Product recommendation device, product recommendation method, and program

Abstract

To provide a technique for recommending a product while reducing customer's time for choosing a product in a case where products of the same food on which a plurality of prices are set are sold.SOLUTION: A product recommendation device includes food identification means that identifies kinds of foods and the amount of each food that a customer desires to purchase on the basis of input information input by the customer, condition identification means that identifies a condition for customer's purchase of a product on the basis of the input information, recommended product identification means that when a food that the customer desires to purchase includes products on which a plurality of prices are set, identifies combination of products among the products on which a plurality of prices are set, so as to satisfy a customer's condition for product purchase, and recommendation information output means that outputs recommendation information which is about the combination of products.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to a product recommendation device, a product recommendation method, and a program. [Background technology]

[0002] There is technology that can present recommended products to customers.

[0003] Patent document 1 describes an ingredient purchasing support device that classifies customers into multiple types based on the customer's past ingredient purchasing history, provides recommended recipes that suit the customer's type, and provides information on ingredient products necessary to make the recipe selected by the customer. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2014-093064 A Summary of the Invention [Problem to be solved by the invention]

[0005] Meanwhile, dynamic pricing may be used for food ingredients sold at retail stores. In dynamic pricing, for example, different prices are set for the same product depending on the expiration date or arrival date. Also, food ingredients sold at retail stores may have different prices depending on the manufacturer or brand, even if the food ingredients are the same. In this way, when multiple prices are set for a product, a customer may intentionally select a product to purchase from multiple products that are the same food ingredient but have different prices. For this reason, when multiple products that are the same food ingredient but have different prices are sold, it may be troublesome for the customer to select a product to purchase from many options.

[0006] One example of the objective of the present disclosure is to provide a technology that makes product suggestions that reduce the effort required for customers to select products when selling products that are the same ingredient and have multiple price settings. [Means for solving the problem]

[0007] A product recommendation device in one aspect of the present disclosure includes an ingredient identification means for identifying the type of ingredients and the quantity of each ingredient that a customer desires to purchase based on input information entered by the customer, a condition identification means for identifying the conditions for the customer's product purchase based on the input information, a recommended product identification means for identifying, when the ingredients that the customer desires to purchase include products with multiple price settings, a combination of products that meet the customer's product purchase conditions from among the products with multiple price settings, and a recommended information output means for outputting recommended information that is information regarding the product combination.

[0008] In one aspect of the present disclosure, a product recommendation method includes a computer that identifies the types of ingredients and the quantities of each ingredient that a customer desires to purchase based on input information entered by the customer, identifies the customer's product purchasing conditions based on the input information, and, if the ingredients that the customer desires to purchase include items with multiple price settings, identifies a combination of items that meet the customer's product purchasing conditions from among the items with multiple price settings, and outputs recommendation information that is information regarding the product combination.

[0009] A product recommendation program in one aspect of the present disclosure causes a computer to execute a process of identifying the types of ingredients and the quantities of each ingredient that a customer desires to purchase based on input information entered by the customer, identifying the customer's conditions for purchasing the products based on the input information, and, if the ingredients that the customer desires to purchase include products with multiple price settings, identifying a combination of products that meets the customer's conditions for purchasing the products from among the products with multiple price settings, and outputting recommendation information that is information regarding the product combination.

[0010] Each program may be stored in a non-transitory computer-readable recording medium. Effect of the Invention

[0011] One example of an effect of the present disclosure is that when selling products that are the same ingredients but have multiple price settings, it is possible to make product suggestions that reduce the effort required for customers to select products. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 is a diagram illustrating a configuration of a system including a product recommendation device. [Diagram 2] 2 is a block diagram showing an example of a functional configuration of the product recommendation device. FIG. [Diagram 3] FIG. 13 is a diagram showing an example of an output screen. [Figure 4] 10 is a flowchart showing an example of an operation of the product recommendation device. [Diagram 5] 13 is a block diagram showing another example of a functional configuration of the product recommendation device. [Figure 6] 13 is a block diagram showing another example of a functional configuration of the product recommendation device. [Figure 7] 13 is a flowchart showing another example of the operation of the product recommendation device. [Figure 8] FIG. 2 is a diagram showing a hardware configuration for realizing the product recommendation device according to the present disclosure using a computer and its peripheral devices. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present disclosure will be described in detail with reference to the drawings.

[0014] [First embodiment] 1 is a diagram illustrating an example of a product recommendation system including the product recommendation device 100 of the present disclosure. The product recommendation system includes, for example, the product recommendation device 100, a terminal device, a customer information database, and a product information database.

[0015] In FIG. 1, a product recommendation device 100 recommends to a customer a combination of products that meets the customer's requirements for purchasing products.

[0016] In the present disclosure, it is assumed that there are products of the same foodstuffs in a retail store such as a supermarket or a convenience store, for which multiple prices are set. For example, the price of a product may vary depending on the freshness, such as lowering the price of a product with a close expiration date or expiry date. In this way, changing the price of a product depending on the freshness of the product is an example of dynamic pricing, which changes the price depending on the demand and supply of the product. Also, for example, the price of soy sauce manufactured by manufacturer A, which manufactures foodstuffs, may differ from the price of soy sauce manufactured by manufacturer B, which manufactures foodstuffs. Also, for example, the price of brand C may differ from the price of strawberries, which are brand D, which are different from the price of foodstuffs. In other words, in the present disclosure, a product with multiple prices set may include one or both of the same product with different prices set by dynamic pricing and the same type but different products.

