JP7337298B1 - Information processing method, information processing program, and information processing apparatus - Google Patents

Abstract

An object of the present invention is to provide an information processing device and the like that can generate lending information using information other than financial indicators. [Solution] The information processing method is based on the consent for the lending due to the aforementioned operator operating a link for applying for lending set on a site where the platformer provides services to the operator. Send the screen to a business terminal, receive consent information of the business operator from the business terminal, and receive registration information of the business operator regarding the lending and information regarding the business operator's use of the platform service. A computer performs a process of acquiring service data and transmitting information regarding lending for the aforementioned merchant to the aforementioned merchant terminal based on the acquired registration information and the service data. [Selection diagram] Figure 12

Description

TECHNICAL FIELD The present invention relates to an information processing method and the like for transmitting information on lending for businesses.

In a company, when cash flow becomes severe, it is practiced to receive a loan from a financial institution. In preparation for financing, it is practiced to always keep track of the funds that can be procured by the company, but for many SMEs, it is easy to grasp the funds that can be procured by the company as a management option. isn't it.

In such a situation, by using one or more financial indicators based on the financial data of the target company together with one or more reliability levels related to the financial data, the company can There has been proposed an information processing device capable of determining loan conditions for possible loans (Patent Document 1).

JP 2020-67682 A

However, the information processing apparatus described in Patent Literature 1 determines loan conditions based on financial indices, and merely determines loan conditions from one aspect of corporate activities. The present invention has been made in view of such circumstances. The purpose is to provide an information processing device or the like that can use information other than financial indicators when generating lending information.

An information processing method according to an aspect of the present application is a method in which the lending is not performed due to the business operator operating a link for applying for lending set on a site where a platformer provides a service to the business operator. to the operator terminal, receive the consent information of the operator from the terminal of the operator , acquire the registration information of the operator, the account deposit and withdrawal information, and the usage information of the accounting system , From a plurality of platformers including the platformer, obtain service data on the use of the service of each platformer, and based on the service data of each platformer, derive an evaluation value of business feasibility evaluation from multiple viewpoints, Based on the obtained registration information , account deposit/withdrawal information, usage information, and service data of each platformer , deriving loan conditions for the business operator, information on lending including the loan conditions , multiple viewpoints and a list of the plurality of platformers to the business operator terminal.

According to one aspect of the present application, it is possible to use information other than financial indicators when generating lending information.

FIG. 2 is an explanatory diagram showing a configuration example of an ecosystem; It is a block diagram which shows the hardware configuration example of a server. 3 is a block diagram showing a hardware configuration example of a terminal; FIG. It is explanatory drawing which shows the example of user DB. FIG. 4 is an explanatory diagram showing an example of a linked account DB; It is explanatory drawing which shows the example of deposit/withdrawal DB. FIG. 4 is an explanatory diagram showing an example of a platformer DB; It is explanatory drawing which shows the example of a representative / business owner DB. FIG. 4 is an explanatory diagram showing an example of a corporate DB; FIG. 4 is an explanatory diagram showing a structural example of a credit model; FIG. 11 is an explanatory diagram showing a structural example of a third model; 7 is a flowchart illustrating an example of the procedure of main processing; 6 is a flowchart illustrating an example of a procedure for data linkage processing; 10 is a flow chart showing an example of the procedure of provisional examination processing; FIG. 11 is an explanatory diagram showing an example of a platformer screen; FIG. 4 is an explanatory diagram showing an example of a login screen; FIG. 4 is an explanatory diagram showing an example of a home screen; FIG. 10 is an explanatory diagram showing an example of a service selection screen; FIG. 11 is an explanatory diagram showing an example of a login information input screen; FIG. 11 is an explanatory diagram showing another example of the home screen; FIG. 11 is an explanatory diagram showing another example of the home screen; FIG. 11 is an explanatory diagram showing another structural example of a credit model; It is explanatory drawing which shows the structure example of a business feasibility evaluation model. It is explanatory drawing which shows the example of evaluation result DB. 6 is a flowchart showing an example of a procedure for business feasibility evaluation processing; FIG. 11 is an explanatory diagram showing another example of the home screen; FIG. 11 is an explanatory diagram showing another example of the home screen; FIG. 9 is an explanatory diagram showing an example of a list screen; FIG. 10 is a flowchart showing an example of a procedure of list screen creation processing; FIG. It is explanatory drawing which shows the example of provision condition DB. FIG. 4 is an explanatory diagram showing an example of a privilege DB; It is explanatory drawing which shows the example of provision log|history DB. 7 is a flow chart showing an example of the procedure of point awarding processing; It is explanatory drawing which shows the example of a provision notification screen. 9 is a flow chart showing an example of a consistency confirmation process procedure; 4 is a table showing an example of consistent related data; FIG. 4 is a graph showing an example of consistent related data; FIG. FIG. 11 is a table showing an example of related data whose consistency is suspected; FIG. FIG. 11 is a graph showing an example of related data whose consistency is suspected; FIG. FIG. 11 is a flow chart showing an example of the procedure of related data consistency confirmation processing; FIG. 10 is a flow chart showing an example procedure of examination processing.

Embodiments will be described below with reference to the drawings. FIG. 1 is an explanatory diagram showing a configuration example of an ecosystem. The ecosystem 100 is operated by a financial institution that lends to businesses. Ecosystem 100 includes server 1 and terminal 2 . Also, the platformer 3 indicates a system capable of data cooperation with the ecosystem 100 . Platformers are businesses that provide and operate services and systems that serve as platforms that connect users and service providers on the Internet. and write down. The accounting system 4 is a system that mainly provides accounting services through cloud services. The financial institution server 5 is a so-called accounting server that manages deposit/withdrawal information of accounts in financial institutions. The financial institution server 5 includes an accounting system server that manages deposit/withdrawal information of accounts in the financial institution that operates the ecosystem 100 . The model vendor 6 is an information provider or a credit information agency that provides services using a scoring model or the like for determining credit risk. The other vendor 7 is a company or the like that provides services used by financial institutions in conducting their business. AML (Anti-Money Laundering) services, eKYC, personal credit information provision services, etc. are assumed as services that financial institutions will use. AML services are services for preventing money laundering. eKYC (electronic Know Your Customer) is a service that provides identity verification online. The personal credit information provision service is a personal credit information provision service by a credit information agency. Personal credit information includes personal annual income, housing information, attribute information such as place of employment, and payment information such as loans and utility bills. A server 1, a terminal 2, a platformer 3, an accounting system 4, a financial institution server 5, a model vendor 6 and other vendors 7 are connected by a network N so as to be able to communicate with each other. Network N is the Internet, a general public network, a mobile phone network, or a leased line network. The network N may be a combination of these multiple types of networks, or a virtual network such as a VPN (Virtual Private Network).

FIG. 2 is a block diagram showing a hardware configuration example of a server. The server 1 includes a control section 11 , a main storage section 12 , an auxiliary storage section 13 , a communication section 15 and a reading section 16 . The control unit 11, the main storage unit 12, the auxiliary storage unit 13, the communication unit 15 and the reading unit 16 are connected by a bus B. The server 1 is composed of a server computer, a workstation, a PC (Personal Computer), and the like. Further, the server 1 may be configured by a multicomputer composed of a plurality of computers, a virtual machine or a quantum computer virtually constructed by software. Furthermore, the functions of the server 1 may be realized by a cloud service.

The control unit 11 has one or more arithmetic processing units such as a CPU (Central Processing Unit), an MPU (Micro-Processing Unit), and a GPU (Graphics Processing Unit). The control unit 11 reads out and executes a control program 1P (program, program product) stored in the auxiliary storage unit 13 to perform various information processing, control processing, etc. related to the server 1, and the first transmission unit, It implements functional units such as a receiving unit, an acquiring unit, and a second transmitting unit.

The main storage unit 12 is SRAM (Static Random Access Memory), DRAM (Dynamic Random Access Memory), flash memory, or the like. The main storage unit 12 mainly temporarily stores data necessary for the control unit 11 to execute arithmetic processing.

The auxiliary storage unit 13 is a hard disk, SSD (Solid State Drive), or the like, and stores the control program 1P and various DBs (Databases) necessary for the control unit 11 to execute processing. The auxiliary storage unit 13 stores a user DB 131, a linked account DB 132, a deposit/withdrawal DB 133, a platformer DB 134, a representative/business owner DB 135, a corporation DB 136, and the like. The auxiliary storage unit 13 also stores a credit model 141 and a business feasibility evaluation model 142 . Further, the auxiliary storage unit 13 may store an evaluation result DB 137, a grant condition DB 138, a privilege DB 139, and a grant history DB 13A. The auxiliary storage unit 13 may be an external storage device that is separate from the server 1 and externally connected. Various DBs and the like stored in the auxiliary storage unit 13 may be stored in a database server or cloud storage different from the server 1 . The database stored in the auxiliary storage unit 13 is not limited to that shown in FIG. 2, and the auxiliary storage unit 13 may store other databases as necessary.

