JP7649235B2 - Purchasing fraud detection business support device, purchasing fraud detection business support method, and purchasing fraud detection business support program - Google Patents

Description

The present invention relates to a purchasing fraud detection business support device, a purchasing fraud detection business support method, and a purchasing fraud detection business support program.

Paragraph 0025 of Patent Document 1 discloses that "The abolition of paper invoices makes it unclear who is responsible, and conventional processing methods are extremely vulnerable to fraud caused by collusion between the supplier and the ordering company's personnel. The sales audit model of this invention replaces invoices with electromagnetic media, and in the event of fraud, not only does it make it possible to subject the supplier to civil liability, such as the return of ill-gotten gains, but it also makes it easier to pursue criminal liability for embezzlement, fraud, etc., providing an extremely powerful deterrent against fraud."

JP 2006-164255 A

As described in the above-mentioned Patent Document 1, it is important for a company to detect fraudulent acts (purchasing fraud) regarding purchasing records that are committed through collusion between its own staff and those of its business partners.

One form of purchasing fraud involves a company's own staff conspiring with a client's staff to record fictitious purchases, pay the amount to the client, and then later having the client's staff give the company's staff a kickback for the amount paid for the fictitious purchases.

This method has the characteristic that the purchase amount for the month in which fictitious purchases were made increases sharply compared to, for example, the purchase amount for the same month in the previous year. Therefore, in order to discover fictitious purchases, it is sufficient to compare the purchase amount for the month being confirmed with the purchase amount for the comparison month that corresponds to the same month in the previous year.

However, in the past, the task of tallying up monthly purchase amounts and comparing them with the purchase amounts for the same month of the previous year was done manually by staff, which meant that discovering purchase fraud took a lot of time and effort.

The present invention has been made in consideration of the above problems, and aims to provide a fraudulent purchasing detection support device, a fraudulent purchasing detection support method, and a fraudulent purchasing detection support program that can support fraudulent purchasing detection operations by automatically calculating the difference between the purchase amount during the confirmation period and the purchase amount during the comparison period, and detecting any abnormally large numerical values among the calculated differences.

In order to solve the above-mentioned problems and achieve the object, the purchasing fraud detection business support device of the present invention is capable of supporting the business of detecting fraud relating to purchasing performance by calculating a purchasing difference amount, which is the difference between the purchasing amount during the period to be confirmed and the purchasing amount during the period to be compared. The purchasing fraud detection business support device is equipped with a control unit and a memory unit, and the memory unit stores purchase accounting data including a purchase date, a business partner identification data, and a purchase amount, and the control unit calculates the purchase amount corresponding to the purchase date belonging to the period to be confirmed from the amount calculated for each business partner identification data, and calculates the purchase amount corresponding to the purchase date belonging to the period to be confirmed from the amount calculated for each business partner identification data. The system is characterized by comprising: a purchase difference amount calculation means for calculating the purchase difference amount for each combination of the confirmation target period and the customer identification data by subtracting the purchase amount corresponding to the purchase date belonging to the period from the total amount for each customer identification data; an abnormal value detection means for detecting, according to a predetermined statistical method, a purchase difference amount calculated by the purchase difference amount calculation means that exceeds an upper limit as an abnormal value; and an abnormality display means for displaying the purchase difference amount detected as an abnormal value by the abnormal value detection means and the combination of the confirmation target period and the customer identification data corresponding to the detected purchase difference amount.

In addition, in the purchasing fraud detection business support device according to the present invention, the anomaly display means also displays the specified statistical method and the upper limit value.

The device for supporting the detection of purchasing fraud according to the present invention is characterized in that the predetermined statistical method is a method using the interquartile range, and the upper limit value is a value calculated by adding the third quartile to a value 1.5 times the interquartile range.

In addition, in the purchasing fraud detection business support device according to the present invention, the storage unit further stores purchase difference amount calculation result data including the amount obtained by aggregating the purchase amount in the purchase recording data for each period and the customer identification data, and the period and the customer identification data, and the control unit further includes a customer-specific graph display means for generating and displaying a line graph showing the progress of the aggregated amount for each period based on the purchase difference amount calculation result data, the number of which corresponds to the number of customer identification data, and the line graph has the aggregated amount in the purchase difference amount calculation result data as the vertical axis and the period in the purchase difference amount calculation result data as the horizontal axis.

In addition, in the purchasing fraud detection business support device according to the present invention, the storage unit further stores the purchase amount in the purchase recording data, which is calculated by aggregating the purchase amount for each period and the customer identification data, and the purchase difference amount calculation result data, which includes the period and the customer identification data. The control unit further stores, based on the purchase difference amount calculation result data, a line graph showing the progress of the aggregated amount for each period in the confirmation target year for the customer identification data corresponding to the purchase difference amount detected by the abnormal value detection means, and The system further includes a graph display means for comparing two years, which generates and displays a line graph showing the trend of the aggregated amount for each period in the comparison year for the supplier identification data corresponding to the purchase difference amount, the year to be confirmed is the year to which the confirmation period corresponding to the purchase difference amount detected as an abnormal value by the abnormal value detection means belongs, the comparison year is the year immediately preceding the confirmation year, and the line graph has the aggregated amount in the purchase difference amount calculation result data as the vertical axis and the month without a year displayed as the horizontal axis.

In addition, in the purchasing fraud detection business support device according to the present invention, the purchase accounting data further includes at least one of the aggregation units of business establishment identification data, department identification data, and person in charge identification data, and the storage unit further stores purchase amount aggregation data including the amount obtained by aggregating the purchase amount in the purchase accounting data for each combination of the period, the customer identification data, and the aggregation unit, the period, the customer identification data, and the aggregation unit, and the control unit further includes a graph display means for each aggregation unit that generates and displays a line graph showing the progress of the aggregated amount for each period for the customer identification data corresponding to the purchase difference amount detected by the abnormal value detection means based on the purchase amount aggregation data, the line graph having the aggregated amount in the purchase amount aggregation data as the vertical axis and the period in the purchase amount aggregation data as the horizontal axis.

In addition, in the purchasing fraud detection business support device according to the present invention, the purchasing accounting data further includes at least one of the aggregation units of business establishment identification data, department identification data, and person in charge identification data, and for purchases made via an order by registering order accounting data, an order number, and the memory unit stores the purchase amounts in the purchasing accounting data, aggregated for each combination of the period, the customer identification data, the aggregation unit, and the presence or absence of the order number, the period, the customer identification data, the aggregation unit, an immediate purchase classification which is a classification when the purchase amounts without the order number are aggregated, and the purchase amounts with the order number are aggregated. The control unit further stores aggregated data by purchasing method including an order route category, which is a category when the difference between the purchase amount and the purchase amount is calculated, and the control unit further includes a graph display means for generating and displaying a bar graph for the number of aggregation units, which shows the aggregated amount for the period corresponding to the purchase difference amount detected by the abnormal value detection means and the trading partner identification data corresponding to the purchase difference amount detected by the abnormal value detection means, based on the aggregated data by purchasing method, and the aggregated amount corresponding to the immediate purchase category and the aggregated amount corresponding to the order route category are distinguishable in the bar graph.

In addition, in the purchasing fraud detection business support device according to the present invention, the storage unit further includes sales accounting data including a sales date and a sales amount, a purchase summary record including an amount obtained by aggregating the purchase amount in the purchase accounting data for each combination of a period and the customer identification data, the period in the purchase accounting data, and the customer identification data in the purchase accounting data, and sales and purchase summary data including a sales summary record including an amount obtained by aggregating the sales amount in the sales accounting data for each period and the period in the sales accounting data. The control unit further includes a sales/purchase comparison graph display unit that generates a line graph showing the progress of the aggregated amount for each period for the supplier identification data corresponding to the purchase difference amount detected by the abnormal value detection unit based on the purchase aggregate record, generates a line graph showing the aggregated amount for each period based on the sales aggregate record, and displays both graphs together, and the line graph has the aggregated amount in the record as the vertical axis and the period in the record as the horizontal axis.

In addition, in the purchasing fraud detection business support device according to the present invention, the storage unit further includes order/purchase-specific summary data including order posting data including an order date and an order amount, a purchase summary record including an amount obtained by summarizing the purchase amount in the purchase posting data for each combination of a period and the customer identification data, the period in the purchase posting data, and the customer identification data in the purchase posting data, and an order/purchase summary record including an amount obtained by summarizing the order amount in the order posting data for each period and the period in the order posting data. The control unit further includes an order/purchase comparison graph display unit that generates a line graph showing the progress of the aggregated amount for each period for the supplier identification data corresponding to the purchase difference amount detected by the abnormal value detection unit based on the purchase aggregate record, generates a line graph showing the aggregated amount for each period based on the order aggregate record, and displays both graphs together, and the line graph has the aggregated amount in the record as the vertical axis and the period in the record as the horizontal axis.

The purchasing fraud detection business support device according to the present invention is characterized in that the confirmation period is a certain month of a certain year, and the comparison period is the certain month of the year prior to the certain year.

In addition, the purchasing fraud detection business support method according to the present invention is a method for supporting the detection of purchasing fraud, which is executed by an information processing device having a control unit and a memory unit, and which can support the business of detecting fraud in purchasing records by calculating a purchasing difference amount, which is the difference between the purchasing amount during the period to be confirmed and the purchasing amount during the period to be compared. The memory unit stores purchase accounting data including a purchase date, customer identification data, and a purchase amount, and the control unit calculates the purchasing amount corresponding to the purchase date belonging to the period to be confirmed from the amount calculated for each customer identification data, and calculates the purchasing amount corresponding to the purchase date belonging to the period to be compared. The method includes a purchase difference amount calculation step for calculating the purchase difference amount for each combination of the confirmation period and the customer identification data by subtracting the purchase amount corresponding to the purchase date from the amount tallied for each customer identification data, an abnormal value detection step for detecting, according to a predetermined statistical method, an abnormal value that exceeds an upper limit among the purchase difference amounts calculated in the purchase difference amount calculation step, and an abnormality display step for displaying the purchase difference amount detected as an abnormal value in the abnormal value detection step and the combination of the confirmation period and the customer identification data corresponding to the detected purchase difference amount.

In addition, the purchasing fraud detection business support program according to the present invention is a purchasing fraud detection business support program executed by an information processing device having a control unit and a memory unit, which can support the business of detecting fraud in purchasing records by calculating a purchasing difference amount, which is the difference between the purchasing amount during the period to be confirmed and the purchasing amount during the period to be compared. The memory unit stores purchase accounting data including a purchase date, customer identification data, and a purchase amount, and the control unit executes the purchasing fraud detection business support program by calculating the purchase difference amount, which is the difference between the purchasing amount during the period to be confirmed and the purchasing amount during the period to be compared. The method includes a purchase difference amount calculation step for calculating the purchase difference amount for each combination of the confirmation period and the customer identification data by subtracting the purchase amount corresponding to the purchase date from the amount aggregated for each customer identification data, an abnormal value detection step for detecting, according to a predetermined statistical method, the purchase difference amount calculated in the purchase difference amount calculation step that exceeds an upper limit as an abnormal value, and an abnormality display step for displaying the purchase difference amount detected as an abnormal value in the abnormal value detection step and the combination of the confirmation period and the customer identification data corresponding to the detected purchase difference amount.

The present invention has the effect of supporting the detection of purchasing fraud by automatically calculating the difference between the purchase amount during the period to be confirmed and the purchase amount during the period to be compared, and automatically detecting any abnormally large numerical values among the calculated differences.

