Program, information processing method, and information processing system

Abstract

[Problem] To provide a technology capable of supporting a determination as to whether, in a credit appraisal, a business operator being appraised is a high-risk business operator. [Solution] One aspect of the present invention provides a program that causes an information processing system to execute the following steps: an acquisition step in which balance information relating to a prescribed business operator is acquired, where the balance information is information relating to an account balance; and an output control step in which high-risk information is output on the basis of the balance information and a high-risk business operator learning model, where the high-risk information is information indicating whether the prescribed business operator is a high risk in a credit appraisal, and the high-risk business operator learning model is a learning model obtained by performing learning using balance information and high-risk information as teacher data.

Description

Program, information processing method and information processing system

[0001] The present invention relates to a program, an information processing method, and an information processing system.

[0002] In the prior art, a deferred payment support device that can reduce the risk of bad debts during deferred payment has been disclosed (Patent Document 1).

[0003] Japanese Patent Application Laid-Open No. 2021-002244

[0004] In recent years, there has been a demand for technology that can support the determination of whether a business operator is a high-risk business operator.

[0005] In view of the above circumstances, the present invention provides a technology that can support the determination of whether a business subject to credit screening is a high-risk business.

[0006] According to one aspect of the present invention, there is provided a program that causes an information processing system to execute the following steps: in an acquisition step, balance information of a specified business is acquired, the balance information being information regarding the account balance; in an output control step, high-risk information is output based on the balance information and a high-risk business learning model, the high-risk information being information indicating whether the specified business is a high risk in credit screening; and the high-risk business learning model is a learning model obtained by learning the balance information and the high-risk information as training data.

[0007] According to the present disclosure, it is possible to provide a technology that can support a determination of whether a business subject to a credit check is a high-risk business in a credit check. Thus, the above-described aspect improves the technical field related to credit checks.

[0008] Furthermore, since the above-described aspect does not require the preparation of special equipment to support the credit screening decision, the functionality of the computer can be improved with a simple configuration. That is, the above-described aspect can improve the functionality of the computer so as to achieve at least one of the following (1) to (4): (1) The computer's processing speed can be increased. (2) The computer's power consumption can be reduced. (3) The computer's communication speed can be increased. (4) The saved resources in the computer can be used for other core functions.

[0009] 1 is a diagram showing an example of a system configuration of an information processing system 1 according to embodiment 1. FIG. 2 is a diagram showing an example of a hardware configuration of a server device 2 according to embodiment 1. FIG. 3 is a diagram showing an example of a hardware configuration of a client device 3 according to embodiment 1. FIG. 4 is an example of a block diagram showing functions realized by a processor 21 of the server device 2. FIG. 5 is a diagram showing an example of an activity showing the flow of information processing executed by the information processing system 1 according to embodiment 1. FIG. 6 is a diagram showing an example of teacher data 4. FIG. 7 is a diagram showing an example for explaining information processing for acquiring a high-risk business operator learning model 5. FIG. 8 is a diagram showing an example for explaining information processing for outputting high-risk information from the high-risk business operator learning model 5. FIG. 9 is a diagram showing an example of output data 7 including high-risk probability information 70 and high-risk occurrence information 71. FIG. 10 is a diagram showing an example of an activity showing the flow of information processing executed by the information processing system 1 according to embodiment 2. FIG. 11 is a diagram showing an example of a screen displayed in A27. FIG. 12 is a diagram showing an example of an activity showing the flow of information processing executed by the information processing system 1 according to embodiment 3. FIG. 13 is a diagram showing an example of a screen displayed in A34.

[0010] Hereinafter, each embodiment of the present invention will be described with reference to the drawings. Various features shown in each embodiment can be combined with each other.

[0011] [Embodiment 1] 1. Definition of Terms A program for realizing software appearing in this embodiment may be provided as a non-transitory computer-readable medium, or may be provided so as to be downloadable from an external server, or may be provided so that the program is started on an external computer and its functions are realized on a client device (so-called cloud computing).

[0012] In this embodiment, the term "unit" may include, for example, a combination of hardware resources implemented by a circuit in the broad sense and software information processing that can be specifically realized by these hardware resources. In addition, various types of information are handled in this embodiment, and this information may be represented by, for example, physical values ​​of signal values ​​representing voltages and currents, high and low signal values ​​as a binary bit set consisting of 0 or 1, or quantum superposition (so-called quantum bits), and communication and calculations may be performed on a circuit in the broad sense.

[0013] Furthermore, a circuit in the broad sense is a circuit realized by at least an appropriate combination of a circuit, circuitry, a processor, a memory, etc. That is, it includes an application specific integrated circuit (ASIC), a programmable logic device (e.g., a simple programmable logic device (SPLD), a complex programmable logic device (CPLD), and a field programmable gate array (FPGA)), etc.

[0014] "Credit screening" refers to the examination of a business's financial strength, repayment ability, and other creditworthiness. Credit screening also includes screening for factoring.

[0015] A "seizure business" is a business that is likely to experience seizure or has experienced seizure. A "seizure is likely" means, for example, that the conditions for seizure are met. Examples of conditions for seizure include: information related to seizure has been recorded; a payment grace period has been approved; emergency financing or guarantees have been received to provide relief from deteriorating business; the occurrence of an issue that raises concerns about the continuation of business (for example, seizure, provisional seizure, or other disposition by public authority; a large amount of unpaid debt; an extremely poor sales situation; etc.); or chronic payment delays.

[0016] From one perspective, a seizing business operator is defined as a "problem party whose default risk should be reduced or avoided," and therefore, in this specification, a seizing business operator may be interpreted approximately as a "defaulting business operator."

[0017] A "defaulting business" is a business that is likely to experience defaults or has experienced defaults.

