JP7327542B2 - 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 setting method, and a program.

BACKGROUND ART In electronic commerce and advertisements on the Internet, there is known a method of extracting so-called recommended information that matches user's taste from a huge amount of information and displaying it on a terminal device used by the user. The recommendation information is also called recommendation information, and corresponds to, for example, product information including recommended products that are likely to match the user's tastes and should be recommended. The recommended information is, for example, specified based on the user's action history such as history information of purchasing products on the website (product purchase history) and history information of browsing product introduction pages (product browsing history). . A function that analyzes the action history and identifies and presents the recommended information that matches the user's preference based on a predetermined algorithm is called a recommendation function, and has been widely used in recent years.

Conventionally, a recommendation device having the recommendation function has been proposed (see Patent Literature 1). The recommendation device displays recommended information that matches the tastes of customers (visitors) on a tablet or the like used by an employee in a store such as a retail store.

Moreover, in recent years, a settlement system using information codes such as bar codes and QR codes (registered trademark) indicating the contents and prices of products has been proposed (see Patent Document 2). Mobile terminals (for example, smartphones) of customers who use the store are equipped with a code reading function that reads information codes attached to products. According to the payment system, the customer uses a mobile terminal to read the information code attached to the product package, etc., put the product in the shopping cart, and then pays the price of the product included in the read information code. make the payment. The payment for the product can be made, for example, at a register that can communicate with the mobile terminal, or can be settled using electronic money registered in the mobile terminal, or can be used as online credit settlement.

JP 2017-215667 A JP 2015-185089 A

However, in stores such as retail stores, the product status differs for each of the multiple types of products they handle. In some cases, the side (seller) cannot present the customer with the product it wants to sell. For example, in a store that sells a certain product A and a similar product B, if the inventory of product B is larger than that of product A, the store wants to sell product B preferentially. The function cannot preferentially present product B to the customer as recommended information. In particular, at supermarkets that handle products with a short product lifespan, such as fresh foods and ready-to-eat foods (processed foods), the product situation changes from moment to moment throughout the day. Suggestions are welcome.

It is an object of the present invention to specify recommended information according to product status, so that the user who receives the recommended information can quickly obtain product information that matches his or her tastes. To provide a recommended information specifying device, a recommended information specifying system, a recommended information specifying method, and a program capable of increasing the purchase rate of commodities to be sold preferentially.

A recommended information specifying device according to one aspect of the present invention is based on a user's action history with respect to products offered at a store and a predetermined priority order regarding sales of a plurality of products offered at the store. A recommended information identifying unit that identifies recommended information including a recommended product according to the user from a plurality of product information related to a plurality of products.

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

A recommended information identification system according to another aspect of the present invention includes a priority acquisition unit, an action history acquisition unit, and a recommended information identification unit. The priority acquisition unit acquires the priority from a priority storage unit that stores priority regarding sales of a plurality of products offered at the store. The action history acquisition unit acquires the action history from an action history storage unit that stores a user's action history with respect to products offered at the store. The recommended information specifying unit is configured to make a recommendation based on the action history acquired by the action history acquisition unit and the priority order acquired by the priority order acquisition unit based on the plurality of product information items. Identify recommendations that include products.

A recommended information identification method according to another aspect of the present invention includes an action history acquisition step of acquiring a user's action history with respect to products offered at a store; and a recommended information specifying step of specifying recommended information including a recommended product according to the user from a plurality of product information related to the plurality of products based on a predetermined priority regarding sales of the product.

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

According to the present invention, by specifying recommended information according to product status, the user presented with the recommended information can quickly obtain product information that matches his or her tastes, and the store side ( Sellers) can increase the purchase rate of products that they want to sell preferentially.

Further, according to the present invention, by providing the order determination unit including the change information acquisition unit and the order learning unit, the priority order regarding sales of a plurality of products is determined based on the change information that changes the priority order. Learned by the rank learning unit. As a result, even if the product situation at the store changes from moment to moment, the recommended information is identified using the priority determined according to the fluctuation, and as a result, the user who is presented with the recommended information can quickly obtain product information that matches one's tastes, and the store side (seller) can increase the purchase rate of products that the store side (seller) 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 that constitutes the information processing system. FIG. 3 is a block diagram showing the configuration of each device that constitutes the information processing system. FIG. 4 is a block diagram showing the configuration of the order determining section of the server device that constitutes the information processing system. FIG. 5 is a block diagram showing the configuration of the discount determination section of the server device that constitutes the information processing system. FIG. 6 is a diagram showing the ranking of each product handled by the store. FIG. 7 is a flow chart showing an example of a procedure of recommendation processing executed in the information processing system. FIG. 8 exemplifies coefficients and recommendation evaluation values when the five extracted candidate products are the products ranked first to fifth in priority order. FIG. 9 is a flow chart showing an example of the learning operation performed by the learning unit of the information processing system.

Hereinafter, embodiments of the present invention will be described with reference to the drawings as appropriate. It should be noted that the embodiment described below is merely an example that embodies 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 (recommended information specifying system of the present invention) according to the embodiment of the present invention includes a server device 11 (an example of a recommended information specifying device of the present invention), a gate device for entering a store. 12, a gate device 13 for exiting the store, a self-type cash register 14, a user terminal 15 (an example of the terminal device of the present invention), and an employee terminal 16 used by an employee of the store 200. Each of the gate device 12, the gate device 13, and the register 14 may be one or plural. In this embodiment, for example, an example in which the information processing system 100 is installed in a store 200 such as a supermarket that handles a wide variety of products will be described. Of course, the store 200 where the information processing system 100 is installed is not limited to a supermarket, and may be a real retail store that handles a plurality of products, or may be a virtual store on the Internet.

The gate device 12 is installed near the entrance of the store 200 , and the gate device 13 is installed near the exit of the store 200 . The user terminal 15 is a terminal device possessed and used by a user (user) who enters the store 200 when purchasing a product displayed in the store 200. For example, a smartphone or the like is used. Mobile terminals such as mobile phones and tablet terminals. The employee terminal 16 is a terminal device used by an employee of the store 200, and is, for example, a notebook computer, a mobile terminal such as a smart phone or a tablet terminal.

Hereinafter, with reference to FIGS. 1 and 2, a series of flows from a user's visit to the store 200 to purchase of merchandise and leaving the store will be illustrated.

