Information processing apparatus, information processing method, and information processing program

Abstract

To properly determine a sales page of a commodity to be recommended to a user.SOLUTION: An information processing apparatus acquires a list of sales page identification information identifying each sales page for one or more commodities associated with each user, converts the list of sales page identification information into a list of commodity identification information identifying a commodity, generates a recommended commodity list including recommended commodity identification information identifying each of one or more commodities to be recommended to the user, based on the list of commodity identification information, determines a sales page for each of the one or more commodities to be recommended to the user, based on one or more pieces of recommended commodity identification information included in the recommended commodity list.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a technique for determining a sales page for a product to be recommended to a user. [Background technology]

[0002] In recent years, electronic commerce (EC), which involves selling products online using the Internet, has become increasingly common. This type of EC is conducted, for example, through EC sites such as EC malls (mall-type EC sites), where multiple stores operate online shopping malls. By accessing EC sites from mobile devices such as personal computers (PCs) or smartphones, users can browse and purchase desired products without having to physically visit multiple stores or worrying about time.

[0003] For the purpose of sales promotion on an e-commerce site, a technology has been proposed that identifies products that a user is interested in from the user's action history on product sales pages, such as purchase history and browsing history, and determines products to recommend to the user based on the identified products. For example, Patent Document 1 discloses that a second product that has a predetermined relationship with a first product identified based on the user's purchase history is determined as a recommended product. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2022-172545 Summary of the Invention [Problem to be solved by the invention]

[0005] When determining recommended products for a user based on the user's action history on a sales page, it is important to appropriately associate action information (action information) with product identification information (product identification information) in order to identify products that interest the user. In an e-commerce mall, when the same product is sold at multiple stores, the same product may be displayed and sold on different sales pages for each store. Even if the same product is sold on different sales pages, there is a possibility that the action information may not be appropriately associated with the product identification information. In other words, the action information on the different sales pages by the user may not be associated with the same product identification information, but with different product identification information, which may result in the action information not being appropriately associated with the product identification information. As a result, there is a risk that the sales page of the product to be recommended to the user may not be appropriately determined.

[0006] The present invention has been made in view of the above-mentioned problems, and aims to provide a technique for appropriately determining a sales page for a product to be recommended to a user. [Means for solving the problem]

[0007] In order to solve the above problem, one aspect of an information processing device according to the present invention comprises an acquisition unit that acquires a list of sales page identification information that identifies each of the sales pages of one or more products associated with each user; a conversion unit that converts the list of sales page identification information into a list of product identification information that identifies the products; a generation unit that generates a recommended product list based on the list of product identification information, the recommended product identification information including recommended product identification information that identifies each of the one or more products recommended to each user; and a determination unit that determines the sales page for each of the one or more products recommended to each user based on one or more of the recommended product identification information included in the recommended product list.

[0008] In order to solve the above problem, one aspect of the information processing method according to the present invention includes an acquisition step of acquiring a list of sales page identification information that identifies each of the sales pages of one or more products associated with each user; a conversion step of converting the list of sales page identification information into a list of product identification information that identifies the products; a generation step of generating a recommended product list based on the list of product identification information, the recommended product list including recommended product identification information that identifies each of the one or more products to be recommended to each user; and a determination step of determining a sales page for each of the one or more products to be recommended to each user based on one or more of the recommended product identification information included in the recommended product list.

[0009] In order to solve the above problem, one aspect of the information processing program of the present invention causes a computer to execute an acquisition process for acquiring a list of sales page identification information that identifies each of the sales pages of one or more products associated with each user; a conversion process for converting the list of sales page identification information into a list of product identification information that identifies the products; a generation process for generating a recommended product list based on the list of product identification information, the recommended product identification information including recommended product identification information that identifies each of the one or more products recommended to each user; and a determination process for determining the sales page for each of the one or more products recommended to each user based on one or more of the recommended product identification information included in the recommended product list. [Effects of the Invention]

[0010] According to the present invention, it is possible to appropriately determine the sales page of the product to be recommended to the user. The above-mentioned objects, aspects, and advantages of the present invention, as well as other objects, aspects, and advantages of the present invention not described above, will be understood by those skilled in the art from the following detailed description of the invention by referring to the accompanying drawings and the claims. [Brief explanation of the drawings]

[0011] [Figure 1] FIG. 1 shows an example of the configuration of an information processing system according to an embodiment. [Figure 2]FIG. 2 shows an example of the functional configuration of the information processing device according to the first embodiment. [Figure 3] Figure 3 shows an example of a list of sales page IDs in chronological order. [Figure 4] FIG. 4 shows an example of the hardware configuration of an information processing device according to an embodiment. [Figure 5] FIG. 5 is a flowchart of a process executed by the information processing device according to the embodiment. [Figure 6] FIG. 6 shows an example of a list of sales page IDs for each user. [Figure 7] FIG. 7 shows an example of a list of product IDs for each user. [Figure 8] FIG. 8 shows an example of a list of recommended product IDs for each user. [Figure 9] FIG. 9 shows an example of a list of recommended sales page IDs for each user. [Figure 10] FIG. 10 shows an example of the functional configuration of an information processing device according to the second embodiment. [Figure 11] FIG. 11 is a flowchart of the genre determination process according to the second embodiment. [Figure 12] FIG. 12 is a conceptual diagram showing the flow of data for determining the top four genre IDs. [Figure 13A] FIG. 13A shows an example of a screen including a sales page for selling multiple products in an EC mall. [Figure 13B] FIG. 13B shows an example of a screen including a sales page for selling products in an EC mall. [Figure 13C] FIG. 13C shows another example screen including a sales page for selling multiple products in an e-commerce mall. [Figure 14] FIG. 14 shows an example of the functional configuration of an information processing device according to the fourth embodiment. [Figure 15] FIG. 15 shows another example of a list of product IDs for each user. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, with reference to the accompanying drawings, an embodiment for carrying out the present invention will be described in detail. Among the components disclosed below, those having the same function will be given the same reference numerals, and their description will be omitted. Note that the embodiment disclosed below is an example of a means for realizing the present invention, and should be appropriately modified or changed depending on the configuration of the device to which the present invention is applied and various conditions. The present invention is not limited to the following embodiment. Furthermore, not all of the combinations of features described in the present embodiment are necessarily essential to the solution of the present invention.

[0013] First Embodiment [Configuration of information processing system] FIG. 1 shows an example configuration of an information processing system 1 according to this embodiment. The information processing system 1 includes an information processing device 10, an EC site server 11, and a user device 12. The information processing device 10, the EC site server 11, and the user device 12 are configured to be able to communicate with each other via a network 13. The network 13 may include the Internet, an intranet, a local area network (LAN), a wide area network (WAN), a mobile communication network, and the like. Although FIG. 1 illustrates one user device 12, the information processing system 1 may include multiple user devices having similar functions to the user device 12. In the present disclosure, the multiple user devices are collectively referred to as the user device 12. The user device 12 is operated by a user 14. In the present disclosure, the terms "user device" and "user" may be understood to be synonymous.

