External auditing equipment, external monitoring system, external auditing method, and external auditing program

Abstract

To provide an external audit apparatus and system thereof, configured to always audit a wide range of transactions of a company without requiring manual work, and report a result of the audio early to prevent unauthorized transactions of the company from spreading.SOLUTION: An external audio apparatus connected to an audited company and a financial institution using a network line includes: a transaction information acquisition unit which acquires transaction information of the audited company; a balance information acquisition unit which acquires balance information of the audited company from the financial institution; an ordinary transaction information generation unit which accumulates the transaction information and the balance information to generate ordinary transaction information of the audited company; a determination unit which determines whether there is the transaction information and / or the balance information that deviates from the ordinary transaction information; an anomaly information generation unit which generates anomaly information when there is a deviation from the ordinary transaction information; and a reporting unit which reports the anomaly information to the audited company.SELECTED DRAWING: Figure 4

Description

Technical Field

[0001] The present invention relates to an external auditing device, an external monitoring system, an external auditing method, and an external auditing program, and particularly to an external auditing device, an external monitoring system, an external auditing method, and an external auditing program for supporting the auditing operations of enterprises.

Background Art

[0002] There still exist falsifications of accounting information (cooking the books) for a company to show good performance, or irregularities such as accounting fraud committed by the company's employees. Moreover, the more sophisticated the concealment is, the later the discovery will be. If irregularities in a company are discovered, many problems will occur, such as the loss of the company's own credibility, deterioration of the relationship with business partners, and delisting if the company is listed, which will affect many stakeholders such as the company's employees, investors in the company, and transaction banks.

[0003] Currently, internal audits within a company and further audits of the company by external auditing firms are being carried out. However, it is not easy to detect irregularities from the huge number of daily transactions and settlements. Also, it is impossible to constantly monitor and audit both internally and externally. For example, it takes time to obtain settlement records from a bank (transaction financial institution) that has transactions with the company. A large amount of manpower is required for internal and external audits. Even if transaction details can be obtained, detailed information cannot be confirmed. Since the transaction financial institution also obtains records manually, there are still some errors. Therefore, in reality, audits are limited to only partially extracted cases rather than audits of all transactions.

[0004] Therefore, measures have been studied to automatically detect irregularities in money management within a company. For example, a user information management device has been proposed that has a detection unit for detecting a transaction with a predetermined transaction pattern as an abnormal transaction from transaction information related to cash transactions and the user information of the user associated with the transaction information (see Patent Document 1).

[0005] According to the user information management device described in Patent Document 1, a certain degree of effectiveness can be expected in detecting abnormal transactions within companies using the device. However, the user information management device described in Patent Document 1 is limited to detecting fraud by employees of the company in question, and does not address the detection of large-scale fraudulent activities by the company itself or by departments within the company. [Prior art documents] [Patent Documents]

[0006] [Patent Document 1] Japanese Patent Publication No. 2019-164662 [Overview of the project] [Problems that the invention aims to solve]

[0007] The present invention has been made in view of the above points, and provides an external auditing device, system, method, and program that does not rely on human labor, enables continuous auditing of a wide range of corporate transactions, and promptly notifies the corporate of the audit results to prevent the expansion of fraudulent corporate transactions. [Means for solving the problem]

[0008] In other words, the external monitoring device of the embodiment is connected to the audited company and the audited company's financial institution via a network line, and is characterized by comprising: a transaction information acquisition unit that acquires transaction information of the audited company from the audited company; a balance information acquisition unit that acquires balance information of the audited company from the financial institution; a normal transaction information generation unit that stores transaction information and balance information to generate normal transaction information of the audited company; a determination unit that determines whether there is any or both of the audited company's transaction information or balance information that deviates from the normal transaction information; an abnormal information generation unit that generates abnormal information when either or both of the audited company's transaction information or balance information deviates from the normal transaction information; and a notification unit that notifies the audited company of the abnormal information.

[0009] Furthermore, the system may be equipped with a blocking detection unit that detects the blocking of access between the audited company and its financial institutions, or the blocking of access between the audited company and external monitoring devices, and the notification unit may notify the audited company of the detected blocking of access as abnormal information.

[0010] Furthermore, the transaction information acquisition unit may be equipped with a change detection unit that detects changes to the transaction information after it has been acquired, and the notification unit may notify the audited company of the detected changes to the transaction information as abnormal information.

[0011] Furthermore, the reporting department may also report any abnormal information to the auditing firm contracted by the audited company.

[0012] Furthermore, the normal transaction information generation unit may generate normal transaction information by accumulating transaction information and balance information over a predetermined period.

[0013] Furthermore, the normal transaction information generation unit may include a machine learning unit, which may accumulate transaction information and balance information over a predetermined period to generate normal transaction information.

[0014] Furthermore, transaction information may include the audited company's settlement records, inventory records, order records, income statement records, or balance sheet records.