FIG. 1 is a block diagram showing an example of the configuration of a purchasing fraud detection operation support device. FIG. 2 is a diagram showing an example of fraud (fictitious purchase) regarding purchase records. FIG. 3 is a diagram showing an example of normal purchase records and abnormal purchase records. FIG. 4 is a diagram showing an example of the initial analysis screen. FIG. 5 is a diagram showing an example of the analysis screen. FIG. 6 is a diagram showing an example of a purchase data registration screen and purchase accounting data in the case of immediate purchase. FIG. 7 is a diagram showing an example of an order data registration screen and order accounting data when purchasing via an order, and a purchase data registration screen and purchase accounting data when purchasing via an order. FIG. 8 is a diagram showing an example of acquisition range condition data for automatic detection execution schedule data and purchase accounting data. FIG. 9 is a diagram showing an example of automatic detection execution schedule data and parameters used for acquiring the data. FIG. 10 is a diagram showing an example of acquisition range condition data for purchase accounting data and parameters used for acquiring the data. FIG. 11 is a diagram showing an example of parameters used for acquiring data from the purchase accounting data. FIG. 12 is a diagram showing an example of the purchase appropriation data and the purchase difference amount calculation result data. FIG. 13 is a diagram showing an example of abnormal value detection using the interquartile range. FIG. 14 is a diagram showing an example of abnormality determination result data, abnormality determination result message data, and abnormality determination result message detail data. FIG. 15 is a diagram showing an example of parameters used to acquire the abnormality determination result message data. FIG. 16 is a diagram showing an example of a summary message displayed on the initial analysis screen. FIG. 17 is a diagram showing an example of a set of base dates on the initial analysis screen. FIG. 18 is a diagram showing an example of a detailed message displayed on the analysis screen. FIG. 19 is a diagram showing an example of an output area of a graph. FIG. 20 is a diagram showing an example of analysis data acquisition range conditions. FIG. 21 is a diagram showing an example of extraction condition initial values. FIG. 22 is a diagram showing an example of a set of extraction conditions on the analysis screen. FIG. 23 is a diagram showing an example in which the aggregation unit "person in charge" and the aggregation target "sales" are specified as parameters used to acquire data that is the basis for generating graphs 1 and 2. FIG. 24 is a diagram showing an example of purchase appropriation data and purchase difference amount calculation result data. FIG. 25 is a diagram showing an example in which the aggregation unit "person in charge" and the aggregation target "sales" are specified as parameters used to acquire data that is the basis for generating graph 3. FIG. 26 is a diagram showing an example of purchase accounting data and purchase amount summary data. FIG. 27 is a diagram showing an example in which the aggregation unit "person in charge" and the aggregation target "sales" are specified as parameters used to acquire data that is the basis for generating graph 4. FIG. 28 is a diagram showing an example of order accounting data, purchase accounting data, and total data by purchase method. FIG. 29 is a diagram showing an example in which the aggregation unit "person in charge" and the aggregation target "sales" are specified as parameters used to acquire data that is the basis for generating graph 5. FIG. 30 is a diagram showing an example of sales recorded data, purchase recorded data, and sales/purchase aggregated data. FIG. 31 is a diagram showing an example of graph 1 in the case where the aggregation unit is "person in charge" and the aggregation target is "sales." FIG. 32 is a diagram showing an example of graph 2 when the aggregation unit is "person in charge" and the aggregation target is "sales." FIG. 33 is a diagram showing an example of graph 3 when the aggregation unit is "person in charge" and the aggregation target is "sales." FIG. 34 is a diagram showing an example of graph 4 when the aggregation unit is "person in charge" and the aggregation target is "sales." FIG. 35 is a diagram showing an example of graph 5 in the case where the aggregation unit is "person in charge" and the aggregation target is "sales." FIG. 36 is a diagram showing an example of a list of graphs 1 to 5 when the tally unit is "person in charge" and the tally target is "sales." FIG. 37 is a diagram showing an example of coloring parameters that need to be set when highlighting a graph by coloring. FIG. 38 is a diagram showing an example in which the aggregation unit "department" and the aggregation target "sales" are specified as parameters used to acquire data that is the basis for generating graphs 1 and 2. FIG. 39 is a diagram showing an example in which the aggregation unit "department" and the aggregation target "sales" are specified as parameters used to acquire data that is the basis for generating Graph 3. FIG. 40 is a diagram showing an example of purchase accounting data and purchase amount summary data. FIG. 41 is a diagram showing an example in which the aggregation unit "department" and the aggregation target "sales" are specified as parameters used to acquire data that is the basis for generating graph 4. FIG. 42 is a diagram showing an example of order accounting data, purchase accounting data, and total data by purchase method. FIG. 43 is a diagram showing an example of graph 3 when the aggregation unit is "department" and the aggregation target is "sales." FIG. 44 is a diagram showing an example of graph 4 when the aggregation unit is "department" and the aggregation target is "sales." FIG. 45 is a diagram showing an example of a list of graphs 1 to 5 when the aggregation unit is "department" and the aggregation target is "sales." FIG. 46 is a diagram showing an example of coloring parameters that need to be set when highlighting a graph by coloring. FIG. 47 is a diagram showing a comparison between an example of graph 3 when the tallying unit is "person in charge" and an example of graph 3 when the tallying unit is "department." FIG. 48 is a diagram showing, in comparison, an example of graph 4 when the tallying unit is "person in charge" and an example of graph 4 when the tallying unit is "department." FIG. 49 is a diagram showing an example in which the aggregation unit "department" and the aggregation target "orders" are specified as parameters used to acquire data that is the basis for generating graphs 1 and 2. FIG. 50 is a diagram showing an example in which the aggregation unit "department" and the aggregation target "orders" are specified as parameters used to acquire data that is the basis for generating graph 5. FIG. 51 is a diagram showing an example of order entry data, purchase entry data, and order/purchase specific aggregate data. FIG. 52 is a diagram showing an example of graph 5 when the aggregation unit is "department" and the aggregation target is "orders." FIG. 53 is a diagram showing an example of a list of graphs 1 to 5 when the aggregation unit is "department" and the aggregation target is "orders." FIG. 54 is a diagram showing an example of graph 5 when the aggregation target is "sales." FIG. 55 is a diagram showing an example of graph 5 when the aggregation target is "orders."

Below, embodiments of the purchasing fraud detection business support device, purchasing fraud detection business support method, and purchasing fraud detection business support program according to the present invention will be described in detail with reference to the drawings. Note that the present invention is not limited to these embodiments.

[1. Overview]
In this section, the background, problems, overview and effects of the present invention will be explained in order.

[1-1. Background]
In recent years, corporate fraud and scandals such as embezzlement by employees and executives (hereinafter collectively referred to as "fraud") have been on the rise. In addition, due to the impact of COVID-19, such fraud is on the rise. As a result, fraud is exceeding the range of what can be detected by humans, and there is a demand for a system that can detect and deal with it early using non-human means.

There are various types of fraud. The first type is fraud related to sales records. This type of fraud inflates sales figures by falsifying sales figures. The second type is fraud related to purchasing records. This type of fraud involves collusion with business partners through fictitious payments. The third type is fraud related to inventory. This type of fraud involves profit manipulation, such as by inflating end-of-period inventory values. This embodiment deals with the second type of fraud, fraud related to purchasing records (hereinafter referred to as "purchasing fraud").

As shown in Figure 2, purchasing fraud is often committed by conspiring with a business partner's staff to make fictitious purchases. The specific flow is as follows: first, as shown in Figure 2 (1), a fictitious transaction is created between one's own company's staff and a business partner's staff to align their understanding. Next, as shown in Figure 2 (2), one's own company's staff records fictitious purchasing data, and as shown in Figure 2 (3), one's own company makes a payment to the business partner. Then, after payment, as shown in Figure 2 (4), the business partner's staff gives a kickback to one's own company's staff for the amount paid for the fictitious transaction.

Purchasing fraud has the following characteristics, for example:
1. Purchase amounts are concentrated in certain business partners.
2. For the business partner who engaged in fraudulent transactions, show an increase in purchase amounts that are clearly unnatural compared to past transactions.
3. There is a combination of business partners and personnel with biased purchasing records.

Here, in most cases, purchasing transactions are basically constant. For this reason, the reasons for an increase in purchasing transactions can be narrowed down to the following, for example. First, an increase in purchasing due to strong sales and increased demand. In this case, the timing of the increase in purchasing is the beginning and end of the term. Second, an increase in purchasing due to the need to secure inventory with the addition of new transactions. In this case, the timing of the increase in purchasing is the beginning and end of the term. Third, an increase in purchasing due to the need to secure inventory with the arrival of a selling season. In this case, the timing of the increase in purchasing is the end of each quarter during the term. The increase in purchasing explained in this paragraph is, so to speak, a normal increase in purchasing that is not due to fraud.

In contrast, when purchasing increases are due to fraud, as shown in Figure 3, the purchase amount often increases sharply compared to the same month of the previous year. Specifically, in the example of supplier A in Figure 3, since no purchasing fraud occurred, there is no significant difference in the purchase amount when comparing the same month of the current year and last year (the difference is at most 1,000 yen). In contrast, in the example of supplier B in Figure 3, there is a difference of 19,500 yen in the purchase amount between August 2021 (purchase amount: 30,000 yen) of the current year when purchasing fraud occurred and August 2020 (purchase amount: 10,500 yen) of the previous year. From this, it can be inferred that there is a high possibility that purchasing fraud occurred in August 2021.

[1-2. Issues]
However, there are problems (1) and (2) below regarding the manual detection of purchasing fraud by personnel in charge.

(1) First, there was a problem with the tabulation process using multiple combinations. That is, in order to discover purchasing fraud, it is first necessary to tabulate the monthly purchase amounts for each tabulation unit for all business partners. However, since there are various possible tabulation units, as explained below, there was a problem that the tabulation took time and tabulation errors were likely to occur. Examples of the tabulation units include "business partner unit," which is used when simply checking the month in which the purchase amount increased abnormally, "(business partner x person in charge) unit," which is used when checking whether there is a concentration of transactions between business partners and a specific person in charge, and "(business partner x organization) unit," which is used when checking whether there is a concentration of transactions between business partners and a specific organization.

(2) There was also a problem with the comparison work that was carried out after the tabulation work. In other words, visually comparing the purchase amounts for a certain month of the current year with the corresponding month of the previous year was time-consuming and prone to overlooking.

[1-3. Overview and Effects]
In this embodiment, in order to solve the above problems (1) and (2), for example, the following function (1) is implemented. In addition, the following functions (2) to (5) are implemented as secondary functions.

(1) First, we have made it possible to automatically detect business partners whose purchase amounts, tallied for each month, have increased dramatically compared to the same month of the previous year. This has made it possible to perform the above-mentioned tabulation and comparison tasks easily, accurately, and quickly.

(2) In addition, the monthly purchase amounts for the fiscal year including the month in which the anomaly was detected and the monthly purchase amounts for the fiscal year immediately prior to that year can be visualized as a graph (Graph 2 described below). This makes it possible to analyze whether the purchase results for the month in which the anomaly was detected show an unnatural increase compared to the purchase results for the same month in the previous fiscal year, and whether the purchase results for the month in which the anomaly was detected are based on abnormal transactions that have never been recorded in the past.

(3) For business partners for which an abnormality has been detected, the actual purchase amounts for each business establishment, department, or person who recorded the purchases can be visualized in a graph (Graph 3 described below). This makes it possible to determine, for example, whether the fraudulent purchases occurred at the organizational level, or at the individual level of a person in charge. In the case of fraudulent purchases at the organizational level, it is further possible to determine whether the fraudulent purchases occurred at the business establishment level, or at the department level. It is also possible to determine which business establishment, department, or person in charge committed the fraudulent purchases.

(4) Furthermore, the purchase amount can now be visualized as a graph (Graph 4 described below) by dividing it into two types, "immediate purchase" and "purchase via order," for each business establishment, department, or person in charge who recorded the purchase. This makes it possible to check whether there is a concentration of purchase amounts recorded using "immediate purchase," which is difficult to check.

Here, we will explain purchasing via order and immediate purchasing. Purchasing via order is used when all steps from ordering, receiving, and purchasing are registered in the data. Generally, more companies use purchasing via order. With purchasing via order, the transaction details are checked before purchasing, such as checking the purchase order when ordering, and the receipt confirmation when receiving, so it is difficult to commit purchasing fraud. In contrast, immediate purchasing is used when ordering, receiving, and purchasing have already been completed, and only the purchasing results are registered in the data. Immediate purchasing is often used by companies that purchase mechanically based on certain transactions. With immediate purchasing, the transaction details are not checked, so it is easy to commit purchasing fraud.

(5) Furthermore, the monthly occurrence status of sales records or order records and purchase records can be visualized as a graph (Graph 5 described below). Here, in the three normal patterns of increased purchase transactions described in [1-1. Background], the purchase records increase as the sales records or order records increase, but if purchase fraud occurs, the purchase records increase even though the sales records or order records do not increase. According to the graph described in this paragraph (Graph 5 described below), it is possible to check whether such unnatural increases in purchase records are occurring. The specific configuration and operation are described below.

2. Configuration
An example of the configuration of the purchasing fraud detection operation support device 100 according to this embodiment will be described with reference to Fig. 1. Fig. 1 is a block diagram showing an example of the configuration of the purchasing fraud detection operation support device 100.

The purchasing fraud detection business support device 100 is a commercially available desktop personal computer. Note that the purchasing fraud detection business support device 100 is not limited to a stationary information processing device such as a desktop personal computer, but may be a portable information processing device such as a commercially available notebook personal computer, PDA (Personal Digital Assistant), smartphone, or tablet personal computer.

The purchasing fraud detection business support device 100 includes a control unit 102, a communication interface unit 104, a memory unit 106, and an input/output interface unit 108. Each unit included in the purchasing fraud detection business support device 100 is connected to each other so that they can communicate with each other via any communication path.

The communication interface unit 104 communicatively connects the purchasing fraud detection business support device 100 to the network 300 via a communication device such as a router and a wired or wireless communication line such as a dedicated line. The communication interface unit 104 has a function of communicating data with other devices via the communication line. Here, the network 300 has a function of connecting the purchasing fraud detection business support device 100 and the server 200 so that they can communicate with each other, and is, for example, the Internet or a LAN (Local Area Network). Note that data such as various masters described below may be stored in the server 200, for example.

An input device 112 and an output device 114 are connected to the input/output interface unit 108. The output device 114 may be a monitor (including a home television), a speaker, or a printer. The input device 112 may be a keyboard, a mouse, a microphone, or a monitor that works with a mouse to provide a pointing device function. In the following, the output device 114 may be referred to as the monitor 114, and the input device 112 may be referred to as the keyboard 112 or the mouse 112.

The storage unit 106 stores various databases, tables, files, and the like. The storage unit 106 records computer programs that work in cooperation with the OS (Operating System) to give instructions to the CPU (Central Processing Unit) to perform various processes. The storage unit 106 can be, for example, a memory device such as a RAM (Random Access Memory) or a ROM (Read Only Memory), a fixed disk device such as a hard disk, a flexible disk, an optical disk, or the like.

The memory unit 106 includes, for example, automatic detection execution schedule data 106a, acquisition range condition data 106b, purchase accounting data 106c, order accounting data 106d, sales accounting data 106e, order accounting data 106f, purchase difference amount calculation result data 106g, purchase amount summary data 106h, summary data by purchase method 106i, summary data by sales/purchase 106j, summary data by order/purchase 106k, abnormality determination result data 106m, abnormality determination result message data 106n, and abnormality determination result message detail data 106p.

Here, we will provide definitions of the terms used in this embodiment and an overview of each piece of data contained in the memory unit 106.

In this embodiment, the purchase difference amount is the difference between the purchase amount during the period to be confirmed and the purchase amount during the period to be compared. The period to be confirmed is, for example, a certain month in a certain year, and the period to be compared is, for example, the certain month in the year prior to the certain year (i.e., the same month of the previous year).

In this embodiment, it is assumed that the business partner and the payee are the same. For this reason, in the following explanation, the "business partner" may be referred to as the "payee."

The automatic detection execution schedule data 106a and the acquisition range condition data 106b belong to the "data for performing anomaly detection" and are pre-configured data for performing anomaly detection.

Purchase accounting data 106c, order accounting data 106d, sales accounting data 106e, and order accounting data 106f belong to "business data" and are assumed data accumulated during business operations.

