JP7505537B2 - Information processing device, information processing system, information processing method, and program - Google Patents

Description

The present invention relates to an information processing device, an information processing system, an information processing method, and a program.

In electronic commerce and advertising on the Internet, a method is known in which so-called recommended information tailored to a user's preferences is extracted from a vast amount of information and displayed on a terminal device used by the user. The recommended information is also called recommendation information, and corresponds to, for example, product information including recommended products that are likely to match the user's preferences and should be recommended. The recommended information is identified based on, for example, a user's behavioral history, such as history information on products purchased on a website (product purchase history) and history information on viewing product introduction pages (product viewing history). A function that analyzes the behavioral history and identifies and presents the recommended information tailored to the user's preferences based on a specified algorithm is called a recommendation function, and has been used frequently in recent years.

Conventionally, a recommendation device equipped with the recommendation function has been proposed (see Patent Document 1). The recommendation device displays recommended information tailored to the preferences of customers (visitors) on tablets used by employees in stores such as retail stores.

In recent years, a payment system has been proposed that uses information codes such as barcodes and QR codes (registered trademarks) that indicate the contents and price of a product (see Patent Document 2). The mobile terminals (e.g., smartphones) used by customers who visit a store are equipped with a code reading function that reads the information codes attached to products. With this payment system, customers use their mobile terminals to read the information codes attached to the product packages, place the products in their shopping cart, and then make payment for the product price included in the read information code. Payment for the product price can be made, for example, at a cash register that can communicate with the mobile terminal, or payment can be made using electronic money registered in the mobile terminal, or online credit card payment can be used.

JP 2017-215667 A JP 2015-185089 A

However, in retail stores and other stores, the product status differs for each of the multiple types of products they handle, so a conventional recommendation function that presents recommended information determined solely by customer preferences may not be able to present to customers the products that the store (seller) wants to sell according to the product status. For example, in a store that handles a certain product A and a similar product B, if the stock of product B is greater than that of product A, the store would like to sell product B preferentially, but a conventional recommendation function cannot present product B preferentially to customers as recommended information. In particular, in supermarkets and other stores that handle products with short product life spans, such as fresh foods and prepared foods (processed foods) such as prepared dishes, the product status changes from moment to moment throughout the day, so it is desirable to propose recommended information that takes product status into account.

The object of the present invention is to provide a recommended information identification device, a recommended information identification system, a recommended information identification method, and a program that identify recommended information according to the product status, thereby enabling users who receive the recommended information to quickly obtain product information that matches their preferences, and enabling stores to increase the purchase rate of products that they wish to prioritize selling.

A recommended information identification device according to one aspect of the present invention includes a recommended information identification unit that identifies recommended information including recommended products tailored to the user from multiple product information related to the multiple products based on a user's behavior history regarding products offered in a store and a predetermined priority order regarding the sales of the multiple products offered in the store.

The recommended information identification device of the present invention further includes a ranking determination unit configured to determine the priority order by machine learning. The ranking determination unit includes a fluctuation information acquisition unit that acquires fluctuation information that causes fluctuations in the priority order, and a ranking learning unit that learns the priority order based on the fluctuation information acquired by the fluctuation information acquisition unit.

A recommended information identification system according to another aspect of the present invention includes a priority order acquisition unit, a behavior history acquisition unit, and a recommended information identification unit. The priority order acquisition unit acquires the priority order from a priority order storage unit that stores priorities for the sale of multiple products offered in a store. The behavior history acquisition unit acquires the behavior history from a behavior history storage unit that stores the behavior history of users with respect to products offered in a store. The recommended information identification unit identifies recommended information including recommended products according to the user from the multiple product information based on the behavior history acquired by the behavior history acquisition unit and the priority order acquired by the priority order acquisition unit.

A method for identifying recommended information according to another aspect of the present invention includes a behavior history acquisition step of acquiring a behavior history of a user regarding products offered in a store, and a recommended information identification step of identifying recommended information including recommended products corresponding to the user from multiple product information regarding the multiple products based on the acquired behavior history and a predetermined priority order regarding the sales of the multiple products offered in the store.

The present invention can also be understood as a program for causing a computer to execute each step of the method for identifying recommended information, or as a computer-readable recording medium on which such a program is non-temporarily recorded.

According to the present invention, by identifying recommended information according to the product status, users who are presented with the recommended information can quickly obtain product information that matches their preferences, and stores (sellers) can increase the purchase rate of products that they want to prioritize.

Furthermore, according to the present invention, by providing a ranking determination unit including a variable information acquisition unit and a ranking learning unit, the ranking learning unit learns the priorities for the sale of multiple products based on variable information that causes fluctuations in the priority order. As a result, even if the product situation in a store changes from moment to moment, recommended information is identified using the priorities determined in response to the fluctuations. As a result, users who are presented with the recommended information can quickly obtain product information that matches their preferences, and the store (seller) can increase the purchase rate of products that the store wants to sell preferentially in real time.

FIG. 1 is a diagram showing the configuration of an information processing system according to an embodiment of the present invention. FIG. 2 is a diagram showing an example of a screen displayed on a mobile terminal constituting the information processing system. FIG. 3 is a block diagram showing the configuration of each device constituting the information processing system. FIG. 4 is a block diagram showing the configuration of a ranking determination unit of a server device constituting the information processing system. FIG. 5 is a block diagram showing the configuration of the discount determination unit of the server device constituting the information processing system. FIG. 6 is a diagram showing the ranking of each product handled by a store. FIG. 7 is a flowchart illustrating an example of a procedure of a recommendation process executed in the information processing system. FIG. 8 illustrates examples of the coefficients and recommendation evaluation values when the five extracted candidate products are products with first to fifth priority orders. FIG. 9 is a flowchart showing an example of a learning operation executed by the learning unit of the information processing system.

Below, an embodiment of the present invention will be described with reference to the drawings as appropriate. Note that the embodiment described below is merely an example of the present invention and does not limit the technical scope of the present invention.

[Information processing system 100]
As shown in FIG. 1, an information processing system 100 according to an embodiment of the present invention (a recommended information identification system according to the present invention) includes a server device 11 (an example of a recommended information identification device according to the present invention), a store entrance gate device 12, a store exit gate device 13, a self-service register 14, a user terminal 15 (an example of a terminal device according to the present invention), and an employee terminal 16 used by an employee of a store 200. The gate device 12, the gate device 13, and the register 14 may each be one or more. In this embodiment, for example, an example in which the information processing system 100 is introduced in a store 200 such as a supermarket that handles a wide variety of products will be described. Of course, the store 200 in which the information processing system 100 is introduced is not limited to a supermarket, and may be any real retail store that handles a variety of products, or may be a virtual store on the Internet.

Gate device 12 is installed near the entrance of store 200, and gate device 13 is installed near the exit of store 200. User terminal 15 is a terminal device carried and used by a user who enters store 200 when purchasing an item displayed in store 200, and is, for example, a mobile terminal such as a smartphone, mobile phone, or tablet terminal. Employee terminal 16 is a terminal device used by an employee of store 200, and is, for example, a laptop computer or a mobile terminal such as a smartphone or tablet terminal.

Below, with reference to Figures 1 and 2, we will explain an example of the sequence of events from when a user visits store 200 to when he or she purchases a product and leaves the store.

