Product recommendation systems, product recommendation methods, and programs

Abstract

To provide a merchandise recommendation system, a merchandise recommendation method, and a program which can easily find out a merchandise suitable for a customer.SOLUTION: In a merchandise recommendation system 10, an article detection part 101 detects an article that a customer visiting a shop wears. A web analysis part 102 analyzes a web, and thereby specifies an article used together with the detected article. A determination part 103 determines a recommendation merchandise from merchandises sold at the shop, on the basis of the specified article. An output control part 104 recommends the determined recommendation merchandise to the customer.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to a product recommendation system and the like.

Background Art

[0002] Customers visit a store to shop for clothing items. For example, a store clerk may propose clothing items and the like to a customer who has visited the store as part of customer service.

[0003] For example, Patent Document 1 describes analyzing a Social Networking Service (hereinafter referred to as SNS), predicting future demand for clothing, identifying the types of clothing included in a user image, estimating the body type of the user from the types of clothing, and selecting clothing that fits the body type of the consumer.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] It may be difficult to find products suitable for customers in a store.

[0006] An example of the object of the present disclosure is to provide a product recommendation system and the like that facilitate finding products suitable for customers.

Means for Solving the Problems

[0007] A product recommendation system in one aspect of the present disclosure includes: an item detection means for detecting an item worn by a customer visiting a store; a web analysis means for identifying an item used together with the detected item by analyzing the web; a determination means for determining a recommended product from the products sold at the store based on the identified item; and an output control means for recommending the determined recommended product to the customer.

[0008] A product recommendation method in one aspect of this disclosure involves detecting items worn by a customer visiting a store, identifying items used together with the detected items by analyzing the web, determining recommended products from the products sold at the store based on the identified items, and recommending the determined recommended products to the customer.

[0009] A program in one aspect of the present disclosure causes a computer to perform the following processes: detect items worn by a customer visiting a store; identify items used together with the detected items by analyzing the web; determine recommended items from the products sold in the store based on the identified items; and recommend the determined recommended items to the customer.

[0010] The program may be stored on a non-temporary storage medium that is readable by the computer. [Effects of the Invention]

[0011] This disclosure makes it easier for customers to find products that are suitable for them. [Brief explanation of the drawing]

[0012] [Figure 1] This is a block diagram showing an example configuration of a product recommendation system according to Embodiment 1. [Figure 2] This flowchart shows an example of the operation of the product recommendation system according to Embodiment 1. [Figure 3] This is an explanatory diagram illustrating an example of a customer visiting a store. [Figure 4] This is an explanatory diagram showing an example of connecting the product recommendation system to other devices. [Figure 5] This is a block diagram showing an example configuration of a product recommendation system according to Embodiment 2. [Figure 6] This is an explanatory diagram showing an example of displaying information about recommended products. [Figure 7] This is an explanatory diagram illustrating an example of displaying information for multiple recommended products. [Figure 8] This is a flowchart illustrating an example of the operation of a product recommendation system according to an embodiment. [Figure 9] This is an explanatory diagram showing an example of a computer hardware configuration. [Modes for carrying out the invention]

[0013] Embodiments of the product recommendation system, product recommendation method, program, and non-temporary recording medium for recording the program relating to this disclosure will be described in detail below with reference to the drawings. These embodiments are not intended to limit the technology of the disclosure.

[0014] (Embodiment 1) First, the basic functions of the product recommendation system will be described in Embodiment 1. Figure 1 is a block diagram showing an example configuration of the product recommendation system according to Embodiment 1. The product recommendation system 10 comprises an item detection unit 101, a web analysis unit 102, a determination unit 103, and an output control unit 104.

[0015] The item detection unit 101 detects items worn by customers in a store. Specifically, for example, the item detection unit 101 acquires an image of the customer captured by an imaging device. Then, the item detection unit 101 detects the items worn by the customer from the image. The imaging device may be installed in the store, for example. The imaging device may be equipped in a portable terminal device.

[0016] For example, as an example where the imaging device is installed in a store, the imaging device may be provided in a signboard or the like. For example, when a customer visits a store and wants to look for products in the store, the customer stands in front of the signboard. Then, the imaging device provided in the signboard may image the customer and the items the customer is wearing. The signboard may transmit the captured image to the product recommendation system 10.

[0017] Also, for example, the imaging device may be provided in a portable terminal device. The portable terminal device may be a store employee's terminal device or a customer's terminal device. For example, the imaging device provided in the terminal device may image the customer at the customer's operation. Or, the imaging device provided in the terminal device may image the customer at the store employee's operation. Then, the terminal device may transmit the captured image to the product recommendation system 10.

[0018] Here, for example, the items the customer is wearing are clothing accessories such as clothes, bags, shoes, hats, and accessories. That the customer is wearing means at least any one of the state of the customer wearing, the state of the customer putting on, the state of the customer holding, and the state of the customer wearing. Here, the items the customer is wearing are also called wearing items.

[0019] Next, the web analysis unit 102 identifies items used in conjunction with the worn items by analyzing the web. Specifically, for example, the web analysis unit 102 identifies items used in conjunction with the worn items by analyzing images on the web. Here, the web being analyzed refers to internet services. Internet services being analyzed are, for example, services that allow images to be posted. For example, services that allow images to be posted include SNS, blogs, homepages, and EC (Electronic Commerce) sites. SNS are community sites where people can build relationships online. For example, images can be posted on SNS. The web being analyzed may also be an internet service used by the customer. The web being analyzed may also be an internet service designated by the store. The web being analyzed may also be an internet service where images of products in the store have been posted. Thus, the web being analyzed is not particularly limited. Hereafter, SNS will be used as an example of a web being analyzed.

