system

Abstract

We provide the system. [Solution] An input means for inputting the user's basic information, A question generation means that generates an initial question based on the basic information received from the input means, A display means for displaying the generated question to the user, A means of receiving user responses, An intent analysis means analyzes the response received from the aforementioned response receiving means and infers the user's purpose for visiting the store, A product recommendation means that recommends appropriate products based on the purpose of visit inferred by the aforementioned intent analysis means, A display method for showing recommended products to the user, A system that includes this.

Description

Technical Field

[0001] The technology of the present disclosure relates to a system.

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, and includes steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In conventional home appliance mass retailers, much time and labor were required to understand the purchase intention of customers. Also, it was difficult for store clerks to ask appropriate questions to customers and quickly grasp their purchase intention. Furthermore, when customers could not clearly express their purchase intention, it was difficult to propose the most suitable products. Thus, in the conventional system, there was a problem that customers could not efficiently find the products they needed, and the purchase process became complicated.

Means for Solving the Problems

[0005] To solve the above problems, the present invention provides a system that includes an input means for inputting basic user information, a question generation means for generating initial questions based on the basic information received from the input means, a display means for displaying the generated questions to the user, an answer receiving means for receiving the user's answers, an intent analysis means for analyzing the answers received from the answer receiving means and inferring the user's purpose for visiting the store, a product recommendation means for recommending appropriate products based on the purpose of visiting the store inferred by the intent analysis means, and a display means for displaying the recommended products to the user. With this system, customers can efficiently and quickly clarify their purchase intentions and receive optimal product suggestions based on those intentions simply by answering appropriate questions.

[0006] "Input means" refers to a device or interface for a user to input basic information.

[0007] A "question generation means" is a device or software that automatically creates appropriate questions based on basic information received from the user.

[0008] "Display means" refers to a device or interface for visually presenting generated questions or recommended products to a user.

[0009] "Response receiving means" refers to a device or interface that receives responses entered by a user.

[0010] "Intention analysis means" refers to a device or software that has the function of analyzing a user's responses to infer their intentions and purpose for visiting the store.

[0011] "Product recommendation means" refers to a device or software that has the function of recommending products suitable for a user based on the purpose of their visit, which is inferred by intent analysis means.

[0012] "System" refers to the entirety of the devices or software that include the aforementioned means and are used to infer the user's purpose for visiting the store and suggest appropriate products. [Brief explanation of the drawing]

[0013] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2, which incorporates an emotion engine. [Figure 14] This is a sequence diagram showing the processing flow of the data processing system in Application Example 2, which combines an emotion engine. [Modes for carrying out the invention]

[0014] Next, an example of an embodiment of the system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0015] First, the terms used in the following description will be explained.

[0016] <00​​​​​​​​​​​​In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0020] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0021] [First Embodiment]

[0022] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0023] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0024] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0025] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0026] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0027] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0028] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0029] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0031] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0032] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0033] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0034] As an embodiment of the present invention, a system that uses AI to predict the purchase purpose of a customer visiting a consumer electronics store will be described. This system has a process of taking the user's basic information as input, generating questions based on that information, and analyzing the user's answers to infer the purpose of the visit.

[0035] Initialization process

[0036] The server reads the product database and the question database, and prepares the necessary information in advance when the system starts up.

[0037] The device displays an initial screen that allows the user to enter basic information.

[0038] The user enters basic information (e.g., age, gender, categories of interest) on the displayed screen.

[0039] User information entry

[0040] The terminal receives basic information entered by the user and sends it to the server.

[0041] The server analyzes the received basic information and prepares to generate appropriate initial questions.

[0042] question generation

[0043] Based on the basic information, the server selects a suitable question from the question database and generates a question to display to the user.

[0044] Example: "What size TV screen are you interested in?"

[0045] The terminal displays the generated question to the user.

[0046] The user answers the question and sends the answer to the server via their device.

[0047] Receiving and analyzing responses

[0048] The server receives the user's response and saves it to its internal database.

[0049] The server analyzes the user's responses and processes them to generate the next question or to specifically infer the purpose of their visit.

[0050] Intent Analysis

[0051] The server uses natural language processing (NLP) techniques to analyze user responses and infer the user's purpose for visiting the store.

[0052] Example: Analyze the intention behind the statement, "I want a 50-inch TV."

[0053] Based on the analysis results, the server selects appropriate products from the product database and generates a list of recommended products.

[0054] Displaying recommended results

[0055] The server sends a list of recommended products along with an estimated reason for the customer's visit to the terminal.

[0056] The terminal displays to the user a list of suggested reasons for visiting the store and recommended products.

[0057] Example: "Your purpose for visiting our store is to purchase a 50-inch television. Please see our recommended products below."

[0058] The user reviews the recommended product list and selects the products they are interested in.

[0059] Ending Phase

[0060] The user can choose to either view the recommended products and decide to purchase them, or answer the questions again.

[0061] Based on the user's selection, the device will either generate further questions or guide them through the purchase process.

[0062] As described above, the system can start with user input, go through AI analysis, and ultimately present recommended products. This system allows customers to efficiently reach their intended purpose for visiting the store, providing a smooth shopping experience.

[0063] The following describes the processing flow.

[0064] Step 1:

[0065] During system initialization, the server establishes connections between the product database and the question database and loads the necessary data. This prepares the information required for system operation.

[0066] Step 2:

[0067] The device displays an initial screen to the user and provides an input form for basic information. This is for entering information such as age, gender, and categories of interest.

[0068] Step 3:

[0069] The user enters their basic information into the input form displayed on this initial screen and clicks the submit button.

[0070] Step 4:

[0071] The terminal receives basic information entered by the user and sends that information to the server.

[0072] Step 5:

[0073] The server analyzes the received basic information and prepares to generate initial questions from the question database. Based on this analysis, it selects appropriate questions related to the user's interests.

[0074] Step 6:

[0075] The server sends the generated initial question to the terminal and displays it to the user.

[0076] Step 7:

[0077] The terminal displays an initial question received from the server to the user. For example, it might display, "What size television are you interested in?"

[0078] Step 8:

[0079] The user answers the displayed questions and sends those answers to the server via their device.

[0080] Step 9:

[0081] The server receives the user's response and stores it in the database. At the same time, it either performs analysis to generate the next question or starts processing to specifically infer the purpose of the visit.

[0082] Step 10:

[0083] The server uses natural language processing (NLP) techniques to analyze user responses and infer user-specific intentions and reasons for visiting the store. For example, it might extract specific objectives such as "I want a 50-inch TV."

[0084] Step 11:

[0085] Based on the analysis results, the server generates a list of products suitable for the user from the product database. This list also includes detailed information about recommended products.

[0086] Step 12:

[0087] The server sends a list of the suspected purpose of the visit and recommended products to the terminal.

[0088] Step 13:

[0089] The terminal displays the user's estimated purpose for visiting the store and a list of recommended products. For example, it might say, "Your purpose for visiting the store is to purchase a 50-inch TV. Please see the recommended products below."

[0090] Step 14:

[0091] The user reviews the displayed list of recommended products and selects the products they are interested in.

[0092] Step 15:

[0093] The user clicks on the selected product to view details. They then choose to proceed with the purchase or answer further questions.

[0094] Step 16:

[0095] The device will determine the next action based on the user's selection. If the user chooses to answer the question again, it will send a request to the server and restart the question generation process. If the user chooses to proceed with the purchase, it will display the appropriate purchase process screen.

[0096] Step 17:

[0097] The server generates new questions and performs processes related to the purchase procedure, returning appropriate feedback based on the user's choices. This completes the entire process.

[0098] Through these steps, the system can efficiently understand the user's purpose for visiting the store and suggest appropriate products.

[0099] (Example 1)

[0100] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0101] In modern consumer electronics stores, quickly and accurately understanding a customer's purpose for visiting is crucial for providing a smooth shopping experience. However, existing systems struggle to efficiently carry out the entire process from user input to question generation, response analysis, intent inference, and product recommendation. This invention aims to solve these problems and provide a system that presents users with appropriate questions, quickly and accurately infers their purpose for visiting, and recommends appropriate products.

[0102] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0103] In this invention, the server includes an input means for inputting the user's basic information, a question generation means for generating an initial question based on the basic information received from the input means, a display means for displaying the generated question to the user, an answer receiving means for receiving the user's answer, an intent analysis means for analyzing the answer received from the answer receiving means and inferring the user's purpose for visiting the store, a product recommendation means for recommending appropriate products based on the purpose of visiting the store inferred by the intent analysis means, a display means for displaying the recommended products to the user, a communication means for performing question generation, answer reception, intent analysis, product recommendation, and display via a terminal and the server, and a next question generation means for generating the next question that the user is likely to be interested in based on the generated list of recommended products. This makes it possible to efficiently infer the purpose of visiting the store from the user's basic information and answers and to quickly recommend appropriate products.

[0104] "User basic information" refers to personal information that users enter, such as age, gender, and categories of interest.

[0105] An "input means" is a device or interface for a user to input basic information.

[0106] A "question generation means" is an algorithm or system that automatically generates appropriate questions based on the user's basic information.

[0107] A "display means" is an interface that visually presents generated questions and recommended products to the user.

[0108] A "response receiving system" is a system that receives responses entered by users and saves them in an appropriate data format.

[0109] A "means of intent analysis" is a system that uses natural language processing technology to analyze user responses and infer their purpose for visiting the store.

[0110] A "product recommendation system" is a system that selects and recommends appropriate products to users based on their purpose for visiting the store, as inferred by intent analysis.

[0111] "Communication methods" refer to network infrastructure and protocols used to send and receive data between a server and a terminal.

[0112] "Next question generation means" refers to an algorithm or system that generates the next appropriate question based on the user's basic information and previous answers.

[0113] A "recommended product list" is a list of products suitable for the user's purpose of visiting the store, and represents a group of product options presented to the user.

[0114] To implement the present invention, a system is needed that takes basic user information as input, uses AI to infer the purpose of the visit based on that information, and recommends appropriate products. This system involves cooperation between a server, a terminal, and the user, and includes the following main components and processes.

[0115] Hardware and software

[0116] Hardware used: A typical server device (e.g., Dell PowerEdge R740) and a terminal device for user operation (e.g., iPad® Pro).

[0117] Software used: Database management system (e.g., MySQL®), Web framework (e.g., Flask for Python), Natural Language Processing library (e.g., NLTK for Python), Frontend framework (e.g., React)

[0118] Specific description of the system's operation

[0119] The server first loads the product database and the question database, preparing the information needed when the system starts up. This includes executing SQL queries such as "SELECT FROM products" and "SELECT FROM questions," for example.

[0120] The device displays an initial screen where the user can enter basic information (age, gender, categories of interest, etc.). This initial screen is built using React and designed to be easy for the user to use.

[0121] The user enters their basic information into the form displayed on the initial screen. For example, the user might enter information such as age "30 years old," gender "female," and category of interest "home appliances."

[0122] The terminal receives the basic information entered by the user and sends it to the server via an HTTP POST request using the Axios library. Specifically, the terminal sends {"age": 30, "gender": "female", "category": "electronics"} in JSON format to "POST / api / userinfo".

[0123] The server analyzes the received basic information and selects an appropriate question from the question database. The technology used here is the Python natural language processing library NLTK. Based on the user's age (30 years old), gender (female), and category (home appliances), the server generates the question, "What size TV are you interested in?"

[0124] The generated question is sent from the server to the terminal, which then displays it to the user. The user then enters an answer to the displayed question. For example, the user might answer "50 inches".

[0125] The terminal receives the user's response and sends it back to the server using an HTTP POST request. The server stores the received response in its internal database and analyzes it using NLTK. This allows the server to infer the user's purpose for visiting the store, such as "the user is looking for a 50-inch television."

[0126] Based on the inferred purpose of the visit, the server selects appropriate products from the product database and generates a list of recommended products. This list is then sent back to the terminal and displayed to the user. Based on the displayed list of recommended products, the user can select products.

[0127] Example prompt statements

[0128] Prompt example: "The user is 30 years old, female, and interested in home electronics. Please generate the first question to infer their purpose for visiting."

[0129] As demonstrated here, the present invention can provide an efficient and smooth purchasing experience through a series of processes that begin with user input, go through server processing, and display the results on the terminal. This makes it possible to quickly identify the customer's purchase purpose and support them in selecting appropriate products.

[0130] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0131] System program processing flow

[0132] Step 1:

[0133] Initialization process

[0134] The server reads the product database and the question database, preparing the information needed when the system starts up. This includes executing SQL queries such as "SELECT FROM products" and "SELECT FROM questions" from the MySQL database to load all the data into memory.

[0135] Input: Database connection information

[0136] Output: Product data and question data loaded into memory

[0137] Step 2:

[0138] Initial screen display

[0139] The device displays an initial screen for the user to enter basic information. This screen is designed using JavaScript® and the React framework.

[0140] Input: UI design information

[0141] Output: Screen displaying an input form to the user

[0142] Step 3:

[0143] Entering basic information

[0144] The user enters their basic information (e.g., age "30", gender "female", category of interest "home appliances") into the form displayed on the initial screen.

[0145] Input: User's basic information

[0146] Output: Basic information data sent to the terminal

[0147] Step 4:

[0148] Receiving and sending basic information

[0149] The terminal receives the basic information entered by the user and sends it to the server using an HTTP POST request. The Axios library is used to send the data in JSON format to "POST / api / userinfo".

[0150] Input: User's basic information

[0151] Output: Basic information sent to the server

[0152] Step 5:

[0153] question generation

[0154] The server analyzes the received basic information and selects a suitable question from the question database. Using Python's NLTK, it selects a question based on information such as "the user is 30 years old, female, and in the home appliance category."

[0155] Input: User's basic information

[0156] Output: Generated question (e.g., "What size TV screen are you interested in?")

[0157] Step 6:

[0158] Submitting and displaying questions

[0159] The server sends the generated question to the terminal.

[0160] The terminal displays the received question to the user.

[0161] Input: Generated Question

[0162] Output: Questions displayed to the user

[0163] Step 7:

[0164] User response input

[0165] The user enters their answers to the questions displayed on the device.