[0015] The transaction information acquisition unit may also acquire transaction information of the audited company's subsidiaries or equity-method affiliates if the audited company has such subsidiaries or equity-method affiliates. [Effects of the Invention]

[0016] The external audit device of the present invention is connected to the audited company and the audited company's financial institutions via a network line and includes: a transaction information acquisition unit that acquires transaction information of the audited company from the audited company; a balance information acquisition unit that acquires balance information of the audited company from the financial institutions; a normal transaction information generation unit that stores transaction information and balance information to generate normal transaction information of the audited company; a determination unit that determines whether there is any or both of the audited company's transaction information or balance information that deviates from the normal transaction information; an abnormal information generation unit that generates abnormal information when either or both of the audited company's transaction information or balance information deviates from the normal transaction information; and a notification unit that notifies the audited company of the abnormal information. Therefore, it is possible to conduct continuous audits of a wide range of a company's transactions without relying on human intervention, and the audit results can be notified to the company early to prevent the expansion of fraudulent transactions. Similar effects can also be expected with external audit systems, external audit methods, and external audit programs. [Brief explanation of the drawing]

[0017] [Figure 1] This is a schematic diagram showing the configuration of an external audit system, including an external audit device. [Figure 2] This is a schematic block diagram showing the configuration of the external auditing system. [Figure 3] This is a schematic block diagram showing the functional components of the external auditing device. [Figure 4] This is a schematic diagram illustrating the acquisition of transaction information and balance information. [Figure 5] This diagram shows the normal transaction information consisting of (A) the number of items sold, (B) inventory records, and (C) balance information, and a schematic representation of the abnormal situation with (D) the number of items sold, (E) inventory records, and (F) balance information. [Figure 6] This is a schematic diagram showing transaction information and balance information. [Figure 7] This is an example of a display screen for reporting abnormal information. [Figure 8] This is a schematic diagram illustrating the blocking of access. [Figure 9] This is a schematic diagram illustrating the generation of normal transaction information by the machine learning unit. [Figure 10] It is a schematic diagram of another embodiment of the audited enterprise. [Figure 11] It is the first flowchart for explaining the processing of the external auditing method. [Figure 12] It is the second flowchart for explaining the processing of the external auditing method. [Figure 13] It is the third flowchart for explaining the processing of the external auditing method. [Figure 14] It is the fourth flowchart for explaining the processing of the external auditing method.

Embodiments for Carrying Out the Invention

[0018] The external auditing system of the embodiment includes an external monitoring device, and audits and monitors irregularities within the enterprise such as transactions and accounting of the audited enterprise from outside the audited enterprise which is the client. Then, at the stage where irregularities are foreseen or at the initial stage of irregularities, a predetermined report is made to the audited enterprise and further to the auditing corporation. Thus, it promotes raising awareness of the risk of irregularities or the enterprise's countermeasures against the initially detected irregularities.

[0019] The schematic configuration of the external auditing system 1 of the embodiment is shown as the schematic diagram in FIG. 1. In the external auditing system 1, the audited enterprise 3 and the transaction financial institution 4 of the audited enterprise 3 are connected by a network line Nt, and further, the audited enterprise 3 and the transaction financial institution 4 are connected to the external auditing device 2 by the network line Nt. Thus, each location is interconnected by the network line Nt.

[0020] A known Internet line is used for the network line Nt connecting each location. In addition, a dedicated line may be installed and operated to ensure the transmission and reception of data among the audited enterprise 3, the transaction financial institution 4, and the external auditing device 2.

[0021] Audited Company 3 is a company that utilizes the services of the External Audit System 1 and has its internal transactions, accounting, and other irregularities audited and monitored through the External Audit Device 2 located outside of Audited Company 3. This applies to both listed and unlisted companies. Furthermore, the form of the company, such as a stock company, limited liability company, or limited company, is also optional. In addition, organizations with legal entities such as incorporated associations, foundations, medical corporations, and agricultural corporations are also included. Moreover, as will be described later, if Audited Company 3 has subsidiaries or equity-method affiliates, those subsidiaries or equity-method affiliates of Audited Company 3 are also included as Audited Company 3.

[0022] Transaction financial institution 4 is a financial institution that handles the current account or savings account of the audited company 3, and is mainly a bank. Furthermore, transaction financial institution 4 may also include financial institutions such as credit unions and credit cooperatives. Note that the number of transaction financial institutions 4 is not limited to one bank, but may include multiple banks.

[0023] The external audit device 2 has the function of obtaining transaction information from the audited company 3, and further obtaining balance information from the transaction financial institution 4, and cross-referencing the information. It also has the function of notifying the audited company 3 if it detects an abnormal transaction. Details of the external audit device 2 will be described in detail using Figure 2 and subsequent figures.

[0024] As shown in the schematic diagram in Figure 1, the audited company 3, the financial institution 4, and the external audit device 2 are connected to each other by a network line Nt. Here, various data such as deposits and withdrawals are transmitted from the audited company 3 to the financial institution 4, and processing such as API (Application Programming Interface) authentication is used to enhance the security of mutual data exchange (data transmission security). Appropriate approval is also set when various data are transmitted from the audited company 3 to the external audit device 2. Then, the financial institution 4 provides the external audit device 2 with the balance information of the audited company 3 recorded at the financial institution 4.