[0018] A "normal business" is a business that has not experienced any seizures or bad debts.

[0019] An "account" is a system for managing the flow of money. Accounts are provided by banks, Japan Post Bank, securities companies, etc., and can include various types of accounts in addition to ordinary accounts and current accounts.

[0020] 2. System Configuration of Information Processing System 1 Next, the system configuration of the information processing system 1 according to the first embodiment will be described with reference to FIG.

[0021] FIG. 1 is a diagram illustrating an example of the system configuration of an information processing system 1 according to a first embodiment. As illustrated in FIG. 1, the information processing system 1 includes a server device 2, a client device 3, and a network N. The server device 2 is configured to be able to communicate with the client device 3 via the network N. This allows the server device 2 and the client device 3 to transmit and receive various information to and from each other. Note that the server device 2 and the client device 3 are examples of information processing devices and are not limited to the present embodiment. The client device 3 may be any of a PC (Personal Computer), a tablet computer, a smartphone, etc. Note that there may be a plurality of server devices 2 and a plurality of client devices 3.

[0022] 3. Hardware Configuration Next, the hardware configuration of the server device 2 according to the first embodiment will be described with reference to FIG. 2, and the hardware configuration of the client device 3 according to the first embodiment will be described with reference to FIG.

[0023] 2 is a diagram illustrating an example of the hardware configuration of the server device 2 according to the first embodiment. As shown in FIG. 2, the server device 2 includes a processor 21, a storage unit 22, and a communication unit 23, and these components are electrically connected via a communication bus inside the server device 2. The server device 2 executes the processing according to the embodiment.

[0024] The processor 21 processes and controls the overall operations related to the server device 2. The processor 21 is, for example, a central processing unit (CPU). The processor 21 reads out a predetermined program stored in the storage unit 22 and executes processing based on the program, thereby realizing various functions related to the server device 2, for example, the processing shown in FIG. 5 described below. Note that the number of processors 21 is not limited to a single processor, and the server device 2 may be implemented with multiple processors 21 for each function. Alternatively, a combination of these may be used.

[0025] The memory unit 22 stores various pieces of information defined above. This may be implemented, for example, as a storage device such as a solid state drive (SSD) that stores various programs related to the server device 2 executed by the processor 21, or as a memory such as a random access memory (RAM) that stores temporarily required information (arguments, arrays, etc.) related to program calculations. The memory unit 22 stores various programs, variables, and data used when the processor 21 executes processes based on the programs related to the server device 2. The memory unit 22 is an example of a storage medium.

[0026] The communication unit 23 is preferably a wired communication means such as USB, IEEE 1394, Thunderbolt (registered trademark), or wired LAN network communication, but may also include wireless LAN network communication, mobile communication such as LTE / 3G / 4G / 5G, or BLUETOOTH (registered trademark) communication as needed. That is, it is more preferable to implement the communication unit 23 as a combination of multiple communication means. That is, the server device 2 may communicate various information from the outside via the communication unit 23.

[0027] 3.2 Hardware Configuration of Client Device 3 Fig. 3 is a diagram illustrating an example of the hardware configuration of the client device 3 according to the first embodiment. As illustrated in Fig. 3, the client device 3 includes a processor 31, a storage unit 32, a communication unit 33, an input unit 34, and an output unit 35, and these components are electrically connected via a communication bus within the client device 3. The client device 3 executes the processing according to the embodiment. For the processor 31, storage unit 32, and communication unit 33 of the client device 3, please refer to the processor 21, storage unit 22, and communication unit 23 of the server device 2.

[0028] The input unit 34 may be included in the housing of the client device 3 or may be externally attached. For example, the input unit 34 may be integrated with the output unit 35 and implemented as a touch panel. A touch panel allows a user to input tapping, swiping, and the like. Of course, a switch button, a mouse, a QWERTY keyboard, or the like may be used instead of a touch panel. In other words, the input unit 34 accepts an input based on an operation performed by the user. The input is transferred as a command signal to the processor 31 via a communication bus, and the processor 31 can execute predetermined control or calculation as necessary.

[0029] The output unit 35 can function as a display unit of the client device 3. The output unit 35 may be included in the housing of the client device 3 or may be externally attached. The output unit 35 displays a graphical user interface (GUI) screen that can be operated by a user. This is preferably implemented by selectively using display devices such as a CRT display, a liquid crystal display, an organic EL display, and a plasma display depending on the type of client device 3.

[0030] 4. System Configuration of Information Processing System 1 In this section, the functional configuration of the processor 21 of this embodiment will be described with reference to FIG. 4. FIG. 4 is an example of a block diagram showing functions realized by the processor 21 of the server device 2. The processor 21 of the server device 2, which is an example of the information processing system 1, includes an information transmitting / receiving unit 210, a learning unit 211, an output control unit 212, a determination unit 213, and a display control unit 214. As described above, information processing by software stored in the storage unit 22 is specifically realized by the processor 21, which is an example of hardware, and can be executed as each functional unit included in the processor 21.

[0031] The information transmitting / receiving unit 210 receives various pieces of information from the client device 3 via the network N and the communication unit 23. The information transmitting / receiving unit 210 transmits various pieces of information to the client device 3 via the communication unit 23 and the network N. The information transmitting / receiving unit 210 executes an acquisition step.

[0032] The learning unit 211 constructs a learning model based on the training data. The learning unit 211 executes a learning step.

[0033] The output control unit 212 controls the output of data using the high-risk business operator learning model. The output control unit 212 executes an output control step.

[0034] The determination unit 213 determines whether or not a business entity is likely to be a high-risk business entity in a credit examination. The determination unit 213 executes a determination step.

