Information processing system, information processing method, and program

Abstract

To make a proposal to a user based on a phrase accumulated as metadata of a commodity by flexibly taking the mood of the user who tries to purchase the commodity at that time.SOLUTION: In the CPU11 of the management server 1, the information management unit 32 manages phrases of commodities based on the viewpoints of consumers, the phrases being defined in advance for each of commodities constituting a commodity group to be sold, the generation unit 34 extracts, from the commodity group in the commodity master DB44, a commodity corresponding to a phrase having a predetermined relevance to the psychological state of a user who is going to purchase any of the commodity group, and generates information to be proposed to the user, and the transmission control unit 35 proposes the generated proposal to the user. Thus, the above problem is solved.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention relates to an information processing system, an information processing device, and a program. [Background technology]

[0002] In the process of producing retail flyers and point-of-purchase (POP) advertisements, phrases based on the consumer's perspective are sometimes attached to products. In addition, technology has been proposed for suggesting products to users who are about to purchase a product based on the user's purchase history (for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2025-048378 Summary of the Invention [Problem to be solved by the invention]

[0004] However, no technology has been proposed that can flexibly grasp the mood of a user who is about to purchase a product and make suggestions to the user based on phrases stored as product metadata.

[0005] The present invention was made in consideration of such circumstances, and aims to flexibly grasp the mood of a user who is about to purchase a product and make suggestions to the user based on phrases stored as product metadata. [Means for solving the problem]

[0006] In order to achieve the above object, one aspect of the present invention is to a management means for managing phrases of products based on a consumer's viewpoint, which are predefined for each product constituting a group of products to be sold; a suggestion means for extracting from the group of products products corresponding to the phrases that have a predetermined relevance to the psychological state of a user who is about to purchase one of the group of products, and suggesting the extracted products to the user; It is an information processing system having the above.

[0007] An information processing device, an information processing method, and a program corresponding to the information processing system according to one aspect of the present invention are also provided as an information processing device, an information processing method, and a program corresponding to the information processing system according to one aspect of the present invention. [Effects of the Invention]

[0008] According to the present invention, it is possible to flexibly grasp the mood of a user who is about to purchase a product at that time, and make suggestions to the user based on phrases stored as product metadata. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a diagram illustrating an example of the overall configuration of an information processing system according to an embodiment of the present invention. [Figure 2] 2 is a block diagram showing an example of a hardware configuration of a management server that constitutes the information processing system of FIG. 1. FIG. [Figure 3] FIG. 2 is a functional block diagram illustrating an example of a functional configuration of a management server. [Figure 4] 10 is a flowchart illustrating an example of a processing flow of a management server. [Figure 5] FIG. 4 is a diagram showing specific examples of phrases stored in the phrase DB of FIG. 3. [Figure 6] 10A and 10B are diagrams showing an example of a screen presented to a user when a selection-type technique is used. [Figure 7] FIG. 10 is a diagram showing another example of a screen presented to a user when a selection-type technique is used. [Figure 8]1A is a diagram showing a specific example of a chat screen presented to a user when the selection-based method is used, and FIG. 1B is a diagram showing a specific example of a chat screen presented to a user when the conversation-based method is used. [Figure 9] 1A is a diagram showing a specific example of a chat screen presented to a user when a search-based method is used, and FIG. 1B is a diagram showing a specific example when a cart-based method is used. [Figure 10] FIG. 2 is a diagram showing a specific example of a prompt that causes a process to estimate a user's psychological state, among the prompts input to the generation AI server that constitutes the information processing system of FIG. 1. [Figure 11] FIG. 10 is a diagram showing a specific example of a prompt input to the generation AI server that causes the server to perform a process to generate information to be suggested to the user. [Figure 12] FIG. 10 is a diagram showing an example of a recommendation screen for phrases (Kototag (registered trademark)) used in this embodiment. [Figure 13] FIG. 2 is a diagram illustrating a phrase generation process in the present embodiment. [Figure 14] FIG. 1 illustrates a phrase value measurement framework. [Figure 15] FIG. 1 illustrates the phrase value measurement framework in more detail. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, this embodiment will be described with reference to the drawings. <Information Processing System S> FIG. 1 is a diagram showing an example of the overall configuration of an information processing system S according to one embodiment of the present invention. The information processing system S is an information processing system configured to include a management server 1, a generation AI server 2, an administrator terminal 3, a user terminal 4, a store terminal 5, and a store cart 6, all connected via a network N. The network N is, for example, the Internet, a LAN (Local Area Network), a VPN (Virtual Private Network), etc.

[0011] The information processing system S is an information processing system that enables the provision of a service (hereinafter referred to as "this service") that suggests products to users who are visiting a store that sells products or to users who have not yet visited the store, based on the user's mood at that time. In this embodiment, users who are visiting the store include those who use this service by operating a user terminal 4 such as a smartphone that they own. Furthermore, users who are visiting the store include those who use this service by operating a store terminal 5 such as a digital signage installed in the store, and those who use this service by operating a store cart 6 such as a shopping cart installed in the store.

[0012] [Management Server 1] The management server 1 that constitutes the information processing system S is an information processing device that serves as a server that manages the entire information processing system S. The management server 1 is capable of executing predetermined application programs that make the information processing system S available. The management server 1 transmits various types of information to the generation AI server 2, administrator terminal 3, user terminal 4, store terminal 5, store cart 6, and external devices, and enables various processes to be executed. The management server 1 also acquires various types of information transmitted from the generation AI server 2, administrator terminal 3, user terminal 4, store terminal 5, store cart 6, and external devices, and enables various processes to be executed.

[0013] For example, the management server 1 manages product phrases based on the consumer's perspective that are predefined for each product that constitutes a product group for sale in a store such as a supermarket. Such phrases define the usage scenarios and meanings of the product from the consumer's perspective, and are used, for example, in flyers, POP displays, etc. as emotive words that stimulate consumers' desire to purchase. The phrases are associated with the product as product metadata and managed. For example, suppose the product is yogurt. In this case, consumer-oriented phrases such as "health management" and "daily diligence" are associated with the product as product metadata and managed. Specific examples of product phrases will be described later with reference to FIG. 5.

[0014] Here, the phrases in this embodiment will be described in detail. The phrases in this embodiment are metadata that verbalize the consumer's usage scenario, psychological state, and emotional value for a product. Unlike conventional product categories and function descriptions, they function as new metadata that express consumer motivations and usage contexts, such as "why choose that product" and "how do you feel when using it?"

[0015] The generation and assignment of phrases can be achieved through the following process: first, an information decomposition process that breaks down product attributes into function, use, and emotional value; second, a scene extraction process that identifies usage situations, psychological states, and time axes; third, a phrase generation process that converts this information into expressions from the consumer's perspective; fourth, an effect measurement process that evaluates value using quantitative indicators; and fifth, an accumulation process that compiles the data into a database in a highly versatile format.