[0025] Furthermore, as can be understood from the schematic diagram in Figure 1, the abnormal information resulting from the judgment of the external audit device 2 is reported to the audit firm 5 contracted by the audited company 3. The abnormal information resulting from the judgment is then communicated to the internal audit personnel 6 of the audited company 3, and further to the certified public accountant 7 belonging to audit firm 5 who is responsible for audited company 3. In the diagram, server 5a is connected to audit firm 5.

[0026] The external audit system 1 of this embodiment is configured with an external audit device 2 and an audit firm 5 separately. In the illustrated external audit system 1, the external audit device 2 of the system provides external audit services to multiple audited companies 3 and multiple audit firms 5. Alternatively, the audit firm 5 itself may be equipped with an external audit device 2, and the audit firm 5 may use the external audit device 2 to audit the audited companies 3 (client companies) it is responsible for.

[0027] As shown in the schematic block diagram of Figure 2, the external audit device 2 is, in terms of hardware, a computer unit 10 that implements an input / output interface 11 (I / O buffer), ROM 12, RAM 13, storage unit 14, CPU 15, etc. In terms of software, it is implemented by an external audit program loaded into main memory. Specifically, the external audit device 2 (computer unit 10) is various electronic computers (computing resources) such as personal computers (PCs), mainframes, workstations, and cloud computing systems.

[0028] When each function of the external audit device 2 (computer unit 10) shown in Figures 1 and 2 is implemented by software, the external audit device 2 is implemented by executing instructions for a program, which is the software that implements each function. The recording medium that stores this program can be a "non-temporary tangible medium," such as a CD, DVD, semiconductor memory, or programmable logic circuit. This program may also be supplied to the external audit device 2 of the external audit system 1 via any transmission medium capable of transmitting the program (communication network, broadcast wave, etc.).

[0029] The storage unit 14 of the external audit device 2 (computer unit 10) is a known storage device such as an HDD or SSD. The storage unit 14 stores various data, information, the external audit program, and various data necessary for the execution of the program. In addition, each functional unit that performs calculations such as calculations and arithmetic operations is an arithmetic element such as a CPU 15. Furthermore, input devices such as a keyboard and mouse, display devices such as a display, and output devices that output data are also connected to the appropriate input / output interface 11 of the computer unit 10.

[0030] The CPU 101 of the external audit device 2 (computer unit 10) comprises, as shown in the schematic block diagram in Figure 3, a transaction information acquisition unit 110, a balance information acquisition unit 120, a normal transaction information generation unit 130, a determination unit 140, a block detection unit 150, a change detection unit 160, an abnormal information generation unit 170, a notification unit 180, and a machine learning unit 190. The operation and execution of the external audit device 2 are realized in software by an external audit program loaded into main memory.

[0031] The external audit device 2 in external audit system 1 will be explained with reference to the diagrams from Figure 4 onwards.

[0032] The transaction information acquisition unit 110 acquires transaction information 30 from the audited company 3. Transaction information 30 is various information arising from the business activities of the audited company 3, and specifically, as shown in the schematic diagram in Figure 4, it includes settlement records 31 of purchases or sales of goods or services (cash, deposits, accounts receivable, accounts payable, prepaid expenses, unearned revenue, etc.), inventory records 32 (inventory assets) located in the audited company 3's warehouses, etc., order records 33 related to goods or services, and accounting documents such as the audited company 3's income statement 34 (P / L) and balance sheet 35 (B / S). In addition, transaction information 30 also includes information on the quantity and amount of purchases, sales, issuances, and issuances of various financial products such as securities (stocks), bonds (government and corporate bonds), and financial derivatives, as well as information on the type and amount of taxes (various taxes such as income tax, corporate tax, consumption tax, and local tax).

[0033] For example, if audited company 3 is a manufacturing company, inventory records 32 would include the quantity of products (final products, semi-finished products, work in progress), the cost of raw materials purchased for product manufacturing, the number of products shipped, and the inventory quantity. The income statement 34 (P / L) and balance sheet 35 (B / S) of audited company 3 would be based on monthly figures for monthly closings, quarterly figures for quarterly closings, semi-annual figures for semi-annual closings, and year-end figures for full-year closings.

[0034] The transaction information 30 of audited company 3 reflects the daily business activities of audited company 3. Therefore, by making the transaction information 30 of audited company 3 the subject of ongoing audits, it is possible to detect fraud in business activities, and even detect it at an early stage.

[0035] The balance information acquisition unit 120 acquires the balance information 40 of the audited company 3 from the transaction financial institution 4. The transaction financial institution 4 is in a position to manage the current account of the audited company 3. Therefore, the transaction financial institution 4 can grasp the settlement information of the audited company 3. Thus, by constantly tracking the balance information 40 of the audited company 3, it becomes possible to monitor any unusual deposits or withdrawals that deviate from normal business activities. Of course, if there are multiple transaction financial institutions 4, the balance information 40 of each bank will be acquired.

