JP2020035417A - Banking business support system, banking business support method and banking business support program - Google Patents

Abstract

To provide a banking business support system, a method and a program for supporting performing right and efficient banking business.SOLUTION: A control section 21 of a support server 20 acquires external situation information, a financial transaction condition, customer situation information and a financial transaction situation of the same time from a back-end system 40; performs machine learning using teacher data having the acquired information as a data set; creates a customer behavior prediction model associating the financial transaction condition with the financial transaction situation; and calculates the financial transaction condition corresponding to a target value of a transaction situation using the customer behavior prediction model for a present external situation and a customer situation.SELECTED DRAWING: Figure 1

Description

  The present disclosure relates to a banking support system, a banking support method, and a banking support program for supporting banking.

  Banks provide various financial services and products such as deposits and loans. Then, technologies for predicting the behavior of a financial institution customer and proposing a service or a product are being studied (for example, see Patent Documents 1 and 2). In the technology described in Patent Document 1, motivation is provided only to customers who may visit the store. The customer behavior management server of Patent Literature 1 predicts the visit date and time of a customer based on the visit history of the customer to the store and the stop-over information of the customer. It is determined whether or not the predicted visit date and time corresponds to the busy time of this store. If it is determined that the customer's visit date and time corresponds to the busy time, the customer is notified of information prompting the customer to visit another store.

  The model creation device described in Patent Literature 2 models customer behavior using the recognition level of a customer's service as a recognition level including recognition of the existence of a service and recognition of a service function. Then, the prediction device predicts which service the customer selects using the customer behavior model.

  In addition, technologies for supporting asset management by clients' securities investment and facilitating lending operations of financial institutions and sound management have been studied (for example, see Patent Document 3). In the technology described in Patent Document 3, a financial institution server issues a deposit account DB that stores information of a customer's deposit account opened at a financial institution and securities that are stored and managed by a securities company. And a depository account account DB that stores information on the depository deposit account of the customer opened at the financial institution in a form in which the depository is borrowed from the customer by the financial institution. Then, it accepts the customer's buying and selling instructions, creates trading agent information, transmits it to the securities company server, receives trading result information from the securities company server, and updates the balances in the deposit account DB and the depositary securities account DB.

  In addition, a technology that supports business by using machine learning or the like has been studied (for example, see Patent Document 4). In the technology described in Patent Literature 4, a receiving unit receives a response to a requested task and a reason for determination from a contractor. The type classification unit classifies the determination reason into a type based on the determination reason and the answer. The type extraction unit extracts a reason for determination belonging to a predetermined type based on the classified type. The providing unit provides the extracted determination reason.

JP-A-176-173649A JP 2005-352728 A JP 2007-47940 A JP-A-17617777

  The technique described in Patent Document 1 predicts the behavior of each customer and does not predict customer behavior that contributes to the entire management. Further, in the technology described in Patent Literature 2, control of customer behavior in a bank is not considered. In the technology described in Patent Literature 3, sound management of financial institutions is considered, but it is not assumed that customer behavior is promoted.

  The banking support system that solves the above problems stores, at each time, financial transaction conditions of financial transactions, customer status information, history information including the status of financial transactions with customers, and external status information. And a control unit connected to the user terminal and the information storage system. Then, the control unit acquires external situation information, financial transaction conditions, customer situation information, and financial transaction situation at the same time from the information storage system, and uses the acquired information as a data set for teacher data. Performs learning, generates a customer behavior prediction model that associates financial transaction conditions with financial transaction status, and uses the customer behavior prediction model for the current external situation and customer situation to correspond to the target value of the transaction situation. Calculate financial transaction conditions.

  According to the present disclosure, it is possible to support accurate and efficient execution of banking business.

FIG. 1 is an explanatory diagram of a banking support system according to the embodiment. FIG. 4 is an explanatory diagram of a hardware configuration example. It is explanatory drawing of the information storage part of this embodiment, (a) is a customer information storage part, (b) is an account information storage part, (c) is a transaction information storage part, (d) is a condition information storage part. FIG. It is explanatory drawing of the information storage part of this embodiment, (a) is an external situation storage part, (b) is a regulation information storage part, (c) is an explanatory view of a management information storage part. Explanatory drawing of the relationship between customer behavior and bank behavior of the present embodiment. Explanatory drawing of the balance sheet of this embodiment. Explanatory drawing of the customer behavior model of this embodiment. Explanatory drawing of the processing procedure of other embodiment. Explanatory drawing of the customer behavior model of other embodiment. Explanatory drawing of the processing procedure of other embodiment.

(First embodiment)
Hereinafter, an embodiment of a banking business support system, a banking business support method, and a banking business support program will be described with reference to FIGS. In the present embodiment, it is assumed that a bank (financial institution) controls financial transaction conditions of services and products provided to a bank customer in order to realize an accurate balance sheet.

