JP2022108349A - Banknote demand prediction program - Google Patents

Abstract

To automatically and accurately predict the demand for banknotes in an automated teller machine.SOLUTION: A banknote demand prediction program which predicts the demand for banknotes to be stored in an automated teller machine causes a computer to execute: an information acquisition step of acquiring region information on a region where the automated teller machine is installed; and a determination step of determining degrees of demand for each of banknotes, on the basis of the region information acquired in the information acquisition step, by referring to three or more stages of associations between reference region information acquired in advance on regions where automated teller machines are installed and a degree of demand for each banknote in the automated teller machines in accordance with the sales.SELECTED DRAWING: Figure 1

Description

The present invention relates to a banknote demand prediction program for predicting the demand for banknotes in automatic teller machines.

Automatic teller machines (ATMs) are devices that allow users to perform operations such as withdrawing and transferring cash from accounts of financial institutions, account transfers, and the like without the intervention of bank employees. ATMs store a large amount of banknotes in advance so that users can withdraw cash. Sometimes. Therefore, it is necessary to replenish banknotes at appropriate timing so as not to run out of banknotes in the ATM. For this reason, there has long been a desire for a system that predicts the demand for banknotes at ATMs and replenishes banknotes at the appropriate timing, even without special skills or know-how, but the current situation has not been proposed yet. .

SUMMARY OF THE INVENTION Accordingly, the present invention has been devised in view of the above-described problems, and an object of the present invention is to provide a banknote capable of automatically and highly accurately predicting the demand for banknotes in an automatic teller machine. To provide a demand forecasting program.

In order to solve the above-described problems, a banknote demand forecasting program according to the present invention is a banknote demand forecasting program for predicting demand for each banknote to be stored in an automatic teller machine. an information acquisition step of acquiring area information about the area where the automatic teller machine is located; pre-obtained reference area information about the area where the automatic teller machine is installed; and a determining step of determining the degree of demand for each banknote based on the area information obtained in the information obtaining step with reference to three or more levels of degree of association with the degree of demand.

Even if there is no special skill or experience, it is possible to automatically and highly accurately predict demand for banknotes in an automated teller machine.

1 is a block diagram showing the overall configuration of a system to which the present invention is applied; FIG. It is a figure which shows the specific structural example of a search apparatus. It is a figure for demonstrating the operation|movement of this invention. It is a figure for demonstrating the operation|movement of this invention. It is a figure for demonstrating the operation|movement of this invention. It is a figure for demonstrating the operation|movement of this invention. It is a figure for demonstrating the operation|movement of this invention. It is a figure for demonstrating the operation|movement of this invention. It is a figure for demonstrating the operation|movement of this invention. It is a figure for demonstrating the operation|movement of this invention. It is a figure for demonstrating the operation|movement of this invention. It is a figure for demonstrating the operation|movement of this invention. It is a figure for demonstrating the operation|movement of this invention.

A banknote demand prediction program to which the present invention is applied will be described in detail below with reference to the drawings.

First Embodiment FIG. 1 is a block diagram showing the overall configuration of a banknote demand forecasting system 1 in which a banknote demand forecasting program to which the present invention is applied is installed. The banknote demand prediction system 1 includes an information acquisition unit 9 , a search device 2 connected to the information acquisition unit 9 , and a database 3 connected to the search device 2 . Data communication may be performed between each of these components via an Internet connection. Any one or more of each of these components may also be implemented within an automated teller machine (ATM).

The information acquisition unit 9 is a device for a person using this system to input various commands and information. The information acquisition unit 9 is not limited to a device for inputting text information, and may be configured by a device such as a microphone that can detect voice and convert it into text information. Further, the information acquisition unit 9 may be configured as an imaging device capable of capturing an image, such as a camera. The information acquisition unit 9 may be configured by a scanner having a function of recognizing a character string from a paper document. Further, the information acquisition unit 9 may be integrated with the discriminating device 2 to be described later. The information acquisition unit 9 outputs the detected information to the discrimination device 2 . Further, the information acquisition unit 9 may be configured by a means for specifying area information by scanning map information. Further, the information acquisition unit 9 may be configured with an illuminance sensor for measuring a temperature sensor, a humidity sensor, and a wind direction sensor. The information acquisition unit 9 may also be configured with a communication interface that acquires weather data from the Meteorological Agency or a private weather forecast company. The information acquisition unit 9 may be composed of a body sensor worn on the body to detect body data, and the body sensor detects, for example, body temperature, heart rate, blood pressure, number of steps, walking speed, and acceleration. It may be configured with a sensor for Also, the body sensor may acquire biological data not only of humans but also of animals. The information acquisition unit 9 may be configured as a device that acquires information such as drawings by scanning or reading from a database. The information acquisition unit 9 may be configured by an odor sensor that detects odors and scents in addition to these.

The database 3 accumulates various reference information necessary for banknote demand forecasting. Details of the reference information will be described later.

The searching device 2 is composed of an electronic device such as a personal computer (PC), for example, but in addition to the PC, it is embodied in any other electronic device such as a mobile phone, a smartphone, a tablet terminal, a wearable terminal, etc. It may be one that is made into. A user can obtain a search solution by this search device 2 .

FIG. 2 shows a specific configuration example of the search device 2. As shown in FIG. This searching device 2 performs wired or wireless communication with a control unit 24 for controlling the entire searching device 2 and an operation unit 25 for inputting various control commands via operation buttons, keyboards, etc. A communication unit 26 for searching, an estimating unit 27 for making various judgments, and a storage unit 28, represented by a hard disk, for storing a program for performing a search to be executed are connected to the internal bus 21, respectively. . Further, the internal bus 21 is connected with a display unit 23 as a monitor for actually displaying information.

The controller 24 is a so-called central control unit for controlling each component implemented in the search device 2 by transmitting control signals via the internal bus 21 . In addition, the control unit 24 transmits commands for various controls through the internal bus 21 according to operations through the operation unit 25 .