[0036] The normal transaction information generation unit 130 stores transaction information 30 and balance information 40 to generate normal transaction information 50 of the audited company. As shown in Figure 4, transaction information 30 (31 to 35) from the audited company 3 and balance information 40 from the transaction financial institution 4 are transmitted (acquired) to the external audit device 2. The normal transaction information 50 referred to here is information that links (associates) the audited company 3's business activities, such as the sale of products and provision of services, and the purchase or sale of assets (movable and immovable property), with the inflow and outflow of money associated with those activities, so that the content of the business activities and the amount of inflow and outflow can be grasped.

[0037] The determination unit 140 determines whether there is any or both of the transaction information 30 or balance information 40 of the audited company 3 that deviates from the normal transaction information 50.

[0038] The abnormal information generation unit 170 generates abnormal information 60 (see Figures 5 and 6) when either or both of the transaction information 30 or balance information 40 of the audited company 3 deviate from normal transaction information 50.

[0039] Reporting unit 180 reports abnormal information 60 to audited company 3.

[0040] When generating normal transaction information 50 in the external audit device 2, the schematic diagram in Figure 5 is referenced. In the configuration of Figure 5, as an example, transaction information 30 and balance information 40 covering one fiscal year (12 months) up to the previous period are accumulated, and the trends in the number of goods sold, the number of goods issued, and the balance are shown. 1Q, 2Q, 3Q, and 4Q in the figure represent quarters.

[0041] Figure 5(A) shows the annual trend of the number of goods sold (sales settlement records 31) included in the transaction information 30 from audited company 3. Figure 5(B) shows the annual trend of inventory records 32. In the case of seasonal goods, production is increased and inventory is increased in anticipation of periods when sales volume increases. Figure 5(C) shows the annual trend of balance information 40. In actual sales, payment is often made by promissory note as accounts receivable rather than cash. Therefore, there is a time lag of about three months between the time of recording sales and the time when the promissory note is actually issued and received. The number of goods sold, inventory records, and balance information (average up to the previous period) in Figures 5(A), (B), and (C) represent the normal transaction information 50 of audited company 3 under normal circumstances. Transaction information 30 and balance information 40 are accumulated over a predetermined period, and for example, the average for each month can be calculated by comparing it with the previous period, the period before that, etc. The same applies to inventory records and balance information. In this way, fluctuations over multiple accounting years are mitigated, and the normal transaction information 50 of audited company 3 is generated.

[0042] Figures 5(D), (E), and (F) correspond to the items of product sales, inventory records, and balance information in Figures 5(A), (B), and (C). Both show the annual trend from the beginning to the end of the period. Figure 5(D) shows the annual trend of product sales (sales settlement record 31) for the current period. Compared to Figure 5(A), it is higher, indicating strong sales. Also, the balance information 40 in Figure 5(F) has increased. In this case, improved performance of audited company 3 can be expected. However, according to the inventory record 32 in Figure 5(E), there is little change compared to Figure 5(B). The dashed lines in Figures 5(D), (E), and (F) represent the normal transaction information in Figures 5(A), (B), and (C), and are displayed to highlight the differences.

[0043] In the case of Figure 5, aside from the price increase of audited company 3's products, there are no significant changes in the inventory records. This suggests that no actual product shipments have taken place, and it is possible that only the invoices have been manipulated. This raises suspicions of fraudulent accounting practices, such as circular transactions, primarily carried out by audited company 3. Although not shown in the diagram, transaction information 30 regarding raw material purchases should also be considered in the verification. It is possible that while the actual product sales volume was the same as in previous years, the sales were recorded by increasing the number of product shipments on the books.

[0044] Therefore, the determination unit 140 compares "the normal transaction information 50 of audited company 3 (averaged over multiple years) when the number of goods sold, inventory records, and balance information in Figures 5(A), (B), and (C) are normal" with "the number of goods sold, inventory records, and balance information in Figures 5(D), (E), and (F) (for which there are doubts about a certain period)". The inventory record (Figure 5(E)) has not changed significantly from the inventory record in the normal transaction information (Figure 5(B)). However, the number of goods sold (Figure 5(D)) and the balance information (Figure 5(F)) have increased significantly. The inventory record should have progressed accordingly. In other words, there is a contradiction between the number of goods sold and the number of inventory (the transaction information deviates from the normal transaction information). Specifically, the difference between the number of goods sold and balance information for the same month included in the normal transaction information 50 and the number of goods sold and balance information in Figures 5(D) and (F) is calculated. Then, a contradiction is discovered because the expected increase in inventory records does not exist.

[0045] From this comparison, the determination unit 140 determined that General The abnormal information generation unit 170 then generates a judgment result indicating "a deviation from normal transaction information." General As evidence of deviations from normal transaction information, the sales figures, inventory records, and balance information in Figures 5(D), (E), and (F) are generated as anomaly information 60 for audited company 3.

