Data processing device, data processing method, and program

Abstract

This invention provides a data processing device, method, and program that contribute to predicting the timing of fundraising. [Solution] A data processing device that acquires business data and accounting data of multiple businesses from an accounting data management device via a network 50 includes a storage unit that stores a learning model in which future monthly cash balances are the target variable, with at least the cash balances for multiple months prior to a base month and the amount corresponding to one or more accounts in the month prior to the base month as explanatory variables; a prediction unit that inputs the cash balances for multiple months prior to a base month of the target business and the amount corresponding to one or more accounts in the month prior to the base month into the learning model and acquires the future monthly cash balances output from the learning model as the predicted cash balances for multiple months in the future of the target business; and an output control unit that controls the output of the predicted cash balances for multiple months in the future of the target business.

Description

Technical Field

[0001] The present invention relates to a data processing device, a data processing method, and a program.

Background Art

[0002] An accounting management system for assisting an operator's financial accounting work has been known (see, for example, Patent Document 1). [Prior Art Document] [Patent Document] [Patent Document 1] Japanese Patent Application Laid-Open No. 2001-014384

Summary of the Invention

[0003] According to an embodiment of the present invention, a data processing device is provided. The data processing device may include a storage unit that stores a learning model having at least explanatory variables including cash balances for a plurality of months prior to a reference month and amounts corresponding to one or more accounting items in the month preceding the reference month, and a target variable being cash balances for a plurality of future months. The data processing device may include a prediction unit that inputs cash balances for a plurality of months prior to a reference month of a target operator and amounts corresponding to one or more accounting items in the month preceding the reference month into the learning model, and acquires, as predicted cash balances for a plurality of future months of the target operator, the cash balances for a plurality of future months output from the learning model. The data processing device may include an output control unit that controls to output the predicted cash balances for a plurality of future months of the target operator acquired by the prediction unit.

[0004] Note that the above summary of the invention does not list all the necessary features of the present invention. Also, sub-combinations of these feature groups can also be inventions.

Brief Description of the Drawings

[0005] [Figure 1] An example of the data processing device 100 is schematically shown. [Figure 2]This is an explanatory diagram illustrating the process of creating a data mart 390 by the data processing device 100. [Figure 3] This is an explanatory diagram illustrating the process of creating a data mart 390 by the data processing device 100. [Figure 4] This is an explanatory diagram illustrating the process of creating a data mart 390 by the data processing device 100. [Figure 5] A schematic example of a cash balance graph 400 is shown below. [Figure 6] Another example of the cash balance graph 400 is shown in general terms. [Figure 7] An example of the functional configuration of the data processing device 100 is shown in a schematic manner. [Figure 8] An example of the processing flow by the data processing device 100 is shown in a schematic manner. [Figure 9] An example of the processing flow by the data processing device 100 is shown in a schematic manner. [Figure 10] A schematic example of the hardware configuration of a computer 1200 that functions as a data processing device 100 is shown. [Modes for carrying out the invention]

[0006] The present invention will be described below through embodiments of the invention, but these embodiments are not intended to limit the invention as defined in the claims. Furthermore, not all combinations of features described in the embodiments are necessarily essential to the solution of the invention.

[0007] Figure 1 schematically shows an example of the data processing device 100. The data processing device 100 according to this embodiment provides technology that contributes to predicting the timing of fundraising. The data processing device 100 has a function that predicts the future cash balance of a target business from the cash balance of multiple target businesses over multiple months, using accounting data of multiple businesses. The data processing device may use AI (Artificial Intelligence) or ML (Machine Learning).

[0008] The data processing device 100 stores business data and accounting data of multiple businesses. For example, the data processing device 100 acquires and stores business data and accounting data of multiple businesses from an accounting data management device 200 that manages business data and accounting data of multiple businesses.

[0009] The data processing device 100 may acquire business data and accounting data of multiple businesses from the accounting data management device 200 via the network 50. The network 50 may include a cloud network. The network 50 may include the internet. The network 50 may include a mobile communication network. The network 50 may include a dedicated network such as a LAN (Local Area Network). The data processing device 100 and the accounting data management device 200 may communicate directly without going through the network 50.

[0010] The accounting data management device 200 may manage accounting data entered by businesses and accounting firms using so-called accounting software, so-called cloud-based accounting software, and related software. The accounting data management device 200 may also manage accounting data provided by businesses and accounting firms without using accounting software, cloud-based accounting software, or related software.

[0011] The accounting data management device 200 may receive and manage the business data and accounting data of a business operator from a communication terminal 80 used by user 82. User 82 may be a business operator. User 82 may be an accounting firm or the like used by a business operator. The accounting data management device 200 may communicate with the communication terminal 80 via the network 50. The communication terminal 80 may be any terminal capable of communication. For example, the communication terminal 80 may be a mobile phone such as a smartphone, a tablet terminal, or a PC (Personal Computer).

[0012] The data processing device 100 and the accounting data management device 200 may be integrated into a single unit. In other words, the data processing device 100 may be configured to include the functions of the accounting data management device 200.

[0013] Business data may include data indicating the attributes of the business. Business data may include the name of the business. Business data may include closing date information indicating the closing date. Business data may include the business description. Business data may include the number of employees. Business data may include the date of establishment. Business data may include the period elapsed since establishment. Business data may include the region of location. Business data may include an industry code indicating the type of business the company is in. The industry code may be a code established by the country. For example, the industry code may be a Japan Standard Industrial Classification code. The industry code may also be a securities code established by the Japan Exchange Group. The industry code is not limited to these and may be any other code that indicates the type of business the company is in. Business data may include company type indicating whether the business is a stock company, a limited liability company, etc. Business data may include data indicating whether the company is a listed company or a private company.