The operation unit 25 is embodied by a keyboard or a touch panel, and a user inputs an execution command for executing a program. The operation unit 25 notifies the control unit 24 when the execution command is input by the user. Upon receiving this notification, the control unit 24 cooperates with each component including the estimation unit 27 to execute desired processing operations. The operation unit 25 may be embodied as the information acquisition unit 9 described above.

The estimation unit 27 estimates a search solution. The estimation unit 27 reads various information stored in the storage unit 28 and various information stored in the database 3 as necessary information in executing the estimation operation. This estimation unit 27 may be controlled by artificial intelligence. This artificial intelligence may be based on any known artificial intelligence technology.

The display unit 23 is composed of a graphic controller that creates a display image based on control by the control unit 24 . The display unit 23 is implemented by, for example, a liquid crystal display (LCD).

When the storage unit 28 is configured with a hard disk, predetermined information is written to each address based on the control by the control unit 24, and this information is read as necessary. The storage unit 28 also stores a program for executing the present invention. This program is read and executed by the control unit 24 .

The operation of the banknote demand prediction system 1 configured as described above will be described.

In the banknote demand forecasting system 1, as shown in FIG. 3A, for example, it is assumed that three or more levels of association between the reference time information and the demand level are preset and acquired. The reference time information is all information related to the time when the number of bills in the automated teller machine was acquired in advance.

The reference time information is information relating to the time when the number of each banknote in the automated teller machine, which has been acquired in advance, was checked. This reference time information may be expressed in minutes, such as "14:24 on December 12, 2020 to 14:25 on December 12, 2020", but is limited to this. Data may be converted into data in any time units such as seconds, minutes, hours, days, weeks, months, periods (eg, first half, quarter), seasons, years. In addition, this reference time information may be represented by a time interval from the start time to the end time, for example, "December 12, 2020 14:24:35". It may be shown at a certain point in time.

Also, the reference time information may be composed of a flow of a plurality of chronological points in time. For example, "December 12, 2020 14:24:35", "December 12, 2020 14:25:35", "December 12, 2020 14:26:35", etc. The measurement time points may be configured in an intermittent time series.

Furthermore, this reference time information may consist of a transaction history including chronological changes in past banknote sales.

The degree of demand may be ranked as "large", "very large", "normal", "small", "very small", etc., or may be displayed as a score, with 1000 being the most demanded. It may be indicated as high, with 0 being the lowest demand. Also, this most sought-after score may be any number other than 1000 points. Further, the degree of demand may be indicated by the number of banknotes that actually need to be replenished. Furthermore, if this demand level is represented by two values of demand and no demand, the two values may be learned as an output solution data set. The degree of demand may be obtained by evaluation based on criteria set for each automatic teller machine, or may be obtained by evaluating demand based on unified criteria. . Moreover, the degree of demand may be calculated, for example, based on the number of subsequent orders, instead of being artificially evaluated. In other words, it is possible to determine that the greater the number of dispensed banknotes, the higher the demand, and on the other hand, it is possible to determine that the lower the number of dispensed banknotes, the lower the demand. As an alternative to the degree of demand, the number of bills to be dispensed may be output for each banknote. In such a case, it is assumed that the number of dispensed bills for each bill is learned as output data.

Such a degree of demand may be extracted from quantity data such as the number of bills stored in the automatic teller machine and the number of remaining bills.

Further, the demand level may refer to deposit/withdrawal data for each ATM. For example, when there are many withdrawals in units of 10,000, such as 60,000 yen and 100,000 yen, the chances of withdrawing 10,000 yen bills are overwhelmingly high, and the demand for 10,000 yen bills is high. In addition, when there are many withdrawals in units of 1,000 yen, such as 67,000 yen, demand for 1,000 yen bills as well as 10,000 yen bills increases. In addition, when there are many withdrawals in units of 5,000 yen, such as 55,000 yen, demand for 5,000 yen bills increases. In addition, when there are many bills with a large withdrawal amount, demand for 10,000-yen bills increases. Such deposit/withdrawal data can also be obtained from an ATM. Then, from the acquired deposit/withdrawal data, the degree of demand for 10,000-yen bills, 5,000-yen bills, and 1,000-yen bills may be linked in advance according to the amount of money, as described above.

Through such reference period information and data set of the demand degree, the actual demand degree of each bill can be known from various data generated in the reference period information. In other words, various points in time described in the reference time information and the degree of demand form a data set. Therefore, by collecting reference period information and a data set of the degree of demand, it is possible to know what the degree of demand of each banknote was in what period in the past.

In the example of FIG. 3(a), it is assumed that the reference time information P01 to P03, for example, is input data for a banknote α (eg, 10,000 yen note) stored in the automated teller machine. The time information for reference as such input data is connected to the output. In this output, the degree of demand is displayed as an output solution.

The timing information for reference is associated with the degree of demand as the output solution through three or more levels of association. The timing information for reference is arranged on the left side through the degree of association, and the demand degree is arranged on the right side through the degree of association. The degree of relevance indicates which degree of demand and degree of relevance to the reference time information arranged on the left side. In other words, this degree of association is an index that indicates to what degree of demand each piece of reference time information is likely to be linked, and is an index for selecting the most probable degree of demand from the reference time information. It shows accuracy. In the example of FIG. 3A, w13 to w19 are shown as association degrees. These w13 to w19 are shown in 10 stages as shown in Table 1 below, and the closer to 10 points, the higher the degree of demand for each combination as an intermediate node and the degree of correlation with each other as an output. , conversely, the closer to one point, the lower the degree of correlation between each combination as an intermediate node and the degree of demand as an output.

The search device 2 acquires in advance the degrees of association w13 to w19 of three or more levels shown in FIG. 3(a). In other words, the search device 2 accumulates past data indicating which of the reference time information and the degree of demand in that case was adopted in determining the actual search solution, and analyzes these data. Thus, the degree of association shown in FIG. 3(a) is created.

By the way, these degrees of association may be created for each type of banknote sold by the automated teller machine. FIG. 3(a) shows the degree of association of banknotes α stored in an automated teller machine, and FIG. is the degree of association. In this way, the degree of association may be learned for each banknote α, β, γ, .