[0046] When generating the normal transaction information 50, instead of using the predetermined period shown in Figure 5, it is possible to record the matching of sequential transaction information 30 and balance information 40 daily and hourly. For example, the schematic diagram in Figure 6 shows the number of goods sold (sales settlement record 31) included in the transaction information 30 of audited company 3 on a certain day (20xx / yy / zz). It also shows the balance information 40 of the transaction financial institution 4 on the same day. In this example, the combination of the number of goods sold (sales settlement record 31) included in the transaction information 30 of audited company 3 on that day (20xx / yy / zz) and the balance information 40 becomes the normal transaction information 50.

[0047] When goods are sold with cash payment, the sales amount (sales price x number of items sold) is added to the balance information 40 at the time of sale. Naturally, the sales price and number of items sold are known. If the corresponding sales amount does not exist in the balance information 40, it may indicate unfair pricing or a fictitious transaction. Specifically, the normal sales amount is known, and the anomaly can be detected by the difference between the balance information, which is expected to show an increase from the normal sales amount, and the actual balance information. Records of this daily matching of transaction information 30 and balance information 40 in business activities are also generated as anomaly information 60 for audited company 3.

[0048] The notification unit 180 of the external audit device 2 notifies the client, audited company 3, of the result of the deviation from the normal transaction information 50 detected by the determination unit 140, as well as the abnormal information 60. In addition to the audited company 3, the notification unit 180 may also notify the auditing firm 5, which has a contract with the audited company 3, of the same result of the deviation and the abnormal information 60. Subsequently, the auditing firm 5 contacts the person in charge of internal audits 6 of the audited company 3, and further, the certified public accountant 7 belonging to the auditing firm 5 who is in charge of the audited company 3.

[0049] As an example of a report from the reporting unit 180 (an example of a display screen), Figure 7(A) corresponds to an example of a report of fraudulent activity discovered through tracking over a specified period as shown in Figure 5 above. The display device (not shown) on the terminal of the internal auditor 6 of audited company 3 or the certified public accountant 7 in charge of audited company 3 displays the message, "Warning! Please check the sales quantity, inventory, and accounts receivable of product PP of company AAA from bb month to cc month of 20aa."

[0050] Furthermore, Figure 7(B) corresponds to an example of a report of fraudulent activity discovered through tracking a single day as shown in Figure 6 above. The display device (not shown) on the terminal of the internal auditor 6 of audited company 3 or the certified public accountant 7 in charge of audited company 3 displays the message, "Warning! Please check the sales quantity of product QQ of BBB Company on 20xx / yy / zz."

[0051] In this way, the facts of the fraud that occurred at audited company 3, as well as signs of fraud, are shared among audited company 3 and other relevant parties, enabling a swift response to the fraud problem. Even if the external audit device 2 itself is installed within audit firm 5, the results of the deviation and the anomaly information 60 will similarly be reported.

[0052] The external audit system 1 (external audit device 2) of this embodiment is equipped with a blockage detection unit 150 and a change detection unit 160.

[0053] The blocking detection unit 150 detects the blocking of access between the audited company 3 and the trading financial institution 4, or the blocking of access between the audited company 3 and the external audit device 2. In the external audit system 1 of this embodiment, the audited company 3, the trading financial institution 4, and the external audit device 2 are connected to each other by a network line Nt (see Figure 1). If, as shown in the schematic diagram of Figure 8, an access blockage (X in the figure) occurs between the audited company 3 and the trading financial institution 4, the records of deposits and withdrawals during that period (balance information 40) will not remain with the trading financial institution 4, and the audited company 3 may have made some deposits or withdrawals that it wanted to keep confidential.

[0054] Furthermore, if access is blocked (X in the diagram) between the audited company 3 and the external audit device 2 (including cases where it is installed within the audit firm 5), the transaction information 30 during that period will not be recorded by the financial institution 4, and the audited company 3 may have conducted some transactions that it wish to keep confidential.

[0055] The very occurrence of access being blocked between these locations can be considered extremely unnatural. Therefore, the reporting unit 180 reports the detected access blockage as a form of abnormal information 60 to the audited company 3 and further to the auditing firm 5. Upon receiving such a report, the audited company 3 or the auditing firm 5 can conduct an internal investigation and take countermeasures at a stage where fraudulent transactions are minimal or only as a warning sign.

[0056] The change detection unit 160 detects changes in the transaction information 30 (31 to 35) after it has been acquired by the transaction information acquisition unit 110. Once a transaction such as the sale or purchase of goods has been completed at the audited company 3, changes to the transaction information 30 should not be permitted without going through the proper procedures such as returns or discounts. Therefore, if a change in the transaction information 30 occurs without processing in accordance with internal regulations and accounting rules, it is highly likely that there has been illegal falsification of transactions and accounting at the audited company 3.

[0057] Therefore, the reporting unit 180 reports the detected change in transaction information 30 as a form of abnormal information 60 to the audited company 3 and further to the auditing firm 5. Upon receiving such a report, the audited company 3 or the auditing firm 5 can conduct an internal investigation and take countermeasures at a stage where illegal falsification of transactions and accounting is minimal or only a warning sign.