[0020] For example, the Web Analysis Unit 102 identifies items used in conjunction with worn items by analyzing images on social media. Images on social media refer to images included in social media posts. For instance, the Web Analysis Unit 102 analyzes social media posts within a specified period, such as recent posts. This allows for the analysis of posts that reflect recent trends.

[0021] More specifically, as a method for identifying items, for example, the web analysis unit 102 analyzes images included in SNS posts. Then, for example, the web analysis unit 102 identifies images containing worn items from the images included in SNS posts. As a method for identifying images containing worn items, the web analysis unit 102 detects images containing worn items by matching images included in SNS posts with images of worn items. For example, if the degree of matching between a partial image of an item in an image included in an SNS post and an image of the worn item is above a predetermined degree, the web analysis unit 102 detects the item in the image included in the post as the same item as the worn item. Therefore, items identified as the same item as the worn item may include not only completely identical items but also items similar to the worn item. Then, the web analysis unit 102 identifies items used together with the worn item in the detected images.

[0022] Furthermore, items used together with the worn item refer to items worn together with the worn item in the posted image. For example, the image may include a coordinated outfit of a white skirt, top, shoes, bag, and hat. For example, if the detected worn item is a white skirt, the web analysis unit 102 will identify items such as a top, shoes, bag, and hat that are worn together with the white skirt from the image.

[0023] The decision unit 103 determines recommended products from the products sold in the store based on the items identified by the web analysis unit 102. For example, the decision unit 103 determines at least one of the products that are the same as the identified item, or a product that is similar to the identified item, as recommended products from the products sold in the store.

[0024] The output control unit 104 then recommends recommended products to the customer, for example. Specifically, as a method of recommendation, the output control unit 104 may notify the customer of recommended products by electronic message or email. Alternatively, as a specific method of recommendation, the output control unit 104 may cause the output device to output information about the recommended products. The information about the recommended products to be output is not particularly limited and may include the name of the recommended product, an image of the recommended product, or the display location of the recommended product in the store. Note that the display location may be the location on a map of the store.

[0025] Furthermore, the output device may be a display device or an audio output device, and is not particularly limited. For example, if the output device is a display device, the output control unit 104 may cause the display device to display information about recommended products. If the output device is an audio output device, the output control unit 104 may cause the audio output device to output information about recommended products as audio. Also, the output device may be, for example, a portable terminal device or a device installed in a store. As mentioned above, the terminal device may be, for example, a customer's terminal device or a store employee's terminal device. For example, if information about recommended products is output to a store employee's terminal device, the recommended products may be recommended to the customer through the store employee. Also, the device installed in the store may be a digital signage display. For example, the output control unit 104 may cause the digital signage display information about recommended products.

[0026] Figure 2 is a flowchart showing an example of the operation of the product recommendation system 10 according to Embodiment 1. The item detection unit 101 detects the items worn by a customer who has come into the store (step S101).

[0027] Next, the web analysis unit 102 identifies items used together with the worn items by analyzing the web (step S102).

[0028] Then, the decision unit 103 determines a recommended product from the products sold in the store based on the items identified by the web analysis unit 102 (step S103). Then, the output control unit 104 recommends the determined recommended product to the customer (step S104). The product recommendation system 10 then terminates its process.

[0029] As described above, in Embodiment 1, the product recommendation system 10 analyzes the web to identify items that are used together with items worn by customers visiting the store. Then, based on the identified items, the product recommendation system 10 recommends recommended products from among the products sold in the store to the customer. In this way, by using the web, it is possible to recommend trendy items that are suitable for the customer. Therefore, it is possible to make it easier for customers to find products that are suitable for them when they visit the store.

[0030] Here, the device that notifies customers of recommended products and the device that captures images of items worn by the customer may be the same or different. For example, the signage may capture images of the customer, and the customer's terminal device may output information about recommended products. In such a case, the signage may biometrically authenticate the customer. The output control unit 104 may then notify the authenticated customer's terminal device of the information about recommended products. Alternatively, for example, the signage may capture images of the customer and output information about recommended products. Furthermore, by allowing customers to check recommended products using their terminal devices, it becomes easier for customers to find products that are suitable for them.

[0031] Furthermore, for example, a device that notifies customers of recommended products may be used to display these recommendations on the employee's terminal or digital signage, allowing the employee to review the recommended products and recommend them to customers. In this way, the product recommendation system 10 may be used to support the customer service provided by the employee.

[0032] (Embodiment 2) Next, Embodiment 2 will be described in detail with reference to the drawings. In Embodiment 2, SNS will be used as an example of the web to be analyzed. Furthermore, Embodiment 2 will describe in more detail an example of recommending products that, when used with the worn items, will result in a more trendy coordinated look. To the extent that the description of Embodiment 2 does not become unclear, explanations that overlap with the above description will be omitted.

[0033] Figure 3 is an explanatory diagram illustrating an example of a customer entering a store. For example, an imaging device 23 is installed in the store. The imaging device 23 photographs customers who enter the store. As a result, the imaging device 23 can photograph items that the customer is wearing. The imaging device 23 also photographs customers while they are shopping in the store. As a result, customer behavior can be analyzed from the captured images and videos.

[0034] For example, the number of imaging devices 23 is not particularly limited. For instance, some imaging devices 23 may be provided on signage or terminal devices, and some imaging devices 23 may be installed in stores.

[0035] Specifically, for example, the product recommendation system according to Embodiment 2 may, as described in Embodiment 1, have an imaging device 23 provided on the signage capture images of customers who have come into the store, and detect the items the customer is wearing from the captured images.