[0035] The display control unit 214 is configured to control screen data to be displayed on the output unit 35 of each information processing device. The display control unit 214 executes a display control step. Note that the screen data may be visual information itself, such as a screen, an image, an icon, or text, generated in a manner that is visible to the user, or may be rendering information for displaying visual information, such as a screen, an image, an icon, or text, on various terminals.

[0036] The information transmitting / receiving unit 210, the learning unit 211, the output control unit 212, the determination unit 213, and the display control unit 214 will be described in detail later.

[0037] 5. Information Processing of Information Processing System 1 Next, an example of a preferable information processing executed by the information processing system 1 of this embodiment will be described.

[0038] In this specification, various data transmitted and received via the network N, the communication unit 23, and the communication unit 33 may be stored in the storage unit 22 and the storage unit 32, respectively.

[0039] Furthermore, in this specification, components that appear in information processing related to the network N, the communication unit 23, the communication unit 33, and the input unit 34 may be omitted. For example, the processor 31 accepting a predetermined input via a user's operation on the input unit 34 may be simply described as the processor 31 accepting a predetermined input from the user. Furthermore, the processor 21 transmitting and receiving information to and from the client device 3 via the communication unit 23, the network N, and the communication unit 33 may be simply described as the processor 21 transmitting information to the client device 3 or simply described as the processor 21 receiving information from the client device 3. The same applies to the case where the processor 31 transmits and receives information to and from the server device 2 via the communication unit 33, the network N, and the communication unit 23.

[0040] 5.1 Overview of Information Processing This section describes an information processing method performed by the information processing system 1 described above. FIG. 5 is a diagram showing an example of an activity showing the flow of information processing executed by the information processing system 1 according to embodiment 1. In FIG. 5, processes A1 to A6 are processes for acquiring the high-risk business operator learning model 5, and processes A7 to A13 are processes for prediction using the high-risk business operator learning model 5.

[0041] In A1, the processor 31 acquires balance information 40, 41, and 42 and high-risk information 43, which become training data 4. The balance information 40, 41, and 42 are information related to account balances and are linked to the high-risk information 43. The high-risk information 43 is information indicating that a business is a high-risk business in a credit screening. The high-risk information 43 may be bad debt information or seizure information. Bad debt information is information indicating whether a business in a credit screening will default. Seizure information is information indicating whether a business in a credit screening meets the conditions for seizure. The high-risk information 43 may be information in an account statement that indicates in an identifiable manner that the conditions for seizure have been met or that there is a high possibility of a bad debt occurring, such as information indicating "seizure" or "bad debt" written on the statement. Training data 4 will be described later using FIG. 6. In A2, the processor 31 receives an instruction from a user to start machine learning. When an instruction to start machine learning is received from the user, the processor 31 transmits the teacher data 4 and a machine learning instruction to the server device 2. In A3, the information transmission / reception unit 210 receives the teacher data 4 and the machine learning instruction from the client device 3. In A4, the learning unit 211 acquires a high-risk business operator learning model 5 using, as features, statistical values ​​calculated from the balances indicated by the balance information 40, 41, and 42 as the teacher data 4 and the high-risk information 43 as the teacher data 4. The acquisition of the high-risk business operator learning model 5 through machine learning will be described later with reference to FIG. 7. In A5, the display control unit 214 transmits screen data to the client device 3 indicating that the high-risk business operator learning model 5 has been acquired. In A6, the processor 31 receives screen data from the server device 2 indicating that the high-risk business operator learning model 5 has been acquired. Thereafter, the processor 31 causes the output unit 35 to display screen data indicating that the high-risk business operator learning model 5 has been acquired. For example, the processor 31 displays on the output unit 35 information indicating that the high-risk business learning model 5 has been acquired, such as "Machine learning has been completed and the learning model has been acquired," based on the screen data.

[0042] The above processes A1 to A6 are processes related to acquiring the high-risk business operator learning model 5. The configurations A1 to A6 make it possible to acquire the high-risk business operator learning model 5, which can determine whether a business operator is a high-risk operator in a credit screening process. Note that, although FIG. 5 describes the processes A1 to A13 as if they were a series of processes, the information processing system 1 may execute the processes A1 to A6 for acquiring the high-risk business operator learning model 5 and the processes A7 to A13 related to predictions using the high-risk business operator learning model 5 at different times, as if they were not a series of processes. For example, if the high-risk business operator learning model 5 has already been acquired, the information processing system 1 may omit processes A1 to A6 and execute only processes A7 to A13.

[0043] In A7, the processor 31 acquires balance information for the specified business operator that is the target of prediction. The processor 31 accepts a prediction instruction using the high-risk business operator learning model 5 from the user. In A8, when a prediction instruction is accepted from the user, the processor 31 transmits the balance information for the specified business operator and the prediction instruction to the server device 2. In A9, the information transmission / reception unit 210 receives the balance information for the specified business operator and the prediction instruction from the client device 3. In A10, the output control unit 212 outputs high-risk information based on the balance information and the high-risk business operator learning model 5. The output of high-risk information using the high-risk business operator learning model 5 will be described later with reference to FIG. 8. In A11, the display control unit 214 generates screen data including high-risk information as display content. The information transmission / reception unit 210 transmits the screen data including high-risk information as display content to the client device 3. In A12, the processor 31 receives screen data including high-risk information as display content from the server device 2. In A13, the processor 31 displays information relating to the high risk information on the output unit 35 based on screen data including the high risk information as display content. The high risk information to be output will be described later with reference to FIG. 9.

[0044] As described above, according to the present disclosure, it is possible to provide a technology that can support the determination of whether a business subject to credit screening is a high-risk business.

[0045] 5.2 Details of Information Processing Next, the details of the information processing outlined above will be explained using Figures 6 to 9. First, the training data 4 will be explained using Figure 6.

