Program and data processing device

Abstract

To provide a program for estimating the purchasing likelihood of advertisement recipients for an advertisement target such as a product or service. [Solution] The program causes a computer to execute a process of acquiring first people flow data representing the flow of people in a specified area of ​​non-purchasers of the advertising target, and estimating the purchasing likelihood of advertising recipients for the advertising target based on the first people flow data and a trained model, and the trained model is generated by machine learning multiple training data that each define the relationship between purchasing data of actual purchasers of the advertising target and second people flow data representing the flow of people of the actual purchasers in the specified area.

Description

[Technical field]

[0001] The present invention relates to a program and a data processing device. [Background technology]

[0002] Manufacturers often want to know what kind of customers are selling their products. For example, if they can find out that the products they manufacture are mostly purchased by women in their 20s, they can use that information to develop new products.

[0003] For this reason, stores may collect ID-POS data that links customer IDs (identifiers) that uniquely identify customers with POS (Point Of Sales) data that indicates product sales results, and provide the data to manufacturers. Also, businesses that operate and provide data platforms (so-called "data platformers") may collect ID-POS data from stores and provide the data to manufacturers (see, for example, Patent Document 1).

[0004] Here, it is known that ID-POS data includes, for example, the date and time when the product was purchased, the customer ID of the customer who purchased the product, the purchased product, the unit price and number of purchased products, and the total amount of the purchased products. Customer data including the customer ID, gender, zip code of the customer's residence, and month of birth is also known (for example, see Patent Document 2). [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2023-107185 A [Patent Document 2] JP 2024-023848 A Summary of the Invention [Problem to be solved by the invention]

[0006] As mentioned above, both ID-POS data and customer data contain a customer ID. Therefore, if the store provides the ID-POS data and customer data to the manufacturer, the manufacturer can infer customer attributes from the ID-POS data based on the customer ID that is common to both the ID-POS data and the customer data. For example, the manufacturer can infer customer attributes such as gender and birth month from the ID-POS data. The manufacturer can also infer the generation in which the product was purchased as a customer attribute based on the date and month of birth when the product was purchased.

[0007] However, for customers who do not have ID-POS data, it is difficult for manufacturers to estimate the above-mentioned customer attributes. In other words, for new products that have not yet been released to the market, ID-POS data is not generated, so it is difficult for manufacturers to estimate the purchasing potential of those who have not yet purchased the new product.

[0008] In addition, even for old products that have been released to the market, ID-POS data is generated by each store, so each store can only collect ID-POS data from a portion of the actual purchasers of the old products. For this reason, it is difficult for each store or manufacturer to accurately estimate the purchase probability of all old products based on ID-POS data alone.

[0009] Furthermore, advertisements for products are often distributed to advertisement distribution targets based on such purchasing likelihood. Therefore, if the purchasing likelihood cannot be estimated, there is a possibility that appropriate advertisements for products will not be distributed to advertisement distribution targets. This possibility is not limited to products, but also applies to services.

[0010] Therefore, in one aspect, an object of the present invention is to provide a program and a data processing device that estimate the purchasing likelihood of advertisement distribution targets for an advertisement target such as a product or service. [Means for solving the problem]

[0012] In one embodiment, the program causes a computer to execute a process of acquiring first people flow data representing the flow of people in a specified area of ​​non-purchasers of the advertising target, and estimating the purchasing likelihood of advertising recipients of the advertising target based on the first people flow data and a trained model, the trained model being generated by machine learning a plurality of training data each defining a relationship between purchase data of actual purchasers of the advertising target and second people flow data representing the flow of people of the actual purchasers in the specified area.The first human flow data, the second human flow data, and the purchase data all include unique advertising identifiers held by a plurality of mobile terminals, and the process includes identifying a portion of the advertising identifier based on the degree of the purchase likelihood. Determine R, It is characterized by: In one embodiment, the program causes a computer to execute a process of acquiring first people flow data representing the flow of people in a specified area of ​​non-purchasers of the advertising target, and estimating the purchasing likelihood of advertising recipients for the advertising target based on the first people flow data and a learned model, the learned model being generated by machine learning multiple training data that each define the relationship between purchasing data of actual purchasers for the advertising target and second people flow data representing the flow of people in the specified area of ​​the actual purchasers, the purchasing data including attributes of the actual purchasers, and the process estimating either the number of purchases or the purchasing frequency of the actual purchaser based on the purchasing data, estimating the purchasing tendency of the actual purchaser based on either the number of purchases or the purchasing frequency, and assigning the purchasing tendency to the non-purchasers who have the attributes in common with the actual purchaser.

[0013] In the above configuration, the purchase data may include attributes of the actual purchaser, and the process may estimate the purchase likelihood of the advertisement distribution target persons who share the attributes with the actual purchaser.

[0015] In one embodiment, the program causes a computer to execute a process of acquiring first people flow data representing the flow of people in a specified area of ​​non-purchasers of the advertising target, generating visit history data of the non-purchasers to the specific facility included in the specified area based on the first people flow data and time data representing the time that the non-purchasers stay in the specific facility included in the specified area, estimating an advertising target identifier of the advertising target to be recommended to the non-purchaser based on the visit history data of the non-purchaser and a trained model, and estimating a purchasing possibility of an advertising delivery target for the advertising target based on an advertising target tag representing the characteristics of the advertising target identified by the advertising target identifier of the advertising target and the attributes of the non-purchaser, and The model is characterized in that it is generated by machine learning multiple training data that respectively define the relationship between tag data combining, based on a common attribute, an advertising target tag representing the characteristics of the advertising target extracted based on the purchasing data of the actual purchaser for the advertising target, a facility tag representing the characteristics of the specific facility extracted based on the visit history data of the actual purchaser, which is visit history data of the actual purchaser for the specific facility generated based on second people flow data representing the people flow in the specified area of ​​the actual purchaser and time data representing the time the actual purchaser stays at the specific facility, and the combined data of the purchaser identifier of the actual purchaser and the advertising target identifier of the advertising target.

[0016] In the above configuration, the first human flow data, the second human flow data, and the purchase data all include unique advertisement identifiers held by a plurality of mobile terminals, and the process includes identifying a part of the advertisement identifier based on the degree of the purchase possibility. Determine The configuration can be such that:

[0017] In the above configuration, the processing may be configured such that, when input or selection of a confirmation item regarding the advertisement of the advertised target is detected via a specified screen, attributes of the non-purchaser corresponding to the confirmation item are searched for, and a search result including a combination of attributes of the non-purchaser and a purchase probability that the advertised target will be purchased by the non-purchaser is output as a purchase possibility of the non-purchaser on the specified screen or on a screen separate from the specified screen.