First, a user who visits the store 200 approaches the gate device 12 for entering the store from the entrance while carrying the 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) and FeliCa (registered trademark). At this time, the user identification information (ID number, name, sex, age, etc.) stored in the user terminal 15 is transmitted to the gate device 12 . The user identification information is information for identifying users 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 . Thereby, the server device 11 recognizes that the user registered in advance has come to the store. When 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 history information of products purchased by the user in the store 200 in the past.

The user terminal 15 is preinstalled with a shopping application that is used in a system for shopping while scanning an information code (bar code, QR code, etc.) attached to a product. When the user terminal 15 communicates with the gate device 12, the user terminal 15 automatically activates the shopping application. At this time, for example, a screen shown in FIG. 2A is displayed on the operation display unit 22 (see FIG. 3) of the user terminal 15. FIG. As a result, the user can use the shopping application to purchase products in the store 200 .

While shopping around the store 200, the user picks up a product that the user wishes to purchase (a product to be purchased) from the display shelf, and reads the information code attached to the product through the imaging unit 24 of the user terminal 15. (See FIG. 3). The shopping application of the user terminal 15 reads the price information of the product (information including the product name, price, and product ID) included in the information code, and displays it on the operation display unit 22 of the user terminal 15 (see FIG. 3). ). 2 ( The screen shown in B) is displayed. The read price information is stored in the user terminal 15, transmitted from the user terminal 15 to the server device 11 through the communication network N1, and stored in the server device 11 as well.

The server device 11 reads the purchase history information (an example of the action history of the present invention) corresponding to the user identification information transferred from the gate device 12 from the storage unit 32 (customer information storage unit 321), and Execute the recommendation process (see FIG. 7). In this recommendation process, recommendation information (an example of the recommended information of the present invention) according to the user is generated from a plurality of product information related to all products (products that can be sold at the store 200) managed in the server device 11. This is the process of specifying. Specifically, product information that is likely to match the user's taste is extracted based on a predetermined algorithm, and the extracted product information is specified as the recommended information. Then, the specified recommendation information is transmitted to the user terminal 15 through the communication network N1. In the user terminal 15, the recommended 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 who is shopping around can easily notice it.

When paying (settlement) for the read product, the user goes to the register 14, holds the user terminal 15 over the communication unit provided in the register 14, and presses "payment" (Fig. 2 (B )). As a result, the user terminal 15 communicates with the register 14 through the near field radio, and the user identification information is transmitted from the user terminal 15 to the register 14 . Further, the user identification information transmitted to the register 14 is transferred from the register 14 to the server device 11 through the communication network N1. After that, in the server device 11 , the price information of the payment object corresponding to the user identification information is extracted and transferred to the register 14 . Then, at the register 14, settlement processing of the amount indicated by the price information is executed based on the settlement method specified by the user. When the payment processing is completed, a screen indicating that the payment is completed is displayed on the user terminal 15 .

Note that settlement processing may be performed in an online state between the user terminal 15 and the server device 11 without using the register 14 . The settlement process is performed in the following manner. For example, when the user selects "payment" on the screen of FIG. 2(B), settlement processing is started. For example, the user terminal 15 transmits the price information of the products a, b, and c to be settled to the server device 11 together with the user identification information. The server device 11 transmits a payment request to a payment server (not shown) connected via the Internet based on the price information. The payment server executes payment processing using a payment system corresponding to a payment business operator. After that, the fact that the settlement process has been executed is transmitted to the user terminal 15 via the server device 11, and a message to that effect is displayed on the operation display section 22 (see FIG. 3).

When the payment processing is completed, the register 14 transmits to the server device 11 the information on the purchased product for which the payment has been completed. Based on the information received from the register 14, the server device 11 updates the number of products in stock managed by the server device 11 and the like. 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 the user leaves the store 200, the user approaches the gate device 13 for leaving the store. 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, together with the user identification information of the user, flag information indicating that the shopping is completed is transmitted to the gate device 13, and the identification information and the flag information are sent from the gate device 13 to the server via the communication network N1. Transferred to device 11 . As a result, the server device 11 recognizes that the shopping has been properly completed and that the user has left the store.

As described above, the user can shop at the store 200 using the shopping application.

As described above, in the information processing system 100 , the recommendation processing for specifying the recommendation information according to the user is executed, and the specified recommendation information is displayed on the user terminal 15 .

A specific configuration of the information processing system 100 for implementing the recommendation process will be described below with reference to FIG.

[User terminal 15]
As shown in FIG. 3, the user terminal 15 includes a control section 21, an operation display section 22, a storage section 23, an imaging section 24, a communication I/F 25, and the like. The user terminal 15 is, for example, a mobile terminal owned by a user of the store 200, such as a smart phone, a mobile phone, or a tablet terminal. The user terminal 15 may be a mobile terminal lent 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 who is shopping in the store 200 captures the image of the information code attached to the product to be purchased by the imaging unit 24 . Price information (product name, price, product ID, etc.) of the product to be purchased is obtained by reading the captured image of the information code by the reading processing unit 211, which will be described later.

The communication I/F 25 connects the user terminal 15 to the communication network N1 by wire or wirelessly, and performs data communication according to a predetermined communication protocol with an external device such as the server device 11 via the communication network N1. It is a communication interface for Further, the communication I/F 25 enables data communication between the user terminal 15 and devices such as the gate devices 12 and 13 and the register 14 through radio (for example, short-range radio 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 types of information, and an operation unit such as a mouse, keyboard, or touch panel that receives user operations. The recommended information corresponding to the user is displayed on the display section of the operation display section 22 .

The storage unit 23 is a non-volatile storage medium such as a flash memory that stores various information. The storage unit 23 stores a control program for causing the control unit 21 to execute various processes.

The storage unit 23 also stores user identification information for identifying users. The user identification information is information for specifying the user, and includes an ID number, name, sex, age, and the like. The user identification information is registered in advance in the server device 11 provided in the store 200, and stored and managed in the storage section 32 (customer information storage section 321) in the server device 11 together with the purchase history information.

Further, the storage unit 23 stores the price information of the product to be purchased read by the reading processing unit 211, which will be described later.

The control unit 21 has control devices such as a CPU, ROM, and RAM. The CPU is a processor that executes various kinds of arithmetic processing. The ROM is a non-volatile storage unit in which control programs such as BIOS and 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 memory (work area) for various processes executed by the CPU. The control unit 21 controls the user terminal 15 by causing the CPU to execute various control programs pre-stored in the ROM or storage unit 23 .