[0016] Using phrases generated through such a systematic process provides the following technical benefits. First, systematic phrase generation based on practical experience over a certain period of time enables product recommendations that accurately capture consumers' latent needs. Second, quality assurance through multifaceted value measurement improves recommendation accuracy. Third, a versatile design makes it possible to deploy the same phrases across multiple media, improving the scalability of the system. Fourth, collaboration with AI technology enables dynamic product recommendations that correspond to the user's psychological state.

[0017] It is preferable to use phrases having the above characteristics in this embodiment. Specifically, for example, Kototag (registered trademark) can be used as the phrase.

[0018] The management server 1 also estimates the psychological state of a user who is about to purchase one of a group of products for sale in a store. For example, the management server 1 can present a question to the user and estimate the user's psychological state based on the user's answer to the question. In this case, the method for presenting the question to the user and the method for the user to answer the question are not particularly limited. For example, a selection-based method may be used in which the question is presented to the user as a set of options and the user is allowed to select an option. Alternatively, for example, a conversation-based method may be used in which the question and answer are a dialogue between a chatbot and the user in the form of text data or voice data. Specific examples of the selection-based method and the conversation-based method will be described later.

[0019] This service, realized by the management server 1, is an item-based recommendation system that uses phrases (Kototag (registered trademark)) associated with products. Specifically, the management server 1 suggests products based on the association between the user's current psychological state and product phrases, without using the user's purchase history data. This makes it possible to suggest products to new users without the cold start problem, and also makes it an excellent system from the perspective of privacy protection, as it does not require the accumulation and management of personal purchase history data.

[0020] Furthermore, the management server 1 can estimate the psychological state of a user based on spontaneous questions from the user. In this case, the method by which the user asks a question is not particularly limited. For example, a search-type method in which text data or voice data that constitutes a spontaneous question is input to a chatbot may be used. Specific examples of search-type methods will be described later.

[0021] The management server 1 can extract products corresponding to phrases having a predetermined relevance to the estimated psychological state of the user from the product group and suggest them to the user. For example, suppose the estimated psychological state of the user is "I want to buy ingredients that will enhance the effects of muscle training." In this case, for example, the product "chicken breast" corresponding to the phrase "rich in protein," which is related to muscle training and muscle building, is extracted from the product group.

[0022] Furthermore, when a user registers one or more products from a group of products for movement to a predetermined physical area, the management server 1 can extract products from the group of products that correspond to phrases that are the same as or have a predetermined relationship with the phrases corresponding to the products, and suggest them to the user. An example of the predetermined physical area is a basket into which products are placed in the store cart 6. Furthermore, registration for movement to a physical area means, for example, reading an identifier (an example of identification information) attached to the product that is identifiable and linked to the product, such as a one-dimensional code attached to the product, when the product is placed in the basket in the store cart 6. In this case, it can also be considered that the management server 1 estimates the user's psychological state from the products registered by the user.

[0023] For example, suppose the product registered in the store cart 6 is yogurt, and the corresponding phrases are "health management" and "daily steady." In this case, other products with the same phrase or other products associated with phrases having a predetermined association, such as vegetable juice, are suggested. Hereinafter, the method of suggesting other products to the user based on phrases corresponding to products registered in the store cart 6 will be referred to as the cart-type method.

[0024] When suggesting products to a user, the management server 1 is able to make suggestions that take into consideration external information (hereinafter referred to as "external information") that may affect the user's psychological state. The external information includes basic context information consisting of weather, season, temperature, date, day of the week, etc., and store context information consisting of events, the intentions of the store manager (e.g., sale items, discounted items, recommended items, etc.). The configuration and processing of the management server 1 will be described in detail later.

[0025] The management server 1 realizes flexible recommendations by combining proprietary data, phrases (Kototag (registered trademark)), with external data (basic context information and store context information). Phrases function as proprietary data for eliciting the user's purpose and product usage scenarios, and by combining them with external data, it is possible to propose products that are more suited to the user's psychological state. In addition, by taking store context information into consideration, it is possible to effectively recommend products that the store wants to sell while understanding user needs.

[0026] External information can be broadly divided into user-oriented context information and store-oriented context information. User-oriented context information includes basic environmental information such as weather, season, temperature, date, and day of the week, as well as product phrases (Kototag®), sale information, etc. Store-oriented context information includes events held at the store, the store manager's intentions (sale items, discounted items, recommended items, etc.), and inventory status. By adjusting the weights of these context information, the management server 1 realizes product proposals that optimize the balance between user satisfaction and store sales targets. As a specific example of weight adjustment, for recommendations in the evening on weekdays, the weight of user-oriented context information can be set to 80% and that of store-oriented context information to 20%. This is based on the judgment that, while users tend to feel more tired and need to save time during this time period, user needs should be prioritized over store promotions. Conversely, on weekend sale days, the weights of user-oriented context information can be adjusted to 60% and store-oriented context information to 40%, respectively, to help achieve store sales targets while maintaining user satisfaction. Furthermore, when the store is about to close, it is possible to prioritize inventory utilisation and increase the weighting for the store up to 50%. This dynamic weighting adjustment allows for the optimal balance of product recommendations to be achieved depending on the time of day, day of the week, events, and other circumstances.

[0027] [Generation AI Server 2] The generation AI server 2, which constitutes the information processing system S, utilizes a large-scale language model (LLM) and deep learning to provide information for the management server 1 to perform the following processes. Specifically, the generation AI server 2 executes the following processes using natural language processing with a large-scale language model based on the input prompt and outputs the processing results. Based on the input prompt, the generation AI server 2 performs processes such as generating questions to be presented to the user, estimating the user's psychological state, determining whether or not there is a correlation between the user's psychological state and phrases, and generating information to suggest to the user, and outputs the results of each process. The generation AI server 2 then transmits the output processing results to the management server 1. The questions and answers generated and output by the generation AI server 2 include natural text data and voice data personalized for each target user.

[0028] [Administrator terminal 3] The administrator terminal 3 constituting the information processing system S is an information processing device operated by the administrator of the store that uses this service. The administrator terminal 3 is configured, for example, as a personal computer, tablet terminal, smartphone, etc. The administrator terminal 3 is capable of executing a predetermined application program that enables the use of this service. The administrator terminal 3 is capable of executing various processes based on various information transmitted from the management server 1, the store terminal 5, the store cart 6, and external sources, as well as various information input by the store administrator. The administrator terminal 3 is also capable of transmitting various information to each of the management server 1, the store terminal 5, the store cart 6, and external sources. For example, the administrator terminal 3 transmits store context information to the management server 1.