First, a user who has come to the store 200 approaches the gate device 12 for entering the store from the entrance while holding a user terminal 15. When the user approaches the gate device 12, the user terminal 15 communicates with the gate device 12 by short-range wireless communication such as Bluetooth (registered trademark) or FeliCa (registered trademark). At this time, the user identification information (ID number, name, gender, age, etc.) stored in the user terminal 15 is transmitted to the gate device 12. The user identification information is information for identifying the user of the store 200. The user identification information is transferred from the gate device 12 to the server device 11 through a communication network N1 such as a wireless LAN (in-store LAN) installed in the store 200. As a result, the server device 11 recognizes that a user who has been registered in advance has come to the store. Note that if the user terminal 15 contains purchase history information of the user at the store 200, the purchase history information is also transmitted to the gate device 12 and transferred to the server device 11 through the communication network N1. Here, the purchase history information is historical information about products that the user previously purchased at the store 200.

A shopping application used in a system in which shopping is performed while scanning information codes (such as barcodes or QR codes) attached to products is pre-installed on the user terminal 15. When the user terminal 15 communicates with the gate device 12, the user terminal 15 automatically starts the shopping application. At this time, the operation display unit 22 (see FIG. 3) of the user terminal 15 displays, for example, a screen as shown in FIG. 2 (A). This enables the user to use the shopping application to purchase products within the store 200.

While shopping around the store 200, the user picks up a product he or she wishes to purchase (a product to be purchased) from the display shelf and captures (scans) the information code attached to the product using the imaging unit 24 (see FIG. 3) of the user terminal 15. The shopping application of the user terminal 15 reads the price information of the product contained in the information code (information including the product name, amount, and product ID) and displays it on the operation display unit 22 (see FIG. 3) of the user terminal 15. For example, if the user scans three products a, b, and c, the screen shown in FIG. 2 (B) containing the price information of the three products a, b, and c selected as products to be purchased is displayed on the user terminal 15. The read price information is stored in the user terminal 15 and is also sent from the user terminal 15 to the server device 11 via the communication network N1 and stored in the server device 11.

The server device 11 reads out the purchase history information (an example of the behavioral history of the present invention) corresponding to the user identification information transferred from the gate device 12 from the memory unit 32 (customer information storage unit 321) and executes the recommendation process (see FIG. 7) described below. This recommendation process is a process of identifying recommendation information (an example of the recommended information of the present invention) corresponding to the user from multiple product information related to all products (products that can be sold at the store 200) managed in the server device 11. Specifically, product information that is likely to match the user's preferences is extracted based on a predetermined algorithm, and the extracted product information is identified as the recommendation information. The identified recommendation information is then transmitted to the user terminal 15 via the communication network N1. In the user terminal 15, the recommendation information is displayed in a pop-up format or the like on the operation display unit 22 (see FIG. 3) of the user terminal 15 so that the user can easily notice it while shopping around.

When making payment (settlement) for the scanned product, the user goes to the cash register 14 and selects "Payment" (see FIG. 2(B)) while holding the user terminal 15 over the communication unit provided on the cash register 14. This causes the user terminal 15 to communicate with the cash register 14 via the short-range wireless communication, and the user identification information is sent from the user terminal 15 to the cash register 14. The user identification information sent to the cash register 14 is also transferred from the cash register 14 to the server device 11 via the communication network N1. The server device 11 then extracts the price information of the payment object corresponding to the user identification information and transfers it to the cash register 14. The cash register 14 then executes a payment process for the amount indicated by the price information based on the payment method specified by the user. When the payment process is completed, a screen indicating that the payment has been completed is displayed on the user terminal 15.

It should be noted that the settlement process may be performed online between the user terminal 15 and the server device 11 without using the cash register 14. The settlement process is performed as follows. For example, when the user selects "Payment" on the screen of FIG. 2 (B), the settlement process is started. For example, the user terminal 15 transmits the price information of the products a, b, and c to be settled together with the user identification information to the server device 11. Based on the price information, the server device 11 transmits a payment request to a settlement server (not shown) connected via the Internet. The settlement server executes the settlement process using a settlement system corresponding to the settlement business operator. After that, a message that the settlement process has been executed is transmitted to the user terminal 15 via the server device 11, and a message indicating this is displayed on the operation display unit 22 (see FIG. 3).

When the payment process is complete, the register 14 sends information about the purchased items for which payment has been completed to the server device 11. The server device 11 updates the inventory quantity of items managed by the server device 11 based on the information received from the register 14. The server device 11 also updates the purchase history information stored and managed for each user based on the actual purchase information received from the register 14.

When a user wants to leave the store 200, the user approaches the exit gate device 13. When the user approaches the gate device 13, the user terminal 15 communicates with the gate device 13 by short-range wireless communication. At this time, flag information indicating that shopping is complete is sent to the gate device 13 together with the user's user identification information, and the identification information and flag information are transferred from the gate device 13 to the server device 11 via the communication network N1. As a result, the server device 11 recognizes that shopping has been completed without any fraudulent activity and that the user has left the store.

In this way, the user can shop at store 200 using the shopping application.

As described above, in the information processing system 100, the recommendation process is executed to identify the recommendation information according to the user, and the identified recommendation information is displayed on the user terminal 15.

Below, the specific configuration of the information processing system 100 for implementing the recommendation processing will be described with reference to Figure 3.

[User terminal 15]
3, the user terminal 15 includes a control unit 21, an operation display unit 22, a storage unit 23, an imaging unit 24, and a communication I/F 25. The user terminal 15 is, for example, a mobile terminal owned by a user of the store 200, such as a smartphone, a mobile phone, or a tablet terminal. The user terminal 15 may be a mobile terminal loaned by the store 200, or may be a cart terminal installed in a shopping cart.

The imaging unit 24 is a digital camera that captures an image of a subject and outputs it as digital image data. A user shopping in the store 200 captures the information code attached to the product to be purchased with the imaging unit 24. The captured image of the information code is read by the reading processing unit 211 described below, thereby obtaining price information (product name, amount, product ID, etc.) of the product to be purchased.

The communication I/F 25 is a communication interface that connects the user terminal 15 to the communication network N1 by wire or wirelessly, and executes data communication with external devices such as the server device 11 via the communication network N1 according to a predetermined communication protocol. The communication I/F 25 also enables data communication between the user terminal 15 and devices such as the gate devices 12, 13 and the cash register 14 wirelessly (for example, short-range wireless such as Bluetooth).

The operation display unit 22 is a user interface that includes a display unit such as a liquid crystal display or an organic EL display that displays various information, and an operation unit such as a mouse, keyboard, or touch panel that accepts user operations. The recommendation information according to the user is displayed on the display unit of the operation display unit 22.

The storage unit 23 is a non-volatile storage medium such as a flash memory that stores various types of information. The storage unit 23 stores control programs that cause the control unit 21 to execute various processes.

The memory unit 23 also stores user identification information for identifying a user. The user identification information is information for identifying a user, such as an ID number, name, gender, and age. The user identification information is preregistered in the server device 11 installed in the store 200, and is stored and managed together with purchase history information in the memory unit 32 (customer information storage unit 321) within the server device 11.

The memory unit 23 also stores the price information of the product to be purchased that is read by the reading processing unit 211 described below.

The control unit 21 has control devices such as a CPU, a ROM, and a RAM. The CPU is a processor that executes various arithmetic operations. The ROM is a non-volatile storage unit in which control programs such as a BIOS and an OS for causing the CPU to execute various processes are stored in advance. The RAM is a volatile or non-volatile storage unit that stores various information, and is used as a temporary storage memory (work area) for the various processes executed by the CPU. The control unit 21 controls the user terminal 15 by having the CPU execute various control programs that are stored in advance in the ROM or the storage unit 23.

Specifically, as shown in FIG. 3, the control unit 21 includes various processing units such as a reading processing unit 211, a display processing unit 212, a notification processing unit 213, and an action result transmission unit 214. The control unit 21 functions as the various processing units by executing various processes according to the control program with the CPU. Some or all of the processing units included in the control unit 21 may be configured with electronic circuits. The control program may be a program for causing multiple processors to function as the various processing units.