[0014] Accounting data may be data relating to the accounting of a business. Accounting data may include so-called journal entry data. In other words, accounting data may include transaction data that is classified and recorded by account.

[0015] Accounting data may include capital stock. Accounting data may include capital reserves. Accounting data may include advance payments. Accounting data may include time deposits. Accounting data may include accounts receivable. Accounting data may include long-term borrowings. Accounting data may include interest income. Accounting data may include sales. Accounting data may include revenue. Accounting data may include ending inventory. Accounting data may include gains on sale of securities. Accounting data may include accounts receivable. Accounting data may include miscellaneous income. Accounting data may include beginning balances. Accounting data may include accounts payable. Accounting data may include cost of goods sold. Accounting data may include accounts receivable. Accounting data may include operating profit. Accounting data may include total assets. Accounting data may include cash and cash equivalents. Accounting data may include trade receivable. Accounting data may include securities. Accounting data may include inventory. Accounting data may include accounts payable. Accounting data may include tangible fixed assets. Accounting data may include intangible fixed assets. Accounting data may include interest-bearing debt. Accounting data may include equity. Accounting data may include trade payables. Accounting data may include depreciation expenses. Accounting data may include goodwill amortization expenses. Accounting data may include selling, general and administrative expenses. Accounting data may include the effective tax rate. Accounting data may include changes in working capital. Accounting data may include capital expenditures. Accounting data may include debt financing costs. Accounting data may include the equity ratio. Accounting data may include the average balance of accounts receivable. Accounting data may include the average balance of inventory. Accounting data may include the average balance of accounts payable. Accounting data may include retirement benefit obligations. Accounting data may include minority interests. Accounting data may include dividend amounts. Accounting data may include annual profit amounts. Accounting data may include net assets. Accounting data may include net income for the period. Accounting data may include the number of outstanding shares. Accounting data may include total assets. Accounting data may include accounts receivable turnover days. Accounting data may include inventory turnover days. Accounting data may omit some of these. Accounting data may include other data.

[0016] The data processing device 100 may generate training data using business data and accounting data from multiple businesses. The data processing device 100 may create a data mart that tracks the monthly cash balance for each of the multiple businesses. Generally, accounting data does not include monthly cash balances. The data processing device 100 may create a data mart that tracks the monthly cash balance by processing the business data and accounting data of the businesses.

[0017] Figures 2, 3, and 4 are explanatory diagrams illustrating an example of how the data processing device 100 can create a data mart. The data processing device 100 may create a data mart using the business operator's accounting data and business operator data. The accounting data and business operator data can be managed in various data structures. In particular, the data structure of the accounting data and business operator data may differ depending on the managing entity. The data processing device 100 may create a data mart in a manner that corresponds to the data structure of the accounting data and business operator data in question.

[0018] Here, we will explain, as an example, how to create a data mart using a transaction details table 330, which contains information for one transaction per record; an accounting settings table 340, which contains business data; and an account balance table 350, which contains the opening balance for each account. In this example, each record in the transaction details table 330 contains information such as the business ID, account ID, and transaction amount.

[0019] The account item ID is a sequence of mechanically generated letters and numbers and has no logical structure on its own. The data processing device 100 uses the account item table 310 and the subsidiary item table 320 to convert the account item ID into account item information with a logical structure. The account item table 310 and the subsidiary item table 320 may be tables showing the correspondence between the account item ID and the account item information. The account item information may have a hierarchical logical structure for each digit. As a specific example, the data processing device 100 converts the account item ID: XXXX-XXXX-0001 into 6-digit account item information: 1-1-0-0-0-0. The first layer of the account item information may represent the major items of the account item, and the subsequent layers may represent the subdivided items. The formats of the account item ID and the account item information are not limited to these and may be other formats. The data processing device 100 combines the account item information with the transaction detail table 330. Note that the account item information may originally be combined with the transaction detail table 330. In this case, the data processing device 100 may not perform the conversion using the account item table 310 and the subsidiary item table 320.

[0020] The data processing device 100 identifies the beginning-of-period year and month and the beginning-of-period balance from the closing date information described in the accounting setting table 340 and the balance in the account balance table 350.

[0021] The data processing device 100 calculates the monthly income and expenditure for each account item for each of a plurality of merchants by grouping and aggregating the records in the transaction detail table 330 by merchant, account item, and month. Also, the data processing device 100 calculates the overall monthly income and expenditure for each of the plurality of merchants. At this time, the data processing device 100 may refer to the credit and debit flags and make the transaction amount negative for credit items. Thereby, offsetting can be expressed.

[0022] The data processing device 100 calculates the monthly cash balance 360 ​​by starting with the specified beginning month and beginning balance and adding the calculated total monthly income and expenses. Figure 3 shows a specific example. In Figure 3, we explain using the example where the target business operator's beginning month is January 2020, the beginning balance 352 is 1,000,000 yen, the income and expenses for January 2020 332 is -100,000 yen, and the income and expenses for February 2020 334 is +300,000 yen. The data processing device 100 calculates the cash balance for February 2020 (900,000 yen) by adding the January income and expenses 332 to the target business operator's beginning balance 352. Furthermore, the data processing device 100 calculates the cash balance for March 2020 (1,200,000 yen) by adding the February income and expenses of 334 to the calculated cash balance for February 2020. By performing such calculations for each month, the data processing device 100 calculates the cash balances 360 for multiple months.