[0014] The EC site server 11 is a server device that operates an EC mall (mall-type EC site) where multiple stores operate an online shopping mall. Specifically, the EC site server 11 operates an EC mall that operates a shopping mall via sales pages (web pages) of products sold by multiple stores (merchants). The EC site server 11 can accept access from user devices 12 via a network 13 and provide various services related to shopping at the EC mall to users 14. For example, when a user accesses the EC mall and performs an action such as purchasing or viewing on the sales page of a product, the EC site server 11 provides services related to the product to the user 14. The EC site server 11 can also acquire (collect) and manage information related to the sales page of the product on which the user 14 performed an action in the EC mall. Note that the EC site server 11 is not limited to a server device and may be realized by a mainframe or the like.

[0015] The user device 12 is operated by a user 14 to access the EC site server 11 and receive various services at the EC mall provided by the EC site server 11. For example, the user 14 can operate the user device 12 to access the EC mall, browse sales pages for various products offered at the EC mall, and make purchases. When using services at the EC mall, the user 14 registers information about the user 14 (hereinafter also referred to as user attributes). For example, the user 14 sets a user ID (user identification information) that identifies the user 14 and is linked to the user attributes of the user 14, logs in using the user ID, and uses services at the EC mall. By setting the user ID, the user 14 can use the services of the EC mall even from a user device other than the user device 12 connected to the network 13.

[0016] User attributes include the user's actual attributes. The user's actual attributes include the IP address of the user's device, the user's address, the user's name, the user's credit card number, and the user's demographic information (such as gender, age, residential area, occupation, and family composition). The user's actual attributes may also include the user's registration number and registered name when using web services, including e-commerce malls (web services related to e-commerce malls). The user's actual attributes may also include information about the user's usage history of web services, including e-commerce malls, search history, product purchase history (including purchase results), and points that can be earned by using services. Thus, the user's actual attributes may include any information, including information related to the user's device or the user himself / herself, and information about the user's use of web services, including e-commerce malls. The user attributes may also include estimated attributes of the user. The estimated attributes of the user may be estimated based on the user's actual attributes, for example, by a trained user attribute estimation model. The estimated attributes of the user may include preferences for products of interest and lifestyle.

[0017] The user device 12 is, for example, a device such as a smartphone or a tablet, and is configured to be able to communicate with the EC site server 11 and the information processing device 10 via the network 13. The user device 12 has a display unit (display surface) such as a liquid crystal display, and a user 14 can perform various operations using a GUI (Graphical User Interface) provided on the display unit. These operations include various operations on content such as images displayed on the screen, such as tapping, sliding, and scrolling using a finger or a stylus. The user device 12 may also have a separate display unit.

[0018] The information processing device 10 acquires information about product sales pages collected by the EC site server 11 and, based on the information, determines a sales page for a product to be recommended to the user 14. Specifically, the information processing device 10 first acquires a list of sales page IDs that identify sales pages for products associated with the user 14 and converts the list of sales page IDs into a list of product IDs that identify the products. Next, the information processing device 10 generates a recommended product list including recommended product IDs that identify products to be recommended to the user 14 based on the list of product IDs, and determines a sales page for the product to be recommended to the user 14 based on the recommended product IDs. Information about the determined product sales pages is provided to the EC site server 11, and the EC site server 11 can provide the sales page to the user device 12 based on the information. Note that, although the information processing device 10 and the EC site server 11 are configured as separate devices in FIG. 1, the information processing device 10 may be configured to include the functions of the EC site server 11.

[0019] FIG. 13A shows an example of a screen 1300 including multiple sales pages in an EC mall provided by the EC site server 11. Each of the multiple sales pages is managed and operated by a different store (merchant). Therefore, each of the multiple sales pages is associated with the store of the product being sold on the sales page. In screen 1300, each sales page includes an image of the product being sold and information about the product (description, price, etc.). Taking sales page 1302 as an example, sales page 1302 is a sales page selling product 1301. Product 1301 is specifically an image of the product being sold on sales page 1302, but in the present disclosure, this image will be referred to as the product. Furthermore, the image on the sales page may be any image that explains the product being sold.

[0020] [Functional configuration of information processing device] 2 shows an example of the functional configuration of the information processing device 10 according to this embodiment. As an example of its functional configuration, the information processing device 10 has a sales page ID (sales page identification information) acquisition unit 201, a product ID (product identification information) list generation unit 202, a recommended product list generation unit 203, a sales page determination unit 204, a table storage unit 210, and a learning model storage unit 220.

[0021] The table storage unit 210 is configured to be able to store a product conversion table 211 for converting sales page IDs into product IDs. The product conversion table 211 is a table that associates product IDs, which identify products sold by the EC mall, with sales page IDs, which identify one or more sales pages that sell the products. The same product ID is assigned to the same product. Since the same product may be sold on different sales pages, the product conversion table 211 may associate different (multiple) sales page IDs with one product ID. In this embodiment, it is assumed that multiple product IDs are not associated with one sales page ID.

[0022] The learning model storage unit 220 is configured to store a trained feature vector extraction model 221. The feature vector extraction model 221 is a machine learning model that extracts (derives) a vector representation of a user having a user ID based on past actions associated with the user ID. Specifically, the feature vector extraction model 221 is a machine learning model configured to input a list of product IDs of products for which the user having the user ID has previously taken an action (e.g., information indicating the list) and output a vector representation of the user having the user ID. The vector representation corresponds to a vectorized representation of the product IDs in which the user has shown interest. The feature vector extraction model 221 is configured based on MF (Matrix Factorization) or natural language processing (NLP). An MF-based model is a model that decomposes an evaluation value matrix consisting of m (number of rows) × n (number of columns) (m and n are integers equal to or greater than 2) into two matrices by performing dimensionality reduction using collaborative filtering. An NLP-based model is, for example, a natural language processing model such as Word2Vec that converts words into vector representations. The feature vector extraction model 221 is re-learned as needed by a processing unit (not shown) in the information processing device 10 or an external device.

[0023] The information processing device 10 may not be entirely provided in one device, but may be provided in separate devices. For example, part of the information processing device 10 may be provided in an external server device such as an EC site server 11. In this case, the functions described in this embodiment are realized by cooperation between the information processing device 10 and the external server device.

[0024] 2 also shows an example of the functional configuration of the EC site server 11 related to this embodiment. The EC site server 11 is connected to the information processing device 10 via a network 13. In FIG. 2, the EC site server 11 has, as functional configurations related to this embodiment, an EC mall operation unit 231, an evaluation index derivation unit 232, a sales page ID database 240, and an evaluation index database 250.

[0025] The EC mall operation unit 231 operates an EC mall where multiple stores sell products online. For example, the EC mall operation unit 231 can create multiple sales pages based on product information provided by the multiple stores and provide them to the user device 12 connected via the network 13. The EC mall operation unit 231 can also provide the user device 12 with a service corresponding to the type of action on the product sales page from the user device 12. The EC mall operation unit 231 can also provide the user device 12 with a sales page specified by the information processing device 10. When a predetermined action is performed on the product sales page by the user device 12, the EC mall operation unit 231 stores a sales page ID associated with the user ID in the sales page ID database 240. Examples of sales page IDs will be described later with reference to FIG. 3. The evaluation index derivation unit 241 derives evaluation indexes (marketing indexes) associated with sales pages in the EC mall provided by the EC mall operation unit 231. The evaluation index represents an evaluation index for the product sold on the sales page or the store (merchant) selling the product, and may include the CVR (conversion rate) for the sales page. In this embodiment, a conversion in the CVR is assumed to be a purchase of the product, but in other embodiments, it may be a click on the product. The evaluation index may also include user evaluation points for the sales page and rankings of the evaluation points (ranking positions). The evaluation points may include numerical values ​​of user evaluations, comments, and recommendation levels for at least one of the sales page, store, and product.