[0018] In the above configuration, the process may be configured to obtain demographic data for the designated area, and calculate a sales volume of the advertising target in the designated area based on the demographic data and the purchase probability.

[0019] In the above configuration, the processing may be configured to obtain unit price data representing the unit price of the advertised target, and calculate the total purchase amount of the advertised target by the non-purchasers in the designated area based on the sales quantity and the unit price data.

[0020] In one embodiment, the data processing device has an acquisition unit that acquires first people flow data representing the flow of people in a specified area of ​​non-purchasers of the advertising target, and an estimation unit that estimates the purchasing possibility of an advertisement delivery target for the advertising target based on the first people flow data and a trained model, and the trained model is generated by machine learning a plurality of training data each of which defines a relationship between purchase data of an actual purchaser of the advertising target and second people flow data representing the flow of people of the actual purchaser in the specified area. the first human flow data, the second human flow data, and the purchase data all include unique advertising identifiers held by a plurality of mobile terminals, and the estimation unit identifies a portion of the advertising identifier based on the degree of the purchase likelihood. It is characterized by: In one embodiment, the data processing device has an acquisition unit that acquires first people flow data representing the flow of people in a specified area of ​​non-purchasers for the advertising target, and an estimation unit that estimates the purchasing possibility of advertising delivery targets for the advertising target based on the first people flow data and a learned model, wherein the learned model is generated by machine learning multiple training data that each define the relationship between purchasing data of actual purchasers for the advertising target and second people flow data representing the flow of people in the specified area of ​​the actual purchasers, the purchasing data includes attributes of the actual purchasers, and the estimation unit estimates either the number of purchases or the purchasing frequency of the actual purchaser based on the purchasing data, estimates the purchasing tendency of the actual purchaser based on either the number of purchases or the purchasing frequency, and assigns the purchasing tendency to the non-purchasers who have the attributes in common with the actual purchaser.

[0021] In one embodiment, the data processing device includes an acquisition unit that acquires first people flow data representing the flow of people in a specified area of ​​non-purchasers for the advertising target, a generation unit that generates visit history data of the non-purchasers for the specific facility included in the specified area based on the first people flow data and time data representing the time that the non-purchasers stay in the specific facility included in the specified area, a first estimation unit that estimates an advertising target identifier of the advertising target to be recommended to the non-purchaser based on the visit history data of the non-purchasers and a trained model, and a second estimation unit that estimates the purchasing possibility of advertising delivery targets for the advertising target based on an advertising target tag representing the characteristics of the advertising target identified by the advertising target identifier of the advertising target and attributes of the non-purchasers. A processing device, characterized in that the trained model is generated by machine learning multiple training data that respectively define the relationship between tag data combining an advertising target tag representing the characteristics of the advertising target extracted based on the purchasing data of the actual purchaser for the advertising target, a facility tag representing the characteristics of the specific facility extracted based on the visit history data of the actual purchaser, which is visit history data of the actual purchaser for the specific facility generated based on second people flow data representing the people flow in the specified area of ​​the actual purchaser and time data representing the time the actual purchaser stays at the specific facility, and a common attribute, and combined data of the purchaser identifier of the actual purchaser and the advertising target identifier of the advertising target. Effect of the Invention

[0022] According to the present invention, it is possible to estimate the purchasing likelihood of advertisement recipients for advertisement targets such as products and services. [Brief description of the drawings]

[0023] [Figure 1] 1 is an example of a data processing system. [Diagram 2] 2 is an example of a hardware configuration of a data processing server. [Diagram 3] 2 is an example of a functional configuration of a data processing server. [Figure 4]This is an example of actual purchaser flow data. [Diagram 5] 1 is an example of visit history data. [Figure 6] 13 is an example of purchaser attribute data. [Figure 7] 1 is an example of product data. [Figure 8] This is an example of ID-POS data. [Figure 9] 1 is an example of POI data. [Figure 10] This is an example of demographic data. [Figure 11] 13 is a flowchart illustrating an example of an operation of the data processing server. [Figure 12] This is an example of assigning purchasing tendencies to human flow data of non-purchasers. [Figure 13] 10 is a flowchart showing another example of the operation of the data processing server. [Figure 14] (a) is an example of a first knowledge graph. (b) is an example of a second knowledge graph. (c) is an example of a knowledge graph collection. [Figure 15] 13 is an example of a purchase potential confirmation screen. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0024] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. In the embodiment described below, a product will be described as an example of the advertisement target, but the advertisement target is not limited to a product, and may be, for example, a service such as a restaurant.

[0025] (First embodiment) 1, the data processing system ST is a computer system including a terminal device 10 and a data processing server 100. The terminal device 10 and the data processing server 100 are connected via a communication network NW. The communication network NW includes either or both of a LAN (Local Area Network) and the Internet.

[0026] In FIG. 1, a PC (Personal Computer) is shown as an example of the terminal device 10, but the terminal device 10 is not limited to a PC. The terminal device 10 may be a smart terminal such as a smartphone or a tablet terminal. In addition, in FIG. 1, a physical server device is shown as an example of the data processing server 100, but the data processing server 100 may be a virtual server device. Furthermore, in FIG. 1, one data processing server 100 is shown as an example, but multiple data processing servers 100 may be provided in the data processing system ST, and various data processing may be distributed among the multiple data processing servers 100.

[0027] The data processing system ST is used by a user 11 belonging to a business company. The business company may be a manufacturer that produces products, or a non-manufacturer that provides services. Manufacturers include, but are not limited to, food manufacturers, cosmetics manufacturers, and shoe manufacturers. Non-manufacturers include, but are not limited to, restaurants, retail stores, and the like.

[0028] A user 11 can use the data processing system ST by operating an input device 12 provided in the terminal device 10 and accessing the data processing server 100. For example, when the user 11 performs a predetermined operation on the input device 12, the control device 13 of the terminal device 10 transmits an instruction corresponding to the predetermined operation to the data processing server 100. Upon receiving the instruction, the data processing server 100 executes various data processing operations based on the received instruction and transmits the processing results to the control device 13.

[0029] Although details will be described later, for example, when the data processing server 100 receives an instruction corresponding to a predetermined operation, it estimates the purchasing likelihood of non-purchasers who have not purchased the new product, which is a product before it is released on the market, and transmits the estimated purchasing likelihood to the control device 13 as a processing result. When the control device 13 receives the processing result, it displays a predetermined screen including the processing result on the display device 14 provided in the terminal device 10. As a result, a predetermined screen appears on the display device 14. By viewing the predetermined screen, the user 11 can grasp the purchasing likelihood of non-purchasers who have not purchased the new product.