This analysis may be performed by artificial intelligence. In such a case, for example, in the case of the reference time information P01, analysis is performed from the data of the past demand level determination results. In the case of the reference time information P01, if there are many instances of the demand degree A (for example, the demand degree score is 760 points), the degree of association leading to the demand degree A is set higher, and the demand degree C ( For example, if there are many cases with a demand degree score of 433 points), the degree of association leading to this demand degree C is set higher. For example, in the example of the reference time information P01, the demand level A and the demand level C are linked. The degree is set to 2 points.

Further, the degree of association shown in FIG. 3 may be composed of nodes of a neural network in artificial intelligence. That is, the weighting coefficients for the outputs of the nodes of this neural network correspond to the degrees of association described above. Moreover, it is not limited to a neural network, and may be composed of all decision-making factors that constitute artificial intelligence.

In such a case, as shown in FIG. 4, reference time information is input as input data, each degree of demand is output as output data, and at least one or more hidden layers are provided between the input node and the output node. , may be machine-learned. Conversely, the demand level may be input and the reference time information may be output.

Such a degree of association becomes learned data in terms of artificial intelligence. After such learned data is created, the above-described learned data is used to predict the demand when actually newly determining the demand. In such a case, the timing information regarding the timing of actually predicting the degree of demand of the automatic teller machine in which banknotes for which the degree of demand is to be newly judged is stored is acquired. This period information is composed of the same kind of data as the reference period information described above.

The time information to be newly acquired is input by the information acquisition unit 9 described above. The information acquisition unit 9 may acquire such time information as electronic data.

Based on the time information newly acquired in this way, the degree of demand that is likely to be actually determined for the time information is searched for. In such a case, reference is made to the degree of association shown in FIG. 3 (Table 1), which has been acquired in advance. For example, when the newly acquired time information is the same as or similar to P02, demand B is associated with w15 and demand C with association w16. In such a case, the demand degree B with the highest degree of association is selected as the optimum solution. However, it is not essential to select the one with the highest degree of association as the optimum solution, and the demand degree C, which has a low degree of association but is recognized as having the association itself, may be selected as the optimum solution. In addition, it is of course possible to select an output solution that is not connected by an arrow, and any other priority may be used as long as it is based on the degree of association.

In this way, the degree of demand to be judged is searched from the newly acquired time information, and the user (the owner of the automatic teller machine, the bank, the employee who manages the automatic teller machine, the consultant, the person in charge of distribution, etc.) , security companies, etc.). By looking at this search result, the user can grasp the searched demand level. Incidentally, in the process of outputting the degree of demand, in addition to simply displaying the degree of demand that has been found, it is also possible to specifically draw attention or generate an alarm based on the degree of demand. Based on this demand level, the user can find out the customer's needs for each banknote, and based on the found needs, can select a lineup of banknotes to be replenished and stored in the automatic teller machine. In addition, the user can know that the higher the demand for banknotes, the fewer the remaining banknotes in the automatic teller machine. Become.

The present invention is not limited to the embodiment described above, and for example, as shown in FIG. 5, the degree of demand may be output as one set through banknotes α, β, and γ. In such a case, a data set of the reference time information and the degree of demand (the number of bills that need to be replenished) is created for each of the banknotes α, β, and γ. For example, the demand level A is composed of a combination of banknotes α: 50, banknotes β: 120, and banknotes γ: 70. The demand level B is composed of a combination of banknotes α: 130, banknotes β: 90, and banknotes γ: 100. By learning such a group from reference period information, the demand level as a search solution is similarly output in combination with the respective demand levels of banknotes α to γ.

In the present invention, instead of the demand level, the amount of deposits and withdrawals based on this deposit/withdrawal data may be learned through a data set together with reference time information. In such a case, as a search solution using the association degree, the deposit/withdrawal amount may be output instead of the demand degree. Alternatively, the deposit/withdrawal amount and the demand level of each bill may be linked in advance, and the demand level of each bill may be output according to the deposit/withdrawal amount output as the search solution. For example, if 600,000 yen is the overwhelmingly large deposit/withdrawal amount as a search solution, by linking 60 10,000 yen bills to 600,000 yen, the demand for 10,000 yen bills corresponding to this is determined. Print degrees.

In the example of FIG. 6, the input data are, for example, reference time information P01 to P03 and reference weather information P14 to P17. An intermediate node shown in FIG. 6 is obtained by combining reference weather information with reference time information as such input data. Each intermediate node is also connected to an output. In this output, each degree of demand is displayed as an output solution.

In the example of FIG. 6, it is assumed that a combination of reference time information and reference weather information is formed. The reference weather information is information about the weather at the time when the reference time information is acquired. For example, in addition to temperature and humidity data, the weather at that time (clear, cloudy, rainy, thunderstorm, snow), wind direction, etc. , typhoon movement, and atmospheric pressure distribution. Such weather information may be acquired manually, or may be acquired by importing past weather data provided by the Meteorological Agency or other private companies. If the weather is bad or cold for a long time, the number of customers going to ATMs may slow down, so these are added as explanatory variables. Incidentally, this reference weather information is not limited to the weather at the time when the reference time information is acquired, and may be the weather when learning data is actually created in relation to the degree of demand.

In the example of FIG. 6, the input data are, for example, reference time information P01 to P03 and reference weather information P14 to P17. An intermediate node shown in FIG. 6 is obtained by combining reference weather information with reference time information as such input data. Each intermediate node is also connected to an output. This output shows the degree of demand as an output solution.

Each combination (intermediate node) of reference time information and reference weather information is associated with the degree of demand as the output solution through three or more degrees of association. The time information for reference and the weather information for reference are arranged on the left side through the degree of association, and the demand degree is arranged on the right side through the degree of association. The degree of relevance indicates the degree of high relevance to each demand degree with respect to the reference time information and the reference weather information arranged on the left side. In other words, this degree of association is an index that indicates with what degree of demand each reference time information and reference weather information is likely to be associated with the degree of demand. It indicates the accuracy in selecting each probable demand degree. In addition to the time information, the degree of demand may actually change depending on the geographical situation, and decisions can be made based on this. Therefore, by combining the reference time information and the reference weather information, the optimum degree of demand is searched for.