[0046] FIG. 6 is a diagram showing an example of training data 4. The training data 4 is data used to construct a high-risk business learning model 5. For example, data on multiple seizing businesses and multiple normal businesses is used as the training data 4. Furthermore, data on multiple defaulting businesses may be used as the training data 4 in addition to or instead of the multiple seizing businesses. The training data 4 includes balance information 40, 41, 42 and high-risk information 43. In the example of FIG. 6, the balance information 40, 41, 42 and the high-risk information 43 are arranged in chronological order from top to bottom (or bottom to top).

[0047] The training data 4 is prepared by a user operating the client device 3. For example, the processor 31 of the client device 3 accesses a database storing pre-processing data for one or more regular businesses, seizing businesses, or bad debt businesses to acquire the pre-processing data for the businesses. The database is provided by an organization capable of managing the account balances of businesses, such as a bank or a Fintech business. The acquired pre-processing data is information indicating the account balance over time, and is information before being converted into training data. Note that the pre-processing data itself may be used as training data, in which case the pre-processing data may be an example of training data. The pre-processing data includes information on a point in time and information indicating the balance amount at that point in time. The pre-processing data may further include information indicating whether the conditions for seizure were met at that point in time, information indicating whether the business owner of the account at that point in time was a seizing business, or information indicating whether the business owner of the account at that point in time was a bad debt business. The processor 31 of the client device 3 performs calculations based on information indicating the balance amount at a certain point in time, which is included in the pre-processing data, to output balance information 40, 41, and 42. Furthermore, the processor 31 links high-risk information 43 to the balance information 40, 41, and 42 through user operation. Furthermore, the high-risk information 43 may be linked to the pre-processing data in advance, and may be, for example, information regarding bad debts included in a statement linked to the balance. In other words, the high-risk information 43 may be managed in a database as a flag linked to the balance, included in the pre-processing data. The information processing for outputting the balance information 40, 41, and 42 and high-risk information 43 described above may be performed by the server device 2.

[0048] The balance information 40, 41, and 42 is information relating to the balance of an account. The balance information 40, 41, and 42 is information calculated from balance information at a certain point in time contained in the pre-processing data. In the example of Figure 6, there are three types of balance information 40, 41, and 42, but it is sufficient if there is one or more types.

[0049] More specifically, the balance information 40, 41, and 42 is information that shows the balance at each predetermined timing in chronological order. The predetermined timing may be every year, month, week, day, hour, etc. The predetermined timing may also be when a transaction related to the account occurs or is completed.

[0050] The balance information 40, 41, 42 includes any value calculated from the account balance. The any value includes a statistical value, a value defined for this information processing, etc. In the embodiment, the average value is used as the balance information 40, but is not limited to this. Any statistical value may be used, for example, the mean, median, minimum value, maximum value, mode, variance, standard deviation, skewness, kurtosis, etc. The balance information 40, 41, 42 may also include information obtained from records of deposits or withdrawals from the account.

[0051] The high-risk information 43 is information indicating whether a business is a high-risk business in credit screening. More specifically, for example, the high-risk information 43 corresponds to at least one of information indicating whether the conditions for seizure were met at the time, information indicating whether the business that owns the account at the time was a seizing business, and information indicating whether the business that owns the account at the time was a defaulting business. More specifically, for example, the high-risk information 43 is information indicating whether the conditions for seizure were met, whether the business was a seizing business, or whether the business was a defaulting business at the time linked to the high-risk information of a certain bank. In the example of Figure 6, the high-risk information 43 is represented by a circle or an X, but it may also be represented by, for example, a "1" or a "0," or whether the characters "default" or "seizing" are included, and is not limited to these.

[0052] 7 is a diagram illustrating an example of information processing related to acquisition of the high-risk business operator learning model 5. FIG. 7 includes training data 4 and the high-risk business operator learning model 5.

[0053] The high-risk business operator learning model 5 is a learning model created using balance information 40, 41, 42 and high-risk information 43 as training data 4. More specifically, for example, the high-risk business operator learning model 5 is a learning model constructed using balance information 40, 41, 42 relating to account balances for a specified period and high-risk information 43 linked to that period (the high-risk information 43 may be included in the account statement) as training data 4.

[0054] This predetermined period is a period prior to the time when the conditions for account seizure were met. More specifically, for example, this predetermined period is (1) a period that does not include a predetermined period going back from the time when the conditions for account seizure were met, and (2) a period that includes a predetermined range of time going back (for example, (1) a period that does not include the period up to seven days prior to the time when the conditions for account seizure were met, and (2) a period that includes a range of 1 to 365 days going back seven days). Note that the conditions (1) and (2) may be set in any time unit, such as year, month, week, day, or hour. This uses balance information up to the time when the conditions for account seizure were met, thereby obtaining a high-risk business operator learning model 5 that can predict the occurrence of high-risk business operators in advance. Furthermore, this allows the high-risk business operator learning model 5 to be obtained that can predict the occurrence of high-risk business operators in advance using information on past high-risk business operators.

[0055] The specified period may also be a period prior to the time the company became a seizing business or a defaulting business. In this case, the condition (2) above remains unchanged, and the condition (1) above is changed to a period that does not include the time from the time the company became a seizing business or a defaulting business up to the predetermined time.

[0056] The explanatory variable serving as a feature of the high-risk business operator learning model 5 is balance information. Specifically, for example, the explanatory variable of the high-risk business operator learning model 5 is balance information including the statistical values ​​described above. Furthermore, the target variable serving as a feature of the high-risk business operator learning model 5 is high-risk information. As a result, by simply providing business operator balance information as an explanatory variable, it is possible to obtain the high-risk business operator learning model 5, which predicts the occurrence of businesses that will become high-risk. Furthermore, by using only balance information and high-risk information, it is possible to predict whether a business is likely to become a defaulting business, without using information such as the business's years of establishment, capital, and number of employees, which are commonly used in credit screening.