The communication unit 15 communicates with the terminal 2 , the platformer 3 , the accounting system 4 , the financial institution server 5 , the model vendor 6 and other vendors 7 via the network N. Alternatively, the control unit 11 may use the communication unit 15 to download the control program 1P from another computer via the network N or the like and store it in the auxiliary storage unit 13 .

The reader 16 reads portable storage media 1a including CD (Compact Disc)-ROM and DVD (Digital Versatile Disc)-ROM. The control unit 11 may read the control program 1P from the portable storage medium 1a via the reading unit 16 and store it in the auxiliary storage unit 13 . Alternatively, the control unit 11 may read the control program 1P from the semiconductor memory 1b.

FIG. 3 is a block diagram showing a hardware configuration example of a terminal. A terminal 2 is a terminal used by an end user. The terminal 2 is composed of a notebook computer, a panel computer, a tablet computer, a smart phone, or the like. The terminal 2 includes a control section 21 , a main memory section 22 , an auxiliary memory section 23 , a communication section 24 , an input section 25 and a display section 26 . Each configuration is connected by a bus B.

The control unit 21 has one or more arithmetic processing units such as CPU, MPU, and GPU. The control unit 21 provides various functions by reading and executing the control program 2P (program, program product) stored in the auxiliary storage unit 23 .

The main storage unit 22 is SRAM, DRAM, flash memory, or the like. The main storage unit 22 mainly temporarily stores data necessary for the control unit 21 to execute arithmetic processing.

The auxiliary storage unit 23 is a hard disk, SSD, or the like, and stores various data necessary for the control unit 21 to execute processing. The auxiliary storage unit 23 may be an external storage device that is separate from the terminal 2 and externally connected. Various DBs and the like stored in the auxiliary storage unit 23 may be stored in a database server or cloud storage.

The communication unit 24 communicates with the server 1 and the platformer 3 via the network N. FIG. Alternatively, the control unit 21 may use the communication unit 24 to download the control program 2P from another computer via the network N or the like and store it in the auxiliary storage unit 23 .

The input unit 25 is a keyboard or mouse. The display unit 26 includes a liquid crystal display panel and the like. The display unit 26 displays the lending information output by the server 1 and the like. Also, the input unit 25 and the display unit 26 may be integrated to form a touch panel display. Note that the terminal 2 may perform display on an external display device.

FIG. 4 is an explanatory diagram showing an example of a user DB. The user DB 131 stores end user information. End users are corporations and sole proprietors, and use the ecosystem 100 to raise funds from financial institutions and the like. The user DB 131 includes a user ID column, a type column, a name/name column, and an e-mail address column. The user ID column stores user IDs that can uniquely identify end users. The type column stores the type by the end user. The type is, for example, a corporation or sole proprietorship. The name/name column stores the name of a corporation or the name of a sole proprietor. The email address column stores the user's contact email address. An e-mail address may be used as the user ID. In this case, the values in the user ID column and the values in the email address column may match.

FIG. 5 is an explanatory diagram showing an example of a linked account DB. The cooperation account DB 132 stores end user information necessary for data cooperation with the platformer 3 . The linked account DB 132 includes a user ID column, a Pfer name column, and a token information column. The User ID column stores the end user's user ID. The Pfer name column stores the name of the platformer 3 with which data is linked. The token information column stores access tokens, refresh tokens, etc. given by the platformer 3 .

FIG. 6 is an explanatory diagram showing an example of a deposit/withdrawal DB. The deposit/withdrawal DB 133 stores deposit/withdrawal information of bank accounts used by end users. The deposit/withdrawal DB 133 includes a user ID column, a bank code column, a branch number column, an account number column, an account type column, a type column, a date and time column, an amount column, a transfer destination column, and a balance column. The user ID column stores user IDs. The bank code column stores bank codes that identify banks. The branch number column stores bank branch numbers. The account number column stores account numbers. The account type column stores the type of account. Account types are, for example, regular, current, term, and savings. The type column stores the type of transaction. Transaction types are, for example, deposit, withdrawal, and transfer. The date and time column stores the date and time when the transaction was made. The amount column stores the amount of the transaction. The transfer destination column stores the transfer destination. The balance column stores account balances. The deposit/withdrawal information stored in the deposit/withdrawal DB 133 is obtained from the financial institution server 5 . The deposit/withdrawal DB 133 may be used as a temporary database, and when the deposit/withdrawal information is required, the necessary deposit/withdrawal information may be obtained from the financial institution server 5 to construct the deposit/withdrawal DB 133 . Also, if the user is using an aggregation service and the aggregation service has deposit/withdrawal information of a plurality of financial institutions, the deposit/withdrawal information may be obtained from the aggregation service.

FIG. 7 is an explanatory diagram showing an example of the platformer DB. The platformer DB 134 stores platformer 3 information provided by the platformer 3 . The platformer DB 134 includes an ID column, a name column, a link column and a tag column. The ID column stores an ID that uniquely identifies the platformer 3 . The name column stores the name of the platformer 3 . The link column stores link information for accessing the platformer 3 . Link information is, for example, a URL (Uniform Resource Locator). The tag column stores tags that characterize the platformer 3 . The platformer DB 134 is an example of a platformer storage unit.

FIG. 8 is an explanatory diagram showing an example of the representative/business owner DB. The representative/business owner DB 135 stores information about the representative of the corporation when the end user is a corporation, and information about the sole proprietor when the end user is a sole proprietor. The representative/business owner DB 135 includes a user ID column, a name column, a phone number column, an email column, and a mobile phone column. The user ID column stores user IDs. The name column stores the name of the representative or sole proprietor. The phone number column stores phone numbers. The mail column stores the mail address of the representative or sole proprietor. The mobile phone column stores mobile phone numbers of representatives or sole proprietors. It should be noted that the contents stored in the representative/business owner DB 135 may be stored in the user DB 131 and the representative/business owner DB 135 may not be provided.

FIG. 9 is an explanatory diagram showing an example of a corporate DB. The corporate DB 136 stores corporate information. The corporate DB 136 includes a corporate number column, a user ID column, an address column and a representative telephone column. The corporate number column stores corporate identification numbers. The identification number is, for example, a 13-digit corporate number given by the National Tax Agency. The identification number is not limited to this, and a company identification code or securities code assigned by a credit bureau may be used. Although it is desirable that the identification number is stored, it may not be stored. The user ID column stores user IDs. The location column stores the location of the corporation. The representative telephone string stores the corporation's representative telephone number. As for the telephone number, numbers other than the representative telephone number may be stored. Note that the contents stored in the corporate DB 136 may be stored in the user DB 131 and the corporate DB 136 may not be provided.

FIG. 10 is an explanatory diagram showing a structural example of a credit model. The credit model 141 inputs the end user's deposit/withdrawal data, accounting data, and service data, and outputs determination results. The service data is the usage history and the like obtained from the platformer 3 . The deposit/withdrawal data is acquired from the deposit/withdrawal DB 133 . Accounting data is obtained from the accounting system 4 . The determination result is whether or not the loan is approved, the credit limit, and the like. For example, a loan facility includes a loan amount, repayment period, and interest rate. Credit model 141 includes first model 1411 , second model 1412 , third model 1413 and combiner 1414 . The first model 1411 uses a predetermined determination logic based on the deposit/withdrawal data to determine whether the end user can be financed. The second model 1412 uses a predetermined determination logic based on the accounting data to determine whether the end user can be financed. A third model 1413 determines from the service data whether the end user can be financed. A combiner 1414 combines the judgment result of the first model 1411, the judgment result of the second model 1412, and the judgment result of the third model 1413, and outputs the final judgment result. Also, when the credit model 141 determines that the loan is possible, the credit limit is set. The credit line may be set by the credit model 141, or may be set separately using a predetermined logic. The first model 1411 , the second model 1412 , and the third model 1413 that make up the credit model 141 may all be provided by the model vendor 6 instead of being provided by the server 1 . In this case, the server 1 acquires the determination result of the first model 1411 , the second model 1412 or the third model 1413 from the model vendor 6 . In addition, the credit model 141 is suitable not only when the three data of deposit/withdrawal data, accounting data, and service data are provided, but also when any two data or only one data is input. It is desirable to be able to make judgments. The credit model 141 includes three models, but is not limited thereto, and may be a single model or may include two models.

FIG. 11 is an explanatory diagram showing a structural example of the third model. The third model 1413 is a neural network whose input is the end-user's service data and whose output is the evaluation score of the end-user's business. The service data is the usage history and the like obtained from the platformer 3 . FIG. 11 shows an example of obtaining service data from three platformers 3 and inputting them to the third model 1413 . Platformer A data is service data obtained from Platformer A. Similarly, platform B data and platform C data are service data obtained from platform B and platform C, respectively. The neural network is, for example, a CNN (Convolution Neural Network), and has an input layer that receives input of service data and an output layer that outputs evaluation points.