[0166] Input: User's response (e.g., "50 inches")

[0167] Output: Response data sent to the terminal

[0168] Step 8:

[0169] Receiving and sending responses

[0170] The terminal receives the user's response and sends it to the server using an HTTP POST request.

[0171] Input: User's response

[0172] Output: Response sent to the server

[0173] Step 9:

[0174] Saving and analyzing responses

[0175] The server stores the received responses in an internal database and then uses NLTK again to analyze and infer the user's intent.

[0176] Input: User's response

[0177] Output: Estimated purpose of visit (e.g., "The user is looking for a 50-inch TV")

[0178] Step 10:

[0179] Product Recommendation

[0180] The server selects appropriate products from the product database based on the analysis results and generates a list of recommended products. The SQL query "SELECT FROM products WHERE category='TV' AND size='50 inches'" is executed.

[0181] Input: Estimated purpose of visit

[0182] Output: Recommended product list

[0183] Step 11:

[0184] Sending and displaying recommended products

[0185] The server sends a list of recommended products and the presumed purpose of the visit to the terminal.

[0186] The device displays a list of recommended products to the user.

[0187] Input: Recommended Conlist

[0188] Output: Recommended product list displayed to the user

[0189] Step 12:

[0190] User choices and next actions

[0191] The user reviews the recommended product list and can choose to select products they are interested in, or choose to answer the questions again.

[0192] Input: User Selection

[0193] Output: Further questions or instructions for the purchase process

[0194] Step 13:

[0195] Continue with your question or purchase

[0196] Based on the user's selection, the device will either generate further questions or guide them through the purchase process.

[0197] Input: User Selection

[0198] Output: Display of the next question or guidance to the purchase page

[0199] This makes the entire process, starting with user input, going through server analysis, and displaying the results on the terminal, concretely clear.

[0200] (Application Example 1)

[0201] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0202] Traditional customer service systems in physical stores have struggled to accurately understand customers' purchase objectives and recommend appropriate products. This often resulted in an inefficient and smooth customer purchasing experience, potentially leading to decreased customer satisfaction. Furthermore, conventional systems sometimes failed to fully utilize basic customer information and past history, resulting in inappropriate question generation. To address these issues, this invention provides a system that uses AI to accurately predict customers' purchase objectives and recommend the most suitable products.

[0203] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0204] In this invention, the server includes an input means for inputting basic user information, a question generation means for generating initial questions based on the basic information received from the input means, a display means for displaying the generated questions to the user, an answer receiving means for receiving the user's answers, an intent analysis means for analyzing the answers received from the answer receiving means and inferring the user's purpose for visiting the store, a product recommendation means for recommending appropriate products based on the purpose of visiting the store inferred by the intent analysis means, a display means for displaying the recommended products to the user, an initialization means for obtaining a product database and a question database from the server via a specially designed computer program, and a recommendation means for analyzing the user's answers using natural language processing technology and recommending products based on the results. This makes it possible to accurately infer the customer's purchase purpose and efficiently recommend appropriate products.

[0205] "User" refers to consumers or users who utilize a system.

[0206] "Basic information" refers to information necessary for generating initial questions, such as the user's age, gender, and areas of interest.

[0207] "Input means" refers to devices or interfaces that users use to input basic information.

[0208] "Question generation means" refers to a system or algorithm that has the function of generating the next question to be displayed based on the user's basic information.

[0209] "Display means" refers to displays, screens, or interfaces used to show generated questions and recommended products to the user.

[0210] "Response receiving means" refers to the device or system used to receive responses entered by the user.

[0211] "Intention analysis methods" refer to natural language processing techniques and algorithms used to analyze user responses and infer their purchase intentions.

[0212] "Product recommendation methods" refer to systems and algorithms that select appropriate products based on the purpose of a customer's visit, as inferred by intent analysis methods, and recommend them to the user.

[0213] "Initialization means" refers to the function of retrieving the product database and question database from the server via a specially designed computer program.

[0214] "Recommendation methods" refer to a function that uses natural language processing technology to analyze user responses and recommends products based on the results.

[0215] A "server" refers to a central control unit that manages and processes data for the entire system.

[0216] A "product database" refers to a collection of data that stores information about products that are sold.

[0217] A "question database" refers to a collection of data that stores questions to be displayed to users.

[0218] This invention is a system that uses AI to predict customers' purchase intentions in physical stores and efficiently recommends appropriate products. Specific embodiments of this invention are shown below.

[0219] Hardware and software to be used

[0220] Server: A central control unit that manages and processes data for the entire system. It manages the product database and the inquiry database.

[0221] Device: Uses a smartphone to interact with users. This includes displaying questions, receiving answers, and showing recommended products.

[0222] software:

[0223] Python: Execute a script.

[0224] requests: Sends HTTP requests and communicates with the API.

[0225] scikit-learn: NLP analysis and database search.

[0226] Natural Language Processing (NLP): Used to analyze user responses and infer their purpose for visiting the store.

[0227] System configuration and operation

[0228] Initialization

[0229] The server retrieves and initializes the product database and question database from the server. This process utilizes the server API via a specially designed computer program.

[0230] Entering user information

[0231] The device provides an interface for users to input basic information (age, gender, and categories of interest). Users input this information through their smartphone's UI.

[0232] question generation

[0233] The server generates an initial question based on the basic information entered. This question is selected from a question database and displayed on the terminal. For example, a question such as "What size television are you interested in?" might be generated.

[0234] Receiving and analyzing responses

[0235] The user answers questions displayed on the device, and the device sends those answers to the server. The server analyzes the received answers and uses natural language processing techniques to infer their intent. For example, the answer "I want a 50-inch TV" is analyzed.

[0236] Product Recommendation

[0237] Based on the intent analysis results, the server selects appropriate products from the product database and generates a list of recommended products. This list of recommended products is displayed to the user via their terminal. For example, it might say, "Your purpose for visiting the store is to purchase a 50-inch television. Please see the recommended products below."

[0238] Display and select from the recommended product list

[0239] The terminal displays a list of recommended products to the user, who reviews them and selects products of interest. The server then generates additional questions as needed and makes further recommendations.

[0240] Examples of specific prompt messages

[0241] User's basic information: Age 30, Gender Male, Interests: Television

[0242] Initial question: What size television screen are you interested in?

[0243] User reply: I want a 50-inch TV.

[0244] AI analysis results: The recommended product is a 50-inch television, "BRAND A 50-inch".

[0245] Based on the above, the present invention provides a system that can accurately predict the purchase purpose of a store visitor and efficiently recommend appropriate products. This makes it possible to improve customer satisfaction and achieve a smooth purchasing experience.

[0246] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0247] Step 1:

[0248] The server loads and initializes the product database and the question database. Specifically, the server retrieves each database via the server API and loads it into memory. At this time, it sends an HTTP request using a specially designed computer program to download the database files. The database format is JSON or CSV, and the loaded data is stored in internal data storage.

[0249] Input: Database URL obtained from the server API

[0250] Output: Product database and question database

[0251] Step 2:

[0252] The device displays an interface for the user to enter basic information. Specifically, a basic information input form (e.g., age, gender, categories of interest) is displayed on the smartphone screen. The user enters their information into these fields and presses the submit button.

[0253] Input: Smartphone basic information input form

[0254] Output: Basic information entered by the user

[0255] Step 3:

[0256] The server receives basic user information sent from the terminal and generates an initial question. The server retrieves an appropriate question from the question database and customizes it based on the user information. Natural language generation technology may be used in this process. For example, if the user is "30 years old and their interest category is television," the server will generate the question, "What size television are you interested in?"

[0257] Input: Basic information sent from the device

[0258] Output: Initial Question

[0259] Step 4:

[0260] The device displays the generated initial question to the user. Specifically, the question text and answer input fields are displayed on the smartphone screen. The user enters their answer to the question and presses the submit button.

[0261] Input: Initial question received from the server

[0262] Output: User's response

[0263] Step 5:

[0264] The server receives user responses sent from the terminal and analyzes them. Natural language processing techniques are used for intent analysis, and the response content is vectorized for analysis. As a result of the analysis, the user's purpose for visiting the store is identified. For example, from the response "I want a 50-inch TV," the intention "to purchase a 50-inch TV" is inferred.

[0265] Input: User's response sent from the device

[0266] Output: Results of identifying the purpose of visit

[0267] Step 6:

[0268] The server selects appropriate products from the product database based on the intent analysis results. Machine learning algorithms are used to search the product database and find the best product for the user's needs. For example, a list of products suitable for a "50-inch television" is generated.

[0269] Input: Results of intent analysis

[0270] Output: Recommended product list

[0271] Step 7:

[0272] The device displays a list of recommended products to the user. Specifically, a list of recommended products is displayed on the smartphone screen. The user can select a product they are interested in from this list and view detailed information.

[0273] Input: Recommended product list received from the server

[0274] Output: Recommended product list displayed on the smartphone screen

[0275] Step 8:

[0276] The user checks the recommended product list and selects whether to determine the product to purchase or answer the question again. The terminal executes the following operations based on the user's selection. If the user answers the question again, the processing after step 3 is executed again.

[0277] Input: User's selection (purchase decision or re-questioning)

[0278] Output: Generation of purchase procedure or next question

[0279] Furthermore, an emotion engine for estimating the user's emotion may be combined. That is, the specific processing unit 290 may estimate the user's emotion using the emotion specific model 59 and perform specific processing using the user's emotion.

[0280] As a form for implementing the present invention, a system for guessing why a customer who comes to a home appliance mass retailer comes to the store using AI and an emotion engine will be described. This system has a process of inputting the user's basic information, generating questions based on it, and analyzing the user's answers and emotions to infer the purpose of coming to the store.

[0281] Initialization process

[0282] At the time of initializing the system, the server establishes a connection to the product database and the question database and reads the necessary data. This also includes the initial settings of the emotion engine.

[0283] The terminal displays an initial screen so that the user can input basic information. This screen also has an interface for acquiring emotions, such as voice input and a face recognition camera.

[0284] The user inputs basic information (e.g., age, gender, category of interest) on the displayed initial screen.

[0285] User information input

[0286] The terminal receives the basic information input by the user and sends it to the server. It also obtains the user's emotion data from voice input and face recognition and sends this as well to the server.

[0287] The server analyzes the received basic information and emotion data and prepares to generate initial questions from the question database.

[0288] Question generation

[0289] Based on the basic information and emotion data, the server selects suitable questions from the question database and generates questions for display to the user.

[0290] Example: "What size of TV are you interested in?"

[0291] The terminal displays the generated questions to the user.

[0292] The user answers the questions and sends the answers to the server via the terminal. The user's expression and voice during answering are also obtained as emotion data.

[0293] Receiving and analyzing the answers

[0294] The server receives the user's answers and stores them in the internal database. Furthermore, it analyzes the user's emotion during answering using the emotion engine.

[0295] The server analyzes the answer content and emotion data and starts the process of generating the next question or specifically inferring the purpose of the visit to the store.

[0296] Intention analysis

[0297] The server uses natural language processing (NLP) technology and the emotion engine to analyze the user's answers and emotion data, and infer the user's specific intentions and the purpose of the visit to the store. For example, it extracts the specific purpose of "wanting a 50-inch TV" together with the emotion data.

[0298] Based on the analysis results, the server generates a list of products suitable for the user from the product database. This includes detailed information on the recommended products.

[0299] Display of Recommendation Results

[0300] The server sends the inferred purpose of the visit and the list of recommended products to the terminal.

[0301] The terminal displays the inferred purpose of the visit and the list of recommended products to the user. For example, it is displayed as "Your purpose of visit is 'Purchase of a 50-inch TV'. Please view the following recommended products."

[0302] The user checks the displayed list of recommended products and selects the products of interest.

[0303] Utilization of Emotional Data

[0304] The server also utilizes the emotion engine while displaying the recommended products and continuously collects emotional data obtained from the user's expressions and voices. This enables a more refined understanding of which products the user is most interested in.

[0305] Based on the emotional data, the server can make optimal additional proposals to the user.

[0306] End Phase

[0307] The user decides whether to purchase after viewing the recommended products or makes a choice to answer the questions again.

[0308] Based on the user's selection, the terminal generates questions again or guides the purchase procedure.

[0309] Specific Example

[0310] For example, if a user answers "50 inches" as the TV size they are interested in, the emotion engine can analyze the user's facial expressions and tone of voice during the response to determine if they are truly interested. As a result, if they show a high level of interest, "50-inch TVs" will be strongly reflected in the recommendation list; if they show a low level of interest, TVs of other sizes or with different characteristics can be recommended again.

[0311] By incorporating user sentiment data in this way, it becomes possible to make more accurate product recommendations and increase user satisfaction.

[0312] The following describes the processing flow.

[0313] Step 1:

[0314] During system initialization, the server establishes connections to the product database and the question database and loads the necessary data. This includes the initial setup of the emotion engine.

[0315] Step 2:

[0316] The device displays an initial screen that allows the user to enter basic information. It also simultaneously displays interfaces for voice input and a facial recognition camera to acquire emotional data.

[0317] Step 3:

[0318] On the initial screen displayed, the user enters basic information (e.g., age, gender, categories of interest). Simultaneously, facial expressions and voice tone are captured through the camera and microphone built into the device.

[0319] Step 4:

[0320] The device receives basic information, facial expression data, and voice data, and sends them to the server.

[0321] Step 5:

[0322] The server analyzes the received basic information and emotional data, and prepares to generate initial questions from the question database. This includes a process that evaluates the user's psychological state based on their facial expressions and tone of voice.

[0323] Step 6:

[0324] The server generates appropriate initial questions based on basic information and analysis results, and sends them to the terminal. Example: "What size television are you interested in?"

[0325] Step 7:

[0326] The terminal displays the questions received from the server to the user.

[0327] Step 8:

[0328] The user answers questions. During this process, the device's camera captures the user's facial expressions, and the microphone records their voice tone.

[0329] Step 9:

[0330] The device sends the user's response along with newly acquired emotional data (facial expressions and voice) to the server.