[0023] The data processing device 100 generates a data mart 390 by combining explanatory variables 370 and dependent variable 380 by combining the income and expenses for each account with the monthly cash balance 360. The data processing device 100 may use the cash balances for multiple months prior to the base month (the year and month in which the forecast is performed) as explanatory variables 370, and the cash balances for multiple future months of the base month as dependent variable 380. For example, the data processing device 100 calculates cash balances for 15 months, using the last 3 months of the time series as dependent variable 380 and the previous 12 months of the time series as explanatory variables 370. The data processing device 100 adds the amounts corresponding to one or more accounts for one or more months prior to the base month to the explanatory variables 370. The amounts corresponding to monthly accounts are the income and expenses of the monthly accounts. The data processing device 100, for example, groups and aggregates records from the transaction details table 330 by business operator, account, and month, and outputs the monthly income and expenses for each account for each of the multiple business operators as a feature table. The data processing device 100 may set the unique keys of the feature table to the business operator ID and the predicted year and month, and each record may have the income and expenses for each account as a column. The data processing device 100 may then generate a data mart 390 by joining the monthly cash balance 360 ​​with the feature table. The data mart 390 contains multiple monthly cash balances and the income and expenses for one or more monthly accounts for each of the multiple business operators.

[0024] Figure 4 schematically shows an example of explanatory variable 370 and dependent variable 380. In the example shown in Figure 4, explanatory variable 370 includes cash balances from one month to 12 months prior to the base month, and dependent variable 380 includes cash balances from one month to three months after the base month. For example, if the base month is April, explanatory variable 370 includes cash balances from April to December of the previous year and from January to March of the same year, and dependent variable 380 includes cash balances for April, May, and June.

[0025] The data processing device 100 may include cash statistics in the explanatory variable 370, as illustrated in Figure 4. Examples of cash statistics include the mean over the past n months and the median over the past n months, as shown in Figure 4, but are not limited to these; they may also include the variance over the past n months, the monthly differences over the past n months, etc. The data processing device 100 may include some of these in the explanatory variable 370, or may include other statistics. The data processing device 100 may include the income and expenses of each account for the previous month in the explanatory variable 370, as illustrated in Figure 4. Examples of accounts include sales and accounts receivable for the previous month, as shown in Figure 4, but are not limited to these; they may also include capital stock, capital reserves, advances, time deposits, accounts receivable, long-term borrowings, interest income, sales, ending inventory, gains on sale of securities, and miscellaneous income, etc. The data processing device 100 may include some of these in the explanatory variable 370, or may include income and expenses of other accounts.

[0026] The data processing device 100 can generate a learning model that takes the cash balances, cash statistics, and the income and expenses of each account for the previous month as inputs, and outputs the cash balances for the next three months, by performing machine learning using the explanatory variables 370 and the target variable 380 as illustrated in Figure 4. The data processing device 100 may use the generated learning model to predict the cash balances of the target business for multiple months in the future. The data processing device 100 may output the predicted cash balances of the target business for multiple months in the future. For example, the data processing device 100 may display and output a graph showing the predicted cash balances of the target business for multiple months in the future.

[0027] Figure 5 schematically shows an example of the cash balance graph 400. The vertical axis of the cash balance graph 400 shows the cash balance in millions of yen. The horizontal axis of the cash balance graph 400 shows the year and month. Here, we illustrate the cash balance graph 400 when the cash balance for April 2025, May 2025, and June 2025 is predicted based on data prior to March 2025, as of the end of March 2025.

[0028] The data processing device 100 may generate a cash balance graph 400 that shows the trend of past cash balances and the trend of predicted future cash balances in different ways, as illustrated in Figure 5. In Figure 5, the trend of past cash balances is shown with a solid line, and the trend of predicted future cash balances is shown with a dashed line.

[0029] The data processing device 100 outputs a cash balance graph 400 as illustrated in Figure 5, thereby reducing the effort required for businesses to plan their cash flow. It also allows businesses to understand when and by how much cash will be insufficient, enabling them to secure funding with ample time.

[0030] Figure 6 schematically shows another example of the cash balance graph 400. Here, we will mainly explain the differences from the cash balance graph 400 in Figure 5.

[0031] In the example shown in Figure 6, the data processing device 100 outputs a baseline trend 410, which is the projected balance trend for multiple months in the future; a high-balance trend 420, which shows that the cash balance for multiple months is greater than that of the baseline trend 410; and a low-balance trend, which shows that the cash balance for multiple months is less than that of the baseline trend.

[0032] The data processing device 100, for example, predicts the cash balances for multiple future months from past data to generate a baseline trend 410, generates a high-balance trend 420 by multiplying the predicted cash balances for multiple months by a predetermined value greater than 1, and generates a low-balance trend 430 by multiplying the predicted cash balances for multiple months by a predetermined value less than 1. Specifically, the data processing device 100 generates a high-balance trend 420 by multiplying the predicted cash balances for multiple months by 1.1, and generates a low-balance trend 430 by multiplying the predicted cash balances for multiple months by 0.9.

[0033] The data processing device 100 may generate a learning model for generating a baseline transition 410, a learning model for generating a high-balance transition 420, and a learning model for generating a low-balance transition 430. For example, the data processing device 100 may generate a learning model for generating a baseline transition 410 using past data as is, generate a learning model for generating a high-balance transition 420 using past data in which the cache balance of the target variable has been increased, and generate a learning model for generating a low-balance transition 430 using past data in which the cache balance of the target variable has been decreased.

[0034] The data processing device 100 outputs a cash balance graph 400 as illustrated in Figure 6, allowing users to understand the changes in cash balance under both a favorable and unfavorable scenario, thus enabling the implementation of a more multifaceted financing plan.