Specifically, 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, as shown in FIG. The control unit 21 functions as the various processing units by executing various types of processing according to the control program by the CPU. Also, part or all of the processing units included in the control unit 21 may be configured by electronic circuits. The control program may be a program for causing a plurality of 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 included in the information code based on the image data of the information code captured by the imaging unit 24 . Hereinafter, for convenience of explanation, the price information read as a purchase target is referred to as purchase target information. The purchase target information that has been read is stored in the storage unit 23 until it is canceled or the payment for the product to be purchased is completed.

The display processing unit 212 causes the display unit of the operation display unit 22 to display the recommendation information transmitted from the server device 11 . For example, the display processing unit 212 displays the recommendation information in a pop-up format on the display unit of the operation display unit 22 . When the recommended information includes product images and product names, the display processing unit 212 displays these information in a pop-up format on the display unit of the operation display unit 22 . The product images and product names are, for example, linked to addresses storing more detailed information about recommended products (for example, product descriptions, product prices, and product display locations in the store 200). When the user touches the product image or product name on the display screen, the display processing unit 212 reads the detailed information stored in the address and displays it on the display unit. Further, when the recommendation information includes a URL of a website including a photograph image and 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 types of 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 and the like to the gate device 12 when the user enters the store 200, and transmits the user identification information to the gate device 13 when the user leaves the store 200. to transmit user identification information, etc. Further, the notification processing unit 213 transmits the purchase target information read during shopping to the server device 11 .

The action result transmission unit 214 transmits the action result indicating 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 to the ranking determination unit 33 of the server device 11 (see FIG. 3). ). As the action, for example, the action of browsing the recommended information displayed on the display unit of the operation display unit 22 (browsing action), the action of reading the product information of the information code of the product indicated by the recommended information by the reading processing unit 211. (reading behavior), or behavior of purchasing a product indicated by the recommendation information displayed on the display unit (purchase behavior). The action result includes a positive result indicating that each action occurred or a negative result indicating that each action did not occur.

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

For example, the control unit 21 determines whether the product image or the product name has been touch-operated during a certain period of time (time-up time) after the recommendation information is displayed on the display unit of the operation display unit 22. Alternatively, a process of determining whether or not the URL has been touch-operated is performed. In this case, when it is determined that the touch operation has been performed within the predetermined period of time, the action result transmission unit 214 sends information of a positive result (action result) indicating that the user has browsed the recommended information on the user terminal 15. is sent to the ranking determination unit 33 of the server device 11 .

In addition, the control unit 21 determines whether or not the product information of the information code of the product indicated by the recommendation information has been read by the reading processing unit 211 until the certain period of time has elapsed or until the settlement process is performed. or not. 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 action result transmission unit 214 determines that the user of the user terminal 15 is planning to purchase the product indicated by the recommendation information. Information of a positive result (behavior result) indicating that there is a result is transmitted to the ranking determination unit 33 of the server device 11 .

Further, the control unit 21 performs processing for 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 product for which the settlement process has been performed, the action result transmission unit 214 sends a positive result indicating that the product indicated by the recommendation information has been purchased. (Action result) information is transmitted to the ranking determination unit 33 of the server device 11 .

Note that the action result transmission unit 214 is configured to detect that the browsing action or the reading action is not performed within the certain period of time, and that the reading action is not performed until the user performs payment processing at the register 14. , when the purchase action is not performed in the settlement process, information of a negative result (behavior result) indicating that each action has not occurred is transmitted to the ranking determining section 33 of the server device 11 .

[Server device 11]
The server device 11 includes a control unit 31, a storage unit 32, a ranking determination unit 33 (an example of the ranking determination unit of the present invention), a discount determination unit 34 (an example of the discount determination unit of the present invention), a communication I/F 35, and the like. A server computer. Note that the server device 11 is not limited to one computer, and may be a computer system in which a plurality of computers operate in cooperation. Various types of processing executed by the server device 11 may be distributed and executed by one or more processors.

The communication I/F 35 connects the server device 11 to the communication network N1 by wire or wirelessly, and communicates with external devices such as the user terminal 15, the gate devices 12 and 13, and the register 14 via the communication network N1. A communication interface for executing data communication according to a communication protocol.

The storage unit 32 is a non-volatile storage unit including a semiconductor memory, HDD (Hard Disk Drive), SSD (Solid State Drive), or the like that stores various information. In this embodiment, a configuration in which the storage unit 32 is provided in the server device 11 is exemplified. It may be stored in an external device such as another server device or a storage device capable of data communication with. In this case, the server device 11 reads necessary information from the external device and stores information in the external device as necessary.

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

The storage unit 32 is divided into a plurality of storage areas. Specifically, the storage unit 32 includes a customer information storage unit 321 (an example of an action history storage unit of the present invention), a product information storage unit 322, a planned purchase product storage unit 323, and a priority storage unit 324 (a behavior history storage unit of the present invention). An example of a priority storage unit) is provided.

The customer information storage unit 321 stores pre-registered information for each user (hereinafter referred to as customer information). The customer information includes the user identification information including the user's ID number, name, gender, age, etc., as well as purchase history information regarding products purchased by the user at the store 200 in the past. In other words, the purchase history information is information indicating the user's action history with respect to the products sold at the store 200 .

The customer information is stored (registered) in the customer information storage section 321 of the storage section 32 as the user identification information, for example, when the user registers as a member of the store 200 . After that, when the user purchases a product in the store 200, the purchase history information including the product information and the date of purchase related to the purchased product is linked to the user identification information of the user who made the purchase. It is overwritten and stored in the information storage unit 321 . That is, the customer information is updated. The purchase history information is used for recommendation extraction processing 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, product price, and product ID (JAN code, etc.). In addition, the product information includes the product type, product lifespan (expiration date, use-by date, use-by date, etc.), discount availability and discount rate, product display location, product photos and descriptions, etc. It is information such as a URL, and includes information that makes shopping at the store 200 more convenient for the user. In addition, if the product is a cooked food (processed food) such as side dishes, the cooking time (including the cooking date and scheduled cooking time), which is the processing date and time of the cooked food, and the state after cooking (processing state) may be included in the product information. In addition, if the product is the target of a sales promotion campaign, for example, the product information may include information for determining whether the product is the target of the sales promotion campaign. Furthermore, the product information may include information that only the seller of the store 200 should be aware of, such as the profit margin of the product, the degree of risk of discarding the product, the number of items in stock, and the like.

In addition, the product information storage unit 322 stores past sales results and past disposal histories for each product. The past sales record includes date and time, price, number of sales, presence/absence of discount, discount rate, etc. when the product was sold in the past. The past disposal history includes whether or not the unsold product was disposed of, the date and time of disposal in the past, the amount of disposal, and the like.