[0036] Furthermore, for example, the product recommendation system according to Embodiment 2 may detect products that customers are interested in by having an imaging device 23 installed in the store capture images of customers shopping in the store and analyzing the customer's behavior from the captured images.

[0037] Figure 4 is an explanatory diagram showing an example of connecting a product recommendation system to other devices. For example, the product recommendation system 20 is connected to the imaging device 23 via a communication network NT. The imaging device 23 may continuously capture images or may capture images in response to the control of the product recommendation system 20. The imaging device 23 may also capture images of customers in order to analyze their behavior when they visit a store. The number of imaging devices 23 is not particularly limited.

[0038] For example, the product recommendation system 20 is connected to terminal devices 21 and the like via a communication network NT. In Figure 4, the number of terminal devices 21 is not particularly limited. Terminal devices 21 may be customer terminal devices 21 or store employee terminal devices 21. Also, terminal devices 21 may be terminal devices 21 installed in the store. The type of terminal device 21 is not particularly limited and can be a smartphone, tablet device, or PC (Personal Computer).

[0039] Furthermore, for example, the product recommendation system 20 is connected to SNS 22 and other devices via the communication network NT. The type and number of SNS 22 devices are not particularly limited.

[0040] In Figure 4, one communication network NT is shown as an example for the sake of simplicity, but the recommended product system 20 may be connected by different communication networks NT for each device, and is not particularly limited.

[0041] Although not shown in the diagram, the product recommendation system 20 may also be connected to signage installed in the store via a communication network.

[0042] Figure 5 is a block diagram showing an example configuration of the product recommendation system 20 according to Embodiment 2. The product recommendation system 20 comprises an item detection unit 201, a web analysis unit 202, a decision unit 203, an output control unit 204, a behavior analysis unit 205, and a system identification unit 206. The product recommendation system 20 according to Embodiment 2 is the product recommendation system 10 according to Embodiment 1 with the addition of a behavior analysis unit 205 and a system identification unit 206.

[0043] The item detection unit 201 has the functions of the item detection unit 101 according to Embodiment 1 as its basic functions. The web analysis unit 202 has the functions of the web analysis unit 102 according to Embodiment 1 as its basic functions. The determination unit 203 has the functions of the determination unit 103 according to Embodiment 1 as its basic functions. The output control unit 204 has the functions of the output control unit 104 according to Embodiment 1 as its basic functions.

[0044] The product recommendation system 20 includes a product database 2001. The product database 2001 stores information about items sold in stores, for example. The product database 2001 stores product identification information and product information in association for each product. The product identification information is not particularly limited as long as it can uniquely identify the product. For this reason, the product identification information may be represented by a code or by a product name. The product information includes the product name, product image, product classification, product design, product price, inventory quantity, and release date. The product image is an image containing the product. For example, the product image may be an image posted on SNS 22. For example, the product classification may be a broad category such as bottoms, tops, bags, shoes, hats, and accessories. Alternatively, the product classification may be a subcategory such as dress shirts and T-shirts within tops. The product design is not particularly limited and may include the product color, pattern or design, material, and shape.

[0045] Next, I will explain each functional part.

[0046] The item detection unit 201 detects items worn by the customer. The method by which the item detection unit 201 detects items is as described above.

[0047] The web analysis unit 202 then identifies items used together with the worn items by analyzing the web. As mentioned above, SNS22 will be used as an example of a web. As explained in Embodiment 1, the web analysis unit 202 detects images containing worn items by analyzing images on the web and identifies items used together with the worn items in the detected images. Here, Embodiment 2 will describe an example in which the classification of items recommended to the customer is the classification of the items the customer is looking for. Alternatively, Embodiment 2 will describe an example in which the design of the items recommended to the customer is the design of the items the customer is looking for.

[0048] <Classification of items that customers are interested in> The behavioral analysis unit 205 detects the classification of products that customers are interested in in a store. Specifically, for example, the behavioral analysis unit 205 analyzes customer behavior in the store. Then, for example, based on the product database 2001, the behavioral analysis unit 205 identifies products that customers picked up or looked at as products that customers are interested in. Then, for example, the behavioral analysis unit 205 can detect the classification of the identified products based on the product database 2001. This makes it possible to detect the classification of products that customers are interested in in a store.

[0049] Furthermore, a product picked up by a customer may be a product the customer has picked up multiple times. A product picked up by a customer may also be a product the customer has tried on. A product that a customer looks at may be a product the customer has looked at for a predetermined amount of time or longer, and a product that a customer looks at may be a product the customer has looked at for a predetermined number of times or longer. In this way, a product that a customer shows interest in may be a product that was detected by analyzing their behavior in more detail.

[0050] The Web Analysis Department 202 analyzes SNS 22 to identify items of a detected category that are used together with the worn items. For example, if the category of the product a customer is interested in is T-shirts, the Web Analysis Department 202 identifies T-shirts among the items used together with the worn items. More specifically, for example, the Web Analysis Department 202 identifies images containing worn items from the images included in the posts. Then, for example, the Web Analysis Department 202 identifies items other than worn items that are classified as T-shirts from the identified images.

[0051] <Designs of items that customers found interesting> The behavioral analysis unit 205 may detect the designs of products that customers are interested in at the store. Examples of how the behavioral analysis unit 205 detects products that customers are interested in are as described above. For example, the behavioral analysis unit 205 can identify the designs of products that customers are interested in based on the product database 2001.

[0052] The Web Analysis Department 202 analyzes SNS 22 to identify items with the detected design that are used together with the worn items. For example, if the design of a product that a customer is interested in is a polka dot pattern, the Web Analysis Department 202 identifies items with a polka dot pattern that are used together with the worn items. More specifically, for example, the Web Analysis Department 202 identifies images containing the worn items from the images included in the post. Then, for example, the Web Analysis Department 202 identifies items other than the worn items that have a polka dot pattern from the identified images.