[0026] The EC mall operation unit 231 adds a sales page ID to the list of sales page IDs stored in the sales page ID database 240 in chronological order each time a user performs a specific action on a product sales page in the EC mall. That is, a list of sales page IDs is generated each time the specific action is performed. Hereinafter, the list of sales page IDs stored in the sales page ID database 240 will be referred to as the "list of chronological sales page IDs." Figure 3 shows an example of a list 30 of chronological sales page IDs stored in the sales page ID database 240. The list 30 of chronological sales page IDs is a list of sales page IDs collected each time three users perform a specific action on a sales page for multiple products in the EC mall. As an example, the list 30 of chronological sales page IDs is configured so that each column contains the following components: user ID 301, sales page ID 302, action type 303, action date and time 304, and identifying genre ID (genre identification information) 305. The genre ID 305 is the genre of the product sold on the sales page identified by the sales page ID 302, and is associated with each product page.

[0027] The user ID 301 is information identifying each of the three users and is one of UID_1, UID_2, and UID_3. Each UID in the user ID 301 is associated with a user attribute. For example, UID_1 is associated with a user attribute registered when the user identified by UID_1 uses the EC mall service. The sales page ID 302 is information identifying a sales page (webpage). For example, the sales page ID 302 is composed of a character string identifying a product sales page. It may be a character string such as that shown in FIG. 3 or a uniform resource locator (URL). The action type 303 indicates the type of a predetermined action on the product sales page. In this embodiment, the predetermined action is one of "view," "favorite registration (add to favorites)," and "purchase." "View" corresponds to the action of accessing (visiting) the sales page and viewing the sales page. "Favorite registration" corresponds to the action of marking and registering a product sold on the sales page online. "Purchase" corresponds to the action of purchasing a product sold on the sales page. If a "purchase" action is performed within a predetermined time after a "view" action, only the "purchase" action may be registered. Action date and time 304 is information indicating the date and time when the action indicated by action type 303 was performed. The list 30 of chronological sales page IDs is configured so that new components are placed at the top of the list according to the date and time. Genre ID 305 is information (genre identification information) that identifies the genre of the product sold on the sales page identified by sales page ID 302. Genre ID 305 is used in the second embodiment.

[0028] With reference to FIG. 13B, the actions of "view," "add to favorites," and "purchase" for a product sold on the sales page will be described. FIG. 13B shows an example of a screen 1310 including a sales page 1303, which is displayed when sales page 1302 is selected (activated) and accessed on screen 1300 in FIG. 13A. Both sales page 1302 in FIG. 13A and sales page 1303 in FIG. 13B are sales pages for selling product 1301. Sales page 1302 and sales page 1303 are examples of sales pages for selling product 1301, and may be configured with other designs or displayed on the screen using different display methods. "View" corresponds to the action of accessing sales page 1303 and viewing information about product 1301. "Add to favorites" corresponds to the action of selecting (activating) favorite button 1304 on screen 1310 to mark and register product 1301 online. "Purchase" corresponds to an action of selecting (activating) the cart button (a button indicating a shopping cart) 1305 on the screen 1310 to purchase the product 1301, and then carrying out the purchase procedure. Note that in this embodiment, the action type is assumed to be one of three actions: "purchase," "view," and "add to favorites," but is not limited to these. For example, the access type may include other access types such as searching or clicking on the product sales page.

[0029] In the information processing device 10, the sales page ID acquisition unit 201 acquires a list of sales page IDs for each user ID from a list of time-series sales page IDs stored in the sales page ID database 240 in the EC site server 11. The product ID list generation unit 202 converts the sales page IDs included in the list of sales page IDs for each user ID acquired by the sales page ID acquisition unit 201 into product IDs, thereby generating a list of product IDs. The recommended product list generation unit 203 determines one or more recommended product IDs for each user ID based on the list of product IDs generated by the product ID list generation unit 202, and generates a list of recommended product IDs. The sales page determination unit 204 determines a sales page for recommended products (hereinafter also referred to as a recommended sales page) for each user ID based on the list of recommended product IDs generated by the recommended product list generation unit 203. Specific processing by the sales page ID acquisition unit 201, product ID list generation unit 202, recommended product list generation unit 203, and sales page determination unit 204 will be described later.

[0030] [Hardware configuration of information processing device] Next, a description will be given of an example of the hardware configuration of the information processing device 10. Fig. 4 is a block diagram showing an example of the hardware configuration of the information processing device 10 according to this embodiment. The information processing device 10 according to this embodiment can be implemented on a single or multiple computers, mobile devices, or any other processing platform. 4, the information processing device 10 is illustrated as being implemented in a single computer, but the information processing device 10 according to this embodiment may be implemented in a computer system including multiple computers. The multiple computers may be connected to each other via a wired or wireless network so as to be able to communicate with each other.

[0031] 4, the information processing device 10 may include a CPU (Central Processing Unit) 401, a ROM (Read Only Memory) 402, a RAM (Random Access Memory) 403, a HDD (Hard Disk Drive) 404, an input unit 405, a display unit 406, a communication I / F (communication unit) (interface) 407, and a system bus 408. The information processing device 10 may also include an external memory. The CPU 401 controls the overall operation of the information processing device 10, and controls each component (402 to 407) via a system bus 408, which is a data transmission path.

[0032] The ROM 402 is a non-volatile memory that stores a control program and the like required for the CPU 401 to execute processing. The program includes instructions (code) for executing the processing according to the above-described embodiment. The program may be stored in a non-volatile memory such as the HDD 404 or an SSD (Solid State Drive), or in an external memory such as a removable storage medium (not shown). The RAM 403 is a volatile memory and functions as the main memory, work area, etc. of the CPU 401. That is, when executing a process, the CPU 401 loads necessary programs, etc. from the ROM 402 into the RAM 403 and executes the programs, etc. to realize various functional operations. The RAM 403 may include the table storage unit 210 and the learning model storage unit 220 shown in FIG. 2.

[0033] The HDD 404 stores, for example, various data and information required when the CPU 401 performs processing using a program. The HDD 404 also stores, for example, various data and information obtained when the CPU 401 performs processing using a program. The input unit 405 is configured with a keyboard and a pointing device such as a mouse. The display unit 406 is configured by a monitor such as a liquid crystal display (LCD), etc. The display unit 406 may be configured in combination with the input unit 405 to function as a GUI (Graphical User Interface).