[0035] Figure 7 schematically shows an example of the functional configuration of the data processing device 100. The data processing device 100 comprises a data storage unit 102, a data acquisition unit 104, a specification unit 106, a data management unit 108, a learning model generation unit 110, a target data acquisition unit 112, a prediction unit 114, and an output control unit 116.

[0036] The data storage unit 102 stores various types of data. The data storage unit 102 also stores data acquired by the data acquisition unit 104.

[0037] The data acquisition unit 104 acquires various types of data. The data acquisition unit 104 acquires business data and accounting data of multiple businesses. The data acquisition unit 104 may acquire business data and accounting data of multiple businesses from the data management device 200.

[0038] The identification unit 106 identifies the business operator's monthly cash balance. The identification unit 106 may identify the business operator's monthly cash balance using the business operator's business data and accounting data. If the business operator's business data and accounting data include the business operator's monthly cash balance, the identification unit 106 identifies the business operator's monthly cash balance by referring to the business operator's data and accounting data. If the business operator's business data and accounting data do not include the business operator's monthly cash balance, the identification unit 106 identifies the business operator's monthly cash balance using the data included in the business operator's data and accounting data.

[0039] The identification unit 106 may identify the cash balance for each month since the beginning of the fiscal year based on the business operator's past opening balances. For example, the identification unit 106 identifies the beginning year and month and opening balance from business operator data and accounting data. For example, the identification unit 106 identifies the beginning year and month from the closing date information included in the business operator data, identifies the opening balance for each account by referring to the account balance table included in the accounting data, and identifies the opening balance by adding these together. The identification unit 106 also calculates the monthly income and expenditure for each account by referring to the accounting data and aggregating the transaction amounts for each account on a monthly basis, and calculates the overall monthly income and expenditure by aggregating the income and expenditure on a monthly basis. The identification unit 106 may also determine whether an transaction is a debit or credit and execute a process to make the transaction amount negative for credit transactions.

[0040] The identification unit 106 calculates the monthly cash balance by starting with the identified beginning year and month and beginning balance, and adding the calculated total monthly income and expenses. This allows the data storage unit 102 to calculate the monthly cash balance for the period in which accounting data is included. For example, if the data storage unit 102 contains 10 years of accounting data, it can calculate the monthly cash balance for 10 years. The identification unit 106 identifies the monthly cash balance for each of the multiple businesses identified by the identification unit 106. The data storage unit 102 stores the monthly cash balance for each of the multiple businesses identified by the identification unit 106.

[0041] The data management unit 108 manages the data stored in the data storage unit 102. The data management unit 108 may generate training data using the data stored in the data storage unit 102. For each of the multiple businesses, the data management unit 108 may generate a dataset that includes the cash balances for multiple months identified by the identification unit 106 and the amounts corresponding to one or more accounts for each of those multiple months. The data storage unit 102 stores the multiple datasets generated by the data management unit 108.

[0042] The data management unit 108 may preprocess the data stored in the data storage unit 102 to generate training data. For example, the data management unit 108 may target only data where the cache stock for any of several months is less than a predetermined threshold. That is, the data management unit 108 excludes data stored in the data storage unit 102 where the cache stock for any of several months is greater than a threshold. For example, a threshold of 100 million yen may be set. Also, for example, the data management unit 108 may remove data stored in the data storage unit 102 that has not changed for the past 12 months. Also, for example, the data management unit 108 may exclude data stored in the data storage unit 102 that falls at the beginning of the month, where the settlement date is not the end of the month.

[0043] Data Management Department 108 may include capital in the dataset. Data Management Department 108 may include advances in the dataset. Data Management Department 108 may include accounts receivable in the dataset. Data Management Department 108 may include long-term borrowings in the dataset. Data Management Department 108 may include interest income in the dataset. Data Management Department 108 may include sales revenue in the dataset. Data Management Department 108 may include ending inventory in the dataset. Data Management Department 108 may include gains on sale of securities in the dataset. Data Management Department 108 may include accounts receivable in the dataset. Data Management Department 108 may include miscellaneous income in the dataset. Among the many existing accounts, these accounts have a significant impact on the cash balance, so including at least one of them in the dataset can contribute to generating a learning model with high accuracy in predicting future cash balances. Data Management Department 108 may also include amounts corresponding to other accounts in the dataset.

[0044] The learning model generation unit 110 uses multiple datasets stored in the data storage unit 102 to generate a learning model in which the cash balances for multiple months prior to the base month and the amounts corresponding to one or more accounts in the month preceding the base month are at least explanatory variables, and the cash balances for multiple future months are the dependent variable. The data storage unit 102 stores the learning model generated by the learning model generation unit 110.

[0045] The dataset contains actual monthly cash balances from the past, showing how cash balances actually changed after a certain trend occurred over a given period. Therefore, by generating a learning model using multiple datasets from multiple businesses, it is possible to generate a learning model that can predict future cash fluctuations based on past cash fluctuation trends over a given period.

[0046] The learning model generation unit 110 may generate a learning model that further uses statistical information of past monthly cash balances as explanatory variables. For example, the learning model generation unit 110 generates a learning model that uses cash balances for several months prior to the base month, statistical information of cash balances for several months, and amounts corresponding to one or more accounts in the month prior to the base month as explanatory variables, and future monthly cash balances as the dependent variable.

[0047] The learning model generation unit 110 may use statistical information on the cash balances of multiple months, including the cash balance for each month. For example, when the learning model generation unit 110 uses monthly cash balances for the past year of the base month, it may use statistical information on the cash balances of the base month for the past year.