[0017] In the above-mentioned case, when there are products with multiple price levels set for the food ingredient that the customer wishes to purchase, the product recommendation device 100 recommends to the customer a combination of products that meets the customer's conditions for purchasing the products.

[0018] 1, the product recommendation device 100 is realized using a server. The product recommendation device 100 outputs information about product combinations to a terminal device used by a customer.

[0019] The terminal device is a terminal device used by a customer. The terminal device displays information about recommended product combinations. The terminal device is, for example, a smartphone or a tablet. The terminal device may be a terminal device owned by the customer, or may be a terminal device used for scan shopping in which the customer moves around the store while scanning products on the sales floor.

[0020] The customer information database is a database in which customer information, which is information about a customer, is stored in association with customer identification information that identifies the customer. The customer information includes, for example, the customer's product purchase history. The purchase history includes information for identifying the products purchased by the customer, the quantity of products purchased by the customer, the price of the products purchased by the customer, and the date and time when the customer purchased the products. The customer information may also include information indicating the attributes of the customer, the customer's website browsing history, etc.

[0021] The product information database is a database in which product information, which is information about a product, is stored in association with product identification information that identifies the product. Product information includes, for example, the price of the product, the number of products in stock, and information indicating which food ingredient the product is classified into. Product information may also include other information such as the name of the product and an image of the product.

[0022] Here, ingredients are materials used in cooking, such as "carrots," "cabbage," "milk," "pork belly," and "salt." In retail stores, products that are the same ingredients but have different prices set may be sold through dynamic pricing. Products that are the same ingredients but have different prices set may have different freshness, for example. That is, even if the product is the same, the price may differ depending on the freshness of the product. The freshness of the product is an example of a condition for changing the price of the product in dynamic pricing. The condition for changing the price of the product in dynamic pricing may be the freshness of the product, the purchase amount of the product, the demand for the product, etc., and is not limited to these examples. In addition, for example, products that are the same ingredients but have different prices may be products that have different origins, brands, manufacturers, etc. In the following explanation, ingredients are used as a general term for products that are the same ingredients but have multiple prices set.

[0023] Each database may be provided inside the product recommendation device 100. Also, each database may be made up of multiple databases.

[0024] Next, the configuration of the product recommendation device 100 in the embodiment will be described.

[0025] 2 is a block diagram showing the configuration of the product recommendation device 100. Referring to FIG. 2, the product recommendation device 100 includes an ingredient specification unit 101, a condition specification unit 102, a recommended product specification unit 103, and a recommended information output unit 104.

[0026] Next, the configuration of the product recommendation device 100 in the first embodiment will be described in detail.

[0027] The ingredient identification unit 101 is an example of an ingredient identification means that identifies the type of ingredients and the quantity of each ingredient that the customer wishes to purchase based on the input information entered by the customer. The ingredient identification unit 101 identifies multiple ingredients that the customer wishes to purchase. In the following description, "ingredients that the customer wishes to purchase" refers to the amount of ingredients and the quantity of each ingredient that the customer wishes to purchase.

[0028] The input information is information that a customer inputs to an application or a website on a terminal device. The input information is, for example, customer identification information. The customer identification information is, for example, a membership number of a customer who is registered as a member of a store, a service, or the like. For example, a customer inputs customer identification information when logging in to an application or a website. Also, for example, the input information may be information that a customer inputs on an input screen such as a questionnaire response screen of an application or a website. For example, a customer inputs information on an input screen of an application or a website by inputting characters, selecting an option, inputting a photographed image, or the like, in accordance with questions on the input screen. The application or website is, for example, an application or website provided by a store to a customer. The application or website provided by a store may be, for example, an application or website that provides information to a customer by a store or that allows a customer to order products from the store, but is not limited to these examples. Note that the application or website may be an application or website that provides information on multiple stores or that allows a customer to order products from multiple stores, and the source of the application or website is not limited to the store.

[0029] An example of a method in which the ingredient specification unit 101 specifies an ingredient that a customer wishes to purchase will be described.

[0030] For example, the ingredient specification unit 101 may specify the input information input by the customer as the type of ingredients and the quantity of each ingredient that the customer wishes to purchase. In this case, for example, the customer inputs information indicating the type of ingredients and the quantity of each ingredient that the customer wishes to purchase into an application or a website.

[0031] Also, for example, the ingredient identification unit 101 may identify information associated with the customer identification information input by the customer as the ingredient the customer wishes to purchase. In this case, for example, the customer registers the ingredient they wish to purchase in advance in an application or a website. Also, for example, in this case, the information associated with the customer identification information may be information indicating the ingredient the customer is presumed to wish to purchase. At this time, the ingredient the customer is presumed to wish to purchase may be presumed based on the customer's purchase history associated with the customer identification information by another device, for example. For example, the ingredient identification unit 101 may presume, based on the purchase history, that the customer frequently purchases as the ingredient the customer wishes to purchase.

[0032] Also, for example, the ingredient identification unit 101 may identify ingredients that the customer wishes to purchase from a recipe that the customer intends to cook. The processing of the ingredient identification unit 101 in this case will be described in the first modification.