[0029] [User terminal 4] The user terminal 4 constituting the information processing system S is an information processing device operated by a user who uses the service. The user terminal 4 is configured, for example, as a smartphone, a tablet terminal, or the like. The user terminal 4 is capable of executing a predetermined application program that enables use of the service. The user terminal 4 is capable of executing various processes based on various information transmitted from the management server 1 and external sources, and various information input by the user. The user terminal 4 is also capable of transmitting various information to both the management server 1 and external sources.

[0030] For example, the user terminal 4 realizes the above-mentioned selection-type method, conversation-type method, and search-type method. That is, the user terminal 4 transmits spontaneous questions input by the user and answers to questions presented to the user to the management server 1. Also, for example, the user terminal 4 acquires questions and answers transmitted from the management server 1 and outputs them as text data or voice data. Also, the user terminal 4 acquires suggestions transmitted from the management server 1 and outputs them as text data or voice data.

[0031] [Store terminal 5] The store terminal 5 constituting the information processing system S is an information processing device operated by a user who visits the store. The store terminal 5 is configured, for example, as a digital signage installed in the store. The store terminal 5 is capable of executing a predetermined application program that enables use of the present service. The store terminal 5 is capable of executing various processes based on various information transmitted from the management server 1, the administrator terminal 3, and external sources, as well as various information input by the user. The store terminal 5 is also capable of transmitting various information to the management server 1, the administrator terminal 3, and external sources.

[0032] For example, the store terminal 5 realizes the above-mentioned selection-type method. That is, the store terminal 5 transmits to the management server 1 an answer to a question input by the user and presented to the user. The store terminal 5 also acquires a proposal transmitted from the management server 1 and outputs it as text data or voice data. The store terminal 5 can also realize the above-mentioned conversation-type method and search-type method.

[0033] [Store Cart 6] The store cart 6, which constitutes the information processing system S, is a so-called smart cart used by users who visit the store. The store cart 6 utilizes RFID (radio frequency identification) and AI technology to include an information processing device that reads identification information such as one-dimensional codes attached to products, performs automatic checkout, manages product inventory, creates shopping lists, and navigates within the store. The store cart 6 is capable of executing a predetermined application program that enables use of the service. The store cart 6 is capable of executing various processes based on various information transmitted from the management server 1, the administrator terminal 3, and external sources, as well as various information input by users. The store cart 6 is also capable of transmitting various information to the management server 1, the administrator terminal 3, and external sources.

[0034] For example, the store cart 6 realizes the cart-type method described above. That is, when a product corresponding to a phrase that is the same as or has a predetermined relevance to a phrase corresponding to a product registered by a user and added to the cart is transmitted from the management server 1, the store cart 6 acquires the product as a suggestion and outputs it as text data or voice data. Note that the store terminal 5 can also realize the conversation-type method, search-type method, and selection-type method described above.

[0035] The above-described processing by each of the management server 1, generation AI server 2, administrator terminal 3, user terminal 4, store terminal 5, and store cart 6 that constitute the information processing system S is merely an example. In other words, the information processing system S as a whole is required to have the functions to realize the above-described processing, and some or all of the functions to realize the above-described processing may be shared or cooperated within the information processing system S.

[0036] For example, some or all of the functions of the management server 1 may be functions of other information processing devices in the information processing system S. Also, some or all of the functions of other information processing devices in the information processing system S may be functions of the management server 1. Furthermore, some or all of the functions of the management server 1 may be transferred to other servers (not shown). This promotes processing in the information processing system S as a whole and also makes it possible for processes to complement each other.

[0037] <Hardware configuration> [Hardware configuration of Management Server 1] FIG. 2 is a block diagram showing an example of the hardware configuration of the management server 1 that constitutes the information processing system S of FIG. The management server 1 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a bus 14, an input / output interface 15, an output unit 16, an input unit 17, a memory unit 18, a communication unit 19, and a drive 20.

[0038] The CPU 11 executes various processes in accordance with programs recorded in the ROM 12 or programs loaded from the storage unit 18 into the RAM 13. The RAM 13 also stores data and the like required for the CPU 11 to execute various processes. The CPU 11, the ROM 12, and the RAM 13 are interconnected via a bus 14. An input / output interface 15 is also connected to this bus 14.

[0039] The input / output interface 15 is connected to an output unit 16, an input unit 17, a storage unit 18, a communication unit 19, and a drive 20. The output unit 16 is composed of a display, a speaker, etc., and outputs various types of information as images, sounds, etc. The input unit 17 is composed of a keyboard, a mouse, a touch panel, etc., and accepts input of various types of information. The storage unit 18 is composed of a hard disk, a DRAM (Dynamic Random Access Memory), etc., and stores various types of data. The communication unit 19 communicates with other devices via the above-mentioned network N, which is composed of the Internet, etc.

[0040] Removable media 21, such as a magnetic disk, optical disk, magneto-optical disk, or semiconductor memory, is appropriately attached to the drive 20. Programs read from the removable media 21 by the drive 20 are installed in the storage unit 18 as needed. The removable media 21 can also store various data stored in the storage unit 18 in the same way as the storage unit 18.

[0041] [Hardware configuration of the administrator terminal 3, user terminal 4, store terminal 5, and store cart 6] The administrator terminal 3, user terminal 4, store terminal 5, and store cart 6 each have a hardware configuration similar to that of the management server 1 shown in Fig. 2. That is, the administrator terminal 3, user terminal 4, store terminal 5, and store cart 6 each have a CPU, ROM, RAM, bus, input / output interface, output unit, input unit, memory unit, communication unit, drive, and removable media (not shown) that correspond to the CPU 11, ROM 12, RAM 13, bus 14, input / output interface 15, output unit 16, input unit 17, memory unit 18, communication unit 19, drive 20, and removable media 21 shown in Fig. 2. The administrator terminal 3, user terminal 4, store terminal 5, and store cart 6 may have an imaging unit (such as a camera) as an input unit and an audio output unit (such as a speaker) as an output unit.

[0042] <Functional configuration of management server 1> FIG. 3 is a functional block diagram showing an example of the functional configuration of the management server 1. As shown in FIG. When the CPU 11 of the management server 1 operates, an information acquisition unit 31, an information management unit 32 as a management means, an estimation unit 33 as an estimation means, a generation unit 34 as a proposal means, and a transmission control unit 35 function.

[0043] Furthermore, various databases are provided in the storage unit 18 of the management server 1. For example, databases such as a basic context DB 41 storing basic context information among the external information, a store context DB 42 storing store context information among the external information, a phrase DB 43 storing phrases, and a product master DB 44 storing master information related to products are provided. In the storage unit 18, the phrase DB 43 and the product master DB 44 are associated in a relational manner.