The abnormality determination result data 106m, the abnormality determination result message data 106n, and the abnormality determination result message detail data 106p belong to the "abnormality determination result data" and are data that store the determination results of the abnormality detection execution.

The contents of each piece of data contained in the memory unit 106 are explained in detail below.

As shown in FIG. 8, the automatic detection execution schedule data 106a includes, for example, a detection ID, a schedule ID, execution conditions, and execution time.

The acquisition range condition data 106b is, for example, data for setting the range when acquiring data from the purchase accounting data 106c. As shown in FIG. 8, the acquisition range condition data 106b includes, for example, the detection ID, the schedule ID, the target column, the FROM condition, the TO condition, and the like.

The purchase accounting data 106c is data that accumulates daily purchase accounting. As shown in FIG. 24, the purchase accounting data 106c includes, for example, a purchase number, a purchase date, an order number, business establishment identification data (business establishment), department identification data (department), person in charge identification data (person in charge), business partner identification data (payee), and a purchase amount. The business establishment identification data (business establishment), department identification data (department), and person in charge identification data (person in charge) are the units of calculation when the purchase amount is calculated.

Here, we will explain how to register purchase accounting data 106c, separately for immediate purchase and purchase via order. The definitions of immediate purchase and purchase via order are as explained in (4) of [1. Overview], [1-3. Overview and Effects].

In the case of immediate purchase, as shown in FIG. 6, purchase accounting data 106c is registered on the purchase data registration screen without entering an order number. Therefore, the generated purchase accounting data 106c corresponding to immediate purchase does not hold an order number, as shown in FIG. 6.

When purchasing via order, first, order accounting data 106d is registered from the order data registration screen, as shown in FIG. 7. Then, when the ordered goods arrive and arrival confirmation is completed, purchase accounting data 106c is registered with the order number (H001) entered from the purchase data registration screen, as shown in FIG. 7. Therefore, the generated purchase accounting data 106c corresponding to purchase via order holds the order number (H001), as shown in FIG. 7.

In this way, if a record in purchase accounting data 106c does not hold an order number, the purchase identified in that record is an immediate purchase. In contrast, if a record in purchase accounting data 106c holds an order number, the purchase identified in that record is a purchase via an order. In other words, it is possible to distinguish between immediate purchases and purchases via an order depending on whether or not an order number is present in a record in purchase accounting data 106c. The information on whether a purchase is immediate or via an order is related to the generation of Graph 4, which will be described later.

Returning to the explanation of the data contained in the memory unit 106, the order accounting data 106d is data that accumulates daily order accounting. As shown in FIG. 28, the order accounting data 106d includes, for example, the order number, the order date, the business establishment identification data (business establishment), the department identification data (department), the person in charge identification data (person in charge), the business partner identification data (payee), the order amount, and the like.

The sales accounting data 106e is data that accumulates daily sales accounting. As shown in FIG. 30, the sales accounting data 106e includes, for example, a sales number, a sales date, the business establishment identification data (business establishment), the department identification data (department), the person in charge identification data (person in charge), and a sales amount.

The order entry data 106f is data that accumulates daily order entries. As shown in FIG. 51, the order entry data 106f includes, for example, the order number, the order date, the business establishment identification data (business establishment), the department identification data (department), the person in charge identification data (person in charge), the order amount, and the like.

The purchase difference amount calculation result data 106g is data including the result of calculating the purchase difference amount by referring to the purchase accounting data 106c. As shown in FIG. 24, etc., the purchase difference amount calculation result data 106g includes, for example, the period (fiscal year and month), the business partner identification data (payee), the purchase amount in the purchase accounting data 106c totaled for each period (fiscal year and month) and business partner identification data (payee) (monthly purchase amount total and monthly purchase amount total for the same month of the previous year), and the purchase difference amount (monthly purchase difference amount total).

The purchase amount summary data 106h includes the result of summarizing the purchase amount according to the aggregation unit with reference to the purchase accounting data 106c. As shown in FIG. 26 and FIG. 40, the purchase amount summary data 106h includes, for example, the period (fiscal year and month), the business partner identification data (payee), the aggregation unit ("person in charge" in FIG. 26, and "department" in FIG. 40), and the amount (purchase amount) obtained by summarizing the purchase amount in the purchase accounting data 106c for each combination of the period (fiscal year and month), the business partner identification data (payee), and the aggregation unit.

The purchase method-specific aggregation data 106i includes the result of aggregating the purchase amount according to the purchase accounting method with reference to the purchase accounting data 106c. As shown in FIG. 28 and FIG. 42, the purchase method-specific aggregation data 106i includes, for example, the purchase accounting method (purchase method), the period (accounting year and month), the customer identification data (payee), the aggregation unit ("person in charge" in FIG. 28, and "department" in FIG. 42), and the amount (purchase amount) obtained by aggregating the purchase amount in the purchase accounting data 106c for each combination of the period (accounting year and month), the customer identification data (payee), the aggregation unit, and the presence or absence of the purchase order number. The purchase accounting method (purchase method) includes two types: an immediate purchase category (immediate purchase), which is a category when the purchase amount without the purchase order number is aggregated, and an order via category (purchase via order), which is a category when the purchase amount with the purchase order number is aggregated.

The sales and purchase aggregated data 106j is data that includes the results of aggregating the sales amount by referring to the sales apportionment data 106e and the results of aggregating the purchase amount by referring to the purchase apportionment data 106c. As shown in FIG. 30, the sales and purchase aggregated data 106j is made up of sales aggregate records that are records including the data classification "sales" indicating that the results are aggregated by referring to the sales apportionment data 106e, and purchase aggregate records that are records including the data classification "purchase" indicating that the results are aggregated by referring to the purchase apportionment data 106c.

The order/purchase summary data 106k includes the results of aggregating the order amount by referencing the order posting data 106f and the results of aggregating the purchase amount by referencing the purchase posting data 106c. As shown in FIG. 51, the order/purchase summary data 106k includes order summary records that are records including the data classification "order" indicating that the results are aggregated by referencing the order posting data 106f, and purchase summary records that are records including the data classification "purchase" indicating that the results are aggregated by referencing the purchase posting data 106c.

The contents of the sales summary record, order summary record, and purchase summary record are explained below.

As shown in FIG. 30, the sales summary record includes, for example, the data classification "Sales," the period (fiscal year and month), and the sales amount in the sales posting data 106e summarized for each period (fiscal year and month).

As shown in FIG. 51, the order summary record includes, for example, the data classification "Order," the period (fiscal year and month), and the order amount in the order posting data 106f totaled for each period (fiscal year and month).

As shown in Figures 30 and 51, the purchase summary record includes, for example, the data classification "Purchase," the period (fiscal year and month), the supplier identification data (payee), and the purchase amount in the purchase accounting data 106c, which is totaled for each combination of the period (fiscal year and month) and the supplier identification data (payee).

Returning to the explanation of the data contained in the memory unit 106, the abnormality determination result data 106m includes, for example, as shown in FIG. 14 etc., an abnormality detection ID (detection ID), which is the detection ID that detected an abnormality in the purchase difference amount (hereinafter referred to as "abnormality detection" in this paragraph), an abnormality detection JOB ID (JOBID), which is an ID for identifying the JOB in which the abnormality was detected, a message ID, which is an ID for identifying the message to be output when the abnormality is detected, the period (fiscal year and month) in which the abnormality was detected, the business partner identification data (payee) for the business partner in which the abnormality was detected, the abnormal purchase difference amount detected, and an upper limit value calculated using a specified statistical method.

As shown in FIG. 14, etc., the abnormality determination result message data 106n includes, for example, the abnormality detection ID (detection ID), the abnormality detection JOBID (JOBID), the message ID, a category (degree of abnormality) for identifying the presence or absence of an abnormality, a definition name specified by the abnormality detection JOBID, an overview, the detected content (detection target), etc. The detected content is the detected abnormal purchase difference amount and the corresponding combination of the period (fiscal year and month) and the customer identification data (payee).

As shown in FIG. 14, etc., the abnormality determination result message detail data 106p includes, for example, the abnormality detection ID (detection ID), the abnormality detection JOBID (JOBID), the message ID, the specified statistical method (detection method), a threshold value, a determination method, the upper limit value, etc.

The control unit 102 is a CPU or the like that performs overall control of the purchasing fraud detection business support device 100. The control unit 102 has an internal memory for storing control programs such as an OS, programs that define various processing procedures, required data, etc., and executes various information processing based on these stored programs.

The control unit 102 has, in terms of its functional concept, for example, (1) a purchase difference amount calculation unit 102a as a purchase difference amount calculation means that calculates the purchase difference amount for each combination of the confirmation target period and the trading partner identification data by subtracting the amount obtained by aggregating the purchase amount corresponding to the purchase date belonging to the comparison target period for each trading partner identification data from the amount obtained by aggregating the purchase amount corresponding to the purchase date belonging to the confirmation target period for each trading partner identification data, (2) an abnormal value detection unit 102b as an abnormal value detection means that detects, according to a predetermined statistical method, an abnormal value that exceeds an upper limit among the purchase difference amounts calculated by the purchase difference amount calculation means, and (3) a purchase difference amount detected as an abnormal value by the abnormal value detection means and a purchase difference amount for the confirmation target period and the trading partner corresponding to the detected purchase difference amount. (4) a supplier-specific graph display unit 102d as supplier-specific graph display means for generating and displaying line graphs showing the trend of the aggregated amount for each period based on the purchase difference amount calculation result data, the number of which corresponds to the number of supplier identification data; and (5) a two-year comparison graph display unit for generating and displaying a line graph showing the trend of the aggregated amount for each period in the confirmation target year for the supplier identification data corresponding to the purchase difference amount detected by the abnormal value detection means, and a line graph showing the trend of the aggregated amount for each period in the comparison target year for the supplier identification data corresponding to the purchase difference amount detected by the abnormal value detection means, based on the purchase difference amount calculation result data. (6) a graph display unit by aggregation unit 102f as a graph display unit by aggregation unit which generates and displays a line graph showing the trend of the aggregated amount for each period for the supplier identification data corresponding to the purchase difference amount detected by the abnormal value detection means based on the purchase amount aggregate data, the number of the aggregation units; (7) a graph display unit by purchase accounting method 102g as a graph display unit by purchase accounting method which generates and displays a bar graph showing the aggregated amount for the period corresponding to the purchase difference amount detected by the abnormal value detection means and the supplier identification data corresponding to the purchase difference amount detected by the abnormal value detection means, the number of the aggregation units, based on the purchase method aggregate data; and (8) a graph display unit by purchase accounting method which generates and displays a line graph showing the trend of the aggregated amount for each period for the supplier identification data corresponding to the purchase difference amount detected by the abnormal value detection means, the number of the aggregation units, based on the purchase aggregation data by purchase method. (9) a sales/purchase comparison graph display unit 102h as a sales/purchase comparison graph display means that generates a line graph showing the trend of the aggregated amount for each period for the supplier identification data corresponding to the purchase difference amount detected by the abnormal value detection means based on the purchase summary record, and displays both graphs together; and (10) an order/purchase comparison graph display unit 102i as an order/purchase comparison graph display means that generates a line graph showing the trend of the aggregated amount for each period for the supplier identification data corresponding to the purchase difference amount detected by the abnormal value detection means based on the purchase summary record, and generates a line graph showing the aggregated amount for each period based on the order summary record, and displays both graphs together.

The purchase difference amount calculation unit 102a refers to the purchase accounting data 106c (see FIG. 12) including the purchase date, the payee, and the purchase amount, and tally up the purchase amount corresponding to the purchase date belonging to the confirmation period (e.g., "2021/03" in the purchase accounting data 106c in FIG. 12) for each payee, while tallying up the purchase amount corresponding to the purchase date belonging to the comparison period (e.g., "2020/03" in the purchase accounting data 106c in FIG. 12) for each payee. The purchase difference amount calculation unit 102a then subtracts the aggregated latter amount (e.g., "5,000 yen" in the purchase accounting data 106c of FIG. 12) from the aggregated former amount (e.g., "15,000 yen" in the purchase accounting data 106c of FIG. 12) to calculate the purchase difference amount (e.g., "10,000 yen") for each combination of the confirmation period (e.g., "2021/03") and the payment recipient.

The abnormal value detection unit 102b detects, according to a predetermined statistical method, any of the purchase difference amounts calculated by the purchase difference amount calculation unit 102a that exceeds an upper limit value as an abnormal value. The predetermined statistical method is not particularly limited, but is, for example, a method using the interquartile range. When the interquartile range is used, the upper limit value is, for example, a value calculated by adding a value obtained by multiplying the interquartile range by 1.5 to the third quartile.

The abnormality display unit 102c displays the purchase difference amount detected as an abnormal value by the abnormal value detection unit 102b and the combination of the confirmation target period and the payment destination corresponding to the detected purchase difference amount. At this time, the abnormality display unit 102c may also display the specified statistical method and the upper limit value.

The graph display unit 102d for each business partner generates Graph 1 (described later) based on the purchase difference amount calculation result data 106g. That is, the graph display unit 102d for each business partner generates and displays line graphs showing the progress of the total monthly purchase amount for each fiscal year and month based on the purchase difference amount calculation result data 106g (see FIG. 24) including the fiscal year and month, the payee, and the total monthly purchase amount (see FIG. 31).

The vertical and horizontal axes of the line graph described in the previous paragraph are as follows:
Vertical axis: Monthly total of purchase amount in the purchase difference amount calculation result data 106g Horizontal axis: Fiscal year and month in the purchase difference amount calculation result data 106g

The graph display unit 102e for comparing two years generates Graph 2, which will be described later, based on the purchase difference amount calculation result data 106g. That is, the graph display unit 102e for comparing two years generates and displays a line graph (the graph of "Purchase Actual" in FIG. 32) showing the trend of the total monthly purchase amount for each fiscal year and month for the payee corresponding to the purchase difference amount detected by the abnormal value detection unit 102b, and a line graph (the graph of "Last Year's Purchase Actual" in FIG. 32) showing the trend of the total monthly purchase amount for each fiscal year and month for the payee corresponding to the purchase difference amount detected by the abnormal value detection unit 102b, based on the purchase difference amount calculation result data 106g (see FIG. 24) including the fiscal year and month, the payee, the monthly purchase amount, and the monthly purchase amount for the same month of the previous year.