[0048] The learning model generation unit 110 may use statistical information on the cash balance of multiple months, including the cash balance for one month, and statistical information on the cash balance of multiple months that are different from the cash balance for another month. For example, when the learning model generation unit 110 uses the monthly cash balance for the past year of the base month, it may use statistical information on the cash balance for the past three months of the base month.

[0049] The learning model generation unit 110 may use at least one of the mean, median, variance, and difference of the cash balances for multiple months as statistical information for the cash balances of multiple months. Verification by the inventor using actual data demonstrated a high correlation between the statistical information of the cash balances for multiple past months of a base month and the cash balances for multiple future months of the base month. The inventor created a learning model using actual data, verified the input and output of the learning model, verified the SHAP (SHapley Additive exPlanations) values ​​at that time, and, based on experience in accounting data management, identified the mean, median, variance, and difference as statistical information with a high correlation to future cash balances. In this way, by adding statistical information of past cash balances with a high correlation to future cash balances as explanatory variables and learning, the accuracy of predicting future cash balances can be improved.

[0050] The dataset further includes actual monthly amounts for one or more accounts from the past. Verification by the inventor using actual data demonstrated a high correlation between the amounts for one or more accounts in the month prior to the base month and the cash balances for several future months of the base month. Based on creating a learning model using actual data, verifying the input and output of the learning model, verifying the SHAP (SHapley Additive exPlanations) values, and experience in accounting data management, the inventor identified the following accounts as having a particularly high correlation with future cash balances: capital stock, advances, accounts receivable, long-term borrowings, interest income, sales revenue, ending inventory, gains on sale of securities, accounts receivable, and miscellaneous income. The learning model generation unit 110 may generate a learning model using the cash balances for several months prior to the base month and at least one of the following as explanatory variables: capital stock, advances, accounts receivable, long-term borrowings, interest income, sales revenue, ending inventory, gains on sale of securities, accounts receivable, and miscellaneous income for the month prior to the base month. The learning model generation unit 110 may also add other account titles as explanatory variables. By adding account titles that have a high correlation with future cash balances as explanatory variables during training, the accuracy of predicting future cash balances can be improved.

[0051] The target data acquisition unit 112 acquires data of the target business operator whose future cash balance is to be predicted. For example, the target data acquisition unit 112 acquires cash balances for multiple months prior to the base month, and amounts corresponding to one or more accounts for the month preceding the base month. The target data acquisition unit 112 may acquire this data from the data stored in the data storage unit 102.

[0052] The prediction unit 114 predicts the future cash balance of the target business. The prediction unit 114 inputs the data acquired by the target data acquisition unit 112 into a learning model stored in the data storage unit 102, and obtains the future monthly cash balances output from the learning model as the projected cash balances for multiple months in the future of the target business.

[0053] The output control unit 116 outputs the projected monthly cash balances for the target business operator, acquired by the forecasting unit 114. The output control unit 116 displays the projected cash balances on, for example, the display of the data processing device 100. The output control unit 116 transmits the projected cash balances to, for example, the communication terminal 80.

[0054] The output control unit 116 may output the projected future cash balance of the target business as a numerical value. The output control unit 116 may output a cash balance graph, as illustrated in Figure 5, which shows the trend of the target business's cash balance over several past months and the trend of the target business's projected future cash balance.

[0055] The output control unit 116 may be controlled to output a baseline trend, which is the trend of the target business operator's projected cash balance for multiple months in the future, obtained by the forecasting unit 114; a high-balance trend, where the cash balance for multiple months is greater than the baseline trend; and a low-balance trend, where the cash balance for multiple months is less than the baseline trend. The output control unit 116 may output a cash balance graph, for example, as illustrated in Figure 6. For example, the output control unit 116 obtains the projected cash balance for multiple months in the future from the forecasting unit 114, generates a baseline trend from the projected cash balance for multiple months, generates a high-balance trend by multiplying the projected cash balance for multiple months by a predetermined value greater than 1, and generates a low-balance trend by multiplying the projected cash balance for multiple months by a predetermined value less than 1.

[0056] In addition, the learning model generation unit 110 may generate a learning model for generating a baseline trend, a learning model for generating a high-balance trend, and a learning model for generating a low-balance trend. For example, the learning model generation unit 110 may generate a learning model for generating a baseline trend using past data as is, generate a learning model for generating a high-balance trend using past data in which the cache balance of the target variable has been increased, and generate a learning model for generating a low-balance trend using past data in which the cache balance of the target variable has been decreased.

[0057] The data processing device 100 outputs a cash balance graph 400 as illustrated in Figure 6, allowing users to understand the changes in cash balance under both successful and unsuccessful circumstances, thus enabling the implementation of a more multifaceted financing plan.

[0058] The data processing device 100 may perform a forecast using the monthly cash balances of the target business operator for any period in the past. The data processing device 100 may use the cash balances of multiple consecutive months in the past of the target business operator, or it may use the cash balances of multiple non-consecutive months in the past of the target business operator. As a specific example, the target data acquisition unit 112 acquires the monthly cash balances of the target business operator for the past year for a base month, and the amounts corresponding to one or more accounts for the month preceding the base month, and the forecasting unit 114 inputs these into the learning model. There is a high possibility that yearly cyclical characteristics will appear in the trends of cash balances. Therefore, by performing a forecast using the monthly cash balances for the past year, it may be possible to improve the accuracy of future cash balance forecasts.

[0059] The data processing device 100 may predict the monthly cash balance of the target business for any future period. The data processing device 100 may predict the cash balance of the target business for several consecutive future months, or for several future non-consecutive months. As a specific example, the learning model generation unit 110 generates a learning model with the monthly cash balance for the next three months of a base month as the target variable. This allows the business to know, for example, the predicted cash balance one month, two months, and three months after the prediction is made, and enables them to raise funds with sufficient time if necessary.