[0030] In this way, since the purchase possibility of non-purchasers can be estimated, the data processing server 100 can deliver appropriate advertisements for new products to advertisement distribution targets. In addition, even for old products that have been released on the market, the data processing server 100 allows each store or manufacturer to accurately estimate the purchase possibility of all old products based only on the ID-POS data. Therefore, appropriate advertisements for old products can also be delivered to advertisement distribution targets.

[0031] The purchasing possibility may be, for example, the number of people in a specified area who can purchase, the probability of purchase, the degree of purchase, or the amount of purchase that can be made in a specified area. The data processing server 100 may also estimate the purchasing tendency of non-purchasers, such as savers or spenders, and transmit the estimated purchasing tendency to the control device 13 as a processing result. In this case, the user 11 can grasp the purchasing tendency of non-purchasers. In this way, the data processing server 100 can estimate the purchasing possibility and purchasing tendency of non-purchasers for new and old products.

[0032] The hardware configuration of the data processing server 100 will be described with reference to Fig. 2. Note that the above-mentioned terminal device 10 basically has the same hardware configuration as the data processing server 100, and therefore a detailed description thereof will be omitted.

[0033] The data processing server 100 includes a CPU (Central Processing Unit) 100A as a processor, and a RAM (Random Access Memory) 100B and a ROM (Read Only Memory) 100C as memories. The data processing server 100 includes a network I / F (Interface) 100D and a HDD (Hard Disk Drive) 100E. Instead of the HDD (Hard Disk Drive) 100E, a SSD (Solid State Drive) may be used.

[0034] The data processing server 100 may include at least one of an input I / F 100F, an output I / F 100G, an input / output I / F 100H, and a drive device 100I, as necessary. The CPU 100A to the drive device 100I are connected to each other by an internal bus 100J. In other words, the data processing server 100 can be realized by a computer.

[0035] An input device 710 is connected to the input I / F 100F. Examples of the input device 710 include a keyboard, a mouse, and a touch panel. A display device 720 is connected to the output I / F 100G. Examples of the display device 720 include a liquid crystal display. A semiconductor memory 730 is connected to the input / output I / F 100H. Examples of the semiconductor memory 730 include a USB (Universal Serial Bus) memory and a flash memory. The input / output I / F 100H reads a program stored in the semiconductor memory 730. The input I / F 100F and the input / output I / F 100H each include a USB port, for example. The output I / F 100G includes a display port, for example.

[0036] A portable recording medium 740 is inserted into the drive device 100I. The portable recording medium 740 may be a removable disk such as a CD (Compact Disc)-ROM or a DVD (Digital Versatile Disc). The drive device 100I reads a program recorded on the portable recording medium 740. The network I / F 100D includes, for example, a LAN port and a communication circuit. The communication circuit includes either or both of a wired communication circuit and a wireless communication circuit. The network I / F 100D is connected to a communication network NW.

[0037] A program stored in at least one of ROM 100C, HDD 100E, and semiconductor memory 730 is temporarily stored in RAM 100B by CPU 100A. A program recorded in portable recording medium 740 is temporarily stored in RAM 100B by CPU 100A. By executing the stored program, CPU 100A realizes various functions described later, and also executes a data processing method including various processes described later. The program may be one according to a flowchart described later.

[0038] The functional configuration of the data processing server 100 will be described with reference to Figures 3 to 10. Note that Figure 3 shows the main functions of the data processing server 100.

[0039] 3, the data processing server 100 includes a storage unit 110, a processing unit 120, and a communication unit 130. The storage unit 110 can be realized by either or both of the RAM 100B and the HDD 100E described above. The processing unit 120 can be realized by the CPU 100A described above. The communication unit 130 can be realized by the network I / F 100D described above.

[0040] The memory unit 110, the processing unit 120, and the communication unit 130 are connected to each other. The memory unit 110 includes a people flow memory unit 111, a visit history memory unit 112, a purchaser attribute memory unit 113, and a product information memory unit 114. The memory unit 110 also includes a purchase history memory unit 115, a POI (Point Of Interest) information memory unit 116, and a demographics memory unit 117. The memory unit 110 uses the people flow memory unit 111, the visit history memory unit 112, the purchaser attribute memory unit 113, the product information memory unit 114, the purchase history memory unit 115, the POI information memory unit 116, and the demographics memory unit 117 to store various data.

[0041] The processing unit 120 includes a teacher data generating unit 121, a model generating unit 122, and a potential estimating unit 123. The processing unit 120 uses the teacher data generating unit 121, the model generating unit 122, and the potential estimating unit 123 to process various data.

[0042] The people flow storage unit 111 stores people flow data representing the people flow in the designated area of ​​actual purchasers of old products. Old products are products that were released on the market before the above-mentioned new products. The people flow storage unit 111 also stores people flow data representing the people flow in the designated area of ​​the above-mentioned non-purchasers. The people flow data of non-purchasers is an example of first people flow data, and the people flow data of actual purchasers is an example of second people flow data.

[0043] For example, as shown in FIG. 4, the human flow data of actual purchasers includes multiple items such as a personal ID (identifier), advertising ID, latitude, longitude, and measurement date and time. A unique identifier that identifies an individual actual purchaser is registered in the personal ID item. A unique identifier that is possessed by the mobile terminal of the actual purchaser and is used only for delivering advertisements within the application software (hereinafter simply referred to as an app) of the mobile terminal is registered in the advertising ID item. The mobile terminal may be a smartphone, tablet terminal, smartwatch, or game terminal.

[0044] In each of the items of latitude, longitude, and measurement date and time, the latitude and longitude of the mobile terminal measured by the GPS (Global Positioning System) function of the mobile terminal are registered together with the measurement date and time. The latitude and longitude of the mobile terminal estimated based on the radio wave intensity of a Bluetooth (registered trademark) beacon wirelessly communicated between the mobile terminal carried by the actual purchaser and a beacon terminal installed in various facilities described later and the installation location of the beacon terminal may be registered. In this way, the latitude and longitude that specify the location of the mobile terminal are periodically measured by the GPS function or the like. Therefore, if the actual purchaser moves while carrying a mobile terminal, the movement status of the actual purchaser is expressed as a people flow. Note that the people flow data of non-purchasers is basically the same as the people flow data of actual purchasers, so a detailed description will be omitted.