The reading processing unit 211 reads the price information (product name, price, product ID, etc.) of the product contained in the information code based on the image data of the information code captured by the imaging unit 24. For ease of explanation, the price information read as a purchase target is referred to as purchase target information below. The read purchase target information is stored in the memory unit 23 until it is canceled or payment for the purchase target product is completed.

The display processing unit 212 displays the recommendation information transmitted from the server device 11 on the display unit of the operation display unit 22. For example, the display processing unit 212 displays the recommendation information on the display unit of the operation display unit 22 in a pop-up format. When the recommendation information includes a product image, a product name, etc., the display processing unit 212 displays this information on the display unit of the operation display unit 22 in a pop-up format. The product image and product name are linked to an address that stores more detailed information about the recommended product (e.g., product description, product price, product display location in the store 200), and when the user touches the display screen of the product image or product name, the display processing unit 212 reads out the detailed information stored in the address and displays it on the display unit. In addition, when the recommendation information includes a URL of a WEB site including a photo image or description of the product, the display processing unit 212 displays the URL on the display unit of the operation display unit 22. In this case, when the user touches the URL, the display processing unit 212 accesses the URL and displays the website on the display unit.

The notification processing unit 213 transmits various information to the server device 11, the gate devices 12 and 13, and the register 14. For example, the notification processing unit 213 transmits the user identification information, etc. to the gate device 12 when the user enters the store 200, and transmits the user identification information, etc. to the gate device 13 when the user leaves the store 200. The notification processing unit 213 also transmits the purchase item information read during shopping to the server device 11.

The action result sending unit 214 sends an action result indicating whether or not the user has taken an action in response to the recommendation information displayed on the display unit of the operation display unit 22 to the ranking determination unit 33 of the server device 11 (see FIG. 3). Examples of the action include an action of viewing the recommendation information displayed on the display unit of the operation display unit 22 (viewing action), an action of reading product information of an information code of a product indicated by the recommendation information by the reading processing unit 211 (reading action), or an action of purchasing a product indicated by the recommendation information displayed on the display unit (purchasing action). The action result includes a positive result indicating that each of the actions has occurred, or a negative result indicating that each of the actions has not occurred.

The control unit 21 performs a process of determining whether or not the user has taken action in response to the recommendation information displayed on the display unit of the operation display unit 22. The control unit 21 that performs such a determination process can be considered as a behavior determination unit that determines whether or not an action has been taken in response to the recommendation information.

For example, the control unit 21 performs a process of determining whether or not the product image or the product name has been touched, or whether or not the URL has been touched, during the period from when the recommendation information is displayed on the display unit of the operation display unit 22 until a certain period of time (time-up time) has elapsed. In this case, if it is determined that a touch operation has been performed within the certain period of time, the behavior result sending unit 214 sends information on a positive result (behavior result) indicating that the user has viewed the recommendation information on the user terminal 15 to the ranking determination unit 33 of the server device 11.

The control unit 21 also performs a process of determining whether or not the reading processing unit 211 has read product information of the information code of the product indicated by the recommendation information during the period until the certain time has elapsed or the payment process is performed. In this case, when it is determined that the product information of the information code of the product indicated by the recommendation information has been read, the behavior result sending unit 214 sends information of a positive result (behavior result) indicating that the user of the user terminal 15 plans to purchase the product indicated by the recommendation information to the ranking determination unit 33 of the server device 11.

The control unit 21 also performs a process of determining whether or not the product indicated by the recommendation information has been purchased. In this case, if the product indicated by the recommendation information is included in the post-purchase products for which the payment has been processed, the behavior result sending unit 214 sends information of a positive result (behavior result) indicating that the product indicated by the recommendation information has been purchased to the ranking determination unit 33 of the server device 11.

The behavior result sending unit 214 sends information on a negative result (behavior result) indicating that the behavior did not occur to the ranking determination unit 33 of the server device 11 if the browsing behavior or reading behavior was not performed within the certain period of time, if the reading behavior was not performed before the user performed the payment process at the register 14, or if the purchasing behavior was not performed during the payment process.

[Server device 11]
The server device 11 is a server computer including a control unit 31, a memory unit 32, a ranking determination unit 33 (an example of a ranking determination unit of the present invention), a discount determination unit 34 (an example of a discount determination unit of the present invention), a communication I/F 35, etc. The server device 11 is not limited to a single computer, and may be a computer system in which multiple computers operate in cooperation with each other. Furthermore, various processes executed by the server device 11 may be executed in a distributed manner by one or multiple processors.

The communication I/F 35 is a communication interface that connects the server device 11 to the communication network N1 by wire or wirelessly, and performs data communication in accordance with a specified communication protocol with external devices such as the user terminal 15, gate devices 12 and 13, and cash register 14 via the communication network N1.

The storage unit 32 is a non-volatile storage unit including a semiconductor memory, HDD (Hard Disk Drive), or SSD (Solid State Drive) that stores various types of information. In the present embodiment, the storage unit 32 is provided in the server device 11, but some or all of the various types of information in the storage unit 32 may be stored in an external device such as another server device or storage device that can communicate with the server device 11 via the communication network N1 or the Internet. In this case, the server device 11 reads out necessary information from the external device or stores information in the external device as necessary.

The storage unit 32 stores control programs for causing the control unit 31 to execute various control processes and recommendation processes (recommendation extraction process, recommendation selection process) shown in FIG. 7 and described below. For example, the control programs are non-temporarily recorded on a computer-readable recording medium such as a CD or DVD, and are read by a reading device (not shown) such as a CD drive or DVD drive provided in the server device 11 and stored in the storage unit 32.

The memory unit 32 is divided into multiple memory areas. Specifically, the memory unit 32 includes a customer information storage unit 321 (an example of a behavioral history storage unit of the present invention), a product information storage unit 322, a product to be purchased storage unit 323, and a priority order storage unit 324 (an example of a priority order storage unit of the present invention).

The customer information storage unit 321 stores information for each user (hereinafter referred to as customer information) that has been registered in advance. The customer information includes the user identification information, which includes the user's ID number, name, gender, age, etc., as well as purchase history information regarding products that the user has previously purchased at the store 200. In other words, the purchase history information is information that indicates the user's behavioral history regarding products sold at the store 200.

The customer information is stored (registered) as the user identification information in the customer information storage unit 321 of the memory unit 32, for example, when the user registers as a member of the store 200. Thereafter, when the user purchases a product in the store 200, purchase history information including product information and the purchase date relating to the purchased product is overwritten and stored in the customer information storage unit 321 while being linked to the user identification information of the purchasing user. In other words, the customer information is updated. The purchase history information is used in the recommendation extraction process by the recommendation extraction processing unit 311, which will be described later.

The product information storage unit 322 stores product information on all products that can be sold at the store 200. The product information is information that can identify the product, and includes, for example, price information including the product name, the price of the product, and the product ID (JAN code, etc.). The product information also includes information such as the type of product, the product lifespan (best-before date, expiry date, use-by date, etc.), whether or not there is a discount and the discount rate, the display location of the product, and the URL of a website that includes a photo and description of the product, and includes information that improves the convenience of users when shopping at the store 200. In addition, if the product is a cooked food (processed food) such as a side dish, the product information may include information indicating the cooking time (including the cooking date and the scheduled cooking time) that is the processing date and time of the cooked food, and the state after cooking (processing state). In addition, if the product is, for example, the target of a sales promotion campaign, the product information may include information for determining whether or not the product is the target of the sales promotion campaign. Furthermore, the product information may include information that only the seller at store 200 should know, such as the product's profit margin, the degree of risk of disposal, and the number of items in stock.

