information regarding the rank in a 1st user's commercial transaction site. The determination unit 134 determines the creditworthiness of the first operator based on information related to the rank of the first user's commercial transaction site.

  As described above, the determination apparatus 100 according to the embodiment can evaluate a business operator from various viewpoints, such as determining the creditworthiness of the business operator based on the rank of the customer's commercial transaction site.

  In addition, the acquisition unit 131 acquires at least one of the first user's age, sex, address, occupation, and language used. The determination unit 134 determines the credit rating of the first operator based on at least one of the age, sex, address, occupation, and language used of the first user.

  As described above, the determination apparatus 100 according to the embodiment can perform the company evaluation based on the tendency of being unlikely to appear in the financial information or the like by determining the creditworthiness of the business operator according to the attribute of the user as the customer. .

  In addition, the determination device 100 according to the embodiment is a business operator that is different from the first business operator among business operators who conduct business transactions on the commercial transaction site, and is a business operator whose credit rating is determined in advance. The apparatus further includes a generation unit 132 that generates a first model for determining the creditworthiness of the business operator based on the correlation between the creditworthiness and the information related to the second user who is a user who has a transaction with the second business operator. The determination unit 134 determines the reliability of the first operator based on the result output from the first model generated by the generation unit 132.

  Thus, the determination apparatus 100 according to the embodiment can perform a determination process with higher accuracy by generating a model based on the correlation between the creditworthiness of the business operator and the information on the user and performing the determination process. it can.

  In addition, the acquisition unit 131 acquires at least one of information on the number of payments at the second user's commercial transaction site, the number of unpaid payments, and the number of use of installments. The generating unit 132 generates a first model based on a correlation between at least one of the number of settlements at the second user's commercial transaction site, the number of unpaid payments or the number of installments, and the creditworthiness of the second operator. Is generated.

  As described above, the determination apparatus 100 according to the embodiment uses the payment information for the generation process, and adjusts so that creditworthiness is determined to be high for a firm customer and a business that makes a transaction steadily. The accuracy of the reliability determination can be further improved, for example, the generated model can be generated.

  In addition, the acquisition unit 131 acquires the number of times or time when the second user browses the news content in the predetermined period and the category of the browsed news content. The generation unit 132 generates the first model based on the number of times or time when the second user browses the news content in a predetermined period, and the correlation between the browsed news content category and the creditworthiness of the second operator. To do.

  As described above, the determination apparatus 100 according to the embodiment uses the browsing history for the generation process, and thus the model adjusted so that the trustworthiness is determined to be high for the business with the customer who often browses the economic news. The accuracy of the reliability determination can be further improved.

  Moreover, the production | generation part 132 produces | generates the 2nd model for determining a user creditworthiness based on the correlation with the information regarding a 2nd user, and the user creditworthiness set with respect to the said 2nd user. The determination unit 134 determines the creditworthiness of the first operator based on the output result when information related to the first user is input to the second model.

  Thus, the determination apparatus 100 according to the embodiment may determine the creditworthiness of the user itself, and may further determine the creditworthiness of a business operator who has a transaction with the user. Thereby, the determination apparatus 100 can determine the creditworthiness of the business operator from various viewpoints such as, for example, a business operator that has many users with high user creditworthiness as customers. Moreover, according to this method, the determination apparatus 100 can determine the creditworthiness of the first operator based on the creditworthiness of the user itself even if the creditworthiness of the second operator is not set in advance.

  In addition, the acquisition unit 131 acquires at least one of information on the number of payments at the second user's commercial transaction site, the number of unpaid payments, and the number of use of installments. The generating unit 132 generates a second model based on a correlation between at least one of the number of payments at the second user's commercial transaction site, the number of unpaid payments or the number of installments, and the user credit of the second user. Is generated.

  As described above, the determination apparatus 100 according to the embodiment uses the payment information for the generation process, so that a solid user who steadily makes a payment is determined to have a high creditworthiness, and is a model based on the actual situation. Can be generated.

  In addition, the acquisition unit 131 acquires the number of times or time when the second user browses the news content in the predetermined period and the category of the browsed news content. The generation unit 132 generates the second model based on the number of times or time when the second user browses the news content in a predetermined period, and the correlation between the browsed news content category and the user credit of the second user. To do.

  As described above, the determination apparatus 100 according to the embodiment evaluates the user from various viewpoints such that the user who often browses the economic news is determined to have high reliability by using the browsing history for the generation process. Model can be generated.

  The embodiments of the present application and some of the modifications of the embodiments have been described in detail with reference to the drawings. However, these are examples, and the knowledge of those skilled in the art including the aspects described in the disclosure section of the invention is included. The present invention can be implemented in other forms based on various modifications and improvements.

  Further, the determination apparatus 100 described above may be realized by a plurality of server computers, and depending on functions, an external platform or the like may be realized by calling an API (Application Programming Interface) or network computing. Can be changed flexibly.

  Further, “section (module, unit)” described in the claims can be read as “means”, “circuit”, and the like. For example, the generation unit can be read as generation means or a generation circuit.