In the example of FIG. 6, w13 to w22 are shown as association degrees. These w13 to w22 are shown in 10 stages as shown in Table 1. The closer to 10 points, the higher the degree of correlation between each combination as an intermediate node and the output. The closer it is, the lower the degree of correlation between each combination as an intermediate node and the output.

The search device 2 acquires in advance three or more degrees of association w13 to w22 shown in FIG. That is, the search device 2 accumulates past data as to which of the reference time information, the reference weather information, and the degree of demand in that case was preferable in determining the actual search solution, and stores these data. By analyzing and analyzing, the degree of association shown in FIG. 6 is created.

This analysis may be performed by artificial intelligence. In such a case, for example, in the case of reference time information P01 and reference weather information P16, the degree of demand is analyzed from past data. If there are many cases of demand A, the degree of association leading to this demand A is set higher. If there are many cases of demand B and few cases of demand A, the association leading to demand B is set. The degree of association leading to the degree of demand A is set to be low. For example, in the example of the intermediate node 61a, the outputs of demand A and demand B are linked. The degree is set to 2 points.

Further, the degree of association shown in FIG. 6 may be composed of nodes of a neural network in artificial intelligence. That is, the weighting coefficients for the outputs of the nodes of this neural network correspond to the degrees of association described above. Moreover, it is not limited to a neural network, and may be composed of all decision-making factors that constitute artificial intelligence.

In the example of the degree of association shown in FIG. 6, the node 61b is a node that combines reference weather information P14 with reference time information P01. is w16. The node 61c is a node of a combination of the reference weather information P15 and P17 with respect to the reference time information P02.

Such a degree of association becomes learned data in terms of artificial intelligence. After such learned data is created, the above-described learned data will be used when actually performing a search for determining the degree of demand. In such a case, the weather information is acquired in addition to the time information from the automatic teller machine that newly determines the degree of demand. This weather information corresponds to the reference weather information described above, and the acquisition method is also the same.

Based on the time information and weather information newly acquired in this way, the optimal demand level is searched for. In such a case, reference is made to the degrees of association shown in FIG. 6 (Table 1) that have been acquired in advance. For example, when the newly acquired time information is the same as or similar to P02, and the weather information is the same as or similar to P17, the node 61d is associated via the degree of association. , and this node 61d is associated with the degree of demand C with w19 and the degree of demand D with the degree of association w20. In such a case, the demand degree C with the highest association degree is selected as the optimum solution. However, it is not essential to select the one with the highest degree of association as the optimum solution, and the degree of demand D, which has a low degree of association but is recognized as having the association itself, may be selected as the optimum solution. In addition, it is of course possible to select an output solution that is not connected by an arrow, and any other priority may be used as long as it is based on the degree of association.

Table 2 below shows examples of degrees of association w1 to w12 extending from the input.

Intermediate nodes 61 may be selected based on degrees of association w1 to w12 extending from this input. In other words, the greater the degree of association w1 to w12, the heavier the weight in selecting the intermediate node 61 may be. However, the degrees of association w1 to w12 may all have the same value, and the weighting in selecting the intermediate nodes 61 may all be the same.

FIG. 7 shows an example in which three or more levels of correlation are set between combinations of the aforementioned reference time information and reference area information, and the degrees of demand for the combinations.

Reference area information is information on the area where the automatic teller machine for which the actual reference period information was acquired is installed, and is specified and classified by country, region, and municipality. may be The area information for reference includes, for example, a school nearby, a construction site nearby, a sports facility nearby, inside a sports facility, whether it is inside a building structure, whether it is inside a public transportation system such as a station, and whether it is windy. , the location of the area, such as heavy snowfall in winter, the presence of a downtown area nearby, various phenomena and events associated therewith, and information on the environment may be added. Also, if there is an event unique to the region, for example, the Aomori Nebuta Festival, such a regional event may be added to the reference region information. If the school is nearby, it is conceivable that there will be more opportunities for students to withdraw cash, and as a result, cash withdrawals in units of 1,000 yen may increase, so this is used as an explanatory variable.

In the example of FIG. 7, it is assumed that the input data are reference time information P01 to P03 and reference area information P18 to 21, for example. An intermediate node shown in FIG. 7 is a combination of reference time information as input data and reference region information. Each intermediate node is also connected to an output. In this output, the degree of demand is displayed as an output solution.

Each combination (intermediate node) of reference time information and reference area information is associated with the degree of demand as this output solution through three or more levels of association. The time information for reference and the area information for reference are arranged on the left side through the degree of association, and the demand degree is arranged on the right side through the degree of association. The degree of relevance indicates the degree of high degree of relevance to the demand for the reference time information and the reference area information arranged on the left side. In other words, this degree of association is an index that indicates the degree of demand that each reference period information and reference area information is likely to be associated with. It indicates the accuracy in selecting each demand degree that seems to be appropriate. In addition to time information, the degree of demand changes according to various natural environments in the area and the type and scale of local events. Therefore, the optimum degree of demand is searched for by combining the reference time information and the reference area information.

The search device 2 acquires in advance three or more degrees of association w13 to w22 shown in FIG. In other words, the search device 2 accumulates past data such as reference time information, reference area information, and the degree of demand in that case, in order to determine an actual search solution. is analyzed to create the degree of association shown in FIG.

This analysis may be performed by artificial intelligence. In such a case, for example, in the case of reference time information P01 and reference area information P20, the degree of demand is analyzed from past data. Also, the degree of association shown in FIG. 7 may be composed of nodes of a neural network in artificial intelligence.

In the example of the degree of association shown in FIG. 7, the node 61b is a node of the combination of the region information for reference P18 with the period information for reference P01. is w16. The node 61c is a node of a combination of the reference area information P19 and P21 with respect to the reference time information P02.

Such a degree of association becomes learned data in terms of artificial intelligence. After creating such learned data, the above-described learned data will be used when actually giving advice. In such a case, in addition to the time information described above, the area information of the automated teller machine for which the new demand forecast is to be made is acquired. The area information corresponds to the reference area information.