The monthly total purchase amount is the amount in the year to be confirmed (e.g., "FY2020"), which is the year to which the confirmation period (e.g., "2021/03") corresponding to the purchase difference amount (e.g., "10,000 yen") detected as an abnormal value by the abnormal value detection unit 102b belongs, among the amounts obtained by aggregating the purchase amounts in the purchase accounting data 106c by fiscal year and month and by payee. On the other hand, the monthly total purchase amount for the same month of the previous year is the amount in the year prior to the year to be confirmed (e.g., "FY2019"), among the amounts obtained by aggregating the purchase amounts in the purchase accounting data 106c by fiscal year and month and by payee.

The vertical and horizontal axes of the line graph described in the previous paragraph are as follows:
Vertical axis: the monthly total of the purchase amount in the purchase difference amount calculation result data 106g or the monthly total of the purchase amount for the same month of the previous year. Horizontal axis: the month without the year displayed (the "year" is not displayed in order to display the graphs for both years).

The graph display unit by aggregation unit 102f generates graph 3 (described later) based on the purchase amount aggregation data 106h. That is, the graph display unit by aggregation unit 102f generates and displays line graphs showing the trend of the aggregated amount (purchase amount) for each fiscal year and month for the payee (e.g., "Payee A") corresponding to the purchase difference amount (e.g., "10,000 yen") detected as an abnormal value by the abnormal value detection unit 102b based on the purchase amount aggregation data 106h (see Figs. 26 and 40) including the fiscal year and month, the payee, the aggregation unit, and the aggregated amount (purchase amount) for the number of aggregation units (see Figs. 33 and 43). Fig. 33 is a graph when the aggregation unit is "person in charge". Fig. 43 is a graph when the aggregation unit is "department".

The vertical and horizontal axes of the line graph described in the previous paragraph are as follows:
Vertical axis: The total amount (purchase amount) in the purchase amount total data 106h
Horizontal axis: The accounting year and month in the purchase amount summary data 106h

The graph display unit 102g by purchasing accounting method generates graph 4, described below, based on the aggregated data by purchasing method 106i. That is, the purchase accounting method-specific graph display unit 102g generates and displays bar graphs showing the aggregated amount (purchase amount) for the fiscal year and month (e.g., "2021/03") corresponding to the purchase difference amount (e.g., "10,000 yen") detected as an abnormal value by the abnormal value detection unit 102b and the purchase difference amount (e.g., "Payee A") corresponding to the purchase difference amount (e.g., "10,000 yen") detected as an abnormal value by the abnormal value detection unit 102b based on the purchase method-specific aggregated data 106i (see Figures 28 and 42) including the purchase method ("immediate purchase" or "purchase via order"), the accounting year and month, the payee, the aggregation unit, and the aggregated amount (purchase amount) for the payment destination (e.g., "Payee A") corresponding to the purchase difference amount (e.g., "10,000 yen") detected as an abnormal value by the abnormal value detection unit 102b, for the number of aggregation units (see Figures 34 and 44). Figure 34 is a graph when the aggregation unit is "person in charge". Figure 44 shows a graph when the aggregation unit is "department."

As shown in Figures 34 and 44, the key point in this bar graph is that the tabulated amount corresponding to the immediate purchase category (the "immediate purchase" part in the bar graph) and the tabulated amount corresponding to the order via category (the "purchase via order" part in the bar graph) are distinguishable by different hatching.

The sales/purchase comparison graph display unit 102h generates graph 5, which will be described later, based on the sales/purchase-specific aggregated data 106j. That is, the sales/purchase comparison graph display unit 102h generates a line graph (the "purchase" graph in FIG. 35) showing the trend of the aggregated amount (purchase amount) for each accounting year/month for the payee (e.g., "Payee A") corresponding to the purchase difference amount (e.g., "10,000 yen") detected as an abnormal value by the abnormal value detection unit 102b, based on the purchase aggregate record (see FIG. 30) including the accounting year/month, the payee, and the aggregated amount (purchase amount). In addition, the sales/purchase comparison graph display unit 102h generates a line graph (the "sales" graph in FIG. 35) showing the trend of the aggregated amount (sales amount) for each accounting year/month, based on the sales aggregate record (see FIG. 30) including the accounting year/month and the aggregated amount (sales amount). The sales/purchase comparison graph display unit 102h then displays both generated graphs together (see Figure 35).

The vertical and horizontal axes of the line graph described in the previous paragraph are as follows:
Vertical axis: The aggregated amount (purchase amount) in the purchase aggregate record or the aggregated amount (sales amount) in the sales aggregate record
Horizontal axis: The accounting year and month in the purchase summary record or sales summary record

The order/purchase comparison graph display unit 102i generates graph 5, which will be described later, based on the order/purchase aggregate data 106k. That is, the order/purchase comparison graph display unit 102i generates a line graph (the "purchase" graph in FIG. 52) showing the trend of the aggregated amount (purchase amount) for each accounting month for the payee (e.g., "Payee A") corresponding to the purchase difference amount (e.g., "10,000 yen") detected as an abnormal value by the abnormal value detection unit 102b, based on the purchase aggregate record (see FIG. 51) including the accounting month, the payee, and the aggregated amount (purchase amount). In addition, the order/purchase comparison graph display unit 102i generates a line graph (the "order" graph in FIG. 52) showing the trend of the aggregated amount (order amount) for each accounting month, based on the order aggregate record (see FIG. 51) including the accounting month and the aggregated amount (order amount). The order/purchase comparison graph display unit 102i then displays both of the generated graphs together (see Figure 52).

The vertical and horizontal axes of the line graph described in the previous paragraph are as follows:
Vertical axis: The aggregated amount in the purchase aggregate record (purchase amount) or the aggregated amount in the order aggregate record (order amount)
Horizontal axis: The accounting year and month in the purchase summary record or the order summary record

[3. Processing Overview and Screen Configuration]
[3-1. Processing Overview]
In this embodiment, "purchase results by payee" is output. In addition, in this embodiment, for example, "purchase results by person in charge or organization recorded as payee" may be output. Information for which an abnormality is detected may be highlighted by changing the color or font size so that the abnormality can be recognized at a glance. The data to be detected is, for example, the difference between the purchase amount of the confirmation month for each combination of month and payee and the purchase amount of the comparison month (for example, the same month of the previous year) for each combination of month and payee. Payees for which the difference is abnormally large are detected.

[3-2. Screen configuration]
In this embodiment, the displayed screen changes in the following order: (1)→(2).

(1) Display of anomaly detection results First, as shown by A in the example screen of Fig. 4, the results of anomaly detection are displayed at the summary message level. The area shown by A in Fig. 4 is an area for displaying messages related to the anomaly detection process. Specifically, this area displays the timing at which the anomaly was detected, the payee, and the difference from the purchase amount for the same month of the previous year. Note that, if there are multiple pieces of information detected as anomalies, the multiple pieces of information are displayed vertically as shown by A in Fig. 4. This area only displays the anomaly detection results on a summary basis, so if you want to see the anomaly detection results on a detailed basis, you need to select the summary message and switch screens.

(2) Display of business data related to the detected abnormal information Next, detailed messages and graphs necessary for analyzing the detected abnormal information are displayed.

The area shown in the lower half of A in Figure 5 is an area that displays detailed messages regarding the anomaly detection results. This area displays information such as the "detection method used to detect the anomaly" and "key information of the data in which the anomaly was detected." If multiple pieces of information are detected as an anomaly, a summary message will be displayed for each anomaly, but a detailed message will only be displayed for the one selected from the displayed summary messages. After the detailed message is displayed, the following graphs B to F are displayed.

The graph shown in area B of Figure 5 is a graph that outputs the purchase amount for each payee. In other words, it is a graph for checking the "monthly purchase amount trends for each payee." This graph makes it possible to compare the purchase amount trends for payees detected as abnormal and normal payees.

The graph shown in area C of Figure 5 is a graph that outputs, for a specific payee, the purchase amount for the year in which the abnormality was detected and the purchase amount for the year prior to the year in which the abnormality was detected. In other words, this graph is used to compare and confirm the trends in the "purchase amount for the year in which abnormality detection is performed" and the "purchase amount for the year prior to the current year." This graph makes it possible to make a comparison with the purchase amount for the same month of the previous year.

The graph shown in area D of Figure 5 is a graph that outputs the purchase amount by employee who recorded purchases to a specific payee. In other words, it is a graph for checking "the trend in monthly purchase amounts by employee." This graph makes it possible to check whether purchase amounts are concentrated in a specific employee. Specifically, if a employee has colluded with a payee, it can be confirmed that the purchase amount for that employee is abnormally high.

The graph shown in area E of Figure 5 is a graph that outputs the purchase amount by "purchase accounting method" for each employee who recorded purchases to a specific payee. In other words, this graph is for checking the "purchase amount corresponding to immediate purchases" and the "purchase amount corresponding to purchases via order" for each employee within a single month. This graph makes it possible to check whether there are any employees who are recording a large amount of purchases as "immediate purchases," where it is difficult to check the transaction details until the purchase transaction is completed.

The graph shown in area F in Figure 5 is a graph that outputs the total sales amount and the purchase amount for a specific payee. In other words, it is a graph for checking the "monthly trends in total sales amount and purchase amount for a specific payee." With this graph, if there is an increase in purchases, it can be checked whether the increase in purchases is due to an increase in sales (i.e., whether it is a normal increase).

Area G in Figure 5 is an area for setting the extraction conditions for data extraction. In other words, this area is used when you want to display the results of data extraction with different conditions for the graphs displayed in areas B to F in Figure 5. Specifically, if you want to display a graph at the organizational level, "office" or "department" is specified, while if you want to display a graph at the person in charge level, "person in charge" is specified. Also, if you want to check the sales amount (= if you want to display a graph of the sales amount in the graph shown in area F in Figure 5), "sales" is specified, while if you want to check the sales amount that will occur in the future as the order amount (= if you want to display a graph of the order amount in the graph shown in area F in Figure 5), "order" is specified.

4. Specific examples of processing
In this section, a specific example of the process according to this embodiment will be described. In [4-1], the process of automatically detecting a payee for which the difference between the purchase amount in the month to be confirmed and the purchase amount in the same month of the previous year is abnormally large by checking the contents of the purchase accounting data 106c in the business data (anomaly detection process) will be described. In [4-2], the process of displaying the detected abnormal data and data related to the abnormal data that was referenced for the detection on the analysis screen (analysis screen display process) will be described.

[4-1: Abnormality detection process]
(1) Pre-setting As the pre-setting, information required for anomaly detection is stored in advance in a table. In this example, it is assumed that the automatic detection execution schedule data 106a and the acquisition range condition data 106b are set as shown in FIG.

(2) Automatic Execution (2-1) Acquisition of Automatic Detection Execution Schedule Data 106a First, the automatic detection execution schedule data 106a set in (1) is automatically acquired. Specifically, if "detection ID: AB001, schedule ID: SH001" is set as parameters as shown in Fig. 9, the automatic detection execution schedule data 106a shown in Fig. 9 is automatically acquired as data having the detection ID and schedule ID.

(2-2) Determining whether it is the timing to detect an abnormality Next, it is determined whether the timing automatically acquired in (2-1) is the timing to detect an abnormality. Specifically, it is assumed that the timing automatically acquired in (2-1) is "2021/04/07 (the fifth business day of April)." It is possible to determine holidays and business days by referencing the calendar master in the business database, and it is assumed that the calendar master is always updated based on the sales business calendar.

Now, referring to the automatic detection execution schedule data 106a in FIG. 9 that was automatically acquired in (2-1), the execution condition is "the fifth business day of every month." In this way, if the timing at which the automatic detection execution schedule data 106a was automatically acquired matches the timing defined by the execution condition set in the automatic detection execution schedule data 106a, the execution decision result will be "execute," and the processing from (2-3) onwards will be executed. On the other hand, if there is no match, the execution decision result will be "do not execute," and the processing will end.

(2-3) Acquisition of range conditions of data for detecting anomalies Next, the acquisition range condition data 106b set in (1) is automatically acquired. Specifically, the acquisition range condition data 106b shown in FIG. 10 is automatically acquired as data having the parameters "detection ID: AB001, schedule ID: SH001" set in (2-1).

(2-4) Detection of a payee with an abnormally high purchase difference amount Finally, by referring to the purchase accounting data 106c in the business data, a payee with an abnormally high purchase difference amount is detected. In this example, the purchase accounting data 106c is set and saved as shown in FIG.

(2-4-1) Calculation of Purchase Difference Amount First, the purchase difference amount calculation unit 102a calculates the purchase difference amount by referring to the purchase appropriation data 106c in Fig. 12. The calculation formula for the purchase difference amount is, for example, "purchase amount of the month to be confirmed - purchase amount of the same month of the previous year".

Specifically, in the purchase accounting data 106c of FIG. 12, if 2021/03 is viewed as the month to be confirmed, the purchase date belonging to 2021/03 is 2021/03/15. On the other hand, the purchase date belonging to 2020/03, which is the same month of the previous year as the month to be confirmed, is 2020/03/15. The purchase amount on 2021/03/15 is 15,000 yen, so the purchase amount for the confirmation month 2021/03 is 15,000 yen. On the other hand, the purchase amount on 2020/03/15 is 5,000 yen, so the purchase amount for the same month of the previous year 2020/03 is 5,000 yen.