[0058] As a modified example of the external audit system 1 (external audit device 2) of the embodiment, the normal transaction information generation unit 130 includes a machine learning unit 190. As shown in Figure 5, the machine learning unit 190 estimates a reasonable transaction volume and balance volume from the transaction activities of the audited company 3 by accumulating transaction information 30 and balance information 40 over a predetermined period. This estimated value is generated as normal transaction information 50. Figure 9 is a schematic diagram showing the outline of the machine learning unit 190.

[0059] Transaction information 30 of the audited company 3 includes settlement records 31, inventory records 32, order records 33, and accounting documents such as income statements 34 and balance sheets 35, which are input to the machine learning unit 190 via the transaction information acquisition unit 110. In addition, balance information 40 of the audited company 3 from the transaction financial institution 4 is input to the machine learning unit 190 via the balance information acquisition unit 120. Based on the various input information, the machine learning unit 190 uses machine learning methods such as support vector machines (SVM), model trees, decision trees, neural networks, multilinear regression, locally weighted regression, and established search methods to generate estimated normal transaction information 50.

[0060] For example, when a neural network is used, normal transaction information 50 is generated as the standard trading activity of the audited company 3 over a specified period, based on transaction information 30 such as settlement records 31, inventory records 32, order records 33, and balance information 40. If a transaction occurs that deviates significantly from this normal transaction information 50, the involvement of some kind of fraudulent transaction is suspected. Furthermore, the normal transaction information 50 derived from Figures 5(A), (B), and (C) above, as well as the examples of suspected anomalies (anomaly information 60) in Figures 5(D), (E), and (F), are also utilized as data for machine learning.

[0061] By running machine learning in the machine learning unit 190, it becomes easier to track individual records of multiple types within the transaction information 30. When generating transaction information 50 under normal circumstances, there is a tendency for bias in the types of records within the transaction information 30. In this respect, machine learning allows for objective estimation, although there are differences in the weighting of the data.

[0062] The corporate structure of audited company 3 is also diversifying; in addition to being a single entity, it may have multiple companies as part of a group. In such cases, audits and monitoring are required not only for audited company 3 but also for the group companies. Currently, due to the control relationship of shares, consolidated financial statements are applied to group companies. Consequently, if fraudulent activity or accounting fraud is discovered in one company belonging to the group, it will affect the entire group of companies.

[0063] Even if we were to include all the group companies of audited company 3, expanding the scope without limit would increase complexity. Therefore, the group companies of audited company 3 are grouped by their equity stake and capital control ratio. Specifically, if audited company 3 has subsidiaries or equity-method affiliates, the transaction information 30 and balance information 40 of those companies are also acquired by the external audit device 2, just as with audited company 3, as described above.

[0064] The schematic diagram in Figure 10 shows a situation where, in addition to the audited company 3, there are also its subsidiary 3a and equity-method affiliate 3b. The audited company 3, subsidiary 3a, and equity-method affiliate 3b are connected to the external audit equipment 2 (including cases where it is installed within the audit firm 5) via a network line Nt. Furthermore, the audited company 3, subsidiary 3a, and equity-method affiliate 3b are each connected to their respective financial institutions 4 via network lines Nt. Therefore, it is possible to audit fraudulent transactions within the audited company 3 and its group companies.

[0065] As described in the series of explanations, the external audit system 1 and external audit device 2 of the embodiment automatically detect and report transactions suspected of being fraudulent, eliminating the need for manual intervention in audit work that is normally done manually. This allows for faster response by companies and audit firms once fraudulent transactions are discovered. In particular, currently, when audit firms request balance confirmation letters from financial institutions, there are problems such as the time required to verify signatures or seals, as well as the loss of mailing time and the possibility of mail being lost. Therefore, real-time verification is difficult with existing audit methods. For this reason, although it is desirable to audit all transactions, the appropriateness of audits has been limited to selecting only some transactions. In contrast, the embodiment enables real-time verification because it is automated. Furthermore, it becomes possible to audit all transactions, which was previously difficult. In particular, in the case of foreign financial institutions, balance confirmation letters are rarely issued, and there is no way to verify transactions with overseas institutions. However, with the introduction of the external audit system 1 and external audit device 2 of the embodiment, the workload on financial institutions is reduced, and the barriers to cooperation in audit work are lowered.

[0066] Next, using the flowcharts in Figures 11 to 14, we will explain both the external audit method and the external audit program in the external audit system 1 (external audit device 2). The external audit method is executed by the CPU 15 of the external audit device 2 (computer unit 10) based on the external audit program. The external audit program causes the external audit device 2 (computer unit 10) shown in Figure 2 to execute various functions, including transaction information acquisition function, balance information acquisition function, normal transaction information generation function, judgment function, abnormal information generation function, notification function, as well as blockage detection function, change detection function, and machine learning function. Since each function overlaps with the explanation of the external audit device 2 above, details will be omitted.