[0057] The high-risk business operator learning model 5 is constructed by any machine learning algorithm.

[0058] 8 is a diagram illustrating an example of information processing for outputting high-risk information from the high-risk business operator learning model 5. FIG. 8 includes input data 6, the high-risk business operator learning model 5, and output data 7.

[0059] The input data 6 is balance information of a business operator whose potential high-risk status is to be predicted in a credit screening. The statistical values ​​may be calculated by a user using the client device 3, or may be processed by the server device 2 that has acquired the input data 6 before inputting the data into the high-risk business operator learning model 5. The input data 6 includes balance information 60, 61, and 62, which will be described later with reference to FIG. 9 .

[0060] The high-risk business operator learning model 5 is a learning model that is trained to output output data 7 in response to balance information being input as input data 6.

[0061] The output data 7 is data output from the high-risk business operator learning model 5. The output data 7 includes high-risk probability information 70 and high-risk occurrence information 71 as shown in Fig. 9. The high-risk probability information 70 and the high-risk occurrence information 71 are examples of high-risk information.

[0062] FIG. 9 is a diagram showing an example of output data 7 including high risk probability information 70 and high risk occurrence information 71. The output data 7 may be data in a format in which the high risk probability information 70 and high risk occurrence information 71 are added to the input data 6. The output data 7 includes balance information 60, 61, 62, high risk probability information 70 linked to the balance information, and high risk occurrence information 71 linked to the balance information. The balance information 60, 61, 62 is information related to the account balance. The balance information 60, 61, 62 is information calculated from the balance information at the relevant time point in the pre-calculation balance information. In the example of FIG. 9, there are three types of balance information, but one or more types may be used.

[0063] The high-risk probability information 70 as output data 7 includes information on the probability of a business being a high-risk business. The output probability may be in any format, such as 0 to 100%, 0 to 1, etc. The output control unit 212 outputs the high-risk probability information 70 for each period indicated by the balance information included in the input data 6, based on the input data 6 and the high-risk business learning model 5. By referring to the high-risk information for the most recent period, it is possible to confirm the probability that a business is likely to be a high-risk business, and this can be used as a reference for examining the business's creditworthiness, such as its repayment ability, during credit screening.

[0064] Furthermore, the high-risk occurrence information 71 as output data 7 includes information indicating whether or not the business operator is likely to be a high-risk business operator. In the example of FIG. 9 , the high-risk occurrence information 71 is represented by a circle or an X, but may also be represented by a "1" or a "0," or whether or not the characters "bad debt" or "seizure" are included, and is not limited to these. The determination unit 213 outputs information indicating that the business operator is likely to be a high-risk business operator when the probability of the business operator being a high-risk business operator indicated by the high-risk probability information 70 output by the output control unit 212 is equal to or greater than a threshold (e.g., 60% or greater when the probability is expressed as 0 to 100%). The determination unit 213 also outputs information indicating that the business operator is unlikely to be a high-risk business operator when the probability of the business operator being a high-risk business operator indicated by the high-risk probability information 70 output by the output control unit 212 is less than a threshold (e.g., less than 60% when the probability is expressed as 0 to 100%). This makes it possible to obtain predictions as to whether a business is likely to become a high-risk business, which can be used as a reference for assessing the business's creditworthiness, such as its financial strength and repayment ability, during credit screening.

[0065] Next, we will explain the case where a seized business is approximately interpreted as a default business. In this case, seizure information regarding whether a business's account will be seized is used as the high-risk information as training data, and bad debt information is used as the high-risk information as output data. This bad debt information includes information on the probability of a business becoming a default or information indicating whether a business will become a default. This makes it possible to obtain a high-risk business learning model and predict the occurrence of default businesses using balance information on seized businesses, which is easier to obtain than balance information on default businesses.

[0066] [Second Embodiment] 6. Information Processing Method Section 6 describes the flow of information processing according to the second embodiment. In the second embodiment, descriptions that overlap with those in the first embodiment will be omitted as appropriate.

[0067] 10 is a diagram showing an example of an activity illustrating the flow of information processing executed by the information processing system 1 according to the second embodiment. In the following embodiments, a factoring service will be used as an example of a finance service, and the terms "finance service" and "factoring service" will be used as appropriate. Here, the explanation begins from the point when business operator A, the owner of the client device 3, has finished inputting the necessary information into the input unit 34 in order to use the factoring service operated by business operator B, the owner of the server device 2.

[0068] In A21, the processor 31 in the client device 3 transmits each piece of information required to apply for the factoring service (hereinafter also referred to as "required information") to the server device 2. Here, the required information includes, for example, the accounts receivable of business A, the desired procurement amount, the desired payment terms and repayment terms, etc. In A21, for example, the following two-stage information processing is executed: (1) The processor 31 reads the required information from the memory unit 32. (2) The processor 31 transmits the required information to the server device 2 via the communication unit 33.

[0069] In A22, the processor 21 in the server device 2 receives the necessary information from the client device 3. In A22, for example, the following two-stage information processing is executed: (1) The communication unit 23 receives the necessary information from the client device 3. (2) The processor 21 stores the necessary information in the storage unit 22.

[0070] In A23, the processor 21 in the server device 2 sets a financing cost to be applied to business operator A (hereinafter also referred to as "used financing cost") based on summary statistics (such as mean, median, or mode; equivalent to "statistics" in the claims) that represent the center of the current financing cost, the high-risk information output in A10, and coefficients according to the payment terms and repayment terms desired by business operator A. Note that the current financing cost may be the financing cost stored in the memory unit 22 as the going rate for financing costs set in the factoring service, and this financing cost may be updated as appropriate.