[0033] Next, the condition specifying unit 102 is an example of a condition specifying means for specifying the conditions for a customer's product purchase based on input information. The conditions for a customer's product purchase may be a plurality of conditions. Furthermore, when there are a plurality of conditions, a priority may be set for each of the conditions. The conditions for a customer's product purchase are constraint conditions used by the recommended product specifying unit 103, which will be described later, to specify the optimal product combination of ingredients that the customer wishes to purchase.

[0034] The conditions for a customer to purchase a product can also be said to be the criteria by which the customer selects the product to purchase. The conditions for a customer to purchase a product can be, for example, that the product is fresh, that the product is cheap, that the product has a specific attribute, etc. A product being fresh can mean that the product is not spoiled, that the product has a long expiration date or best before date, etc. Furthermore, a product having a specific attribute can mean that the product attributes, such as place of origin, brand, producer, manufacturer, distributor, production method, manufacturing method, etc., are specific attributes desired by the customer.

[0035] Furthermore, the conditions for a customer's product purchase may be set for each type of food ingredient or for each category of food ingredient. The category of food ingredients is a category for classifying food ingredients. The category of food ingredients may be, for example, a classification indicating whether or not a food ingredient is fresh food. This is because fresh food ingredients and food ingredients other than fresh food ingredients generally differ in how easily the ingredients spoil, and the conditions for a customer's product purchase may differ between fresh food ingredients that spoil easily and other food ingredients that do not spoil easily.

[0036] The condition specifying unit 102 may specify, for example, the customer's input information as the condition for the customer's product purchase. In this case, for example, the customer inputs the condition for the product purchase in an application or website of the terminal device.

[0037] Also, for example, the condition specification unit 102 may specify information associated with customer identification information, which is input information of the customer, as a condition for the customer's product purchase. In this case, for example, the customer registers the conditions for the product purchase in advance in an application of the terminal device or on a website.

[0038] Also, for example, the condition specifying unit 102 may specify the conditions for the customer's product purchase based on the customer's purchase history associated with the customer's identification information, which is input information of the customer. It is considered that the purchase history includes products that satisfy the conditions for the customer's product purchase. In other words, it is considered that the customer consciously or unconsciously selects and purchases products that satisfy the conditions for the customer's product purchase. Therefore, it can be said that the purchase history is information that can estimate the conditions for the customer's product purchase. For this reason, the condition specifying unit 102 can specify the conditions for the customer's product purchase from the purchase history. For example, if the ratio of discounted products in the customer's purchase history is high, the condition specifying unit 102 may specify that the price of the product is low as a condition for the customer's product purchase. Note that the conditions for the customer's product purchase and the method of specifying the conditions are not limited to this example.

[0039] Furthermore, the condition specifying unit 102 may specify the conditions for the customer's product purchase based on the purchase history and the price and stock quantity of the product that is the same food ingredient as the product included in the purchase history at the time of purchase. The price and stock quantity of the product that is the same food ingredient as the product included in the purchase history at the time of purchase are used to estimate which product the customer selected from the product included in the purchase history. For example, even if a customer wants to prioritize freshness of the product over the price of the product, if the only food ingredient they want to purchase is a spoiled product that is cheap, they may purchase the spoiled product at the cheap price. In this case, the condition specifying unit 102 may refer to information stored in the product information database as the history of product information of products sold to determine the price and stock quantity of the product that is the same food ingredient as the product included in the purchase history at the time of purchase. Alternatively, the product purchased by the customer and the price and stock quantity of the product that is the same food ingredient as the product may be recorded in the customer's purchase history.

[0040] The timing when the condition specifying unit 102 specifies the conditions for the customer's product purchase based on the purchase history is not particularly limited. For example, the condition specifying unit 102 may specify the conditions for the customer's product purchase when a new purchase history is added to the customer's purchase history or at a predetermined frequency. In addition, when the condition specifying unit 102 specifies the conditions for the customer's product purchase based on the purchase history, the specified conditions for the customer's product purchase may be associated with customer identification information and stored in the customer information database.

[0041] In addition, the purchase history used by the condition specifying unit 102 to specify the conditions for the customer's product purchase is not limited to the purchase history in one store, but may be the purchase history in multiple stores. The purchase history used by the condition specifying unit 102 to specify the conditions for the customer's product purchase may be the purchase history for a predetermined period, such as the most recent year.

[0042] Next, the recommended product identification unit 103 is an example of a recommended product identification means that, when the food ingredients desired to be purchased by a customer include products with multiple price settings, identifies a combination of products that meets the customer's conditions for purchasing products from among the products with multiple price settings. For example, the recommended product identification unit 103 identifies a combination of products that meets more of the customer's conditions for purchasing products as a combination of products to be recommended to the customer. For example, the recommended product identification unit 103 identifies a combination of products to be recommended to the customer by using a combinatorial optimization problem.

[0043] Also, for example, the recommended product identification unit 103 identifies a combination of products that satisfies the conditions for the customer's product purchase by selecting an action using a reward function estimated by inverse reinforcement learning of the customer's purchase history. Inverse reinforcement learning estimates a reward function when a problem and behavioral data that maximizes the reward function are given. In the present disclosure, the problem is to select a product that maximizes the reward function from the price and inventory quantity at the time of purchase of each product that is the same food ingredient as the product purchased by the customer. The reward function is estimated to be large when the conditions for the customer's product purchase are satisfied, and is estimated to be small when the conditions for the customer's product purchase are not satisfied. The behavioral data is information indicating the products purchased by the customer, that is, the purchase history. For example, the recommended product identification unit 103 identifies a combination of products that maximizes the reward function identified by the above-mentioned inverse reinforcement learning. The reward function in inverse reinforcement learning is an example of a method of expressing the conditions for the customer's product purchase.