[0331] Step 10:

[0332] The server receives user responses and sentiment data and stores them in a database. Simultaneously, it generates the next question or begins analysis to specifically infer the user's purpose for visiting the store.

[0333] Step 11:

[0334] The server uses natural language processing (NLP) techniques and an emotion engine to analyze the user's responses and emotional data. For example, if a user responds "A 50-inch TV is interesting" while smiling, the server infers their intention as "They have a strong interest in 50-inch TVs."

[0335] Step 12:

[0336] Based on the analysis results, the server generates a list of products suitable for the user from the product database. The generated list also takes into account the user's emotional state.

[0337] Step 13:

[0338] The server sends a list of the suspected purpose of visit and recommended products to the terminal.

[0339] Step 14:

[0340] The terminal displays the user's estimated purpose for visiting the store and a list of recommended products. Example: "Your purpose for visiting the store is to purchase a 50-inch TV. Please see the recommended products below."

[0341] Step 15:

[0342] The user reviews the displayed list of recommended products and selects items that interest them. At this point, the device continues to monitor the user's facial expressions and voice, collecting emotional data.

[0343] Step 16:

[0344] Based on the user's choices, the server analyzes further sentiment data to generate the next question or proceed with the purchase process.

[0345] Step 17:

[0346] Depending on the user's choice, the device will either display the question again or show a screen for the purchase process. If the user chooses to proceed with the purchase, the necessary options and input fields will be displayed.

[0347] Step 18:

[0348] The user proceeds with the purchase process, and the purchase is completed. During the purchase process, the device collects emotional data, allowing for monitoring of the user's satisfaction level.

[0349] This enables more accurate product recommendations that take user emotions into account, resulting in a more personalized and smoother purchasing experience throughout the entire process.

[0350] (Example 2)

[0351] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0352] Conventional user information collection systems generate questions and perform intent analysis based on basic user information and response history, but this alone makes it difficult to accurately grasp the user's true intentions and emotions. As a result, appropriate product recommendations may not be made, potentially lowering user satisfaction.

[0353] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a question generation means that generates an initial question based on the user's basic information and emotional data, an answer receiving means that receives the user's answer and emotional data, and an intent analysis means that analyzes the answer and emotional data to infer the user's purpose for visiting the store. By incorporating the user's emotional data into the analysis, it becomes possible to accurately grasp the user's intentions and emotions and make appropriate product suggestions.

[0354] "Basic information" refers to identifying information such as the user's age, gender, and categories of interest.

[0355] "Emotional data" refers to information about a user's emotions, obtained from their facial expressions, voice tone, and other sources.

[0356] A "question generation means" is a means for generating initial and subsequent questions based on the user's basic information and sentiment data.

[0357] "Display means" refers to an interface for displaying questions generated for the user and recommended products.

[0358] "Response receiving means" refers to a means of receiving user responses and sentiment data.

[0359] "Intention analysis methods" are techniques that analyze received responses and emotional data to infer the user's purpose for visiting the store.

[0360] A "product recommendation method" is a means of recommending appropriate products based on the purpose of the customer's visit, which is inferred by intent analysis methods.

[0361] "Natural language processing technology" is a technology that analyzes a user's language response and understands its meaning.

[0362] An "emotion engine" is software or a system used to analyze a user's emotional data.

[0363] This invention relates to a system that uses AI and an emotion engine to analyze the purpose of a user's visit to a consumer electronics store and recommend appropriate products. Based on the basic information and emotion data entered by the user, this system generates questions, receives the answers, analyzes the answers and emotion data to clarify the user's intentions, and makes optimal product suggestions.

[0364] System Configuration

[0365] 1. The server establishes connections to the product database and the question database during system initialization and loads the necessary data. Specifically, it uses MySQL as the database management system and IBM Watson® Emotion Recognition as the sentiment analysis engine. This establishes the initial settings of the engine and the database connection.

[0366] 2. The device displays a screen for the user to enter basic information. This screen includes interfaces that utilize voice input (Google® Speech-to-Text API) and a facial recognition camera (Azure® Face API). This makes it possible to obtain emotional data from the user's facial expressions and voice.

[0367] 3. The user enters basic information (age, gender, categories of interest, etc.) on the initial screen that is displayed. The entered basic information is sent to the server by the device.

[0368] 4. The device transmits not only basic information but also emotional data obtained from voice input and facial recognition to the server.

[0369] 5. The server analyzes the received basic information and sentiment data to generate an initial question from the question database. For example, a specific question such as "What size television are you interested in?" is generated.

[0370] 6. The terminal displays questions sent from the server to the user. The user answers the displayed questions. When answering, facial expressions and voice are also captured through the terminal and simultaneously sent to the server as emotion data.

[0371] 7. The server stores the received responses in a database and uses an emotion engine to analyze the user's emotions at the time of the response. This allows the server to infer the user's specific intentions and purpose for visiting the store.

[0372] 8. Based on the analysis results, the server generates an appropriate product list from the product database. For example, if the intention "I want a 50-inch TV" is extracted, a product list specifically for 50-inch TVs will be generated.

[0373] 9. The terminal displays a list of recommended products sent from the server to the user. Sentimental data continues to be collected even after the display, making it possible to understand more precisely which products the user is interested in.

[0374] 10. The user reviews the displayed list of recommended products and selects items of interest. In some cases, they may answer further questions or proceed with the purchase.

[0375] Specific example

[0376] For example, if a user enters basic information such as "30 years old, male, interested in home appliances" and answers the question "I want a 50-inch TV," the server uses an emotion engine to analyze the user's facial expressions and tone of voice during the response. Based on this analysis, if the server determines that the user is genuinely very interested in a 50-inch TV, a 50-inch TV will be strongly reflected in the recommendation list. If the user shows little interest, a TV of a different size or with different characteristics will be recommended instead.

[0377] Example of a prompt

[0378] An example of a prompt to input into a generative AI model is: "A user showed interest in a 50-inch TV when they visited the store. Please explain the process for recommending products that match this user's interests."

[0379] By incorporating user emotional data into the analysis in this way, it becomes possible to provide a system that makes more accurate and appropriate product recommendations.

[0380] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0381] Step 1:

[0382] The server establishes connections to the product database and the question database during system initialization. Specifically, it uses a database management system (e.g., MySQL) to load the necessary data (e.g., product characteristics, question templates). It also performs the initial setup of the sentiment analysis engine (e.g., IBM Watson Emotion Recognition). The input is the database connection information and the sentiment engine configuration information, and the output is the connected database and the initialized sentiment engine.

[0383] Step 2:

[0384] The device displays an initial screen for the user to enter basic information. This screen also includes voice input (e.g., Google Speech-to-Text API) and a facial recognition camera (e.g., Azure Face API), providing an interface for acquiring emotional data from the user's facial expressions and voice. Input consists of basic information provided by the user (e.g., age, gender, categories of interest), and output consists of the user's basic information and the display of the initial screen.

[0385] Step 3:

[0386] The user enters basic information on the initial screen. This information includes age, gender, and categories of interest. The input is basic information, and the output is basic information sent to the device.

[0387] Step 4:

[0388] The terminal receives basic information entered by the user and sends it to the server. It also simultaneously collects emotional data obtained from voice input and facial recognition and sends this data to the server as well. The input consists of the user's basic information and emotional data, while the output is the data sent to the server.

[0389] Step 5:

[0390] The server analyzes the received basic information and sentiment data and generates an initial question from the question database. Specifically, it selects an appropriate question based on the received basic information and sentiment data, and generates a concrete question such as, "What size television are you interested in?" The input is basic information and sentiment data, and the output is the generated question.

[0391] Step 6:

[0392] The terminal displays the initial question sent from the server to the user. The input is the initial question from the server, and the output is the display of the question to the user.

[0393] Step 7:

[0394] The user answers the displayed questions. During the answering process, facial expressions and voice are also collected through a facial recognition camera and voice input function. Input consists of the questions, user answers, facial expressions, and voice data, while output consists of the answers and emotion data sent to the device.

[0395] Step 8:

[0396] The terminal receives user responses and sentiment data and sends them to the server. The input is the user's responses and sentiment data, and the output is the data sent to the server.

[0397] Step 9:

[0398] The server stores user responses in a database and uses an emotion engine to analyze the user's emotions at the time of their response. The input consists of the user's response and emotion data, while the output consists of the stored data and the analysis results.

[0399] Step 10:

[0400] The server analyzes the user's intent based on the response content and sentiment data using natural language processing technology (e.g., Google Natural Language API) and a sentiment engine. For example, it extracts a specific objective, such as "I want a 50-inch TV," along with sentiment data. The input is the user's response and sentiment data, and the output is the analyzed user intent.

[0401] Step 11:

[0402] The server generates a list of appropriate products from the product database based on the analysis results. The input is the analysis results, and the output is the generated product list.

[0403] Step 12:

[0404] The server sends the generated product list to the terminal. The input is the generated product list, and the output is the data sent to the terminal.

[0405] Step 13:

[0406] The terminal displays a list of recommended products to the user. For example, it might display, "Your purpose for visiting is to purchase a 50-inch television. Please see the recommended products below." The input is a list of recommended products from the server, and the output is the displayed product list.

[0407] Step 14:

[0408] Even while displaying recommended products, the device utilizes an emotion engine to continuously collect emotional data obtained from the user's facial expressions and voice, and transmits it to the server. The input is the user's facial expressions and voice data, and the output is the data transmitted to the server.

[0409] Step 15:

[0410] The server provides users with optimal additional suggestions based on collected sentiment data. The input is continuously collected sentiment data, and the output is data for additional suggestions.

[0411] Step 16:

[0412] The user can choose to view the recommended products and decide whether to purchase them, or to answer another question. The input is a list of recommended products or a new question, and the output is the user's choice or answer.

[0413] Step 17:

[0414] The terminal either generates new questions or guides the user through the purchase process based on their selection. The input is the user's selection, and the output is a new question or a purchase process screen.

[0415] (Application Example 2)

[0416] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0417] Existing consumer electronics retailers face challenges in accurately understanding what customers want to buy, hindering efficient product recommendations. Furthermore, recommending products without understanding user emotions or specific intentions can lead to decreased customer satisfaction. Moreover, with the rise of online shopping, there's a need to enhance the unique customer experience offered by physical stores. To address these challenges, a system is needed that utilizes not only basic customer information but also emotional data to accurately predict purchasing intent and provide precise product recommendations.

[0418] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes an input means for inputting the user's basic information, a question generation means for generating initial questions based on the basic information, a display means for displaying the generated questions to the user, an answer receiving means for receiving the user's answers and emotional data, an intent analysis means for analyzing the answers and emotional data to infer the purpose of the visit, a product recommendation means for recommending appropriate products based on the inferred purpose of the visit, and a display means for displaying the recommended products. This makes it possible to accurately grasp the user's emotions and intentions and make product suggestions that match the purpose of the visit, thereby improving customer satisfaction.

[0419] A "user" is a customer who visits an electronics retail store and uses the system.

[0420] "Basic information" refers to data that users enter, such as age, gender, and categories of interest.

[0421] "Input method" refers to an interface that allows users to input basic information, and includes devices such as smartphones and smart glasses.

[0422] "Question generation method" refers to the process of creating appropriate questions based on the basic information entered by the user.

[0423] "Display means" refers to displays or screens used to present generated questions and recommended products to the user.

[0424] A "response receiving method" refers to an interface that receives responses and sentiment data entered by users in response to questions.

[0425] "Emotional data" refers to emotional information obtained from the user's facial expressions, voice, and other sources.

[0426] "Intention analysis means" refers to a function that analyzes user responses and emotional data to infer the user's purpose for visiting the store.

[0427] "Natural language processing technology" refers to artificial intelligence technology used to understand and interpret user text data.

[0428] An "emotion engine" refers to an algorithm and software that analyzes a user's emotions from their facial expressions and voice.

[0429] "Product recommendation methods" refer to the process of selecting and recommending appropriate products based on the user's purpose for visiting the store.

[0430] A "server" is a central device that manages the entire system and performs tasks such as data analysis, storage, and query generation.

[0431] To implement the present invention, terminals such as smartphones and smart glasses, and servers connected to them, are used in consumer electronics stores. The main components of the system include input means for inputting basic user information, question generation means for generating questions, display means for displaying the generated questions, answer receiving means for receiving user answers and sentiment data, intent analysis means for analyzing user answers and sentiment data, product recommendation means for recommending appropriate products, and display means for displaying recommended products.

[0432] The server connects to the product database and question database during system initialization and performs the initial setup of the emotion engine. It uses "EmotionEngine" as the emotion engine and "NlpProcessor" as the natural language processing technology. The server generates questions based on the user's basic information and past answer history and sends them to the terminal.

[0433] The device takes the form of a smartphone or smart glasses and displays an initial screen for the user to input basic information. The device also includes features such as voice input and a facial recognition camera, which are used to collect emotional data. The basic information entered by the user is sent to a server, which then generates appropriate initial questions based on this information.

[0434] Users respond to questions displayed on their device using voice or touch input. The device sends these responses to a server, which analyzes the emotional data acquired along with the responses. By using an emotion engine to read the user's emotions from their facial expressions and voice, and by analyzing the content of their responses using NLP technology, the server infers the user's purpose for visiting the store.

[0435] The server selects appropriate products from its product database based on the inferred purpose of the visit, generates a list of recommended products, and sends it to the terminal. The terminal displays this list to the user, who then selects products that interest them. Throughout this process, the server continuously collects sentiment data and provides recommendations tailored to the user's preferences.

[0436] For example, when a user responds that they are "looking for a large refrigerator," the system analyzes the user's facial expressions and tone of voice to estimate whether their interest is genuine. Based on these results, "refrigerators of 500 liters or more" are strongly reflected in the recommendation list.

[0437] Examples of prompt statements used as input to a generative AI model are as follows:

[0438] Please generate questions to infer the user's purpose of purchase. The basic user information is as follows:

[0439] Age: 35

[0440] Gender: Male

[0441] Categories of interest: Home appliances

[0442] Please generate questions based on the following data:

[0443] Emotional data: Joy (high)

[0444] Example questions to output:

[0445] "What size refrigerator in liters are you interested in?"