[0053] Here, the classification and design of items of interest to the customer may be combined. The method of combination is not particularly limited. For example, the web analysis unit 202 may, by analyzing SNS22, identify items that are used with the worn item, have a detected design, and belong to the detected classification. Alternatively, the web analysis unit 202 may, by analyzing SNS22, identify at least one of the items that are used with the worn item, has a detected design, and belongs to the detected classification.

[0054] <Recommendations based on fashion style> Customers may shop according to their preferred fashion style. Therefore, the style identification unit 206 identifies the customer's fashion style from the items the customer is wearing in the store. Here, the fashion style can be conservative, traditional, trendy, casual, etc., and is not particularly limited. For example, the style identification unit 206 may identify multiple styles from combinations including the items worn. Specifically, for example, coordinations for each fashion style may be created in advance in a table, and the style identification unit 206 can identify styles from the table that are similar to the combination of items the customer is wearing. The combination of items may also be a combination of item designs.

[0055] The web analysis unit 202 may also identify posts that match the identified fashion style. The web analysis unit 202 may detect posts from SNS 22 that include the identified style as a tag, for example. For example, the web analysis unit 202 may identify posts from SNS 22 that contain images corresponding to the identified fashion style.

[0056] <Recommendations based on level of attention> To promote trendy fashion coordinates, it is acceptable to recommend items that are the same as or similar to items featured in highly popular posts or items that appear in many posts.

[0057] For example, the Web Analysis Department 202 analyzes SNS22 to identify items used together with worn items based on the level of attention given to posts on SNS22. The level of attention given to a post is represented by the number of expressions of empathy or favor towards the post, the number of views the post has received, the number of followers the poster has received, the number of times the post has been saved, and the number of times the post has been screenshotted. Here, expressions of empathy or favor are represented by the number of times a button indicating such expressions of empathy or favor, implemented on SNS22, is pressed. The number of expressions of empathy or favor is represented by the number of times this button has been pressed. Expressions of empathy or favor are also called "likes," and the number of expressions of empathy or favor is also called the number of "likes." For example, the higher the number of "likes," the number of views the post has received, the number of followers the poster has received, the number of times the post has been saved, and the number of times the post has been screenshotted, the higher the level of attention the post is presumed to have received.

[0058] Specifically, for example, the Web Analysis Department 202 analyzes SNS 22 to identify items used together with worn items that are included in posts on SNS 22 that have a certain level of attention or higher. The predetermined level of attention can be predetermined. For example, let's take the case where the level of attention is the number of "likes". If the number of "likes" on a post containing an image of worn items exceeds a predetermined number, the Web Analysis Department 202 identifies items used together with worn items in that image. This makes it possible to recommend products that can be used together with the items worn by the customer to create a highly attention-grabbing outfit.

[0059] <Recommendations based on the number of posts> For example, products that, when combined with items worn by customers, create outfits that are frequently posted on social media (SNS22) may be recommended. Products that create trendy fashion coordinates can also be recommended.

[0060] The web analysis unit 202 analyzes SNS 22 to identify items used in conjunction with the worn items based on the number of posts on SNS 22. Specifically, for example, the web analysis unit 202 identifies items used in conjunction with the worn items that have a predetermined number of posts on SNS 22 or more.

[0061] This concludes the explanation of Web Analysis Unit 202.

[0062] Next, the decision unit 203 determines recommended products from the products sold in the store based on the items identified by the web analysis unit 202. Specifically, for example, the decision unit 203 matches the image of the item identified by the web analysis unit 202 with the image of the product included in the product database 2001. The decision unit 203 may then determine, for example, that a product with a matching degree equal to or greater than a predetermined degree is a recommended product. Based on this matching, the recommended product may be the same product as the item identified by the web analysis unit 202, or it may be a product similar to the identified item. For this reason, for example, the decision unit 103 determines at least one of the products that are the same as the identified item and a product similar to the identified item from the products sold in the store as recommended products. Not limited to image matching, the decision unit 203 may also determine the same or similar product as the identified item based on its name or design as a recommended product.

[0063] Furthermore, if the web analysis unit 202 identifies multiple items, the decision unit 203 may appropriately narrow down the items to recommend from among the multiple items based on factors such as price and release date.

[0064] Next, the output control unit 204 recommends recommended products to the customer. As a specific recommendation method, as described in Embodiment 1, the output control unit 204 may recommend recommended products to the customer by outputting information about recommended products to an output device. The output control unit 204 may also recommend recommended products to the customer by notifying the terminal device 21 of the information about recommended products. As a notification method, the output control unit 204 may notify the customer of the information about recommended products using email, electronic messages, etc. Note that the recipients of emails, electronic messages, etc., may be pre-registered for each customer in a customer database or the like.

[0065] Furthermore, the output control unit 204 may output information on recommended products and posts related to the recommended products. Posts related to recommended products are posts about items identified by the web analysis unit 202.

[0066] Figure 6 is an explanatory diagram illustrating an example of displaying recommended product information. In Figure 6, the output control unit 204 causes the terminal device 21 to display recommended product information and posts related to the recommended product. In Figure 6, the screen displays recommended product information such as the image of the recommended product, the price of the recommended product, and the name of the recommended product. Specifically, in Figure 6, the screen displays the name of the recommended product, "striped shirt," the price, "2500 yen," and an image. The screen also displays posts related to the recommended product and the number of "likes" on those posts. Specifically, in Figure 6, the screen displays an image included in a post with "10,000 likes" related to the recommended product. The image included in the post includes a striped shirt that is the same as or similar to the recommended product. The image also includes a white skirt worn by the customer. This makes it possible to recommend products that match the items the customer is wearing when they visit the store. For example, it is possible to recommend products that, when used with the items the customer is wearing when they visit the store, will create a trendy coordinated look.