[0044] In addition, the inverse reinforcement learning may be so-called intention learning. Intention learning can learn which viewpoints a person emphasizes based on the person's decision-making history. That is, in intention learning, a weighting factor for each viewpoint is determined. In addition, the viewpoints learned by intention learning can be applied to combinatorial optimization. That is, the viewpoints learned by intention learning are used as constraints for combinatorial optimization. In the present disclosure, the decision-making history is the purchase history of the customer. In addition, the decision-making history may include the price and inventory quantity at the time of purchase of each product that is the same ingredient as the product purchased by the customer, in order to understand from what options the customer selected the product. For example, the viewpoints include, but are not limited to, the above-mentioned conditions for the customer's product purchase, such as the product being fresh, the product being cheap, and the product having a specific attribute. Therefore, the recommended product identification unit 103 can identify a combination of products that the customer himself or herself would normally select, using the viewpoints learned by intention learning and the weighting factor for each viewpoint. The viewpoints and the weighting factor for each viewpoint in intention learning are an example of a method of expressing conditions for the customer's product purchase.

[0045] In addition, when the recommended product specification unit 103 uses inverse reinforcement learning or intention learning, the condition specification unit 102 may not be provided. In this case, the process of performing inverse reinforcement learning or intention learning can be said to correspond to the process of specifying the conditions for the customer's product purchase.

[0046] In addition, when the customer wishes to purchase two or more ingredients of a certain type, the recommended product identification unit 103 may identify a product combination in which a different price is set for each ingredient of a certain type. For example, when the customer wishes to purchase three carrots, the recommended product identification unit 103 may identify a combination of one carrot for 60 yen and two carrots for 100 yen.

[0047] Furthermore, when the ingredients desired by the customer do not include an item with multiple prices set, that is, when the ingredients desired by the customer are only items with a certain price, the recommended product identification unit 103 may identify one item as included in the combination of products. The recommended product identification unit 103 may determine whether the ingredients desired by the customer include an item with multiple prices set by referring to the number of items in stock in the product information database.

[0048] The recommended information output unit 104 is an example of a recommended information output means that outputs recommended information, which is information related to a specified combination of products. The recommended information output unit 104 outputs the recommended information to a terminal device used by a customer. For example, the recommended information output unit 104 may output the recommended information at a timing when a customer opens an application or a website, when a customer comes to a store, or at a preset time. The timing when the recommended information output unit 104 outputs the recommended information is not limited to these examples.

[0049] The recommendation information is information about the combination of products identified. The recommendation information output unit 104 may output a list of products included in the combination of products identified as the recommendation information. The list of products may include at least information such as the name and image of the product for the customer to identify each product, and the quantity of each product. In addition, when there are products that are the same but have different prices, the list of products may further include the price of the product as information for the customer to identify each product. The recommendation information may include information for encouraging the customer to purchase the product included in the combination of products identified. The recommendation information includes, for example, information for the customer to purchase the product included in the combination of products. The information for the customer to purchase the product may include the name of the product, the image of the product, the price of the product, and information indicating the place where the product is sold. The information for the customer to purchase the product included in the combination of products is a part or all of the product information. The information indicating the place where the product is sold may be information indicating the location of the sales floor in a physical store or the URL of the purchase page in an online supermarket. The information for the customer to purchase the product may be included in the list of products described above. Also, for example, when a customer selects a certain product from a list of products displayed on a terminal device, the recommended information output unit 104 may output product information of the selected product so as to be displayed.

[0050] Furthermore, the recommendation information may further include information on products other than the products included in the specified product combination, which is the food ingredient that the customer wishes to purchase. Information on products other than the products included in the specified product combination may include, for example, information on other food ingredients related to the food ingredient that the customer wishes to purchase. Other food ingredients related to the food ingredient that the customer wishes to purchase may be, for example, a type of food ingredient similar to the food ingredient that the customer wishes to purchase, or an ingredient that is used together with the food ingredient that the customer wishes to purchase. By outputting information on products other than the products included in the specified product combination together with information on the product combination, the customer may be able to make a more comprehensive selection of products to purchase.

[0051] Fig. 3 is an example of an output screen of recommended information. In Fig. 4, the output screen of recommended information includes, as information about one of the products included in the specified combination of products, the product name, the product's place of origin, the product's price, and the arrival date indicating the product's freshness. In addition, the output screen of recommended information includes, as information about products other than the products included in the specified combination of products, the product's place of origin, price, and arrival date of products other than the products included in the specified combination of products. Note that the output screen of recommended information is not limited to the example of Fig. 4.

[0052] In addition, if the food ingredient that the customer wishes to purchase is not on sale, that is, if the food ingredient product that the customer wishes to purchase is out of stock or the food ingredient product that the customer wishes to purchase is not available, the recommended information output unit 104 may output a message that the food ingredient that the customer wishes to purchase is not on sale.

[0053] An example of the operation of the product recommendation device 100 configured as above will be described with reference to the flowchart of FIG.