[0044] The information acquisition unit 31 acquires various types of information. For example, the information acquisition unit 31 acquires various types of information transmitted to the management server 1 from the generation AI server 2, the administrator terminal 3, the user terminal 4, the store terminal 5, the store cart 6, and external sources. Among these, information transmitted from the generation AI server 2 to the management server 1 includes, for example, generated questions and answers, and generated information to be proposed to users. Information transmitted from the administrator terminal 3 to the management server 1 includes, for example, store context information. Information transmitted from the user terminal 4 and the store terminal 5 to the management server 1 includes, for example, questions and answers. Information transmitted from the store cart 6 to the management server 1 includes, for example, information for registering products. Information transmitted from external sources to the management server 1 includes, for example, external information.

[0045] The information management unit 32 stores and manages various types of information in the database of the storage unit 18. For example, the information management unit 32 stores and manages basic context information acquired by the information acquisition unit 31 in a basic context DB 41. The information management unit 32 also stores and manages store context information acquired by the information acquisition unit 31 in a store context DB 42. The information management unit 32 also stores and manages phrases in a phrase DB 43. The information management unit 32 also stores and manages master information related to products in a product master DB 44.

[0046] The estimation unit 33 estimates the psychological state of a user who is about to purchase one of a group of products for sale in a store. The estimation of the user's psychological state by the estimation unit 33 is performed using the generation AI server 2. For example, the estimation unit 33 estimates the user's psychological state based on the user's answer to a question presented to the user. In this case, the estimation unit 33 is able to estimate the user's psychological state based on the user's answer (selection of an option) to a question presented to the user using a selection-based method.

[0047] The estimation unit 33 can estimate the user's psychological state based on the user's response to a question presented to the user through a conversational method (a dialogue of text data or voice data).The estimation unit 33 can also estimate the user's psychological state based on a question from the user through a search method (input information of text data or voice data).

[0048] The generation unit 34 generates questions to be presented to the user. The generation unit 34 generates questions using the generation AI server 2. For example, the generation unit 34 generates questions to be presented to the user using a selection-based method based on external information (basic context information and store context information). Also, for example, the generation unit 34 generates questions to be presented to the user using a conversational method.

[0049] Furthermore, the generation unit 34 generates information to be suggested to the user. The generation unit 34 generates information to be suggested to the user using the generation AI server 2. For example, the generation unit 34 extracts products corresponding to phrases that have a predetermined relevance to the psychological state of the user estimated by the estimation unit 33 from the group of products. Then, the generation unit 34 generates information to be suggested to the user based on the extracted products.

[0050] Here, we will explain in more detail the processing by the estimation unit 33 and the generation unit 34. The information processing system of this embodiment employs a three-step processing configuration, from the user's "mind" through "motivation" to "product recommendation."

[0051] The first step is to understand the user's "mind." The estimation unit 33 estimates the user's current psychological state (mind) from spontaneous questions from the user, answers to questions presented to the user, or registered information when the user moves a product into a physical area. For example, psychological states such as "I'm tired," "I want to relax," or "I want to have fun with my family" are estimated.

[0052] The second step is to identify "motivation." The estimation unit 33 identifies the user's motivation for purchasing based on the estimated mindset. For example, from the mindset of "I'm tired," motivations such as "I want to easily replenish my nutrients" and "I want to eat something delicious and easy" are derived. This motivation is used to determine the relevance with phrases stored in the phrase DB 43.

[0053] The third step is "product recommendation." The generation unit 34 extracts products that have phrases associated with the identified motivation from the product master DB 44 and recommends them to the user. This three-step configuration enables highly accurate product recommendations based on the user's deep psychology.

[0054] For example, when a user registers to move one or more products from a product group to a predetermined physical area, the generation unit 34 extracts products from the product group that correspond to phrases that are the same as or have a predetermined relationship with the phrases corresponding to the products. The generation unit 34 then generates information to be proposed to the user based on the extracted products. Furthermore, when generating information to be proposed to the user, the generation unit 34 can generate information that takes into account external information (basic context information and store context information). For example, the generation unit 34 can generate coupon information that matches the user's needs as an indirect PDP (personal dynamic pricing) based on the user's psychological state estimated from the user's response results. This indirect PDP realizes personalized price proposals based only on the user's immediate response results, without directly using the user's personal information or purchase history. When generating information to be proposed to the user, the generation unit 34 can adjust the weighting of user-oriented context information and store-oriented context information, thereby optimizing the balance between user satisfaction and store sales targets.

[0055] Furthermore, the generation unit 34 may use the following method when generating information to be suggested to the user. That is, the generation unit 34 may vectorize each of the text data indicating the user's psychological state (mind) and purchasing willingness (motivation) and the master information related to the product, calculate the similarity (for example, cosine similarity) between the obtained vectors, and extract and suggest the top n products (n is an integer value of 1 or more) that are similar to the selected mind and motivation.

[0056] Here, we will explain in more detail the matching algorithm executed by the estimation unit 33. The estimation unit 33 calculates the cosine similarity between a psychological state vector obtained by vectorizing text data indicating the psychological state of the user and a phrase vector obtained by vectorizing phrases for each product stored in the product master DB 44.

[0057] Specifically, the estimation unit 33 extracts products whose calculated cosine similarity is equal to or greater than a predetermined threshold (e.g., 0.75) as products that match the user's psychological state. This threshold can be adjusted based on the system's operational status and user feedback.

[0058] Furthermore, the generation unit 34 performs weighting processing on the extracted products, taking into consideration external information. Specifically, the generation unit 34 applies a weather coefficient (for example, a range of 0.1 to 0.3) based on weather information acquired from the basic context DB 41, a time period coefficient (for example, a range of 0.1 to 0.3) based on time period information, and a sale coefficient (for example, a range of 0.1 to 0.3) based on sale information acquired from the store context DB 42.

[0059] The final recommendation score is calculated as the sum of the values ​​obtained by multiplying the cosine similarity by each coefficient. For example, it can be calculated using a formula such as: Final score = Cosine similarity x (1 + Weather coefficient + Time of day coefficient + Sale coefficient). This process dynamically suggests products with phrases that best fit the user's current psychological state.

[0060] The transmission control unit 35 controls the communication unit 19 (see FIG. 2) to transmit various types of information to the generation AI server 2, the administrator terminal 3, the user terminal 4, the store terminal 5, the store cart 6, and external devices. For example, the transmission control unit 35 controls the transmission of prompts to the generation AI server 2 to generate and output questions to be presented to the user. For example, the transmission control unit 35 controls the transmission of prompts to the generation AI server 2 to estimate and output the user's psychological state. For example, the transmission control unit 35 controls the transmission of prompts to the generation AI server 2 to determine whether or not there is a correlation between the user's psychological state and a phrase. For example, the transmission control unit 35 controls the transmission of prompts to the generation AI server 2 to generate information to be suggested to the user. The prompts are transmitted in the form of text data or audio data.