The product information storage unit 322 also stores past sales records and past disposal history for each product. The past sales records include the date and time when the product was sold in the past, the price, the number of units sold, whether or not it was discounted, the discount rate, etc. The past disposal history includes whether or not unsold products were disposed of, the date and time of past disposals, the amount disposed of, etc.

The product information stored in the product information storage unit 322 is updated whenever a change occurs. For example, the number of products in stock is updated immediately by the control unit 31 of the server device 11 when a user makes a payment (settlement) for the product. In addition, the product price, number in stock, whether or not there is a discount, the discount rate, cooking time, whether or not there is a promotional campaign, waste risk, increase in stock due to new arrivals, etc. are updated by the control unit 31 based on information entered by employees from the employee terminal 16.

The purchase candidate information is temporarily stored in the purchase candidate storage unit 323, which is sequentially transmitted from the user terminal 15 each time the reading processing unit 211 performs a reading process. This purchase candidate storage unit 323 functions as a virtual shopping basket. The purchase candidate information is stored in the purchase candidate storage unit 323 until it is canceled or payment for the purchase candidate product is completed.

The priority order storage unit 324 stores multiple product information items as well as sales priorities for the multiple products. FIG. 6 shows a ranking list TB1 indicating the priority order for each of the multiple products. Ranking list TB1 includes multiple products (hereinafter referred to as priority products) that the store 200 (seller) wishes to prioritize selling among all the products stored in the product information storage unit 322, and a sales priority order is set for each priority product. The priority order indicates the order of products to be promoted, and the product ranked first is the product that the store 200 wishes to most likely promote. The priority order may be set for each of all the products stored in the product information storage unit 322.

The priority of each priority product in the ranking list TB1 is determined based on variable information related to variable factors that cause fluctuations in the priority. The variable information is, for example, the price, type, lifespan, whether or not there is a discount, the discount rate, the cooking time of the priority product if the product is a prepared food such as a side dish, whether or not it is subject to the promotional campaign, the profit margin, the degree of waste risk, the number of items in stock, past sales performance, and past waste history. In addition, in the case where the priority product is a prepared food such as "yakitori" which is a representative side dish, the variable information includes the cooking date, cooking time, and state after cooking. In addition, the variable information includes the temperature, humidity, air pressure, and weather at the location of the store 200. In addition, the variable information includes the current time, the number of customers at the store 200 on the day, and the customer history. The customer history is statistical information indicating the number of customers by month or day in the past. In addition, the number of customers includes the expected number of customers predicted from the customer history. Such variable information can be said to be information that causes changes in the circumstances surrounding the product (product situation), including the condition of the product, the user's willingness to purchase, etc. In determining the priority order of the priority products, the variable information used in the determination process may include information on one or more of the specific variable factors described above.

The priority order varies depending on the product situation and various factors on the store 200 side (seller). For example, products with a large stock are assigned a higher priority, and products with a small stock are assigned a lower priority. This is because the more stock there is, the more likely it is that unsold products will occur. Products with a short product lifespan are assigned a higher priority, and those with a long product lifespan are assigned a lower priority. Products with a high profit margin are assigned a higher priority, and products with a low profit margin are assigned a lower priority. Products with a high risk of waste are assigned a higher priority, and products with a low profit margin are assigned a lower priority. The priority order also varies depending on the type of product. For example, ingredients and groceries that are used as cooking materials, foods (including cooked foods), and beverages are assigned a higher priority, and detergents and stationery are assigned a lower priority. Even for cooked foods such as prepared foods, the priority order varies depending on the time of cooking or the condition after cooking. If the time elapsed after cooking is long, if the waiting time until cooking is long, or if the condition after cooking is poor, the priority order is assigned a higher priority, and if the time elapsed after cooking is short, if the waiting time is short, or if the condition after cooking is good, the priority order is assigned a lower priority.

Factors that change the priority order include the price of the product, whether or not it is discounted, the discount rate, past sales, past disposal history, the temperature, humidity, and air pressure at the location of the store 200, as well as the weather, the number of customers, and the customer history. When multiple pieces of variable information exist, the above-mentioned multiple pieces of variable information may affect each other and change the priority order. For example, since many customers are expected to visit the store on a sunny day, in this case, the order of each priority product is determined by placing more importance on the impact of profit margins than on the impact of inventory levels. When the weather is good but the temperature is high, the order is determined by placing more importance on the type of product. For example, in the case of cooked foods, heated foods and cooked foods kept warm that are estimated to be difficult to sell based on past sales are placed at the top, and cold dishes that are estimated to be popular are placed at the bottom.

In this embodiment, the priority order is determined or updated by the ranking determination unit 33, which will be described later, based on the multiple pieces of change information described above. The ranking determination unit 33 will be described later.

The control unit 31 has control devices such as a CPU, a ROM, and a RAM. The CPU is a processor that executes various arithmetic processes. The ROM is a non-volatile storage unit in which control programs such as a BIOS and an OS for causing the CPU to execute various arithmetic processes are stored in advance. The RAM is a volatile or non-volatile storage unit that stores various information, and is used as a temporary storage memory (work area) for various processes executed by the CPU. The control unit 31 controls the server device 11 by having the CPU execute various control programs that are stored in advance in the ROM or the storage unit 32.

Specifically, as shown in FIG. 3, the control unit 31 includes various processing units such as a recommendation extraction processing unit 311, a recommendation selection processing unit 312, a recommendation information transmission unit 313 (an example of a recommended information output unit of the present invention), and an information acquisition unit 314 (an example of a priority acquisition unit and a behavior history acquisition unit of the present invention). Here, the recommendation extraction processing unit 311 and the recommendation selection processing unit 312 are examples of a recommended information identification unit of the present invention. The control unit 31 functions as the various processing units by executing various processes according to the control program with the CPU. In addition, some or all of the processing units included in the control unit 31 may be configured with electronic circuits. In addition, the control program may be a program for causing multiple processors to function as the various processing units.

The recommendation extraction processing unit 311 executes a recommendation extraction process to extract recommendation information that matches the user's preferences from multiple product information based on the user's purchase history information stored in the customer information storage unit 321. As a control program for executing the recommendation extraction process, the memory unit 32 stores a recommendation engine (recommendation program) that extracts the recommendation information using a predetermined, conventionally known recommendation algorithm that uses the purchase history information. The recommendation extraction processing unit 311 uses the recommendation engine to extract a preset number of two or more recommended product candidates (recommended product candidates, equivalent to candidate information in the present invention) from multiple product information. In other words, the recommendation extraction processing unit 311 executes the recommendation extraction process based on a predetermined algorithm using the recommendation engine stored in the memory unit 32.

The recommendation selection processing unit 312 executes a recommendation selection process to further select a specific recommended product candidate from the multiple recommended product candidates extracted by the recommendation extraction processing unit 311. A control program for executing the recommendation selection process is stored in the storage unit 32. In this embodiment, the recommendation selection processing unit 312 selects the specific recommended product candidate from the multiple recommended product candidates extracted by the recommendation extraction processing unit 311, the specific recommended product candidate being identified based on the priority order of the prioritized product stored in the priority order storage unit 324.

In this embodiment, a configuration in which the recommendation information identification unit of the present invention is realized by the recommendation extraction processing unit 311 and the recommendation selection processing unit 312 is illustrated, but the present invention is not limited to such a configuration. For example, it is also applicable to a configuration in which, when the recommendation extraction processing is executed, the specific recommendation information is extracted based on both the user's purchase history information stored in the customer information storage unit 321 and the priority of the prioritized product stored in the priority storage unit 324. In this case, a program that realizes the recommendation selection processing is incorporated into a recommendation engine that uses the above-mentioned conventionally known predetermined recommendation algorithm.