[0067] As shown in the flowchart of Figure 11, the processing of the external audit device 2 (computer unit 10) includes a transaction information acquisition step (S110), a balance information acquisition step (S120), a normal transaction information generation step (S130), a judgment step (S140), an abnormal information generation step (S170), and a notification step (S180). Furthermore, the external audit device 2 (computer unit 10) includes various steps such as a blockage detection step (S150) as shown in the flowchart of Figure 12, a change detection step (S160) as shown in the flowchart of Figure 13, and a machine learning step (S190) as shown in the flowchart of Figure 14. Of course, the various steps necessary for the operation of the external audit device 2 (computer unit 10) itself are naturally included.

[0068] The transaction information acquisition function acquires transaction information 30 of audited company 3 from audited company 3 (S110; transaction information acquisition step). The balance information acquisition function acquires balance information 40 of audited company 3 from transaction financial institution 4 (S120; balance information acquisition step). The normal transaction information generation function accumulates the transaction information 30 and balance information 40 to generate normal transaction information 50 of audited company 3 (S130; normal transaction information generation step). The determination function determines whether there is any or both of the transaction information 30 or balance information 40 of audited company 3 that deviates from the normal transaction information 50. It determines whether either or both of the transaction information 30 or balance information 40 of audited company 3 are outside the normal transaction information 50 (S140; determination step). The abnormal information generation function generates abnormal information 60 if either or both of the transaction information 30 or balance information 40 of audited company 3 are outside the normal transaction information 50 (S170; abnormal information generation step). The notification function notifies the audited company 3 of the abnormal information 60 (S180; notification step). Each of these functions is executed by the CPU 15 of the external audit device 2 (computer unit 10).

[0069] The blocking detection function detects the blocking of access between the audited company 3 and the trading financial institution 4, or the blocking of access between the audited company 3 and the external audit device 2 (S150; blocking detection step). The change detection function detects changes to the transaction information 30 after it has been acquired by the transaction information acquisition unit 110 (S160; change detection step). The machine learning function accumulates the transaction information 30 over a predetermined period and also accumulates the balance information 40 over a predetermined period to generate normal transaction information 50. Each of these functions is executed by the CPU 15 of the external audit device 2 (computer unit 10).

[0070] The computer program of the present invention described above may be recorded on a processor-readable recording medium, and as the recording medium, a "non-temporary tangible medium" such as tape, disk, card, semiconductor memory, or programmable logic circuit can be used.

[0071] The above computer program can be implemented using, for example, scripting languages ​​such as ActionScript and JavaScript®, object-oriented programming languages ​​such as Objective-C and Java®, and markup languages ​​such as HTML5. [Industrial applicability]

[0072] The external audit device and external audit system of the present invention enable continuous audits of a company's transactions and prompt notification of audit results to the company, allowing the company to quickly address fraudulent transactions and problematic accounting practices. Similar effects can be expected with external audit methods and external audit programs. [Explanation of Symbols]

[0073] 1. External audit system 2. External auditing device 3. Audited Company 4. Transaction financial institutions 5. Audit Firm 6. Internal Audit Officer 7. Certified Public Accountant 10. Computer Department 11. Input / Output Interfaces 12 ROM 13 RAM 14 Storage section 15 CPU 30 Transaction Information 40 Balance Information 50 Regular Transaction Information 60 Abnormal information 110 Transaction Information Acquisition Department 120 Balance Information Acquisition Unit 130 Normal Transaction Information Generation Department 140 Judgment section 150 Interruption detection unit 160 Change detection unit 170 Abnormal information generation section 180 Reporting Department 190 Machine Learning Department

Claims

1. An external monitoring device connected via a network line to the audited company, the audited company's financial institutions, and the auditing firm contracted by the audited company, A transaction information acquisition unit that acquires transaction information from the audited company, including the audited company's settlement records, inventory records, order records, profit and loss statement records, and balance sheet records. A balance information acquisition unit that acquires balance information of the audited company from the aforementioned financial institution, A normal transaction information generation unit accumulates the aforementioned transaction information and balance information over a predetermined period and generates normal transaction information that allows the audited company to understand the content of its business activities and the amount of its cash inflows and outflows by linking the audited company's business activities, such as the sale of products or provision of services, the purchase or sale of assets, and the cash inflows and outflows associated with those activities. An abnormal information generation unit that generates abnormal information when either or both of the transaction information or balance information of the audited company deviate from the average of the audited company's normal transaction information under normal circumstances, The system includes a notification unit that notifies the audited company and the auditing firm of the abnormal information and displays the notification on a display device of the terminal of the person in charge of internal audits at the audited company or the certified public accountant of the auditing firm in charge of the audited company, The aforementioned normal transaction information generation unit includes a machine learning unit. The machine learning unit accumulates the transaction information and balance information over a predetermined period, estimates the transaction volume and balance from the audited company's transaction activities, and generates the normal transaction information. An external auditing device characterized by the following features.