The degree of demand is searched based on the time information and area information newly acquired in this way. In such a case, reference is made to the degrees of association shown in FIG. 7 (Table 1) that have been acquired in advance. For example, when the newly acquired time information is the same as or similar to P02 and the region information is P21, the node 61d is associated via the degree of association, and this node 61d is associated with the degree of demand C with w19 and the degree of demand D with the degree of association w20. In such a case, the demand degree C with the highest association degree is selected as the optimum solution. However, it is not essential to select the one with the highest degree of association as the optimum solution, and the degree of demand D, which has a low degree of association but is recognized as having the association itself, may be selected as the optimum solution. In addition, it is of course possible to select an output solution that is not connected by an arrow, and any other priority may be used as long as it is based on the degree of association.

In the present invention, as an alternative to the reference area information shown in FIG. 7, reference traffic information shown below may be learned in association with the degree of relevance in combination with reference time information. good.

This reference traffic information indicates any information about the means of transportation to the automated teller machine or any information about the means of transportation to the area to which the automated teller machine belongs. If there is a train as one of the means of transportation to the automatic teller machine, it will show all kinds of information about the train. If additional services are provided in response to events in other areas connected by the train, information about that is also included. In addition to this, the reference traffic information includes information on delays in transportation means, information on accidents, information on congestion, and the like. In some cases, such as shopping malls located in the suburbs, automobiles may be the only means of transportation. There is

In such a case, combinations having reference time information and reference traffic information, and three or more degrees of association with the degree of demand are obtained in advance. At the time of actual judgment, the traffic information regarding the automated teller machine for which the new demand forecast is to be made is acquired. This traffic information corresponds to the above-described reference traffic information, and the acquisition method is also the same. After referring to the association degree, the demand degree is determined in the same manner as described above based on the newly acquired time information and traffic information.

In the present invention, as an alternative to the reference region information shown in FIG. 7, the reference external environment information shown below is learned in association with the degree of relevance in combination with the reference time information. good too.

The reference external environment information referred to here is various information related to the external environment information. The external environment information here includes economic data (GDP, employment statistics, industrial production index, capital investment, labor force survey, etc.), household data (household consumption survey, household data, average working hours per week, savings amount, etc.). statistical data, annual income statistical data, etc.), real estate data (office vacancy rate, unit price per tsubo, rent market, land price, vacant house data, etc.), natural environment data (disaster data, temperature data, precipitation data, wind direction data, humidity data) etc.). The external environment information includes all external information of the automated teller machine, in addition to the information reflecting part or all of these data. The external environment information for reference may categorize the external environment itself. For example, it may be classified by separating data in employment statistics. In addition, they may be categorized according to patterns (for example, patterns such as whether the growth rate of GDP increases rapidly or gradually).

In such a case, three or more levels of correlation between a combination of reference time information and reference external environment information and the degree of demand are acquired in advance. At the time of actual determination, external environment information regarding the outside of the automated teller machine subject to new demand prediction is acquired. This external environment information corresponds to the reference external environment information described above, and the acquisition method is also the same. Then, after referring to the association degree, the demand degree is determined in the same manner as described above based on the newly acquired time information and external environment information.

In the present invention, as an alternative to the reference area information shown in FIG. 7, the reference market information shown below may be learned in association with the degree of relevance in combination with the reference time information. good.

Reference market information is various information about market conditions. The market conditions referred to here may target any range, such as a single company, the entire industry including the company, the entire Japan, or the entire world. Examples of this reference market information include interest rates, exchange rates, stock prices of various brands, crude oil, futures, precious metals, bitcoins, and other price movements. This reference market condition information may be displayed as a time-series chart, a line graph, or the like for these targets. Further, information such as Bollinger band, MACD, moving average line, etc. may be attached. In addition, this market information may include the fundamental indicators of the company of each brand. Return on equity) and other indicators may be included. Information such as charts, Bollinger bands, MACD, moving averages, etc. showing price movements between currencies may also be attached to exchange rates. The reference market condition information may categorize the market potential itself. The reference market information may be information classified by type. For example, it may be classified by whether or not the stock price growth rate is 0% or more per year. In addition, they may be categorized according to patterns (for example, patterns such as whether the growth rate of the stock price increases rapidly or gradually).

In such a case, three or more levels of correlation between a combination of reference time information and reference market information and the degree of demand are acquired in advance. At the time of actual determination, the market information at the time of acquisition of the period information is acquired. This market condition information corresponds to the reference market condition information described above, and the acquisition method is also the same. Then, after referring to the association degree, the demand degree is determined in the same manner as described above based on the newly acquired time information and market information.

In the present invention, as a substitute for the reference area information shown in FIG. 8, reference traffic information shown below is learned in association with the degree of relevance in combination with reference time information. good too.

The reference traffic volume information here is information relating to the traffic volume around the automatic teller machine at the time when the reference time information is acquired. The area around the automated teller machine as used herein may be the amount of traffic on the sidewalk or aisle in front of the bank or building in which the automated teller machine is installed. The traffic volume is basically the traffic volume of people per unit time, but is not limited to this, and may be the traffic volume of vehicles. This reference traffic volume information may be manually counted traffic volume per unit time, but is not limited to this. You may make it Incidentally, this reference traffic information is not limited to the data at the time when the reference time information is obtained, and may be data used when learning data is actually created in relation to the demand level.

In such a case, a combination of reference time information and reference traffic information and three or more degrees of association with the demand level are acquired in advance. At the time of actual determination, traffic information at the time of acquisition of time information is acquired. This traffic amount information corresponds to the reference traffic amount information described above, and the acquisition method is also the same. Then, after referring to the association degree, the demand degree is determined in the same manner as described above based on the newly acquired time information and traffic volume information.

In the present invention, as an alternative to the reference area information shown in FIG. 8, reference attribute information shown below may be learned in association with the degree of relevance in combination with reference time information. good.

The reference attribute information referred to here is the occupation, student, and class of student determined from the customer's age, sex, and appearance. In actually creating learning data, a customer is photographed with a camera, and the features are extracted by image analysis. This feature extraction is obtained by image analysis of the image data as required. If necessary, deep learning technology may be used to automatically discriminate the customer's age, gender, occupation, student, and student type based on the feature values of the analysis image, and convert them into data. Combinations having such reference attribute information and reference time information related to customer attributes and three or more degrees of association with the degree of demand are acquired in advance. In the case of students, cash withdrawals in units of 1,000 yen may increase, so this is used as an explanatory variable.