The product information stored in the product information storage unit 322 is updated each time a change occurs. For example, the number of products in stock is immediately updated by the control unit 31 of the server device 11 when the user makes a payment (settlement) for the product. In addition, the price of the product, the number of items in stock, the presence or absence of discounts, the discount rate, the cooking time, the presence or absence of sales promotion campaigns, the risk of disposal, the increase in the number of items in stock due to arrival, etc. are based on the information input by the employee from the employee terminal 16. is updated by the control unit 31.

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

The priority storage unit 324 stores priority regarding sales of the plurality of products together with information on the plurality of products. FIG. 6 is an order list TB1 showing the priority order of each of a plurality of commodities. Among all the products stored in the product information storage unit 322, the order list TB1 contains a plurality of products (hereinafter referred to as priority products) that the store 200 side (seller) wishes to sell preferentially. Each priority product has its own sales priority. The order of priority indicates the order of products whose sales are desired to be promoted. The priority order may be defined for each of all products stored in the product information storage unit 322 .

The priority of each priority product in the priority list TB1 is determined based on the variation information regarding the variable factors that cause variation in the priority. The variable information includes, for example, the price, type, lifespan, presence/absence of discount, discount rate of the priority product, cooking time when the product is cooked food such as side dishes, whether the product is subject to the sales promotion campaign, The information includes the profit rate, the degree of disposal risk, the number of inventories, the past sales results, the past disposal history, and the like. Further, when the priority product is a cooked food such as "grilled chicken" typified by side dishes, the variable information includes the cooking date, cooking time, and state after cooking. The variation information also includes temperature, humidity, air pressure, and weather at the location of the store 200 . Further, the fluctuation information includes the current time, the number of visitors to the store 200 on the day, and visitor history. The visitor history is statistical information indicating the number of past visitor numbers for each month or day of the week. Also, the number of visitors includes the expected number of visitors predicted from the history of visitors. Such change information can be said to be information that causes a change in the circumstances surrounding the product (merchandise situation), including the condition of the product and the user's willingness to buy. Incidentally, in determining the priority order of the priority products, the variation information used for the determination process may include any one or more of the above-described specific variation factors.

The priority varies depending on the product status, various factors on the side of the store 200 (seller), and the like. For example, a product with a large number of items in stock is given a higher priority, and a product with a smaller number of items is given a lower priority. This is because unsold items are more likely to occur as the number of items in stock increases. A product with a short product life is assigned a higher rank, and a product with a longer product life is assigned a lower rank. Products with high profit margins are placed at the top, and products with low profit margins are placed at the bottom. Products with a high disposal risk are ranked high, and products with a low risk of disposal are ranked low. In addition, the order of priority also varies depending on the type of product. For example, foodstuffs, foodstuffs, foods (including cooked foods), and beverages that are ingredients for cooking are set at the top, while detergents and stationery are set at the bottom. Determined. In addition, even with cooked foods such as side dishes, the priority varies depending on what time it is cooked or the state after cooking. If the waiting time is long or the condition after cooking is poor, it is assigned a higher rank.

In addition, the price of the product, whether or not there is a discount, discount rate, past sales performance, past disposal history, temperature, humidity, air pressure and weather at the location of the store 200, the number of visitors, the history of visitors, etc. is a factor that changes Also, when a plurality of pieces of variation information exist, the plurality of pieces of variation information described above may affect each other and the priority may change. For example, many customers are expected to come to the store on a sunny day, so in this case, the priority items are ranked with more emphasis on the profit rate than on the number of items in stock. In addition, when the weather is fine but the temperature is high, the order is determined with emphasis on the type of product. Cooked food in a warm state is set at the top, and cold dishes that are estimated to sell well are set at the bottom.

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

The control unit 31 has control devices such as a CPU, ROM, and RAM. The CPU is a processor that executes various kinds of arithmetic processing. The ROM is a non-volatile storage unit in which control programs such as BIOS and 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 memory (work area) for various processes executed by the CPU. The control unit 31 controls the server device 11 by causing the CPU to execute various control programs pre-stored in the ROM or storage unit 32 .

Specifically, as shown in FIG. 3, the control unit 31 includes 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), an information acquisition unit 314 ( It includes various processing units such as an example of a priority acquisition unit and an action history acquisition unit of the present invention). Here, the recommendation extraction processing unit 311 and the recommendation selection processing unit 312 are examples of the recommended information specifying unit of the present invention. The control unit 31 functions as the various processing units by executing various types of processing according to the control program by the CPU. Also, part or all of the processing units included in the control unit 31 may be configured by electronic circuits. The control program may be a program for causing a plurality of processors to function as the various processing units.

The recommendation extraction processing unit 311 executes recommendation extraction processing for extracting recommendation information that matches the user's preference from a plurality of 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 storage unit 32 stores a recommendation engine (recommendation program) that extracts the recommendation information using a conventionally known predetermined recommendation algorithm that uses the purchase history information. ing. 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 of the present invention) from a plurality of product information. That is, the recommendation extraction processing unit 311 uses the recommendation engine stored in the storage unit 32 to execute the recommendation extraction processing based on a predetermined algorithm.

The recommendation selection processing unit 312 executes recommendation selection processing for selecting a specific recommended product candidate from the plurality of 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 the present embodiment, the recommendation selection processing unit 312 identifies from the plurality of recommended product candidates extracted by the recommendation extraction processing unit 311 based on the priority order of the priority products stored in the priority storage unit 324. selects the above-mentioned specific recommended product candidate.

In this embodiment, a configuration in which the recommendation extraction processing unit 311 and the recommendation selection processing unit 312 implement the recommended information specifying unit of the present invention is illustrated, but the present invention is not limited to such a configuration. For example, when the recommendation extraction process is executed, both the user's purchase history information stored in the customer information storage unit 321 and the priority order of the priority products stored in the priority order storage unit 324 It is applicable even if it is the structure which extracts the said specific recommendation information based on. In this case, a program that implements the recommendation selection process is incorporated into the recommendation engine that uses the conventionally known predetermined recommendation algorithm described above.