[0045] Returning to FIG. 3, the visit history storage unit 112 stores visit history data indicating the history of visits made by actual purchasers to POIs. POIs are an example of specific facilities, and include, for example, commercial facilities such as shops including restaurants and retail stores. POIs may include public facilities such as parks, libraries, and stations, sports facilities such as baseball fields and soccer stadiums, medical facilities such as hospitals and clinics, and roads such as sidewalks and roadways. Roads may be expressways (for example, national expressways and urban expressways) including SAs (Service Areas) and PAs (Parking Areas), or general roads other than expressways including intersections and T-junctions. POIs are not limited to such man-made objects, and may include natural objects such as mountains, rivers, and lakes. In this way, POIs correspond to specific features on map information, such as man-made objects and natural objects.

[0046] As shown in Figure 5, the visit history data contains multiple items such as personal ID, POI name, POI tag, visit date, visit start time, visit end time, and visit duration. A unique identifier that identifies an individual actual purchaser or non-purchaser is registered in the personal ID item. The POI name item is registered with the name of the POI visited by the actual purchaser or non-purchaser. The POI tag item is registered with the characteristics of the POI associated with the POI as a POI tag. For example, if word-of-mouth information has been posted about the POI on the Internet, some words contained in the word-of-mouth information are registered as POI tags. The visit date item is registered with the date on which the actual purchaser or non-purchaser visited the POI.

[0047] The start time of stay, end time of stay, and duration of stay are registered as the time when the stay at the POI started, the time when the stay at the POI ended, and the duration of stay at the POI. If an actual purchaser or a non-purchaser stays at a specific location for a certain period of time, it is estimated that the actual purchaser or non-purchaser stays at the POI set at that specific location for a certain period of time. The visit history data is generated based on the people flow data of the actual purchaser or non-purchaser and the POI data described later, and is stored in the visit history storage unit 112.

[0048] Returning to FIG. 3, the purchaser attribute storage unit 113 stores purchaser attribute data representing the attributes of the actual purchaser. As shown in FIG. 6, the purchaser attribute data includes a plurality of items such as a purchaser ID, gender, age group, and occupation. The items of the purchaser attribute data may include the postal code of the residential area, the date of birth, and the nationality. The purchaser ID item registers a unique identifier for identifying the actual purchaser or the non-purchaser. The gender item registers the gender of the actual purchaser or the non-purchaser. The age item registers an age group based on the date of birth of the actual purchaser or the non-purchaser. The occupation item registers the occupation of the actual purchaser or the non-purchaser. The purchaser attribute data is generated based on, for example, information entered when issuing a membership card that can be used at a store or information entered into a member app, and is stored in the purchaser attribute storage unit 113.

[0049] Returning to FIG. 3, the product information storage unit 114 stores product data related to the new product and the old product. As shown in FIG. 7, the product data includes a number of items, such as product ID, product name, product description, product tag, manufacturer, unit price, and price range. A unique identifier for identifying the new product or the old product is registered in the product ID item. For example, a JAN (Japanese Article Number) code may be used as the identifier registered in the product ID item. The product name item is registered with the name of the new product or the old product. The product description item is registered with a sentence explaining the characteristics of the new product or the old product.

[0050] In the product tag field, some of the words contained in the sentence registered in the product description are registered as product characteristics. The number of product characteristics registered in the product tag field may be one or more. In the manufacturer field, the name of the manufacturer that produces the new product or old product is registered. In the unit price field, the unit price of the new product or old product is registered. In the price range field, the price range of the new product or old product, such as a high price range or a low price range, is registered.

[0051] Returning to Fig. 3, the purchase history storage unit 115 stores ID-POS data that links the purchaser ID with POS data that indicates the sales record of the old product. The ID-POS data is an example of purchase data of an actual purchaser. As shown in Fig. 8, the ID-POS data includes multiple items such as purchase date and time, purchaser ID, advertisement ID, product ID, unit price, quantity, and total amount. The purchase date and time item is registered with the date and time when the actual purchaser identified by the purchaser ID purchased the old product.

[0052] The purchaser ID field registers a unique identifier that identifies the actual purchaser. The advertising ID field registers a unique identifier that is held by the actual purchaser's mobile device and is used only to deliver advertisements within the app on that mobile device. The product ID field registers a unique identifier that identifies the old product. The unit price field registers the unit price of the old product. The quantity field registers the quantity of the old product purchased by the actual purchaser. The total amount field registers the product of the unit price of the old product registered in the unit price field and the quantity of the old product registered in the quantity field.

[0053] Returning to Fig. 3, the POI information storage unit 116 stores POI data related to the above-mentioned POIs. As shown in Fig. 9, the POI data includes a number of items such as a POI-ID, a POI name, a POI tag, a latitude range, a longitude range, and a price range. A unique identifier for identifying a POI is registered in the POI-ID item. The name of the POI is registered in the POI name item. The above-mentioned POI tag is registered in the POI tag item.

[0054] The latitude range in which the POI is located is registered in the latitude range field. The longitude range in which the POI is located is registered in the longitude range field. The latitude range and longitude range registered in the latitude range and latitude range fields uniquely identify the POI's occupied area on the map information. Therefore, if the latitude and longitude included in the people flow data are included within the POI's occupied area, it is presumed that actual purchasers or non-purchasers located at that latitude and longitude have visited the POI. The price range of old and new products handled at the POI is registered in the price range field.

[0055] Returning to FIG. 3, the demographics storage unit 117 stores demographic data representing the population statistics in the designated area. As shown in FIG. 10, the demographics data includes a plurality of items such as a postal code, an area name, population by age group, and a male-female ratio. In the postal code item, a postal code that identifies the designated area is registered. In the area name item, an area name (such as a city, town, or village name) of the designated area is registered. Instead of the area name item, a station name item may be used. In this case, the station name of the designated station is registered in the station name item. In the population by age group item, the population by age group in the designated area is registered. In FIG. 10, the population in the 30s is shown as an example, and the other age groups are omitted. In the male-female ratio item, the male-female ratio in the designated area is registered. Note that the population by age group and the male-female ratio in the designated area may be obtained by using information provided by the local government that has jurisdiction over the designated area.

[0056] Returning to Fig. 3, the teacher data generation unit 121 acquires people flow data (see Fig. 4) of actual purchasers in the specified area from the people flow memory unit 111. The teacher data generation unit 121 also acquires ID-POS data (see Fig. 8) from the purchase history memory unit 115. When the teacher data generation unit 121 acquires the people flow data and ID-POS data of actual purchasers, it generates multiple teacher data that each define the relationship between the people flow data of actual purchasers and the ID-POS data.