[0067] <Example of recommending multiple products> Multiple recommended products may be determined. In such cases, the output control unit 204 recommends the multiple recommended products in a predetermined order. The predetermined order is not particularly limited and may be based on popularity, the number of posts, price, inventory, or a combination thereof. The predetermined order may be switchable by user operations such as those of customers or store staff.

[0068] The predetermined order will be explained using the example of an order based on attention level. An order based on attention level can be in order of highest attention level, or in order of lowest attention level, etc. For example, the output control unit 204 recommends multiple recommended products in order of highest attention level. Attention level is as described above. For example, if attention level is represented by the number of "likes", the output control unit 204 recommends multiple products in order of the number of "likes". In this way, when recommending multiple recommended products, arranging them in order of the attention level of posts related to the recommended products makes it easier to check trends on SNS22.

[0069] Furthermore, the predetermined order will be explained using the example of an order based on the number of posts. An order based on the number of posts may be in descending order of the number of posts, or descending order of the number of posts, etc. For example, the output control unit 204 may recommend multiple recommended products in descending order of the number of posts. In this way, when recommending multiple recommended products, arranging them in descending order of the number of posts related to the recommended products makes it easier for users such as customers and store clerks to check trends on SNS22.

[0070] Furthermore, the predetermined order will be explained using the example of an order based on price. An order based on price can be in descending order of price, descending order of price, etc. For example, the output control unit 204 recommends multiple recommended products in descending order of price based on the product database 2001. Or, for example, the output control unit 204 recommends multiple recommended products in descending order of price based on the product database 2001.

[0071] Furthermore, the predetermined order will be explained using an example where the order is based on the number of items in stock. The order based on the number of items in stock may be in descending order of quantity, or descending order of quantity. For example, the output control unit 204 recommends multiple recommended products in descending order of quantity based on the product database 2001. Alternatively, for example, the output control unit 204 recommends multiple recommended products in descending order of quantity based on the product database 2001.

[0072] Here, using Figure 7, we will explain an example in which the output control unit 204 recommends multiple products in order of the number of "likes".

[0073] Figure 7 is an explanatory diagram illustrating an example of displaying information for each of multiple recommended products. In Figure 7, the output control unit 204 displays information for each of the multiple recommended products on the terminal device 21 in order of the number of "likes" received.

[0074] In Figure 7, the screen displays information on recommended products in descending order of the number of "likes," and is scrollable. For example, in Figure 7, the screen displays information for each recommended product, including the image of the recommended product, the price of the recommended product, and the name of the recommended product. The screen also displays posts related to the recommended product and the number of "likes" on those posts, for each recommended product. The screen also displays images included in posts related to the recommended product. Each image includes the recommended product or similar items, as well as items worn by the customer. In Figure 7, the screen displays information on the recommended product, including the name of the recommended product, "striped shirt," the price, "2500 yen," and an image. The screen also displays an image included in a post with "10,000 likes" related to the recommended product. The image included in this post includes a striped shirt that is the same as or similar to the recommended product. The image also includes a white skirt worn by the customer. Next, in Figure 7, the screen displays information on the recommended product, including the name of the recommended product, "checked shirt," the price, "3000 yen," and an image. The screen displays images from a post with 4,000 likes related to the recommended product. These images include a checkered shirt that is the same as or similar to the recommended product. Additionally, the image includes a white skirt worn by the customer. As mentioned earlier, the screen is scrollable; for example, scrolling will display the next recommended product in order of the number of likes.

[0075] (flowchart) Figure 8 is a flowchart illustrating an example of the operation of the product recommendation system 20 according to an embodiment. The item detection unit 201 detects the items worn by a customer who has come into the store (step S201). The style identification unit 206 then identifies the style of the customer's fashion (step S202). The behavioral analysis unit 205 then detects the classification of products that the customer is interested in (step S203).

[0076] The web analysis unit 202 identifies items used together with the detected worn items by analyzing SNS 22 (step S204). In step S204, the web analysis unit 202 identifies items of the detected classification that are used together with the worn items from posts corresponding to the identified lineage.

[0077] Then, the decision unit 203 determines a recommended product from the products sold in the store based on the identified item (step S205). In step S205, the decision unit 203 determines a recommended product from the products sold in the store based on the product DB 2001, which is the same product as the identified item or a product similar to the identified item.

[0078] Then, the output control unit 204 recommends a product (step S206). The product recommendation system 20 then terminates the process.

[0079] Furthermore, in step S203, the behavioral analysis unit 205 may detect the design of a product that the customer is interested in, as described above. In such a case, in step S204, the web analysis unit 202 may identify from posts corresponding to the identified lineage an item that is used together with the worn item and has the detected design.

[0080] Furthermore, the order in which steps S201 to S203 are processed is not particularly limited.

[0081] As described above, in Embodiment 2, the product recommendation system 20 detects items worn by customers visiting the store and identifies items that are used together with the detected items by analyzing the web. Then, the product recommendation system 20 determines recommended products from the products sold in the store based on the identified items and recommends the determined recommended products to the customer. This makes it easier for customers to find products that are suitable for them.

[0082] Furthermore, the product recommendation system 20 detects the categories of products that customers are interested in in stores and identifies items of the detected categories that are used together with the items being worn. This allows the product recommendation system 20 to recommend products of categories that customers are likely to purchase.

[0083] The product recommendation system 20 detects the design of the product a customer is looking for in a store and, by analyzing the web, identifies items that are used together with the item being worn and that have the detected design. This allows the product recommendation system 20 to recommend products with designs that the customer is likely to purchase.