[0446] This makes it possible to accurately understand the user's emotions and intentions, and by suggesting products that match their purpose for visiting the store, customer satisfaction can be improved.

[0447] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0448] Step 1:

[0449] During system initialization, the server establishes connections to the product database and question database and performs initial setup of the sentiment engine. At this time, the server reads product information and question templates and sets initial parameters for sentiment analysis.

[0450] Input: Product database, question database, emotion engine configuration information

[0451] Output: System state after initial setup is complete

[0452] Specific actions: Connecting to the database, loading configuration files.

[0453] Step 2:

[0454] The device displays an initial screen where the user can enter basic information. The device is also equipped with voice input and a facial recognition camera, providing an interface to acquire user emotion data.

[0455] Input: System initial settings

[0456] Output: Basic Information Input Screen

[0457] Specific actions: rendering the initial screen, voice input, and camera activation.

[0458] Step 3:

[0459] The user enters basic information (age, gender, and categories of interest) using the initial screen displayed. The device also acquires emotional data through voice input and facial recognition.

[0460] Input: User's basic information, voice and facial recognition data

[0461] Output: User basic information, sentiment data

[0462] Specific operations: Collection of user input data, real-time acquisition of sentiment data.

[0463] Step 4:

[0464] The terminal analyzes the collected basic information and emotional data and sends it to the server. The server receives this data and prepares it for analysis.

[0465] Input: User's basic information, sentiment data

[0466] Output: Data ready for analysis

[0467] Specific actions: Packing input data and sending it to the server.

[0468] Step 5:

[0469] The server generates an initial question from the question database based on the user's basic information and sentiment data.

[0470] Input: User's basic information, sentiment data

[0471] Output: Initial Question

[0472] Specific actions: Execute database queries, apply query templates.

[0473] Step 6:

[0474] The device displays the generated questions to the user. The user enters answers to the questions and expresses emotions through voice and facial expressions.

[0475] Input: Initial Question

[0476] Output: User responses, sentiment data

[0477] Specific actions: Displaying questions, providing an interface for entering answers.

[0478] Step 7:

[0479] The device sends the user's responses and sentiment data to the server, which receives, stores, and then analyzes this data.

[0480] Input: User responses, sentiment data

[0481] Output: Analysis results

[0482] Specific actions: Packing data, sending it to the server, and saving it to the database.

[0483] Step 8:

[0484] The server uses natural language processing technology and an emotion engine to analyze the user's responses and emotion data, and to infer the user's purpose for visiting the store.

[0485] Input: User responses, sentiment data

[0486] Output: Estimated results of the purpose of visit

[0487] Specific actions: Application of NLP algorithms, performance of sentiment analysis.

[0488] Step 9:

[0489] Based on the inferred purpose of the customer's visit, the server selects appropriate products from the product database and generates a list of recommended products.

[0490] Input: Estimated result of the purpose of visit

[0491] Output: Recommended product list

[0492] Specific actions: Execute database queries, select recommended products.

[0493] Step 10:

[0494] The terminal displays a list of recommended products to the user, who then selects products of interest. The server continuously collects sentiment data and provides optimal additional suggestions.

[0495] Input: Recommended product list, user selection

[0496] Output: Optimal product proposal

[0497] Specific actions: Rendering of recommended products, real-time analysis of sentiment data.

[0498] Step 11:

[0499] When a user selects a product or answers additional questions, the device sends data back to the server, which then asks the next most appropriate question or suggests a product as needed.

[0500] Input: User selection, answers to additional questions

[0501] Output: Next question or product suggestion

[0502] Specific actions: data retransmission, reapplication of analysis algorithms.

[0503] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0504] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0505] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0506] [Second Embodiment]

[0507] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0508] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0509] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0510] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0511] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0512] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0513] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0514] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0515] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0516] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0517] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0518] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0519] As an embodiment of the present invention, a system that uses AI to predict the purchase purpose of a customer visiting a consumer electronics store will be described. This system has a process of taking the user's basic information as input, generating questions based on that information, and analyzing the user's answers to infer the purpose of the visit.

[0520] Initialization process

[0521] The server reads the product database and the question database, and prepares the necessary information in advance when the system starts up.

[0522] The device displays an initial screen that allows the user to enter basic information.

[0523] The user enters basic information (e.g., age, gender, categories of interest) on the displayed screen.

[0524] User information entry

[0525] The terminal receives basic information entered by the user and sends it to the server.

[0526] The server analyzes the received basic information and prepares to generate appropriate initial questions.

[0527] question generation

[0528] Based on the basic information, the server selects a suitable question from the question database and generates a question to display to the user.

[0529] Example: "What size TV screen are you interested in?"

[0530] The terminal displays the generated question to the user.

[0531] The user answers the question and sends the answer to the server via their device.

[0532] Receiving and analyzing responses

[0533] The server receives the user's response and saves it to its internal database.

[0534] The server analyzes the user's responses and processes them to generate the next question or to specifically infer the purpose of their visit.

[0535] Intent Analysis

[0536] The server uses natural language processing (NLP) techniques to analyze user responses and infer the user's purpose for visiting the store.

[0537] Example: Analyze the intention behind the statement, "I want a 50-inch TV."

[0538] Based on the analysis results, the server selects appropriate products from the product database and generates a list of recommended products.

[0539] Displaying recommended results

[0540] The server sends a list of recommended products along with an estimated reason for the customer's visit to the terminal.

[0541] The terminal displays to the user a list of suggested reasons for visiting the store and recommended products.

[0542] Example: "Your purpose for visiting our store is to purchase a 50-inch television. Please see our recommended products below."

[0543] The user reviews the recommended product list and selects the products they are interested in.

[0544] Ending Phase

[0545] The user can choose to either view the recommended products and decide to purchase them, or answer the questions again.

[0546] Based on the user's selection, the device will either generate further questions or guide them through the purchase process.

[0547] As described above, the system can start with user input, go through AI analysis, and ultimately present recommended products. This system allows customers to efficiently reach their intended purpose for visiting the store, providing a smooth shopping experience.

[0548] The following describes the processing flow.

[0549] Step 1:

[0550] During system initialization, the server establishes connections between the product database and the question database and loads the necessary data. This prepares the information required for system operation.

[0551] Step 2:

[0552] The device displays an initial screen to the user and provides an input form for basic information. This is for entering information such as age, gender, and categories of interest.

[0553] Step 3:

[0554] The user enters their basic information into the input form displayed on this initial screen and clicks the submit button.

[0555] Step 4:

[0556] The terminal receives basic information entered by the user and sends that information to the server.

[0557] Step 5:

[0558] The server analyzes the received basic information and prepares to generate initial questions from the question database. Based on this analysis, it selects appropriate questions related to the user's interests.

[0559] Step 6:

[0560] The server sends the generated initial question to the terminal and displays it to the user.

[0561] Step 7:

[0562] The terminal displays an initial question received from the server to the user. For example, it might display, "What size television are you interested in?"

[0563] Step 8:

[0564] The user answers the displayed questions and sends those answers to the server via their device.

[0565] Step 9:

[0566] The server receives the user's response and stores it in the database. At the same time, it either performs analysis to generate the next question or starts processing to specifically infer the purpose of the visit.

[0567] Step 10:

[0568] The server uses natural language processing (NLP) techniques to analyze user responses and infer user-specific intentions and reasons for visiting the store. For example, it might extract specific objectives such as "I want a 50-inch TV."

[0569] Step 11:

[0570] Based on the analysis results, the server generates a list of products suitable for the user from the product database. This list also includes detailed information about recommended products.

[0571] Step 12:

[0572] The server sends a list of the suspected purpose of the visit and recommended products to the terminal.

[0573] Step 13:

[0574] The terminal displays the user's estimated purpose for visiting the store and a list of recommended products. For example, it might say, "Your purpose for visiting the store is to purchase a 50-inch TV. Please see the recommended products below."

[0575] Step 14:

[0576] The user reviews the displayed list of recommended products and selects the products they are interested in.

[0577] Step 15:

[0578] The user clicks on the selected product to view details. They then choose to proceed with the purchase or answer further questions.

[0579] Step 16:

[0580] The device will determine the next action based on the user's selection. If the user chooses to answer the question again, it will send a request to the server and restart the question generation process. If the user chooses to proceed with the purchase, it will display the appropriate purchase process screen.

[0581] Step 17:

[0582] The server generates new questions and performs processes related to the purchase procedure, returning appropriate feedback based on the user's choices. This completes the entire process.

[0583] Through these steps, the system can efficiently understand the user's purpose for visiting the store and suggest appropriate products.

[0584] (Example 1)

[0585] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0586] In modern consumer electronics stores, quickly and accurately understanding a customer's purpose for visiting is crucial for providing a smooth shopping experience. However, existing systems struggle to efficiently carry out the entire process from user input to question generation, response analysis, intent inference, and product recommendation. This invention aims to solve these problems and provide a system that presents users with appropriate questions, quickly and accurately infers their purpose for visiting, and recommends appropriate products.

[0587] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0588] In this invention, the server includes an input means for inputting the user's basic information, a question generation means for generating an initial question based on the basic information received from the input means, a display means for displaying the generated question to the user, an answer receiving means for receiving the user's answer, an intent analysis means for analyzing the answer received from the answer receiving means and inferring the user's purpose for visiting the store, a product recommendation means for recommending appropriate products based on the purpose of visiting the store inferred by the intent analysis means, a display means for displaying the recommended products to the user, a communication means for performing question generation, answer reception, intent analysis, product recommendation, and display via a terminal and the server, and a next question generation means for generating the next question that the user is likely to be interested in based on the generated list of recommended products. This makes it possible to efficiently infer the purpose of visiting the store from the user's basic information and answers and to quickly recommend appropriate products.

[0589] "User basic information" refers to personal information that users enter, such as age, gender, and categories of interest.

[0590] An "input means" is a device or interface for a user to input basic information.

[0591] A "question generation means" is an algorithm or system that automatically generates appropriate questions based on the user's basic information.

[0592] A "display means" is an interface that visually presents generated questions and recommended products to the user.

[0593] A "response receiving system" is a system that receives responses entered by users and saves them in an appropriate data format.

[0594] A "means of intent analysis" is a system that uses natural language processing technology to analyze user responses and infer their purpose for visiting the store.

[0595] A "product recommendation system" is a system that selects and recommends appropriate products to users based on their purpose for visiting the store, as inferred by intent analysis.

[0596] "Communication methods" refer to network infrastructure and protocols used to send and receive data between a server and a terminal.

[0597] "Next question generation means" refers to an algorithm or system that generates the next appropriate question based on the user's basic information and previous answers.

[0598] A "recommended product list" is a list of products suitable for the user's purpose of visiting the store, and represents a group of product options presented to the user.

[0599] To implement the present invention, a system is needed that takes basic user information as input, uses AI to infer the purpose of the visit based on that information, and recommends appropriate products. This system involves cooperation between a server, a terminal, and the user, and includes the following main components and processes.

[0600] Hardware and software

[0601] Hardware used: A typical server device (e.g., Dell PowerEdge R740) and a terminal device for user operation (e.g., iPad Pro)

[0602] Software used: Database management system (e.g., MySQL), Web framework (e.g., Flask for Python), Natural Language Processing library (e.g., NLTK for Python), Frontend framework (e.g., React)

[0603] Specific description of the system's operation

[0604] The server first loads the product database and the question database, preparing the information needed when the system starts up. This includes executing SQL queries such as "SELECT FROM products" and "SELECT FROM questions," for example.

[0605] The device displays an initial screen where the user can enter basic information (age, gender, categories of interest, etc.). This initial screen is built using React and designed to be easy for the user to use.

[0606] The user enters their basic information into the form displayed on the initial screen. For example, the user might enter information such as age "30 years old," gender "female," and category of interest "home appliances."

[0607] The terminal receives the basic information entered by the user and sends it to the server via an HTTP POST request using the Axios library. Specifically, the terminal sends {"age": 30, "gender": "female", "category": "electronics"} in JSON format to "POST / api / userinfo".

[0608] The server analyzes the received basic information and selects an appropriate question from the question database. The technology used here is the Python natural language processing library NLTK. Based on the user's age (30 years old), gender (female), and category (home appliances), the server generates the question, "What size TV are you interested in?"

[0609] The generated question is sent from the server to the terminal, which then displays it to the user. The user then enters an answer to the displayed question. For example, the user might answer "50 inches".

[0610] The terminal receives the user's response and sends it back to the server using an HTTP POST request. The server stores the received response in its internal database and analyzes it using NLTK. This allows the server to infer the user's purpose for visiting the store, such as "the user is looking for a 50-inch television."

[0611] Based on the inferred purpose of the visit, the server selects appropriate products from the product database and generates a list of recommended products. This list is then sent back to the terminal and displayed to the user. Based on the displayed list of recommended products, the user can select products.

[0612] Example prompt statements

[0613] Prompt example: "The user is 30 years old, female, and interested in home electronics. Please generate the first question to infer their purpose for visiting."

[0614] As demonstrated here, the present invention can provide an efficient and smooth purchasing experience through a series of processes that begin with user input, go through server processing, and display the results on the terminal. This makes it possible to quickly identify the customer's purchase purpose and support them in selecting appropriate products.

[0615] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0616] System program processing flow

[0617] Step 1:

[0618] Initialization process

[0619] The server reads the product database and the question database, preparing the information needed when the system starts up. This includes executing SQL queries such as "SELECT FROM products" and "SELECT FROM questions" from the MySQL database to load all the data into memory.

[0620] Input: Database connection information

[0621] Output: Product data and question data loaded into memory

[0622] Step 2:

[0623] Initial screen display

[0624] The device displays an initial screen for the user to enter basic information. This screen is designed using JavaScript and the React framework.

[0625] Input: UI design information

[0626] Output: Screen displaying an input form to the user

[0627] Step 3:

[0628] Entering basic information

[0629] The user enters their basic information (e.g., age "30", gender "female", category of interest "home appliances") into the form displayed on the initial screen.