The input layer has a plurality of neurons that receive service data as input, and passes input item values to the intermediate layer. The hidden layer has a configuration in which a convolution layer that convolves each value input from the input layer and a pooling layer that maps the values convolved in the convolution layer are alternately connected, and compresses the input information. while extracting the evaluation value. The output layer outputs evaluation points based on the evaluation values output from the intermediate layer. The evaluation points take values from 0 to 1, for example. The evaluation points may be discrete values, and may take integer values from 1 to 5, for example.

In the following description, it is assumed that the third model 1413 is CNN, but the third model 1413 is not limited to CNN, and is constructed by other learning algorithms such as neural networks other than CNN, Bayesian networks, and decision trees. It may be a model that has been

The third model 1413 is generated as follows using training data consisting of service data and labels (evaluation points for correct answers). The server 1 inputs service data included in the training data to the input layer, performs arithmetic processing in the intermediate layer, and obtains evaluation points from the output layer. The server 1 compares the evaluation score output from the output layer with the label included in the training data, that is, the correct value, and sets the parameters used in the arithmetic processing in the intermediate layer so that the output value of the output node approaches the correct value. Optimize. The parameters are, for example, weights (coupling coefficients) between neurons, coefficients of activation functions used in each neuron, and the like. Although the parameter optimization method is not particularly limited, for example, the server 1 optimizes various parameters using the error backpropagation method. The server 1 performs the above-described learning process using all the training data, and generates a learned third model 1413 .

Next, information processing performed in the ecosystem 100 will be described. FIG. 12 is a flow chart showing an example of the procedure of main processing. The main process starts when the end user selects a link for starting lending on the screen of the platformer 3 . The control unit 21 of the terminal 2 transmits a start request to the server 1 based on the link information (step S1). The control unit 11 of the server 1 receives the start request (step S2). The control unit 11 transmits a login screen to the terminal 2 (step S3). The control unit 21 of the terminal 2 receives the login screen (step S4). The control unit 21 displays the login screen on the display unit 26 (step S5). If the end user is already registered in the ecosystem 100, he or she enters an ID and password to perform a login operation. If the end user is not registered in the ecosystem 100, he or she performs a registration operation. The control unit 21 determines whether or not the user's operation is registration (step S6). When the control unit 21 determines that the user's operation is registration (YES in step S6), it performs user registration processing (step S7). The user registration process is a process performed in cooperation with the server 1, but since it is a well-known technique, a description thereof will be omitted. When the control unit 21 determines that the user's operation is not registration (NO in step S6), or after the user registration process is completed, the control unit 21 transmits a login request to the server 1 (step S8). The login request includes an ID and password. The control unit 11 of the server 1 performs user authentication using the ID and password included in the login request (step S9). If the end user permits data linkage with the platformer 3, it is desirable to perform authentication using a one-time password at the time of user authentication. The purpose is to further reduce the risk of information leakage due to unauthorized access. The control unit 11 transmits the home screen to the terminal 2 (step S10). The control unit 21 of the terminal 2 receives the home screen (step S11). The control unit 21 displays the home screen on the display unit 26 (step S12). The end user selects the desired processing on the home screen. The processes assumed here are data linkage processing and provisional examination processing. The provisional examination process cannot be executed unless the data linkage process has been executed at least once before. This is because the data necessary for the examination cannot be obtained. The control unit 21 determines whether or not the end user has selected data linkage processing (step S13). When the control unit 21 determines that the end user has selected data linkage processing (YES in step S13), data linkage processing is performed (step S14). After that, the control unit 21 returns the processing to step S13. The details of the data linkage process will be described later. When the control unit 21 determines that the end user has not selected the data linkage process (NO in step S13), it determines whether the end user has selected the provisional examination process (step S15). When the control unit 21 determines that the end user has selected the provisional examination process (YES in step S15), the provisional examination process is performed (step S16). After that, the control unit 21 returns the processing to step S13. When the control unit 21 determines that the end user has not selected the provisional examination process (NO in step S15), the main process ends.

FIG. 13 is a flow chart showing an example of the procedure of data linkage processing. The control unit 21 of the terminal 2 transmits a data linkage processing start request to the server 1 (step S31). Here, it is assumed that there are three types of data linkage, and information on the type selected by the end user is included in the start request. The three types are data linkage of deposit/withdrawal information of financial institutions, data linkage of accounting information, linkage of usage data of other services provided by the platformer 3, and the like. The control unit 11 of the server 1 receives the start request (step S32). The control unit 11 transmits the selection screen to the terminal 2 (step S33). The selection screen is a screen corresponding to the type. In the case of data linkage of deposit/withdrawal information of financial institutions, this screen is for selecting a financial institution. In the case of data linkage of accounting information, this screen is for selecting an accounting service. In the case of linking the usage data of the platformer 3, etc., this is the screen for selecting the platformer 3 that can be linked. The control unit 21 of the terminal 2 receives the selection screen and displays it on the display unit 26 (step S34). The end user selects the data linkage destination on the selection screen. The control unit 21 transmits a data linkage request to the platformer 3 selected by the end user as a data linkage destination (step S35). The platformer 3 receives the request for data linkage (step S36). The platformer 3 transmits the input screen to the terminal 2 (step S37). The input screen is a screen for inputting information to log in to the platformer 3 . In many cases, it is a screen requesting input of a user ID and a password. The control unit 21 of the terminal 2 receives and displays the input screen (step S38). The end user inputs a user ID, password, etc. via the input unit 25 . The control unit 21 transmits login information such as a user ID and a password to the platformer 3 (step S39). The platformer 3 receives the login information (step S40). After user authentication, the platformer 3 issues a token for cooperation and transmits it to the server 1 (step S41). The control unit 11 of the server 1 receives the token and stores it in the cooperation account DB 132 (step S42). The control unit 11 uses the token to transmit the end user's data request to the platformer 3 (step S43). The request contains the end user's user ID. Platformer 3 receives the request (step S44). The platformer 3 extracts end user data based on the user ID (step S45). The platformer 3 transmits the extracted data to the server 1 (step S46). When the type of cooperation is deposit/withdrawal information of a financial institution, the system of the financial institution performs steps S36 and S37, S40 and S41, and S44 to S46. If the type of cooperation is accounting information, the accounting information system performs steps S36 and S37, S40 and S41, and S44 to S46. The control unit 11 of the server 1 receives the data and stores it in the auxiliary storage unit 13 (step S47). The control unit 11 transmits the updated home screen to the terminal 2 (step S48). The control unit 21 of the terminal 2 receives and displays the home screen (step S49). After that, the control unit 21 returns the processing to the caller of the data linkage. Note that the control unit 11 stores the user ID and password that have been input to the input screen in the cooperation account DB 132 .

FIG. 14 is a flow chart showing an example of the procedure of provisional examination processing. The control unit 21 of the terminal 2 transmits a provisional examination request to the server 1 (step S61). The control unit 11 of the server 1 receives the request (step S62). The control unit 11 acquires end user data (step S63). The control unit 11 acquires at least one data from deposit/withdrawal data, accounting data, or service data. The control unit 11 inputs the acquired deposit/withdrawal data, accounting data, and service data to the model (step S64). The control unit 11 inputs deposit/withdrawal data, accounting data, and service data to the credit model 141 . The control unit 11 acquires an evaluation from the credit model 141 (step S65). Based on the evaluation from the credit model 141, the control unit 11 generates financing terms for the end user (step S66). The control unit 11 generates a result screen including loan terms and transmits it to the terminal 2 (step S67). The control unit 21 of the terminal 2 receives and displays the result screen (step S68). The control unit 21 returns the process to the calling source.

Next, screens displayed on the terminal 2 will be described. FIG. 15 is an explanatory diagram showing an example of the platformer screen. The platformer screen d01 is a screen displayed on the terminal 2 when the end user uses the service provided by the platformer 3 . The platformer screen d01 displays a guidance display d011 to the effect that guidance is provided for pre-screening as preparation for applying for a loan to a cooperating financial institution, and a reference button d012 for referring to the guidance. FIG. 15 shows an example of the screen of the platformer 3 that provides crowdfunding, but the platformer 3 that provides services other than crowdfunding similarly displays a guidance display d011 and a reference button d012. When the end user clicks the reference button d012, a guide page for preliminary examination is displayed. It should be noted that if the business operator that operates the platformer 3 has acquired a bank agency business license based on the Banking Act, the reference button d012 may be used as a pre-examination application button. In this case, when the end user clicks the application button, a request to start preliminary examination is transmitted from the terminal 2 to the server 1, and the main processing described above is started. For example, a URL for requesting the server 1 to start the preliminary examination is set as a hyperlink to the application button. If the terminal 2 has already logged into the ecosystem 100, when the end user clicks the application button, a request to start the preliminary examination is made. When the terminal 2 has not logged into the ecosystem 100, a login screen is displayed, and when the login is completed, a request to start the preliminary examination is made.