Here, as shown in FIG. 1, a banking business system including a user terminal 10, a support server 20, and a back-end system 40 (business system) is used.
First, the concept of the present embodiment will be described with reference to FIG. It is assumed that customer behavior A1 is predicted using machine learning. The customer behavior A1 includes depositing funds in a bank and borrowing funds from a bank. The customer behavior A1 depends on the customer situation D01 according to the segment to which the customer attribute belongs, and the external situation D02 such as the price and the market situation. In this case, the bank can prompt the customer behavior A1 by controlling the deposit condition D03 and the lending condition D04 as financial transaction conditions. Then, the borrowing (loan) and the deposit by the customer behavior A01 determine the cash D05 of the bank.

  The bank action A2 is performed according to the cash D05. This bank action A2 includes financing and investment. In the present embodiment, it is assumed that securities are purchased, for example, as investments. This bank action A2 depends on external situations D11 such as market conditions and interbank interest rates, and regulations D12 such as restrictions.

Then, the balance sheet BS1 of the bank is determined based on the cash D05, financing and investment.
Next, the bank balance sheet BS1 will be described with reference to FIG. The balance sheet BS1 includes cash, deposits, loans, securities, and other items as debits. The credit is composed of customer deposits and net assets. In this case, each item constituting the balance sheet BS1 has a priority or a restriction.

  For example, as the priority (1), there is “the cash is a% of the customer deposit (for example, about 20%)”. Cash is to prepare for the termination of customer deposits, but the profitability is low, so it is better to have a small percentage of cash.

  As the priority (2), there is “b% of loans and securities are customer deposits (for example, about 70%)”. Loans and securities have high profitability, but have a credit risk, resulting in a decrease in the equity ratio. Therefore, it is necessary to control the amount by the lending interest rate and the lending period.

  As the priority (3), there is “c% (for example, about 10%) excluding the priority (1) and (2) of securities”. Although securities are subordinate to loans in terms of profitability, government bonds have no credit risk. Therefore, it is used as surplus cash in cash and deposits.

  In addition, as the constraint (1), there is a regulation regarding the “equity ratio”. This capital adequacy ratio regulation is a capital adequacy ratio for risk assets, and is a regulation that makes the Tier 1 ratio of common stock etc. about 10%.

As the restriction (2), there is a regulation on “net assets”. The net assets are, for example, about 10% of the total assets. This net asset is used as a buffer when a loss occurs.
As the constraint (3), there is a regulation on the “liquidity coverage ratio”. The liquidity coverage ratio is to secure assets with high liquidity in response to cash outflow during stress for 30 days.

As the restriction (4), there is a regulation on the “stable procurement ratio”. This stable funding ratio is to secure stable funding over the medium to long term for assets that are difficult to sell in the short term.
The constraint (5) relates to “IRRRBB (interest rate risk of the bank)”. Specifically, the loss of interest rate risk (for example, US dollar / Euro: 200 bp, Japanese yen: 100 bp when interest rate rises) is to be limited to 15% of Tier 1.

The user terminal 10 is a computer terminal used by a person in charge of a bank.
The support server 20 includes a control unit 21 and a learning result storage unit 22. The control unit 21 includes a support management unit 211, a learning unit 212, a prediction unit 213, and an analysis unit 214.

  The support management unit 211 executes processing for supporting banking operations. Specifically, the support management unit 211 acquires various information from the back-end system 40 and provides the information to the learning unit 212 and the prediction unit 213. Further, it obtains a prediction result from the prediction unit 213 and outputs it to the user terminal 10. The support management unit 211 holds a condition candidate list. In this condition candidate list, data on a plurality of deposit condition candidates and lending condition candidates used for the support processing are recorded.

  The learning unit 212 performs machine learning by using teacher data including variables constituting the input layer and the output layer, and executes a process of generating a prediction model. The learning unit 212 generates a prediction model used for prediction from various types of information by multivariate analysis, machine learning, and deep learning.

The prediction unit 213 performs a process of predicting each variable using the prediction model generated by the learning unit 212.
The analysis unit 214 executes a process of predicting an action that maximizes profit in a business operation of the company. The analysis unit 214 holds data on a reference ratio for evaluating the cash and deposit. Further, the analysis unit 214 executes an investment simulation and a procurement simulation according to the cash and customer balances. In the investment simulation and the procurement simulation, the simulation is performed in consideration of the above-described regulations (priorities and restrictions).

  In the investment simulation, a simulation for purchasing securities is performed using surplus cash and deposits. For example, forecast the profit and loss of purchasing government bonds with the current surplus.