[0057] For example, the teacher data generating unit 121 generates multiple teacher data based on an identifier of an actual purchaser that is commonly included in the person flow data of the actual purchaser and the ID-POS data. The teacher data generating unit 121 may generate multiple teacher data based on an identifier of an advertisement that is commonly included in the person flow data of the actual purchaser and the ID-POS data. In this way, the teacher data generating unit 121 can generate multiple teacher data that each define the relationship between the person flow data of the actual purchaser and the ID-POS data.

[0058] The model generation unit 122 generates a trained model by machine learning the multiple training data generated by the training data generation unit 121. Here, there is a high possibility that there is a correlation between the flow of real purchasers and the purchase of old products by the real purchasers. Therefore, the model generation unit 122 adjusts and calculates a coefficient that satisfies this correlation. This allows the model generation unit 122 to estimate the correlation between the flow of real purchasers and the purchase of old products by the real purchasers.

[0059] The potential estimation unit 123 acquires people flow data of non-purchasers in the specified area from the people flow storage unit 111. Upon acquiring the people flow data of non-purchasers, the potential estimation unit 123 estimates the purchase possibility of non-purchasers for a new product based on the people flow data of non-purchasers and the trained model generated by the model generation unit 122. Specifically, the potential estimation unit 123 acquires purchaser attribute data (see FIG. 6) from the purchaser attribute storage unit 113, and estimates the purchase possibility of non-purchasers who have attributes in common with actual purchasers.

[0060] The potential estimation unit 123 may identify a part of the advertisement identifier based on the degree of purchase likelihood, and may use the part of the identified advertisement identifier as human flow data of non-purchasers. In this way, for example, the potential estimation unit 123 can appropriately deliver an advertisement for a new product to non-purchasers who have a high degree of purchase likelihood. As a result, the purchase motivation of non-purchasers for the new product is stimulated. Note that the degree of purchase likelihood may be a degree of purchase likelihood, such as high or low purchase likelihood, or may be a numerical value such as purchase probability.

[0061] Furthermore, the potential estimation unit 123 can estimate the number of purchases and the purchasing frequency of the actual purchaser based on the ID-POS data. Therefore, the potential estimation unit 123 can estimate the purchasing tendency of a non-purchaser defined by various purchasing factors, such as saver or spender, based on either the estimated number of purchases of the actual purchaser or the estimated purchasing frequency of the actual purchaser. Note that the purchasing factors are not particularly limited to saver or spender. Various terms expressing purchasing habits may be used for the purchasing factors.

[0062] The operation of the data processing server 100 will be described with reference to FIG.

[0063] First, the teacher data generating unit 121 generates teacher data (step S1). For example, when the teacher data generating unit 121 receives an instruction corresponding to a predetermined operation, the teacher data generating unit 121 acquires the people flow data and ID-POS data of actual purchasers, and generates multiple teacher data each defining the relationship between the people flow data and the ID-POS data of actual purchasers.

[0064] After the teacher data generation unit 121 generates the teacher data, the model generation unit 122 then generates a trained model by machine learning (step S2). As described above, the model generation unit 122 generates a trained model by machine learning multiple teacher data. After the model generation unit 122 generates the trained model, the potential estimation unit 123 then acquires human flow data of non-purchasers from the human flow storage unit 111 (step S3).

[0065] When the person flow data of the non-purchaser is acquired, the potential estimation unit 123 estimates the possibility of the non-purchaser purchasing the new product based on the person flow data of the non-purchaser and the learned model (step S4). For example, as shown in Fig. 12, the potential estimation unit 123 can assign the purchasing tendency of the non-purchaser, such as saver or spender, to the person flow data of the non-purchaser identified by an identifier different from the identifier of the actual purchaser. When the purchase possibility of the non-purchaser is estimated, the potential estimation unit 123 ends the process.

[0066] Thus, according to the first embodiment, the data processing server 100 can generate a trained model by machine learning a plurality of training data each defining the relationship between the people flow data of actual purchasers and the ID-POS data. In addition, when the trained model is generated, the data processing server 100 can estimate the purchase possibility of non-purchasers for a new product based on the people flow data of non-purchasers and the trained model. By using the data processing server 100, the user 11 can grasp the purchase possibility of a new product, and a new product that is effective in the market can be developed without waste.

[0067] Second embodiment A second embodiment of the present invention will be described with reference to Fig. 13 to Fig. 15. In the first embodiment, it has been described that the purchase likelihood of non-purchasers and the purchasing tendency of non-purchasers defined by various purchasing factors such as being a saver or a spender are estimated. In the second embodiment, it will be described that the number of non-purchasers who may purchase a new product in a specified area, the purchase probability of a non-purchaser purchasing the new product, the planned purchase quantity of the new product in the specified area, etc. are estimated.

[0068] First, as shown in Fig. 13, the teacher data generation unit 121 extracts a product tag (step S11). More specifically, the teacher data generation unit 121 first acquires ID-POS data (see Fig. 8) from the purchase history storage unit 115, and extracts the identifier of the old product registered in the ID-POS data together with the individual identifier of the actual purchaser. After extracting the identifier of the old product, the teacher data generation unit 121 next acquires product data (see Fig. 7) from the product information storage unit 114, and extracts the product characteristics of the old product associated with the identifier of the old product as a product tag.

[0069] Here, the product tags may be defined by different words. However, even if the product tags are defined by different words, they may have a common concept or meaning. For this reason, the teacher data generation unit 121 attempts to unify the product tags having a common concept based on, for example, LLM (Large Language Models). For example, the teacher data generation unit 121 identifies one of the product tags having a common concept as a representative product tag. This reduces the number of product tags and improves the processing speed of subsequent processes.

[0070] After extracting the product tag, the teacher data generating unit 121 then extracts the POI tag (step S12). More specifically, the teacher data generating unit 121 first acquires the visit history data (see FIG. 5) generated based on the people flow data of the actual purchaser from the visit history storage unit 112, and extracts the POI tag registered in the visit history data together with the individual identifier of the actual purchaser.

[0071] Even if POI tags are defined by different words, they may have a common concept or meaning. For this reason, the teacher data generating unit 121 unifies POI tags that have a common concept based on the LLM and identifies a representative POI tag. This reduces the number of POI tags and improves the processing speed of subsequent processes. Note that POI tags are an example of facility tags.

[0072] After extracting the POI tag, the teacher data generating unit 121 then combines the product tag and the POI tag (step S13). For example, the teacher data generating unit 121 combines the product tag and the POI tag based on a predetermined rule defined in advance, such as CDR (Cross Domain Recommendation). For example, the teacher data generating unit 121 can combine the product tag and the POI tag based on the redundancy (overlap degree) of words included in the product tag and the POI tag, respectively. The teacher data generating unit 121 may find the word redundancy based on LLM and combine the product tag and the POI tag.