[0034] The communication I / F 407 is an interface that controls communication between the information processing device 10 and an external device. The communication I / F 407 provides an interface with a network and executes communication with the external device via the network. Various data, parameters, and the like are transmitted and received between the information processing device 10 and the external device via the communication I / F 407. In this embodiment, the communication I / F 407 may execute communication via a wired LAN (Local Area Network) or a dedicated line that conforms to a communication standard such as Ethernet (registered trademark). However, the network that can be used in this embodiment is not limited to this and may be configured as a wireless network. This wireless network includes wireless PANs (Personal Area Networks) such as Bluetooth (registered trademark), ZigBee (registered trademark), and UWB (Ultra Wide Band). It also includes wireless LANs (Local Area Networks) such as Wi-Fi (Wireless Fidelity) (registered trademark) and wireless MANs (Metropolitan Area Networks) such as WiMAX (registered trademark). It also includes wireless WANs (Wide Area Networks) such as 4G and 5G. The network may be any network that connects devices to each other so that they can communicate with each other, and the communication standard, scale, and configuration are not limited to those described above.

[0035] At least some of the functions of each element of the information processing device 10 shown in Fig. 2 can be realized by the CPU 401 executing a program. However, at least some of the functions of each element of the information processing device 10 shown in Fig. 2 may be configured to operate as dedicated hardware. In this case, the dedicated hardware operates under the control of the CPU 401.

[0036] The hardware configuration of the EC site server 11 is also the same as the configuration shown in Fig. 4. Here, the RAM 403 can include the sales page ID database 240 and the evaluation index database 250 shown in Fig. 2.

[0037] [Recommended sales page determination process] Next, a recommended sales page determination process executed by the information processing device 10 according to this embodiment will be specifically described. Fig. 5 is a flowchart of the process executed by the information processing device 10 according to this embodiment. The process shown in Fig. 5 can be performed by the CPU 401 executing a control program stored in the information processing device 10.

[0038] In S51, the sales page ID acquisition unit 201 acquires a list of sales page IDs for each user ID from the list of time-series sales page IDs collected by the EC site server 11 and stored in the sales page ID database 240. The list of sales page IDs for each user ID corresponds to a list of sales page IDs that identify one or more sales pages associated with a user identified by the user ID. The list of sales page IDs for each user ID acquired by the sales page ID acquisition unit 201 is also referred to as a list of sales page IDs for each user ID to distinguish it from the time-series sales page IDs stored in the sales page ID database 240. Figure 6 shows an example of a list 60 of sales page IDs for each user acquired by the sales page ID acquisition unit 201. The list 60 of sales page IDs for each user is generated by sorting the list 30 of sales page IDs shown in Figure 3 by user ID. That is, the list 60 of sales page IDs is configured to include information such as a sales page ID 302, an action type 303, and an action date and time 304 for each user ID 301.

[0039] The list of sales page IDs for each user 60 is a list of sales page IDs from the list of sales page IDs 30 that are included within a predetermined period of time from a predetermined timing. This predetermined timing may be when the sales page ID acquisition unit 201 accesses the EC site server 11 to acquire the list of chronological sales page IDs, or a timing arbitrarily set by the sales page ID acquisition unit 201. Alternatively, the list of sales page IDs for each user may include sales page IDs corresponding to a predetermined number of actions for each user. In other words, the sales page ID acquisition unit 201 may generate and acquire a list of sales page IDs for each user so that it includes a predetermined number of sales page IDs corresponding to the predetermined number of actions for each user. In this case, the same number of sales page IDs are listed for each user ID. Note that the list of sales page IDs for each user 60 is sorted by user ID, but the list of sales page IDs for each user may also be generated without sorting.

[0040] Next, in S52, the product ID list generation unit 202 converts the sales page IDs included in the sales page ID list acquired by the sales page ID acquisition unit 201 into product IDs and generates a product ID list. In this embodiment, the product ID list generation unit 202 converts the sales page IDs included in the sales page ID list into product IDs using a product conversion table 211 stored in the table storage unit 210. The product conversion table 211 is a table that associates product IDs of products sold by the EC mall with sales page IDs that identify one or more sales pages selling the products.

[0041] In an e-commerce mall, the same product may be sold on different sales pages. Therefore, in the product conversion table 211, one product ID may correspond to different sales page IDs. When different sales page IDs correspond to one item ID, the product ID list generation unit 202 converts the different sales page IDs into the same product ID. Figure 13C shows an example of a screen 1320 in which the same product is sold on different sales pages. In screen 1320, sales page 1306 and sales page 1307 are sales pages selling the same product 1308. In such a case, the product ID list generation unit 202 converts the sales page IDs of sales page 1306 and sales page 1307 into the product ID of product 1308, respectively.

[0042] FIG. 7 shows an example of a list 70 of product IDs for each user ID, which is converted by the product ID list generation unit 202 from the list 60 of sales page IDs for each user shown in FIG. 6. The list 70 of product IDs is configured to include a product ID 701 for each user ID 301. In the list 60 of sales page IDs for each user, it is assumed that the sales page IDs 302 for user ID 301 = UID_1, "abc121," "bcd942," "bcd125," and "kyr580," correspond to product IDs "ABC," "BCD," "BCD," and "KYR," respectively, according to the product conversion table 211. In this case, the product ID list generation unit 202 lists product IDs 701 = "ABC," "BCD," "BCD," and "KYR" for user ID 301 = UID_1. Here, sales page IDs "bcd942" and "bcd125" correspond to one product ID "BCD," but "BCD" is listed for each sales page ID.

[0043] In this embodiment, the product ID list generation unit 202 converts the sales page ID into a product ID by consulting the product conversion table 211, but the conversion process may be performed using other means. For example, the product ID list generation unit 202 may perform the conversion process using a trained machine learning model.

[0044] In S53, the recommended product list generation unit 203 determines product IDs (recommended product IDs) of products to be recommended for each user ID based on the list of product IDs generated by the product ID list generation unit 202, and generates a list of recommended product IDs. In this embodiment, the recommended product list generation unit 203 inputs a list of product IDs of products for which a user of the user ID has previously performed an action (information indicating the list) into the feature vector extraction model 221, and acquires a vector representation of the user of the user ID (hereinafter referred to as a user vector representation). The recommended product list generation unit 203 then embeds the acquired user vector representation in a common vector space. The vector space contains embedded vector representations of product IDs sold on the e-commerce site. The recommended product list generation unit 203 selects (extracts) product IDs whose vector representations have a high degree of similarity to the acquired user vector representation in the vector space. For example, cosine similarity is used as the similarity. Cosine similarity is a measure of the similarity between two vectors in a common vector space. The recommended product list generation unit 203 can determine that the greater the cosine value (-1 to +1) of the angle formed by two vectors, the user vector representation and the vector representation of the product ID, the higher the similarity. The recommended product list generation unit 203 calculates the cosine similarity between the acquired user vector representation and the vector representation of any product ID in the vector space, and determines a certain number of product IDs with high cosine similarity (for example, equal to or greater than a predetermined threshold) as recommended product IDs to be included in the list of recommended product IDs.

[0045] 7, for example, the recommended product list generation unit 203 inputs a list of product IDs = {ABC, BCD, KYR} for a user with user ID = UID_1 to the feature vector extraction model 221. The feature vector extraction model 221 extracts and outputs a user vector representation of the user with user ID = UID_1 from the input list of product IDs. That is, the recommended product list generation unit 203 converts the list of product IDs into a user vector representation using the feature vector extraction model 221. Next, the recommended product list generation unit 203 determines, as a recommended product ID, a product ID corresponding to a vector representation having a high cosine similarity (for example, higher than a predetermined threshold) with respect to the user vector representation in the vector space. The recommended product list generation unit 203 performs this process for each user ID and generates a list of recommended product IDs for each user ID.