  In the procurement simulation, a simulation for procuring cash is performed using the shortage of cash and deposits. For example, the profit and loss when cash is raised using the interbank interest rate with the current surplus is predicted.

  In the learning result storage unit 22, prediction model data for predicting customer behavior is recorded. This prediction model data is recorded when learning is performed based on the customer situation, external situation, and interest rate. In the learning result storage unit 22, prediction model data relating to the customer is recorded.

  The back-end system 40 functions as an information storage system. Therefore, the back-end system 40 includes a customer information storage unit 42, an account information storage unit 43, a transaction information storage unit 44, a condition information storage unit 45, an external situation storage unit 46, a regulation information storage unit 47, and a management information storage unit 48. Is provided.

  As shown in FIG. 3A, the customer information storage unit 42 stores a customer management record for a bank customer. The customer management record is registered when the customer opens an account at the bank, and is updated when it is changed. The customer management record records customer identifiers such as customer numbers, customer information, customer attributes, and account identifier data.

In the customer identifier data area, data relating to an identifier for specifying each bank customer is recorded.
In the customer information data area, data on personal information (name, account identifier, contact information, address, etc.) of the customer is recorded.

  In the customer attribute data area, data on the attributes (age, gender, occupation, etc.) of this customer is recorded. The segment to which the customer belongs is determined by the customer attribute. This customer attribute is stored in association with the registration time and the change time.

In the account identifier data area, data relating to identifiers (head office code, type code, account number) for specifying the account held by the customer is recorded.
As shown in FIG. 3B, the account information storage unit 43 records an account management record 430 related to a bank account opened in a bank. The account management record 430 is recorded when the balance of an account opened in a bank is statistically evaluated. In the account management record 430, data relating to the evaluation time, segment, deposit type, and total amount of deposit balance are recorded.

In the evaluation time data area, data relating to the time (year, month, and day) at which the account opened in the bank was evaluated is recorded.
In the segment data area, data relating to an identifier for specifying a segment of a customer attribute is recorded.

In the deposit type data area, data relating to an identifier for specifying the type of deposit (for example, ordinary deposit, time deposit, etc.) is recorded.
In the deposit balance total amount data area, data on the total amount of balances of all accounts of the deposit type held by customers belonging to this segment is recorded.

  As shown in FIG. 3C, the transaction information storage unit 44 records a transaction management record 440 relating to a transaction with a customer (loaned in this embodiment). The transaction management record is recorded when the status of a transaction with a customer is statistically evaluated. In this transaction management record, data on the evaluation time, segment, lending type, and total lending balance are recorded.

In the evaluation time data area, data relating to the time (year, month, and day) at which the loan was evaluated at the bank is recorded.
In the segment data area, data relating to an identifier for specifying a segment of a customer attribute is recorded.

In the lending type data area, data relating to an identifier for specifying the type of lending (for example, a home loan, a card loan, etc.) is recorded.
In the loan balance total data area, data relating to the total amount of the balance of all loans of the loan type to the customers belonging to this segment is recorded.

  As shown in FIG. 3D, the condition information storage unit 45 stores a deposit condition management record 450 and a lending condition management record 451. The deposit condition management record 450 and the loan condition management record 451 are recorded when financial transaction conditions (deposit conditions and loan conditions) to be provided to the customer are determined.

The deposit condition management record 450 records data relating to a change time, a deposit type, and a deposit condition.
In the change time data area, data relating to the date when the deposit condition was changed is recorded.

In the deposit type data area, data relating to an identifier for specifying the type of deposit is recorded.
In the deposit condition data area, data on the changed deposit condition (deposit interest rate, deposit period) is recorded.

The lending condition management record 451 records data on the change time, the lending type, and the lending condition.
In the change time data area, data on the date when the lending condition was changed is recorded.

  In the lending type data area, data relating to an identifier for specifying the type of lending is recorded. Here, the lending type according to the segment of the lending customer or the credit score may be recorded in the lending type data area.

In the lending condition data area, data on the changed lending condition (lending interest rate, lending period) is recorded.
As shown in FIG. 4A, data relating to various external situations is recorded in the external situation storage unit 46. In the present embodiment, economic index information 461, market information 462, and interest rate information 463 are used as external situation information.

  In the economic index information 461, information on the economic index is recorded in association with the time (year, month, and day) when the economic index information was acquired. As this economic index, for example, information on GDP, prices, and the like can be used.

  In the market information 462, market information is recorded in association with the time (date) when the market information was acquired. As the market information, for example, index information on government bonds (domestic / overseas), stocks (domestic / overseas), foreign exchange, and the like can be used.

  The interest rate information 463 records the interest rate information in association with the time (date) when the interest rate information was acquired. As the interest rate information, for example, a self-bank interest rate, an overseas interest rate, another bank interest rate, an interbank interest rate, and the like can be used.