[0630] Input: User's basic information

[0631] Output: Basic information data sent to the terminal

[0632] Step 4:

[0633] Receiving and sending basic information

[0634] The terminal receives the basic information entered by the user and sends it to the server using an HTTP POST request. The Axios library is used to send the data in JSON format to "POST / api / userinfo".

[0635] Input: User's basic information

[0636] Output: Basic information sent to the server

[0637] Step 5:

[0638] question generation

[0639] The server analyzes the received basic information and selects a suitable question from the question database. Using Python's NLTK, it selects a question based on information such as "the user is 30 years old, female, and in the home appliance category."

[0640] Input: User's basic information

[0641] Output: Generated question (e.g., "What size TV screen are you interested in?")

[0642] Step 6:

[0643] Submitting and displaying questions

[0644] The server sends the generated question to the terminal.

[0645] The terminal displays the received question to the user.

[0646] Input: Generated Question

[0647] Output: Questions displayed to the user

[0648] Step 7:

[0649] User response input

[0650] The user enters their answers to the questions displayed on the device.

[0651] Input: User's response (e.g., "50 inches")

[0652] Output: Response data sent to the terminal

[0653] Step 8:

[0654] Receiving and sending responses

[0655] The terminal receives the user's response and sends it to the server using an HTTP POST request.

[0656] Input: User's response

[0657] Output: Response sent to the server

[0658] Step 9:

[0659] Saving and analyzing responses

[0660] The server stores the received responses in an internal database and then uses NLTK again to analyze and infer the user's intent.

[0661] Input: User's response

[0662] Output: Estimated purpose of visit (e.g., "The user is looking for a 50-inch TV")

[0663] Step 10:

[0664] Product Recommendation

[0665] The server selects appropriate products from the product database based on the analysis results and generates a list of recommended products. The SQL query "SELECT FROM products WHERE category='TV' AND size='50 inches'" is executed.

[0666] Input: Estimated purpose of visit

[0667] Output: Recommended product list

[0668] Step 11:

[0669] Sending and displaying recommended products

[0670] The server sends a list of recommended products and the presumed purpose of the visit to the terminal.

[0671] The device displays a list of recommended products to the user.

[0672] Input: Recommended Conlist

[0673] Output: Recommended product list displayed to the user

[0674] Step 12:

[0675] User choices and next actions

[0676] The user reviews the recommended product list and can choose to select products they are interested in, or choose to answer the questions again.

[0677] Input: User Selection

[0678] Output: Further questions or instructions for the purchase process

[0679] Step 13:

[0680] Continue with your question or purchase

[0681] Based on the user's selection, the device will either generate further questions or guide them through the purchase process.

[0682] Input: User Selection

[0683] Output: Display of the next question or guidance to the purchase page

[0684] This makes the entire process, starting with user input, going through server analysis, and displaying the results on the terminal, concretely clear.

[0685] (Application Example 1)

[0686] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0687] Traditional customer service systems in physical stores have struggled to accurately understand customers' purchase objectives and recommend appropriate products. This often resulted in an inefficient and smooth customer purchasing experience, potentially leading to decreased customer satisfaction. Furthermore, conventional systems sometimes failed to fully utilize basic customer information and past history, resulting in inappropriate question generation. To address these issues, this invention provides a system that uses AI to accurately predict customers' purchase objectives and recommend the most suitable products.

[0688] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0689] In this invention, the server includes an input means for inputting basic user information, a question generation means for generating initial questions based on the basic information received from the input means, a display means for displaying the generated questions to the user, an answer receiving means for receiving the user's answers, an intent analysis means for analyzing the answers received from the answer receiving means and inferring the user's purpose for visiting the store, a product recommendation means for recommending appropriate products based on the purpose of visiting the store inferred by the intent analysis means, a display means for displaying the recommended products to the user, an initialization means for obtaining a product database and a question database from the server via a specially designed computer program, and a recommendation means for analyzing the user's answers using natural language processing technology and recommending products based on the results. This makes it possible to accurately infer the customer's purchase purpose and efficiently recommend appropriate products.

[0690] "User" refers to consumers or users who utilize a system.

[0691] "Basic information" refers to information necessary for generating initial questions, such as the user's age, gender, and areas of interest.

[0692] "Input means" refers to devices or interfaces that users use to input basic information.

[0693] "Question generation means" refers to a system or algorithm that has the function of generating the next question to be displayed based on the user's basic information.

[0694] "Display means" refers to displays, screens, or interfaces used to show generated questions and recommended products to the user.

[0695] "Response receiving means" refers to the device or system used to receive responses entered by the user.

[0696] "Intention analysis methods" refer to natural language processing techniques and algorithms used to analyze user responses and infer their purchase intentions.

[0697] "Product recommendation methods" refer to systems and algorithms that select appropriate products based on the purpose of a customer's visit, as inferred by intent analysis methods, and recommend them to the user.

[0698] "Initialization means" refers to the function of retrieving the product database and question database from the server via a specially designed computer program.

[0699] "Recommendation methods" refer to a function that uses natural language processing technology to analyze user responses and recommends products based on the results.

[0700] A "server" refers to a central control unit that manages and processes data for the entire system.

[0701] A "product database" refers to a collection of data that stores information about products that are sold.

[0702] A "question database" refers to a collection of data that stores questions to be displayed to users.

[0703] This invention is a system that uses AI to predict customers' purchase intentions in physical stores and efficiently recommends appropriate products. Specific embodiments of this invention are shown below.

[0704] Hardware and software to be used

[0705] Server: A central control unit that manages and processes data for the entire system. It manages the product database and the inquiry database.

[0706] Device: Uses a smartphone to interact with users. This includes displaying questions, receiving answers, and showing recommended products.

[0707] software:

[0708] Python: Execute a script.

[0709] requests: Sends HTTP requests and communicates with the API.

[0710] scikit-learn: NLP analysis and database search.

[0711] Natural Language Processing (NLP): Used to analyze user responses and infer their purpose for visiting the store.

[0712] System configuration and operation

[0713] Initialization

[0714] The server retrieves and initializes the product database and question database from the server. This process utilizes the server API via a specially designed computer program.

[0715] Entering user information

[0716] The device provides an interface for users to input basic information (age, gender, and categories of interest). Users input this information through their smartphone's UI.

[0717] question generation

[0718] The server generates an initial question based on the basic information entered. This question is selected from a question database and displayed on the terminal. For example, a question such as "What size television are you interested in?" might be generated.

[0719] Receiving and analyzing responses

[0720] The user answers questions displayed on the device, and the device sends those answers to the server. The server analyzes the received answers and uses natural language processing techniques to infer their intent. For example, the answer "I want a 50-inch TV" is analyzed.

[0721] Product Recommendation

[0722] Based on the intent analysis results, the server selects appropriate products from the product database and generates a list of recommended products. This list of recommended products is displayed to the user via their terminal. For example, it might say, "Your purpose for visiting the store is to purchase a 50-inch television. Please see the recommended products below."

[0723] Display and select from the recommended product list

[0724] The terminal displays a list of recommended products to the user, who reviews them and selects products of interest. The server then generates additional questions as needed and makes further recommendations.

[0725] Examples of specific prompt messages

[0726] User's basic information: Age 30, Gender Male, Interests: Television

[0727] Initial question: What size television screen are you interested in?

[0728] User reply: I want a 50-inch TV.

[0729] AI analysis results: The recommended product is a 50-inch television, "BRAND A 50-inch".

[0730] Based on the above, the present invention provides a system that can accurately predict the purchase purpose of a store visitor and efficiently recommend appropriate products. This makes it possible to improve customer satisfaction and achieve a smooth purchasing experience.

[0731] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0732] Step 1:

[0733] The server loads and initializes the product database and the question database. Specifically, the server retrieves each database via the server API and loads it into memory. At this time, it sends an HTTP request using a specially designed computer program to download the database files. The database format is JSON or CSV, and the loaded data is stored in internal data storage.

[0734] Input: Database URL obtained from the server API

[0735] Output: Product database and question database

[0736] Step 2:

[0737] The device displays an interface for the user to enter basic information. Specifically, a basic information input form (e.g., age, gender, categories of interest) is displayed on the smartphone screen. The user enters their information into these fields and presses the submit button.

[0738] Input: Smartphone basic information input form

[0739] Output: Basic information entered by the user

[0740] Step 3:

[0741] The server receives basic user information sent from the terminal and generates an initial question. The server retrieves an appropriate question from the question database and customizes it based on the user information. Natural language generation technology may be used in this process. For example, if the user is "30 years old and their interest category is television," the server will generate the question, "What size television are you interested in?"

[0742] Input: Basic information sent from the device

[0743] Output: Initial Question

[0744] Step 4:

[0745] The device displays the generated initial question to the user. Specifically, the question text and answer input fields are displayed on the smartphone screen. The user enters their answer to the question and presses the submit button.

[0746] Input: Initial question received from the server

[0747] Output: User's response

[0748] Step 5:

[0749] The server receives user responses sent from the terminal and analyzes them. Natural language processing techniques are used for intent analysis, and the response content is vectorized for analysis. As a result of the analysis, the user's purpose for visiting the store is identified. For example, from the response "I want a 50-inch TV," the intention "to purchase a 50-inch TV" is inferred.

[0750] Input: User's response sent from the device

[0751] Output: Results of identifying the purpose of visit

[0752] Step 6:

[0753] The server selects appropriate products from the product database based on the intent analysis results. Machine learning algorithms are used to search the product database and find the best product for the user's needs. For example, a list of products suitable for a "50-inch television" is generated.

[0754] Input: Results of intent analysis

[0755] Output: Recommended product list

[0756] Step 7:

[0757] The device displays a list of recommended products to the user. Specifically, a list of recommended products is displayed on the smartphone screen. The user can select a product they are interested in from this list and view detailed information.

[0758] Input: Recommended product list received from the server

[0759] Output: Recommended product list displayed on the smartphone screen

[0760] Step 8:

[0761] The user reviews the recommended product list and chooses whether to purchase the product or answer the questions again. The terminal then performs the following actions based on the user's choice. If the user answers the questions again, the process from step 3 onwards is repeated.

[0762] Input: User selection (purchase decision or follow-up question)

[0763] Output: Purchase procedure or generation of next question

[0764] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0765] As an embodiment of the present invention, a system that uses AI and an emotion engine to predict the purchase purpose of a customer visiting a consumer electronics store is described. This system has a process of taking the user's basic information as input, generating questions based on that information, and analyzing the user's answers and emotions to infer the purpose of the visit.

[0766] Initialization process

[0767] During system initialization, the server establishes connections to the product database and the question database and loads the necessary data. This includes the initial setup of the emotion engine.

[0768] The device displays an initial screen that allows the user to enter basic information. This screen also includes interfaces for capturing emotions, such as voice input and a facial recognition camera.

[0769] The user enters basic information (e.g., age, gender, categories of interest) on the initial screen that is displayed.

[0770] User information entry

[0771] The terminal receives basic information entered by the user and sends it to the server. It also obtains user emotion data from voice input and facial recognition and sends this to the server as well.

[0772] The server analyzes the received basic information and sentiment data, and prepares to generate initial questions from the question database.

[0773] question generation

[0774] Based on basic information and sentiment data, the server selects appropriate questions from the question database and generates questions to display to the user.

[0775] Example: "What size TV screen are you interested in?"

[0776] The terminal displays the generated question to the user.

[0777] The user answers questions and sends their answers to the server via their device. The user's facial expressions and voice during the answering process are also captured as emotion data.

[0778] Receiving and analyzing responses

[0779] The server receives the user's response and stores it in an internal database. Furthermore, it uses an emotion engine to analyze the user's emotions at the time of the response.

[0780] The server analyzes the responses and sentiment data, and then starts processing to generate the next question or to specifically infer the customer's purpose for visiting.

[0781] Intent Analysis

[0782] The server uses natural language processing (NLP) technology and an emotion engine to analyze user responses and emotion data, inferring the user's specific intentions and reasons for visiting the store. For example, it might extract a specific purpose, such as "I want a 50-inch TV," along with emotion data.

[0783] Based on the analysis results, the server generates a list of products suitable for the user from the product database. This list also includes detailed information about recommended products.

[0784] Displaying recommended results

[0785] The server sends a list of the suspected purpose of the visit and recommended products to the terminal.

[0786] The terminal displays the user's estimated purpose for visiting the store and a list of recommended products. For example, it might say, "Your purpose for visiting the store is to purchase a 50-inch TV. Please see the recommended products below."

[0787] The user reviews the displayed list of recommended products and selects the products they are interested in.

[0788] Utilization of emotional data

[0789] The server utilizes its emotion engine even while displaying recommended products, continuously collecting emotional data obtained from the user's facial expressions and voice. This allows for a more precise understanding of which products the user is most interested in.

[0790] The server can use emotional data to provide users with the most suitable additional suggestions.

[0791] Ending Phase

[0792] The user can choose to either view the recommended products and decide to purchase them, or answer the questions again.

[0793] Based on the user's selection, the device will either generate further questions or guide them through the purchase process.

[0794] Specific example

[0795] For example, if a user answers "50 inches" as the TV size they are interested in, the emotion engine can analyze the user's facial expressions and tone of voice during the response to determine if they are truly interested. As a result, if they show a high level of interest, "50-inch TVs" will be strongly reflected in the recommendation list; if they show a low level of interest, TVs of other sizes or with different characteristics can be recommended again.

[0796] By incorporating user sentiment data in this way, it becomes possible to make more accurate product recommendations and increase user satisfaction.

[0797] The following describes the processing flow.

[0798] Step 1:

[0799] During system initialization, the server establishes connections to the product database and the question database and loads the necessary data. This includes the initial setup of the emotion engine.

[0800] Step 2:

[0801] The device displays an initial screen that allows the user to enter basic information. It also simultaneously displays interfaces for voice input and a facial recognition camera to acquire emotional data.

[0802] Step 3:

[0803] On the initial screen displayed, the user enters basic information (e.g., age, gender, categories of interest). Simultaneously, facial expressions and voice tone are captured through the camera and microphone built into the device.

[0804] Step 4:

[0805] The device receives basic information, facial expression data, and voice data, and sends them to the server.