[0061] The transmission control unit 35 also controls the transmission of the question generated by the generation unit 34 to the user terminal 4. The transmission control unit 35 also controls the transmission of information generated by the generation unit 34 to be proposed to the user to the user terminal 4.

[0062] <Processing flow of Management Server 1> Fig. 4 is a flowchart showing an example of the processing flow of the management server 1. Note that the example shown in Fig. 4 shows an example in which a multiple-choice method is used as an example of a method for presenting a question to a user and a method for the user to answer the question. The management server 1 generates options for questions to be presented to the user based on external information (basic context information and store context information) and phrases stored in the phrase DB 43 (see FIG. 3 ) (step S1). At this time, various context data (basic context information such as weather, season, temperature, date, and day of the week, and store context information such as store inventory information, sales promotion information, and event information) acquired by the information acquisition unit 31 are utilized. The estimation unit 33 analyzes this context data and generates options for questions appropriate to the current situation. For example, options for psychological states according to the situation are generated, such as "I want to warm my wet body" on rainy days, "I want to feel refreshed" on extremely hot days, and "I want to enjoy the weekend" on Friday nights. The management server 1 presents the generated options to the user (step S2). Specifically, the management server 1 generates options using the generation AI server 2 and presents them to the user by displaying them on the display of the user terminal 4, the store terminal 5, or the store cart 6.

[0063] When one option is selected from the options presented to the user (YES in step S3), the management server 1 acquires the input information (step S4). Here, for example, assume that the option "I'm busy, but I want to get some nutrition" is selected. In this case, the input information is acquired by the management server 1. On the other hand, when one option is not selected from the options presented to the user (NO in step S3), the management server 1 repeats the determination process of step S3.

[0064] Then, the management server 1 estimates the user's psychological state based on the input information acquired in step S4 and the phrases stored in the phrase DB 43 (step S5), and presents the estimation result to the user in a selectable format (step S6). Specifically, the management server 1 estimates the user's psychological state using the generation AI server 2, and presents the estimation result to the user by displaying it in a selectable format on the display of the user terminal 4, the store terminal 5, or the store cart 6, etc.

[0065] When one psychological state is selected from the psychological states of the user presented in a selectable manner (YES in step S7), the management server 1 acquires the input information (step S8). Here, for example, assume that the psychological state "getting nutrition while warming up with soup" is selected. In this case, the input information is acquired by the management server 1. On the other hand, when one psychological state is not selected from the psychological states presented to the user (NO in step S7), the management server 1 repeats the determination process of step S7.

[0066] Next, the management server 1 generates information to be suggested to the user based on the input information acquired in step S8, the phrases stored in the phrase DB 43, and the master information on products stored in the product master DB 44 (step S9). Specifically, the management server 1 uses the generation AI server 2 to extract products corresponding to phrases that have a predetermined relevance to the user's psychological state from the master information on products. Then, the management server 1 generates information to be suggested to the user based on the extracted products.

[0067] Next, the management server 1 proposes the information generated in step S9 to the user (step S10). Specifically, the management server 1 proposes the generated information to the user by displaying it on the display of the user terminal 4, the store terminal 5, or the store cart 6. For example, a product called "Our Miso Soup with 4 Types of Vegetables" is proposed to the user. This causes the management server 1 to end the process (END). Thereafter, the user who has received the product proposal determines whether or not to purchase the proposed product.

[0068] <Example> FIG. 5 shows specific examples of phrases stored in the phrase DB 43 of FIG. As shown in Figure 5, the phrase DB43 in the memory unit 18 of the management server 1 associates examples of phrases such as "#hashtag1" and "#hashtag2," "single item" which is identification information that can uniquely identify the product, the "manufacturer, brand, and place of origin" of the product, and the "advertised product name" which is the content of the product.

[0069] For example, a product whose "advertising product name" is "Manufacturer A XX Yogurt YY Plain Unsweetened 1 pack (280g)" is associated with the phrases "Health Management" (#hashtag1) and "Diligently Every Day" (#hashtag2). Also, for example, a product whose "advertising product name" is "Manufacturer A XX Jelly Peach & Muscat 1 pack (70g x 4)" is associated with the phrases "Dessert Time" (#hashtag1) and "Relaxation Time" (#hashtag2). Other specific examples of phrases associated with products are shown in Figure 5.

[0070] Here, we will explain a specific example of phrase generation. For example, consider a product called frozen dumplings. Conventional product classifications have been limited to functional and categorical expressions such as "frozen food," "Chinese food," and "quick cooking."

[0071] In contrast, the phrase generation process of this embodiment first analyzes usage scenarios. For frozen dumplings, usage scenarios such as "busy weekday dinner," "family get-together," and "as a companion to alcohol" are extracted. Next, the user's psychological state corresponding to these usage scenarios is estimated. For example, psychological states such as "tired," "want to take it easy but want to eat something delicious," and "want to have fun with everyone" are estimated.

[0072] Based on the results of these analyses, phrases from the consumer's perspective are generated. For example, phrases such as "A dish to say thank you for your hard work" (a simple recipe for when you're tired), "Family smiles" (a scene of family time together), and "Evening drinks time" (a drink to go with alcohol) are generated. These phrases are not simply descriptions of product functions, but function as metadata that express the emotional value linked to the user's psychological state and usage scenario.

[0073] The phrases generated in this way are stored in the phrase DB 43 and are used to recommend products according to the user's psychological state. For example, if the user's psychological state is estimated to be "tired today," frozen dumplings are recommended with the phrase "a dish to show your appreciation for your hard work."

[0074] 6A and 6B are diagrams showing an example of a screen presented to the user when the selection-type technique is used. When a selection-based method is used as an example of a method for presenting questions to a user and a method for the user to answer the questions, a screen such as that shown in FIG. 6(A) may be displayed on each of the user terminal 4, the store terminal 5, and the store cart 6. On the screen shown in FIG. 6(A), options are displayed in a display area F1 below the words "Recommended for you." The options displayed in the display area F1 are questions generated based on the estimation results of the user's psychological state.

[0075] Specifically, the display area F1 displays options such as "Spring-colored menus you'll want to try soon," "The tough season is coming... start before the season!", and "Easy! Easy! Easy lunches." When one of the options displayed in the display area F1 is selected, a product corresponding to the selected option is suggested. The product suggestion may be for only one product, but may also be in a format that allows multiple products to be selected and purchased, as shown in FIG. 6(B). In this case, the user can check out products that interest them from the suggested products and save them in a "shopping memo."