FIG. 16 is an explanatory diagram showing an example of a login screen. The login screen d02 is a screen for logging into the ecosystem 100. FIG. The login screen d02 includes an ID entry field d021, a password entry field d022, a login button d023, and a registration button d024. The ID input field d021 is a field for inputting a login ID in the ecosystem 100. FIG. As the login ID, an ID given by the ecosystem 100 or an e-mail address used by the end user is used. The password entry field d022 is a field for entering a login password. When the end user selects the login button d023, a login request is sent to the server 1. An end user who has not completed registration with the ecosystem 100 selects the registration button d024 to perform the registration procedure.

FIG. 17 is an explanatory diagram showing an example of the home screen. The home screen d03 is a screen displayed after logging into the ecosystem 100. FIG. The home screen d03 includes a provisional examination button d031, an account linkage display d032, an accounting linkage display d033, a PFer linkage display d034, an account linkage button d035, an accounting linkage button d036, a PFer linkage button d037, an application button d038, a list button d039, and an amount display d03A. including. When the end user selects the provisional examination button d031, the above-described provisional examination process is activated, and a request for provisional examination is transmitted from the terminal 2 to the server 1. FIG. In FIG. 17, since it is not possible to apply for a provisional examination, the characters are displayed in a thin font and the outline of the button is indicated by a dotted line. In this case, even if the provisional examination button d031 is selected, it is not possible to apply for provisional examination. The account linkage display d032 displays the status of linkage with financial institutions. FIG. 17 shows a situation in which cooperation with any financial institution is not performed. The accounting linkage display d033 displays the status of linkage with the accounting service. FIG. 17 shows a situation in which cooperation with any accounting service is not performed. In the present embodiment, provisional examination and borrowing application cannot be performed unless at least data linkage with a financial institution or accounting service is performed, so the provisional examination button d031 does not function as described above. state. The PFer linkage display d034 displays the status of linkage with the platformer 3. FIG. FIG. 17 shows a situation in which no cooperation with any platformer 3 is performed. The account linkage button d035 is a button for setting linkage with a financial institution. The accounting cooperation button d036 is a button for setting cooperation with an accounting service. The PFer cooperation button d037 is a button for setting cooperation with the platformer 3 . When the account linkage button d035, the accounting linkage button d036, or the PFer linkage button d037 is selected, the data linkage processing described above is activated, and a start request is transmitted from the terminal 2 to the server 1. FIG. When the end user selects the application button d038, the terminal 2 sends a borrowing application request to the server 1. FIG. In FIG. 17, since it is not possible to apply for a loan, the characters are displayed in a thin font, and the outline of the button is a dotted line. In this case, even if the application button d038 is selected, the borrowing application cannot be made. When the list button d039 is selected, a list of financial institutions, accounting services, and platformers 3 with which data linkage is possible is displayed. The amount display d03A displays the amount that the end user can borrow. In FIG. 17, the amount is not displayed because the provisional examination is not performed. It should be noted that the provisional examination button d031 may always function regardless of whether or not cooperation is being performed. In this case, when the provisional examination button d031 is selected, if data linkage is required to proceed to the provisional examination, it is desirable to display a message to the effect that data linkage is necessary and prompt the end user to set the linkage. . Also, the provisional examination button d031 may not be included in the components of the screen, and the provisional examination may be executed when data linkage is set or when the end user logs in.

FIG. 18 is an explanatory diagram showing an example of the service selection screen. The service selection screen d04 is a screen for selecting a service for data linkage. The service selection screen d04 includes a selection button d041. Although four selection buttons d041 are displayed in FIG. 18, the number may be three or less, or five or more. Also, if there are many services that can be linked, it will be difficult to select a desired service. Therefore, it may be possible to search for a service name or narrow down by service content. Furthermore, the services displayed on the service selection screen d04 do not include services for which data linkage has already been set by the end user. Services for which data linkage has already been set can be determined by referring to the linkage account DB 132 . The service selection screen is an example of a screen displaying a list of platformers 3 .

FIG. 19 is an explanatory diagram showing an example of the login information input screen. A login information input screen (hereinafter referred to as an "input screen") d05 is a screen for inputting login information for logging into the platformer 3 of the data cooperation destination. The input screen d05 includes an email address input field d051, a password input field d052, an accept button d053, and a cancel button d054. The e-mail address used as the login ID is entered in the e-mail address input field d051. For platformers 3 that cannot use an e-mail address as a login ID, there is a column for entering a login ID instead of an e-mail address. A login password is entered in the password entry field d052. When the end user selects the consent button d053, it is assumed that the end user has consented to collecting data from the platformer 3. Further, login information is transmitted to the platformer 3, and user authentication is performed. When the platformer 3 succeeds in authentication, it issues a token for access and transmits it to the server 1 . The server 1 requests data from the platformer 3 . When the cancel button d054 is selected, data linking is not performed and the screen returns to the home screen. The input screen d05 is an example of a consent screen. Information indicating that the accept button d053 has been selected is transmitted to the server 1. FIG. This information is an example of consent information. A token is issued when the end user selects the consent button d053, and the token is received by the server 1, so the token can also be regarded as consent information.

FIG. 20 is an explanatory diagram showing another example of the home screen. The home screen d06 is a home screen when it is possible to apply for provisional examination. The home screen d06 includes a provisional examination button d061, an account linkage display d062, an accounting linkage display d063, a PFer linkage display d064, an account linkage button d065, an accounting linkage button d066, a PFer linkage button d067, an application button d068, a list button d069, and an amount display d06A. including. Each of the provisional examination button d061 to the amount display d06A is the same as the provisional examination button d031 to the amount display d03A on the home screen d03. In the following description, mainly the configuration different from that of FIG. 17 will be described. In FIG. 20, the provisional examination button d061 is displayed so that it can be selected because it is possible to apply for provisional examination. As described above, the provisional examination button d061 may be displayed so that it can always be selected. The end user can select the provisional examination button d061 to activate the provisional examination process described above. In FIG. 20, the data linkage with the financial institution has been completed. Therefore, the end user's evaluation value, here 70, based on the deposit/withdrawal information in the linked account is displayed in the account linkage display d062. In FIG. 20 as well, no evaluation value is displayed in the accounting linkage display d063 because no linkage with any accounting service is performed. In FIG. 20, since cooperation with at least one platformer 3 has been completed, the end user's evaluation value, here 10, is displayed in the PFer cooperation display d064. In FIG. 20, the application button d068 is displayed so as not to operate, as in FIG. 17, because the provisional examination has not been completed. Also, as in the case of FIG. 17, since the provisional examination has not been carried out, the amount is still not displayed in the amount display d06A.

FIG. 21 is an explanatory diagram showing another example of the home screen. The home screen d07 is a home screen when a borrowing application is possible. The home screen d07 includes an account linkage display d072, an accounting linkage display d073, a PFer linkage display d074, an account linkage button d075, an accounting linkage button d076, a PFer linkage button d077, an application button d078, a list button d079, and an amount display d07A. Each of the account linkage display d072 to the amount display d07A is the same as the account linkage display d062 to the amount display d06A on the home screen d06. In the following description, mainly the configuration different from that of FIG. 20 will be described. In FIG. 21, since the provisional examination has been completed, the provisional examination button d071 is not displayed. In FIG. 21, since the provisional examination has been completed, the application button d078 is displayed so as to be selectable. The end user can apply for borrowing by selecting the application button d078. In addition, the amount display d07A displays the amount determined to be borrowable as a result of the provisional examination. The amount display d07A on the home screen d07 is an example of loan terms and an example of information on lending.

In FIG. 21, when the end user operates the account linkage button d075, the accounting linkage button d076, or the PFer linkage button d077 to add a linkage destination, the loan conditions may change. In this case, even if the end user does not apply for the provisional examination, the home screen d07 reflecting the provisional examination result may be displayed after the provisional examination is executed after the data linkage. Alternatively, a provisional examination button d071 may be displayed on the home screen d07 displayed after data linkage, so that the end user can apply for provisional examination.

Further, when the end user logs out of the ecosystem 100 after undergoing a provisional examination or after applying for a loan, and then logs back in several days later, the following processing is performed. When the home screen is displayed, provisional examination is performed based on the latest deposit/withdrawal information and service information (latest data) from the platformer 3, and the home screen reflecting the results is displayed. Incidentally, every time the end user logs in, provisional examination based on the latest data may be performed without requiring the end user's operation, and the home screen d07 reflecting the result may be displayed. Also, even if the end user does not log in, the server 1 may execute the provisional examination process based on the latest data once a month, for example, by batch processing. The home screen referred to here is assumed to be the home screen d07 shown in FIG.