[0054] First, the ingredient specification unit 101 specifies the type of ingredients and the quantity of each ingredient that the customer wishes to purchase based on the input information entered by the customer (step S101). The ingredient specification unit 101 may start processing when the customer inputs the ingredients that the customer wishes to purchase, when the customer opens an application or website, or at a predetermined time.

[0055] Next, the condition specifying unit 102 specifies the conditions for the customer's product purchase based on the input information (step S102).

[0056] Next, if the ingredients desired to be purchased by the customer identified in step S101 include products with multiple price settings, the recommended product identification unit 103 identifies a combination of products that meets the product purchase conditions of the customer identified in step S102 from the ingredients desired to be purchased by the customer and that have multiple price settings (step S103).

[0057] Then, the recommended information output unit 104 outputs information about the combination of products identified in step S103 (step S104).

[0058] With this, the product recommendation device 100 ends a series of operations.

[0059] The product recommendation device 100 in the above-described embodiment includes an ingredient specification unit 101, a condition specification unit 102, a recommended product specification unit 103, and a recommended information output unit 104. The ingredient specification unit 101 specifies the type of ingredients and the quantity of each ingredient that the customer desires to purchase based on the input information input by the customer. The condition specification unit 102 specifies the conditions for the customer's product purchase based on the input information. When the ingredients that the customer desires to purchase include products with multiple prices, the recommended product specification unit 103 specifies a combination of products that meet the customer's conditions from among the products with multiple prices. Then, the recommended information output unit 104 outputs recommended information that is information about the combination of products. This allows the product recommendation device 100 to recommend products that meet the customer's conditions for product purchase to the customer. Products that meet the customer's conditions for product purchase are products that the customer is likely to purchase.

[0060] As a result, in a case where a product having the same ingredient and multiple prices is sold, the product recommendation device 100 in this embodiment can make product suggestions that reduce the effort required for the customer to select a product.

[0061] [Variation 1] The product recommendation device 100 may include a recipe identification unit 105 that identifies a recipe that the customer wishes to cook.

[0062] 5 is a block diagram showing the configuration of the product recommendation device 100 A. The product recommendation device 100 A may include a recipe specification unit 105 in addition to the configuration of the product recommendation device 100 .

[0063] The recipe identification unit 105 is an example of a recipe identification means that identifies a recipe that the customer wishes to cook based on the input information entered by the customer. The recipe identification unit 105 identifies one or more recipes that the customer wishes to cook. In this case, the ingredient identification unit 101 identifies ingredients that the customer wishes to purchase based on the identified recipe.

[0064] An example of a method in which the recipe specifying unit 105 specifies a recipe that a customer wishes to cook will be described.

[0065] For example, recipe identification unit 105 may identify the input information entered by the customer as the recipe the customer intends to cook. In this case, for example, the customer inputs information indicating the recipe the customer intends to cook in an application or website. The information indicating the recipe the customer intends to cook may be text data indicating the recipe, or may be a uniform resource locator (URL) where the recipe is displayed, or the name of a dish, etc. If the information indicating the recipe the customer intends to cook is the name of a dish, recipe identification unit 105 may identify the recipe from the name of the dish by using a database or website.

[0066] Also, for example, the recipe identification unit 105 may identify information associated with the customer identification information input by the customer as the recipe that the customer intends to cook. In this case, for example, the customer registers the recipe that the customer intends to cook in advance in an application or a website. Also, for example, in this case, the information associated with the customer identification information may be information indicating the recipe that is estimated to be cooked by the customer. At this time, the recipe that is estimated to be cooked by the customer may be estimated based on the customer's purchase history that is associated with the customer identification information by another device, for example.

[0067] Based on the recipe that the customer intends to cook, identified as described above, the ingredient identification unit 101 may identify ingredients that the customer wishes to purchase. For example, the ingredient identification unit 101 may identify ingredients contained in the text data of the recipe as ingredients that the customer wishes to purchase.

[0068] [Variation 2] The product recommendation device 100 may identify a combination of products to be recommended to a customer from products sold in multiple stores. In this case, a product that is the same food ingredient and has multiple prices includes a product sold in multiple stores.

[0069] For example, a customer may purchase food ingredients from multiple stores. The availability of a product, its price, inventory, and freshness may vary depending on the store. In this case, the product recommendation device 100 identifies a combination of products from products sold in multiple stores and outputs recommendation information, making it easier for the customer to select products to purchase.

[0070] In this case, the condition specification unit 102 may specify further conditions for the customer's product purchase, such as whether it is acceptable to shop at multiple stores, and if so, what is the acceptable distance between the multiple stores.

[0071] The recommended product specification unit 103 may specify a combination of products from among a plurality of products that are ingredients that the customer wishes to purchase and are sold in a plurality of stores, by referring to product information databases of a plurality of stores.

[0072] The recommended information output unit 104 further outputs, as information related to the combination of products, information related to the store that sells the products included in the identified combination of products.

[0073] [Second embodiment] Next, a second embodiment of the present disclosure will be described in detail with reference to the drawings. In the following, the description of the second embodiment will be omitted unless the description of the present embodiment is unclear.

[0074] In the second embodiment, the product recommendation device 200 outputs business support information based on the conditions for product purchases by customers, which are specified by the condition specification unit 102. The product recommendation device 200 in the second embodiment aims to enable sales of products with high appeal to customers who use the store by outputting business support information based on the conditions for product purchases by customers who use the store.