[0073] When the product tag and the POI tag are combined, the teacher data generating unit 121 generates teacher data (step S14). More specifically, the teacher data generating unit 121 generates multiple teacher data that respectively define the relationship between tag data combining the product tag and the POI tag and combination data combining the identifier of an individual actual purchaser and the identifier of an old product. This associates individual actual purchasers who have both a product domain representing the area of ​​the old product and a POI domain representing the area of ​​the POI, and identifies the purchasing preferences of the individual actual purchasers.

[0074] When the training data is generated, the model generation unit 122 generates a knowledge graph of the POIs by machine learning the training data (step S15). By using such a knowledge graph as a trained model, a product domain is inferred from a POI domain.

[0075] For example, based on the ID-POS data (see FIG. 8) and the product data (see FIG. 7), the actual purchaser P1 identified by the purchaser ID "P#001" purchased the old product "Chicken Tenderloin Bar". Also, based on the visit history data (see FIG. 5), the actual purchaser P1 identified by the personal ID "P#001" visited the POI "Fitness Gym". Here, the old product "Chicken Tenderloin Bar" is associated with the product tag "health-conscious". Also, the POI "Fitness Gym" is associated with the POI tag "health-conscious". As a result, when the model generation unit 122 generates a knowledge graph, a first knowledge graph KG1 related to the actual purchaser P1 is generated, as shown in FIG. 14(a).

[0076] Similarly, based on the ID-POS data (see FIG. 8) and the product data (see FIG. 7), the actual purchaser P2 identified by the purchaser ID "P#002" purchased the old product "cosmetics." Also, based on the visit history data (see FIG. 5), the actual purchaser P2 identified by the personal ID "P#002" visited the POI "department store." Here, the old product "cosmetics" is associated with the product tag "beautiful skin." Also, the POI "department store" is associated with the POI tag "skirt." Therefore, when the model generation unit 122 generates a knowledge graph, a second knowledge graph KG2 related to the actual purchaser P2 that is different from the first knowledge graph KG1 is generated, as shown in FIG. 14(b).

[0077] In addition, based on the visit history data, if a non-purchaser P51 identified by personal ID "P#501" visits the POI "fitness gym," the POI "fitness gym" of the non-purchaser P51 is associated with the first knowledge graph KG1, as shown in FIG. 14(a). The same is true for non-purchaser P61. Also, based on the visit history data, if a non-purchaser P52 identified by personal ID "P#502" visits the POI "department store," the POI "department store" of the non-purchaser P52 is associated with the second knowledge graph KG2, as shown in FIG. 14(b).

[0078] When the model generation unit 122 generates various knowledge graphs such as the first knowledge graph KG1 and the second knowledge graph KG2, it embeds the nodes and edges in the knowledge graphs into a three-dimensional vector space. For example, TransE is known as an embedding method. Here, in the knowledge graph, knowledge is expressed in a format called a triple, such as "for s (subject), the value (object) of r (predicate) is o." The subject (s) and the object (o) are called entities, and the predicate (r) is called a relation. An entity corresponds to a node, and a relation corresponds to an edge. For example, in this embodiment, actual purchasers P1 and P2, the POI "fitness gym," and the POI "department store" correspond to entities (or nodes). The association between entities, such as purchases and visits, corresponds to a relation.

[0079] A triple is symbolically represented as [s,r,o], and the three elements of a triple are each represented by three vectors in the embedding space. Embedding involves representing knowledge as triple data, and entities and relations as vectors. By embedding the nodes and edges in a knowledge graph into a three-dimensional vector space, a knowledge graph aggregation KB is generated that stores ternary relationships (triplets) in the knowledge graph, as shown in Figure 14(c). This type of embedding makes it possible to predict unknown triples.

[0080] Returning to Fig. 13, when the model generation unit 122 generates the knowledge graph, the potential estimation unit 123 then acquires people flow data of non-purchasers (step S16) and generates visit history data of non-purchasers (step S17). More specifically, the potential estimation unit 123 acquires people flow data of non-purchasers from the people flow storage unit 111 and acquires POI data from the POI information storage unit 116. Upon acquiring the people flow data and POI data of non-purchasers, the potential estimation unit 123 generates visit history data of non-purchasers to the POI based on time data indicating the time that non-purchasers stay at the POI, the latitude and longitude registered in the people flow data, and the latitude range and longitude range registered in the POI data.

[0081] When the visit history data of the non-purchasers is generated, the potential estimation unit 123 estimates the product ID of the new product to be recommended (step S18). More specifically, the potential estimation unit 123 estimates the product ID of the new product to be recommended to the non-purchasers P51, P52, and P61 based on the visit history data of the non-purchasers, the knowledge graph aggregate KB, and a known KGAT (Knowledge Graph Attention Network) model. The KGAT can output the purchase probability of the non-purchasers P51, P52, and P61 for the new product. The KGAT model can be referred to Non-Patent Document 1 below. <Non-Patent Document 1> Fumiyo Ito and 3 others, "A Study on Purchasing Behavior Analysis Model Based on Knowledge Graph Attention Network", Transactions of Information Processing Society of Japan Vol.63, No.1, pp.205-217, 2022-01

[0082] Once the product ID is estimated, the potential estimation unit 123 extracts a product tag for the product ID (step S19). For example, the potential estimation unit 123 accesses the product information storage unit 114 and extracts a product tag corresponding to the estimated product ID for each product ID. The potential estimation unit 123 may extract a single product tag for each product ID, or may extract multiple product tags. Once the product tags are extracted, the potential estimation unit 123 complements the knowledge graph aggregate KB based on the extracted product tags and the KGAT model.

[0083] Here, when the potential estimation unit 123 detects an input or selection of a confirmation item related to the advertisement of a product based on the operation of the user 11, it generates a matching rule (step S20). For example, as shown in FIG. 15, when a product is selected on the purchase potential confirmation screen displayed on the display device 14 and a search is instructed by the pointer Pt, the terminal device 10 transmits the confirmation item related to the advertisement of the selected product to the potential estimation unit 123. Note that the purchase potential confirmation screen is an example of a predetermined screen. Furthermore, the confirmation item is not limited to the selection of a product, and may be, for example, the selection or input of a brand name or JAN code of a product, or the selection or input of an event name or an event venue of various events. The confirmation item may be the selection or input of the age or sex of the advertisement delivery target, the selection or input of the facility name including the store selling the product, or the selection or input of the name of a region or area.