This embodiment has the following effects. It is possible to generate lending information such as the amount of money that can be borrowed by using not only financial information but also information such as account deposit/withdrawal information of a financial institution, accounting data of an accounting system, usage history of the platformer 3, and the like.

(Modification)
Although an example of the credit model 141 is shown in FIG. 10, a model that integrates the three models that constitute the credit model 141 may be used for provisional examination. FIG. 22 is an explanatory diagram showing another structural example of the credit model. The credit model 141 is a neural network that inputs the end user's deposit/withdrawal data, accounting data, service data, etc., and outputs the end user's evaluation score. The service data is the usage history and the like obtained from the platformer 3 . The neural network is, for example, a RNN (Recurrent Neural Network), a long short-term memory (LSTM: Long Short Term Memory) network, or a Transformer. and an output layer that outputs points. Note that the credit model 141 may be a model constructed by other learning algorithms such as a Bayesian network, a decision tree, etc., other than the neural network.

It should be noted that instead of inputting accounting data, which is usage information of an accounting system, to the credit model 141, cash flow obtained from accounting data may be input. The deposit/withdrawal data may include other information about the account (account), such as the number of years since the account was opened, the average balance for each predetermined period, and the like. Further, the service data to be input to the credit model 141 may be a business feasibility evaluation value obtained from the usage history instead of the usage history of the platformer 3 or the like. For example, the business feasibility is evaluated based on the usage frequency and usage amount obtained from the usage history.

The input layer has a plurality of neurons that accept deposit/withdrawal data, accounting data, and service data as inputs, and passes input item values to the intermediate layer. The hidden layer has a configuration in which a convolution layer that convolves each value input from the input layer and a pooling layer that maps the values convolved in the convolution layer are alternately connected, and compresses the input information. while extracting the evaluation value. The output layer outputs evaluation points based on the evaluation values output from the intermediate layer. The evaluation points are, for example, default probabilities, and take values from 0 to 1. Alternatively, the evaluation points may be values corresponding to ratings.

The credit model 141 is generated as follows using training data consisting of deposit/withdrawal data, accounting data, service data, and labels (evaluation points for correct answers). The server 1 inputs the deposit/withdrawal data, accounting data, and service data included in the training data to the input layer, performs arithmetic processing in the intermediate layer, and acquires evaluation points from the output layer. The server 1 compares the evaluation score output from the output layer with the label included in the training data, that is, the correct value, and sets the parameters used in the arithmetic processing in the intermediate layer so that the output value of the output node approaches the correct value. Optimize. The parameters are, for example, weights (coupling coefficients) between neurons, coefficients of activation functions used in each neuron, and the like. Although the parameter optimization method is not particularly limited, for example, the server 1 optimizes various parameters using the error backpropagation method. The server 1 performs the above-described learning process using all training data to generate a learned credit model 141 .

When the credit model 141 is used in provisional examination, the control unit 11 of the server 1 acquires the evaluation points or rating values output by the credit model 141 . The control unit 11 determines the loan amount or the like based on the evaluation score or rating value, the end user's account balance, changes in the account balance, sales, capital, and the like.

In the credit model 141 shown in FIG. 22, four types of data, namely deposit/withdrawal data, accounting data, platform A data, and platform B data, are used as inputs. It is desirable to learn the credit model 141 as obtained. In addition, with respect to the data of the platformer, not only two types of data, but also one type or three or more types of data are input to the credit model 141, so that appropriate output can be obtained. It is desirable to learn model 141 .

(Embodiment 2)
The present embodiment relates to a form in which the end user's business feasibility is evaluated from multiple viewpoints. FIG. 23 is an explanatory diagram showing a structural example of a business feasibility evaluation model. The business feasibility evaluation model 142 is a neural network that inputs data of the platformer 3 used by the end user, evaluates the business feasibility of the end user from a plurality of viewpoints, and outputs an evaluation value for each viewpoint. The input data is the usage history of the platformer 3 obtained from the platformer 3, and the like. The neural network is, for example, a CNN, and has an input layer that receives data input and an output layer that outputs evaluation values for each viewpoint. The points of view are, for example, management efficiency, trading opportunities, competitive environment, market size, need for financing, and the like. Other sets of viewpoints may include product appeal, productivity, sales force, customer base, organizational management ability, and the like. The input data may include deposit/withdrawal data and accounting data.

The input layer has a plurality of neurons that receive various data as inputs, and transfers the input values to the intermediate layer. The hidden layer has a configuration in which a convolution layer that convolves each value input from the input layer and a pooling layer that maps the values convolved in the convolution layer are alternately connected, and compresses the input information. while extracting the evaluation value. The output layer outputs an evaluation value for each viewpoint based on the evaluation values output from the intermediate layer. The evaluation value may be a discrete value such as 1, 2, 3, 4, or 5, or may be a continuous value from 0 to 1.

In the following description, the business feasibility evaluation model 142 will be described as being a CNN, but the business feasibility evaluation model 142 is not limited to CNN, and other learning algorithms such as neural networks other than CNN, Bayesian networks, decision trees, etc. It may be a model constructed with

The business feasibility evaluation model 142 is generated as follows using training data consisting of input various data and labels (evaluation values of correct answers) for each viewpoint. The server 1 inputs various data included in the training data to the input layer, performs arithmetic processing in the intermediate layer, and acquires evaluation values of each viewpoint from the output layer. The server 1 compares the evaluation value of each viewpoint output from the output layer with the label of each viewpoint included in the training data, that is, the correct value of each viewpoint, and adjusts the intermediate value so that the output value of the output node approaches the correct value. Optimize the parameters used for computations in layers. The parameters are, for example, weights (coupling coefficients) between neurons, coefficients of activation functions used in each neuron, and the like. Although the parameter optimization method is not particularly limited, for example, the server 1 optimizes various parameters using the error backpropagation method. The server 1 performs the above learning process using all the training data, and generates a learned business feasibility evaluation model 142 . Note that the business feasibility evaluation model 142 may be used as the third model 1413 that constitutes the credit model 141 .

FIG. 24 is an explanatory diagram showing an example of the evaluation result DB. The evaluation result DB 137 stores the results of business feasibility evaluation for each end user. The evaluation result DB 137 includes a user ID column, a comprehensive evaluation score column, a management efficiency column, a transaction opportunity column, a competitive environment column, a market size column, and a funding necessity column. The user ID column stores user IDs. The comprehensive evaluation score string stores the comprehensive score of the business feasibility evaluation. A total score is calculated based on the score for each viewpoint. For example, the total score is the average score of the evaluation of each viewpoint. The total score may be the total score of the evaluation for each viewpoint. The management efficiency column stores evaluation points for management efficiency. The Trading Opportunity column stores ratings for trading opportunities. The competitive environment column stores the score for the competitive environment. The market size column stores ratings for market size. The funding need column stores ratings for funding needs.

FIG. 25 is a flow chart showing an example of a procedure for business feasibility evaluation processing. The business feasibility evaluation process is a process of evaluating the end user's business feasibility from the data of the platformer 3 . The control unit 11 of the server 1 acquires data (step S81). The data to be acquired is the data of the linked platformer 3 . The control unit 11 inputs the acquired data to the business feasibility evaluation model 142 (step S82). The control unit 11 acquires an evaluation value for each viewpoint from the business feasibility evaluation model 142 (step S83). The control unit 11 calculates a total evaluation point from the evaluation value for each viewpoint (step S84). The control unit 11 stores the evaluation value for each viewpoint and the total evaluation score in the evaluation result DB 137 (step S85), and ends the process. The business feasibility evaluation process is executed when the end user logs in. In addition, it is periodically executed by batch processing or the like.

Next, screens displayed by the terminal 2 will be described. Most of the screens displayed by the terminal 2 are the same as those in the first embodiment, but the home screen different from that in the first embodiment will be described. FIG. 26 is an explanatory diagram showing another example of the home screen. The home screen d08 shown in FIG. 26 is a screen that is displayed when data linkage and provisional examination are finished. The home screen d08 includes a deposit/withdrawal information display d081, a deposit/withdrawal evaluation value d082, an accounting information display d083, an accounting evaluation value d084, a business feasibility evaluation display d085, a business feasibility evaluation value d086, a bank display d087, an accounting service display d088, and an application button d089. , a list button d08A, and an amount display d08B. A deposit/withdrawal information display d081, an accounting information display d083, and a business evaluation display d085 indicate end-user evaluation items. A deposit/withdrawal evaluation value d082, an accounting evaluation value d084, and a business evaluation value d086 indicate end-user evaluation values based on deposit/withdrawal information, accounting information, and service information obtained from the platformer 3, respectively. The bank display d087 indicates the bank with which the deposit/withdrawal information is linked. Accounting service display d088 indicates an accounting service whose accounting information is linked. When the end user selects the application button d089, the terminal 2 sends a borrowing application request to the server 1. FIG. When the list button d08A is selected, a list of financial institutions, accounting services, and platformers 3 with which data linkage is possible is displayed. The amount display d08B displays the amount that the end user can borrow.