[0075] Fig. 6 is a block diagram showing the configuration of the product recommendation device 200. In Fig. 4, the product recommendation device 200 includes a trend identification unit 206, a generation unit 207, and a business support information output unit 208 in addition to the configuration of the first embodiment.

[0076] The trend identification unit 206 is an example of a trend identification means for identifying a trend of conditions for product purchases by customers at a store, based on the conditions for product purchases by multiple customers who use the store. The trend identification unit 206 may identify, for example, the ratio of conditions that customers who use the store place importance on, as the trend of conditions for product purchases by customers who use the store. In this case, for example, the trend identification unit 206 may identify the ratio of the sum of values ​​obtained by multiplying the conditions for product purchases by the weights of each customer who uses the store. Also, for example, the trend identification unit 206 may identify the ratio of conditions that customers who use the store place most importance on.

[0077] The customers for whom the trend identification unit 206 identifies the trend of conditions may be customers who frequently use the store. For example, the trend identification unit 206 may identify the trend based on the conditions for product purchases of customers who use the store with a frequency of use of the store equal to or greater than a predetermined value among a plurality of customers who use the store. Also, for example, the trend identification unit 206 may identify the trend based on the conditions for product purchases of customers who use the store with a frequency of use of the store equal to or greater than a predetermined value among a plurality of customers who use the store. Also, for example, the trend identification unit 206 may identify the trend based on the conditions for product purchases of customers who use the store for a period of time equal to or greater than a predetermined value among a plurality of customers who use the store. These predetermined values ​​are not particularly limited, and may be set to values ​​that can extract customers who frequently use the store.

[0078] The generating unit 207 is an example of a generating means that generates business support information related to products sold in the store based on the trend identified by the trend identifying unit 206. The business support information related to products sold in the store may be, for example, at least one of information suggesting products to be sold, information suggesting criteria for changing the prices of products, and information suggesting the purchase quantity of products.

[0079] For example, the generating unit 207 generates information proposing products for sale based on the trend. The information proposing products for sale may be information proposing attributes and prices of the products for sale. For example, the generating unit 207 may propose products that meet a condition that is highly valued by customers as products for sale. For example, the generating unit 207 generates information proposing a food product that meets the condition "domestic origin" that is highly valued by customers, and that is not available as a product that meets the condition "domestic origin", as a product for sale. Also, for example, the generating unit 207 may generate information proposing a price of a product for sale so that the price meets the condition "low price" that is highly valued by customers.

[0080] Furthermore, the generating unit 207 may generate information proposing a criterion for changing the price of a product based on a trend. The criterion for changing the price of a product is, for example, a criterion for lowering the price according to freshness. For example, the generating unit 207 may generate information proposing to raise the freshness as a criterion for lowering the price when a ratio of customers who attach importance to the condition "low price" becomes high to a certain extent. Furthermore, for example, the generating unit 207 may generate information proposing a time period for lowering the price according to freshness based on the trend of customers for each time period.

[0081] Furthermore, the generating unit 207 may generate information that suggests the purchase amount of a product based on the trend. For example, the generating unit 207 may generate information that suggests the purchase amount of each different product of the same ingredient in a ratio according to the ratio of the condition that the customer values. For example, when the ratio of the condition "domestic origin" is higher than the condition "low price" shown in the trend, the generating unit 207 may generate information that suggests the purchase amount of a product that is domestically produced and expensive is to be greater than the purchase amount of a product that is the same ingredient but is produced abroad and is cheap.

[0082] The business support information output unit 208 is an example of a business support information output means that outputs, as business support information, information on the trend identified by the trend identification unit 206. The information on the trend may be information indicating the trend itself, or may be business support information generated by the generation unit 207.

[0083] The business support information output unit 208 outputs the business support information to, for example, a terminal device used by a store clerk.

[0084] An example of the operation of the product recommendation device 200 configured as above will be described with reference to the flowchart of Fig. 7. In Fig. 7, it is assumed that conditions for customers' product purchases are stored in advance.

[0085] First, the trend identifying unit 206 identifies a trend in the conditions for product purchases made by customers who visit the store (step S201).

[0086] Next, the generating unit 207 generates business support information based on the tendency of conditions for product purchases of customers who use the store, identified in step S201 (step S202).

[0087] Next, the business support information output unit 208 outputs the business support information generated in step S202 (step S203).

[0088] With this, the product recommendation device 200 ends a series of operations.

[0089] The product recommendation device 200 in the above-described embodiment includes a trend identification unit 206, a generation unit 207, and a business support information output unit 208. The trend identification unit 206 identifies a trend of product purchase conditions of customers at the store based on the product purchase conditions of multiple customers who use the store. The generation unit 207 generates business support information related to products sold at the store based on the trend. Then, the business support information output unit 208 outputs information related to the identified trend as business support information. Therefore, the store clerk can grasp the product purchase trends of customers who use the store. In addition, the store clerk can thereby determine the type, quantity, price, etc. of products to be sold at the store according to the product purchase trends of customers who use the store.

[0090] As a result, the product recommendation device 200 in this embodiment can make it possible to sell products that are highly appealing to customers who visit the store.

[0091] [Hardware configuration] Some or all of the components of each device or system in each embodiment of the present disclosure described above are realized by any combination of an information processing device 1000 and a program as shown in Fig. 8. The information processing device 1000 includes, as an example, the following configuration.