[0071] In other words, in the output control step, a used finance cost, which is a finance cost to be applied when a predetermined business uses a finance service, is output based on a statistical value calculated from the market price of finance costs set in the finance service and high-risk information. According to this aspect, an appropriate finance cost can be applied to a business using the finance service in accordance with the risk of default or seizure.

[0072] In A23, for example, the following three-stage information processing is executed. (1) The processor 21 reads out the current financing cost, high-risk information, and coefficients according to the payment terms and repayment terms from the storage unit 22. (2) The processor 21 executes a setting process to set the used financing cost from the summary statistics representing the center of the current financing cost after the calculation process, the high-risk information, and the coefficients. (3) The processor 21 stores the used financing cost in the storage unit 22.

[0073] In A24, the processor 21 in the server device 2 waits until the examination results of the application by the business operator A are stored in the memory unit 22. That is, when the examination results are received by the communication unit 23 and the received signal is transmitted to the processor 21 via the communication bus, the processor 21 stores the examination results in the memory unit 22 and proceeds to the process of A25.

[0074] In A25, the processor 21 in the server device 2 transmits information on the set used finance cost to the client device 3. In A25, for example, the following two-stage information processing is executed: (1) The processor 21 reads the used finance cost from the memory unit 22. (2) The processor 21 transmits the information on the used finance cost to the client device 3 via the communication unit 23.

[0075] In A26, the processor 31 in the client device 3 receives information on the used finance cost from the server device 2. In A26, for example, the following two-stage information processing is executed: (1) The communication unit 33 receives information on the used finance cost from the server device 2. (2) The processor 31 stores the information on the used finance cost in the memory unit 32.

[0076] In A27, the processor 31 in the client device 3 displays information on the used finance cost on the output unit 35. In A27, for example, the following three-stage information processing is executed: (1) The processor 31 reads information on the used finance cost from the memory unit 32. (2) The processor 31 displays the used finance cost on the output unit 35.

[0077] 11 is a diagram showing an example of a screen displayed in A27. Fig. 11 illustrates a case of a factoring service, which is one of the financing services. The output section 35 displays the results of the screening, and more specifically, an area 81 and a confirmation button 82.

[0078] Area 81 displays the result of the main screening as "Main screening passed." It also displays "5 million yen" as the amount that can be raised. Furthermore, the terms and conditions of this application are displayed as "Payment terms: 60 days," "Total fee: 100,000 yen," and "Fees rate: 2.0% / month." The fees displayed here are an example of financing costs, and the fees rate corresponds to the "used financing cost" in the explanation of A27. The "used financing cost" here is set based on coefficients corresponding to the payment terms and repayment terms so that the total financing cost when the used financing cost is applied is at the same level as the total financing cost when the market price of financing costs is applied. According to this embodiment, an appropriate financing cost can be applied while keeping in line with the market price of financing costs for financing services.

[0079] The confirmation button 82 is a button that is clicked after the displayed result of the main examination is confirmed. Clicking the confirmation button 82 triggers a transition to an input screen for using a factoring service, for example.

[0080] [Embodiment 3] 7. Information Processing Method Section 7 describes the flow of information processing according to embodiment 3. In embodiment 3, descriptions that overlap with embodiment 1 or embodiment 2 will be omitted as appropriate.

[0081] 12 is a diagram showing an example of an activity showing the flow of information processing executed by the information processing system 1 according to the third embodiment. Each of the subsequent activities may be executed as appropriate after the information processing method according to the first embodiment. Here, the owner of the client device 3 is assumed to be a business operator C.

[0082] In A31, the processor 21 in the server device 2 inputs the balance information of business operator C received in A9 and the high-risk information of business operator C output in A10 into a second learning model (corresponding to "reference information" in the claims), and causes the second learning model to output the results of the preliminary screening of business operator C in the case where it is assumed that the factoring service will be used (hereinafter also simply referred to as the "results of the preliminary screening of business operator C"). Here, the results of the preliminary screening are information including, for example, information on whether or not the preliminary screening was passed, an estimate of the amount that can be procured in the case where the main screening is passed, and the like.

[0083] In other words, in the output control step, the results of a preliminary screening for a specified business assuming that the business will use a factoring service are output based on the balance information, high-risk information, and reference information. Here, the reference information, exemplified by the second learning model, is information indicating the relationship between the balance information, high-risk information, and the results of the preliminary screening. That is, the second learning model is a model trained using training data on the relationship between the business's balance information, the business's high-risk information, and the results of the business's preliminary screening. Thus, the second learning model is configured to output the results of the preliminary screening when the balance information and the high-risk information are input. According to this aspect, information on screening documents (e.g., the business's financial statements, more specifically, financial statements, corporate tax returns, account item breakdowns, etc.) that were previously required is no longer necessary, making it possible to appeal to inactive businesses in financial services, which can contribute to improving the screening application rate for financial services.

[0084] In A31, for example, the following three-stage information processing is executed: (1) The processor 21 reads out the balance information of business operator C, the high-risk information of business operator C, and the second learning model from the memory unit 22. (2) The processor 21 inputs the balance information and the high-risk information into the second learning model. (3) The processor 21 stores the results of the preliminary screening of business operator C output from the second learning model in the memory unit 22.