[0046] 8 shows an example of a list 80 of recommended product IDs for each user ID generated by this process. The list 80 of recommended product IDs is configured to include a recommended product ID 801 for each user ID 301. For example, in the list 80 of recommended product IDs, three recommended product IDs, "ABE," "BCP," and "KYQ," are listed in the recommended product ID 801 for user ID=UID_1.

[0047] In S54, the sales page determination unit 204 determines a sales page (recommended sales page) for each of one or more products to be recommended to each user based on the recommended product IDs included in the recommended product list generated by the recommended product list generation unit 203. The sales page determination unit 204 may determine multiple recommended sales pages for each of one or more products to be recommended to each user, but in this embodiment, it determines a unique (i.e., one) recommended sales page. The sales page determination unit 204 first converts one or more recommended product IDs included in the recommended product list into one or more sales page IDs by consulting the product conversion table 211. That is, the sales page determination unit 204 uses the product conversion table 211, which is configured for converting sales page IDs into product IDs, to convert the recommended product IDs into sales pages. The sales page determination unit 204 generates a list of recommended sales page IDs including the converted one or more sales page IDs.

[0048] Figure 9 shows an example of a list 90 of recommended sales page IDs corresponding to recommended product IDs for each user ID, generated based on the list 80 of recommended product IDs shown in Figure 8. The list 90 of recommended sales page IDs is configured to include a recommended product ID 801, a sales page ID 901, and a recommended sales page ID 902 for each user ID 301. The sales page determination unit 204 identifies one or more sales page IDs (candidates for recommended sales page IDs) for one recommended product ID, and determines a unique recommended sales page ID from the identified one or more sales page IDs.

[0049] If one sales page ID (candidate for recommended sales page ID) exists for one recommended product ID, that sales page ID is determined as the recommended sales page ID. For example, in the list 90 of recommended sales page IDs, the recommended sales page ID 902 = "abe528" is determined for the recommended product ID 801 = "ABE" of user ID = UID_1. Also, the recommended sales page ID 902 = "bcp182" is determined for the recommended product ID 801 = "BCP" of user ID = UID_1, and the recommended sales page ID 902 = "kyq490" is determined for the recommended product ID 801 = "KYQ". In this case, the sales page determination unit 204 determines the sales pages identified by the recommended sales page IDs = "abe528", "bcp182", and "kyq490" as the recommended sales pages to be recommended (presented) to the user identified by user ID = UID_1.

[0050] On the other hand, if multiple sales page IDs exist for one recommended product ID, one sales page ID selected from the multiple sales page IDs is determined as the recommended sales page ID. For example, when the user ID is UID_2, three sales page IDs exist corresponding to the recommended product ID 801 = "CDT." In this case, when multiple sales page IDs correspond to a recommended product ID, the sales page determination unit 204 may determine one recommended sales page ID based on the evaluation index associated with the sales page collected by the EC site server 11 and stored in the evaluation index database 250. For example, if the CVR for the sales page is used as the evaluation index, the sales page determination unit 204 can select a sales page ID with a high CVR as the recommended sales page ID. A store selling products on a sales page identified by a sales page ID with a high CVR is considered to be a highly reliable purchasing destination. Therefore, determining a recommended sales page ID based on the CVR can lead to improved usability.

[0051] Furthermore, if multiple sales page IDs exist for one recommended product ID, the sales page determination unit 204 may determine the recommended sales page ID based on an evaluation index other than CVR. For example, the sales page determination unit 204 may determine the recommended sales page ID based on the evaluation score for the sales page as the evaluation index. Alternatively, the sales page determination unit 204 may calculate a recommendation score from the evaluation score and CVR, and determine the recommended sales page ID based on the recommendation score.

[0052] As described above, according to this embodiment, the information processing device 10 converts information on the sales page of a product associated with each user in the EC mall into product information in the EC mall. That is, the information processing device 10 appropriately associates information on the sales page on which each user has performed an action with product information in the EC mall. Then, the information processing device 10 determines a recommended product based on the product information and determines a unique sales page for selling the recommended product. This makes it possible to appropriately determine the sales page of the product to be recommended to each user.

[0053] Second Embodiment The information processing device according to the first embodiment determines the sales page of the product to be recommended to each user without considering the genre (category) of the product. The information processing device according to this embodiment estimates one or more genres that each user is interested in and determines the recommended sales page by limiting it to that genre. This embodiment will be described below, but the same configuration and features as the first embodiment will not be described.

[0054] Fig. 10 shows an example of the functional configuration of an information processing device 1000 according to this embodiment. Compared to the information processing device 10 according to the first embodiment described with reference to Fig. 2, the information processing device 1000 additionally includes a genre determination unit 205 and a target genre database 260. Furthermore, the learning model storage unit 220 additionally includes an MF-based prediction model 222, which is a matrix factorization (MF)-based genre prediction model, and an NLP-based prediction model 223, which is a natural language processing (NLP)-based genre prediction model. Although not shown in Fig. 10, the information processing device 1000 is connected to the EC site server 11 via the network 13, similar to Fig. 2.

[0055] The target genre database 260 stores information about a plurality of target genres determined in advance based on predetermined criteria (hereinafter referred to as target genre information). The target genre information stored in the target genre database 260 is generated by a processing unit (not shown) in the information processing device 1000 or an external device, and is updated as needed. The target genre information may be target genre features that represent the features of each of the plurality of target genres, or target genre IDs that identify each of the plurality of target genres.

[0056] The MF-based prediction model 222 is a model that decomposes an evaluation value matrix consisting of m (number of rows) × n (number of columns) (m and n are integers equal to or greater than 2) into two matrices by performing dimensionality reduction using collaborative filtering. In this embodiment, the MF-based prediction model 222 is configured to decompose an evaluation value matrix consisting of m user IDs and n genre IDs into a matrix representing user features and a matrix representing genre features. The NLP-based prediction model 223 is a natural language processing model that converts words into vector representations, such as Word2Vec. In this embodiment, the NLP-based prediction model 223 is configured to convert user IDs and genre IDs into user features and genre features. The MF-based prediction model 222 and the NLP-based prediction model 223 are each trained models and can be trained by a learning unit (not shown) in the information processing device 10.

[0057] The processing executed by the genre determination unit 205 will be described with reference to Figs. 11 and 12. In this embodiment, the genre determination unit 205 determines, for each user, the genre IDs of the top four genres in which the user is interested (top four genre IDs). Fig. 11 is a flowchart of the genre determination processing executed by the genre determination unit 205 according to this embodiment. Fig. 12 is a conceptual diagram showing the flow of data for determining the top four genre IDs. In this embodiment, the processing by the genre determination unit 205 to determine the top four genre IDs will be described, but the number of top genre IDs to be determined is not limited to four.