[0076] FIG. 7 is a diagram showing another example of a screen presented to the user when the selection-type technique is used. When the selection-based method is used, a screen such as that shown in FIG. 7 may be displayed on each of the user terminal 4, the store terminal 5, and the store cart 6. The screen shown in FIG. 7 displays a selection button B11 below the phrase "Please select your motivation" to allow the user to select their own psychological state. In the example of FIG. 7, "Let's all have a fun barbecue!" is selected in the selection button B11. The selection button B11 is a question generated based on the estimation result of the user's psychological state. Furthermore, when the selection button B11 is selected, an image screen representing the content of the selected selection button B11 may be displayed. In the example of FIG. 7, an image screen showing many people enjoying a barbecue is displayed.

[0077] The screen shown in Figure 7 also displays the details of the products suggested to the user. Specifically, under the notation "Recommended top related products!", products such as "0 XXX Light Salt 1 piece (75g)," "1 XXX 1 bag (1kg)," and "2 XXX Medium Spicy 1 stick (400g)" are displayed.

[0078] Fig. 8(A) is a diagram showing a specific example of a chat screen presented to a user when the selection-based method is used, and Fig. 8(B) is a diagram showing a specific example of a chat screen presented to a user when the conversation-based method is used. The chat screen shown in Figure 8(A) displays the following options as questions from the chatbot: "How are you feeling? A. I don't want to think about it, so please make a suggestion. B. I want to find a bargain. C. I'm wondering what to have for dinner." The options displayed on the chat screen are questions generated to estimate the user's psychological state.

[0079] The chat screen shown in FIG. 8(A) also displays a message "C" as the user's answer to the question from the chatbot above. That is, the user is answering that they are worried about what to have for dinner. The chat screen shown in FIG. 8(A) also displays a further question from the chatbot in response to the above answer from the user, with the options "Please select what interests you: A. Spring-colored menu B. Healthy eating habits C. Take a break from time to time." The options displayed on the chat screen are questions generated to suggest products to the user.

[0080] The chat screen shown in Figure 8(A) also displays a message "A" as the user's response to the further question from the chatbot. In other words, the user is responding by asking for suggestions for a "spring color menu." The chat screen shown in Figure 8(A) also displays a message "Then we recommend the following: ·XXX ·XXX" as a product suggestion from the chatbot in response to the user's response.

[0081] The chat screen shown in Figure 8(B) displays a conversation between the chatbot and the user. Specifically, the chatbot asks a question: "Good morning. How are you feeling?" This message is generated to estimate the user's psychological state.

[0082] The chat screen shown in Figure 8(B) displays the message "Refreshing" as the user's response to the question from the chatbot. The chat screen shown in Figure 8(B) also displays the message "That's great! How's your appetite?" as a further question from the chatbot in response to the user's response. This message is a question generated to suggest a product to the user.

[0083] The chat screen shown in Figure 8(B) displays the message "Quite a few" as the user's response to a further question from the chatbot. The chat screen shown in Figure 8(B) also displays the message "I understand. In that case, I recommend this: A. Yogurt drink type B. Bananas from XX C. XX sandwich" as a product suggestion from the chatbot in response to the user's response.

[0084] Fig. 9(A) is a diagram showing a specific example of a chat screen presented to a user when a search-based method is used, and Fig. 9(B) is a diagram showing a specific example when a cart-based method is used. The chat screen shown in Figure 9(A) displays a conversation between a user and a chatbot. Specifically, the message displayed is a question from the user: "I'm going to start working out today. What should I buy?"

[0085] The chat screen shown in Figure 9(A) displays the following message as a response from the chatbot to the above user's question: "That's great! If you're planning on doing some muscle training, I recommend these: A.YY broccoli, B.ZZ chicken breast, C.XX protein." This message is a question generated to suggest products to the user.

[0086] Figure 9(B) shows a specific example of a case where the cart-type method is used, in which the product registered in the store cart 6 is yogurt and the corresponding phrases are "health management" and "daily steady." In the example of Figure 9(B), vegetable juice is proposed as another product that shares a phrase with yogurt.

[0087] FIG. 10 is a diagram showing a specific example of a prompt that causes the generation AI server 2 constituting the information processing system S of FIG. 1 to perform a process of estimating the psychological state of the user, among the prompts input to the generation AI server 2 of FIG. The prompt shown in Figure 10 is an example of a prompt such as, "- Prompt: "You are a product advisor at a supermarket. Below is a unique list of hashtags (Kototags (registered trademark)) that indicate the characteristics of products. # Hashtag list: {unique_tags_list}When a customer comes to your store, please briefly think of {mind_count} minds (35 characters or less each) that will help you understand their current state of mind, taking into account the hashtag list and the following four perspectives. The four perspectives are as follows: 1. Regarding dates and events..."

[0088] FIG. 11 shows a specific example of a prompt input to the generation AI server 2 that causes the server 2 to perform a process for generating information to be proposed to the user. The prompt shown in Figure 11 is as follows: "- Prompt: "You are a product advisor at a supermarket. Below are the purchasing mindsets of customers who visit your store. # Purchasing mindset: {mind} Below is a unique list of hashtags that describe the characteristics of the product. # Hashtag list: {unique_tags_list}..."

[0089] The phrases in this embodiment will now be described in more detail with reference to Figures 12 to 14. The phrases in this embodiment are not simply descriptions of the product, but function as important elements that connect the consumer's psychological state with the product.

[0090] FIG. 12 is a diagram showing an example of a recommendation screen for phrases (Kototag (registered trademark)) used in this embodiment. As shown in FIG. 12, buttons are arranged at the top of the screen that allow users to select psychological states such as "when tired," "relax," and "family time." Each product is assigned a phrase that verbalizes the consumer's usage scenario, psychological state, and emotional value, such as "a great ally when tired," "energy recharge," "reset the fatigue of the day," and "a treat for yourself." Unlike traditional product categories and function descriptions, these phrases function as new metadata that express consumer motivations and usage contexts, such as "why choose that product" and "how do you feel when using it?"

[0091] FIG. 13 is a diagram illustrating the phrase generation process in this embodiment. As shown in FIG. 13, product information (e.g., frozen dumplings) is first decomposed into function, purpose, and emotional value, such as "frozen food, Chinese food, time-saving," in the information decomposition process. Next, the scene extraction process identifies usage situations, psychological states, and time axes, such as "busy weekday dinner, family gathering, companion to alcohol." In the phrase generation process, this information is converted into expressions from the consumer's perspective, such as "a dish to show appreciation for hard work, smiles from the family, evening drinks." After that, in the effect measurement process, value is evaluated using quantitative indicators such as "click rate, purchase rate, and satisfaction," and finally, in the database creation process, the data is stored in a highly versatile format.