[0084] Furthermore, the product recommendation system 20 identifies the customer's fashion style based on the items worn, and then identifies items that are used together with the worn items based on the identified style. This allows the product recommendation system 20 to recommend products that match the customer's fashion preferences.

[0085] Furthermore, the product recommendation system 20 identifies items that are used together with the detected items based on their popularity on the web. This allows the product recommendation system 20 to recommend items that can be combined with the worn items in outfits that are popular on the web.

[0086] Furthermore, the product recommendation system 20 identifies items that are used together with the worn items based on the number of posts on the web. This allows the product recommendation system 20 to recommend items that can be used to create trendy outfits with the worn items on the web.

[0087] Furthermore, if multiple recommended products are determined, the product recommendation system 20 recommends the products in an order based on at least one of the following: their popularity on the web and the number of posts about them on the web. For example, the product recommendation system 20 can recommend multiple products in order of the popularity of posts related to the recommended products, or in order of the number of posts related to the recommended products. This makes it easier for users such as customers and store staff to check trends on the web.

[0088] This concludes the description of Embodiment 2. Each embodiment may be used in a modified form. Modifications are shown below.

[0089] <Example 1> Embodiment 2 describes an example in which the behavioral analysis unit 205 detects products of interest to a customer by performing behavioral analysis using captured images. For example, the product recommendation system 20 may receive information about products of interest to a customer by inputting that information into the terminal device 21. For example, a reception unit (not shown) may receive information about products of interest to a customer through the operation of a store employee or the customer via the terminal device 21. Here, the product information may include product identification information, product name, product classification, product design, etc. The web analysis unit 202 may then identify items that correspond to the classification or design identified by the received product information by analyzing the web.

[0090] <Modification 2> In Embodiment 2, an example was described in which the system identification unit 206 identifies the customer's fashion system based on the items worn. For example, the system identification unit 206 may identify the customer's fashion system from the customer's purchase history.

[0091] This concludes the explanation of the modifications. Each embodiment and each modification may be used in combination as appropriate. In addition, although each embodiment uses SNS as an example of an internet service, other internet services such as blog services may be used.

[0092] Furthermore, each embodiment is not limited to the examples described above and can be modified in various ways. Also, the configuration of the product recommendation system in each embodiment is not particularly limited. For example, the product recommendation systems 10 and 20 may be implemented by a single device (e.g., a product recommendation device), such as a terminal device 21 or a server. Alternatively, the product recommendation systems 10 and 20 in each embodiment may be implemented by different devices for different functions or data. For example, the product recommendation systems 10 and 20 may be composed of multiple servers.

[0093] In each embodiment, each piece of information and each database may include a portion of the aforementioned information. Furthermore, each piece of information and each database may include information other than that described above. Each piece of information and each database may be further divided into multiple databases or multiple pieces of information. Thus, the method of implementing each piece of information and each database is not particularly limited.

[0094] Furthermore, each screen is merely an example and is not particularly limited. Buttons, lists, checkboxes, information display fields, input fields, etc., not shown in the illustrations may be added to each screen. Also, the background color of the screen may be changed.

[0095] Furthermore, for example, in each embodiment, the process of generating screen information to be displayed on the terminal device 21 may be performed by the output control units 104 and 204 of the product recommendation systems 10 and 20. Alternatively, this process may be performed by the terminal device 21. Similarly, when a digital signage is used as the output device, the process of generating screen information to be displayed on the digital signage may be performed by the output control units 104 and 204 of the product recommendation systems 10 and 20. Alternatively, this process may be performed by the digital signage.

[0096] (Example of computer hardware configuration) Next, we will describe hardware configuration examples when each device, such as the product recommendation systems 10 and 20 and the terminal device 21 described in each embodiment, is implemented using a computer. Figure 9 is an explanatory diagram showing an example of computer hardware configuration. For example, some or all of each device can be implemented using any combination of computer 80 and program, as shown in Figure 9.

[0097] Computer 80 includes, for example, a processor 801, a ROM (Read Only Memory) 802, a RAM (Random Access Memory) 803, and a storage device 804. Computer 80 also includes a communication interface 805 and an input / output interface 806. Each component is connected, for example, via a bus 807. The number of components is not particularly limited, and each component may be one or more.

[0098] The processor 801 controls the entire computer 80. The processor 801 may include, for example, a CPU (Central Processing Unit), a DSP (Digital Signal Processor), or a GPU (Graphics Processing Unit). The computer 80 has memory units such as ROM 802, RAM 803, and a storage device 804. The storage device 804 may include, for example, semiconductor memory such as flash memory, an HDD (Hard Disk Drive), or an SSD (Solid State Drive). For example, the storage device 804 stores OS (Operating System) programs, application programs, and programs related to each embodiment. Alternatively, ROM 802 stores application programs and programs related to each embodiment. The RAM 803 is used as the work area for the processor 801.

[0099] The processor 801 also loads programs stored in the memory device 804, ROM 802, etc. Then, the processor 801 executes each process coded in the program. The processor 801 may also download various programs via the communication network NT. Furthermore, the processor 801 functions as part or all of the computer 80. The processor 801 may also execute processes or instructions in the illustrated flowchart based on the program.

[0100] The communication interface 805 is connected to a communication network NT, such as a LAN (Local Area Network) or WAN (Wide Area Network), via a wireless or wired communication line. The communication network NT may be composed of multiple communication networks NT. This allows the computer 80 to connect to external devices and external computers 80 via the communication network NT. The communication interface 805 manages the interface between the communication network NT and the internal workings of the computer 80. Furthermore, the communication interface 805 controls the input and output of data from external devices and external computers 80.