At the time of actual judgment, an image of a customer who is going to purchase banknotes at an automated teller machine is imaged, and the image is analyzed in the same manner as described above to extract the characteristics of the customer. This feature extraction is obtained by image analysis of the image data as required. Attribute information is acquired by using deep learning technology as necessary to automatically determine the customer's age, gender, occupation, student, and student type based on the feature values of the analyzed image. The attribute information corresponds to the reference attribute information described above, and the acquisition method is also the same. After referring to the association degree, the demand degree is determined in the same manner as described above based on the newly acquired time information and attribute information.

In the present invention, as an alternative to the reference area information shown in FIG. 8, reference event information shown below may be learned in association with the degree of relevance in combination with reference time information. good.

The reference event information here means events such as bargain sales, bargains, and stock clearance sales held at retail stores, supermarkets, etc. near the ATM, and when there is an off-track betting booth near the ATM. All events are included, such as events such as G1 races. It also includes various events such as Christmas, the end of the year, entrance celebrations, Tanabata, and Halloween, regardless of whether they are in the vicinity of the ATM.

In such a case, three or more levels of correlation between a combination of reference time information and reference event information and the degree of demand are acquired in advance. At the time of actual determination, the event information at the time of acquisition of the period information is acquired. This event information corresponds to the reference event information described above, and the acquisition method is also the same. After referring to the association degree, the demand degree is determined in the same manner as described above based on the newly acquired time information and event information.

In addition to time information, when acquiring any two or more of weather information, attribute information, traffic volume information, market information, external environment information, regional information, event information, etc., the two or more information to be acquired 2 or more reference information (reference weather information, reference attribute information, reference traffic information, reference market information, reference external environment information, reference regional information, reference event information, etc.) By creating learning data consisting of three or more levels of correlation between a combination with reference time information and a demand level for the combination, it is possible to similarly search for a demand level solution.

In addition to time information, in addition to any one or more of weather information, attribute information, traffic volume information, market information, external environment information, regional information, event information, etc., the same applies when acquiring other information. In addition, reference weather information, reference attribute information, reference traffic information, reference market information, reference external environment information, reference regional information, reference event information, etc., according to the information to be acquired, and other A similar solution search can be performed by preparing learning data consisting of three or more levels of association between a combination of reference information corresponding to the information to be acquired and the degree of demand for the combination.

Further, according to the present invention, as shown in FIG. 8, each degree of demand is determined based on the degree of relevance of a combination of two or more kinds of information, ie, information for reference U and information for reference V. FIG. This reference information U is reference time information, and reference information V is other reference information (reference weather information, reference attribute information, reference traffic information, reference market information, reference external environment information, etc.). , area information for reference, event information for reference, etc.).

At this time, as shown in FIG. 8, the output obtained for the reference information U may be directly used as input data and associated with the output via an intermediate node 61 in combination with the reference information V. FIG. For example, after providing an output solution for the reference information U, it may be used as an input as it is, and the degree of association with other reference information V may be used to search for the output.

In the above-mentioned degree of relevance, the degree of relevance is expressed by a 10-level evaluation, but it is not limited to this, and it may be expressed by a degree of relevance of 3 or more stages. For example, 100 steps or 1000 steps may be used. On the other hand, this degree of association does not include those expressed in two stages, that is, whether or not they are associated with each other, either 1 or 0.

According to the present invention configured as described above, anyone can easily search for the degree of demand of a company that is considering financing, even without special skills or experience. Further, according to the present invention, it is possible to judge the search solution with a higher degree of accuracy than a human being does. Furthermore, by constructing the above-mentioned degree of association with artificial intelligence (neural network, etc.), it is possible to further improve the discrimination accuracy by learning this.

Since the above-described input data and output data may not be exactly the same in many cases during the learning process, information obtained by classifying these input data and output data according to type may be used. That is, the information P01, P02, . . . P15, 16, . A data set may be created between the data and the output data for learning.

Moreover, according to the present invention, it is characterized in that the optimum solution search is performed through the degree of association set to three or more stages. The degree of association can be described by a numerical value of, for example, 0 to 100% in addition to the 10 stages described above, but is not limited to this, and can be described by a numerical value of 3 or more stages. may be configured.

Based on the degree of association represented by numerical values of 3 or more levels, it is possible to discriminate the degree of demand with high credibility and low misidentification, so that multiple candidates for the possibility of search solutions can be considered. , it is also possible to search and display in descending order of the degree of association.

In addition to this, according to the present invention, it is possible to make a judgment without overlooking even a very low output discrimination result such as a correlation degree of 1%. Remind the user that even discrimination results with an extremely low degree of association are connected as slight signs, and that once in dozens or hundreds of times, the discrimination results may be useful. be able to.

Furthermore, according to the present invention, there is an advantage that a search policy can be determined by setting thresholds by performing a search based on three or more degrees of association. If the threshold value is lowered, even if the degree of association is 1%, it can be picked up without omission. be. On the other hand, if the threshold is set high, there is a high possibility that the optimal search solution can be detected with a high probability. Sometimes the solution is overlooked. It is possible to decide which one to place emphasis on based on the way of thinking of the user side and the system side, and it is possible to increase the degree of freedom in selecting such points to place emphasis on.

Furthermore, in the present invention, the degree of association described above may be updated. This update may reflect information provided via a public communication network such as the Internet, for example. Also, when knowledge, information, and data related to each piece of information are acquired, the association degree is increased or decreased according to these.

In other words, this update corresponds to learning in terms of artificial intelligence. Since new data is acquired and reflected in the learned data, it can be said that it is a learning act.

In addition to updating the degree of association based on information that can be obtained from the public communication network, the system side or the user side based on the contents of research data, papers, conference presentations, newspaper articles, books, etc. by experts It may be updated manually or automatically. Artificial intelligence may be used in these update processes.