[0092] Figure 14 is a diagram showing the phrase value measurement framework. As shown in Figure 14, value is evaluated based on four perspectives: numerical effect, reproducibility with other companies, sustainability, and versatility, centered on phrases (Kototag (registered trademark)). Numerical effect enables evaluation using quantitative indicators and contributes to improving recommendation accuracy. Reproducibility with other companies enables accurate capture of consumers' latent needs through systematic phrase generation based on a predetermined number of years of work experience. Sustainability ensures quality through multifaceted value measurement, and versatility design makes it possible to deploy the same phrase across multiple media. By integrating these elements in conjunction with AI technology, dynamic product recommendations based on the user's psychological state are realized.

[0093] Figure 15 shows the phrase value measurement framework in more detail. As shown in Figure 15, the phrase (Kototag (registered trademark)) placed at the center is evaluated multilaterally based on four main evaluation axes.

[0094] First, quantitative effects such as a 30% increase in click rate and an increase in purchase rate are measured. This allows for an objective evaluation of the phrase's actual business impact. Second, to ensure reproducibility with other companies, a systematic generation process has been established based on a unique phrase system and knowledge accumulated over a certain period of time. This makes it possible to consistently generate high-quality phrases.

[0095] Third, in terms of versatility, cross-media deployment and multi-channel compatibility are realized. The same phrase can be used across multiple media, such as store terminals, user terminals, and store carts, improving the efficiency of the entire system. Fourth, in terms of sustainability, value assessment is performed through the accumulation of performance data and an improvement cycle. This makes it possible to continuously improve the quality of phrases.

[0096] By using phrases evaluated using such a multifaceted value measurement framework, the information processing system of this embodiment realizes highly accurate and sustainable product recommendations that correspond to the user's psychological state.

[0097] <Advantageous Effects of the Present Embodiment> According to the above-described embodiment, it is possible to flexibly grasp the mood of a user who is about to purchase a product, and make suggestions to the user based on phrases stored as product metadata. Users will be able to find products that suit their current psychological state without being tied down by historical data, and will also be able to purchase products that meet their needs in bulk. Store managers can evolve their stores into ones that offer higher customer satisfaction by drawing out the user's psychological state and suggesting products. It also becomes possible to directly grasp customer needs by utilizing accumulated user selection history. Secondary personalized recommendations (e.g., email newsletters) based on the identified customer needs become possible. Furthermore, by incorporating store context, it becomes possible to efficiently sell products that the store wants to sell while adapting to the user's psychological state. As a result, customer satisfaction can be improved. An increase in the number of items purchased simultaneously can also be expected, leading to an increase in the unit purchase price. Furthermore, store context can improve the efficiency of inventory utilization. Manufacturers who provide cash registers to stores can strengthen the sales promotion function of smart carts and other retail media. They can also strengthen the PDP advertising space by displaying coupons. As a result, they can improve the added value of their products and increase advertising revenue. Product manufacturers who provide products to stores can track the psychological state of users that led to product purchases. In addition, advertising efficiency can be improved by displaying advertisements to users who have a need for a product. As a result, product planning that matches customer needs can be carried out by utilizing user selection history. In addition, sales from customers who match customer needs can be increased.

[0098] Hereinafter, this embodiment will be described in comparison with conventional technology. Conventional recommendation systems mainly rely on mechanical recommendations based on product categories and purchase histories, and do not have a systematic method for linking products with users' usage scenarios or psychological states. In particular, classifications that focus too much on the functional aspects of products make it difficult to link products with users' emotions and specific usage scenarios, and the systematization of accumulated marketing knowledge is also insufficient.

[0099] In contrast, the phrases in this embodiment are a technical implementation of knowledge that has been systematized in the retail and marketing fields over a certain period of time (25 years in the case of Kototag (registered trademark)). Specifically, they are based on practical experience in systematizing the process of expressing product information in phrases such as "spring outing," "drinking at home," and "time-saving" according to the user's lifestyle, and establishing a method of expression that converts things into experiences. This knowledge has been structured as a phrase database, and this invention has been realized as a technical means of linking the user's psychological state with products.

[0100] In conventional technology, there was no mechanism for systematically managing such consumer-oriented phrases and dynamically recommending products according to the user's psychological state, and this has been realized for the first time by the present invention.

[0101] Furthermore, according to this embodiment, the following quantitative effects are expected.

[0102] First, it allows for the creation of new relationships with customers that eliminate excessive recommendation fatigue. Unlike conventional mechanical recommendations based on purchase history, suggestions based on the user's psychological state make users feel that their feelings are understood, improving their trust in the system. This increases receptivity to recommendations and is expected to improve long-term customer engagement.

[0103] Second, the improvement in product discovery rate promotes new purchases. Products that would be classified as "frozen foods" in the conventional category classification can be newly discovered as an option for when you're tired by the phrase "a dish for thank you for your hard work." Such suggestions that transcend categories can uncover users' latent needs and significantly improve product discovery rates compared to the past.

[0104] Third, operational efficiency is improved by reusing phrases. Once a phrase is created, it can be reused for multiple products and multiple media (user devices, store devices, store carts, etc.), eliminating the need to create content for each new marketing initiative. This significantly reduces operational costs.

[0105] Fourth, by utilizing a database of phrases accumulated over a set number of years (25 years in the case of Kototag (registered trademark)), a strong competitive advantage can be secured. Phrases accumulated and verified over a long period of time become assets that cannot be easily imitated due to their quality and comprehensiveness. This acts as a strong barrier to entry, as it would take a similar amount of time for a new entrant to build a phrase database of the same quality.

[0106] <Other> Although one embodiment of the present invention has been described above, the present invention is not limited to the above-described embodiment, and modifications, improvements, etc. within the scope of achieving the object of the present invention are included in the present invention.

[0107] For example, the specific examples shown in each of FIGS. 5 to 11 are merely examples for achieving the object of the present invention, and are not particularly limited.

[0108] Furthermore, the above-described series of processes can be executed by hardware or software. In other words, the above-described functional configuration is merely an example and is not particularly limited. In other words, it is sufficient for the information processing system to be provided with a function that can execute the above-described series of processes as a whole, and the type of functional block used to realize this function is not particularly limited to the above-described example.

[0109] The location of the functional blocks is not particularly limited and may be arbitrary. For example, the functional blocks of the management server 1 may be transferred to another device, or the functional blocks of another device may be transferred to a server. Furthermore, one functional block may be configured as a single piece of hardware, a single piece of software, or a combination of both.

[0110] When a series of processes is executed by software, the programs constituting the software are installed onto a computer or the like from a network or a recording medium. The computer may be a computer incorporated into dedicated hardware. The computer may also be a computer capable of executing various functions by installing various programs, such as a server, a general-purpose smartphone, or a personal computer.