[0084] When the potential estimation unit 123 receives the confirmation item, it detects the selection of the confirmation item, and extracts the attributes of the non-purchaser from the purchaser attribute storage unit 113 based on the identifier of the individual non-purchaser registered in the human flow data of the non-purchaser not associated with the ID-POS data. When the potential estimation unit 123 extracts the attributes of the non-purchaser, it associates the attributes of the non-purchaser with the extracted product tag, the product name corresponding to the confirmation item, and the purchase probability of the new product output by the KGAT, narrows down the weight of the edge in the knowledge graph by a threshold value, and then adds the edge narrowed down to all users and purchasable products to the original knowledge graph. When the potential estimation unit 123 adds the edge to the original knowledge graph, it generates a matching rule based on the Apriori algorithm and a Bayesian network, and outputs it to the purchase potential confirmation screen. Note that the weight of the edge represents the importance of the relevance between the user and the product (for example, see FIG. 3 in Non-Patent Document 1), and the threshold value is set in advance. In addition, all users are made up of users who have visit history data but no ID-POS data, and users who have both visit history data and ID-POS data.

[0085] As a result, as shown in FIG. 15, the matching rule appears on the purchase potential confirmation screen as a matching result. For example, for a combination of attributes of a non-purchaser for the product "shoes" and a product tag indicating the characteristics of the product, the user 11 can grasp a specific product name and the purchase probability of a non-purchaser for the product with that product name. If, for example, an event name is input as a confirmation item, a combination of multiple items related to the event name appears on the purchase potential confirmation screen as a matching rule. The potential estimation unit 123 may output one optimal matching rule as a matching result, or may output multiple matching rules with high advertising effectiveness. The user 11 can grasp various matching rules and use the matching rules themselves to consider how to deliver advertisements.

[0086] After generating the matching rule, the potential estimation unit 123 then estimates the purchase possibility (step S21). For example, as shown in Fig. 15, when any of the matching results is selected by the pointer Pt on the purchase potential confirmation screen displayed on the display device 14, the terminal device 10 transmits an area designation display instruction to the potential estimation unit 123.

[0087] When the potential estimation unit 123 receives the display instruction, it outputs an area designation field for designating one of a plurality of areas to the purchasing potential confirmation screen, as shown in Fig. 15. When one of the areas is designated by the pointer Pt on the purchasing potential confirmation screen and a search is instructed by the pointer Pt, the terminal device 10 transmits an instruction to the potential estimation unit 123 to calculate the number of people who can make a purchase in the designated area.

[0088] When the potential estimation unit 123 receives the calculation instruction, it accesses the demographics storage unit 117 and calculates the number of potential purchasers by multiplying the population of the age group corresponding to the designated area by the male / female ratio based on the gender, age group, and designated area included in the selected matching result. After calculating the number of potential purchasers, the potential estimation unit 123 calculates the number of potential purchasers in the designated area by betting the purchase probability included in the matching result on the number of potential purchasers. After calculating the number of potential purchasers, the potential estimation unit 123 outputs the number of potential purchasers to a purchasing potential confirmation screen.

[0089] As a result, the number of people who can purchase in the designated area appears on the purchase potential confirmation screen as a purchasing possibility, as shown in Fig. 15. Note that the potential estimation unit 123 may multiply the number of people who can purchase by a predetermined coefficient to calculate and output the possible purchase quantity and possible purchase amount, etc. This allows the user 11 to grasp the number of people who can purchase in the designated area, the possible purchase quantity, the possible purchase amount, etc.

[0090] Furthermore, by preparing in advance past sales history of products belonging to the same category as the selected product, the potential estimation unit 123 may calculate a future sales forecast for the selected product in a specified period based on the number of possible purchasers, the sales history, and a known method. The potential estimation unit 123 outputs the future sales forecast to a purchase potential confirmation screen, allowing the user 11 to grasp the future sales of the selected product. Note that, for example, the following non-patent document 2 can be referred to as an example of a known method. <Non-Patent Document 2> Kenji Tanaka, "A sales forecasting model for new-released and nonlinear sales trend products", Expert Systems with Applications, Vol. 37, Issue. 11, pp. 7387-7393, Nov 2010

[0091] Additionally, the potential estimation unit 123 may output the processing result on a screen other than the purchase potential confirmation screen. Furthermore, when the potential estimation unit 123 detects a press of the pointer Pt on an advertisement ID download button Bt provided on the purchase potential confirmation screen, the potential estimation unit 123 may generate a list of advertisement IDs of advertisement distribution targets corresponding to the possible purchase number and download it to the terminal device 10. This enables pinpoint advertisement distribution to mobile terminals having the advertisement IDs.

[0092] As described above, according to the second embodiment, the data processing server 100 combines the product tag of the old product extracted based on the ID-POS data with the POI tag extracted based on the people flow data and visit history data of the actual purchaser based on a common attribute. The data processing server 100 also generates combination data of the purchaser ID of the actual purchaser and the product ID of the old product. The data processing server 100 can generate a knowledge graph aggregate KB by machine learning a plurality of training data that respectively define the relationship between the tag data combining the product tag and the POI tag and the combination data.

[0093] When the knowledge graph aggregate KB is generated, the data processing server 100 generates visit history data of non-purchasers to POIs based on the people flow data of non-purchasers in the designated area and the time data of non-purchasers staying at POIs included in the designated area. Then, the data processing server 100 estimates the product ID of a new product to be recommended to non-purchasers based on the visit history data and the knowledge graph aggregate KB, and can estimate the purchasing possibility of the advertisement distribution target based on the product tag of the new product identified by the product ID and the attributes of the non-purchaser. By using the purchasing possibility of the advertisement distribution target through the data processing server 100, the user 11 can grasp the number of non-purchasers for the new product, the purchase quantity, the purchase amount, etc., and can develop a new product that is effective in the market without waste. In addition, by using the purchasing possibility of the advertisement distribution target, the user 11 can utilize this purchasing possibility in the promotional activities of the entire product including the old product and the new product.