2. The audited company, the transaction financial institution, and the external audit device are connected to each other by a network line, and the system includes a blockage detection unit that detects the blockage of access to the network line between the audited company and the transaction financial institution, or the blockage of access to the network line between the audited company and the external monitoring device. The external audit device according to claim 1, wherein the notification unit notifies the audited company and the auditing firm of the detected blocking of access as abnormal information.

3. The audited company, the transaction financial institution, and the external audit device are connected to each other by a network line, and the device includes a change detection unit that detects changes to the transaction information after it has been acquired by the transaction information acquisition unit via the network line without processing in accordance with the audited company's internal regulations and accounting rules, The external audit device according to claim 1 or 2, wherein the notification unit notifies the audited company and the auditing firm of the detected change in the transaction information as abnormal information.

4. The external audit device according to any one of claims 1 to 3, wherein the transaction information acquisition unit acquires transaction information of the audited company's subsidiary or equity-method affiliate if the audited company has such a subsidiary or equity-method affiliate.

5. An external audit method in an external monitoring device in which the audited company, the audited company's financial institutions, and the auditing firm contracted by the audited company are connected by a network line, The computer unit of the aforementioned external monitoring device is: A transaction information acquisition step of acquiring transaction information from the audited company, including the audited company's settlement records, inventory records, order records, profit and loss statement records, and balance sheet records. A balance information acquisition step, in which balance information of the audited company is obtained from the aforementioned financial institution, A normal transaction information generation step involves accumulating the aforementioned transaction information and balance information over a predetermined period to generate normal transaction information that allows for the understanding of the content of the audited company's business activities and the amount of cash inflows and outflows by linking the audited company's business activities, such as the sale of products or provision of services, and the purchase or sale of assets, with the cash inflows and outflows associated with those activities. An abnormal information generation step that generates abnormal information when either or both of the transaction information or balance information of the audited company deviate from the average of the audited company's normal transaction information under normal circumstances, The notification step involves notifying the audited company and the auditing firm of the abnormal information and displaying the notification on the display device of the terminal of the person in charge of internal audits at the audited company or the certified public accountant of the auditing firm in charge of the audited company. The aforementioned normal transaction information generation step includes a machine learning step. The machine learning step involves accumulating the transaction information and balance information over a predetermined period, estimating the transaction volume and balance from the audited company's transaction activities, and generating the normal transaction information. An external audit method characterized by the following features.

6. An external audit program in an external monitoring device connected via a network line to the audited company, the audited company's financial institutions, and the auditing firm contracted by the audited company, The computer unit of the external monitoring device, A transaction information acquisition function that acquires transaction information from the audited company, including the audited company's settlement records, inventory records, order records, profit and loss statement records, and balance sheet records. A balance information acquisition function that acquires balance information of the audited company from the aforementioned financial institution, A normal transaction information generation function that accumulates the aforementioned transaction information and balance information over a predetermined period and links the audited company's business activities, such as the sale of products or provision of services, the purchase or sale of assets, and the inflow and outflow of money associated with those activities, to generate normal transaction information that allows for an understanding of the audited company's business activities and the amount of money flowing in and out. An abnormal information generation function that generates abnormal information when either or both of the audited company's transaction information or balance information deviate from the average of the audited company's normal transaction information under normal circumstances, The system provides a notification function that notifies the audited company and the auditing firm of the aforementioned abnormal information and displays the notification on the display device of the terminal of the internal audit officer of the audited company or the certified public accountant of the auditing firm in charge of the audited company. The aforementioned normal transaction information generation function is equipped with machine learning capabilities. The machine learning function accumulates the transaction information and balance information over a predetermined period, estimates the transaction volume and balance from the audited company's transaction activities, and generates the normal transaction information. An external audit program characterized by the following features.

7. An external monitoring system comprising an audited company, the audited company's financial institution, and an auditing firm contracted by the audited company, all connected via a network line to the audited company's external monitoring device, The aforementioned external monitoring device, A transaction information acquisition unit that acquires transaction information of the audited company from the audited company, A balance information acquisition unit that acquires balance information from the aforementioned financial institution, including settlement records, inventory records, order records, profit and loss statement records, and balance sheet records of the audited company. A normal transaction information generation unit accumulates the aforementioned transaction information and balance information over a predetermined period and generates normal transaction information that allows the audited company to understand the content of its business activities and the amount of its cash inflows and outflows by linking the audited company's business activities, such as the sale of products or provision of services, the purchase or sale of assets, and the cash inflows and outflows associated with those activities. An abnormal information generation unit that generates abnormal information when either or both of the transaction information or balance information of the audited company deviate from the average of the audited company's normal transaction information under normal circumstances, The system includes a notification unit that notifies the audited company and the auditing firm of the abnormal information and displays the notification on a display device of the terminal of the person in charge of internal audits at the audited company or the certified public accountant of the auditing firm in charge of the audited company, The aforementioned normal transaction information generation unit includes a machine learning unit. The machine learning unit accumulates the transaction information and balance information over a predetermined period, estimates the transaction volume and balance from the audited company's transaction activities, and generates the normal transaction information. An external audit system characterized by the following features.

Images