[0111] The recording medium containing such a program may be configured as a removable medium distributed separately from the device main body in order to provide the program to users, etc., or may be configured as a recording medium etc. that is pre-installed in the device main body and provided to users, etc. Since the program can be distributed via a network, the recording medium may be installed in or accessible to a computer that is connected or connectable to the network.

[0112] In this specification, the steps describing the program recorded on the recording medium include not only processes that are performed in chronological order, but also processes that are not necessarily performed in chronological order but are performed in parallel or individually. Also, in this specification, the term "system" means an overall device composed of multiple devices or multiple means, etc.

[0113] In other words, the information processing system to which the present invention is applied can take various forms having the following configurations. (1) That is, the information processing system S to which the present invention is applied is an information processing system having a management means (e.g., information management unit 32 in FIG. 3) that manages product phrases based on the consumer's perspective that are predefined for each product constituting a group of products for sale, and a proposal means (e.g., generation unit 34 and transmission control unit 35 in FIG. 3) that extracts products from the group of products that correspond to the phrases that have a predetermined relevance to the psychological state of a user who is about to purchase one of the group of products, and proposes the products to the user. This allows the system to flexibly grasp the mood of a user who is about to purchase a product at that time and make suggestions to the user based on phrases stored as product metadata.

[0114] (2) The system may further include an estimation means (e.g., estimation unit 33 in FIG. 3) for estimating the psychological state of the user based on either a spontaneous question from the user or a response from the user to a question presented to the user. This allows us to estimate the mood of the user who is about to purchase a product, which can be used for phrase-based suggestions.

[0115] (3) The presented question may be a presentation of options, and the answer may be a selection of an option. This allows the mood of the user who is about to purchase the product at that time to be estimated from the options selected by the user.

[0116] (4) The presented question and the answer may be a text or voice dialogue between the chatbot and the user. This allows the mood of a user who is about to purchase a product to be inferred from the content of the conversation between the chatbot and the user.

[0117] (5) Furthermore, when the user registers (e.g., reads the identification information attached to the product) one or more products from the product group to move them to a predetermined physical area (e.g., a basket into which products are placed in the store cart 6 in Figure 1), the suggestion means can extract products from the product group that correspond to phrases that are the same as or have a predetermined relationship with the phrase corresponding to the product, and suggest them to the user. This allows the user's mood at that time to be flexibly grasped from the product that the user has moved to a predetermined physical area, and makes suggestions to the user based on phrases stored as product metadata.

[0118] (6) The physical area may be a shopping cart used by the user. This allows the system to flexibly grasp the user's mood at the time from the products the user has added to their shopping cart, and make suggestions to the user based on phrases stored as product metadata.

[0119] (7) Furthermore, the suggestion means can make the suggestion taking into consideration external information that may affect the psychological state of the user. This allows suggestions to be made that take into account external information that may affect the user's psychological state, making it possible to make suggestions that are more suited to the user's mood.

[0120] (8) The external information may also include one or more of events being held at a store selling the product group, the intentions of the store manager, the weather on the day, the season, the temperature, the date, and the day of the week. This allows suggestions to be made that take into account external information that may affect the user's psychological state, making it possible to make suggestions that are more suited to the user's mood.

[0121] (9) Furthermore, the information processing method to which the present invention is applied can take various forms having the following configurations. In other words, the information processing method to which the present invention is applied is an information processing method that includes the steps of managing product phrases based on the consumer's perspective that are predefined for each product that constitutes a group of products for sale, and extracting products from the group of products that correspond to the phrases that have a predetermined relevance to the psychological state of a user who is planning to purchase one of the group of products, and proposing them to the user. This allows the system to flexibly grasp the mood of a user who is about to purchase a product at that time and make suggestions to the user based on phrases stored as product metadata.

[0122] (10) Furthermore, the program to which the present invention is applied can take various forms having the following configurations. In other words, the program to which the present invention is applied is a program for causing a computer that controls an information processing device (for example, the management server 1 in Figure 1) to execute control processing including the steps of managing product phrases based on the consumer's perspective that are predefined for each product that constitutes a group of products for sale, and extracting from the group of products products those products that correspond to the phrases that have a predetermined relevance to the psychological state of a user who is about to purchase one of the group of products, and proposing them to the user. This allows the system to flexibly grasp the mood of a user who is about to purchase a product at that time and make suggestions to the user based on phrases stored as product metadata. [Explanation of symbols]

[0123] 1: Management server, 2: Generation AI server, 3: Administrator terminal, 4: User terminal, 5: Store terminal, 6: Store cart, 11: CPU, 16: Output unit, 17: Input unit, 18: Memory unit, 19: Communication unit, 31: Information acquisition unit, 32: Information management unit, 33: Estimation unit, 34: Generation unit, 35: Transmission control unit, 41: Basic context DB, 42: Store context DB, 43: Phrase DB, 44: Product master DB, N: Network, S: Information processing system

Claims

1. a management means for managing phrases of products based on a consumer's viewpoint, which are predefined for each product constituting a group of products to be sold; a suggestion means for extracting from the group of products products corresponding to the phrases that have a predetermined relevance to the psychological state of a user who is about to purchase one of the group of products, and suggesting the extracted products to the user; An information processing system having the above.

2. The system further includes an estimation means for estimating a psychological state of the user based on either a spontaneous question from the user or an answer from the user to a question presented to the user. The information processing system according to claim 1 .

3. The presented question is a presentation of options, and the answer is a selection of an option; The information processing system according to claim 2 .

4. The presented questions and the answers are a text or voice dialogue between the chatbot and the user. The information processing system according to claim 2 .

5. When the user registers to move one or more products from the group of products to a predetermined physical area, the suggestion means extracts products from the group of products that correspond to phrases that are the same as or have a predetermined relevance to the phrases corresponding to the products, and suggests the products to the user. The information processing system according to claim 1 .

6. The physical area is a shopping cart used by the user. The information processing system according to claim 5 .

7. the suggestion means makes the suggestion taking into consideration external information that may affect the psychological state of the user. The information processing system according to claim 1 .

8. The external information includes one or more of events being held at a store selling the product group, the intention of the store manager, the weather on the day, the season, the temperature, the date, and the day of the week. The information processing system according to claim 7 .

9. An information processing method executed by an information processing device, A step of managing phrases of products based on a consumer's viewpoint, which are predefined for each product constituting a group of products to be sold; extracting from the group of products products corresponding to the phrases having a predetermined relevance to the psychological state of the user who is about to purchase any one of the group of products, and proposing the extracted products to the user; An information processing method including:

10. A computer that controls an information processing device A step of managing phrases of products based on a consumer's viewpoint, which are predefined for each product constituting a group of products to be sold; extracting from the group of products products corresponding to the phrases having a predetermined relevance to the psychological state of the user who is about to purchase any one of the group of products, and proposing the extracted products to the user; A program for executing control processing including:

Images