As shown in FIG. 4B, the regulation information storage unit 47 records regulation information on various regulations. This regulation information is recorded when the regulation is registered by the person in charge.
In the regulation information, information on regulations (priority, restrictions, etc.), that is, priority information 471 and restriction information 472, is recorded in association with the time (year, month, day) when the restrictions were set.

  As shown in FIG. 4C, in the management information storage unit 48, various types of information on each element constituting a balance sheet explaining the business state of the bank are recorded. In the present embodiment, held securities information 481, financing information 482, and net asset information 483 are recorded.

The held securities information 481 records information on securities (brand, quantity, price) held by the bank.
In the fund procurement information 482, information on the fund procured by the bank (procurement source, amount) is recorded.

The net asset information 483 records information on net assets (assets and prices) held by the bank.
(Example of hardware configuration)
FIG. 2 is a hardware configuration example of an information processing device H10 functioning as the user terminal 10 to the back-end system 40 and the like.

  The information processing device H10 includes a communication device H11, an input device H12, a display device H13, a storage unit H14, and a processor H15. Note that this hardware configuration is an example, and may have other hardware.

  The communication device H11 is an interface that establishes a communication path with another device to transmit and receive data, and is, for example, a network interface card or a wireless interface.

The input device H12 is a device that receives an input from a user or the like, and is, for example, a mouse or a keyboard.
The display device H13 is a display or a touch panel that displays various information.

  The storage unit H14 is a non-temporary computer-readable medium or storage device that stores data and various programs for executing various functions of the user terminal 10 to the back-end system 40 (for example, the learning result storage unit 22, the advance preparation information, The storage unit 23, the customer information storage unit 42 to the management information storage unit 49). Examples of the storage unit H14 include a ROM, a RAM, and a hard disk.

  The processor H15 controls each process in the user terminal 10 to the back-end system 40 using programs and data stored in the storage unit H14. Examples of the processor H15 include, for example, a CPU and an MPU. The processor H15 expands a program stored in a ROM or the like into a RAM and executes various processes corresponding to various processes. For example, when the banking support program is started, the processor H15 operates a process for executing each process shown in FIG. 7 described below.

(Support processing)
The support processing will be described with reference to FIGS.
Here, as shown in FIG. 7, a customer behavior prediction model for predicting a borrow increase / decrease and a deposit increase / decrease is calculated from the deposit conditions (before and after the change), the lending conditions (before and after the change), the external situation, and the customer situation.

  As shown in FIG. 8, the control unit 21 of the support server 20 executes a machine learning process of the customer behavior (Step S1-1). Specifically, the support management unit 211 of the control unit 21 specifies, as the evaluation time, the date when the deposit interest rate or the lending interest rate is changed from the condition information storage unit 45 (interest change time). Then, the support management unit 211 acquires, from the condition information storage unit 45, the deposit conditions and the lending conditions before and after the interest rate change time. Next, the support management unit 211 acquires the external situation information at the evaluation time from the external situation storage unit 46. Here, economic index information 461, market information 462, and interest rate information 463 are acquired as external situations. Further, the support management unit 211 acquires the customer attribute (segment) recorded in the customer management record 420 at the evaluation time from the customer information storage unit 42, and calculates the distribution of the customer attribute as the customer situation. Next, the support management unit 211 acquires the total amount of account balance recorded in the account information storage unit 43 and the total amount of loan balance recorded in the transaction information storage unit 44 before and after the evaluation time. Then, the support management unit 211 calculates an increase / decrease (money amount) of the deposit by subtracting the total amount of the account balance before the evaluation period from the total amount of the account balance after the evaluation period. Further, the support management unit 211 calculates the loan increase / decrease (money amount) by subtracting the total loan balance before the evaluation period from the total loan balance after the evaluation period.

  Then, the support management unit 211 creates a data set (teacher data) including, for each evaluation time, deposit conditions (before and after the evaluation time), lending conditions (before and after the evaluation time), external status, customer status, borrowing increase / decrease, and deposit increase / decrease. I do. Next, the learning unit 212 calculates a customer behavior prediction model by performing machine learning using the deposit condition and the customer status as input layers, and using the increase / decrease in borrowing and the increase / decrease in deposits as an output layer.

  Next, the control unit 21 of the support server 20 executes a process of recording a learning result (Step S1-2). Specifically, the learning unit 212 of the control unit 21 records the generated customer behavior prediction model in the learning result storage unit 22.