[0085] Here, if business C is a business with a probability of passing the screening for factoring services above a threshold, it is preferable to output the results of the preliminary screening for business C. In other words, in the output control step, the results of the preliminary screening are output for businesses with a probability of passing the screening for financing services above a threshold. That is, for example, when business C's high-risk information is read in (1) above, a comparison with a preset threshold may be performed, and if the probability of passing the screening is above the threshold (if business C is expected to pass the screening), the process may proceed to information processing (2) above. This aspect can prevent a user experience in which a business that applied for the main screening is rejected due to the results of the preliminary screening.

[0086] In A32, the processor 21 in the server device 2 transmits the results of the preliminary screening of business operator C to the client device 3. In A32, for example, the following two-stage information processing is executed: (1) The processor 21 reads out the results of the preliminary screening of business operator C from the memory unit 22. (2) The processor 21 transmits the results of the preliminary screening of business operator C to the client device 3 via the communication unit 23.

[0087] In A33, the processor 31 in the client device 3 receives the results of the preliminary screening of business operator C from the server device 2. In A33, for example, the following two-stage information processing is executed: (1) The communication unit 33 receives the results of the preliminary screening of business operator C from the server device 2. (2) The processor 31 stores the results of the preliminary screening of business operator C in the memory unit 32.

[0088] In A34, the processor 31 in the client device 3 displays the results of the preliminary screening of business operator C on the output unit 35. In A34, for example, the following two-stage information processing is executed: (1) The processor 31 reads out the results of the preliminary screening of business operator C from the memory unit 32. (2) The processor 31 displays the results of the preliminary screening of business operator C on the output unit 35.

[0089] 13 is a diagram showing an example of a screen displayed in A34. The output section 35 displays a message indicating that the factoring service pre-screening has been passed, and more specifically, an area 91 and a confirmation button 92 are displayed.

[0090] Area 91 displays an estimate of the amount that can be procured if the screening is passed. Area 91 displays "5 million yen" as the maximum amount. Here, it is preferable that the maximum amount, i.e., the upper limit amount, be set in advance as the estimate of the amount that can be procured. By setting the upper limit amount in advance, it is possible to prevent a significant discrepancy from the amount actually procured through the factoring service.

[0091] The confirmation button 92 is a button that is clicked after the displayed preliminary screening results are confirmed. Clicking the confirmation button 92 triggers a transition to, for example, a home screen for the factoring service.

[0092] [Modifications] Modifications of this embodiment will be described below. The following modifications can be combined as appropriate.

[0093] The program is a program that causes one or more computers to execute each functional unit (step). The information processing system 1 includes one or more computers that execute the program. With respect to the information processing system 1 according to the above-described embodiment, the program may be a program that causes a computer to function as the processor 21 of the information processing system 1. Alternatively, the program may be an information processing method executed by the information processing system 1 (or the processor 21 of the server device 2).

[0094] At least one of the devices included in the information processing system 1 may be installed outside Japan. For example, the server device 2 may be installed outside Japan, and each terminal device may be installed inside Japan. Similarly, a user may access the server device 2 installed inside Japan from outside Japan using their own terminal device. According to such an embodiment, a more convenient experience can be provided to the user through various management forms.

[0095] The server device 2 may be an on-premise server or a cloud server. The cloud server device 2 may provide the above functions and processes in the form of, for example, SaaS (Software as a Service) or cloud computing.

[0096] In the above embodiment, the server device 2 performs various storage and control operations, but multiple external devices may be used instead of the server device 2. That is, various information and programs may be distributed and stored in multiple external devices using blockchain technology or the like.

[0097] In the above embodiment, the information processing is performed by at least two devices, the server device 2 and the client device 3. However, in a modified example, the information processing may be performed by the client device 3 alone.

[0098] In A23 of the second embodiment, summary statistics (such as the mean, median, and mode) that represent the center of the current finance costs are used, but the present invention is not limited to these. In addition to the mean, any statistical value, such as the minimum value, maximum value, mode, variance, standard deviation, skewness, and kurtosis, may also be used.

[0099] In the above embodiment, a factoring service has been described as an example of a finance service, but the present invention is not limited to this. Finance services refer to services such as fund procurement, fund use, fund management, and fund utilization, and are a concept that includes services such as finance and loans.

[0100] In the above embodiment, the financing cost is exemplified by a fee and a fee rate, but is not limited to this. The financing cost is a concept that includes, for example, a financing interest rate in addition to a fee and a fee rate.

[0101] [Others] Furthermore, the present invention may be provided in the following aspects.

[0102] (1) A program that causes an information processing system to execute the following steps: in an acquisition step, balance information of a specified business is acquired, the balance information being information regarding the account balance; in an output control step, high-risk information is output based on the balance information and a high-risk business learning model, the high-risk information being information indicating whether the specified business is a high risk in credit screening; and the high-risk business learning model is a learning model obtained by learning the balance information and high-risk information as training data.

[0103] (2) In any one of the programs described in (1) above, the high-risk information as the training data is information regarding whether the business's account will be seized.

[0104] (3) In the program described in (2) above, the high-risk business operator learning model is a learning model obtained by learning balance information regarding the balance and details of an account for a specified period as the training data, and the specified period is a period prior to when the conditions for seizure of the account are met.

[0105] (4) In the program described in any one of (1) to (3) above, the feature of the high-risk business operator learning model includes a statistical value calculated from the balance indicated by the balance information.

[0106] (5) A program according to any one of (1) to (4) above, wherein the high-risk information includes information on the probability that the business operator may become a high-risk business operator.

[0107] (6) In the program described in (5) above, the information processing system further executes a judgment step, and the high-risk information includes information indicating whether the business operator has the potential to become a high-risk business operator. In the judgment step, if the probability that the business operator has the potential to become a high-risk business operator indicated by the high-risk information output in the output control step is greater than or equal to a threshold, the information processing system outputs information indicating that the business operator has the potential to become a high-risk business operator. In addition, if the probability that the business operator has the potential to become a high-risk business operator indicated by the high-risk information output in the output control step is less than a threshold, the information processing system outputs information indicating that the business operator has the potential to become a high-risk business operator.