[0806] Step 5:

[0807] The server analyzes the received basic information and emotional data, and prepares to generate initial questions from the question database. This includes a process that evaluates the user's psychological state based on their facial expressions and tone of voice.

[0808] Step 6:

[0809] The server generates appropriate initial questions based on basic information and analysis results, and sends them to the terminal. Example: "What size television are you interested in?"

[0810] Step 7:

[0811] The terminal displays the questions received from the server to the user.

[0812] Step 8:

[0813] The user answers questions. During this process, the device's camera captures the user's facial expressions, and the microphone records their voice tone.

[0814] Step 9:

[0815] The device sends the user's response along with newly acquired emotional data (facial expressions and voice) to the server.

[0816] Step 10:

[0817] The server receives user responses and sentiment data and stores them in a database. Simultaneously, it generates the next question or begins analysis to specifically infer the user's purpose for visiting the store.

[0818] Step 11:

[0819] The server uses natural language processing (NLP) techniques and an emotion engine to analyze the user's responses and emotional data. For example, if a user responds "A 50-inch TV is interesting" while smiling, the server infers their intention as "They have a strong interest in 50-inch TVs."

[0820] Step 12:

[0821] Based on the analysis results, the server generates a list of products suitable for the user from the product database. The generated list also takes into account the user's emotional state.

[0822] Step 13:

[0823] The server sends a list of the suspected purpose of visit and recommended products to the terminal.

[0824] Step 14:

[0825] The terminal displays the user's estimated purpose for visiting the store and a list of recommended products. Example: "Your purpose for visiting the store is to purchase a 50-inch TV. Please see the recommended products below."

[0826] Step 15:

[0827] The user reviews the displayed list of recommended products and selects items that interest them. At this point, the device continues to monitor the user's facial expressions and voice, collecting emotional data.

[0828] Step 16:

[0829] Based on the user's choices, the server analyzes further sentiment data to generate the next question or proceed with the purchase process.

[0830] Step 17:

[0831] Depending on the user's choice, the device will either display the question again or show a screen for the purchase process. If the user chooses to proceed with the purchase, the necessary options and input fields will be displayed.

[0832] Step 18:

[0833] The user proceeds with the purchase process, and the purchase is completed. During the purchase process, the device collects emotional data, allowing for monitoring of the user's satisfaction level.

[0834] This enables more accurate product recommendations that take user emotions into account, resulting in a more personalized and smoother purchasing experience throughout the entire process.

[0835] (Example 2)

[0836] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0837] Conventional user information collection systems generate questions and perform intent analysis based on basic user information and response history, but this alone makes it difficult to accurately grasp the user's true intentions and emotions. As a result, appropriate product recommendations may not be made, potentially lowering user satisfaction.

[0838] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a question generation means that generates an initial question based on the user's basic information and emotional data, an answer receiving means that receives the user's answer and emotional data, and an intent analysis means that analyzes the answer and emotional data to infer the user's purpose for visiting the store. By incorporating the user's emotional data into the analysis, it becomes possible to accurately grasp the user's intentions and emotions and make appropriate product suggestions.

[0839] "Basic information" refers to identifying information such as the user's age, gender, and categories of interest.

[0840] "Emotional data" refers to information about a user's emotions, obtained from their facial expressions, voice tone, and other sources.

[0841] A "question generation means" is a means for generating initial and subsequent questions based on the user's basic information and sentiment data.

[0842] "Display means" refers to an interface for displaying questions generated for the user and recommended products.

[0843] "Response receiving means" refers to a means of receiving user responses and sentiment data.

[0844] "Intention analysis methods" are techniques that analyze received responses and emotional data to infer the user's purpose for visiting the store.

[0845] A "product recommendation method" is a means of recommending appropriate products based on the purpose of the customer's visit, which is inferred by intent analysis methods.

[0846] "Natural language processing technology" is a technology that analyzes a user's language response and understands its meaning.

[0847] An "emotion engine" is software or a system used to analyze a user's emotional data.

[0848] This invention relates to a system that uses AI and an emotion engine to analyze the purpose of a user's visit to a consumer electronics store and recommend appropriate products. Based on the basic information and emotion data entered by the user, this system generates questions, receives the answers, analyzes the answers and emotion data to clarify the user's intentions, and makes optimal product suggestions.

[0849] System Configuration

[0850] 1. The server establishes connections to the product database and the question database during system initialization and loads the necessary data. Specifically, it uses MySQL as the database management system and IBM Watson Emotion Recognition as the sentiment analysis engine. This establishes the initial engine settings and database connections.

[0851] 2. The device displays a screen for the user to enter basic information. This screen also includes interfaces that utilize voice input (Google Speech-to-Text API) and a facial recognition camera (Azure Face API). This makes it possible to obtain emotional data from the user's facial expressions and voice.

[0852] 3. The user enters basic information (age, gender, categories of interest, etc.) on the initial screen that is displayed. The entered basic information is sent to the server by the device.

[0853] 4. The device transmits not only basic information but also emotional data obtained from voice input and facial recognition to the server.

[0854] 5. The server analyzes the received basic information and sentiment data to generate an initial question from the question database. For example, a specific question such as "What size television are you interested in?" is generated.

[0855] 6. The terminal displays questions sent from the server to the user. The user answers the displayed questions. When answering, facial expressions and voice are also captured through the terminal and simultaneously sent to the server as emotion data.

[0856] 7. The server stores the received responses in a database and uses an emotion engine to analyze the user's emotions at the time of the response. This allows the server to infer the user's specific intentions and purpose for visiting the store.

[0857] 8. Based on the analysis results, the server generates an appropriate product list from the product database. For example, if the intention "I want a 50-inch TV" is extracted, a product list specifically for 50-inch TVs will be generated.

[0858] 9. The terminal displays a list of recommended products sent from the server to the user. Sentimental data continues to be collected even after the display, making it possible to understand more precisely which products the user is interested in.

[0859] 10. The user reviews the displayed list of recommended products and selects items of interest. In some cases, they may answer further questions or proceed with the purchase.

[0860] Specific example

[0861] For example, if a user enters basic information such as "30 years old, male, interested in home appliances" and answers the question "I want a 50-inch TV," the server uses an emotion engine to analyze the user's facial expressions and tone of voice during the response. Based on this analysis, if the server determines that the user is genuinely very interested in a 50-inch TV, a 50-inch TV will be strongly reflected in the recommendation list. If the user shows little interest, a TV of a different size or with different characteristics will be recommended instead.

[0862] Example of a prompt

[0863] An example of a prompt to input into a generative AI model is: "A user showed interest in a 50-inch TV when they visited the store. Please explain the process for recommending products that match this user's interests."

[0864] By incorporating user emotional data into the analysis in this way, it becomes possible to provide a system that makes more accurate and appropriate product recommendations.

[0865] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0866] Step 1:

[0867] The server establishes connections to the product database and the question database during system initialization. Specifically, it uses a database management system (e.g., MySQL) to load the necessary data (e.g., product characteristics, question templates). It also performs the initial setup of the sentiment analysis engine (e.g., IBM Watson Emotion Recognition). The input is the database connection information and the sentiment engine configuration information, and the output is the connected database and the initialized sentiment engine.

[0868] Step 2:

[0869] The device displays an initial screen for the user to enter basic information. This screen also includes voice input (e.g., Google Speech-to-Text API) and a facial recognition camera (e.g., Azure Face API), providing an interface for acquiring emotional data from the user's facial expressions and voice. Input consists of basic information provided by the user (e.g., age, gender, categories of interest), and output consists of the user's basic information and the display of the initial screen.

[0870] Step 3:

[0871] The user enters basic information on the initial screen. This information includes age, gender, and categories of interest. The input is basic information, and the output is basic information sent to the device.

[0872] Step 4:

[0873] The terminal receives basic information entered by the user and sends it to the server. It also simultaneously collects emotional data obtained from voice input and facial recognition and sends this data to the server as well. The input consists of the user's basic information and emotional data, while the output is the data sent to the server.

[0874] Step 5:

[0875] The server analyzes the received basic information and sentiment data and generates an initial question from the question database. Specifically, it selects an appropriate question based on the received basic information and sentiment data, and generates a concrete question such as, "What size television are you interested in?" The input is basic information and sentiment data, and the output is the generated question.

[0876] Step 6:

[0877] The terminal displays the initial question sent from the server to the user. The input is the initial question from the server, and the output is the display of the question to the user.

[0878] Step 7:

[0879] The user answers the displayed questions. During the answering process, facial expressions and voice are also collected through a facial recognition camera and voice input function. Input consists of the questions, user answers, facial expressions, and voice data, while output consists of the answers and emotion data sent to the device.

[0880] Step 8:

[0881] The terminal receives user responses and sentiment data and sends them to the server. The input is the user's responses and sentiment data, and the output is the data sent to the server.

[0882] Step 9:

[0883] The server stores user responses in a database and uses an emotion engine to analyze the user's emotions at the time of their response. The input consists of the user's response and emotion data, while the output consists of the stored data and the analysis results.

[0884] Step 10:

[0885] The server analyzes the user's intent based on the response content and sentiment data using natural language processing technology (e.g., Google Natural Language API) and a sentiment engine. For example, it extracts a specific objective, such as "I want a 50-inch TV," along with sentiment data. The input is the user's response and sentiment data, and the output is the analyzed user intent.

[0886] Step 11:

[0887] The server generates a list of appropriate products from the product database based on the analysis results. The input is the analysis results, and the output is the generated product list.

[0888] Step 12:

[0889] The server sends the generated product list to the terminal. The input is the generated product list, and the output is the data sent to the terminal.

[0890] Step 13:

[0891] The terminal displays a list of recommended products to the user. For example, it might display, "Your purpose for visiting is to purchase a 50-inch television. Please see the recommended products below." The input is a list of recommended products from the server, and the output is the displayed product list.

[0892] Step 14:

[0893] Even while displaying recommended products, the device utilizes an emotion engine to continuously collect emotional data obtained from the user's facial expressions and voice, and transmits it to the server. The input is the user's facial expressions and voice data, and the output is the data transmitted to the server.

[0894] Step 15:

[0895] The server provides users with optimal additional suggestions based on collected sentiment data. The input is continuously collected sentiment data, and the output is data for additional suggestions.

[0896] Step 16:

[0897] The user can choose to view the recommended products and decide whether to purchase them, or to answer another question. The input is a list of recommended products or a new question, and the output is the user's choice or answer.

[0898] Step 17:

[0899] The terminal either generates new questions or guides the user through the purchase process based on their selection. The input is the user's selection, and the output is a new question or a purchase process screen.

[0900] (Application Example 2)

[0901] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0902] Existing consumer electronics retailers face challenges in accurately understanding what customers want to buy, hindering efficient product recommendations. Furthermore, recommending products without understanding user emotions or specific intentions can lead to decreased customer satisfaction. Moreover, with the rise of online shopping, there's a need to enhance the unique customer experience offered by physical stores. To address these challenges, a system is needed that utilizes not only basic customer information but also emotional data to accurately predict purchasing intent and provide precise product recommendations.

[0903] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes an input means for inputting the user's basic information, a question generation means for generating initial questions based on the basic information, a display means for displaying the generated questions to the user, an answer receiving means for receiving the user's answers and emotional data, an intent analysis means for analyzing the answers and emotional data to infer the purpose of the visit, a product recommendation means for recommending appropriate products based on the inferred purpose of the visit, and a display means for displaying the recommended products. This makes it possible to accurately grasp the user's emotions and intentions and make product suggestions that match the purpose of the visit, thereby improving customer satisfaction.

[0904] A "user" is a customer who visits an electronics retail store and uses the system.

[0905] "Basic information" refers to data that users enter, such as age, gender, and categories of interest.

[0906] "Input method" refers to an interface that allows users to input basic information, and includes devices such as smartphones and smart glasses.

[0907] "Question generation method" refers to the process of creating appropriate questions based on the basic information entered by the user.

[0908] "Display means" refers to displays or screens used to present generated questions and recommended products to the user.

[0909] A "response receiving method" refers to an interface that receives responses and sentiment data entered by users in response to questions.

[0910] "Emotional data" refers to emotional information obtained from the user's facial expressions, voice, and other sources.

[0911] "Intention analysis means" refers to a function that analyzes user responses and emotional data to infer the user's purpose for visiting the store.

[0912] "Natural language processing technology" refers to artificial intelligence technology used to understand and interpret user text data.

[0913] An "emotion engine" refers to an algorithm and software that analyzes a user's emotions from their facial expressions and voice.

[0914] "Product recommendation methods" refer to the process of selecting and recommending appropriate products based on the user's purpose for visiting the store.

[0915] A "server" is a central device that manages the entire system and performs tasks such as data analysis, storage, and query generation.

[0916] To implement the present invention, terminals such as smartphones and smart glasses, and servers connected to them, are used in consumer electronics stores. The main components of the system include input means for inputting basic user information, question generation means for generating questions, display means for displaying the generated questions, answer receiving means for receiving user answers and sentiment data, intent analysis means for analyzing user answers and sentiment data, product recommendation means for recommending appropriate products, and display means for displaying recommended products.

[0917] The server connects to the product database and question database during system initialization and performs the initial setup of the emotion engine. It uses "EmotionEngine" as the emotion engine and "NlpProcessor" as the natural language processing technology. The server generates questions based on the user's basic information and past answer history and sends them to the terminal.

[0918] The device takes the form of a smartphone or smart glasses and displays an initial screen for the user to input basic information. The device also includes features such as voice input and a facial recognition camera, which are used to collect emotional data. The basic information entered by the user is sent to a server, which then generates appropriate initial questions based on this information.

[0919] Users respond to questions displayed on their device using voice or touch input. The device sends these responses to a server, which analyzes the emotional data acquired along with the responses. By using an emotion engine to read the user's emotions from their facial expressions and voice, and by analyzing the content of their responses using NLP technology, the server infers the user's purpose for visiting the store.

[0920] The server selects appropriate products from its product database based on the inferred purpose of the visit, generates a list of recommended products, and sends it to the terminal. The terminal displays this list to the user, who then selects products that interest them. Throughout this process, the server continuously collects sentiment data and provides recommendations tailored to the user's preferences.