The recommendation information transmission unit 313 outputs recommendation information including product information of the specific recommended product candidate selected by the recommendation selection processing unit 312 to the user terminal 15 through the communication network N1. Upon receiving the transmitted recommendation information, the user terminal 15 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 above-mentioned priority regarding sales of a plurality of products provided at the store 200 from the priority storage unit 324 . Specifically, the information acquisition unit 314 acquires the priority of the necessary product from the priority storage unit 324 . More specifically, when the recommendation extraction processing by the recommendation extraction processing unit 311 selects product information of a plurality of recommended product candidates that are estimated to match the user's preference, the information acquisition unit 314 The priority order of a plurality of product information items is obtained from the priority order storage unit 324 . The obtained priority order is used in the recommendation selection processing unit 312 .

The information acquisition unit 314 also acquires the purchase history information of the user for the products provided at the store 200 from the customer information storage unit 321 . Specifically, when the user's user identification information is sent from the gate device 12 when the user comes to the store, the information acquisition unit 314 acquires the customer information corresponding to the user identification information. It extracts from the 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 unit 33]
The order determination unit 33 will be described below with reference to FIG. The order determination unit 33 is a learning device configured to determine the priority by machine learning, and has a learning model for determining the priority. Here, 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 variation information observation unit 331 (an example of the variation information acquisition unit of the present invention), an action result acquisition unit 332 (an example of the action result acquisition unit of the present invention), and a learning It includes a unit 40 (an example of the 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 arithmetic processing, for example, GPGPU (General-Purpose computing on Graphics Processing Units), a large-scale PC cluster, etc. are used. is desirable.

The change information observation unit 331 acquires the change information used for determining the priority of each priority product shown in the order list TB1 of FIG. The fluctuation information includes product price, product type, product lifespan (expiration date, expiry date, expiration date, etc.), discount availability and discount rate, cooking time or scheduled cooking time for cooked food, target of sales promotion campaign In the case of discriminative information as to whether or not, the profit margin of the product, the degree of disposal risk, the number of inventories, the past sales performance, and the past disposal history, the fluctuation information observation unit 331 stores the product information in the product information storage unit 322 are monitored, and real-time information is always acquired from the product information storage unit 322 .

In addition, when the fluctuation information is the temperature, humidity, and atmospheric pressure at the location of the store 200, the fluctuation information observation unit 331 is output from the temperature sensor, humidity sensor, and atmospheric pressure sensor provided indoors or outdoors of the store 200. It receives the sensor output value from the sensor, and obtains the real-time temperature, humidity, and atmospheric pressure indicated by the sensor output value. When the change information is the weather around the store 200, the control unit 31 downloads the weather information from the weather information database connected to the Internet with the server device 11, and the change information observation unit 331 receives the weather information from the control unit 31. receive the weather information;

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

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

The learning unit 40 learns the priority order based on the variation information acquired by the variation information observation unit 331 . In the present embodiment, the learning unit 40 uses not only the acquired variation information but also the action results acquired by the action result acquisition unit 332 to determine the priority order of products that the store 200 wants to sell preferentially. machine learning. Note that 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 storage unit 324, specifically, each of the priority products shown in the order list TB1 of FIG. , update the priority to the learned order.

The learning unit 40 analyzes and extracts useful rules, knowledge expressions, judgment criteria, etc. from the set of the input variation information, outputs the judgment results, and learns knowledge. It has functionality. Machine learning includes algorithms such as supervised learning, unsupervised learning, and reinforcement learning. A technique called “deep learning” is used.

Here, supervised learning is an algorithm that learns the "relationship between input and output" from pre-input data. Input values and correct answers to the data are added to the input data. By supplying a large amount of such data to the learning unit 40, the learning unit 40 learns the features of the data. , to estimate the output (result) from the input. If the relationship between given input and output data can be learned, the output can be predicted by applying that relationship to unknown data.

On the other hand, unsupervised learning is an algorithm that learns feature quantities such as data structure, characteristics, and new knowledge from a large amount of given input data without being given correct output data. It differs from supervised learning in that correct answers are not given to the data that is the source of learning.

In addition, reinforcement learning is not learning based on fixed and clear data like supervised learning or unsupervised learning. The program itself observes the given environment (current state) and It is an algorithm that evaluates the behavior of the robot and learns the appropriate behavior based on the interaction that the behavior gives to the environment, that is, the behavior that maximizes the reward that can be obtained in the future. TD learning and Q learning are known as representative methods.

In reinforcement learning, the consequences of 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. The order may not be the order desired by the store 200 side. Therefore, it is desirable to perform pre-learning by supervised learning, set the pre-learned priority as an 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) of the action value function Q(s, a) is represented by the following expression (1).

In equation (1), s _t represents the environment at time t, and a _t represents the action at time t. Action a _t changes the environment to s _t+1 . r _t+1 is the reward given by the change in the environment. The term with max is the Q value obtained by choosing the action a with the highest Q value known at that time under the environment s _t+1 multiplied by γ. γ is called a discount rate and is a parameter satisfying 0<γ≦1. α is a learning coefficient in the range of 0<α≦1. When Q-learning is applied, behaviors such as browsing the recommended information displayed on the user terminal 15, reading the information code of the product indicated by the recommended information, and purchasing the product (browsing behavior, reading behavior, purchasing behavior) becomes the behavior _at .

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

As described above, the change information causes the priority to change, and this change information itself also changes from moment to moment. Therefore, when the variation information changes, the priority is also updated. Whether or not the changed priority order is appropriate can be evaluated depending on whether the action result of each action of viewing action, reading action, and purchasing action is positive or negative. For this reason, for example, when Q-learning is applied, the action result (positive result, negative result) is substantially the _action at in formula (1), and the action value function Q(s , a) as the value function for changing the behavior _at .

In this embodiment, the reward calculation unit 401 outputs a positive reward when the action result is a positive result indicating that the recommended product indicated by the recommendation information has been browsed or purchased, and the action result indicates that the recommended product has been purchased. output a negative reward if the result is negative indicating that the user did not browse or purchase the The absolute value of the output reward may vary depending on the content of the action result. For example, the reward for the action result of browsing action may be set to be the lowest, and the reward corresponding to the action result of purchase action may be set to be the highest. Also, a weighting factor for the reward may be set for each user. For example, when emphasizing the reward based on the action result of the user who frequently uses the store 200, the weighting factor to be multiplied by the reward is the frequency of use. It may be set to a proportional value.

[Discount determination unit 34]
The discount determination unit 34 will be described below with reference to FIG. The discount judging unit 34 automatically determines whether or not to apply a discount to 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. A learning device configured to determine by learning. 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 the same as the configuration of the ranking determination unit 33 except for the determination result notification unit 341. Therefore, only different configurations will be described in detail below, and similar configurations will be explained in the ranking determination unit 33. , and the detailed description thereof will be omitted.