[0108] (7) In the program described in any one of (1) to (6) above, the output control step outputs a utilization finance cost, which is the finance cost applied when the specified business uses the finance service, based on a statistical value calculated from the market price of finance costs set in the finance service and the high risk information.

[0109] According to this aspect, it is possible to apply appropriate financing costs to businesses that use financing services in accordance with the risk of bad debts and seizures.

[0110] (8) In the program described in (7) above, the used financing cost is set based on a coefficient corresponding to the payment terms and repayment terms so that the total amount of financing costs when the used financing cost is applied is at the same level as the total amount of financing costs when the market price of the financing cost is applied.

[0111] According to this embodiment, it is possible to apply an appropriate financing cost in accordance with the market price of financing services.

[0112] (9) In the program described in any one of (1) to (8) above, the output control step outputs the results of a preliminary screening for the specified business operator assuming that a financial service is used based on the balance information, the high risk information, and reference information, and the reference information is information indicating the relationship between the balance information, the high risk information, and the results of the preliminary screening.

[0113] According to this aspect, the information on the type of screening that was previously required is no longer necessary, making it possible to appeal to businesses that are not yet active in financial services, which can contribute to improving the rate of applications for screening for financial services.

[0114] (10) In the program described in (9) above, the output control step outputs the results of the preliminary screening for businesses whose likelihood of passing the screening for the financial service is above a threshold.

[0115] According to this aspect, it is possible to prevent a user experience in which a business that applied for the main review is rejected due to the results of the preliminary review.

[0116] (11) An information processing method executed by an information processing system, the information processing method comprising the steps of the program described in any one of (1) to (10) above.

[0117] (12) An information processing system comprising one or more processors that execute the program according to any one of (1) to (10) above. Of course, this is not a limitation.

[0118] Finally, while various embodiments of the present invention have been described, they are presented by way of example only and are not intended to limit the scope of the invention. The novel embodiments may be embodied in various other forms, and various omissions, substitutions, and modifications may be made without departing from the spirit of the invention. Such embodiments and modifications are intended to be included within the scope and spirit of the invention, as well as within the scope of the invention and its equivalents as defined in the accompanying claims.

[0119] 1: Information processing system, 2: Server device, 3: Client device, 4: Training data, 5: High-risk business operator learning model, 6: Input data, 7: Output data, 21: Processor, 22: Memory unit, 23: Communication unit, 31: Processor, 32: Memory unit, 33: Communication unit, 34: Input unit, 35: Output unit, 40: Balance information, 41: Balance information, 42: Balance information, 43: High-risk information, 60: Balance information, 61: Balance information, 62: Balance information, 70: High-risk probability information, 71: High-risk occurrence information, 81: Area, 82: Confirmation button, 91: Area, 92: Confirmation button, 210: Information transmission / reception unit, 211: Learning unit, 212: Output control unit, 213: Determination unit, 214: Display control unit, N: Network

Claims

1. A program that causes an information processing system to execute the following steps: in an acquisition step, balance information of a specified business is acquired, the balance information being information regarding an account balance; in an output control step, high risk information is output based on the balance information and a high-risk business learning model, the high risk information being information indicating whether the specified business is a high risk in credit screening, and the high-risk business learning model is a learning model obtained by learning using balance information and high risk information as training data.

2. A program according to any one of claims 1, wherein the high-risk information as the training data is information regarding whether the business's account will be seized.

3. A program as claimed in claim 2, wherein the high-risk business learning model is a learning model obtained by learning balance information regarding the balance and details of an account for a specified period as the training data, and the specified period is a period prior to when the conditions for seizure of the account are met.

4. A program according to any one of claims 1 to 3, wherein the feature of the high-risk business learning model includes a statistical value calculated from the balance indicated by the balance information.

5. A program according to any one of claims 1 to 4, wherein the high-risk information includes information on the probability that the business entity may become a high-risk business entity.

6. A program as described in claim 5, wherein the information processing system further executes a judgment step, wherein the high risk information includes information indicating whether the business operator is likely to become a high-risk business operator, and in the judgment step, when the probability that the business operator is likely to become a high-risk business operator indicated by the high risk information output in the output control step is equal to or greater than a threshold, outputting information indicating that the business operator is likely to become a high-risk business operator, and when the probability that the business operator is likely to become a high-risk business operator indicated by the high risk information output in the output control step is less than a threshold, outputting information indicating that the business operator is not likely to become a high-risk business operator.

7. A program as claimed in any one of claims 1 to 6, wherein the output control step outputs a utilization financing cost, which is the financing cost applied when the specified business uses the financing service, based on a statistical value calculated from the market price of financing costs set in the financing service and the high risk information.

8. A program according to claim 7, wherein the used financing cost is set based on a coefficient corresponding to payment terms and repayment terms so that the total amount of financing cost when the used financing cost is applied is at the same level as the total amount of financing cost when the market price of the financing cost is applied.

9. A program as claimed in any one of claims 1 to 8, wherein in the output control step, the results of a preliminary screening for the specified business operator in the case where a financial service is assumed to be used are output based on the balance information, the high risk information and reference information, and the reference information is information indicating the relationship between the balance information, the high risk information and the results of the preliminary screening.

10. A program according to claim 9, wherein the output control step outputs the results of the preliminary screening for businesses whose likelihood of passing the screening for the financial service is equal to or greater than a threshold.

11. An information processing method executed by an information processing system, comprising each step of the program according to any one of claims 1 to 10.

12. An information processing system comprising one or more processors that execute the program according to any one of claims 1 to 10.

Images