  Next, the control unit 21 of the support server 20 executes a setting process of the current situation (Step S1-3). Specifically, the support management unit 211 of the control unit 21 acquires the current external situation from the external situation storage unit 46. Further, the support management unit 211 acquires the customer attribute (segment) recorded in the current customer management record 420 from the customer information storage unit 42, and calculates the distribution of the customer attribute as the customer situation. Then, the prediction unit 213 sets the current external situation and the customer situation in the input layer of the customer behavior prediction model.

  Next, the control unit 21 of the support server 20 executes a condition provisional setting process (step S1-4). Specifically, the support management unit 211 of the control unit 21 acquires new deposit condition candidates and loan condition candidates from the condition candidate list with respect to the current deposit conditions and loan conditions. Then, the prediction unit 213 sets the current deposit conditions, lending conditions, and the acquired candidate deposit conditions and lending conditions, in the input layer of the customer behavior prediction model.

  Next, the control unit 21 of the support server 20 executes an increase / decrease prediction process (step S1-5). Specifically, the prediction unit 213 of the control unit 21 calculates an increase / decrease in deposits and an increase / decrease in borrowings of the output layer using a customer behavior prediction model.

  Next, the control unit 21 of the support server 20 performs an evaluation process of the cash and customer deposits (Step S1-6). Specifically, the support management unit 211 of the control unit 21 updates the current balance of the customer deposit and the current deposit based on the change in the deposit and the change in the borrowing calculated by the prediction unit 213. Then, the analysis unit 214 calculates the ratio of the cash deposit to the customer deposit and compares it with the reference ratio.

  Here, when the cash deposit ratio is equal to or more than the reference ratio, the control unit 21 of the support server 20 executes an investment simulation process (step S1-7). Specifically, the analysis unit 214 of the control unit 21 acquires securities information that can be purchased from an external site, and calculates funds when an investment is made. In this case, the investment amount is calculated so as to satisfy various regulations (priorities and restrictions) recorded in the regulation information storage unit 47.

  On the other hand, when the cash deposit ratio is less than the reference ratio, the control unit 21 of the support server 20 executes a procurement simulation process (step S1-8). Specifically, the analysis unit 214 of the control unit 21 acquires available procurement information from an external site, and calculates funds when procurement is performed. In this case, a procurement simulation is performed so as to satisfy various regulations (priorities and restrictions) recorded in the regulation information storage unit 47.

  Next, the control unit 21 of the support server 20 executes a balance sheet evaluation process (Step S1-9). Specifically, the support management unit 211 of the control unit 21 applies the simulation result to the balance sheet. Further, the support management unit 211 sets various information recorded in the management information storage unit 48 on the balance sheet. Next, the support management unit 211 applies the simulation results of the investment simulation processing (step S1-7) and the procurement simulation processing (step S1-8) to the balance sheet. Then, the support management unit 211 temporarily stores the calculated balance sheet in association with the deposit condition and the loan condition.

  Next, the control unit 21 of the support server 20 executes a determination process as to whether or not to end (Step S1-10). Specifically, the support management unit 211 of the control unit 21 determines whether all the deposit condition candidates and the loan condition candidates recorded in the condition candidate list have been used in the provisional condition setting process (step S1-4). Determines that the evaluation process has ended. On the other hand, if there are remaining deposit condition candidates and loan condition candidates that have not been used yet in the condition candidate list, it is determined that the process has not been completed.

When it is determined that the processing is not to be terminated (in the case of “NO” in step S1-10), the control unit 21 of the support server 20 repeats the processing after the provisional setting processing of the condition (step S1-4).
When it is determined that the processing is to be ended (in the case of “YES” in step S1-10), the control unit 21 of the support server 20 performs an output process (step S1-11). Specifically, the support management unit 211 of the control unit 21 specifies a balance sheet that satisfies the priority and the constraint in the temporarily stored balance sheet, and outputs the deposit condition and the loan condition in this case to the user terminal 10. . Then, using the output balance sheet, a financial transaction condition that has a target value (maximum profit) is specified.

As described above, according to the present embodiment, the following effects can be obtained.
(1-1) In the present embodiment, the control unit 21 of the support server 20 executes a machine learning process of customer behavior (Step S1-1). For each evaluation period, a data set (teacher data) including deposit conditions, lending conditions, external conditions, customer conditions, increase / decrease in borrowings, and increase / decrease in deposits is created. Next, the learning unit 212 calculates a customer behavior prediction model by performing machine learning using the deposit condition to customer status as an input layer, borrowing increase / decrease, and deposit increase / decrease as an output layer. Thus, by using the deposit conditions and the lending conditions for the input layer, it is possible to predict an increase or decrease in borrowing and an increase in deposits when the financial transaction conditions are changed.

By using the external situation as the input layer, it is possible to predict the increase / decrease in borrowings and deposits in consideration of the external situation.
By using the customer status as the input layer, it is possible to predict a change in borrowing and a change in deposit in consideration of the customer status.