The recommendation information transmission unit 313 outputs recommendation information including product information of the specific recommended product candidates selected by the recommendation selection processing unit 312 to the user terminal 15 via the communication network N1. When the user terminal 15 receives the transmitted recommendation information, it displays the recommendation information on the operation display unit 22 of the user terminal 15. The recommendation information transmission unit 313 also transmits the recommendation information to the employee terminal 16.

The information acquisition unit 314 acquires the priority order for the sale of multiple products provided at the store 200 from the priority order storage unit 324. Specifically, the information acquisition unit 314 acquires the priority order of the required product from the priority order storage unit 324. More specifically, when the recommendation extraction process by the recommendation extraction processing unit 311 selects product information of multiple recommended product candidates that are estimated to match the user's preferences, the information acquisition unit 314 acquires the priority order of the selected multiple product information from the priority order storage unit 324. The acquired priority order is used by the recommendation selection processing unit 312.

In addition, the information acquisition unit 314 acquires the user's purchase history information for products offered at the store 200 from the customer information storage unit 321. Specifically, when the user identification information of the user is sent from the gate device 12 when the user visits the store, the information acquisition unit 314 extracts the customer information corresponding to the user identification information from the customer information storage unit 321, and extracts the purchase history information included in the customer information. The acquired purchase history information is transferred to the recommendation extraction processing unit 311.

[Ranking determination unit 33]
The ranking determination unit 33 will be described below with reference to Fig. 4. The ranking determination unit 33 is a learning device configured to determine the priority order by machine learning, and has a learning model for determining the priority order. Fig. 4 is a block diagram showing the configuration of the ranking determination unit 33, and shows an example in which reinforcement learning is applied as a machine learning algorithm.

As shown in FIG. 4, the ranking determination unit 33 includes a fluctuation information observation unit 331 (an example of a fluctuation information acquisition unit of the present invention), an action result acquisition unit 332 (an example of an action result acquisition unit of the present invention), a learning unit 40 (an example of a ranking learning unit of the present invention), and a data update processing unit 333. The ranking determination unit 33 can use a general-purpose CPU, but in order to enable faster calculation processing, it is desirable to use, for example, a GPGPU (General-Purpose computing on Graphics Processing Units) or a large-scale PC cluster.

The variable information observation unit 331 acquires the variable information used to determine the priority order of each priority product shown in the ranking list TB1 in FIG. 6. If the variable information is the product price, the product type, the product lifespan (best before date, expiry date, use by date, etc.), whether or not there is a discount and the discount rate, the cooking time or scheduled cooking time for cooked food, information on whether or not it is subject to a sales promotion campaign, the product's profit margin, the degree of waste risk, the number of items in stock, past sales performance, and past waste history, the variable information observation unit 331 monitors the product information in the product information storage unit 322 and constantly acquires real-time information from the product information storage unit 322.

If the fluctuation information is the temperature, humidity, and air pressure at the location of the store 200, the fluctuation information observation unit 331 receives sensor output values output from temperature sensors, humidity sensors, and air pressure sensors installed indoors or outdoors in the store 200, and obtains the real-time temperature, humidity, and air pressure indicated by the sensor output values. If the fluctuation information is the weather around the store 200, the control unit 31 downloads weather information from a weather information database connected to the server device 11 via the Internet, and the fluctuation information observation unit 331 receives the weather information from the control unit 31.

The acquired variation information is used for learning by the learning unit 40 and is stored in a memory unit (not shown) in the ranking determination unit 33. The variation information is input from the memory unit to the learning unit 40.

The behavior result acquisition unit 332 acquires the behavior result information transmitted from the behavior result transmission unit 214 of the user terminal 15. In other words, the behavior result acquisition unit 332 acquires a behavior result (positive result, negative result) indicating whether or not the user viewed or purchased the recommended product indicated by the recommendation information. The acquired behavior result information is information used for learning by the learning unit 40, and is stored in a memory unit (not shown) in the ranking determination unit 33. The behavior result is input from the memory unit to the learning unit 40.

The learning unit 40 learns the priority order based on the fluctuation information acquired by the fluctuation information observation unit 331. In this embodiment, the learning unit 40 uses not only the acquired fluctuation information but also the behavioral results acquired by the behavioral result acquisition unit 332 to machine-learn the priority order of products that the store 200 wants to sell with priority. The learning unit 40 will be described later.

The data update processing unit 333 refers to the priority order learned by the learning unit 40 and updates the priority order stored in the priority order storage unit 324, specifically, the priority order of each of the priority products shown in the ranking list TB1 in FIG. 6, to the ranking order after learning.

The learning unit 40 has the function of analyzing and extracting useful rules, knowledge expressions, judgment criteria, etc. from the set of inputted variable information, outputting the judgment results, and learning knowledge. Machine learning includes algorithms such as supervised learning, unsupervised learning, and reinforcement learning, and furthermore, to realize these methods, a method called "deep learning" is used, which learns to extract the features themselves.

Here, supervised learning is an algorithm that learns the "relationship between input and output" from data that has been input in advance. The input data is given the correct answer for that data along with the input value, and by providing a large amount of such data to the learning unit 40, the learning unit 40 learns the characteristics of the data and estimates the output (result) from the input. If the relationship between the given input and output data can be learned, it becomes possible to predict the output by applying that relationship to unknown data.

Unsupervised learning is an algorithm that learns features such as the structure, characteristics, and new knowledge of data from a large amount of input data, without being given correct output data. It differs from supervised learning in that the original data from which learning is done is not given a correct answer.

Furthermore, reinforcement learning is not based on fixed and clear data, as in supervised and unsupervised learning, but is an algorithm in which the program itself observes a given environment (current state), evaluates a series of consecutive actions, and learns appropriate actions based on the interactions between the actions and the environment, that is, actions that will maximize future rewards. Typical methods include TD learning and Q-learning.

In reinforcement learning, the results caused by actions are unknown. Therefore, when reinforcement learning is adopted as the machine learning of the learning unit 40, learning starts from a state in which the priority of each product to be learned is completely unknown, and in the initial stage, the priority obtained by learning may not be the priority desired by the store 200. Therefore, it is desirable to perform pre-learning using supervised learning, set the pre-learned priority as the initial state, and then perform reinforcement learning.

For example, when the reinforcement learning in the learning unit 40 is Q-learning, a general update formula (action value table) for the action value function Q(s, a) is expressed by the following formula (1):

In addition, in formula (1), s _t represents the environment at time t, and a _t represents the action at time t. The environment changes to s _t+1 due to the action a _t . r _t+1 is the reward obtained due to the change in the environment. The term with max is the Q value multiplied by γ when the action a with the highest Q value known at that time is selected under the environment s _t+1 . γ is also called the discount rate and is a parameter of 0<γ≦1. α is a learning coefficient in the range of 0<α≦1. When Q-learning is applied, the actions of browsing the recommendation information displayed on the user terminal 15, reading the information code of the product indicated by the recommendation information, and purchasing the product by the user (viewing action, reading action, purchasing action) are the action a _t .

As shown in FIG. 4, the learning unit 40 has a reward calculation unit 401 (an example of a reward output unit of the present invention) and a value function update unit 402 (an example of a learning model update unit of the present invention). The reward calculation unit 401 calculates and outputs a reward based on the fluctuation information acquired by the fluctuation information observation unit 331. Furthermore, the value function update unit 402 updates a value function related to the priority order based on the fluctuation information and the reward calculated by the reward calculation unit 401. This value function corresponds to the learning model.