This embodiment has the following effects. By using the business feasibility evaluation model 142, it is possible to evaluate the end user's business feasibility from multiple common viewpoints (common evaluation axis) using service data acquired from multiple platformers 3. Become.

(Utilization of business evaluation)
Functions provided by the ecosystem 100 for end users to utilize the results of the business feasibility evaluation will be described. FIG. 27 is an explanatory diagram showing another example of the home screen. In the configuration of the home screen d09 shown in FIG. 27, the reference numerals and descriptions of the configurations that are the same as those already described are omitted. The home screen d09 shown in FIG. 27 shows the results of the business feasibility evaluation by viewpoint. The home screen d09 includes an evaluation result d091 and a linked service display d092. The evaluation result d091 indicates the business feasibility evaluation for each viewpoint. In the example shown in FIG. 27, there are five viewpoints of business feasibility evaluation: management efficiency, transaction opportunities, competitive environment, market size, and need for financing. These five viewpoints are consistent with the business feasibility evaluation model 142 . The evaluation value of each viewpoint is based on the output from the business feasibility evaluation model 142 . Each viewpoint of business feasibility evaluation is evaluated on a scale of 5, with 1 being the lowest evaluation and 5 being the highest evaluation. A black star indicates an evaluation value. For example, management efficiency indicates two stars and an evaluation value of two. In FIG. 27, the size of the market cannot be evaluated, so in order to show it, the characters are displayed in a thinner font and smaller than other viewpoints. The linked service display d092 indicates a service provided by the platformer 3 that allows data linkage. In the linked service display d092, normal displays indicate services for which data linkage has already been set by the end user. Services displayed in small characters in a thin font indicate services for which data linkage has not been set. The linked service display d092 as a service for which data linkage is not set may select and display a service recommended for data linkage and use to the end user based on the end user's business feasibility evaluation result. Note that the display method of the evaluation value for each viewpoint is not limited to that shown in FIG. For example, it may be indicated by a radar chart. Further, a plurality of evaluation values for each viewpoint may be stored in association with evaluation points, and the transition of the evaluation values may be displayed in a line graph or the like.

FIG. 28 is an explanatory diagram showing an example of the list screen. The list screen d10 displays a list of financial institutions, platforms 3, or services provided by the platform 3 with which data linkage is possible. The list screen d10 includes a cooperation destination name field d101, a type field d102, a legend display d103, a process selection field d104, a radio button d105, and an add button d106. The cooperation destination name column d101 displays the name of a financial institution and the name of a service that can be linked with data. The type column d102 displays the type of information obtained by data linkage or the viewpoint of business feasibility evaluation that can be evaluated. The legend display d103 displays the legend of the type column d102. In the type column d102, the corresponding viewpoint is displayed in reversed black and white. For example, if data is linked with IJ Net Service, it is possible to obtain data for evaluating the business feasibility of the market size and the need for financing. Data for setting the type column d102 is stored in the platformer DB 134 in advance. The process selection field d104 is a field for selecting a process for each data link destination. In FIG. 28, three kinds of processing of new use, cooperation, and cancellation are displayed. New use is selected when you wish to open an account or register as a user. Link is selected when data linkage is desired. If you do not have an account at the selected financial institution or have not completed user registration for the selected service, if you select Link, the login information input screen will transition to a screen for opening an account or registering as a user. The hyperlink will be selected by the end user. Data linkage can be canceled by selecting Cancel. The financial institutions and the services provided that have already performed data linkage are displayed so that new use and linkage cannot be selected, and only cancellation is displayed as selectable. The radio button d105 selects a target financial institution or service when starting data linkage with a plurality of financial institutions or platformers 3 . When the add button d106 is selected, the setting process for performing data linkage with the financial institution or platformer 3 selected with the radio button d105, and the above-described data linkage process are executed. In the example of FIG. 28, the EF net service, IJ net service, KL net service, MN net service, and OP net service are selected by the radio button d105, so when the add button d106 is selected, data about these five services will be displayed. Coordination processing is executed sequentially. The type of information displayed in the type column d102 and the viewpoint of business feasibility evaluation are examples, and other types and other viewpoints can also be adopted.

FIG. 29 is a flow chart showing a procedure example of list screen creation processing. The control unit 11 of the server 1 receives a request for a list of data linkage destinations from the terminal 2 (step S101). The control unit 11 acquires services (linked services) with which the end user has already performed data linkage (step S102). A linked service can be acquired from the linked account DB 132 . The control unit 11 selects a service (candidate service) that is a candidate for data linkage (step S103). The control unit 11 extracts services that can be linked from the platformer DB 134 . Services extracted by the control unit 11 excluding linked services are candidate services. The control unit 11 narrows down. For each candidate service, the viewpoint of business feasibility evaluation that can be evaluated through data linkage is acquired from the platform DB 134 . Also, an evaluation value for each viewpoint of end-user business feasibility evaluation is acquired from the evaluation result DB 137 or the like. The control unit 11 rearranges the evaluation values in ascending order and selects the top two viewpoints. The control unit 11 narrows down the candidate services to services corresponding to the selected point of view (step S104), and creates a list of the narrowed down services (step S105). The control unit 11 transmits the created list to the terminal 2 (step S106), and terminates the process.

In this way, when displaying a list of candidate services for which data linkage is set, by using the end user's business feasibility evaluation results, it is possible to list the services provided by the platformer 3 that contribute to the end user's growth. becomes possible. The list display is an example of recommended information. As a result, it is expected that end users will start using the services of the platformer 3, and the evaluation of the business feasibility will improve. As a result, the ecosystem 100 makes it possible to build a win-win relationship between the end user and the platformer 3 .

(Give incentives)
Incentives may be provided to encourage end users to perform data linkage with the platformer 3 . An example of giving incentives will be described below.

FIG. 30 is an explanatory diagram showing an example of the granting condition DB. The granting condition DB 138 stores incentives, here conditions for granting points to end users. The provision condition DB 138 includes a condition ID column, a business evaluation column, a provision interval column, an expiration date column, and a point number column. The condition ID column stores condition IDs that can uniquely identify the grant conditions. The business feasibility evaluation column stores conditions of business feasibility evaluation for receiving grants. The business feasibility evaluation column further includes a comprehensive evaluation score column, a management efficiency column, a transaction opportunity column, a competitive environment column, a market size column, and a financing need column. The comprehensive evaluation score string stores the comprehensive score of the business feasibility evaluation. The management efficiency column, transaction opportunity column, competitive environment column, market size column, and funding need column are respectively from the perspective of management efficiency, from the perspective of transaction opportunities, from the perspective of competitive environment, and from the market size. , and an evaluation value for the viewpoint of the necessity of raising funds. The award interval column stores intervals for awarding points. The expiration date column stores the expiration date of the points. The point number column stores the number of points to be given.

FIG. 31 is an explanatory diagram showing an example of the privilege DB. The privilege DB 139 stores information on privileges obtained by using points. The privilege DB 139 includes a privilege ID column, a usage target column, a content column, and a point number column. The privilege ID column stores privilege IDs that can uniquely specify privileges. The use object column stores the ID of the financial institution or platformer 3 that can use the privilege. The contents column stores the contents of the privilege. The point number column stores the number of points required to use the privilege.

FIG. 32 is an explanatory diagram showing an example of the grant history DB. The grant history DB 13A stores a history of point grants to end users. The grant history DB 13A includes a user ID column, a condition ID column, a grant date column, a deadline column, a status column, and a point column. The user ID column stores the user ID of the end user who gave the points. The condition ID column stores the condition ID of the granting condition that triggered point granting. The grant date column stores the date when points were granted to the end user. The expiry column stores the expiry date of the points. The state column stores the state of the point. The states are, for example, unused and used. The point column stores the amount of points given to the end user.

FIG. 33 is a flow chart showing a procedure example of point giving processing. The point giving process is executed when the end user logs in, performs a provisional examination, or the like. The control unit 11 of the server 1 acquires the end user's business feasibility evaluation result (step S121). The control unit 11 refers to the granting condition DB 138 and extracts granting conditions that match the result of the business feasibility evaluation (step S122). The control unit 11 selects one of the extracted provision conditions (step S123). The control unit 11 acquires the latest point award date according to the selected award condition (step S124). The control unit 11 determines whether or not to give points (step S125). Based on the acquired grant date and current date, when the grant interval condition included in the grant condition is satisfied, the control unit 11 determines to grant points. If the condition of the award interval is not satisfied, the control unit 11 determines not to award points. When the control unit 11 determines to give points (YES in step S125), the points to be given are stored in the giving history DB 13A (step S126). After step S126, or when the control unit 11 determines not to award points (NO in step S125), it determines whether or not there is an unprocessed awarding condition (step S127). When determining that there is an unprocessed provision condition (YES in step S127), the control unit 11 returns the process to step S123 to process the unprocessed provision condition. When the control unit 11 determines that there is no unprocessed granting condition (NO in step S127), it refers to the privilege DB 139 and extracts privileges for which the granted points can be used (step S128). The control unit 11 transmits to the terminal 2 a grant message indicating that points have been granted together with the information on the available privilege (step S129), and ends the process.