[0101] Furthermore, the input / output interface 806 is connected to at least one of the input device, output device, and input / output device. The connection method may be wireless or wired. Examples of input devices include keyboards, mice, and microphones. Examples of output devices include display devices, lighting devices, and speakers that output sound. Examples of input / output devices include touch panel displays. The input device, output device, and input / output device may be built into the computer 80 or may be external.

[0102] The hardware configuration of computer 80 is an example. Computer 80 may have some of the components shown in Figure 9. Computer 80 may have components other than those shown in Figure 9. For example, computer 80 may have a drive device. The processor 801 may read programs and data stored on a recording medium attached to the drive device into RAM 803. Examples of non-temporary tangible recording media include optical discs, flexible discs, magneto-optical discs, and USB (Universal Serial Bus) memory. Also, as mentioned above, computer 80 may have input devices such as a keyboard and a mouse. Computer 80 may have output devices such as a display. Furthermore, computer 80 may have input devices, output devices, and input / output devices, respectively.

[0103] Furthermore, the computer 80 may have various sensors (not shown). The type of sensor is not particularly limited. Also, the computer 80 may be equipped with an imaging device capable of capturing images or videos.

[0104] This concludes the description of the hardware configuration of each device. Furthermore, there are various variations in how each device can be implemented. For example, each device may be implemented by any combination of different computers and programs for each component. Alternatively, the multiple components of each device may be implemented by any combination of a single computer and program.

[0105] Furthermore, some or all of the components of each device may be implemented by application-specific circuits. Alternatively, some or all of the components of each device may be implemented by general-purpose circuits, including processors such as FPGAs (Field Programmable Gate Arrays). Furthermore, some or all of the components of each device may be implemented by a combination of application-specific circuits and general-purpose circuits. These circuits may also be a single integrated circuit, or they may be divided into multiple integrated circuits. These multiple integrated circuits may be connected via a bus or the like.

[0106] Furthermore, if some or all of the components of each device are implemented by multiple computers or circuits, these computers or circuits may be centrally located or distributed.

[0107] The product recommendation method described in each embodiment is implemented by a product recommendation system. Alternatively, the product recommendation method can be implemented by a computer, such as a server or terminal device, executing a pre-prepared program. The program described in each embodiment is recorded on a computer-readable recording medium such as an HDD, SSD, flexible disk, optical disk, magneto-optical disk, or USB memory. The program is then executed by being read from the recording medium by a computer. The program may also be distributed via a communication network NT.

[0108] Each component of the product recommendation system in the embodiments described above may be implemented in hardware, such as a computer. Alternatively, each component may be implemented in a computer or firmware based on program control.

[0109] The present disclosure has been described above with reference to the embodiments described herein, but the present disclosure is not limited to the embodiments described above. The structure and details of each present disclosure may include embodiments that apply various modifications that can be grasped by those skilled in the art within the scope of the present disclosure. The present disclosure may include embodiments that combine or substitute the matters described herein as appropriate. For example, matters described using a particular embodiment may be applied to other embodiments to the extent that they do not cause a contradiction. For example, although multiple operations are described sequentially in the form of a flowchart, the order in which they are described does not limit the order in which the multiple operations are performed. Therefore, when implementing each embodiment, the order of the multiple operations may be changed to the extent that it does not impair the content.

[0110] Some or all of the above embodiments may also be described as follows. However, some or all of the above embodiments are not limited to the following.

[0111] (Note 1) An item detection means for detecting items worn by customers who visit a store, A web analysis means that identifies articles used together with the detected article by analyzing the web, A decision-making means for determining recommended products from the products sold at the store based on the identified articles, Output control means for recommending the determined recommended product to the customer, A product recommendation system equipped with the following features. (Note 2) The aforementioned website, which is the subject of the analysis, is a service that allows images to be posted. The web analysis means identifies articles used together with the detected article by analyzing images on the web. Product recommendation system as described in Appendix 1. (Note 3) The determination means determines, from among the products sold at the store, at least one of the products that is the same as the specified article and one that is similar to the specified article, as the recommended product. Product recommendation system as described in Appendix 1 or 2. (Note 4) A behavioral analysis means for detecting the classification of products that the customer is interested in at the store. Equipped with, The web analysis means, by analyzing the web, identifies the items of the detected classification among the items used together with the detected item. The product recommendation system described in any of the appendices 1 to 3. (Note 5) The behavioral analysis means detects at least one of the products the customer has picked up and the products the customer is looking at as products the customer is interested in, and detects the classification of the detected products. Product recommendation system as described in Appendix 4. (Note 6) A behavioral analysis means for detecting the design of a product that the customer is interested in at the store. Equipped with, The web analysis means, by analyzing the web, identifies articles of the detected design that are used together with the detected article. The product recommendation system described in any of the appendices 1 to 3. (Note 7) The behavioral analysis means detects at least one of the products the customer has picked up and the products the customer is looking at as products the customer is interested in, and detects the design of the detected products. Product recommendation system as described in Appendix 6. (Note 8) A system identification means for identifying the fashion style of the customer based on the detected article, Equipped with, The web analysis means, by analyzing the web, identifies articles used together with the detected articles based on the identified system. Product recommendation system as described in any of the appendices 1 through 7. (Note 9) The web analysis means analyzes the web and, based on the level of attention on the web, identifies articles used together with the detected article. The product recommendation system described in any of the appendices 1 through 8. (Note 10) The web analysis means analyzes the web to identify articles that are used together with the detected article and are included in posts that have a level of attention on the web that is above a predetermined level of attention. Product recommendation system as described in Appendix 9. (Note 11) The web analysis means analyzes the web and, based on the number of posts on the web, identifies articles used together with the detected articles. The product recommendation system described in any of the appendices 1 through 10. (Note 12) The web analysis means, by analyzing the web, identifies articles used together with the detected article, which have a number of posts equal to or greater than a predetermined number. Product recommendation system as described in Appendix 11. (Note 13) When the determination means determines a plurality of recommended products, the output control means recommends the plurality of recommended products in an order based on at least one of the popularity on the web and the number of posts on the web. The product recommendation system described in any of the appendices 1 through 10. (Note 14) The system detects items worn by customers who enter the store. By analyzing the web, we can identify the articles used together with the detected articles. Based on the identified items, recommended products are determined from the products sold at the store. The determined recommended products are recommended to the customer. Recommended method. (Note 15) On the computer, The system detects items worn by customers who enter the store. By analyzing the web, we can identify the articles used together with the detected articles. Based on the identified items, recommended products are determined from the products sold at the store. The determined recommended products are recommended to the customer. A program that executes a process. (Note 16) On the computer, The system detects items worn by customers who enter the store. By analyzing the web, we can identify the articles used together with the detected articles. Based on the identified items, recommended products are determined from the products sold at the store. The determined recommended products are recommended to the customer. A non-temporary recording medium readable by the computer, which records a program that executes a process. [Explanation of Symbols]