As described above, the variable information causes a change in the priority order, and the variable information itself changes from moment to moment. Therefore, when the variable information changes, the priority order is also updated. Whether the changed priority order is appropriate or not can be evaluated based on whether the behavioral results of each of the browsing behavior, reading behavior, and purchasing behavior are positive or negative. Therefore, for example, when Q-learning is applied, the behavioral results (positive result, negative result) are essentially the behavior a _t in formula (1), and the behavior value function Q(s, a) expressed by formula (1) is used as a value function for changing the behavior a _t .

In this embodiment, the reward calculation unit 401 outputs a positive reward when the behavior result is a positive result indicating that the recommended product indicated by the recommendation information was viewed or purchased, and outputs a negative reward when the behavior result is a negative result indicating that the recommended product was not viewed or purchased. The absolute value of the reward that is output may vary depending on the content of the behavior result. For example, the reward for the behavior result of browsing behavior may be set to the lowest, and the reward corresponding to the behavior result of purchasing behavior may be set to the highest. In addition, a weighting coefficient for the reward may be set for each user. For example, when placing importance on the reward for the behavior result of a user who frequently uses the store 200, the weighting coefficient by which the reward is multiplied may be set to a value proportional to the frequency of use.

[Discount determination unit 34]
The discount determination unit 34 will be described below with reference to FIG. 5. The discount determination unit 34 is a learning device configured to determine by machine learning whether to discount a low-ranked product whose priority is lower than a predetermined reference rank (10th in this embodiment) among a plurality of products of the same type. Here, FIG. 5 is a block diagram showing the configuration of the discount determination unit 34, and shows an example in which reinforcement learning is applied as a machine learning algorithm. Note that the configuration of the discount determination unit 34 is similar to the configuration of the ranking determination unit 33 except for the judgment result notification unit 341, so that only the different configuration will be described in detail below, and the same configuration will be denoted by the same reference numerals as the ranking determination unit 33, and detailed description thereof will be omitted.

In this embodiment, the discount does not only include a further reduction in the selling price of the target product, but also includes an action equivalent to a substantial discount on the target product. For example, the provision of a coupon (voucher, discount voucher) when the target product is purchased, or the provision of points that can be exchanged for products, also corresponds to the discount.

As shown in FIG. 5, the discount determination unit 34 includes a fluctuation information observation unit 331 (an example of a deterioration information acquisition unit of the present invention), an action result acquisition unit 332 (an example of a action result acquisition unit of the present invention), a learning unit 40 (an example of a judgment learning unit of the present invention), and a judgment result notification unit 341. In other words, the discount determination unit 34 differs from the ranking determination unit 33 in that it includes a judgment result notification unit 341 instead of a data update processing unit 333.

For example, consider a case in which, in the ranking list TB1 of FIG. 6, there are two products of the same type, "yakitori," ranked within the top 10 (reference ranking), and one product is also ranked 19th, which is lower than the 10th place. For example, the product "yakitori" ranked within the top 10 has a high priority, so it is likely to be notified to users shopping around the store 20 as the recommendation information, and is more likely to be purchased by users than the products "yakitori" with other rankings. On the other hand, the product "yakitori" ranked 19th, which is lower than the reference ranking, (corresponding to a low-ranked product) has a low ranking and is therefore less likely to be notified to users as the recommendation information. In this case, even if the product with a low ranking (low-ranked product) is not currently a product that the store 200 (seller) wants to sell with priority, depending on the content of the change in the change information, it may be necessary to improve sales without waiting for the ranking to be increased or decreased (ranking change) after the ranking is determined by the ranking determination unit 33.

For this reason, in the discount determination unit 34, the fluctuation information observation unit 331 acquires fluctuation information as described above, but acquires the fluctuation information as deterioration information that causes fluctuations in the deterioration of the product. In addition, the learning unit 40 performs machine learning based on the fluctuation information acquired by the fluctuation information observation unit 331 whether or not to implement a discount on a product (low-ranked product) that has a lower priority than the reference priority among multiple products of the same type.

Incidentally, the products handled in the store 200 include short-life products that cannot be stored for a long time, such as fresh foods and prepared foods. In addition, products with long expiration dates or expiration dates, such as household goods such as detergents and stationery, processed foods such as canned foods and instant noodles that can be stored for a long time, confectioneries, luxury items, seasonings, and alcoholic beverages such as liqueurs, are also included. Therefore, even if the low-ranked products have a long life, such as household goods, there is little need to discount them even if they are in large stock. On the other hand, for short-life products such as fresh foods and prepared foods, there is a high need to sell them out even if they are close to their end of life, even if the stock is small, in order to avoid the risk of unsold products being discarded. Therefore, it is preferable that the discount determination unit 34 extracts short-life products with a life shorter than a predetermined threshold from the low-ranked products and performs machine learning to determine whether or not to discount the short-life products. That is, in this embodiment, the discount determination unit 34 performs machine learning to determine whether or not to discount the low-ranked products by excluding products whose life span is longer than the predetermined threshold. Note that the predetermined threshold is set arbitrarily depending on the target product, and is set to, for example, one day for prepared foods and several days for fresh foods depending on the type of food.

The judgment result notification unit 341 performs a process of transmitting the judgment result to the employee terminal 16 by referring to the judgment result learned by the learning unit 40. The employee terminal 16 receives the judgment result and displays it on the display unit of the terminal 16. Therefore, by checking the displayed judgment result, the employee of the store 200 can recognize which products should be discounted and can quickly discount the products, that is, update (replace) the information code.

In the above example, the reference ranking is set to 10th place, but for example, the discount determination unit 34 may learn whether or not to discount each product other than the product with the highest ranking among multiple products of the same type. In this case, the ranking of the product with the highest ranking among multiple products of the same type corresponds to the reference ranking of the present invention.

In addition, the above example illustrates a discount on a product (item) among multiple products of the same type whose priority is lower than the reference priority, but for example, the discount determination unit 34 may learn whether or not to discount each of the products (low-priority products) among all products whose priority is lower than the reference priority.

In addition to determining whether or not to discount the product, the learning unit 40 of the discount determination unit 34 may learn, for example, the discount rate and the timing of discount implementation if it is determined that a discount should be implemented.

The following describes the recommendation process executed by the control unit 31 of the server device 11, as well as the recommended information identification method of the present invention, using the flowchart in FIG. 7. In the following description, it is assumed that the recommendation process is performed between the time the user enters the store 200 and passes through the gate device 12, and the time the user leaves the store and passes through the gate device 13. In other words, the recommendation process is performed while the user is shopping around the store 200. In FIG. 7, S11, S12, ... indicate the numbers (step numbers) of the processing procedure.

First, in step S11, the control unit 31 acquires purchase history information of the user who entered the store from the customer information storage unit 321. This process is performed by the information acquisition unit 314 of the control unit 31. Step S11 corresponds to the behavior history acquisition step of the present invention. In addition, in the next step S12, the control unit 31 extracts the user's behavior details from the acquired purchase history information.

Next, in step S13, the control unit 31 uses the recommendation engine to perform a process of extracting multiple candidates for recommended products that are likely to match the user's preferences based on the behavioral content. This process is performed by the recommendation extraction processing unit 311 of the control unit 31. In this embodiment, the recommendation extraction processing unit 311 determines, for example, five candidates.

In the next step S14, the control unit 31 determines whether the multiple candidates extracted in step S13 include a top-ranked product with a high priority in the ranking list TB1 of FIG. 6. For example, the top-ranked product may be a product whose priority is within a predetermined reference ranking (10th in this embodiment).