[0058] In S111, the genre determination unit 205 acquires a list of time-series sales page IDs (see FIG. 3) stored in the sales page ID database 240 of the EC site server 11. Next, in S112, the genre determination unit 205 acquires a user ID and a genre ID (see FIG. 3) from the list of sales page IDs acquired in S111. As described above, the user ID is associated with the user attributes of the user identified by the user ID. Furthermore, as shown in FIG. 3, the genre ID is an ID that identifies the genre of the product sold on the sales page identified by the corresponding sales page ID 302. In FIG. 12, the user ID and genre ID acquired by the genre determination unit 205 are represented by user ID 1200 and genre ID 1201, respectively. Although FIG. 12 shows user ID 1200 and genre ID 1201, the genre determination unit 205 acquires multiple user IDs and multiple genre IDs by sequentially performing the acquisition process.

[0059] In S113, the genre determination unit 205 inputs the user ID 1200 and multiple genre IDs 1201 into a trained genre prediction model to acquire user features 1202 that express the user's interests and genre features 1203 that express the genre features. The genre prediction model used here is the MF-based prediction model 222 or the NLP-based prediction model 223. The user features 1202 and genre features 1203 can be associated with the user ID 1200 and the genre IDs 1201, respectively.

[0060] When using the MF-based prediction model 222, the genre determination unit 205 first generates an evaluation value matrix consisting of m user IDs and n genre IDs as input data. In the example of FIG. 3, the number of user IDs is 3 (UID_1 to UID_3) and the number of genre IDs is 5 (g_1 to g_5), so a 3 × 5 evaluation value matrix is ​​generated. The values ​​of the elements in the evaluation value matrix can be set by referring to the list of time-series sales page IDs acquired in S111. For example, the genre determination unit 205 refers to the list of time-series sales page IDs acquired in S111 and sets a predetermined value to the genre ID corresponding to each user ID, and sets zero (null value) to the genre ID that does not correspond to each user ID. Referring to FIG. 3, the genre determination unit 205 can set the values ​​of the elements of genre IDs = g_1, g_3, and g_4 to a predetermined value and the values ​​of the elements of genre IDs = g_2 and g_5 to zero in the column of user ID = UID_1. Note that a different value may be set for the genre ID corresponding to each user ID depending on the action type 303 in Fig. 3. For example, different values ​​(for example, weight 702 shown in Fig. 15, which will be described later) may be set for "purchase," "add to favorites," and "view."

[0061] Next, the genre determination unit 205 inputs the generated rating value matrix to the MF-based prediction model 222. The MF-based prediction model 222 reduces the dimensions of the input rating value matrix to generate and output two matrices, namely, a matrix representing user features and a matrix representing genre features. The matrix representing the user features and the matrix representing the genre features can be divided into user features 1202 for each user ID and genre features 1203 for each genre ID. The user features 1202 and genre features 1203 are vector representations embedded in a common vector space and correspond to user embedding and genre embedding, respectively.

[0062] On the other hand, when the NLP-based prediction model 223 is used, the genre determination unit 205 inputs the user ID 1200 and the genre ID 1201 to the NLP-based prediction model 223. The NLP-based prediction model 223 converts the user ID and the genre ID into vector representations and outputs them as user features and genre features. The user features and the genre features are vector representations embedded in a common vector space and correspond to user embedding and genre embedding, respectively.

[0063] In S114, the genre determination unit 205 determines the target genre features 1204 based on the genre features 1203 and the target genre information stored in the target genre database 260. In this example, to ultimately determine the top four genre IDs, target genre information of five or more target genres, which is more than the number of the top genre IDs, is used. The genre determination unit 205 filters (selects) the genre features 1203 based on the target genre information to determine the target genre features 1204. For example, if the target genre information contains five target genre IDs, the genre determination unit 205 determines the genre features 1203 associated with the five target genre IDs as the target genre features 1204.

[0064] In S115, the genre determination unit 205 determines the genre IDs 1205 of the top four based on the user features 1202 and the target genre features 1204. The genre determination unit 205 determines the genre IDs 1205 of the top four according to the similarity between the user features 1202 and the target genre features 1203. As the similarity, for example, cosine similarity is used. As described above, cosine similarity is a measure that represents the similarity between two vectors in a common vector space. The genre determination unit 205 can determine that the greater the cosine value (-1 to +1) of the angle formed by the two vectors of the user features 1202 and the target genre features 1204, the higher the similarity.

[0065] The top four genre IDs 1205 for each user determined by the genre determination unit 205 are used to determine a sales page. For example, the top four genre IDs 1205 may be transmitted to the sales page ID acquisition unit 201. In this case, the sales page ID acquisition unit 201 may exclude genre IDs other than the top four genre IDs when acquiring a list of sales page IDs for each user as shown in FIG. 6. Alternatively, the top four genre IDs 1205 may be transmitted to the sales page determination unit 204. In this case, the sales page determination unit 204 may exclude genre IDs other than the top four genre IDs 1205 when generating a list of sales page IDs as shown in FIG. 9. As a result, recommended sales pages for genre IDs included in the top four genre IDs 1205 are determined for each user.

[0066] In this way, according to this embodiment, the recommended sales page of the product to be recommended is determined for one or more genres that each user is estimated to be more interested in. This makes it possible to appropriately determine the recommended sales page of the product to be recommended to each user by limiting it to one or more genres that are in line with the interests of each user.

[0067] Third Embodiment In the above embodiment, the information processing device determines the sales page of a product to be recommended to each user based on information collected in response to predetermined actions taken by multiple users on a sales page deployed in an EC mall. In this embodiment, a process for determining a sales page to be presented as a default to a user (hereinafter referred to as a new user) who accesses (visits) the EC mall for the first time will be described. Note that the process according to this embodiment can be implemented by any of the information processing devices of the above embodiments, and therefore will be described with reference to the information processing device 10 shown in FIG. 2, which was described in the first embodiment.

[0068] The sales page ID acquisition unit 201 acquires a list of chronological sales page IDs collected by the EC site server 11 and stored in the sales page ID database 240. In this embodiment, the list is not sorted by user, but rather, for example, a list of chronological sales page IDs acquired by the EC site server 11 during a predetermined period (for example, a period from the time of acquisition to a predetermined time before). An example of the list of acquired sales page IDs is shown in Figure 3. The sales page ID acquisition unit 201 may acquire the list of chronological sales page IDs at predetermined time intervals, or may acquire it when a new user accesses the EC mall.

[0069] Next, the product ID list generation unit 202 generates a list of product IDs from the list of time-series sales page IDs acquired by the sales page ID acquisition unit 201. The procedure for generating the list of product IDs is as described in the first embodiment, and can be performed using the product conversion table 211. In the process up to this point, a list of product IDs is generated that identifies products corresponding to sales pages on which multiple users performed actions within a predetermined period.

[0070] The sales page determination unit 204 determines a recommended sales page for each of one or more products to be recommended to the new user based on the product IDs included in the list of product IDs generated by the product ID list generation unit 202. For example, the sales page determination unit 204 converts the product IDs included in the list of product IDs into one or more sales page IDs using the product conversion table 211, and generates a list of recommended sales page IDs for the new user. Then, the sales page determination unit 204 determines the sales page identified by the recommended sales page ID included in the list of recommended sales page IDs as the sales page to be recommended (presented) to the new user.