[0060] It is not necessarily required that the data processing device 100 includes all of the following: data storage unit 102, data acquisition unit 104, identification unit 106, data management unit 108, learning model generation unit 110, target data acquisition unit 112, prediction unit 114, and output control unit 116. For example, the data processing device 100 may have the function to generate a learning model but not the function to perform predictions using the learning model. In this case, the data processing device 100 includes the data storage unit 102, data acquisition unit 104, identification unit 106, data management unit 108, and learning model generation unit 110. For example, the data processing device 100 may not have the function to generate a learning model but may have the function to perform predictions using the learning model. In this case, the data processing device 100 includes the data storage unit 102, target data acquisition unit 112, prediction unit 114, and output control unit 116. In this case, another device generates the learning model, and the data processing device 100 acquires and uses the learning model from that other device.

[0061] Figure 8 schematically shows an example of the processing flow by the data processing device 100. Here, we show an example of the flow when the output control unit 116 controls the output of a warning when the predicted monthly cash balances of the target business operator acquired by the prediction unit 114 are less than a predetermined threshold.

[0062] In step 102 (sometimes abbreviated as S), the target data acquisition unit 112 acquires data from the target business operator. In step 104, the prediction unit 114 inputs the data from the target business operator acquired by the target data acquisition unit 112 in step 102 into a learning model stored in the data storage unit 102 to predict the future cash balance of the target business operator.

[0063] In S106, the output control unit 116 determines whether the projected monthly cash balances of the target business operator, as predicted by the forecasting unit 114 in S104, are below a predetermined threshold. This threshold can be arbitrarily set and may be changed. For example, the output control unit 116 may determine YES if at least one of the projected monthly cash balances is below the threshold, and NO if all of the projected monthly cash balances are above the threshold. The output control unit 116 may also determine YES if the projected monthly cash balance of the last month is below the threshold, and NO if the projected monthly cash balance of the last month is not below the threshold. If the result is YES, the process proceeds to S108; if the result is NO, the process proceeds to S110.

[0064] In S108, the output control unit 116 outputs the prediction result predicted by the prediction unit 114 in S104, along with a warning. The output control unit 116 may output the prediction result and the warning to, for example, the target business operator's communication terminal 80. The warning may include content encouraging fundraising. The warning may include content notifying that funds are insufficient. In S110, the output control unit 116 outputs the prediction result predicted by the prediction unit 114 in S104. The output control unit 116 may output the prediction result to, for example, the target business operator's communication terminal 80.

[0065] In this way, by issuing warnings that prompt fundraising when the projected cash balance falls below a threshold, it is possible to mitigate the risk of profitable businesses going bankrupt due to cash shortages and support the stabilization and further growth of the businesses.

[0066] Figure 9 schematically shows an example of the processing flow by the data processing device 100. Here, we show an example of the flow in which the output control unit 116 controls the output of a warning when the balance obtained by applying the corresponding monthly accounts receivable and accounts payable to each of the projected monthly cash balances of the target business operator acquired by the forecasting unit 114 is less than a predetermined threshold. Note that here we will mainly explain the differences from Figure 8.

[0067] In S202, the target data acquisition unit 112 acquires data on the target business operator. In S204, the prediction unit 114 inputs the data on the target business operator acquired by the target data acquisition unit 112 in S202 into a learning model stored in the data storage unit 102 to predict the future cash balance of the target business operator.

[0068] In S206, the output control unit 116 adds the accounts receivable for each of the projected monthly cash balances of the target business operator, which were projected by the forecasting unit 114 in S204, and subtracts the accounts payable for that month, and compares the result to a predetermined threshold. This threshold can be set arbitrarily and may be changed. For example, 0 may be set as the threshold. The output control unit 116 may determine YES if at least one of the "cash balance + accounts receivable - accounts payable" for multiple months is less than or equal to the threshold, and NO if all of them are greater than the threshold. The output control unit 116 may also determine YES if the "cash balance + accounts receivable - accounts payable" for the last month of the multiple months is less than or equal to the threshold, and NO if the "cash balance + accounts receivable - accounts payable" for the last month is not less than or equal to the threshold. If the result is YES, the process proceeds to S208; if the result is NO, the process proceeds to S210.

[0069] In S208, the output control unit 116 outputs the prediction result predicted by the prediction unit 114 in S204, along with a warning. In S210, the output control unit 116 outputs the prediction result predicted by the prediction unit 114 in S204.

[0070] To prevent profitable bankruptcies, it is necessary to consider accounts payable and accounts receivable in addition to the cash balance. The data processing device 100 performs the processing shown in Figure 9, which makes it possible to determine whether or not there will be a cash shortage after considering accounts payable and accounts receivable. This reduces the risk of profitable bankruptcies due to cash shortages for the target business and supports the stabilization and further growth of the target business.

[0071] Figure 10 schematically shows an example of the hardware configuration of a computer 1200 that functions as a data processing device 100. A program installed on the computer 1200 can cause the computer 1200 to function as one or more "parts" of the apparatus according to this embodiment, or to cause the computer 1200 to execute operations associated with the apparatus according to this embodiment or such one or more "parts", and / or to cause the computer 1200 to execute a process or a stage of such process according to this embodiment. Such a program may be executed by the CPU 1212 to cause the computer 1200 to execute specific operations associated with some or all of the blocks in the flowcharts and block diagrams described herein.