  (1-2) In the present embodiment, the control unit 21 of the support server 20 executes a provisional setting process of the condition (step S1-4) and a prediction process of increase / decrease (step S1-5). This makes it possible to predict an increase or decrease in deposits and loans using the financial transaction condition candidates.

  (1-3) In the present embodiment, the control unit 21 of the support server 20 executes a cash and customer deposit evaluation process (step S1-6). When the ratio of cash and deposit is high, the control unit 21 of the support server 20 executes an investment simulation process (Step S1-7). On the other hand, when the ratio of the cash and deposit is low, the control unit 21 of the support server 20 executes a procurement simulation process (step S1-8). This makes it possible to optimize the cash balance.

  In this case, a simulation is performed so as to satisfy various regulations (priorities and restrictions) recorded in the regulation information storage unit 47. As a result, a simulation can be performed in consideration of management priority and bank regulations.

  (1-4) In the present embodiment, the control unit 21 of the support server 20 executes a balance sheet evaluation process (Step S1-9) and an output process (Step S1-11). Thereby, it is possible to evaluate the influence on the balance sheet when the financial transaction conditions are changed.

(Second embodiment)
Next, a second embodiment that embodies a banking business support system, a banking business support method, and a banking business support program will be described with reference to FIGS. 9 and 10. Note that the second embodiment has a configuration in which only the method of predicting customer behavior in the first embodiment is changed, and a detailed description of similar portions will be omitted. In the first embodiment, a customer behavior prediction model is calculated by performing machine learning using the deposit condition to customer status as an input layer, the borrowing increase / decrease, and the deposit increase / decrease as an output layer.

  In the second embodiment, as shown in FIG. 9, in terms of a deposit condition (before change), a lending condition (before change), an external situation, and a customer situation, a target increase / decrease in borrowing and a deposit condition for obtaining a deposit increase / decrease ( After the change), a customer behavior prediction model for predicting the lending condition (after the change) is calculated.

  As shown in FIG. 10, the control unit 21 of the support server 20 executes a machine learning process of customer behavior (Step S2-1). Specifically, as in step S1-1, the support management unit 211 of the control unit 21 determines from the back-end system 40, based on information acquired at each evaluation time, deposit conditions, loan conditions, external conditions, Create a data set (teacher data) consisting of status, borrowing change, and deposit change. Next, the learning unit 212 uses the deposit conditions (before change), the loan conditions (before change), the borrowing increase / decrease, the deposit increase / decrease, the external situation, and the customer situation as input layers, and deposit conditions (after change), loan conditions (after change). ) Is used as an output layer to perform machine learning to calculate a customer behavior prediction model.

First, the control unit 21 of the support server 20 executes a process of recording a learning result as in step S1-2 (step S2-2).
Next, the control unit 21 of the support server 20 executes a balance sheet evaluation process (Step S2-3). Specifically, the analysis unit 214 of the control unit 21 acquires various information from the management information storage unit 48 and inputs the information to the balance sheet. Next, the analysis unit 214 adjusts each item of the balance sheet with a predetermined fluctuation range, and creates a target balance sheet (target balance sheet). The target balance sheet is, for example, a balance sheet that satisfies the priority items and restrictions recorded in the regulation information storage unit 47 and maximizes the profit.

  Next, the control unit 21 of the support server 20 executes a target amount setting process for cash and deposits from the balance sheet (step S2-4). Specifically, the analysis unit 214 of the control unit 21 calculates the increase / decrease of the deposit from the difference between the customer deposit on the target balance sheet and the customer deposit on the current balance sheet. Further, the analysis unit 214 calculates the increase or decrease in borrowing from the difference between the loan on the target balance sheet and the loan on the current balance sheet.

Next, the control unit 21 of the support server 20 executes a current situation setting process as in step S1-3 (step S2-5).
Next, the control unit 21 of the support server 20 executes a financial transaction condition prediction process (Step S2-6). Specifically, the prediction unit 213 of the control unit 21 inputs the change in deposit and the change in borrowing calculated in step S2-4 to the customer behavior prediction model. Then, the prediction unit 213 calculates a deposit condition and a loan condition using the customer behavior prediction model.

  Next, the control unit 21 of the support server 20 executes an output process (Step S2-7). Specifically, the support management unit 211 of the control unit 21 outputs the calculated deposit conditions (after the change) and the loan conditions (after the change) to the user terminal 10.

As described above, according to the present embodiment, the following effects can be obtained.
(2-1) In the present embodiment, the control unit 21 of the support server 20 executes a machine learning process of customer behavior (Step S2-1). Here, the learning unit 212 uses the deposit conditions (before change), the loan conditions (before change), the borrowing increase / decrease, the deposit increase / decrease, the external situation, and the customer situation as input layers, and deposit condition (after change), loan condition (after change). ) Is used as an output layer to perform machine learning to calculate a customer behavior prediction model. Thereby, financial transaction conditions for realizing the target balance sheet can be predicted.