[0921] For example, when a user responds that they are "looking for a large refrigerator," the system analyzes the user's facial expressions and tone of voice to estimate whether their interest is genuine. Based on these results, "refrigerators of 500 liters or more" are strongly reflected in the recommendation list.

[0922] Examples of prompt statements used as input to a generative AI model are as follows:

[0923] Please generate questions to infer the user's purpose of purchase. The basic user information is as follows:

[0924] Age: 35

[0925] Gender: Male

[0926] Categories of interest: Home appliances

[0927] Please generate questions based on the following data:

[0928] Emotional data: Joy (high)

[0929] Example questions to output:

[0930] "What size refrigerator in liters are you interested in?"

[0931] This makes it possible to accurately understand the user's emotions and intentions, and by suggesting products that match their purpose for visiting the store, customer satisfaction can be improved.

[0932] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0933] Step 1:

[0934] During system initialization, the server establishes connections to the product database and question database and performs initial setup of the sentiment engine. At this time, the server reads product information and question templates and sets initial parameters for sentiment analysis.

[0935] Input: Product database, question database, emotion engine configuration information

[0936] Output: System state after initial setup is complete

[0937] Specific actions: Connecting to the database, loading configuration files.

[0938] Step 2:

[0939] The device displays an initial screen where the user can enter basic information. The device is also equipped with voice input and a facial recognition camera, providing an interface to acquire user emotion data.

[0940] Input: System initial settings

[0941] Output: Basic Information Input Screen

[0942] Specific actions: rendering the initial screen, voice input, and camera activation.

[0943] Step 3:

[0944] The user enters basic information (age, gender, and categories of interest) using the initial screen displayed. The device also acquires emotional data through voice input and facial recognition.

[0945] Input: User's basic information, voice and facial recognition data

[0946] Output: User basic information, sentiment data

[0947] Specific operations: Collection of user input data, real-time acquisition of sentiment data.

[0948] Step 4:

[0949] The terminal analyzes the collected basic information and emotional data and sends it to the server. The server receives this data and prepares it for analysis.

[0950] Input: User's basic information, sentiment data

[0951] Output: Data ready for analysis

[0952] Specific actions: Packing input data and sending it to the server.

[0953] Step 5:

[0954] The server generates an initial question from the question database based on the user's basic information and sentiment data.

[0955] Input: User's basic information, sentiment data

[0956] Output: Initial Question

[0957] Specific actions: Execute database queries, apply query templates.

[0958] Step 6:

[0959] The device displays the generated questions to the user. The user enters answers to the questions and expresses emotions through voice and facial expressions.

[0960] Input: Initial Question

[0961] Output: User responses, sentiment data

[0962] Specific actions: Displaying questions, providing an interface for entering answers.

[0963] Step 7:

[0964] The device sends the user's responses and sentiment data to the server, which receives, stores, and then analyzes this data.

[0965] Input: User responses, sentiment data

[0966] Output: Analysis results

[0967] Specific actions: Packing data, sending it to the server, and saving it to the database.

[0968] Step 8:

[0969] The server uses natural language processing technology and an emotion engine to analyze the user's responses and emotion data, and to infer the user's purpose for visiting the store.

[0970] Input: User responses, sentiment data

[0971] Output: Estimated results of the purpose of visit

[0972] Specific actions: Application of NLP algorithms, performance of sentiment analysis.

[0973] Step 9:

[0974] Based on the inferred purpose of the customer's visit, the server selects appropriate products from the product database and generates a list of recommended products.

[0975] Input: Estimated result of the purpose of visit

[0976] Output: Recommended product list

[0977] Specific actions: Execute database queries, select recommended products.

[0978] Step 10:

[0979] The terminal displays a list of recommended products to the user, who then selects products of interest. The server continuously collects sentiment data and provides optimal additional suggestions.

[0980] Input: Recommended product list, user selection

[0981] Output: Optimal product proposal

[0982] Specific actions: Rendering of recommended products, real-time analysis of sentiment data.

[0983] Step 11:

[0984] When a user selects a product or answers additional questions, the device sends data back to the server, which then asks the next most appropriate question or suggests a product as needed.

[0985] Input: User selection, answers to additional questions

[0986] Output: Next question or product suggestion

[0987] Specific actions: data retransmission, reapplication of analysis algorithms.

[0988] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0989] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0990] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0991] [Third Embodiment]

[0992] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0993] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0994] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0995] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0996] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0997] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0998] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0999] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1000] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1001] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1002] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1003] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[1004] As an embodiment of the present invention, a system that uses AI to predict the purchase purpose of a customer visiting a consumer electronics store will be described. This system has a process of taking the user's basic information as input, generating questions based on that information, and analyzing the user's answers to infer the purpose of the visit.

[1005] Initialization process

[1006] The server reads the product database and the question database, and prepares the necessary information in advance when the system starts up.

[1007] The device displays an initial screen that allows the user to enter basic information.

[1008] The user enters basic information (e.g., age, gender, categories of interest) on the displayed screen.

[1009] User information entry

[1010] The terminal receives basic information entered by the user and sends it to the server.

[1011] The server analyzes the received basic information and prepares to generate appropriate initial questions.

[1012] question generation

[1013] Based on the basic information, the server selects a suitable question from the question database and generates a question to display to the user.

[1014] Example: "What size TV screen are you interested in?"

[1015] The terminal displays the generated question to the user.

[1016] The user answers the question and sends the answer to the server via their device.

[1017] Receiving and analyzing responses

[1018] The server receives the user's response and saves it to its internal database.

[1019] The server analyzes the user's responses and processes them to generate the next question or to specifically infer the purpose of their visit.

[1020] Intent Analysis

[1021] The server uses natural language processing (NLP) techniques to analyze user responses and infer the user's purpose for visiting the store.

[1022] Example: Analyze the intention behind the statement, "I want a 50-inch TV."

[1023] Based on the analysis results, the server selects appropriate products from the product database and generates a list of recommended products.

[1024] Displaying recommended results

[1025] The server sends a list of recommended products along with an estimated reason for the customer's visit to the terminal.

[1026] The terminal displays to the user a list of suggested reasons for visiting the store and recommended products.

[1027] Example: "Your purpose for visiting our store is to purchase a 50-inch television. Please see our recommended products below."

[1028] The user reviews the recommended product list and selects the products they are interested in.

[1029] Ending Phase

[1030] The user can choose to either view the recommended products and decide to purchase them, or answer the questions again.

[1031] Based on the user's selection, the device will either generate further questions or guide them through the purchase process.

[1032] As described above, the system can start with user input, go through AI analysis, and ultimately present recommended products. This system allows customers to efficiently reach their intended purpose for visiting the store, providing a smooth shopping experience.

[1033] The following describes the processing flow.

[1034] Step 1:

[1035] During system initialization, the server establishes connections between the product database and the question database and loads the necessary data. This prepares the information required for system operation.

[1036] Step 2:

[1037] The device displays an initial screen to the user and provides an input form for basic information. This is for entering information such as age, gender, and categories of interest.

[1038] Step 3:

[1039] The user enters their basic information into the input form displayed on this initial screen and clicks the submit button.

[1040] Step 4:

[1041] The terminal receives basic information entered by the user and sends that information to the server.

[1042] Step 5:

[1043] The server analyzes the received basic information and prepares to generate initial questions from the question database. Based on this analysis, it selects appropriate questions related to the user's interests.

[1044] Step 6:

[1045] The server sends the generated initial question to the terminal and displays it to the user.

[1046] Step 7:

[1047] The terminal displays an initial question received from the server to the user. For example, it might display, "What size television are you interested in?"

[1048] Step 8:

[1049] The user answers the displayed questions and sends those answers to the server via their device.

[1050] Step 9:

[1051] The server receives the user's response and stores it in the database. At the same time, it either performs analysis to generate the next question or starts processing to specifically infer the purpose of the visit.

[1052] Step 10:

[1053] The server uses natural language processing (NLP) techniques to analyze user responses and infer user-specific intentions and reasons for visiting the store. For example, it might extract specific objectives such as "I want a 50-inch TV."

[1054] Step 11:

[1055] Based on the analysis results, the server generates a list of products suitable for the user from the product database. This list also includes detailed information about recommended products.

[1056] Step 12:

[1057] The server sends a list of the suspected purpose of the visit and recommended products to the terminal.

[1058] Step 13:

[1059] The terminal displays the user's estimated purpose for visiting the store and a list of recommended products. For example, it might say, "Your purpose for visiting the store is to purchase a 50-inch TV. Please see the recommended products below."

[1060] Step 14:

[1061] The user reviews the displayed list of recommended products and selects the products they are interested in.

[1062] Step 15:

[1063] The user clicks on the selected product to view details. They then choose to proceed with the purchase or answer further questions.

[1064] Step 16:

[1065] The device will determine the next action based on the user's selection. If the user chooses to answer the question again, it will send a request to the server and restart the question generation process. If the user chooses to proceed with the purchase, it will display the appropriate purchase process screen.

[1066] Step 17:

[1067] The server generates new questions and performs processes related to the purchase procedure, returning appropriate feedback based on the user's choices. This completes the entire process.

[1068] Through these steps, the system can efficiently understand the user's purpose for visiting the store and suggest appropriate products.

[1069] (Example 1)

[1070] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1071] In modern consumer electronics stores, quickly and accurately understanding a customer's purpose for visiting is crucial for providing a smooth shopping experience. However, existing systems struggle to efficiently carry out the entire process from user input to question generation, response analysis, intent inference, and product recommendation. This invention aims to solve these problems and provide a system that presents users with appropriate questions, quickly and accurately infers their purpose for visiting, and recommends appropriate products.

[1072] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1073] In this invention, the server includes an input means for inputting the user's basic information, a question generation means for generating an initial question based on the basic information received from the input means, a display means for displaying the generated question to the user, an answer receiving means for receiving the user's answer, an intent analysis means for analyzing the answer received from the answer receiving means and inferring the user's purpose for visiting the store, a product recommendation means for recommending appropriate products based on the purpose of visiting the store inferred by the intent analysis means, a display means for displaying the recommended products to the user, a communication means for performing question generation, answer reception, intent analysis, product recommendation, and display via a terminal and the server, and a next question generation means for generating the next question that the user is likely to be interested in based on the generated list of recommended products. This makes it possible to efficiently infer the purpose of visiting the store from the user's basic information and answers and to quickly recommend appropriate products.

[1074] "User basic information" refers to personal information that users enter, such as age, gender, and categories of interest.

[1075] An "input means" is a device or interface for a user to input basic information.

[1076] A "question generation means" is an algorithm or system that automatically generates appropriate questions based on the user's basic information.

[1077] A "display means" is an interface that visually presents generated questions and recommended products to the user.

[1078] A "response receiving system" is a system that receives responses entered by users and saves them in an appropriate data format.

[1079] A "means of intent analysis" is a system that uses natural language processing technology to analyze user responses and infer their purpose for visiting the store.

[1080] A "product recommendation system" is a system that selects and recommends appropriate products to users based on their purpose for visiting the store, as inferred by intent analysis.

[1081] "Communication methods" refer to network infrastructure and protocols used to send and receive data between a server and a terminal.

[1082] "Next question generation means" refers to an algorithm or system that generates the next appropriate question based on the user's basic information and previous answers.

[1083] A "recommended product list" is a list of products suitable for the user's purpose of visiting the store, and represents a group of product options presented to the user.

[1084] To implement the present invention, a system is needed that takes basic user information as input, uses AI to infer the purpose of the visit based on that information, and recommends appropriate products. This system involves cooperation between a server, a terminal, and the user, and includes the following main components and processes.

[1085] Hardware and software

[1086] Hardware used: A typical server device (e.g., Dell PowerEdge R740) and a terminal device for user operation (e.g., iPad Pro)

[1087] Software used: Database management system (e.g., MySQL), Web framework (e.g., Flask for Python), Natural Language Processing library (e.g., NLTK for Python), Frontend framework (e.g., React)

[1088] Specific description of the system's operation

[1089] The server first loads the product database and the question database, preparing the information needed when the system starts up. This includes executing SQL queries such as "SELECT FROM products" and "SELECT FROM questions," for example.

[1090] The device displays an initial screen where the user can enter basic information (age, gender, categories of interest, etc.). This initial screen is built using React and designed to be easy for the user to use.

[1091] The user enters their basic information into the form displayed on the initial screen. For example, the user might enter information such as age "30 years old," gender "female," and category of interest "home appliances."

[1092] The terminal receives the basic information entered by the user and sends it to the server via an HTTP POST request using the Axios library. Specifically, the terminal sends {"age": 30, "gender": "female", "category": "electronics"} in JSON format to "POST / api / userinfo".

[1093] The server analyzes the received basic information and selects an appropriate question from the question database. The technology used here is the Python natural language processing library NLTK. Based on the user's age (30 years old), gender (female), and category (home appliances), the server generates the question, "What size TV are you interested in?"

[1094] The generated question is sent from the server to the terminal, which then displays it to the user. The user then enters an answer to the displayed question. For example, the user might answer "50 inches".

[1095] The terminal receives the user's response and sends it back to the server using an HTTP POST request. The server stores the received response in its internal database and analyzes it using NLTK. This allows the server to infer the user's purpose for visiting the store, such as "the user is looking for a 50-inch television."

[1096] Based on the inferred purpose of the visit, the server selects appropriate products from the product database and generates a list of recommended products. This list is then sent back to the terminal and displayed to the user. Based on the displayed list of recommended products, the user can select products.

[1097] Example prompt statements

[1098] Prompt example: "The user is 30 years old, female, and interested in home electronics. Please generate the first question to infer their purpose for visiting."

[1099] As demonstrated here, the present invention can provide an efficient and smooth purchasing experience through a series of processes that begin with user input, go through server processing, and display the results on the terminal. This makes it possible to quickly identify the customer's purchase purpose and support them in selecting appropriate products.