If it is determined in step S14 that the top product is included in the multiple candidates (Yes side of S14), then in the next step S15, the top product is selected from the multiple candidates and the selected product is identified as the recommended product. This process is performed by the recommendation selection processing unit 312 of the control unit 31. Step S15 corresponds to the recommended information identification step of the present invention. Thereafter, the recommendation information including the recommended product is transmitted to the user terminal 15 of the user by the recommendation information transmission unit 313. In the user terminal 15, the received recommendation information is displayed on the display unit of the operation display unit 22. The recommendation information is also transmitted to the employee terminal 16.

On the other hand, if it is determined in step S14 that the top product is not included in the multiple candidates (No in S14), the control unit 31 again uses the recommendation engine to perform a process of extracting multiple candidates different from the candidates extracted in step S13 based on the behavioral content (step S17). Thereafter, the process of step S14 is executed, and the processes of steps S14 and S17 are repeated until it is determined that the top product is included in the multiple candidates.

As described above, in this embodiment, the recommendation information that matches the user's preferences is displayed on the user terminal 15, so that the user who is presented with the recommendation information can quickly obtain product information (information on recommended products) that matches his or her preferences. In addition, the recommendation information output to the user terminal 15 includes information on recommended products that takes into account the priority order of products that the store 200 (seller) wants to sell preferentially. This increases the likelihood that the recommended products will be viewed or purchased, and as a result, the store (seller) can increase the purchase rate of products that it wants to sell preferentially.

The recommendation information is also sent to the employee terminal 16, so that the employee can check the recommendation information for each user that he or she has received.

In the above embodiment, an example in which a recommended product is identified by performing the processes of steps S14 and S15 has been described, but the present invention is not limited to such an example. For example, the product with the highest recommendation evaluation value, which will be described later, may be identified as the recommended product from among the multiple candidates extracted in step S13.

The recommendation evaluation value can be, for example, as shown in FIG. 8, a numerical value Z1·Z2 obtained by multiplying a predetermined compatibility coefficient Z1 and a ranking coefficient Z2 for each of the five extracted candidate products A1 to A5. Here, the compatibility coefficient Z1 is a numerical representation of the compatibility (preference) between the user and each of the candidate products A1 to A5, and is calculated, for example, by Probabilistic Latent Semantic Analysis (PLSA) based on data such as the user's purchase history information and product information.

The ranking coefficient Z2 is an arbitrary coefficient determined based on a predetermined rule. In this embodiment, an arbitrary value "0.50" is set for the first-ranked product in the ranking list TB1 (see FIG. 6), a value "0.25" that is half the coefficient of the first-ranked product is set for the second-ranked product, a value "0.16" that is one-third the coefficient of the first-ranked product is set for the third-ranked product, a value "0.13" that is one-quarter the coefficient of the first-ranked product is set for the fourth-ranked product, and a value "0.10" that is one-fifth the coefficient of the first-ranked product is set for the fifth-ranked product. The rule for setting the ranking coefficient Z2 is not limited to the above rule, as long as the ranking coefficient Z2 is set to a smaller value from higher to lower. Note that each of the above-mentioned values is displayed by rounding down to the third decimal place. Here, FIG. 8 illustrates each coefficient and the recommendation evaluation value when the five candidate products extracted in step S13 are products with priority rankings of 1 to 5.

In the recommendation process, instead of the processes of steps S14 and S15, the control unit 31 identifies the product with the largest recommendation evaluation value determined as described above as the recommended product. By identifying the recommended product in this manner, it is possible to prevent the same product from being presented as a recommended product to the same user when there is no change in the ranking of the top products.

In addition, instead of identifying the product with the highest recommendation evaluation value as the recommended product, for example, for each of the five extracted candidate products A1 to A5, the higher the recommendation evaluation value, the higher the selection rate may be set, and the recommended product may be identified probabilistically based on each selection rate.

The recommendation evaluation value may be obtained by multiplying the compatibility coefficient Z1 and the ranking coefficient Z2 to obtain a value Z1·Z2, and further multiplying the value Z1·Z2 by a noise coefficient Z3 randomly determined for each of the five candidate products A1 to A5. This noise coefficient Z3 is a coefficient for varying the recommendation evaluation value. The noise coefficient Z3 is multiplied by the values Z1·Z2, which also affects the recommended products identified. In other words, the recommended products identified may change before and after the noise coefficient Z3 is multiplied. The timing for multiplying the noise coefficient Z3 may be, for example, when one or more of the following conditions are met: the cumulative number of times the user has entered the store has reached a predetermined number; the number of times the user has entered the store in the most recent set period has reached a predetermined number; the user has revisited the store after a predetermined period has passed; or the user has not taken a positive action (viewing, reading, or purchasing) for the recommended products indicated by the presented recommendation information a predetermined number of times in a row. By adding such a noise coefficient Z3 to the recommendation evaluation value, it is possible to add diversity to the recommendation information including the identified recommended products.

Such diverse recommendation information does not necessarily include the recommended products reflected in the user's purchase history. However, when a user views recommendation information that does not include a recommended product, that recommendation information triggers the user to purchase a product indicated by the recommendation information, and the purchase record of such a product is overwritten in the user's purchase history. As a result, at the next timing, the recommendation information including the recommended product is identified using the overwritten purchase history information, and the consistency between the recommended products indicated by the recommendation information and the products preferred by the user is improved.

In addition, instead of using the compatibility coefficient Z1, the recommendation evaluation value may be a value Z2·Z3 obtained by multiplying the ranking coefficient Z2 by the noise coefficient Z3.

Below, an example of the learning operation flow of the learning unit 40 of the ranking determination unit 33 using reinforcement learning will be described with reference to the flowchart in FIG. 9. In FIG. 9, S21, S22, ... indicate numbers (step numbers) of the processing procedure. The learning operation of the learning unit 40 in the discount determination unit 34 is the same as the learning operation flow shown in FIG. 9, and therefore will not be described.

First, in step S21, the behavior result acquisition unit 332 of the ranking determination unit 33 acquires the behavior results (behavior results of browsing behavior, reading behavior, and purchasing behavior) transmitted from the user terminal 15. In this step S21, the behavior results of all users who use the store 200 are transmitted from each of the user terminals 15.

In step S22, the reward calculation unit 401 of the learning unit 40 calculates a judgment evaluation value for judging whether the current priority order is appropriate based on the behavioral results acquired in step S21. The judgment evaluation value is calculated by adding up the evaluation points set for each content of the behavioral results. For example, positive evaluation points are set for the positive results of browsing, reading, and purchasing, and negative evaluation points are set for the negative results. The judgment evaluation value is calculated by adding up the evaluation points for all behavioral results. Note that if a weighting factor is set for the evaluation points for each user, for example, if a weighting factor proportional to the frequency of use is set for the evaluation points for the behavioral results of users who frequently use the store 200, the judgment evaluation value may be calculated by taking that weighting factor into account.

In the next step S23, it is determined whether the judgment evaluation value is within a predetermined allowable range. If it is determined that the judgment evaluation value is within the allowable range, the current priority is deemed appropriate, and the process proceeds to step S24, where a positive reward is set. On the other hand, if it is determined that the judgment evaluation value is outside the allowable range, the current priority is deemed inappropriate, and the process proceeds to step S25, where a negative reward is set.

Then, in step S26, the value function update unit 402 of the learning unit 40 updates the value function related to the priority order, i.e., the action value function Q(s, a), and the process returns to step S21 to repeat the same process.

In the above embodiment, an information processing system 100 applied to a real store 200 is exemplified, but the present invention can also be applied to a virtual store on the Internet where users shop via a user terminal 15.