Note that, in the present embodiment, the discount includes not only a further reduction in the selling price of the target product, but also an act equivalent to a substantial discount on the target product. For example, providing a coupon (cash voucher, discount ticket) when a target product is purchased, or providing points redeemable for a product, etc., correspond to the discount.

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

For example, in the ranking list TB1 of FIG. 6, the same type of product "Yakitori" is ranked within the 10th place (reference ranking) two times, and it also exists in the 19th place, which is lower than the 10th place. do. For example, the product "yakitori" ranked within the top 10 ranks has a high priority, so it is highly likely that the user who is shopping around the store 20 will be notified as the recommendation information. The possibility that it will be purchased by a person is higher than the other ranked product “Yakitori”. On the other hand, the 19th-ranked product "Yakitori" (corresponding to a low-ranked product), which is lower than the reference ranking, has a low ranking, and is unlikely to be notified to the user as the recommended information. In this case, at this point, even if the product with a lower rank (lower rank product) is not a product that the store 200 side (seller) wants to sell preferentially, depending on the variation content of the variation information, the ranking determination unit 33 In some cases, it may be necessary to improve the sales situation without waiting for rank up or rank down (change in rank) after the ranking is determined.

Therefore, in the discount determination unit 34, the change information observation unit 331 acquires the change information as described above, and acquires the change information as the deterioration information that causes the deterioration of the product to fluctuate. In addition, based on the variation information acquired by the variation information observation unit 331, the learning unit 40, among the plurality of products of the same type, the priority order is lower than the reference order (low order product) Machine learning determines whether or not to implement a discount.

By the way, the products handled in the store 200 include short-lived products that cannot be stored for a long period of time, such as perishable foods and side dishes. In addition, daily necessities such as detergents and stationery, processed foods such as canned foods and instant noodles that can be stored for a long time, sweets and luxury goods, seasonings, products with long expiration dates such as alcoholic beverages such as liqueurs, etc. is also included. Therefore, even if the low-ranked products are low-ranked products, there is little need to reduce the price of long-life products such as daily necessities, even if the stock quantity is large. On the other hand, even if the number of items in stock is small, short-lived products such as perishables and side dishes should be sold out at a discounted price to avoid the risk of unsold disposal. Therefore, the discount determination unit 34 extracts short-lived products whose lifespan is shorter than a predetermined threshold among the low-ranked products, and machine-learns whether or not to implement a discount for the short-lived products. is preferred. In other words, in the present embodiment, the discount determination unit 34 performs machine learning on whether or not to implement a discount by excluding products with a lifespan longer than the predetermined threshold from the low-rank products. The predetermined threshold is arbitrarily set according to the target product. For example, it is set to about one day for prepared foods, and several days for fresh foods depending on the type of food. be.

The determination result notification unit 341 refers to the determination result learned by the learning unit 40 and transmits the determination result to the employee terminal 16 . At the employee terminal 16 that receives this determination result, the determination result is displayed on the display section of the terminal 16 . Therefore, the employee of the store 200 recognizes which product should be discounted by confirming the displayed judgment result, and quickly discounts the product, that is, updates (replaces) the information code. It can be carried out.

In the above example, the reference ranking is set to the 10th place. It may be learned whether or not. In this case, the rank of the product with the highest rank among the plurality of products of the same type corresponds to the reference rank of the present invention.

Further, in the above example, the discount is applied to a product (product) whose priority is lower than the reference order among a plurality of products of the same type. It may also be possible to learn whether or not to apply a discount to each product (low-rank product) ranked lower than the reference rank.

Further, the learning unit 40 of the discount determination unit 34 is made to learn not only whether or not to implement a discount on the product, but also, for example, when it is determined to implement a discount, the discount rate and the timing of implementing the discount. can be

Hereinafter, the recommended information identifying method of the present invention will be described together with the recommendation processing executed by the control unit 31 of the server device 11, using the flowchart of FIG. In the following description, it is assumed that the recommendation process is performed from the time the user enters the store 200 through the gate device 12 to the time the user leaves the store 200 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 processing procedure numbers (step numbers).

First, in step S<b>11 , the control unit 31 acquires the purchase history information of the user who entered the store from the customer information storage unit 321 . Such processing is performed by the information acquisition unit 314 of the control unit 31 . Step S11 corresponds to the action history acquisition step of the present invention. Also, in the next step S12, the control unit 31 extracts the action content of the user from the acquired purchase history information.

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

In the next step S14, the control unit 31 determines whether or not a high-priority product belonging to the order list TB1 of FIG. 6 is included in the plurality of candidates extracted in step S13. For example, the high-ranking product can be a product whose priority is within a predetermined reference order (10th in this embodiment).

In step S14, if it is determined that the high-ranking product is included in the plurality of candidates (Yes side of S14), in the next step S15, the highest-ranking product is selected from the plurality of candidates. Then, the selected product is specified as a recommended product. Such processing is performed by the recommendation selection processing unit 312 of the control unit 31 . Step S15 corresponds to the recommended information specifying 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 section of the operation display section 22 . The recommendation information is also transmitted to the employee terminal 16 .

On the other hand, in step S14, when it is determined that the high-ranking product is not included in the plurality of candidates (No in S14), the control unit 31 again uses the recommendation engine to Based on this, a process of extracting a plurality of candidates different from the candidates extracted in step S13 is performed (step S17). After that, the process of step S14 is executed, and the processes of steps S14 and S17 are repeatedly performed until it is determined that the high-ranking product is included in the plurality of candidates.

As described above, in this embodiment, the user terminal 15 displays the recommended information that matches the user's preference. Information (information on recommended products) can be obtained quickly. In addition, the recommendation information output to the user terminal 15 includes recommended product information in consideration of the order of priority of products that the store 200 side (seller) wants to sell preferentially. Therefore, the possibility that the recommended product will be viewed or purchased increases, and as a result, the store side (seller) can improve the purchase rate of the product that it prefers to sell.

In addition, since the recommendation information is also transmitted to the employee terminal 16, the employee can confirm the received recommendation information of each user.

In addition, although the above-mentioned embodiment demonstrated the example which specifies a recommended product by performing the process of step S14 and S15, this invention is not limited to such an example. For example, a product with the highest recommendation evaluation value, which will be described later, may be specified as the recommended product from among the plurality of candidates extracted in step S13.