[0112] 10,20 Product Recommendation System 101,201 Item detection unit 102,202 Web Analysis Department 103,203 Decision Section 104,204 Output control unit 205 Behavior Analysis Department 206 System identification part 2001 Product DB 21 Terminal device 23 Imaging device 80 Computers 801 Processor 802 ROM 803 RAM 804 Storage device 805 Communication Interface 806 Input / Output Interface 807 Bus NT communication network

Claims

1. An item detection means for detecting items worn by customers who visit a store, A means for identifying the fashion style of the customer based on the detected article, A web analysis means that identifies articles used together with the detected article by analyzing the web, A decision-making means for determining recommended products from the products sold at the store based on the identified articles, Output control means for recommending the determined recommended product to the customer, Equipped with, The web analysis means, by analyzing the web, identifies articles used together with the detected articles based on the identified system. Product recommendation system.

2. An item detection means for detecting items worn by customers who visit a store, A web analysis means that identifies articles used together with the detected article by analyzing the web, A decision-making means for determining recommended products from the products sold at the store based on the identified articles, Output control means for recommending the determined recommended product to the customer, Equipped with, The web analysis means analyzes the web and, based on the level of attention on the web, identifies articles used together with the detected article. Product recommendation system.

3. An item detection means for detecting items worn by customers who visit a store, A web analysis means that identifies articles used together with the detected article by analyzing the web, A decision-making means for determining recommended products from the products sold at the store based on the identified articles, Output control means for recommending the determined recommended product to the customer, Equipped with, The web analysis means analyzes the web and, based on the number of posts on the web, identifies articles used together with the detected articles. Product recommendation system.

4. An item detection means for detecting items worn by customers who visit a store, A web analysis means that identifies articles used together with the detected article by analyzing the web, A decision-making means for determining recommended products from the products sold at the store based on the identified articles, Output control means for recommending the determined recommended product to the customer, Equipped with, When the determination means determines a plurality of recommended products, the output control means recommends the plurality of recommended products in an order based on at least one of the popularity on the web and the number of posts on the web. Product recommendation system.

5. The aforementioned website, which is the subject of the analysis, is a service that allows images to be posted. The web analysis means identifies articles used together with the detected article by analyzing images on the web. The product recommendation system according to claim 1.

6. The determination means determines, from among the products sold at the store, that at least one of the products identical to the identified item and products similar to the identified item be the recommended product. The product recommendation system according to claim 1.

7. A behavioral analysis means for detecting the classification of products that the customer is interested in at the store. Equipped with, The web analysis means, by analyzing the web, identifies the items of the detected classification among the items used together with the detected items. The product recommendation system according to claim 1.

8. The behavioral analysis means detects at least one of the products the customer has picked up and the products the customer is looking at as products the customer is interested in, and detects the classification of the detected products. The product recommendation system according to claim 7.

9. A behavioral analysis means for detecting the design of a product that the customer is interested in at the store. Equipped with, The web analysis means, by analyzing the web, identifies articles of the detected design that are used together with the detected article. The product recommendation system according to claim 1.

10. The behavioral analysis means detects at least one of the products the customer has picked up and the products the customer is looking at as products the customer is interested in, and detects the design of the detected products. The product recommendation system according to claim 9.

11. The web analysis means analyzes the web to identify articles that are used together with the detected article and are included in posts that have a level of attention on the web that is above a predetermined level of attention. The product recommendation system according to claim 2.

12. The web analysis means, by analyzing the web, identifies articles used together with the detected article, which have a number of posts equal to or greater than a predetermined number. The product recommendation system according to claim 3.

13. A computer, The system detects items worn by customers who enter the store. Based on the detected items, the customer's fashion style is identified. By analyzing the web, we can identify the articles used together with the detected articles. Based on the identified items, recommended products are determined from the products sold at the store. The determined recommended products are recommended to the customer. Execute the process, In the analysis process described above, by analyzing the web, based on the identified system, the articles used together with the detected articles are identified. Product recommendation method.

14. On the computer, The system detects items worn by customers who enter the store. Based on the detected items, the customer's fashion style is identified. By analyzing the web, we can identify the articles used together with the detected articles. Based on the identified items, recommended products are determined from the products sold at the store. The determined recommended products are recommended to the customer. Execute the process, In the analysis process described above, by analyzing the web, based on the identified system, the articles used together with the detected articles are identified. program.

Images