[0092] ·CPU(Central Processing Unit)1001 ROM (Read Only Memory) 1002 ·RAM(Random Access Memory)1003 Program 1004 loaded into RAM 1003 A storage device 1005 for storing a program 1004 A drive device 1007 for reading and writing data from the recording medium 1006 A communication I / F 1008 that connects to a communication network 1009 Input / output I / F 1010 for inputting and outputting data A bus 1011 connecting each component Note that I / F is an abbreviation for Interface.

[0093] Each component of each device or system in each embodiment is realized by CPU 1001 acquiring and executing a program that realizes these functions. The program that realizes the function of each component of each device is stored in storage device 1005 or RAM 1003 in advance, for example, and is read by CPU 1001 as necessary. Program 1004 may be supplied to CPU 1001 via a communication network, or may be stored in recording medium 1006 in advance, and drive device 1007 may read the program and supply it to CPU 1001.

[0094] There are various variations in the method of implementing each device. For example, each device or system may be implemented by any combination of a separate information processing device 1000 and a program for each component. Also, multiple components included in each device may be implemented by any combination of a single information processing device 1000 and a program.

[0095] Further, a part or all of each component of each device or system is realized by a general-purpose or dedicated circuit including a processor or the like, or a combination of these. The circuit is, for example, a CPU, a GPU (Graphics Processing Unit), an FPGA (Field Programmable Gate Array), or an LSI (Large Scale Integration) for AI (Artificial Intelligence) processing. These may be configured by a single chip, or may be configured by multiple chips connected via a bus. A part or all of each component of each device may be realized by a combination of the above-mentioned circuits and a program.

[0096] When a part or all of each component of each device or system is realized by a plurality of information processing devices, circuits, etc., the plurality of information processing devices, circuits, etc. may be centrally arranged or distributed. For example, the information processing devices, circuits, etc. may be realized as a client-server system, a cloud computing system, etc., in a form in which each is connected via a communication network.

[0097] Although the present invention has been described above with reference to the embodiments, the present invention is not limited to the above-mentioned embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.

[0098] For example, the product recommendation device 200 of the second embodiment may be configured not to include the generating unit 207. Also, for example, the product recommendation device 200 of the second embodiment may be modified to a business support device including the condition specifying unit 102, the trend specifying unit 206, the generating unit 207, and the business support information output unit 208.

[0099] In addition, although the operations are described in a sequence in the form of a flowchart, the sequence does not limit the order in which the operations are performed. Therefore, when implementing each embodiment, the sequence of the operations may be changed as long as it does not affect the content.

[0100] A part or all of the above-described embodiments can be described as, but is not limited to, the following supplementary notes.

[0101] [Appendix 1] A food ingredient specification means for specifying the type of food ingredient and the quantity of each food ingredient that the customer wishes to purchase based on input information input by the customer; A condition specification means for specifying a condition for the customer's product purchase based on the input information; a recommended product specification means for specifying a combination of products that meet the requirements of the customer among the products with multiple prices set for the food ingredients, when the food ingredients include products with multiple prices set; a recommendation information output means for outputting recommendation information which is information regarding the combination of products; A product recommendation device comprising:

[0102] [Appendix 2] The input information includes identification information for identifying the customer, the condition specifying means specifies the condition based on a purchase history of the customer associated with the identification information. Item recommendation device according to appended claim 1.

[0103] [Appendix 3] The input information includes identification information for identifying the customer, the recommended product specification means specifies the combination of products by using conditions for product purchases of the customer specified by performing inverse reinforcement learning on the purchase history of the customer associated with the identification information; Item recommendation device according to appended claim 1.

[0104] [Appendix 4] The condition specifying means specifies the condition for each ingredient. 4. A product recommendation device according to any one of claims 1 to 3.

[0105] [Appendix 5] The condition specifying means specifies the condition for each category of ingredients. Item recommendation device according to appended claim 1.

[0106] [Appendix 6] The category is a classification indicating at least whether the food ingredient is a fresh food or not. Item 5. A product recommendation device as described in Appendix 5.

[0107] [Appendix 7] A recipe specifying means for specifying a recipe that the customer wishes to cook based on the input information, The ingredient identification means identifies the ingredient based on the recipe. 7. A product recommendation device according to any one of claims 1 to 6.

[0108] [Appendix 8] The recommendation information further includes information regarding products other than the products included in the combination of products that are the ingredients. 8. A product recommendation device according to any one of claims 1 to 7.

[0109] [Appendix 9] The product with multiple prices includes a product sold at multiple stores. 9. A product recommendation device according to any one of claims 1 to 8.

[0110] [Appendix 10] The recommendation information further includes, as information about the combination of products, information about a store that sells the products included in the combination of products. 10. A product recommendation device as described in Appendix 9.

[0111] [Appendix 11] A trend identifying means for identifying a trend of the conditions in the store based on the conditions of a plurality of the customers who use the store; and a business support information output means for outputting information relating to the trend as business support information. 11. A product recommendation device according to any one of claims 1 to 10.

[0112] [Appendix 12] the trend identification means identifies the trend based on the condition of customers who use the store with a frequency of use of the store equal to or greater than a predetermined value, among the plurality of customers who use the store; 12. A product recommendation device as described in appended claim 11.

[0113] [Appendix 13] the trend identification means identifies the trend based on the condition of a customer who visits the store a predetermined number of times or more among the plurality of customers who visit the store; 13. A product recommendation device according to claim 11 or 12.