The above-described embodiment includes the following additional features.
(1) A recommended information identification device comprising: a recommended information identification unit that identifies recommended information including recommended products tailored to a user from multiple product information related to multiple products offered in a store, based on a user's behavior history regarding products offered in the store and predetermined priorities regarding the sales of the multiple products offered in the store.
(2) further comprising a ranking determination unit configured to determine the priority order by machine learning;
The ranking determination unit is
A fluctuation information acquisition unit that acquires fluctuation information that causes a fluctuation in the priority order;
The recommended information identifying device according to (1), further comprising: a rank learning unit that learns the priority order based on the variation information acquired by the variation information acquisition unit.
(3) The variation information is:
The recommended information identification device described in (2) includes at least one of the following: the price, type, lifespan, whether or not the product is discounted, the discount rate, profit margin, waste risk, inventory quantity, past sales performance, and past waste history of the product; if the product is a processed food, the processing date and time and processing state, the current time, the temperature, humidity, air pressure, and weather at the location of the store, and the number of customers at the store.
(4) The ranking determination unit:
The method further includes: acquiring a behavior result indicating whether the user has viewed or purchased the recommended product indicated by the recommendation information;
The recommended information identification device described in (2) or (3), wherein the ranking learning unit learns the priority order based on the variation information acquired by the variation information acquisition unit and the behavioral results acquired by the behavioral result acquisition unit.
(5) the ranking learning unit has a learning model for learning the priority order,
The ranking learning unit includes:
a reward output unit that outputs a positive reward when the behavioral result indicates that the recommended product has been viewed or purchased, and outputs a negative reward when the behavioral result indicates that the recommended product has not been viewed or purchased;
The recommended information identification device according to (4), further comprising: a learning model update unit that updates the learning model based on the reward output by the reward output unit.
(6) A recommended information identification device described in any one of (1) to (5), wherein the recommended information identification unit identifies, from a plurality of candidate information extracted from the plurality of product information based on the behavioral history, information that includes the recommended product with the highest priority as the recommended information.
(7) The recommended information identification unit identifies, from the plurality of candidate information, the recommended information that includes the recommended product with the highest evaluation value obtained by multiplying the priority of each of the plurality of candidate information by a noise coefficient randomly determined for each of the plurality of candidate information.
(8) The recommended information identifying device according to any one of (1) to (7), further comprising a recommended information output unit that outputs the recommended information identified by the recommended information identifying unit to a terminal device used by the user when purchasing a product at the store.
(9) A discount determination unit configured to determine by machine learning whether to provide a discount to a low-ranked product whose priority is lower than a predetermined reference rank,
The discount determination unit is
a deterioration information acquisition unit for acquiring deterioration information that changes the deterioration of the product;
The recommended information identification device described in any of (1) to (8), further comprising a judgment learning unit that learns to judge whether or not to discount the low-ranked products based on the deterioration information acquired by the deterioration information acquisition unit.
(10) The degradation information is
The recommended information identification device described in (9) includes at least one of the following: the price, type, lifespan, whether or not the product is discounted, the discount rate, profit margin, waste risk, inventory quantity, past sales performance, and past waste history of the product; if the product is a processed food, the processing date and time and processing state, the current time, the temperature, humidity, air pressure, and weather at the location of the store, and the number of customers at the store.
(11) a priority order acquisition unit that acquires priorities from a priority order storage unit that stores priorities regarding sales of a plurality of products provided in a store;
a behavior history acquisition unit that acquires a behavior history from a behavior history storage unit that stores a behavior history of a user with respect to a product provided in a store;
and a recommended information identification unit that identifies recommended information including recommended products corresponding to the user from the plurality of product information based on the behavioral history acquired by the behavioral history acquisition unit and the priority order acquired by the priority order acquisition unit.
(12) A ranking determination unit configured to determine the priority order by machine learning,
The ranking determination unit is
A fluctuation information acquisition unit that acquires fluctuation information that causes a fluctuation in the priority order;
The recommended information identification system according to (11), further comprising: a rank learning unit that learns the priority order based on the variation information acquired by the variation information acquisition unit.
(13) a behavior history acquisition step of acquiring a behavior history of a user with respect to products provided in a store;
and a recommendation information identification step of identifying recommended information including recommended products tailored to the user from multiple product information related to the multiple products based on the acquired behavioral history and predetermined priorities regarding the sales of multiple products offered in a store.
(14) A program for causing a computer to execute each step of the recommended information specifying method according to (13) above.

11: Server device 12: Gate device 13: Gate device 14: Cash register 15: User terminal 16: Employee terminal 21: Control unit 22: Operation display unit 23: Storage unit 24: Imaging unit 25: Communication I/F
31: Control unit 32: Storage unit 33: Ranking determination unit 34: Discount determination unit 35: Communication I/F
40: Rank learning unit 100: Information processing system 200: Store 211: Reading processing unit 212: Display processing unit 213: Notification processing unit 214: Behavior result transmission unit 311: Recommendation extraction processing unit 312: Recommendation selection processing unit 313: Recommendation information transmission unit 314: Information acquisition unit 321: Customer information storage unit 322: Product information storage unit 323: Purchase planned product storage unit 324: Priority order storage unit 331: Change information observation unit 332: Behavior result acquisition unit 333: Rank update processing unit 401: Reward calculation unit 402: Value function update unit TB1: Rank list

Claims (8)

An information processing system including a mobile terminal device used by an employee of a store where a predetermined product is provided, and an information processing device,
The information processing device includes:
a determination unit for determining a discount rate to be applied to the predetermined product displayed and on sale in the store ;
a determination result notifying unit that transmits the discount rate determined by the determining unit to the mobile terminal device,
the determination unit is configured to determine the discount rate based on variable information including at least the number of customers visiting the store and product information regarding the predetermined product acquired from the mobile terminal device including at least the number of the predetermined products displayed and on sale inputted into the mobile terminal device by an employee of the store;
An information processing system, wherein the mobile terminal device is configured to receive the discount rate transmitted by the determination result notifying unit. 2. The information processing system according to claim 1, wherein the mobile terminal device receives the discount rate transmitted from the determination result notification unit and displays it on a display unit of the mobile terminal device. A variation information acquisition unit that acquires the variation information;
A product information acquisition unit that acquires the product information,
The determination unit is
The information processing system according to claim 1 or 2 , wherein the discount rate is determined based on the fluctuation information and the product information using a learning model that has been machine-learned in advance to determine the discount rate. An information processing system according to any one of claims 1 to 3, wherein the fluctuation information further includes, in addition to the number of customers, one or more of past sales performance of the specified product, past disposal history of the specified product, or weather in the location of the store. The variable information further includes a current time,
The determination unit determines the discount rate and the timing of implementing the discount based on the fluctuation information and the product information,
the determination result notification unit transmits the discount rate and the implementation timing to the mobile terminal device;
5. The information processing system according to claim 1 , wherein the mobile terminal device receives the discount rate and the implementation timing transmitted from the determination result notification unit, and displays them on a display unit of the mobile terminal device. An information processing system as described in any one of claims 1 to 5, wherein the product information further includes, in addition to the number of the specified products on display and for sale , at least one of the following items related to the specified products: product price, product ID, product lifespan, production time, promotional campaign target information, profit margin, and degree of waste risk. a determination step of determining a discount rate to be applied to a predetermined product displayed and on sale in the store;
a transmission step of transmitting the discount rate determined by the determination step to a mobile terminal device used by an employee of the store;
a receiving step of receiving the discount rate transmitted by the transmitting step at the mobile terminal device,
The determining step includes:
acquiring variable information including at least the number of customers visiting the store and product information including at least the number of the predetermined products inputted into the mobile terminal device by an employee of the store , and determining the discount rate based on the variable information and the product information;
An information processing method in which each of the steps is executed by one or more processors. A program for causing one or more processors to execute each step of the information processing method according to claim 7 .

Images