[0094] Although the preferred embodiment of the present invention has been described above in detail, the present invention is not limited to the specific embodiment, and various modifications and changes are possible within the scope of the gist of the present invention described in the claims. For example, this case may be used in a situation where a service of providing Western food is being developed as a new service in the market, when a service of providing Japanese food is already being developed as an old service in the market at a restaurant. [Explanation of symbols]

[0095] ST Data Processing System 100 Data Processing Servers 110 Storage section 111 Human flow memory department 112 Visit history memory unit 113 Purchaser attribute storage unit 114 Product information storage section 115 Purchase history memory section 116 POI information storage section 117 Demographics Memory Department 120 Processing section 121 Teacher Data Generation Unit 122 Model Generation Unit 123 Potential Estimation Unit

Claims

1. Acquire first people flow data representing the flow of people in a designated area of ​​non-purchasers of an advertising target; Estimating a purchase likelihood of an advertisement delivery target for the advertisement target based on the first people flow data and the trained model; The process is executed by a computer, The trained model is generated by machine learning a plurality of training data each of which defines a relationship between purchase data of an actual purchaser of the advertising target and second people flow data representing the people flow of the actual purchaser in the specified area; The first crowd flow data, the second crowd flow data, and the purchase data all include unique advertising identifiers respectively possessed by a plurality of mobile terminals; The process identifies a portion of the advertising identifier based on the likelihood of purchase. A program characterized by:

2. Acquire first people flow data representing the flow of people in a designated area of ​​non-purchasers of an advertising target; Estimating a purchase likelihood of an advertisement delivery target for the advertisement target based on the first people flow data and the trained model; The process is executed by a computer, The trained model is generated by machine learning a plurality of training data each of which defines a relationship between purchase data of an actual purchaser of the advertising target and second people flow data representing the people flow of the actual purchaser in the specified area; The purchase data includes attributes of the actual purchaser, The process estimates either the number of purchases or the frequency of purchases of the actual purchaser based on the purchase data, estimates a purchasing tendency of the actual purchaser based on either the number of purchases or the frequency of purchases, and assigns the purchasing tendency to the non-purchasers who have the same attributes as the actual purchaser. A program characterized by:

3. The purchase data includes attributes of the actual purchaser, The process estimates the purchasing possibility of the advertisement delivery target persons who have the attribute in common with the actual purchaser.

3. The program according to claim 1 or 2.

4. Obtaining first pedestrian flow data representing pedestrian flow in a specified area of ​​non-purchasers of the advertisement target; generating visit history data of the non-purchaser to the specific facility based on the first people flow data and time data representing the time that the non-purchaser stays at the specific facility included in the designated area; Estimating an advertisement target identifier of the advertisement target to be recommended to the non-purchaser based on the visit history data of the non-purchaser and the trained model; estimating a purchasing possibility of the advertisement distribution target for the advertisement target based on an advertisement target tag indicating a characteristic of the advertisement target identified by the advertisement target identifier of the advertisement target and the attributes of the non-purchasers; The process is executed by a computer, The trained model is generated by machine learning a plurality of training data each of which defines the relationship between tag data combining, based on a common attribute, an advertising target tag representing the characteristics of the advertising target extracted based on the purchase data of the actual purchaser for the advertising target, a facility tag representing the characteristics of the specific facility extracted based on the visit history data of the actual purchaser, which is generated based on second crowd flow data representing the crowd flow of the actual purchaser in the specified area and time data representing the time the actual purchaser stays at the specific facility, and the combined data of the purchaser identifier of the actual purchaser and the advertising target identifier of the advertising target, A program characterized by:

5. The first crowd flow data, the second crowd flow data, and the purchase data all include unique advertising identifiers respectively possessed by a plurality of mobile terminals; The process identifies a portion of the advertising identifier based on the likelihood of purchase.

5. The program according to claim 4.

6. When the process detects input or selection of a confirmation item related to the advertisement of the advertising target via a predetermined screen, the process searches for attributes of the non-purchaser corresponding to the confirmation item, and outputs a search result including a combination of attributes of the non-purchaser and a purchase probability that the advertising target will be purchased by the non-purchaser on the predetermined screen or on a screen separate from the predetermined screen as a purchase possibility of the non-purchaser.

5. The program according to claim 4.

7. The process includes acquiring demographic data for the designated area, and calculating the sales volume of the advertising target in the designated area based on the demographic data and the purchase probability.

7. The program according to claim 6 .

8. The process acquires unit price data representing a unit price of the advertising target, and calculates a total purchase amount of the advertising target by the non-purchasers in the designated area based on the sales volume and the unit price data.

8. The program according to claim 7 .

9. An acquisition unit that acquires first people flow data representing a flow of people in a specified area of ​​non-purchasers of an advertisement target; An estimation unit that estimates a purchase possibility of an advertisement delivery target for the advertisement target based on the first people flow data and a trained model; A data processing device comprising: The trained model is generated by machine learning a plurality of training data each of which defines a relationship between purchase data of an actual purchaser of the advertising target and second people flow data representing the people flow of the actual purchaser in the specified area; The first crowd flow data, the second crowd flow data, and the purchase data all include unique advertising identifiers respectively possessed by a plurality of mobile terminals; The estimation unit identifies a part of the advertisement identifier based on the degree of the purchase possibility.

23. A data processing device comprising:

10. An acquisition unit that acquires first people flow data representing the flow of people in a designated area of ​​non-purchasers of an advertising target; An estimation unit that estimates a purchase possibility of an advertisement delivery target for the advertisement target based on the first people flow data and a trained model; A data processing device comprising: The trained model is generated by machine learning a plurality of training data each of which defines a relationship between purchase data of an actual purchaser of the advertising target and second people flow data representing the people flow of the actual purchaser in the specified area; The purchase data includes attributes of the actual purchaser, The estimation unit estimates either the number of purchases or the frequency of purchases of the actual purchaser based on the purchase data, estimates a purchasing tendency of the actual purchaser based on either the number of purchases or the frequency of purchases, and assigns the purchasing tendency to the non-purchasers who have the attribute in common with the actual purchaser.

23. A data processing device comprising:

11. An acquisition unit that acquires first people flow data representing a flow of people in a specified area of ​​non-purchasers of an advertisement target; a generating unit that generates visit history data of the non-purchaser to the specific facility based on the first people flow data and time data representing the time that the non-purchaser stays in the specific facility included in the designated area; A first estimation unit that estimates an advertisement target identifier of the advertisement target to be recommended to the non-purchaser based on the visit history data of the non-purchaser and a trained model; A second estimation unit that estimates a purchase possibility of an advertisement distribution target person for the advertisement target based on an advertisement target tag that indicates a characteristic of the advertisement target identified by the advertisement target identifier of the advertisement target and an attribute of the non-purchaser; A data processing device comprising: The trained model is generated by machine learning a plurality of training data each of which defines the relationship between tag data combining, based on a common attribute, an advertising target tag representing the characteristics of the advertising target extracted based on the purchase data of the actual purchaser for the advertising target, a facility tag representing the characteristics of the specific facility extracted based on the visit history data of the actual purchaser, which is generated based on second crowd flow data representing the crowd flow of the actual purchaser in the specified area and time data representing the time the actual purchaser stays at the specific facility, and the combined data of the purchaser identifier of the actual purchaser and the advertising target identifier of the advertising target, 23. A data processing device comprising:

Images