Here, based on the "Acquisition range condition data 106b for purchase accounting data" shown in FIG. 10, which was automatically acquired in (2-3), the parameters "Extraction condition column: accounting year/month, FROM condition: 2020/04, TO condition: 2021/03" are automatically set as shown in FIG. 11. In this case, the purchase difference amount calculation unit 102a uses the above calculation formula to calculate the purchase difference amount for each payee for the confirmation target months 2020/04 to 2021/03. For example, taking the confirmation target month 2021/03 as an example, the purchase difference amount calculation unit 102a calculates the purchase difference amount for the confirmation target month 2021/03 and payee A as 10,000 yen by subtracting the purchase amount of 5,000 yen for the same month of the previous year 2020/03 from the purchase amount of 15,000 yen for the confirmation target month 2021/03.

Figure 12 shows an example of purchase difference calculation result data 106g generated by calculating the purchase difference amount for each combination of the confirmation month and payment destination.

(2-4-2) Detection of Abnormal Values Next, an upper limit is obtained from the purchase difference amount calculation result data 106g using the "quartile range," which is one of the statistical methods, and an abnormal value is detected based on the upper limit. FIG. 13 is a diagram showing how to obtain the upper limit (5,000 yen) when focusing on payee A in the purchase difference amount calculation result data 106g of FIG. 12. As shown in FIG. 13, when the purchase difference amount for payee A is plotted in a box-and-whisker plot, the purchase difference amount of 10,000 yen for March 2021 exceeds the upper limit value of 5,000 yen. Therefore, the abnormal value detection unit 102b detects the purchase difference amount of 10,000 yen for March 2021 as an abnormal value.

In the example in the previous paragraph, for simplicity's sake, only the detection of anomalies for payee A was described, but in actual processing, the anomaly detection unit 102b detects anomalies for all payees included in the purchase difference amount calculation result data 106g. Also, in this embodiment, when the amount to be paid decreases (when the purchase amount is less in the month being checked than in the same month of the previous year), there is no need to check, so no lower limit is determined when detecting anomalies.

Here, three representative statistical methods will be described. First, standard deviation is a method of taking the distribution of values at a certain point, setting a confidence interval, and judging normality/abnormality. Standard deviation is an analysis that does not include the concept of a time axis, so it is not suitable for cross-sectional data analysis of "tendencies from the past". Secondly, the interquartile range used in this example is a method of determining the analysis range of data, aggregating it on the vertical axis, determining the quartile range, and specifying the range magnification that can be judged as normal. The interquartile range is compatible with cross-sectional data analysis because it treats both time series data and point data as vertical axis elements. Thirdly, moving average + outlier is a method of determining the range of the average value and calculating it for each time series. For moving average + outlier, it is necessary to set a threshold value in advance to judge abnormality from the average value, but it is possible to detect abnormal increases and decreases in the transition of the time series. In this embodiment, the interquartile range and moving average + outlier are raised as candidates, but for moving average + outlier, the user needs to set a threshold value of the average value to judge abnormality. For this reason, in this embodiment, it is preferable to use the interquartile range, which can automatically calculate a threshold and determine abnormality.

(2-4-3) Storage of Abnormal Values Finally, the abnormal value detection unit 102b stores information such as the abnormality determination result (accounting year and month: 2021/03, payee: payee A, purchase difference amount: 10,000 yen) in a table, and at the same time, stores a message to be displayed on the analysis screen. Abnormality determination result data 106m, abnormality determination result message data 106n, and abnormality determination result message detail data 106p generated by storing the abnormality determination result are shown in FIG. 14.

[4-2: Analysis screen display process]
(1) Display of summary message Next, information that is automatically detected as an anomaly from the purchase accounting data 106c in the business data is output to the initial analysis screen. The acquisition and display of the initial analysis screen will be described in detail below in (1-1) to (1-3).

(1-1) Acquiring a result message of an abnormality detection First, as shown in FIG. 15, "JOBID: purchase alert" is set as a parameter. Note that in order to display the detected information in a list, the detection ID is assumed to be in an unset state as shown in FIG. 15. In this case, the abnormality display unit 102c acquires the abnormality determination result data 106m, abnormality determination result message data 106n, and abnormality determination result message detail data 106p in FIG. 14, which were generated in (2-4-3) of [4-1], as data having "JOBID: purchase alert". These acquired data are also shown in FIG. 16.

(1-2) Displaying on screen a message resulting from the detection of an anomaly Next, the anomaly display unit 102c displays the degree of anomaly "x", the definition name "Purchase Rate", the summary "Payee A detected", and the detection target "2021/03 Payee A 10,000 yen" in the anomaly determination result message data 106n of Fig. 16 acquired in (1-1) on the initial analysis screen shown in Fig. 16. The displayed degree of anomaly, definition name, summary, and detection target are collectively referred to as a summary message.

This allows the operator to know that there was a payee with an abnormally large purchase difference amount, and which payee that was (Payee A in this example). At this time, the operator can also know the accounting month and year in which the abnormality occurred (2021/03 in this example) and the numerical value of the purchase difference amount that was detected as an abnormal value (10,000 yen in this example).

(1-3) Setting the reference date After the abnormality is displayed, the date when the initial analysis screen was launched (2021/04/10) is set as the reference date in the extraction condition area on the initial analysis screen, as shown in Figure 17.

(2) Display of detailed message and graphs Next, the detailed message and five graphs are output to the analysis screen. The display of the detailed message and the acquisition and display of the five graphs will be explained below in (2-1) to (2-7).

(2-1) Displaying detailed messages and securing output area for graphs 1. First, when the summary message displayed in (1-2) is selected, an analysis screen is launched. The anomaly display unit 102c displays the detection method "interquartile range," the threshold "1.5 times normal range," and the upper limit "5,000 yen" in the anomaly determination result message detailed data 106p acquired in (1-1) on the analysis screen shown in Fig. 18. The displayed detection method, threshold, and upper limit are collectively referred to as a detailed message.

In this way, by displaying the summary message explained in (1-2) and the detailed message explained in this section (2-1), the operator can understand the subject (payee) for which an anomaly was detected, the detection method (calculation method), the benchmark value (purchase difference amount) for which an anomaly was determined, etc. This allows the operator to perform an analysis while knowing which payees have an abnormal purchase difference amount. The purchase amount of payees with abnormal purchase difference amounts and the past performance that is the basis for determining an anomaly can be analyzed using the five graphs explained below.

2. Next, an output area for the graphs is secured. In this embodiment, five graphs are output, so an output area for five graphs is secured, as shown by the five blank spaces on the analysis screen in FIG. 19.

(2-2) Setting Extraction Conditions Next, in order to output a graph, it is necessary to extract the data to be output from the analysis data. Prior to the extraction, however, the extraction conditions are set using the method described in this section (2-2).

1. First, the acquisition range of the analysis data is obtained based on the accounting year and month (2021/03) held by the anomaly determination result data 106m in FIG. 16 that is linked to the summary message displayed in (1-2).

(i) Specifically, the accounting year and month within a range of one year before and after the accounting year and month in which the anomaly was detected (2021/03) is acquired as the acquisition range. In other words, the accounting year and month one year ago is 2020/03, and the accounting year and month one year later is 2022/03.

The setting of acquiring accounting dates within a range of one year before and after is in accordance with the condition of detection within a one-year range in the "Acquisition range condition data for purchase accounting data 106b," which is data provided in advance for anomaly detection. If you wish to expand the range for analysis, you can do so by changing the start date and end date of the period in the extraction conditions area on the analysis screen.

(ii) Based on the acquired acquisition range (2020/03 to 2022/03), the analysis data acquisition range condition is set by narrowing down the range in which the analysis data exists. In this example, it is assumed that the analysis data exists only in the range of 2020/04 to 2021/03. In this case, the analysis data acquisition range condition is as follows, as shown in FIG. 20:
・Extraction condition column: Fiscal year and month ・FROM condition: 2020/04 (taken from the month after 2020/03, which is the fiscal year and month one year ago)
・TO condition: 2021/03 (hiring from the month before 2022/03, which is the fiscal year one year later)
It becomes.

2. Next, the analysis data acquisition range conditions set in "1." are set as initial values, which become the extraction condition initial values (base date, period start, and period end). The extraction condition initial values are shown in FIG. 21. These extraction condition initial values are then set in the extraction condition area on the analysis screen, as shown in FIG. 22. Note that the period start date and period end date set in this area can be manually changed by the operator.

When the display switches to the graph display state, the extraction items for extracting data for graph output are displayed. As shown in FIG. 22, the "person in charge" is set as the aggregation unit, and "sales" is set as the aggregation target, and displayed as the fixed initial value. In this example (2), these are assumed to remain the fixed initial values and are not changed.

Next, the data for analysis is obtained based on the extraction conditions set in (2-2). Below, each graph is defined as follows. A line graph showing the trend in monthly purchase amounts for each payee is referred to as "Graph 1". A line graph showing the purchase amounts for the previous year and the current year for a specific payee is referred to as "Graph 2". A line graph showing the trend in monthly purchase amounts for each organization or person in charge for a specific payee is referred to as "Graph 3". A bar graph showing the purchase amounts by purchase accounting method for each organization or person in charge for a specific month and a specific payee is referred to as "Graph 4". A line graph showing total sales or total orders and a line graph showing the purchase amounts for a specific payee are referred to as "Graph 5".

The acquisition of the data that is the basis for generating graphs 1 and 2 is explained below in (2-3), the acquisition of the data that is the basis for generating graph 3 is explained below in (2-4), the acquisition of the data that is the basis for generating graph 4 is explained below in (2-5), and the acquisition of the data that is the basis for generating graph 5 is explained below in (2-6).

(2-3) Acquisition of Data to Generate Graphs 1 and 2 Based on the extraction conditions set in (2-2) (shown again in the example screen of FIG. 23), parameters for acquiring Graphs 1 and 2 are set as shown in FIG. 23. In the parameters shown in FIG. 23, the start of the period "2020/04", the end of the period "2021/03", and the aggregation mode "Person in Charge" are set based on the extraction conditions set in (2-2).

Then, based on the set parameters shown in FIG. 23, the purchase difference amount calculation result data 106g, which is the data from which graphs 1 and 2 are generated, is generated as follows.

That is, the purchase difference amount calculation unit 102a calculates the purchase difference amount by referring to the purchase accounting data 106c in FIG. 24. The calculation formula for the purchase difference amount is, for example, "purchase amount for the month to be confirmed - purchase amount for the same month of the previous year".

Specifically, in the purchase accounting data 106c in FIG. 24, if 2021/03 is viewed as the month to be confirmed, the purchase date belonging to 2021/03 is 2021/03/15. On the other hand, the purchase date belonging to 2020/03, which is the same month of the previous year as the month to be confirmed, is 2020/03/15. The purchase amounts on 2021/03/15 are 2,200 yen, 2,500 yen, and 10,300 yen, so the purchase amount for the confirmation month 2021/03 is the sum of these, which is 15,000 yen. On the other hand, the purchase amounts on 2020/03/15 are 1,800 yen, 1,600 yen, and 1,600 yen, so the purchase amount for the same month of the previous year 2020/03 is the sum of these, which is 5,000 yen.

As shown in the parameters in FIG. 23, in this example, the start of the period is set to "2020/04" and the end of the period is set to "2021/03", so the purchase difference amount calculation unit 102a uses the above calculation formula to calculate the purchase difference amount for each payment destination for the confirmation target months 2020/04 to 2021/03. For example, taking the confirmation target month 2021/03 as an example, the purchase difference amount calculation unit 102a calculates the purchase difference amount for the confirmation target month 2021/03 and payment destination A to be 10,000 yen by subtracting the purchase amount of 5,000 yen for the same month of the previous year 2020/03 from the purchase amount of 15,000 yen for the confirmation target month 2021/03.

Figure 24 shows an example of purchase difference calculation result data 106g generated by calculating the purchase difference amount (total monthly purchase difference amount) for each combination of the confirmation month and payment destination.

When generating the purchase difference amount calculation result data 106g, the purchase difference amount calculation unit 102a includes in the purchase difference amount calculation result data 106g the amount obtained by aggregating the purchase amount for the current year (2020) for each combination of month and payee (monthly total purchase amount), and the amount obtained by aggregating the purchase amount for the previous year (2019) for each combination of month and payee (monthly total purchase amount for the same month of the previous year).

In this example, as shown in the purchase difference amount calculation result data 106g in FIG. 24, an example has been described in which the monthly total purchase amount, monthly total purchase amount for the same month of the previous year, and monthly total purchase difference amount are calculated for each combination of month and payee, but these amounts may also be calculated for each combination of month, payee, and person in charge.

(2-4) Acquisition of Data to be Used to Generate Graph 3 Based on the extraction conditions set in (2-2) (shown again in the example screen of FIG. 25), parameters for acquiring Graph 3 are set as shown in FIG. 25. In the parameters shown in FIG. 25, the start of the period "2020/04", the end of the period "2021/03", and the aggregation mode "Person in Charge" are set based on the extraction conditions set in (2-2).

Then, based on the set parameters shown in FIG. 25, purchase amount summary data 106h, which is the data from which graph 3 is generated, is generated as follows.

That is, by referring to the purchase accounting data 106c in FIG. 26, the purchase amount is tallied for each combination of accounting month, payee, and person in charge for the period from 2020/04 to 2021/03. The accounting month, payee, person in charge, and tallied purchase amount are stored, and the purchase amount summary data 106h in FIG. 26 is generated.

(2-5) Acquisition of Data to be Used to Generate Graph 4 Based on the extraction conditions set in (2-2) (shown again in the example screen of FIG. 27), parameters for acquiring Graph 4 are set as shown in FIG. 27. In the parameters shown in FIG. 27, the period start "2020/04", period end "2021/03", and aggregation mode "Person in Charge" are set based on the extraction conditions set in (2-2), and the designated accounting year and month "2021/03" is set based on the anomaly determination result data 106m.

Then, based on the set parameters shown in FIG. 27, the aggregated data by purchasing method 106i, which is the data from which Graph 4 is generated, is generated as follows.