[0071] When the sales page ID acquisition unit 201 acquires a list of chronological sales page IDs at predetermined intervals, the sales page determination unit 204 sequentially generates a list of recommended sales page IDs. The sales page determination unit 204 can then determine a sales page to recommend to the new user based on the list of recommended sales page IDs generated at a timing close to the timing when the new user accessed the EC mall. Also, when the sales page ID acquisition unit 201 acquires a list of chronological sales page IDs at the timing when the new user accessed the EC mall, the sales page determination unit 204 generates a list of recommended sales page IDs corresponding to that timing. The sales page determination unit 204 can then determine a sales page to recommend to the new user based on the list of recommended sales page IDs.

[0072] The sales page determination unit 204 may identify as a new user a user who is not included in the browsing history information of the EC mall acquired by the EC site server 11. Alternatively, the sales page determination unit 204 may identify as a new user a user other than the users included in the list of time-series sales page IDs acquired by the sales page ID acquisition unit 201, without limiting the new user to a user who has accessed the EC mall for the first time.

[0073] The list of time-series sales page IDs acquired by the sales page ID acquisition unit 201 is not limited to the list of time-series sales page IDs acquired by the EC site server 11 within a predetermined period. For example, the sales page ID acquisition unit 201 may acquire a list of time-series sales page IDs limited to the number of times of one or more predetermined action types. Specifically, the sales page ID acquisition unit 201 may acquire a list of time-series sales page IDs associated with a predetermined number of "purchase" actions from the EC site server 11, limited to the "purchase" action. In addition, the sales page ID acquisition unit 201 may generate a list of time-series sales page IDs acquired for one or more users who have user attributes similar to those of the new user from the list of time-series sales page IDs acquired by the EC site server 11, and output the list to the product ID list generation unit 202.

[0074] In this way, according to this embodiment, it is possible to present recommended sales pages to new users who access the EC mall for the first time, based on sales pages associated with the action histories of other users collected in the past. This allows the new users to view sales pages of products for which actions such as purchases have been performed more frequently on the EC mall, which may increase their desire to purchase.

[0075] <Fourth embodiment> In the first embodiment, the recommended product list generation unit 203 generated a list of recommended product IDs using the feature vector extraction model 221 stored in the learning model storage unit 220, but in this embodiment, the recommended product list generation unit 203 generates this list on a rule basis. Specifically, the recommended product list generation unit 203 generates a list of recommended product IDs using a recommended product conversion table 212 described below. This embodiment will be described below, but descriptions of the same configurations and features as in the first embodiment will be omitted. In addition, the second and third embodiments described above can also be applied to this embodiment.

[0076] FIG. 14 illustrates an example of the functional configuration of an information processing device 1400 according to this embodiment. Compared to the information processing device 10 according to the first embodiment described with reference to FIG. 2, the information processing device 1400 includes a table storage unit 210 having a recommended product conversion table 212. The recommended product conversion table 212 is a table that associates product IDs with recommended product IDs (recommended product identification information) that identify one or more products recommended based on the product identified by the product ID. The recommended product conversion table 212 associates one or more recommended product IDs with product IDs based on past performance. For example, the recommended product conversion table 212 associates the product IDs of products purchased by a user group having similar user attributes in an e-commerce mall with the product IDs (recommended product IDs) of one or more other products purchased by users belonging to the group simultaneously with or after the purchase of the product. If there are many such other products, the product IDs of a predetermined number of products that are purchased at a high rate by the user group (i.e., purchased by many users of the user group) may be listed as recommended product IDs. The product conversion table 211 and the recommended product conversion table 212 are generated by a processing unit (not shown) in the information processing device 10 or an external device, and are updated as needed. The information processing device 1400 may be configured to simultaneously perform the processing described in the first embodiment. In this case, the information processing device 1400 may determine the sales page for one or more recommended product IDs determined by both processes.

[0077] The processing executed by the information processing device 10 according to this embodiment will be described with reference to Fig. 5. In Fig. 5, the processing of S51 and S52 is the same as that described in the first embodiment. In S53, the recommended product list generation unit 203 generates a recommended product list based on the recommended product conversion table 212. For example, the recommended product list generation unit 203 determines recommended product IDs corresponding to each product ID by referencing each product ID included in the product ID list 70 shown in Fig. 7 with the recommended product conversion table 212, and generates a list of recommended product IDs.

[0078] As an optional process, if the product ID list generation unit 202 assigns a weight indicating the user's level of interest according to the action type to each product ID in S52, the recommended product list generation unit 203 may generate a list of recommended product IDs based on the weight in S53. For example, the action of "purchasing" a product is an action of actually obtaining the product, and can be said to be an action that interests the user more than the actions of "viewing" and "add to favorites." Furthermore, the action of "add to favorites" a product is an action that makes it easy to access the sales page for the product again, and can be said to be an action that interests the user more than the action of "viewing." In this way, a weight according to the user's level of interest can be associated with each product ID.

[0079] FIG. 15 shows an example of a list 1500 in which weights 702 according to action types are added to the list 70 of product IDs for each user shown in FIG. 7. In the example of FIG. 15, weights of 1.0 for "purchase," 0.3 for "favorite," and 0.2 for "view" are assigned to the corresponding product IDs according to the user's interest level. Note that different weights may be assigned to the same action type depending on the action date and time 304 in the list 60 of sales page IDs for each user corresponding to the list 70 of product IDs. For example, a "purchase" action performed at a newer date and time may be considered to have a higher user interest level than a "purchase" action performed at an older date and time. Therefore, a higher weight may be assigned to a "purchase" action performed at a newer date and time than to a "purchase" action performed at an older date and time. For example, the weight for actions performed within a first period going back from the date and time when the list of sales page IDs 30, which is the basis for the list of sales page IDs 60, was acquired may be multiplied by a coefficient of 1, the weight for actions performed within a second period going back from the first period may be multiplied by a coefficient of 0.8, and the weight for actions performed before the second period may be multiplied by a coefficient of 0.5. Also, when multiple identical product IDs are listed for the same user and a weight is assigned to each, such as product ID 701="BCD" in the list of product IDs 70, the total weight may be assigned to the product ID.

[0080] When a weight is assigned according to the action type (weight 702 shown in FIG. 15), the recommended product list generating unit 203 can determine a recommended product ID based on the weight. For example, when multiple recommended product IDs correspond to one product ID in the recommended product conversion table 212, the recommended product list generating unit 203 may determine more recommended product IDs for a product ID with a higher weight.

[0081] <Modification> In the above embodiment, the sales page ID acquisition unit 201 acquires a list of time-series sales page IDs, and the product ID list generation unit 202 generates a list of product IDs from the list, but the list of product IDs may be generated by the EC site server 11. In this case, the product ID list generation unit 202 may acquire the product ID list generated by the EC site server 11.

[0082] In the fourth embodiment, when a weight is assigned according to the action type, the recommended product list generating unit 203 determines the recommended product ID based on the weight. However, the weight may be used in another process. For example, the sales page determining unit 204 may determine the recommended sales pages, and may instruct the EC site server 11 to display the recommended sales pages with a high weight in a different display mode from other recommended sales pages (for example, to have a more visual effect).