[0072] The computer 1200 according to this embodiment includes a CPU 1212, RAM 1214, and a graphics controller 1216, which are interconnected by a host controller 1210. The computer 1200 also includes input / output units such as a communication interface 1222, a storage device 1224, a DVD drive 1226, and an IC card drive, which are connected to the host controller 1210 via an input / output controller 1220. The DVD drive 1226 may be a DVD-ROM drive and a DVD-RAM drive, etc. The storage device 1224 may be a hard disk drive and a solid-state drive, etc. The computer 1200 also includes legacy input / output units such as a ROM 1230 and a keyboard, which are connected to the input / output controller 1220 via an input / output chip 1240.

[0073] The CPU 1212 operates according to the programs stored in the ROM 1230 and RAM 1214, thereby controlling each unit. The graphics controller 1216 acquires the image data created by the CPU 1212 and stores it in the frame buffer provided in RAM 1214 or within itself, and enables the image data to be displayed on the display device 1218.

[0074] The communication interface 1222 communicates with other electronic devices via a network. The storage device 1224 stores programs and data used by the CPU 1212 in the computer 1200. The DVD drive 1226 reads programs or data from a DVD-ROM 1227, etc., and provides them to the storage device 1224. The IC card drive reads programs and data from an IC card and / or writes programs and data to an IC card.

[0075] The ROM 1230 stores boot programs and / or hardware-dependent programs of the computer 1200, which are executed by the computer 1200 upon activation. The input / output chip 1240 may also connect various input / output units to the input / output controller 1220 via USB ports, parallel ports, serial ports, keyboard ports, mouse ports, etc.

[0076] The program is provided on a computer-readable storage medium such as a DVD-ROM 1227 or an IC card. The program is read from the computer-readable storage medium and installed on a storage device 1224, RAM 1214, or ROM 1230, which are examples of computer-readable storage media, and executed by the CPU 1212. The information processing described within these programs is read by the computer 1200, resulting in coordination between the program and the various types of hardware resources described above. The apparatus or method may be configured to realize the operation or processing of information in accordance with the use of the computer 1200.

[0077] For example, when communication is performed between a computer 1200 and an external device, the CPU 1212 may execute a communication program loaded into RAM 1214 and, based on the processing described in the communication program, instruct the communication interface 1222 to perform communication processing. Under the control of the CPU 1212, the communication interface 1222 reads transmission data stored in a transmission buffer area provided in a recording medium such as RAM 1214, storage device 1224, DVD-ROM 1227, or IC card, transmits the read transmission data to the network, or writes received data received from the network to a reception buffer area or the like provided on the recording medium.

[0078] Furthermore, the CPU 1212 may read all or necessary parts of files or databases stored on external recording media such as the storage device 1224, DVD drive 1226 (DVD-ROM 1227), or IC card into the RAM 1214, and perform various types of processing on the data in the RAM 1214. The CPU 1212 may then write the processed data back to the external recording media.

[0079] Various types of information, such as various types of programs, data, tables, and databases, may be stored on the recording medium and subjected to information processing. The CPU 1212 may perform various types of processing on the data read from RAM 1214, including various types of operations, information processing, conditional judgments, conditional branching, unconditional branching, information retrieval / replacement, etc., as described throughout this disclosure and specified by the program instruction sequence, and write the results back to RAM 1214. The CPU 1212 may also retrieve information in files, databases, etc., within the recording medium. For example, if multiple entries are stored in the recording medium, each having an attribute value of a first attribute associated with an attribute value of a second attribute, the CPU 1212 may search among the multiple entries for an entry that matches the specified condition for the attribute value of the first attribute, read the attribute value of the second attribute stored in that entry, and thereby obtain the attribute value of the second attribute associated with the first attribute that satisfies the predetermined condition.

[0080] The program or software module described above may be stored on or near the computer 1200 in a computer-readable storage medium. Alternatively, a recording medium such as a hard disk or RAM provided within a server system connected to a dedicated communication network or the Internet can be used as a computer-readable storage medium, thereby providing the program to the computer 1200 via the network.

[0081] In this embodiment, blocks in the flowchart and block diagram may represent a stage in a process in which an operation is performed or a "part" of a device that has the role of performing an operation. A particular stage and "part" may be implemented by a dedicated circuit, a programmable circuit supplied with computer-readable instructions stored on a computer-readable storage medium, and / or a processor supplied with computer-readable instructions stored on a computer-readable storage medium. The dedicated circuit may include digital and / or analog hardware circuits, and may include integrated circuits (ICs) and / or discrete circuits. The programmable circuit may include reconfigurable hardware circuits, such as field-programmable gate arrays (FPGAs) and programmable logic arrays (PLAs), which include logical AND, logical OR, exclusive OR, negated AND, negated OR, and other logical operations, flip-flops, registers, and memory elements.

[0082] A computer-readable storage medium may include any tangible device capable of storing instructions to be executed by a suitable device, and as a result, a computer-readable storage medium having instructions stored therein will comprise a product that includes instructions that can be executed to create means for performing operations specified in a flowchart or block diagram. Examples of computer-readable storage media may include electronic storage media, magnetic storage media, optical storage media, electromagnetic storage media, semiconductor storage media, etc. More specific examples of computer-readable storage media may include floppy disks, diskettes, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), electrically erasable programmable read-only memory (EEPROM), static random access memory (SRAM), compact disk read-only memory (CD-ROM), digital multipurpose disc (DVD), Blu-ray® disc, memory stick, integrated circuit card, etc.

[0083] Computer-readable instructions may include assembler instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk®, Java®, C++, and traditional procedural programming languages ​​such as the C programming language or similar programming languages.