That is, by referring to the purchase accounting data 106c in FIG. 28, the purchase amount is tallied for each combination of purchase method, accounting year and month, payee, and person in charge in the range of 2020/04 to 2021/03. Here, for records in which no purchase order number is entered in the purchase accounting data 106c in FIG. 28, the purchase method is tallied as "immediate purchase", whereas for records in which a purchase order number is entered in the purchase accounting data 106c in FIG. 28, the purchase method is tallied as "purchase via order". Note that for purchase records in which a purchase order number is entered in the purchase accounting data 106c, a record of an order corresponding to the purchase order number exists in the order accounting data 106d in FIG. 28. The purchase method, the accounting year and month, the payee, the person in charge, and the tallied purchase amount are stored, and the purchase method-specific tallied data 106i in FIG. 28 is generated.

(2-6) Obtaining Data to Generate Graph 5 Based on the extraction conditions set in (2-2) (shown again in the example screen of FIG. 29), parameters for obtaining Graph 5 are set as shown in FIG. 29. In the parameters shown in FIG. 29, the start of the period "2020/04", the end of the period "2021/03", and the aggregation mode "Sales" have been set based on the extraction conditions set in (2-2), and the designated payee "Payee A" has been set based on the anomaly determination result data 106m.

Then, based on the set parameters shown in FIG. 29, the sales and purchase aggregate data 106j, which is the data from which Graph 5 is generated, is generated as follows.

That is, by referencing the purchase accounting data 106c in FIG. 30, purchase amounts are tallied for each combination of accounting year/month and payee in the range of 2020/04 to 2021/03. Note that this tally is only performed for the payee specified in the parameters shown in FIG. 29 (in this example, "Payee A"). The data classification "Purchases", the accounting year/month, the payee, and the tallied purchase amount are stored in the sales/purchases-specific tallied data 106j in FIG. 30.

In addition, because the aggregation mode specified in the parameters shown in FIG. 29 is "Sales," the sales amount is aggregated for each accounting year and month in the range of 2020/04 to 2021/03 by referencing the sales recording data 106e in FIG. 30. The data classification "Sales," the accounting year and month, and the aggregated sales amount are stored in the sales and purchase aggregated data 106j in FIG. 30.

In this way, the sales and purchase aggregated data 106j shown in Figure 30 is generated.

(2-7) Graph output (display)
Next, five graphs are output based on the data generated in (2-3) to (2-6).

1. Data binding information used when binding data to graph 1 and the graph 1 that is generated are shown in FIG. 31. The graph display unit 102d by business partner generates and displays line graphs with the total monthly purchase amount in the purchase difference amount calculation result data 106g in FIG. 24 on the vertical axis and the accounting year and month in the purchase difference amount calculation result data 106g in FIG. 24 on the horizontal axis for the number of payees.

By looking at Graph 1, the operator can check the purchase amount generation status of the payee for which an anomaly was detected. In addition, by comparing by month, the operator can compare the difference in the purchase amount generation status between the month in which the anomaly was detected and other months. Graph 1 in Figure 31 shows that the purchase amount for payee A in March 2021 was abnormally high.

2. The data binding information used when binding data to graph 2 and the generated graph 2 are shown in FIG. 32. The graph display unit 102e for comparing two years generates a line graph (the line graph of "Purchase Actual Results" in FIG. 32) for the current year (2020) and payee A, with the total monthly purchase amount in the purchase difference amount calculation result data 106g in FIG. 24 as the vertical axis and the month as the horizontal axis. The graph display unit 102e for comparing two years generates a line graph (the line graph of "Last Year's Purchase Actual Results" in FIG. 32) for last year (2019) and payee A, with the total monthly purchase amount in the same month of the previous year in the purchase difference amount calculation result data 106g in FIG. 24 as the vertical axis and the month as the horizontal axis. The graph display unit 102e for comparing two years displays both of the generated graphs together as graph 2. In addition, in graph 2, in order to compare the current fiscal year with last year, only "month" is displayed on the horizontal axis, not "year."

In Graph 2, "FY2020" displayed in the top header is based on the accounting year and month "2021/03" held by the anomaly judgment result data 106m in FIG. 16. In other words, by referencing the monthly period master in the business database, it is determined that the accounting year to which the accounting year and month "2021/03" belongs is "FY2020." Note that the monthly period master is always kept up to date.

In graph 2, "Payee A" displayed in the top header is based on the payee "Payee A" held in the anomaly determination result data 106m in Figure 16.

By looking at graph 2, the operator can compare the purchase amount in the year in which the abnormality was detected with the purchase amount of the previous year from the perspective of the year in which the abnormality was detected. In other words, by comparing by month, it is possible to confirm the difference in the occurrence situation with the purchase amount for the same month of the previous year. From graph 2 in Figure 32, it can be seen that the purchase amount in March 2021 for the year in which the abnormality was detected (FY2020, which corresponds to the line graph of "Purchase results") is significantly higher than the purchase amount in March 2020 for the previous year (FY2019, which corresponds to the line graph of "Last year's purchase results").

3. The data binding information used when binding data to graph 3 and the graph 3 that is generated are shown in FIG. 33. The graph display unit by aggregation 102f generates and displays a line graph with the purchase amount in the purchase amount aggregation data 106h in FIG. 26 (i.e., the "aggregated purchase amount" in (2-4)) on the vertical axis and the accounting year and month in the purchase amount aggregation data 106h in FIG. 26 on the horizontal axis for each person in charge.

In graph 3, "Payee A" displayed in the top header is based on the payee "Payee A" held in the anomaly determination result data 106m in Figure 16.

By looking at Graph 3, the operator can check the purchasing records for each employee who recorded purchases for the payee for which an anomaly was detected. In other words, the purchasing records for each employee can be checked for the month in which the anomaly was detected. If the purchasing records of a specific employee are abnormally high compared to the purchasing records of other employees, this suggests the possibility that the specific employee is colluding with the payee for which an anomaly was detected. Graph 3 in Figure 33 shows that the purchasing records of employee C in March 2021 are abnormally high compared to the purchasing records of other employees A and B, which suggests the possibility that employee C is colluding with payee A.

4. The data binding information used when binding data to graph 4 and the graph 4 that is generated are shown in FIG. 34. The purchase accounting method graph display unit 102g generates and displays a bar graph with the purchase amount in the purchase method aggregation data 106i in FIG. 28 (i.e., the "aggregated purchase amount" in (2-5)) on the vertical axis for each person in charge.

The key point in graph 4 is that it is possible to distinguish between the purchase amount corresponding to "immediate purchase" and the purchase amount corresponding to "purchase via order."

In Graph 4, "2021/03" displayed in the top header is based on the accounting year and month "2021/03" held in the anomaly determination result data 106m in FIG. 16. In addition, "Payee A" displayed in the top header is based on the payee "Payee A" held in the anomaly determination result data 106m in FIG. 16.

By looking at graph 4, the operator can check the purchasing records for each payment method for each employee for the month in which the abnormality was detected and for the payee. Here, with immediate purchases that do not involve an order, no confirmation process is carried out after the order is placed, so fraud is less likely to be discovered when immediate purchases are used. For this reason, employees who record high purchase amounts using immediate purchases are more likely to be committing fraud. Specifically, if the purchasing records of a specific employee and other employees are compared by purchasing method, and the proportion of immediate purchases is high only for a specific employee, this indicates a high possibility that the specific employee is committing fraud. Graph 4 in Figure 34 shows that the proportion of immediate purchases is abnormally high only for employee C, suggesting the possibility that employee C is committing fraud using immediate purchases.

5. Data binding information used when binding data to graph 5 and the graph 5 that is generated are shown in FIG. 35. The sales/purchase comparison graph display unit 102h generates a line graph (the "Purchase" line graph in FIG. 35) for payee A, with the amount in the sales/purchase aggregated data 106j in FIG. 30 (i.e., the "aggregated purchase amount" in (2-6)) as the vertical axis and the accounting year/month in the sales/purchase aggregated data 106j in FIG. 30 as the horizontal axis. The sales/purchase comparison graph display unit 102h also generates a line graph (the "Sales" line graph in FIG. 35) with the amount in the sales/purchase aggregated data 106j in FIG. 30 (i.e., the "aggregated sales amount" in (2-6)) as the vertical axis and the accounting year/month in the sales/purchase aggregated data 106j in FIG. 30 as the horizontal axis. The sales/purchase comparison graph display unit 102h then displays both generated graphs together as graph 5.

In graph 5, the line graph for "Purchases" is a graph based on the amount for the payee "Payee A" held in the anomaly determination result data 106m in FIG. 16.

By looking at Graph 5, the operator can check the sales amount and the purchase amount of the payee for which an anomaly was detected. In other words, the difference in the trend of the sales amount and purchase amount between months in which an anomaly was detected and months in which an anomaly was not detected can be confirmed. Here, when there is purchase fraud, the increase in purchase amount is often not accompanied by an increase in sales amount. Graph 5 in Figure 35 shows that although the sales amount for 2021/03 is the same as other accounting years and months, the purchase amount for 2021/03 has increased, which suggests the possibility of purchase fraud occurring in 2021/03.

6. The generated graphs 1 to 5 are displayed in a list on the analysis screen, as shown in Figure 36.

7. Data for which an abnormality has been detected may be highlighted (for example, displayed in a different color) based on the payee information included in the abnormality determination result data 106m linked to the abnormality determination result message detail data 106p. For example, when displaying Graph 1, as shown in FIG. 37, a setting may be made such that for payees for which an abnormality has been detected (= payees included in the abnormality determination result data 106m), the color (red) of the abnormality flag in the legend indicates that the payee is true, and for payees other than those for which an abnormality has been detected, the color (gray) of the abnormality flag in the legend indicates that the payee is false. In this example, in Graph 1, the line graph for payee A for which an abnormality has been detected is displayed in red.

Furthermore, when displaying graph 4, the purchase amount corresponding to "immediate purchase" is displayed with the abnormality flag in the legend in the color True (red), while the purchase amount corresponding to "purchase via order" is displayed with the abnormality flag in the legend in the color False (gray). This is because "immediate purchase", which does not go through the order confirmation flow, is a form of purchase that is easily used fraudulently, and so needs to be highlighted.

(3) Switching the display of a graph Next, switching the display of an analysis graph will be explained in the following (3-1) and (3-2). In the explanation so far, a case where "person in charge" is specified as the aggregation unit and "sales" is specified as the aggregation target has been explained. In contrast, in the following (3-1), a case where "department" is specified as the aggregation unit and "sales" is specified as the aggregation target will be explained. In the following (3-2), a case where "department" is specified as the aggregation unit and "orders" is specified as the aggregation target will be explained.

(3-1) Display of analysis graphs when "department" is specified as the aggregation unit and "sales" is specified as the aggregation target In this section (3-1), we will mainly explain an example of confirming the amount aggregated by department, which is a type of organizational unit, using graphs 3 and 4. Note that the processing in this section (3-1) is assumed to be performed following the processing in (2).

1. First, as shown in the example screen of FIG. 38, in the area for setting extraction conditions on the analysis screen, "department" is selected as the aggregation unit. In other words, the aggregation unit is changed from "person in charge" to "department." Note that, as shown in the example screen of FIG. 38, the aggregation target remains unchanged at "sales." Furthermore, as shown in the example screen of FIG. 38, the start of the period remains unchanged at "2020/04," and the end of the period remains unchanged at "2021/03."

2. Based on the set extraction conditions, the parameters for obtaining graphs 1 and 2 are set as shown in FIG. 38. In the parameters shown in FIG. 38, the start of the period "2020/04", the end of the period "2021/03", and the aggregation mode "department" are set based on the extraction conditions described in the previous paragraph.

Then, based on the set parameters shown in FIG. 38, purchase difference amount calculation result data 106g is generated, which is the data from which graphs 1 and 2 are generated. However, the method of generation and the contents of the generated data are the same as those explained in (2-3) with reference to FIG. 24, so explanations will be omitted.

3. Next, based on the extraction conditions set in "1", the parameters for obtaining Graph 3 are set as shown in FIG. 39. In the parameters shown in FIG. 39, the start of the period "2020/04", the end of the period "2021/03", and the aggregation mode "Department" are set based on the extraction conditions described in "1".

Then, based on the set parameters shown in FIG. 39, purchase amount summary data 106h, which is the data from which graph 3 is generated, is generated as follows.

That is, by referring to the purchase accounting data 106c in FIG. 40, the purchase amount is tallied for each combination of accounting month, payee, and department in the range of 2020/04 to 2021/03. The accounting month, payee, department, and tallied purchase amount are stored, and the purchase amount summary data 106h in FIG. 40 is generated.

In (2-4), the purchase amount was tallied for each combination of accounting month, payee, and "person in charge," whereas in this item "3," the purchase amount was tallied for each combination of accounting month, payee, and "department."

4. Next, based on the extraction conditions set in "1", the parameters for obtaining graph 4 are set as shown in FIG. 41. In the parameters shown in FIG. 41, the period start "2020/04", the period end "2021/03", and the aggregation mode "department" are set based on the extraction conditions described in "1", and the designated accounting year and month "2021/03" is set based on the anomaly determination result data 106m.

Then, based on the set parameters shown in FIG. 41, the aggregated data by purchasing method 106i, which is the data from which Graph 4 is generated, is generated as follows.

In other words, by referring to the purchase accounting data 106c in FIG. 42, the purchase amount is tallied for each combination of purchase method, accounting year and month, payee, and department in the range of 2020/04 to 2021/03. Here, for records in which no purchase order number is entered in the purchase accounting data 106c in FIG. 42, the purchase method is tallied as "immediate purchase", whereas for records in which a purchase order number is entered in the purchase accounting data 106c in FIG. 42, the purchase method is tallied as "purchase via order". Note that for purchase records in which a purchase order number is entered in the purchase accounting data 106c, a record of an order corresponding to the purchase order number exists in the order accounting data 106d in FIG. 42. The purchase method, the accounting year and month, the payee, the department, and the tallied purchase amount are stored, and the purchase method-specific tallied data 106i in FIG. 42 is generated.