As the recommendation evaluation value, for example, as shown in FIG. 8, numerical values Z1 and Z2 obtained by multiplying predetermined compatibility coefficient Z1 and ranking coefficient Z2 for each of the five extracted candidate products A1 to A5 are used. can be done. Here, the compatibility coefficient Z1 is a numerical representation of compatibility (preference) with the user for each of the candidate products A1 to A5. is calculated by Probabilistic Latent Semantic Analysis (PLSA).

Also, the ranking coefficient Z2 is an arbitrary coefficient determined based on a predetermined rule. In this embodiment, the first-ranked product in the ranking list TB1 (see FIG. 6) is given an arbitrary numerical value “0.50”, and the second-ranked product is given a numerical value “0.50” which is half the coefficient of the first-ranked product. 0.25", 1/3 of the coefficient of the 1st place for the 3rd place product "0.16", 1/4 of the coefficient of the 1st place for the 4th place product "0.13", The fifth-ranked product has a numerical value of 0.10, which is one-fifth of the coefficient of the first-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 numerical value as the ranking decreases from higher to lower. In addition, each numerical value mentioned above is displayed by omitting the third decimal place. Here, FIG. 8 exemplifies each coefficient and the recommendation evaluation value when the five candidate products extracted in step S13 are the products ranked first to fifth in priority order.

In the recommendation process, the control unit 31 specifies, as the recommended product, the product with the largest recommendation evaluation value among the recommendation evaluation values determined as described above, instead of the processing of steps S14 and S15. By specifying the recommended product in this way, it is possible to prevent the same product from being presented to the same user as the recommended product when there is no change in the order of the higher products.

In addition, instead of specifying 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 numerical value of the recommendation evaluation value, the higher. A selection rate may be set and the recommended product may be identified probabilistically based on each selection rate.

The recommendation evaluation value is obtained by multiplying numerical values Z1 and Z2 obtained by multiplying the compatibility coefficient Z1 and the rank coefficient Z2 by a noise coefficient Z3 randomly determined for each of the five candidate products A1 to A5. may be used. This noise coefficient Z3 is a coefficient for varying the recommendation evaluation value. Multiplying the numerical values Z1 and Z2 by this noise coefficient Z3 also affects the specified recommended product. In other words, the specified recommended product may change before and after the noise coefficient Z3 is multiplied. The timing for multiplying the noise coefficient Z3 is, for example, when the number of times the user has entered the store has reached a predetermined number of times, or when the number of times the user has entered the store has reached a predetermined number of times during the most recent set period. , that the user has returned to the store after a predetermined period of time has elapsed, and that the recommended product indicated by the presented recommendation information has been positively performed a predetermined number of times in a row (browsing behavior, reading behavior, (Purchase behavior) can be set as the timing at which any one or more of the following conditions are satisfied. By adding such a noise coefficient Z3 to the recommendation evaluation value, it is possible to provide diversity to the recommendation information including the specified recommended product.

Such diverse recommendation information does not necessarily include the recommended products reflected by the user's purchase history. However, when a user browses recommendation information that does not include a recommended product and purchases a product indicated by the recommendation information with the recommendation information as a trigger, the purchase history of such products is overwritten on the user's purchase history. be. As a result, at the next timing, the recommendation information including the recommended product is specified using the overwritten purchase history information, and the match between the recommended product indicated by the recommendation information and the product preferred by the user is improved.

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

An example of the learning operation flow of the learning unit 40 of the ranking determining unit 33 using reinforcement learning will be described below with reference to the flowchart of FIG. In FIG. 9, S21, S22, . . . indicate processing procedure numbers (step numbers). Since 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, the description thereof will be omitted.

First, in step S<b>21 , the action result acquisition unit 332 of the ranking determination unit 33 acquires the action results (browsing action, reading action, and purchase action results) transmitted from the user terminal 15 . In this step S21, the action results of all the users who use the store 200 are transmitted from each of the user terminals 15. FIG.

In step S22, the reward calculation unit 401 of the learning unit 40 calculates a judgment evaluation value for judging whether the current priority is appropriate based on the action result acquired in step S21. The determination evaluation value is calculated by adding evaluation points determined for each content of the action result. For example, positive evaluation points are assigned to positive results of browsing behavior, reading behavior, and purchasing behavior, and negative evaluation points are assigned to negative results. The determination evaluation value is calculated by adding evaluation points for all action results. When a weighting factor is set for the evaluation points for each user, for example, a numerical weighting factor proportional to the frequency of use is set to the evaluation points for the action result of the user who frequently uses the store 200. If so, the determination evaluation value may be calculated with the weighting factor taken into consideration.

In the next step S23, it is determined whether or not the evaluation value is within a predetermined allowable range. If it is determined that the determination evaluation value is within the allowable range, the current priority is determined to be appropriate, and the process proceeds to step S24 to set a plus reward. On the other hand, if it is determined that the determination evaluation value exceeds the allowable range, the current priority order is determined to be inappropriate, and the process proceeds to step S25 to set a negative reward.

Thereafter, in step S26, the value function related to priority, that is, the action value function Q(s, a) is updated by the value function updating unit 402 of the learning unit 40, and the process returns to step S21 to repeat the same processing. executed.