Also, the process of initially building a trained model and the above-described updating may use not only supervised learning but also unsupervised learning, deep learning, reinforcement learning, and the like. In the case of unsupervised learning, instead of loading and learning datasets of input and output data, learning is performed by loading information corresponding to the input data, from which the degree of association related to the output data is self-formed. You can let it run.

Second Embodiment In the second embodiment, the degree of demand for banknotes is searched for the installation position of an automated teller machine.

For example, as shown in FIG. 9, three or more degrees of association are used between the reference area information about the area where the automatic teller machine is installed and the demand level actually determined in the past.

Such a degree of association becomes learned data in terms of artificial intelligence. After creating such learned data, the above-described learned data will be used when actually newly determining the degree of demand for banknotes. In such a case, area information is newly acquired. The area information is obtained by manual input, GPS measurement, acquisition of information stored in a server or the like, and the acquisition method thereof is the same as in the first embodiment.

Based on the area information newly acquired in this way, the degree of demand is determined for each bill in the automatic teller machine. In such a case, reference is made to the degrees of association shown in FIG. 9 that have been acquired in advance. A specific determination method is the same as that of the first embodiment, and thus description thereof will be omitted.

FIG. 10 shows an example in which three or more levels of association are set between combinations of the above-described reference area information and reference attribute information, and the degrees of demand for the combinations. The details of the reference attribute information are the same as in the first embodiment.

In the example of FIG. 10, it is assumed that the input data are reference area information P01-P03 and reference attribute information P18-21. An intermediate node shown in FIG. 10 is a combination of reference area information as such input data and reference attribute information. Each intermediate node is also connected to an output. In this output, the degree of demand is displayed as an output solution.

Each combination (intermediate node) of the reference area information and the reference attribute information is associated with the degree of demand as the output solution through three or more degrees of association. The area information for reference and the attribute information for reference are arranged on the left side through the degree of association, and the demand degree is arranged on the right side through the degree of association. The degree of relevance indicates the degree of high degree of relevance to the degree of demand for the reference area information and the reference attribute information arranged on the left side. In other words, this degree of association is an index that indicates the degree of demand that each reference area information and reference attribute information is likely to be associated with, and is the most reliable from the reference area information and reference attribute information. It indicates the accuracy in selecting each demand degree that seems to be appropriate. In addition to regional information, the degree of demand changes according to the customer's facial expression and behavior. Therefore, the optimum degree of demand is searched for by combining these reference area information and reference attribute information.

The search device 2 acquires in advance the degrees of association w13 to w22 of three or more stages shown in FIG. In other words, the search device 2 accumulates past data as to which of the reference area information and the reference attribute information and the degree of demand in that case was preferable in determining the actual search solution. is analyzed to create the degree of association shown in FIG.

This analysis may be performed by artificial intelligence. In such a case, for example, in the case of the reference area information P01 and the reference attribute information P20, the degree of demand is analyzed from the past data. Further, the degree of association shown in FIG. 10 may be composed of nodes of a neural network in artificial intelligence.

In the example of the degree of association shown in FIG. 10, the node 61b is a node of the combination of the attribute information for reference P18 with respect to the regional information for reference P01. is w16. The node 61c is a node of a combination of the reference attribute information P19 and P21 with respect to the reference area information P02.

Such a degree of association becomes learned data in terms of artificial intelligence. After creating such learned data, when actually giving advice from now on, the above-described learned data will be used. In such a case, in addition to the area information described above, the attribute information of the automatic teller machine targeted for new demand forecast is acquired. The attribute information corresponds to reference attribute information.

Based on the area information and attribute information newly acquired in this way, the degree of demand is searched for. In such a case, reference is made to the degrees of association shown in FIG. 10 (Table 1) that have been acquired in advance. For example, when the newly acquired area information is the same as or similar to P02 and the attribute information is P21, the node 61d is associated via the degree of association, and this node 61d is associated with the degree of demand C with w19 and the degree of demand D with the degree of association w20. In such a case, the demand degree C with the highest association degree is selected as the optimum solution. However, it is not essential to select the one with the highest degree of association as the optimum solution, and the degree of demand D, which has a low degree of association but is recognized as having the association itself, may be selected as the optimum solution. In addition, it is of course possible to select an output solution that is not connected by an arrow, and any other priority may be used as long as it is based on the degree of association.

In the present invention, instead of the reference attribute information shown in FIG. 10, the above reference information (reference weather information, reference attribute information, reference traffic information, reference market information, reference external environment information, reference region information, reference event information, etc.) may be learned in association with the degree of relevance in combination with the reference region information.

In such a case, a combination of reference area information and other reference information and three or more degrees of association with demand are obtained in advance. At the time of actual determination, weather information, attribute information, traffic information, market information, external environment information, area information, event information, etc. corresponding to the reference information are acquired in addition to the area information. Then, after referring to the degree of association, based on the newly acquired regional information and other information (weather information, attribute information, traffic information, market information, external environment information, regional information, event information, etc.), demand degree is determined in the same manner as described above.

Moreover, both the first and second embodiments are not limited to the above-described embodiments. For example, as shown in FIG. A degree of association may be used. In such a case, the solution search is performed based on three or more levels of association between the reference information corresponding to the newly acquired information and the demand. The basic reference information includes all of the above-mentioned reference information (reference time information, reference area information, reference weather information, reference attribute information, reference traffic information, reference market information, reference external environment information, reference area information, reference event information, etc.) can be applied.

In these cases, similarly, when information corresponding to reference information used as learning data is input, solution search is performed based on the above-described method.

The search solution obtained through the degree of association may be further modified or weighted based on other reference information.

The other reference information referred to here corresponds to any reference information other than the basic reference information when any of the reference information described above is used as the basic reference information.