In (2-5), the purchase amount was tallied for each combination of purchase method, accounting month, payee, and "person in charge," whereas in this item "4," the purchase amount was tallied for each combination of purchase method, accounting month, payee, and "department."

5. Next, based on the extraction conditions set in "1", the parameters for obtaining graph 5 are set as shown in FIG. 29. In the parameters shown in FIG. 29, the period start "2020/04", the period end "2021/03", and the aggregation mode "Sales" are set based on the extraction conditions described in "1", and the designated payee "Payee A" is set based on the anomaly determination result data 106m.

Then, based on the set parameters shown in FIG. 29, sales and purchase aggregate data 106j, which is the data from which Graph 5 is generated, is generated. However, the method of generation and the contents of the generated data are the same as those described in (2-6) with reference to FIG. 30, so a description thereof will be omitted.

6. Next, five graphs are output based on the data generated in "2" to "5".

The data binding information used when binding data to graph 1 and the contents of the generated graph 1 are the same as those explained in (2-7) "1" with reference to Figure 31, so we will not explain them here.

The data binding information used when binding data to graph 2 and the contents of the generated graph 2 are the same as those explained in (2-7) "2" with reference to Figure 32, so we will not explain them here.

The data binding information used when binding data to graph 3 and the graph 3 that is generated are shown in FIG. 43. The graph display unit by aggregation 102f generates and displays line graphs with the purchase amount in the purchase amount aggregation data 106h in FIG. 40 (i.e., the "aggregated purchase amount" in "3") on the vertical axis and the accounting year and month in the purchase amount aggregation data 106h in FIG. 40 on the horizontal axis for the number of departments.

The key point is that Graph 3 in Figure 33 explained in (2-7) “3” was a line graph “by person in charge,” whereas Graph 3 in Figure 43 explained in this section is a line graph “by department.”

The data binding information used to bind data to graph 4 and the graph 4 that is generated are shown in FIG. 44. The purchase accounting method graph display unit 102g generates and displays a bar graph with the purchase amount in the purchase method aggregation data 106i in FIG. 42 (i.e., the "aggregated purchase amount" in "4") on the vertical axis for each department.

The key point is that graph 4 in Figure 34 explained in (2-7) "4" is a bar graph "by person in charge," whereas graph 4 in Figure 44 explained in this section is a bar graph "by department."

The data binding information used when binding data to graph 5 and the contents of the generated graph 5 are the same as those explained in (2-7) “5” with reference to FIG. 35, so we will not explain them here.

7. The generated graphs 1 to 5 are displayed in a list on the analysis screen, as shown in Figure 45.

8. The method for setting up the highlighting of graphs 1 and 4 is the same as that of "7" in (2-7), so a detailed explanation will be omitted. Figure 46 shows the coloring parameter settings for highlighting payment recipients for which an abnormality has been detected in graph 1.

9. Here we will explain the effect of looking at the "by department" graph. If a specific department is generating a large amount of purchases, it is possible that fraud is occurring on a departmental level as multiple employees within that department cancel transactions. The person in charge of checking can become aware of this possibility by looking at graphs 3 and 4.

Below, we will explain in detail how to discover fraud on a departmental basis using figures that compare graphs by person in charge and by department for Graph 3 and Graph 4. Note that Graphs 3 and 4 used in the following explanation are separate graphs that are not related to the specific examples given so far.

First, Figure 47 shows Graph 3 (graph by person in charge) when purchasing amounts are tabulated by person in charge, and Graph 3 (graph by department) when purchasing amounts are tabulated by department. Let's say that fraud is occurring at the department level in Department B. In the graph by person in Figure 47, all of the graphs for persons A to D in Department B show a similar trend of increasing purchasing amounts in March 2021, so at first glance it appears that no purchasing fraud is occurring. In contrast, in the graph by department in Figure 47, only the graph for Department B shows an unnatural increase in purchasing amounts in March 2021. For this reason, by looking at the graph by department, the person in charge of checking can notice the possibility that, for example, persons A to D in Department B may each be recording a small amount of fraudulent purchasing amounts so that the purchasing fraud is less likely to be revealed.

Next, Figure 48 shows graph 4 (graph by employee) when purchasing amounts are tabulated by employee, and graph 4 (graph by department) when purchasing amounts are tabulated by department. Let's say that fraud is occurring at the department level in department B. In the graphs by employee in Figure 48, all of the graphs for employees A to D in department B record immediate purchases on average, and the proportion of immediate purchases for a specific employee is not extremely high, so at first glance it appears that no fraudulent purchasing is occurring. In contrast, in the graphs by department in Figure 48, only the graph for department B shows a high proportion of immediate purchases. For this reason, the person in charge of checking can notice the possibility that employees A to D in department B are each recording a small amount of fraudulent purchasing amounts as immediate purchases so that the fraudulent purchasing will not be revealed.

(3-2) Display of analysis graph when "department" is specified as the aggregation unit and "order" is specified as the aggregation target In this section (3-2), we will mainly explain an example of confirming the sales amount generation status at the time of order (i.e., order amount) using graph 5. Note that the processing in this section (3-2) is assumed to be performed following the processing in (3-1).

1. First, as shown in the example screen of FIG. 49, "Orders" is selected as the aggregation target in the area for setting extraction conditions on the analysis screen. In other words, the aggregation target is changed from "Sales" to "Orders." Note that, as shown in the example screen of FIG. 49, the aggregation unit remains unchanged, remaining at "Department." Furthermore, as shown in the example screen of FIG. 49, the start of the period remains unchanged, remaining at "2020/04," and the end of the period remains unchanged, remaining at "2021/03."

2. Based on the set extraction conditions, the parameters for obtaining graphs 1 and 2 are set as shown in FIG. 49. In the parameters shown in FIG. 49, the start of the period "2020/04", the end of the period "2021/03", and the aggregation mode "department" are set based on the extraction conditions described in the previous paragraph.

Then, based on the set parameters shown in FIG. 49, purchase difference amount calculation result data 106g is generated, which is the data from which graphs 1 and 2 are generated. However, the method of generation and the contents of the generated data are the same as those explained in (2-3) with reference to FIG. 24, so explanations will be omitted.

3. Next, based on the extraction conditions set in "1", the parameters for obtaining Graph 3 are set as shown in FIG. 39. In the parameters shown in FIG. 39, the start of the period "2020/04", the end of the period "2021/03", and the aggregation mode "Department" are set based on the extraction conditions described in "1".

Then, based on the set parameters shown in FIG. 39, purchase amount summary data 106h is generated, which is the data from which Graph 3 is generated. However, the method of generation and the contents of the generated data are the same as those explained in (3-1) "3" with reference to FIG. 40, so explanations are omitted.

4. Next, based on the extraction conditions set in "1", the parameters for obtaining graph 4 are set as shown in FIG. 41. In the parameters shown in FIG. 41, the period start "2020/04", the period end "2021/03", and the aggregation mode "department" are set based on the extraction conditions described in "1", and the designated accounting year and month "2021/03" is set based on the anomaly determination result data 106m.

Then, based on the set parameters shown in FIG. 41, the data used to generate Graph 4, aggregated data by purchasing method 106i, is generated. However, the method of generation and the contents of the generated data are the same as those explained in (3-1) "4" with reference to FIG. 42, so explanations will be omitted.

5. Next, based on the extraction conditions set in "1", the parameters for obtaining Graph 5 are set as shown in FIG. 50. In the parameters shown in FIG. 50, the period start "2020/04", period end "2021/03", and aggregation mode "Order" are set based on the extraction conditions described in "1", and the designated payee "Payee A" is set based on the anomaly determination result data 106m.

Then, based on the set parameters shown in FIG. 50, the order/purchase aggregate data 106k, which is the data from which Graph 5 is generated, is generated as follows.

That is, by referencing the purchase accounting data 106c in FIG. 51, purchase amounts are tallied for each combination of accounting year/month and payee in the range of 2020/04 to 2021/03. Note that this tally is only performed for the payee specified in the parameters shown in FIG. 50 (in this example, "Payee A"). The data classification "Purchases", the accounting year/month, the payee, and the tallied purchase amount are stored in the order/purchase-specific tallied data 106k in FIG. 51.

In addition, because the aggregation mode specified in the parameters shown in FIG. 50 is "Orders," the order amounts are aggregated for each accounting month and year in the range of 2020/04 to 2021/03, with reference to the order posting data 106f in FIG. 51. The data classification "Orders," the accounting month and year, and the aggregated order amounts are stored in the order/purchase aggregate data 106k in FIG. 51.

In this way, the order/purchase aggregated data 106k shown in Figure 51 is generated.

The key point here is that in (2-6), the "sales amount" was the subject of the calculation, whereas in item (3-2) "5", the "order amount" was the subject of the calculation.

6. Next, five graphs are output based on the data generated in "2" to "5".

The data binding information used when binding data to graph 1 and the contents of the generated graph 1 are the same as those explained in (2-7) "1" with reference to Figure 31, so we will not explain them here.

The data binding information used when binding data to graph 2 and the contents of the generated graph 2 are the same as those explained in (2-7) "2" with reference to Figure 32, so we will not explain them here.

The data binding information used when binding data to graph 3 and the contents of the generated graph 3 are the same as those explained in (3-1) "6" with reference to Figure 43, so the explanation will be omitted.

The data binding information used when binding data to graph 4 and the contents of the generated graph 4 are the same as those explained in (3-1) "6" with reference to Figure 44, so the explanation will be omitted.

The data binding information used to bind data to graph 5 and the graph 5 generated are shown in FIG. 52. The order/purchase comparison graph display unit 102i generates a line graph (the "purchase" line graph in FIG. 52) for payee A, with the amount in the order/purchase aggregated data 106k in FIG. 51 (i.e., the "aggregated purchase amount" in "5") as the vertical axis and the accounting year/month in the order/purchase aggregated data 106k in FIG. 51 as the horizontal axis. The order/purchase comparison graph display unit 102i also generates a line graph (the "order" line graph in FIG. 52) with the amount in the order/purchase aggregated data 106k in FIG. 51 (i.e., the "aggregated order amount" in "5") as the vertical axis and the accounting year/month in the order/purchase aggregated data 106k in FIG. 51 as the horizontal axis. The order/purchase comparison graph display unit 102i then displays both generated graphs together as graph 5.

Graph 5 in Figure 35, explained in (2-7) "5," consists of a line graph of purchases and a line graph of "sales," allowing you to compare purchase amounts and "sales amounts," whereas graph 5 in Figure 52, explained in this section, consists of a line graph of purchases and a line graph of "orders," allowing you to compare purchase amounts and "order amounts."

7. The generated graphs 1 to 5 are displayed in a list on the analysis screen, as shown in Figure 53.

8. The method for setting up highlighting for graphs 1 and 4 is the same as that in (2-7) "7", so a detailed explanation will be omitted.

9. Here, we will explain what effect looking at the "Orders" graph in Graph 5 has.

If sales occur in the same month as orders, the person in charge of checking should select "Sales" in the extraction conditions section on the analysis screen. For example, if purchases, orders, and sales for a certain transaction are all recorded in March 2021, the person in charge of checking can discover purchase fraud by comparing the purchase amount graph and the sales amount graph as of March 2021, as shown in the dotted frame in the graph in Figure 54.

However, when sales are recorded across multiple months, it may not be possible to detect abnormalities in the purchase amount by comparing sales and purchases for a single month. In this case, the person in charge of checking can select "Orders" in the extraction conditions section on the analysis screen, compare the order information before sales with the purchase information, and detect abnormalities in the purchase amount.

For example, if the purchases and orders for a certain transaction are recorded in March 2021, but the sales are recorded the following month in April 2021, there is no point in comparing the purchase amount graph and the sales amount graph at the same time. In contrast, as shown by the dotted frame within the graph in Figure 55 (not outside the graph), the purchase amount graph and the order amount graph can be compared at the same time of March 2021, allowing the person in charge of checking to discover any purchasing fraud.

The person checking the data can choose whether to look at the sales amount or the order amount as the comparison target for the purchase amount by specifying it in the extraction conditions section on the analysis screen.

[5. Summary of the present embodiment]
In this way, as mainly explained in [4-1], the fraudulent purchasing detection support device 100 of this embodiment can support the fraudulent purchasing detection work by automatically calculating the difference between the purchase amount during the confirmation period (a certain accounting month) and the purchase amount during the comparison period (the same month of the previous year), and detecting any of the calculated differences that are abnormally large.

In addition, according to the purchasing fraud detection business support device 100 of this embodiment, as mainly explained in [4-2], by referring to the five graphs displayed on the analysis screen, it is also possible to analyze the cause of the abnormality in the purchasing price difference amount.

Conventionally, for purchase transactions that occur periodically and in large numbers, a person in charge would manually check the purchase records, including the previous year's records, in order to detect any abnormalities in the transactions. However, this method requires specialized knowledge to collect various data and perform calculations before a purchase fraud can be discovered, which results in a lot of work time and a high risk of operational errors and oversights.

In this embodiment, for example, data registered due to fraudulent purchases can be automatically detected and notified within transaction data in sales operations. Specifically, anomalies can be detected by calculating the difference between the actual purchases for a single month and the actual purchases for the same month of the previous year. In addition, a screen can be output that allows analysis of the timing of the increase in amount compared to the actual purchases for the same month of the previous year, the person who recorded the purchase, and the recording method, etc.

In other words, in this embodiment, for example, purchasing data can be aggregated and purchasing data deemed to be fraudulent can be classified, making it possible to visualize the fraudulent situation. In addition, in this embodiment, for example, the aggregation method and the alert method have been devised. And, in this embodiment, for example, by making it possible to aggregate purchasing records by organization unit, it is possible to detect purchasing fraud at the organization unit that cannot be detected by just looking at the purchasing records of individual employees.

As a result, this embodiment provides the following effects, for example. First, it is now possible to periodically and automatically detect purchasing fraud from a huge amount of purchasing information. In other words, even people without specialized knowledge can notice purchasing fraud. In addition, the system automatically processes content that requires data to be compiled and compared under various conditions (conditions such as by person in charge, by purchasing accounting method, and by month of last year's purchasing results), thereby avoiding the risk of operational errors. Furthermore, a screen can be output that analyzes and confirms the abnormality and its basis, making it possible to detect abnormalities with high reliability.