FIG. 34 is an explanatory diagram showing an example of a grant notification screen. The provision notification screen d11 is displayed as a pop-up screen overlaid on the home screen or the like, for example. The grant notification screen d11 includes a grant message d111, an available privilege display d112, and a close button d113. The awarding message d111 is a message notifying the awarding of points, and includes the number of awarded points. The available privilege display d112 displays privileges that can be used with the given points. When the end user selects the close button d113, the control unit 21 of the terminal 2 closes the grant notification screen.

By using the service provided by the platformer 3, the end user can expect an increase in the evaluation of business feasibility and acquisition of points accordingly. As a result, the end users are motivated to increase the number of uses of the platformer 3 and to improve the evaluation.

The result of the point giving process may be notified to the platformer 3 used by the end user. Then, when the end user logs in to the platformer 3, the above-described grant notification screen may be displayed. In addition, it is desirable that the platformer 3 can refer to the grant history DB 13A, and the end user can check the number of held points on the user screen of the platformer 3 as well. It is desirable to display a list of privileges that can be received using points on the user screen of the platformer 3 as well.

The method of giving points is not limited to the one described above. Regardless of the results of the business feasibility evaluation, points may be given periodically, for example, monthly, according to the number of platformers 3 with which data is linked and the period of time. Benefits are not limited to the case where they are made available according to points, but the use of benefits may be permitted to end users who are performing data linkage with the platformer 3 .

(Confirmation of consistency of personal data)
Consistency verification of principal data is the following functions. This function is intended to check the consistency between the user information held by the server 1 and the user information held by the linked platformer 3, thereby increasing the reliability of the user information. Specifically, when the above-described data linking process is executed, the server 1 can acquire user information from the platformer 3, so the process of confirming the consistency of the personal data is executed.

FIG. 35 is a flow chart showing an example of the consistency confirmation process procedure. The control unit 11 of the server 1 acquires the personal information (step S131). The control unit 11 acquires personal information from a platformer for which data linkage processing has been completed in advance, and stores it in the auxiliary storage unit 13 . The control unit 11 confirms the consistency between the end-user's identity information acquired from the platformer and the end-user's information stored in the user DB 131 (step S132). The control unit 11 confirms consistency between the items common to the received personal information and the user information stored in the user DB 131, such as name, email address, location address, and the like. The control unit 11 stores the consistency confirmation result (step S133), and terminates the process. For example, the confirmation result is "consistent" or "not consistent".

The confirmation result may be scored. For end users who have completed data linkage with a plurality of platformers, points are determined according to the number of platformers' identity information that is determined to be consistent with the identity information stored in the user DB 131 . In addition, when obtaining personal information from the platformer 3, information on whether or not confirmation has been completed by eKYC or the like may also be obtained, and if the personal identification at the platformer 3 has not been executed, the score may be reduced.

It is assumed that the consistency confirmation process will be executed in the final examination conducted after the provisional examination. Therefore, the result of consistency confirmation is not notified to the end user in the consistency confirmation process. In addition, if the confirmation result is "no consistency" and the "main examination" is terminated at that point, the end user may be notified to that effect.

If the platformer 3 performs account aggregation, etc., and has deposit/withdrawal data of the financial institution account used by the end user, the platformer 3 sends the personal information linked to the account to the server 1 may be sent. Since financial institutions are particularly cautious in verifying identity, the reliability of information will increase if it is consistent with the data held by financial institutions.

(Consistency check of related data)
Relevant data consistency check is the following function. When the end user has set data linkage with multiple platformers 3, a function to check whether each data is consistent by using the relationship between the data provided by each platformer 3 is. If the data are consistent, it can be said that the owner of the data linked from each platformer 3 is the same person. For example, inventory data from the first platformer 3, ordering data from the second platformer 3, account data from the third platformer 3, and the server 1 can refer to these three types of data. Relevance can be used to check the consistency of data. The following description describes an example of checking the consistency of inventory data, accounting data, and account data.

FIG. 36 is a table showing an example of consistent related data. The rows of "inventory decrease (sales)", "inventory increase (purchase)", and "inventory amount (end of month)" are inventory data. Each row of "sales amount" and "purchase amount" is accounting data. "Payment amount" and "payment amount" are account data. For example, inventory decreased due to sales in January 2022, and the value of the “Inventory Decrease (Sales)” row is 100. Correspondingly, sales of 120 are recorded in the "sales amount" row. A payment 120 corresponding to sales in March 2022 is recorded in the "amount received" row, and it can be confirmed that the inventory data, accounting data, and account data are consistent. In addition, it can be seen from the value in the "inventory decrease (sales)" row that the inventory decreased by 120 due to sales in February 2022. Correspondingly, 120 purchases were made in March 2022, which can be read from the value in the "Inventory Increase (Purchase)" row. Accordingly, 120 is recorded in the "Purchase Amount" row. Then, in May 2022, the payment was made for the purchase, and 120 is recorded in the "Amount Paid" row. Even in such a flow, it can be confirmed that the inventory data, accounting data, and account data are consistent.

FIG. 37 is a graph showing an example of consistent related data. The vertical axis is the amount, and the unit is, for example, one million yen. The horizontal axis is year and month. FIG. 37 shows the temporal transition of the amount of inventory reduction (sales) and the amount of money received. As described above, there is a time lag between the time of sale and the receipt of payment, and the amounts do not match, but it can be visually confirmed from the shape of the graph that both data match.

FIG. 38 is a table showing an example of related data whose consistency is suspected. In FIG. 38 , the relationship between “inventory decrease (sales)” and “sales amount” changes from September 2022 onwards. Since "sales amount" and "deposit amount" are related, it is possible to detect changes in the relationship between "inventory decrease (sales)" and "deposit amount". Data integrity is questionable.

FIG. 39 is a graph showing an example of related data whose consistency is suspected. The axes and indicated values are the same as in FIG. As mentioned above, there is a time lag between "inventory decrease (sales)" and "deposit amount", and although the amounts do not match, the shape of the graph should be similar. , the two graph shapes are clearly different, and it is clear that the data consistency is questionable.

FIG. 40 is a flow chart showing an example of the procedure of related data consistency confirmation processing. The control unit 11 of the server 1 acquires data items (hereinafter referred to as "linkage items") obtained from the platformer 3 through data linkage (step S151). Linkage items are stored in the auxiliary storage unit 13 . The control unit 11 searches for a rule that can be used to confirm consistency using the linked items (for example, data items included in each of inventory data, accounting data, and account data) (step S152). The rules are stored in the auxiliary storage unit 13 in advance. The control unit 11 determines whether or not there is a rule that can be used (step S153). The control unit 11 determines that there is a rule if there is a hit in the search, and that there is no rule if there is no hit in the search. If the control unit 11 determines that there is a rule that can be used (YES in step S153), it acquires data for checking consistency based on the rule (step S154). The control unit 11 confirms the consistency of the acquired data (step S155). For example, "sales/inventory decrease amount" is calculated for each month. Regarding the calculated value, the average of all companies (or the same industry) is 1.2 (standard deviation is 0.1) as a reference value, and there are 3 months in a row that do not fall within 1.0 to 1.4. If so, it is determined that there is an inconsistency. The control unit 11 stores the consistency confirmation result (step S156), and terminates the process. For example, the confirmation result is "consistent" or "not consistent". If the control unit 11 determines that there is no rule that can be used (NO in step S153), it sets "impossible" indicating that confirmation is not possible (step S157), and terminates the process.

It is assumed that the related data consistency confirmation process will be executed in the main examination conducted after the provisional examination. Therefore, the result of consistency confirmation is not notified to the end user in the related data consistency confirmation process. In addition, if the confirmation result is "no consistency" and the "main examination" is terminated at that point, the end user may be notified to that effect. Also, end users may be restricted from using the ecosystem 100 .

(Function to integrate results from multiple models)
As shown in FIG. 10, the credit model 141 may be configured to combine the results of a plurality of models and output examination results such as approval or disapproval of financing. The function that integrates the results of multiple models will be described below. In FIG. 10, reference numeral 1414 denotes a combiner, and it is stated that the combiner 1414 combines the determination results of the first model 1411, the second model 1412, and the third model 1413, and outputs the final determination result. However, the content described below is called "integration" rather than "combination" because the expression "integration" is more appropriate.