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

The embodiments described above include features to be added below.
(1) Based on the user's action history with respect to the products offered at the store and a predetermined priority order regarding sales of the products offered at the store, the user selects from the product information of the plurality of products. A recommended information identification device, comprising: a recommended information identification unit that identifies recommended information including recommended products according to the conditions.
(2) further comprising an order determining unit configured to determine the order of priority by machine learning;
The ranking determination unit
a change information acquisition unit that acquires change information that causes a change in the priority;
The recommended information identifying device according to (1) above, further comprising: a ranking learning section that learns the priority ranking based on the variation information acquired by the variation information acquiring section.
(3) The fluctuation information is
Price, type, lifespan, presence/absence of discounts, discount rate, profit margin, disposal risk, inventory quantity, past sales performance, and past disposal history of said product, processing date and processing status if said product is processed food , current time, temperature, humidity, air pressure, and weather at the location of the store, and the number of visitors to the store.
(4) The ranking determination unit
further comprising an action result acquisition unit that acquires an action result indicating whether the user viewed or purchased the recommended product indicated by the recommended information;
wherein the order learning unit learns the priority based on the change information acquired by the change information acquisition unit and the action result acquired by the action result acquisition unit; ) recommended information identification device described in.
(5) the ranking learning unit has a learning model for learning the priorities;
The ranking learning unit
If the action result indicates that the recommended product has been viewed or purchased, a positive reward is output, and if the action result indicates that the recommended product has not been viewed or purchased, a negative reward is output. a reward output unit that outputs a reward for
The recommended information identifying device according to (4) above, further comprising: a learning model updating unit that updates the learning model based on the reward output by the reward output unit.
(6) The recommended information specifying unit specifies, as the recommended information, information including the recommended product with the highest priority among a plurality of candidate information extracted from the plurality of product information based on the action history. , the recommended information identification device according to any one of (1) to (5).
(7) The recommended information identifying unit multiplies the priority of each of the plurality of candidate information by a noise coefficient randomly determined for each of the plurality of candidate information, and the evaluation value obtained by multiplying the priority of each of the plurality of candidate information is: The recommended information identifying device according to (6) above, which identifies information including the highest recommended product as the recommended information.
(8) from (1) to ( 7) The recommended information identification device according to any one of the above.
(9) further comprising a discount determination unit configured to determine by machine learning whether or not to implement a discount for low-ranked products whose priority is lower than a predetermined reference rank;
The discount determination unit
a deterioration information acquisition unit that acquires deterioration information that causes fluctuations in the deterioration of the product;
(1) to (8) above, further comprising a determination learning unit that learns, based on the deterioration information acquired by the deterioration information acquisition unit, whether or not to implement a discount for the low-ranking product. The recommended information identification device according to any one of the above.
(10) The deterioration information is
Price, type, lifespan, presence/absence of discounts, discount rate, profit margin, disposal risk, inventory quantity, past sales performance, and past disposal history of said product, processing date and processing status if said product is processed food , current time, temperature, humidity, air pressure, and weather at the location of the store, and the number of visitors to the store.
(11) a priority acquisition unit that acquires the priority from a priority storage unit that stores the priority regarding sales of a plurality of products offered at the store;
an action history acquisition unit that acquires the action history from an action history storage unit that stores the user's action history with respect to products provided at the store;
Based on the action history acquired by the action history acquisition unit and the priority acquired by the priority acquisition unit, recommended information including a recommended product for the user is specified from the plurality of product information. and a recommended information identification unit.
(12) further comprising a priority determination unit configured to determine the priority by machine learning;
The ranking determination unit
a change information acquisition unit that acquires change information that causes a change in the priority;
The recommended information specifying system according to (11) above, further comprising a ranking learning unit that learns the priority based on the variation information acquired by the variation information acquisition unit.
(13) an action history acquisition step of acquiring a user's action history with respect to products offered at the store;
Recommendation including a recommended product according to the user from a plurality of product information related to the plurality of products based on the acquired action history and a predetermined priority order regarding sales of the plurality of products provided at the store. and a recommended information identifying step of identifying information.
(14) A program for causing a computer to execute each step of the recommended information identifying method described in (13) above.

11: Server device 12: Gate device 13: Gate device 14: 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 : Action 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 storage unit 331 : Fluctuation information observation unit 332 : Action result acquisition unit 333 : Rank update processing unit 401: reward calculator 402: value function updater TB1: ranking list

Claims (10)

A discount determination unit configured to determine by machine learning whether or not to implement a discount for a predetermined product among a plurality of products offered at the store,
The discount determination unit
a deterioration information acquisition unit that acquires deterioration information that causes fluctuations in deterioration of the product and life information that indicates the life of the product;
A short-lived product having a life shorter than a predetermined threshold among the plurality of products from a product information storage unit that stores a plurality of product information related to the plurality of products based on the lifespan information acquired by the deterioration information acquisition unit. as the predetermined product , and learns whether or not to discount the extracted short-life product based on a predetermined machine learning algorithm using the deterioration information; , an information processing device. The deterioration information is
Price, type, profit margin, disposal risk, number of items in stock, past sales performance, and past disposal history of the product, processing date and processing status if the product is processed food, current time, location of the store The information processing apparatus according to claim 1, comprising at least one or more of temperature, humidity, air pressure, weather, and the number of visitors to said store. 3. The information processing apparatus according to claim 1, wherein said discount determination unit further determines a discount rate or a discount implementation timing when determining to implement a discount for said short-life product . 4. The method according to any one of claims 1 to 3, further comprising a determination result notification unit that transmits a determination result by the discount determination unit to a terminal device used in the store and displays the determination result on a display unit of the terminal device. The information processing device according to . a product information storage unit that stores a plurality of product information related to a plurality of products offered at the store;
a discount determination unit configured to determine, by machine learning, whether or not a predetermined product among the plurality of products related to the plurality of product information stored in the product information storage unit is to be discounted. ,
The discount determination unit
a deterioration information acquisition unit that acquires deterioration information that causes fluctuations in deterioration of the product and life information that indicates the life of the product;
Based on the lifespan information acquired by the deterioration information acquisition unit, a short-life product having a life shorter than a predetermined threshold value is extracted from the product information storage unit as the predetermined product , and extracted. and a determination learning unit that learns whether or not to discount the short-life product based on a predetermined machine learning algorithm using the deterioration information. A discount determination step for determining, by machine learning, whether or not to implement a discount for a predetermined product among a plurality of products offered at the store,
The discount determination step includes:
a deterioration information acquisition step of acquiring deterioration information that causes fluctuations in the deterioration of the product and life information that indicates the life of the product;
A short-lived product having a life shorter than a predetermined threshold among the plurality of products from a product information storage unit that stores a plurality of product information related to the plurality of products, based on the lifespan information acquired in the deterioration information acquiring step. is extracted as the predetermined commodity , and for the extracted short-life commodity, a decision learning step of learning whether or not to implement a discount based on a predetermined machine learning algorithm using the deterioration information; , and
An information processing method in which one or more processors execute each of the steps . The deterioration information is
Price, type, profit margin, disposal risk, number of items in stock, past sales performance, and past disposal history of the product, processing date and processing status if the product is processed food, current time, location of the store 7. The information processing method according to claim 6, including at least one or more of temperature, humidity, atmospheric pressure, weather, and the number of visitors to said store. 8. The information processing method according to claim 6, wherein said determining step further determines a discount rate or a discount execution timing when it is determined to implement a discount for said short-life product . 9. The method according to any one of claims 6 to 8, further comprising a determination result notification step of transmitting a determination result by said discount determination unit to a terminal device used in said store and displaying said determination result on a display unit of said terminal device. The information processing method described in . A program for causing one or a plurality of processors to execute each step of the information processing method according to any one of claims 6 to 9.