This embodiment can be implemented with the following modifications. The present embodiment and the following modified examples can be implemented in combination with each other within a technically consistent range.
In the first embodiment, an optimized balance sheet is searched by changing deposit conditions and lending conditions. Alternatively, bank behavior for calculating a better balance sheet may be predicted by machine learning. In this case, the bank behavior is linked with the bank behavior pattern and the balance sheet evaluation result using teacher data including the external situation information of the bank, cash and deposit, bank behavior pattern (financing or investment), and balance sheet evaluation result. Generate a prediction model. Further, a bank behavior prediction model may be generated by using regulation information of financial transactions in an input layer.

  In each of the above embodiments, financial transactions (deposits and borrowings) as customer behavior are predicted using financial transaction conditions. Financial transactions are not limited to deposits and borrowings, but may be combined with one or the other. For example, customer behavior may be predicted using products and services provided by a bank. In this case, a service information storage unit and a product information storage unit are provided in the back-end system 40, and history information on provision conditions and provision status of services and products is recorded. Then, machine learning is performed by using the provision conditions and provision status information of services and products as teacher data as a data set, and a customer behavior model is calculated.

  In each of the above embodiments, teacher data is created using information of the own bank. In addition to this, machine learning may be performed including information of other lines. Here, information that cannot be obtained from another bank is complemented using information of the own bank. For example, as for the customer information, the position information of the other bank branch is acquired, and is substituted by the customer information of the own bank branch located near the other bank branch.

In the above embodiments, the external situation information is recorded in the back-end system 40. The external situation information may be obtained from an external system such as an information providing site.
In the above embodiments, the user terminal 10, the support server 20, and the back-end system 40 are used. The hardware configuration is not limited to this. For example, these may be integrally configured.

  The control unit 21 is not limited to performing software processing for all processing executed by itself. For example, the control unit 21 may include a dedicated hardware circuit (for example, an application specific integrated circuit (ASIC)) that performs hardware processing for at least a part of the processing executed by the control unit 21. That is, the control unit 21 includes [1] one or more processors that operate according to a computer program (software), [2] one or more dedicated hardware circuits that execute at least a part of various processes, or [3] It can be configured as a circuit including a combination thereof. The processor includes a CPU and a memory such as a RAM and a ROM, and the memory stores program codes or instructions configured to cause the CPU to execute processing. Memory or computer readable media includes any available media that can be accessed by a general purpose or special purpose computer.

Next, technical ideas that can be grasped from the above embodiment and other examples will be additionally described below.
[A] an information storage system for storing, at each time, financial transaction conditions of financial transactions, customer status information, history information including the status of financial transactions with customers, and external status information;
A banking business support system comprising a user terminal and a control unit connected to the information storage system,
The control unit includes:
From the information storage system, external situation information at the same time, financial transaction conditions, customer situation information, financial transaction status,
Perform machine learning using teacher data as a data set the acquired information, generate a customer behavior prediction model to associate the financial transaction conditions and financial transaction status,
A banking support system for calculating a financial transaction condition corresponding to a target value of a transaction status using the customer behavior prediction model for a current external situation and a customer situation.
[B] The banking support system according to [a], wherein an interest rate is used as the financial transaction condition.
[C] the financial transaction includes at least one of deposit and lending,
The banking business support system according to [a] or [b], wherein the management status of the bank is predicted based on the deposit status and the lending status as the financial transaction status.
[D] further comprising a management information storage unit for storing the asset status of the bank;
The control unit includes:
Obtain the amount of funding and investment calculated based on the asset status stored in the management information storage unit,
The banking support system according to any one of [a] to [c], wherein a balance sheet is calculated based on the amounts of the financing and the investment.
[E] the control unit includes:
Obtain a target balance sheet, identify the amount of funding or investment in the target balance sheet,
The banking support system according to any one of [a] to [d], wherein a target value of the financial transaction status is calculated based on the amount of money.
[F] The control unit generates a bank behavior prediction model that associates the bank behavior pattern with the balance sheet using teacher data including external situation information of the bank, cash and deposit, bank behavior pattern, and evaluation results of the balance sheet. The banking support system according to claim [e], wherein:
[G] the information storage system further stores regulatory information on financial transactions;
The banking support according to any one of [a] to [f], wherein the control unit predicts banking behavior using regulation information on financial transactions stored in the information storage system. system.
[H] an information storage system that stores financial transaction conditions of financial transactions, customer status information, history information including the status of financial transactions with customers, and external status information for each time period,
A method for supporting banking business using a banking business support system including a user terminal and a control unit connected to the information storage system,
The control unit includes:
From the information storage system, external situation information at the same time, financial transaction conditions, customer situation information, financial transaction status,
Perform machine learning using teacher data as a data set the acquired information, generate a customer behavior prediction model to associate the financial transaction conditions and financial transaction status,
A banking business support method for calculating a financial transaction condition corresponding to a target value of a transaction status using the customer behavior prediction model with respect to a current external situation and a customer situation.
[I] an information storage system that stores, at each time, financial transaction conditions of financial transactions, customer status information, history information including the status of financial transactions with customers, and external status information;
A program for supporting banking business using a banking business support system including a user terminal and a control unit connected to the information storage system,
The control unit,
From the information storage system, external situation information at the same time, financial transaction conditions, customer situation information, financial transaction status,
Perform machine learning using teacher data as a data set the acquired information, generate a customer behavior prediction model to associate the financial transaction conditions and financial transaction status,
A banking business support program that functions as means for calculating a financial transaction condition corresponding to a target value of a transaction situation using the customer behavior prediction model with respect to a current external situation and a customer situation.