In the following description, an example of integrating the results of two models, the A model and the B model, will be shown. The two models may or may not be the accounting model, deposit/withdrawal model, and service model described above. The A model and the B model output determination results upon input of end-user accounting data, deposit/withdrawal information in financial accounts, inventory data, and the like. It is assumed that the determination result includes a PD (Probability of Default) value and a loanable amount. A larger PD value means a higher probability of default, and a smaller PD value means a lower probability of default. Let PD _A be the PD value output by the A model, and PD _B be the PD value output by the B model. At this time, the PD value of the end user is calculated as follows.

PD value = α × PD _A + β × PD _B
0≤α≤1, 0≤β≤1

α and β are coefficients. α and β may vary depending on the type of end user and the value range of PD _A and PD _B.

Next, the method of determining the interest rate will be explained. First, calculate the base interest rate. For example, the base interest rate is the sum of the PD value, the expense ratio, and the profit margin. Expense ratios and profit margins are determined in advance, but it is desirable to review them periodically.

Next, calculate the preferential interest rate. The number of data (number of types of provided data) obtained by the end user by allowing data linkage between other systems and the ecosystem 100, and the number of business feasibility evaluations performed by data linkage (business feasibility evaluation score number) determines the value of the preferential interest rate. For example:

X=r1×(number of provided data types)
Y = r2 x (number of business evaluation scores)
0<r1<1, 0<r2<1
preferential interest rate = min {upper limit, X + Y}

min {value 1, value 2} means that the smaller of value 1 and value 2 is used. The loan interest rate shall be the base interest rate minus the preferential interest rate.

In addition, if the PD value output by the learning model (here, model A and model B) can be evaluated as a value after sufficiently considering the number of types of provided data and the number of business evaluation scores, the preferential interest rate No calculation is required.

Explain how the loanable amount is determined. For example, similarly to the PD value, the loanable amount is obtained by weighting and adding the loanable amounts output by the two learning models.

Loanable amount = γ × (Model A loanable amount) + δ × (Model B loanable amount)
0≦γ≦1, 0≦δ≦1, provided that γ≠0 or δ≠0

γ and δ are coefficients. γ and δ may vary depending on the end user's industry and the value range of PD _A and PD _B. If the learning model does not output the loanable amount, the loanable amount may be calculated based on the data input to the learning model. In addition, if the end user has borrowed money between the reference time when the learning model determined the loanable amount and the processing time, the amount obtained by subtracting the borrowing amount from the loanable amount output by the learning model is , the loanable amount determined by the learning model.

Next, the examination process using the integration function of the results of multiple models will be explained. FIG. 41 is a flow chart showing an example of the examination process procedure. The examination process is a process executed by the server 1, and is executed in response to a request from the terminal 2 or the like. The control unit 11 of the server 1 collects data necessary for examination (step S171). The control unit 11 confirms whether or not the data obtained from the data-linked platform 3 is the latest, and updates the data if it is not the latest. The control unit 11 inputs the input data required by each model to a plurality of learning models, and acquires the PD value output by each learning model (step S172). The control unit 11 determines one PD value (usage default probability) from a plurality of PD values (step S173). The control unit 11 uses the determined PD value to calculate the loan interest rate (step S174). The control unit 11 calculates the loanable amount (step S175). The control unit 11 outputs the result (interest rate, loanable amount) to the terminal 2 that requested the examination (step S176), and ends the process.

The technical features (components) described in each embodiment can be combined with each other, and new technical features can be formed by combining them.
It should be considered that the embodiments disclosed this time are illustrative in all respects and not restrictive. The scope of the present invention is indicated by the scope of the claims rather than the meaning described above, and is intended to include all modifications within the scope and meaning equivalent to the scope of the claims.

100 Ecosystem 1 Server 11 Control Unit 12 Main Storage Unit 13 Auxiliary Storage Unit 131 User DB
132 Linked Account DB
133 Deposit and withdrawal DB
134 Platform DB
135 Representative/Employer DB
136 Corporate database
137 Evaluation result DB
138 grant condition DB
139 Privilege DB
13A grant history DB
141 credit model 142 business evaluation model 15 communication unit 16 reading unit 1P control program 1a portable storage medium 1b semiconductor memory 2 terminal 21 control unit 22 main storage unit 23 auxiliary storage unit 24 communication unit 25 input unit 26 display unit 2P control program 3 Platformer B Bus N Network

Claims (14)

Sending a lending consent screen to the business operator's terminal due to the business operator's operation of a link for applying for lending set on a site where the platform provides services to the business operator,
receiving the consent information of the business operator from the business terminal;
Acquiring registration information, account deposit/withdrawal information, and accounting system usage information of the business operator ,
Acquiring service data related to the use of services of each platformer from a plurality of platformers including the platformer;
Based on the service data of each platformer, deriving an evaluation value of business feasibility evaluation from multiple viewpoints,
Deriving loan conditions for the business operator based on the acquired registration information , account deposit/withdrawal information, usage information and service data of each platform ,
A computer performs a process of transmitting information on lending including the loan conditions , each evaluation value of the business feasibility evaluation from a plurality of viewpoints, and a list of the plurality of platformers to the business operator terminal. Information processing methods. Sending an integrated evaluation value obtained by integrating each evaluation value to the business operator terminal
The information processing method according to claim 1 . 3. The information processing method according to claim 1, wherein said loan conditions are determined based on said business operator's cash flow, account, or said business feasibility evaluation. receive login information for said business;
According to the latest data of the registration information , the account deposit/withdrawal information, the usage information, and the service data, the loan terms are updated and transmitted to the business operator terminal. 2. The information processing method according to 2 . Identifying the platform used by the business operator including the platform,
referring to the platform storage unit in which information of multiple platforms is stored, creating a list of platformers excluding the specified platformer;
3. The information processing method according to claim 1 , further comprising transmitting the created list to the business operator terminal. Select a platformer that contributes to the growth of the business operator based on the business feasibility evaluation,
3. The information processing method according to claim 1 , wherein recommended information of the selected platformer is transmitted to the business operator terminal. In addition to the service data, acquire the service data of the platformer specified by the business operator,
3. The information processing method according to claim 1 or 2, wherein the information on lending is updated based on the acquired plural types of services. 6. The information processing method according to claim 5 , wherein the list includes link information for transitioning to a usage registration screen of the platformer. Based on the business feasibility evaluation, determine the benefits to be received from the bank or the platform,
3. The information processing method according to claim 1 , further comprising: transmitting the determined privilege to the business operator terminal. Sending a lending consent screen to the business operator's terminal due to the business operator's operation of a link for applying for lending set on a site where the platform provides services to the business operator,
receiving the consent information of the business operator from the business terminal;
Acquiring registration information, account deposit/withdrawal information, and accounting system usage information of the business operator ,
Acquiring service data related to the use of services of each platformer from a plurality of platformers including the platformer;
Based on the service data of each platformer, deriving an evaluation value of business feasibility evaluation from multiple viewpoints,
Deriving loan conditions for the business operator based on the acquired registration information , account deposit/withdrawal information, usage information and service data of each platform ,
sending information on lending including the loan conditions , each evaluation value of the business feasibility evaluation from a plurality of viewpoints, and a list of the plurality of platformers to the business operator terminal; information processing program. Sending a consent screen for lending to the business operator's terminal due to the business operator's operation of a link for applying for lending set on a site where the platform provides services to the business operator 1 transmitter;
a receiving unit that receives consent information of the business operator from the business terminal;
a first acquisition unit that acquires the business operator's registration information, account deposit/withdrawal information, and accounting system usage information ;
a second acquisition unit that acquires service data relating to use of services of each platformer from a plurality of platformers including the platformer;
a first derivation unit for deriving an evaluation value of business feasibility evaluation from a plurality of viewpoints based on the service data of each of the platformers;
a second derivation unit that derives loan conditions for the business operator based on the obtained registration information , the account deposit/withdrawal information, the usage information, and the service data of each of the platformers ;
a second transmission unit configured to transmit information on lending including the loan conditions , each evaluation value of the business feasibility evaluation from a plurality of viewpoints, and a list of the plurality of platformers to the business operator terminal; and information processing equipment. Acquiring the personal information of the business operator held by the platformer,
confirming consistency of common data item values between the obtained personal information and pre-stored user information;
3. The information processing method according to claim 1 , further comprising: outputting a confirmation result. Obtaining a plurality of related data item values included in the service data obtained from different platforms;
confirming the consistency of the obtained plurality of data item values based on the relevance;
3. The information processing method according to claim 1 , further comprising: outputting a confirmation result. inputting the obtained registration information or the service data into a plurality of learning models that output a default probability when the registration information or the service data is input;
calculating a used default probability to be used based on the default probability output by each of the plurality of learning models;
3. The information processing method according to claim 1 , wherein a loan interest rate included in said loan conditions is calculated based on said calculated use default probability.

Images