[0114] [Appendix 14] the trend identification means identifies the trend based on the condition of a customer who visits the store for a period of time equal to or longer than a predetermined value, among the plurality of customers who visit the store; 14. A product recommendation device according to any one of appendix 11 to 13.

[0115] [Appendix 15] A generating unit is further provided for generating business support information related to products sold in the store based on the trend. 15. A product recommendation device according to any one of appendix 11 to 14.

[0116] [Appendix 16] The generating means generates information proposing a price of the product based on the trend. 16. A product recommendation device as described in appended claim 15.

[0117] [Appendix 17] The generating means generates information proposing a criterion for changing the price of the product based on the trend. 17. A product recommendation device according to claim 15 or 16.

[0118] [Appendix 18] The generating means generates information for proposing a purchase amount of the product based on the trend. 18. The product recommendation device according to any one of appendix 15 to 17.

[0119] [Appendix 19] The computer Identifying the types of ingredients and the quantity of each ingredient that the customer wishes to purchase based on the input information entered by the customer; Identifying conditions for the customer's product purchase based on the input information; If the food ingredients include products with multiple price settings, a combination of products that meet the customer's requirements is identified from among the products with multiple price settings; outputting recommendation information which is information regarding the combination of products; Product recommendation methods.

[0120] [Appendix 20] Identifying the types of ingredients and the quantity of each ingredient that the customer wishes to purchase based on the input information entered by the customer; Identifying conditions for the customer's product purchase based on the input information; If the food ingredients include products with multiple price settings, a combination of products with multiple price settings that satisfies the condition of the customer is identified, outputting recommendation information which is information regarding the combination of products; A program that causes a computer to carry out processing.

[0121] □ In addition, some or all of the configurations described in Supplementary Notes 2 to 18 that are subordinate to the product recommendation device of Supplementary Note 1 described above may also be subordinate to the product recommendation method of Supplementary Note 19 and the program of Supplementary Note 20 in a similar subordinate relationship to Supplementary Note 2 to Supplementary Note 18. Furthermore, not limited to Supplementary Note 1, Supplementary Note 19, and Supplementary Note 20, some or all of the configurations described as supplementary notes may also be subordinated to various hardware, software, various recording means for recording software, or systems within the scope of each of the above-mentioned embodiments. [Explanation of symbols]

[0122] 100 Product Recommendation Device 101 Food Identification Department 102 Condition specification part 103 Recommended Product Identification Department 104 Recommended Information Output Section 100A Product Recommended Equipment 105 Recipe Specifications 200 Product Recommendation Device 206 Trend Identification Department 207 Generation part 208 Business Support Information Output Unit 1000 Information processing device 1001 CPU 1002 ROM 1003 RAM 1004 Program 1005 Storage device 1006 Recording media 1007 Drive device 1008 Communication I / F 1009 Communication Network 1010 Input / Output Interface 1011 Bus

Claims

1. Food identification means for identifying the types of food ingredients the customer wishes to purchase and the quantity of each food ingredient based on the input information entered by the customer; Condition identification means for identifying the conditions in the customer's product purchase based on the input information; When the food ingredients include products with multiple set prices, recommended product identification means for identifying combinations of products that meet the customer's conditions among the products with multiple set prices respectively; Recommended information output means for outputting recommended information which is information regarding the combination of the products; A product recommendation device comprising the above.

2. The input information includes identification information for identifying the customer, and the condition identification means identifies the conditions based on the customer's purchase history associated with the identification information. The product recommendation device according to Claim 1.

3. The input information includes identification information for identifying the customer, and the recommended product identification means identifies the combination of the products using the conditions in the customer's product purchase identified by performing inverse reinforcement learning on the customer's purchase history associated with the identification information. The product recommendation device according to Claim 1.

4. The condition identification means identifies the conditions for each category of food ingredients. The product recommendation device according to Claim 1.

5. The category is a classification indicating at least whether the food ingredient is fresh food. The product recommendation device according to Claim 4.

6. The product recommendation device further comprises recipe identification means for identifying the recipe the customer intends to cook based on the input information, and the food identification means identifies the food ingredients based on the recipe. The product recommendation device according to Claim 1.

7. The products with multiple set prices include products sold in multiple stores. The product recommendation device according to Claim 1.

8. Trend identification means for identifying the trend of the conditions in the store based on the conditions of the multiple customers using the store, and business support information output means for outputting information regarding the trend as business support information. The product recommendation device according to Claim 1.

9. A computer, food identification means for identifying the types of food ingredients the customer wishes to purchase and the quantity of each food ingredient based on the input information entered by the customer; condition identification means for identifying the conditions in the customer's product purchase based on the input information; When the food ingredients include products with multiple set prices, recommended product identification means for identifying a combination of products that meet the customer's conditions among the products with multiple set prices respectively; Outputting recommended information which is information regarding the combination of the products; A product recommendation method.

10. Food ingredient identification means for identifying food ingredients that the customer wishes to purchase based on input information input by the customer; Condition identification means for identifying conditions in the customer's product purchase based on the input information; When the food ingredients include products with multiple set prices, recommended product identification means for identifying a combination of products that meet the customer's conditions among the products with multiple set prices respectively; Outputting recommended information which is information regarding the combination of the products; A program for causing a computer to execute processing.

Images