For example, it is assumed that in certain reference time information P14, which is one of the other reference information, the demand level B was often discriminated in the past. When time information corresponding to such reference time information P14 is newly acquired, the search solution B as the degree of demand is weighted, in other words, it is connected to the search solution B as the degree of demand. It is set in advance so as to perform processing to

For example, the other reference information G is an analysis result that suggests the search solution C as a higher demand level, and the reference information F is an analysis result that suggests a search solution D as a higher demand level. Assume that there is After the setting with the reference information in this way, when the actually acquired information is the same as or similar to the reference information G, processing to increase the weighting of the degree of demand C is performed. On the other hand, when the actually acquired information is the same as or similar to the reference information F, a process of increasing the weighting of the demand degree D is performed. In other words, the degree of association itself leading to the degree of demand may be controlled based on the reference information FH. Alternatively, after determining the degree of demand based only on the degree of association described above, the obtained search solution may be corrected based on the reference information FH. In the latter case, how and with what weight the degree of demand as a search solution is modified based on the reference information F to H is to reflect what is designed on the system side each time.

The reference information is not limited to any one type, and the solution search may be performed based on two or more types of reference information. Similarly, in such a case, the discrimination type as a search solution obtained through the degree of association may be corrected to a higher value for a case that leads to the degree of demand suggested by the reference information.

Similarly, as shown in FIG. 12, even in the case of forming the degree of association with the demand level for a combination of reference information that is the keynote and other reference information, the reference information that is the keynote is Any reference information in the first and second embodiments (reference time information, reference area information, reference weather information, reference attribute information, reference traffic information, reference market information, reference external environment information, local information for reference, event information for reference, etc.) are also applicable. Other reference information includes any reference information in the first and second embodiments other than the basic reference information.

At this time, if the basic reference information is the reference area information, the other reference information includes any reference information in the first embodiment and the second embodiment.

In such a case as well, the degree of demand can be estimated by searching for solutions in the same way. At this time, as shown in FIG. 11 described above, for the search solution obtained through the degree of association, the degree of demand is corrected through other information for reference (information for reference F, G, H, etc.). can be

Also in the second embodiment, the degree of association may be learned by combining not only one piece of other reference information but also two or more pieces of reference information.

Further, as shown in FIG. 13, the degree of association may be formed between only the reference information that is the keynote and the degree of demand. This basic reference information is any reference information in the first and second embodiments (reference time information, reference area information, reference weather information, reference attribute information, reference traffic information, market information for reference, external environment information for reference, regional information for reference, event information for reference, etc.) are also applicable. Description of the solution search method in FIG. 13 will be omitted by citing the descriptions of FIGS.

1 banknote demand prediction system 2 search device 21 internal bus 23 display unit 24 control unit 25 operation unit 26 communication unit 27 estimation unit 28 storage unit 61 node

Claims (10)

In the banknote demand prediction program that predicts the demand for each banknote stored in an automatic teller machine,
an information acquisition step of acquiring regional information about the region where the automatic teller machine is installed;
Reference area information obtained in advance regarding the area where the automatic teller machine is installed and the degree of correlation of three or more levels according to the number of sales of each banknote in the automatic teller machine. and a determination step of determining the degree of demand for each of the banknotes based on the area information acquired in the information acquisition step. The information acquisition step further acquires attribute information analyzed from an image of the consumer captured through a camera installed in the automated teller machine,
In the determination step, the information acquisition step refers to the degree of association of three or more levels with a combination having the reference area information and the reference attribute information analyzed from the image of the consumer captured through the camera. 2. The banknote demand forecasting program according to claim 1, wherein the degree of demand for each banknote is determined based on the area information and the attribute information acquired in step. The information acquisition step further acquires weather information related to the weather at the time when the demand for banknotes is predicted,
In the determining step, referring to the degree of association of three or more levels of combinations having the reference area information and the reference weather information related to weather, and based on the area information and weather information obtained in the information obtaining step The banknote demand forecasting program according to claim 1, wherein the banknote demand prediction program determines the degree of demand for each banknote. The information acquisition step further acquires market information regarding market conditions at the time when the demand for banknotes is predicted,
In the determining step, referring to the degree of association in three or more stages with a combination having the reference area information and the reference market information related to market conditions, and based on the area information and the market information obtained in the information obtaining step , determining the degree of demand for the banknote. The information acquisition step further acquires external environment information related to the external environment at the time when the demand for banknotes is predicted,
In the determination step, the degree of association of three or more levels of combinations having the reference area information and the reference external environment information related to the external environment is referred to, and the area information and the external environment information acquired in the information acquisition step are referred to. The banknote demand forecasting program according to claim 1, wherein the degree of demand for the banknote is determined based on. The information acquisition step further acquires event information related to an event to be held at the time when demand for banknotes is predicted,
In the determination step, the degree of association of three or more levels of combinations having the reference area information and the reference event information related to the held event is referred to, and the area information and the event information obtained in the information obtaining step are referred to. The banknote demand forecasting program according to claim 1, wherein the degree of demand for the banknote is determined based on. In a banknote demand forecasting program that forecasts the demand for banknotes in automatic teller machines,
an information acquisition step of acquiring traffic volume information relating to the traffic volume at the installation position of the automated teller machine;
Refers to three or more levels of correlation between the reference traffic volume information obtained in advance regarding the traffic volume at the installation location of the automated teller machine and the degree of demand according to the number of sales of each banknote in the automated teller machine. and a judgment step of judging the degree of demand for each bill based on the traffic information acquired in the information acquisition step. 8. The banknote demand forecasting program according to any one of claims 1 to 7, wherein the determining step uses the degree of association corresponding to a weighting factor for each output of a neural network node in artificial intelligence. In a deposit/withdrawal amount prediction program for predicting the deposit/withdrawal amount in an automatic teller machine,
an information acquisition step of acquiring regional information about the region where the automatic teller machine is installed;
The area acquired in the information acquisition step by referring to the reference area information obtained in advance regarding the area where the automatic teller machine is installed and the degree of association of three or more levels of the deposit/withdrawal amount in the automatic teller machine. A deposit/withdrawal amount prediction program for causing a computer to execute a prediction step of predicting a deposit/withdrawal amount based on the information. In the prediction step, three or more degrees of association between the reference area information and the degree of demand for each banknote linked to the amount of deposits and withdrawals in the automatic teller machine are referred to. 10. The deposit/withdrawal amount prediction program according to claim 9, wherein the degree of demand for each banknote is determined based on the area information acquired in the information acquisition step.

Images