[0084] Computer-readable instructions may be provided to a processor or programmable circuit of a general-purpose computer, a special-purpose computer, or other programmable data processing device, either locally or via a local area network (LAN), the Internet, or other wide area network (WAN), so that the processor or programmable circuit of a general-purpose computer, a special-purpose computer, or other programmable data processing device may execute the instructions to create means for the operation specified in a flowchart or block diagram. Examples of processors include computer processors, processing units, microprocessors, digital signal processors, controllers, microcontrollers, and the like.

[0085] Although the present invention has been described above using embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments. It will be apparent to those skilled in the art that various modifications or improvements can be made to the above embodiments. It will be clear from the claims that such modified or improved forms may also be included in the technical scope of the present invention.

[0086] It should be noted that the execution order of operations, procedures, steps, and stages in the apparatus, systems, programs, and methods described in the claims, specifications, and drawings is not explicitly stated as "before" or "prior to," and that these can be implemented in any order unless the output of a previous process is used in a later process. Even if the operation flow in the claims, specifications, and drawings is described using phrases such as "first," and "next," for convenience, this does not mean that it is essential to perform the operations in that order. [Explanation of Symbols]

[0087] 50 Network, 80 Communication terminal, 82 User, 100 Data processing unit, 102 Data storage unit, 104 Data acquisition unit, 106 Identification unit, 108 Data management unit, 110 Learning model generation unit, 112 Target data acquisition unit, 114 Prediction unit, 116 Output control unit, 200 Data management device, 310 Account table, 320 Sub-account table, 330 Transaction details table, 332 January balance, 334 February balance, 340 Accounting settings table, 350 Account balance table, 352 Beginning balance, 360 Monthly cash balance, 370 Explanatory variable, 380 Dependent variable, 390 Data mart, 400 Cash balance graph, 410 Baseline trend, 420 High balance trend, 430 Low balance trend, 1200 Computer, 1210 Host controller, 1212 CPU, 1214 RAM, 1216 Graphics controller, 1218 Display device, 1220 Input / Output controller, 1222 Communication interface, 1224 Storage device, 1226 DVD drive, 1227 DVD-ROM, 1230 ROM, 1240 Input / Output chip

Claims

1. A storage unit that stores a learning model in which the cash balances for multiple months prior to a base month and the amounts corresponding to one or more accounts in the month preceding the base month are at least explanatory variables, and the cash balances for multiple future months are the dependent variable, A prediction unit inputs the cash balances for multiple months prior to a base month of the target business operator, and the amounts corresponding to one or more accounts for the month preceding the base month, into the learning model, and obtains the future cash balances for multiple months output from the learning model as the projected cash balances for multiple months of the target business operator. An output control unit controls the output of the projected monthly cash balances of the target business operator acquired by the prediction unit. A data processing device equipped with the following features.

2. The data processing device according to claim 1, wherein the prediction unit inputs the monthly cash balances of the target business operator for the past year for a reference month and the amounts corresponding to one or more accounts for the month preceding the reference month into the learning model, and obtains the future monthly cash balances output from the learning model as the projected cash balances for the target business operator for multiple future months.

3. The data processing device according to claim 1 or 2, wherein the output control unit controls the output to output a warning when the predicted monthly cash balance of the target business operator obtained by the prediction unit is less than a predetermined threshold.

4. The data processing device according to claim 1 or 2, wherein the output control unit controls the output to output a warning if the balance obtained by applying the corresponding monthly accounts receivable and accounts payable to each of the multiple monthly predicted cash balances of the target business operator acquired by the prediction unit is less than a predetermined threshold.

5. The data processing apparatus according to claim 1 or 2, wherein the output control unit controls the output to output a baseline trend, which is the predicted trend of the target business operator's cash balance for multiple months in the future, obtained by the prediction unit; a high-balance trend, where the cash balance for the multiple months is greater than the baseline trend; and a low-balance trend, where the cash balance for the multiple months is less than the baseline trend.

6. The data processing device according to claim 1 or 2, wherein the learning model further uses statistical information of cash balances for multiple past months as explanatory variables.

7. The data processing device according to claim 1 or 2, wherein the learning model uses at least one of the following as explanatory variables: capital stock, advance payments, accounts receivable, long-term borrowings, interest income, sales revenue, ending inventory, gains on sale of securities, accounts receivable, and miscellaneous income.

8. A data processing method performed by a computer, A prediction step is to input the cash balances of the target business operator for multiple months prior to the base month and the amounts corresponding to one or more accounts in the month prior to the base month into a learning model, which has at least the cash balances for multiple months prior to the base month and the amounts corresponding to one or more accounts in the month prior to the base month as explanatory variables, and the cash balances for multiple months in the future as the dependent variable, into the learning model, and obtain the future cash balances for multiple months output from the learning model as the predicted cash balances for multiple months in the future for the target business operator. An output control stage that controls the output of the projected monthly cash balances of the target business operator obtained in the forecasting stage, and A data processing method comprising the following features.

9. On the computer, A prediction step is to input the cash balances of the target business operator for multiple months prior to the base month and the amounts corresponding to one or more accounts in the month prior to the base month into a learning model, which has at least the cash balances for multiple months prior to the base month and the amounts corresponding to one or more accounts in the month prior to the base month as explanatory variables, and the cash balances for multiple months in the future as the dependent variable, into the learning model, and obtain the future cash balances for multiple months output from the learning model as the predicted cash balances for multiple months in the future for the target business operator. An output control stage that controls the output of the projected monthly cash balances of the target business operator obtained in the forecasting stage, and A program to execute.

Images