  Reference Signs List 10 user terminal, 20 support server, 21 control unit, 22 learning result storage unit, 211 support management unit, 212 learning unit, 213 prediction unit, 214 analysis unit, 40 backend system, 42 ... customer information storage unit, 43 ... account information storage unit, 44 ... transaction information storage unit, 45 ... condition information storage unit, 46 ... external situation storage unit, 47 ... regulation information storage unit, 48 ... management information storage unit.

Claims (8)

For each time period, an information storage system that stores financial transaction conditions of financial transactions, customer status information, history information including the status of financial transactions with customers, and external status information,
A banking business support system comprising a user terminal and a control unit connected to the information storage system,
The control unit includes:
From the information storage system, external situation information at the same time, financial transaction conditions, customer situation information, financial transaction status,
Perform machine learning using teacher data as a data set the acquired information, generate a customer behavior prediction model to associate the financial transaction conditions and financial transaction status,
A banking support system for calculating a financial transaction condition corresponding to a target value of a transaction status using the customer behavior prediction model for a current external situation and a customer situation.   The banking support system according to claim 1, wherein an interest rate is used as the financial transaction condition. It further comprises a management information storage unit for storing the bank's asset status,
The control unit includes:
Obtain the amount of funding and investment calculated based on the asset status stored in the management information storage unit,
3. The banking support system according to claim 1, wherein a balance sheet is calculated based on the amounts of the fund procurement and the investment. The control unit includes:
Obtain a target balance sheet including a customer deposit and a loan and a current balance sheet, compare the target balance sheet with the current balance sheet, and calculate a funding amount from a difference between the customer deposits. Together with calculating the amount of investment from the difference between the loans,
The banking support according to any one of claims 1 to 3, wherein the financial transaction conditions are calculated for the difference between the customer deposits and the difference between the loans using the customer behavior prediction model. system.   The controller generates a bank behavior prediction model that associates the bank behavior pattern with the balance sheet evaluation result using teacher data including external situation information of the bank, cash and deposit, bank behavior pattern, and balance sheet evaluation result. 5. The banking support system according to claim 4, wherein: The information storage system further stores regulatory information on financial transactions,
The banking support system according to any one of claims 1 to 5, wherein the control unit predicts a banking behavior using regulation information on financial transactions stored in the information storage system. For each time period, an information storage system that stores financial transaction conditions of financial transactions, customer status information, history information including the status of financial transactions with customers, and external status information,
A method for supporting banking business using a banking business support system including a user terminal and a control unit connected to the information storage system,
The control unit includes:
From the information storage system, external situation information at the same time, financial transaction conditions, customer situation information, financial transaction status,
Perform machine learning using teacher data as a data set the acquired information, generate a customer behavior prediction model to associate the financial transaction conditions and financial transaction status,
A banking business support method for calculating a financial transaction condition corresponding to a target value of a transaction status using the customer behavior prediction model with respect to a current external situation and a customer situation. For each time period, an information storage system that stores financial transaction conditions of financial transactions, customer status information, history information including the status of financial transactions with customers, and external status information,
A program for supporting banking business using a banking business support system including a user terminal and a control unit connected to the information storage system,
The control unit,
From the information storage system, external situation information at the same time, financial transaction conditions, customer situation information, financial transaction status,
Perform machine learning using teacher data as a data set the acquired information, generate a customer behavior prediction model to associate the financial transaction conditions and financial transaction status,
A banking business support program that functions as means for calculating a financial transaction condition corresponding to a target value of a transaction situation using the customer behavior prediction model with respect to a current external situation and a customer situation.

Images