DESCRIPTION OF SYMBOLS 1 Determination system 10 User terminal 20 Provider terminal 30 Web server 100 Determination apparatus 110 Communication part 120 Storage part 121 Operator information storage part 122 Feature information storage part 123 Usage history table 124 Browsing history table 125 Settlement information table 126 Attribute information table 127 Model storage unit 130 Control unit 131 Acquisition unit 132 Generation unit 133 Reception unit 134 Determination unit

Claims (14)

An acquisition unit for acquiring information about a user who uses a predetermined commercial transaction site on the network;
Of the information about the user acquired by the acquisition unit, the first operator based on the first user who conducts business transactions on the commercial transaction site and the action history on the network of the first user who is a user with a transaction history. A determination unit for determining the creditworthiness of
A determination apparatus comprising: The acquisition unit
As the action history on the network of the first user, a browsing history that the first user has browsed the content is acquired,
The determination unit
Based on the tendency of the browsing history of the first user, the creditworthiness of the first operator is determined.
The determination apparatus according to claim 1, wherein: The acquisition unit
Acquiring the number or time of browsing the news content by the first user in a predetermined period, and the category of the browsed news content;
The determination unit
Based on the number of times or time when the first user browses the news content in a predetermined period and the tendency of the category of the browsed news content, the credit rating of the first operator is determined.
The determination apparatus according to claim 2, wherein: The acquisition unit
Obtaining at least one of information on the number of settlements in the commercial transaction site of the first user, the number of unpaid payments in the settlement, or the number of use of installments;
The determination unit
Determining the creditworthiness of the first operator based on at least one of information on the number of payments at the commercial transaction site of the first user, the number of unpaid payments, and the number of times of use of installments;
The determination apparatus according to any one of claims 1 to 3, wherein The acquisition unit
Obtaining information about the rank of the first user on the commerce site;
The determination unit
Based on the information about the rank of the first user on the commercial transaction site, the credit rating of the first operator is determined.
The determination apparatus according to claim 1, wherein: The acquisition unit
Obtaining information on at least one of the age, sex, address, occupation or language of the first user;
The determination unit
Determining the creditworthiness of the first operator based on information on at least one of the age, sex, address, occupation or language of the first user;
The determination apparatus according to claim 1, wherein: Of the business operators that conduct business transactions on the commercial transaction site, the business operator is different from the first business operator, and the creditworthiness of the second business operator, the business operator whose creditworthiness is determined in advance, and the second business operator A generating unit that generates a first model for determining the creditworthiness of a business operator based on correlation with information on a second user who is a user with a transaction,
Further comprising
The determination unit
Based on the result output from the first model generated by the generating unit, the credit rating of the first operator is determined.
The determination apparatus according to claim 1, wherein The acquisition unit
Obtaining information on at least one of the number of payments at the commerce site of the second user, the number of unpaid payments, and the number of times of installment payments;
The generator is
Based on the correlation between at least one of the number of payments at the commercial transaction site of the second user, the number of unpaid payments or the number of installments, and the credit rating of the second operator, the first model is Generate,
The determination apparatus according to claim 7. The acquisition unit
Acquiring the number or time of browsing the news content by the second user in a predetermined period, and the category of the browsed news content;
The generator is
The first model is generated based on the number of times or time when the second user browses the news content in a predetermined period, and the correlation between the browsed news content category and the creditworthiness of the second operator.
The determination apparatus according to claim 7 or 8, wherein The generator is
Based on the correlation between the information about the second user and the user credit set for the second user, a second model for determining the user credit is generated,
The determination unit
Based on the output result when the information about the first user is input to the second model, the creditworthiness of the first operator is determined.
The determination apparatus according to any one of claims 7 to 9, wherein The acquisition unit
Obtaining information on at least one of the number of payments at the commerce site of the second user, the number of unpaid payments, and the number of times of installment payments;
The generator is
Based on the correlation between at least one of the number of payments at the commercial transaction site of the second user, the number of unpaid payments, or the number of use of installments, and the user credit of the second user, the second model is Generate,
The determination apparatus according to claim 10. The acquisition unit
Acquiring the number or time of browsing the news content by the second user in a predetermined period, and the category of the browsed news content;
The generator is
The second model is generated based on the number of times or time when the second user browses the news content in a predetermined period, and the correlation between the browsed news content category and the user credit of the second user.
The determination apparatus according to claim 10 or 11, wherein A determination method executed by a computer,
An acquisition step of acquiring information about a user who uses a predetermined commercial transaction site on the network;
Of the information on the user acquired in the acquisition step, the first operator based on the first user who conducts business transactions on the commercial transaction site and the action history on the network of the first user who is a user with a transaction history. A determination step of determining the creditworthiness of
The determination method characterized by including. An acquisition procedure for acquiring information about a user who uses a predetermined commercial transaction site on the network;
Of the information about the user acquired by the acquisition procedure, the first operator based on the behavior history on the network of the first user who is a user who has a transaction history with the first operator who conducts a commercial transaction at the commercial transaction site. A judgment procedure for judging the creditworthiness of
The determination program characterized by causing a computer to execute.

Images