[6. Contribution to the United Nations-led Sustainable Development Goals (SDGs)]
This embodiment can contribute to improving business efficiency and promoting appropriate management decisions by companies, thereby making it possible to contribute to goals 8 and 9 of the SDGs.

In addition, this embodiment can contribute to reducing waste and promoting paperless and electronic systems, which can contribute to the achievement of SDGs goals 12, 13, and 15.

In addition, this embodiment can contribute to strengthening control and governance, making it possible to contribute to Goal 16 of the SDGs.

7. Other embodiments
The present invention may be embodied in various different embodiments other than those described above within the scope of the technical concept set forth in the claims.

For example, among the processes described in the embodiments, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically using a known method.

In addition, the processing procedures, control procedures, specific names, registered data for each process, information including parameters such as search conditions, screen examples, and database configurations shown in this specification and drawings may be changed as desired unless otherwise specified.

In addition, with regard to the purchasing fraud detection business support device 100, each component shown in the figure is a functional concept, and does not necessarily have to be physically configured as shown in the figure.

For example, the processing functions of the purchasing fraud detection business support device 100, particularly the processing functions performed by the control unit, may be realized in whole or in part by a CPU and a program interpreted and executed by the CPU, or may be realized as hardware using wired logic. The program is recorded on a non-transient computer-readable recording medium that contains programmed instructions for causing the information processing device to execute the processes described in this embodiment, and is mechanically read by the purchasing fraud detection business support device 100 as needed. That is, a computer program is recorded in a storage unit such as a ROM or HDD (Hard Disk Drive) to give instructions to the CPU in cooperation with the OS and perform various processes. This computer program is executed by being loaded into RAM, and cooperates with the CPU to form the control unit.

This computer program may also be stored in an application program server connected to the purchasing fraud detection business support device 100 via any network, and it may be possible to download all or part of it as needed.

In addition, the program for executing the processing described in this embodiment may be stored on a non-transitory computer-readable recording medium, and may also be configured as a program product. Here, the term "recording medium" refers to a memory card, a USB (Universal Serial Bus) memory, a SD (Secure Digital) card, a flexible disk, a magneto-optical disk, a ROM, an EPROM (Erasable Programmable Read Only Memory), an EEPROM (registered trademark) (Electrically Erasable and Programmable Read Only Memory), a CD-ROM (Compact Disk Read Only Memory), an MO (Magneto-Optical disk), a DVD (Digital Versatile This includes any "portable physical media" such as a Blu-ray (registered trademark) Disc, and a Blu-ray (registered trademark) Disc.

A "program" is a data processing method written in any language or description method, and may be in any format, such as source code or binary code. Note that a "program" is not necessarily limited to a single configuration, but also includes a distributed configuration consisting of multiple modules or libraries, and a program that works in conjunction with a separate program, such as an OS, to achieve its function. Note that the specific configuration and reading procedure for reading a recording medium in each device shown in the embodiments, as well as the installation procedure after reading, can use well-known configurations and procedures.

The various databases stored in the memory unit are storage devices such as RAM, ROM, etc., fixed disk devices such as hard disks, flexible disks, and optical disks, and store various programs, tables, databases, and web page files used for various processes and providing websites.

The purchasing fraud detection business support device 100 may be configured as an information processing device such as a known personal computer or workstation, or may be configured as the information processing device to which any peripheral device is connected. The purchasing fraud detection business support device 100 may be realized by implementing software (including programs or data, etc.) that causes the device to realize the processing described in this embodiment.

Furthermore, the specific form of distribution and integration of the devices is not limited to that shown in the figures, and all or part of them can be functionally or physically distributed and integrated in any unit depending on various additions or functional loads. In other words, the above-mentioned embodiments may be implemented in any combination, or the embodiments may be implemented selectively.

The present invention is useful, for example, in any industry or business type where purchasing history management is required.

100 Purchasing fraud detection business support device 102 Control unit
102a Purchase difference amount calculation unit
102b Abnormal value detection unit
102c Abnormality display unit
102d Graph display section by customer
102e Graph display section for comparing two years
102f Graph display section by aggregation unit
102g Graph display by purchase accounting method
102h Sales and purchase comparison graph display section
102i Order/purchase comparison graph display unit 104 Communication interface unit 106 Storage unit
106a Automatic detection execution schedule data
106b Acquisition range condition data
106c Purchase accounting data
106d Order entry data
106e Sales Record Data
106f Order entry data
106g Purchase difference amount calculation result data
106h Purchase amount summary data
106i Aggregated data by purchasing method
106j Sales and purchase aggregate data
106k Order/Purchase Data
106m Abnormality Judgment Result Data
106n Abnormality determination result message data
106p Abnormality determination result message detail data 108 Input/output interface unit 112 Input device 114 Output device 200 Server 300 Network

Claims (12)

A purchasing fraud detection business support device that is equipped with a control unit and a storage unit and that can support a business of detecting fraud regarding purchasing records by calculating a purchasing difference amount, which is the difference between a purchasing amount during a period to be confirmed and a purchasing amount during a period to be compared,
The storage unit includes:
Purchase accounting data including purchase date, supplier identification data, and purchase amount is stored,
The control unit is
a purchase difference amount calculation means for calculating the purchase difference amount for each combination of the confirmation period and the customer identification data by subtracting the purchase amount corresponding to the purchase date belonging to the comparison period, which is calculated for each customer identification data, from the purchase amount corresponding to the purchase date belonging to the confirmation period, which is calculated for each customer identification data;
an abnormal value detection means for detecting, according to a predetermined statistical method, an abnormal value that exceeds an upper limit among the purchase difference amounts calculated by the purchase difference amount calculation means;
an abnormality display means for displaying the purchase difference amount detected as an abnormal value by the abnormal value detection means and the combination of the confirmation target period and the customer identification data corresponding to the detected purchase difference amount;
To have
A purchasing fraud detection business support device characterized by the above. The abnormality display means is
Displaying the predetermined statistical method and the upper limit value together;
2. The purchasing fraud detection business support device according to claim 1, The predetermined statistical method is a method using an interquartile range,
the upper limit is calculated by adding 1.5 times the interquartile range to the third quartile;
3. The purchasing fraud detection business support device according to claim 1 or 2, The storage unit includes:
The data storage device further stores a purchase difference amount calculation result data including an amount obtained by aggregating the purchase amount in the purchase appropriation data for each period and the customer identification data, the period, and the customer identification data;
The control unit is
a business partner graph display means for generating and displaying a line graph showing a trend of the aggregated amount for each period based on the purchase difference amount calculation result data, the line graph being equal to the number of business partner identification data;
The line graph is:
The aggregated amount in the purchase difference amount calculation result data is set as the vertical axis,
The time period in the purchase difference amount calculation result data is plotted on the horizontal axis;
4. The purchasing fraud detection business support device according to claim 1, The storage unit includes:
The data storage device further stores a purchase difference amount calculation result data including an amount obtained by aggregating the purchase amount in the purchase appropriation data for each period and the customer identification data, the period, and the customer identification data;
The control unit is
a line graph showing the trend of the aggregated amount for each period in a year to be confirmed for the supplier identification data corresponding to the purchase difference amount detected by the abnormal value detection means, based on the purchase difference amount calculation result data, and a line graph showing the trend of the aggregated amount for each period in a year to be compared for the supplier identification data corresponding to the purchase difference amount detected by the abnormal value detection means,
The confirmation target fiscal year is a fiscal year to which the confirmation target period corresponding to the purchase difference amount detected as an abnormal value by the abnormal value detection means belongs,
The comparison year is the year immediately preceding the confirmation year,
The line graph is:
The aggregated amount in the purchase difference amount calculation result data is set as the vertical axis,
The horizontal axis should show months without years.
5. The purchasing fraud detection business support device according to claim 1, The purchase accounting data further includes at least one aggregation unit selected from among business establishment identification data, department identification data, and person in charge identification data;
The storage unit includes:
the purchase amount in the purchase recording data is further stored in the data storage unit, and purchase amount summary data including the period, the customer identification data, and the summary unit is stored in the data storage unit;
The control unit is
a graph display unit for displaying a line graph showing a change in the amount of the transaction for each period for the customer identification data corresponding to the difference in purchase amount detected by the abnormal value detection unit based on the purchase amount summary data, the line graph being generated for each of the number of the summary units;
The line graph is:
The aggregated amount in the purchase amount aggregate data is set as the vertical axis,
The time period in the purchase amount summary data is plotted on the horizontal axis;
6. The purchasing fraud detection business support device according to claim 1, The purchase accounting data further includes at least one of the business establishment identification data, department identification data, and person in charge identification data as a counting unit, and an order number for purchases made via an order by registering the order accounting data;
The storage unit includes:
The data further includes an amount obtained by aggregating the purchase amount in the purchase accounting data for each combination of the period, the customer identification data, the aggregation unit, and the presence or absence of the purchase order number, and an aggregated data by purchase method including the period, the customer identification data, the aggregation unit, an immediate purchase classification which is a classification when the purchase amount without the purchase order number is aggregated, and an order route classification which is a classification when the purchase amount with the purchase order number is aggregated,
The control unit is
a purchase accounting method-specific graph display means for generating and displaying bar graphs showing the aggregated amounts for the period corresponding to the purchase difference amount detected by the abnormal value detection means and the customer identification data corresponding to the purchase difference amount detected by the abnormal value detection means, the number of bar graphs being the same as the number of aggregation units, based on the purchase method-specific aggregated data;
In the bar graph,
The total amount corresponding to the immediate purchase category and the total amount corresponding to the order transit category are identifiable;
7. The purchasing fraud detection business support device according to claim 1, The storage unit includes:
Sales recording data including a sales date and a sales amount;
a purchase summary record including an amount obtained by summarizing the purchase amount in the purchase appropriation data for each combination of a period and the customer identification data, the period in the purchase appropriation data, and the customer identification data in the purchase appropriation data; and sales and purchase summary data including a sales summary record including an amount obtained by summarizing the sales amount in the sales appropriation data for each period and the period in the sales appropriation data;
is further stored,
The control unit is
a sales/purchase comparison graph display means for generating a line graph showing a trend of the aggregated amount for each period for the supplier identification data corresponding to the purchase difference amount detected by the abnormal value detection means based on the purchase aggregate record, generating a line graph showing the aggregated amount for each period based on the sales aggregate record, and displaying both graphs together;
The line graph is:
The aggregated amount in the record is set as the vertical axis,
The time period in the record is plotted on the horizontal axis;
8. The purchasing fraud detection business support device according to claim 1, The storage unit includes:
Order entry data including an order date and an order amount;
a purchase summary record including an amount obtained by summarizing the purchase amount in the purchase apportionment data for each combination of a period and the customer identification data, the period in the purchase apportionment data, and the customer identification data in the purchase apportionment data; and order/purchase summary data including an order summary record including an amount obtained by summarizing the order amount in the order apportionment data for each period and the period in the order apportionment data;
is further stored,
The control unit is
a line graph showing the trend of the aggregated amount for each period for the supplier identification data corresponding to the purchase difference amount detected by the abnormal value detection means based on the purchase aggregate record, a line graph showing the aggregated amount for each period based on the order aggregate record, and a graph display means for comparing order and purchase that displays both graphs together;
The line graph is:
The aggregated amount in the record is set as the vertical axis,
The time period in the record is plotted on the horizontal axis;
9. The purchasing fraud detection business support device according to claim 1, The confirmation period is a certain month of a certain year,
The comparison period is the certain month of the year preceding the certain year;
10. The purchasing fraud detection business support device according to claim 1, A method for supporting the detection of purchase fraud, which is executed by an information processing device having a control unit and a storage unit, and which can support a task of detecting fraud in purchase records by calculating a purchase difference amount, which is the difference between a purchase amount during a period to be confirmed and a purchase amount during a period to be compared, comprising:
The storage unit includes:
Purchase accounting data including purchase date, supplier identification data, and purchase amount is stored,
Executed by the control unit,
a purchase difference amount calculation step of calculating the purchase difference amount for each combination of the confirmation period and the customer identification data by subtracting the purchase amount corresponding to the purchase date belonging to the comparison period, which is calculated for each customer identification data, from the purchase amount corresponding to the purchase date belonging to the confirmation period, which is calculated for each customer identification data;
an abnormal value detection step of detecting, according to a predetermined statistical method, an abnormal value that exceeds an upper limit among the purchase difference amounts calculated in the purchase difference amount calculation step;
an abnormality display step of displaying the purchase difference amount detected as an abnormal value in the abnormal value detection step and a combination of the confirmation target period and the customer identification data corresponding to the detected purchase difference amount;
containing,
A purchasing fraud detection business support method characterized by the above. A purchasing fraud detection support program executed by an information processing device having a control unit and a storage unit, which can support a task of detecting fraud in purchasing records by calculating a purchasing difference amount, which is the difference between a purchasing amount during a period to be confirmed and a purchasing amount during a period to be compared,
The storage unit includes:
Purchase accounting data including purchase date, supplier identification data, and purchase amount is stored,
The control unit ,
a purchase difference amount calculation step of calculating the purchase difference amount for each combination of the confirmation period and the customer identification data by subtracting the purchase amount corresponding to the purchase date belonging to the comparison period, which is calculated for each customer identification data, from the purchase amount corresponding to the purchase date belonging to the confirmation period, which is calculated for each customer identification data;
an abnormal value detection step of detecting, according to a predetermined statistical method, an abnormal value that exceeds an upper limit among the purchase difference amounts calculated in the purchase difference amount calculation step;
an abnormality display step of displaying the purchase difference amount detected as an abnormal value in the abnormal value detection step and a combination of the confirmation target period and the customer identification data corresponding to the detected purchase difference amount;
To carry out the
A program to assist in the detection of purchasing fraud.

Images