[0083] The effects of the above embodiment and the above modified example will be described with reference to FIGS. 13A and 13C. In screen 1300 in FIG. 13A, multiple products are sold on different sales pages. On the other hand, in screen 1320 in FIG. 13C, product 1308 is sold on sales page 1306 and sales page 1307. In such a case, whether a user purchases product 1308 on sales page 1306 or on sales page 1307, the purchase action is associated with product 1308. This makes it possible to accurately aggregate actions for each product, making it possible to determine appropriate products to recommend to the user and sales pages for the recommended products.

[0084] Furthermore, when multiple recommended sales page candidates are determined for one product, one (unique) recommended sales page is determined for the product based on an evaluation index such as CVR. By determining the recommended sales page based on an evaluation index, a sales page for a store that is highly rated by multiple users is determined, which can improve usability and user willingness to purchase.

[0085] The disclosure of this embodiment includes the following configuration. [1] An information processing device having: an acquisition unit that acquires a list of sales page identification information that identifies each of the sales pages of one or more products associated with each user; a conversion unit that converts the list of sales page identification information into a list of product identification information that identifies the products; a generation unit that generates a recommended product list including recommended product identification information that identifies each of one or more products to be recommended to each user based on the list of product identification information; and a determination unit that determines a sales page for each of the one or more products to be recommended to each user based on one or more of the recommended product identification information included in the recommended product list.

[0086] [2] The information processing device according to [1], wherein the determination unit determines a unique sales page for each of one or more products identified by one or more pieces of recommended product identification information.

[0087] [3] The information processing device described in [2], wherein the determination unit identifies one or more sales pages that sell each of one or more products identified by one or more of the recommended product identification information, and determines the unique page based on the one or more sales pages.

[0088] [4] The information processing device described in [3], wherein the determination unit, when identifying one sales page that sells the product identified by the recommended product identification information, determines the identified sales page as the unique sales page; and, when identifying multiple sales pages that sell the product identified by the recommended product identification information, determines the unique sales page from among the multiple identified sales pages based on evaluation indexes associated with the multiple sales pages.

[0089] [5] The information processing device described in [4], wherein the evaluation index includes a CVR (conversion rate) for each of the identified multiple sales pages.

[0090] [6] The information processing device according to [4] or [5], wherein the evaluation index includes user evaluation points for each of the identified sales pages.

[0091] [7] An information processing device described in [1] to [6], further comprising a conversion unit that converts, for each user, product identification information included in the list of product identification information into a user vector representation using a machine learning model, and the generation unit determines, as the recommended product identification information, one or more product identification information corresponding to one or more vector representations whose cosine similarity with the user vector representation is higher than a predetermined threshold in a common vector space in which vector representations of multiple product identification information exist.

[0092] [8] An information processing device described in any of [1] to [7], further comprising a selection unit that selects one or more genres in which each of the users is interested, and the acquisition unit acquires a list of the sales page identification information for the one or more genres selected for each of the users.

[0093] [9] The information processing device described in [8], wherein the selection unit selects one or more genres for each user using a matrix factorization (MF)-based machine learning model.

[0094]

[10] The information processing device described in [8], wherein the selection unit selects one or more genres for each user using a natural language processing-based machine learning model.

[0095]

[11] The information processing device described in any of [1] to

[10] , wherein the determination unit further determines the sales page for one or more products to be recommended to users other than each of the users based on the list of product identification information. [Explanation of symbols]

[0096] 10: Information processing device, 11: EC site server, 12: User device, 13: Network, 14: User, 201: Sales page ID acquisition unit, 202: Product ID list generation unit, 203: Recommended product list generation unit, 204: Sales page determination unit, 210: Table storage unit, 211: Product conversion table, 212: Recommended product conversion table, 220: Learning model storage unit, 221: Feature vector extraction model, 222: MF (Matrix Factorization) based genre prediction model, 223: Natural language processing (NLP) based genre prediction model, 231: EC mall operation unit, 232: Evaluation index derivation unit, 240: Sales page ID database, 250: Evaluation index database, 260: Target genre database

Claims

1. an acquisition unit that acquires a list of sales page identification information that identifies each of the sales pages of one or more products associated with each user; a conversion unit that converts the list of sales page identification information into a list of product identification information that identifies products; a generating unit that generates a recommended product list including recommended product identification information that identifies one or more products to be recommended to each of the users based on the list of product identification information; a determination unit that determines a sales page for each of the one or more products to be recommended to each user based on the one or more pieces of recommended product identification information included in the recommended product list; An information processing device having the above.

2. The information processing device according to claim 1 , wherein the determination unit determines a unique sales page for each of one or more products identified by one or more pieces of recommended product identification information.

3. The information processing device according to claim 2, wherein the determination unit identifies one or more sales pages that sell each of the one or more products identified by the one or more pieces of recommended product identification information, and determines the unique page based on the identified one or more sales pages.

4. The determination unit If one sales page selling the product identified by the recommended product identification information is identified, the identified sales page is determined to be the unique sales page; When a plurality of sales pages selling the product identified by the recommended product identification information are identified, the unique sales page is determined from the plurality of identified sales pages based on evaluation indexes associated with the plurality of sales pages. The information processing device according to claim 3 .

5. The information processing device according to claim 4 , wherein the evaluation index includes a CVR (conversion rate) for each of the identified plurality of sales pages.

6. The information processing device according to claim 4 , wherein the evaluation index includes an evaluation score by a user for each of the plurality of identified sales pages.

7. a conversion unit that converts, for each user, the product identification information included in the product identification information list into a user vector representation using a machine learning model; The information processing device according to claim 1, wherein the generation unit determines, as the recommended product identification information, one or more product identification information corresponding to one or more vector expressions having a cosine similarity with the user vector expression higher than a predetermined threshold in a common vector space in which vector expressions of multiple product identification information exist.

8. a selection unit for selecting one or more genres in which each of the users is interested; The acquisition unit acquires a list of the sales page identification information for one or more genres selected for each of the users. The information processing device according to claim 1 .

9. The information processing device according to claim 8 , wherein the selection unit selects one or more genres for each of the users using a matrix factorization (MF)-based machine learning model.

10. The information processing device according to claim 8 , wherein the selection unit selects one or more genres for each of the users using a machine learning model based on natural language processing.

11. The information processing apparatus according to claim 1 , wherein the determining unit further determines sales pages for one or more products to be recommended to users other than the respective users based on the list of product identification information.

12. An information processing method executed by an information processing device, an acquisition step of acquiring a list of sales page identification information that identifies each of the sales pages of one or more products associated with each user; a conversion step of converting the list of sales page identification information into a list of product identification information that identifies products; a generating step of generating a recommended product list including recommended product identification information that identifies one or more products to be recommended to each of the users based on the list of product identification information; a determining step of determining a sales page for each of the one or more products to be recommended to each user based on the one or more pieces of recommended product identification information included in the recommended product list; An information processing method, including:

13. An information processing program for causing a computer to execute information processing, the program including: an acquisition process for acquiring a list of sales page identification information that identifies each of the sales pages of one or more products associated with each user; A conversion process for converting the list of sales page identification information into a list of product identification information for identifying products; a generation process for generating a recommended product list including recommended product identification information for identifying one or more products to be recommended to each of the users based on the list of product identification information; a determination process for determining a sales page for each of the one or more products to be recommended to each user based on one or more pieces of recommended product identification information included in the recommended product list, Information processing program.

Images