[1100] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1101] System program processing flow

[1102] Step 1:

[1103] Initialization process

[1104] The server reads the product database and the question database, preparing the information needed when the system starts up. This includes executing SQL queries such as "SELECT FROM products" and "SELECT FROM questions" from the MySQL database to load all the data into memory.

[1105] Input: Database connection information

[1106] Output: Product data and question data loaded into memory

[1107] Step 2:

[1108] Initial screen display

[1109] The device displays an initial screen for the user to enter basic information. This screen is designed using JavaScript and the React framework.

[1110] Input: UI design information

[1111] Output: Screen displaying an input form to the user

[1112] Step 3:

[1113] Entering basic information

[1114] The user enters their basic information (e.g., age "30", gender "female", category of interest "home appliances") into the form displayed on the initial screen.

[1115] Input: User's basic information

[1116] Output: Basic information data sent to the terminal

[1117] Step 4:

[1118] Receiving and sending basic information

[1119] The terminal receives the basic information entered by the user and sends it to the server using an HTTP POST request. The Axios library is used to send the data in JSON format to "POST / api / userinfo".

[1120] Input: User's basic information

[1121] Output: Basic information sent to the server

[1122] Step 5:

[1123] question generation

[1124] The server analyzes the received basic information and selects a suitable question from the question database. Using Python's NLTK, it selects a question based on information such as "the user is 30 years old, female, and in the home appliance category."

[1125] Input: User's basic information

[1126] Output: Generated question (e.g., "What size TV screen are you interested in?")

[1127] Step 6:

[1128] Submitting and displaying questions

[1129] The server sends the generated question to the terminal.

[1130] The terminal displays the received question to the user.

[1131] Input: Generated Question

[1132] Output: Questions displayed to the user

[1133] Step 7:

[1134] User response input

[1135] The user enters their answers to the questions displayed on the device.

[1136] Input: User's response (e.g., "50 inches")

[1137] Output: Response data sent to the terminal

[1138] Step 8:

[1139] Receiving and sending responses

[1140] The terminal receives the user's response and sends it to the server using an HTTP POST request.

[1141] Input: User's response

[1142] Output: Response sent to the server

[1143] Step 9:

[1144] Saving and analyzing responses

[1145] The server stores the received responses in an internal database and then uses NLTK again to analyze and infer the user's intent.

[1146] Input: User's response

[1147] Output: Estimated purpose of visit (e.g., "The user is looking for a 50-inch TV")

[1148] Step 10:

[1149] Product Recommendation

[1150] The server selects appropriate products from the product database based on the analysis results and generates a list of recommended products. The SQL query "SELECT FROM products WHERE category='TV' AND size='50 inches'" is executed.

[1151] Input: Estimated purpose of visit

[1152] Output: Recommended product list

[1153] Step 11:

[1154] Sending and displaying recommended products

[1155] The server sends a list of recommended products and the presumed purpose of the visit to the terminal.

[1156] The device displays a list of recommended products to the user.

[1157] Input: Recommended Conlist

[1158] Output: Recommended product list displayed to the user

[1159] Step 12:

[1160] User choices and next actions

[1161] The user reviews the recommended product list and can choose to select products they are interested in, or choose to answer the questions again.

[1162] Input: User Selection

[1163] Output: Further questions or instructions for the purchase process

[1164] Step 13:

[1165] Continue with your question or purchase

[1166] Based on the user's selection, the device will either generate further questions or guide them through the purchase process.

[1167] Input: User Selection

[1168] Output: Display of the next question or guidance to the purchase page

[1169] This makes the entire process, starting with user input, going through server analysis, and displaying the results on the terminal, concretely clear.

[1170] (Application Example 1)

[1171] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1172] Traditional customer service systems in physical stores have struggled to accurately understand customers' purchase objectives and recommend appropriate products. This often resulted in an inefficient and smooth customer purchasing experience, potentially leading to decreased customer satisfaction. Furthermore, conventional systems sometimes failed to fully utilize basic customer information and past history, resulting in inappropriate question generation. To address these issues, this invention provides a system that uses AI to accurately predict customers' purchase objectives and recommend the most suitable products.

[1173] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1174] In this invention, the server includes an input means for inputting basic user information, a question generation means for generating initial questions based on the basic information received from the input means, a display means for displaying the generated questions to the user, an answer receiving means for receiving the user's answers, an intent analysis means for analyzing the answers received from the answer receiving means and inferring the user's purpose for visiting the store, a product recommendation means for recommending appropriate products based on the purpose of visiting the store inferred by the intent analysis means, a display means for displaying the recommended products to the user, an initialization means for obtaining a product database and a question database from the server via a specially designed computer program, and a recommendation means for analyzing the user's answers using natural language processing technology and recommending products based on the results. This makes it possible to accurately infer the customer's purchase purpose and efficiently recommend appropriate products.

[1175] "User" refers to consumers or users who utilize a system.

[1176] "Basic information" refers to information necessary for generating initial questions, such as the user's age, gender, and areas of interest.

[1177] "Input means" refers to devices or interfaces that users use to input basic information.

[1178] "Question generation means" refers to a system or algorithm that has the function of generating the next question to be displayed based on the user's basic information.

[1179] "Display means" refers to displays, screens, or interfaces used to show generated questions and recommended products to the user.

[1180] "Response receiving means" refers to the device or system used to receive responses entered by the user.

[1181] "Intention analysis methods" refer to natural language processing techniques and algorithms used to analyze user responses and infer their purchase intentions.

[1182] "Product recommendation methods" refer to systems and algorithms that select appropriate products based on the purpose of a customer's visit, as inferred by intent analysis methods, and recommend them to the user.

[1183] "Initialization means" refers to the function of retrieving the product database and question database from the server via a specially designed computer program.

[1184] "Recommendation methods" refer to a function that uses natural language processing technology to analyze user responses and recommends products based on the results.

[1185] A "server" refers to a central control unit that manages and processes data for the entire system.

[1186] A "product database" refers to a collection of data that stores information about products that are sold.

[1187] A "question database" refers to a collection of data that stores questions to be displayed to users.

[1188] This invention is a system that uses AI to predict customers' purchase intentions in physical stores and efficiently recommends appropriate products. Specific embodiments of this invention are shown below.

[1189] Hardware and software to be used

[1190] Server: A central control unit that manages and processes data for the entire system. It manages the product database and the inquiry database.

[1191] Device: Uses a smartphone to interact with users. This includes displaying questions, receiving answers, and showing recommended products.

[1192] software:

[1193] Python: Execute a script.

[1194] requests: Sends HTTP requests and communicates with the API.

[1195] scikit-learn: NLP analysis and database search.

[1196] Natural Language Processing (NLP): Used to analyze user responses and infer their purpose for visiting the store.

[1197] System configuration and operation

[1198] Initialization

[1199] The server retrieves and initializes the product database and question database from the server. This process utilizes the server API via a specially designed computer program.

[1200] Entering user information

[1201] The device provides an interface for users to input basic information (age, gender, and categories of interest). Users input this information through their smartphone's UI.

[1202] question generation

[1203] The server generates an initial question based on the basic information entered. This question is selected from a question database and displayed on the terminal. For example, a question such as "What size television are you interested in?" might be generated.

[1204] Receiving and analyzing responses

[1205] The user answers questions displayed on the device, and the device sends those answers to the server. The server analyzes the received answers and uses natural language processing techniques to infer their intent. For example, the answer "I want a 50-inch TV" is analyzed.

[1206] Product Recommendation

[1207] Based on the intent analysis results, the server selects appropriate products from the product database and generates a list of recommended products. This list of recommended products is displayed to the user via their terminal. For example, it might say, "Your purpose for visiting the store is to purchase a 50-inch television. Please see the recommended products below."

[1208] Display and select from the recommended product list

[1209] The terminal displays a list of recommended products to the user, who reviews them and selects products of interest. The server then generates additional questions as needed and makes further recommendations.

[1210] Examples of specific prompt messages

[1211] User's basic information: Age 30, Gender Male, Interests: Television

[1212] Initial question: What size television screen are you interested in?

[1213] User reply: I want a 50-inch TV.

[1214] AI analysis results: The recommended product is a 50-inch television, "BRAND A 50-inch".

[1215] Based on the above, the present invention provides a system that can accurately predict the purchase purpose of a store visitor and efficiently recommend appropriate products. This makes it possible to improve customer satisfaction and achieve a smooth purchasing experience.

[1216] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1217] Step 1:

[1218] The server loads and initializes the product database and the question database. Specifically, the server retrieves each database via the server API and loads it into memory. At this time, it sends an HTTP request using a specially designed computer program to download the database files. The database format is JSON or CSV, and the loaded data is stored in internal data storage.

[1219] Input: Database URL obtained from the server API

[1220] Output: Product database and question database

[1221] Step 2:

[1222] The device displays an interface for the user to enter basic information. Specifically, a basic information input form (e.g., age, gender, categories of interest) is displayed on the smartphone screen. The user enters their information into these fields and presses the submit button.

[1223] Input: Smartphone basic information input form

[1224] Output: Basic information entered by the user

[1225] Step 3:

[1226] The server receives basic user information sent from the terminal and generates an initial question. The server retrieves an appropriate question from the question database and customizes it based on the user information. Natural language generation technology may be used in this process. For example, if the user is "30 years old and their interest category is television," the server will generate the question, "What size television are you interested in?"

[1227] Input: Basic information sent from the device

[1228] Output: Initial Question

[1229] Step 4:

[1230] The device displays the generated initial question to the user. Specifically, the question text and answer input fields are displayed on the smartphone screen. The user enters their answer to the question and presses the submit button.

[1231] Input: Initial question received from the server

[1232] Output: User's response

[1233] Step 5:

[1234] The server receives user responses sent from the terminal and analyzes them. Natural language processing techniques are used for intent analysis, and the response content is vectorized for analysis. As a result of the analysis, the user's purpose for visiting the store is identified. For example, from the response "I want a 50-inch TV," the intention "to purchase a 50-inch TV" is inferred.

[1235] Input: User's response sent from the device

[1236] Output: Results of identifying the purpose of visit

[1237] Step 6:

[1238] The server selects appropriate products from the product database based on the intent analysis results. Machine learning algorithms are used to search the product database and find the best product for the user's needs. For example, a list of products suitable for a "50-inch television" is generated.

[1239] Input: Results of intent analysis

[1240] Output: Recommended product list

[1241] Step 7:

[1242] The device displays a list of recommended products to the user. Specifically, a list of recommended products is displayed on the smartphone screen. The user can select a product they are interested in from this list and view detailed information.

[1243] Input: Recommended product list received from the server

[1244] Output: Recommended product list displayed on the smartphone screen

[1245] Step 8:

[1246] The user reviews the recommended product list and chooses whether to purchase the product or answer the questions again. The terminal then performs the following actions based on the user's choice. If the user answers the questions again, the process from step 3 onwards is repeated.

[1247] Input: User selection (purchase decision or follow-up question)

[1248] Output: Purchase procedure or generation of next question

[1249] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1250] As an embodiment of the present invention, a system that uses AI and an emotion engine to predict the purchase purpose of a customer visiting a consumer electronics store is described. This system has a process of taking the user's basic information as input, generating questions based on that information, and analyzing the user's answers and emotions to infer the purpose of the visit.

[1251] Initialization process

[1252] During system initialization, the server establishes connections to the product database and the question database and loads the necessary data. This includes the initial setup of the emotion engine.

[1253] The device displays an initial screen that allows the user to enter basic information. This screen also includes interfaces for capturing emotions, such as voice input and a facial recognition camera.

[1254] The user enters basic information (e.g., age, gender, categories of interest) on the initial screen that is displayed.

[1255] User information entry

[1256] The terminal receives basic information entered by the user and sends it to the server. It also obtains user emotion data from voice input and facial recognition and sends this to the server as well.

[1257] The server analyzes the received basic information and sentiment data, and prepares to generate initial questions from the question database.

[1258] question generation

[1259] Based on basic information and sentiment data, the server selects appropriate questions from the question database and generates questions to display to the user.

[1260] Example: "What size TV screen are you interested in?"

[1261] The terminal displays the generated question to the user.

[1262] The user answers questions and sends their answers to the server via their device. The user's facial expressions and voice during the answering process are also captured as emotion data.

[1263] Receiving and analyzing responses

[1264] The server receives the user's response and stores it in an internal database. Furthermore, it uses an emotion engine to analyze the user's emotions at the time of the response.

[1265] The server analyzes the responses and sentiment data, and then starts processing to generate the next question or to specifically infer the customer's purpose for visiting.

[1266] Intent Analysis

[1267] The server uses natural language processing (NLP) technology and an emotion engine to analyze user responses and emotion data, inferring the user's specific intentions and reasons for visiting the store. For example, it might extract a specific purpose, such as "I want a 50-inch TV," along with emotion data.

[1268] Based on the analysis results, the server generates a list of products suitable for the user from the product database. This list also includes detailed information about recommended products.

[1269] Displaying recommended results

[1270] The server sends a list of the suspected purpose of the visit and recommended products to the terminal.

[1271] The terminal displays the user's estimated purpose for visiting the store and a list of recommended products. For example, it might say, "Your purpose for visiting the store is to purchase a 50-inch TV. Please see the recommended products below."

[1272] The user reviews the displayed list of recommended products and selects the products they are interested in.

[1273] Utilization of emotional data

[1274] The server utilizes its emotion engine even while displaying recommended products, continuously collecting emotional data obtained from the user's facial expressions and voice. This allows for a more precise understanding of which products the user is most interested in.

[1275] The server can use emotional data to provide users with the most suitable additional suggestions.

[1276] Ending Phase

[1277] The user can choose to either view the recommended products and decide to purchase them, or answer the questions again.

[1278] Based on the user's selection, the device will either generate further questions or guide them through the purchase process.

[1279] Specific example

[1280] For example, if a user answers "50 inches" as the TV size they are interested in, the emotion engine can analyze the user's facial expressions and tone of voice during the response to determine if they are truly interested. As a result, if they show a high level of interest, "50-inch TVs" will be strongly reflected in the recommendation list; if they show a low level of interest, TVs of other sizes or with different characteristics can be recommended again.

[1281] By incorporating user sentiment data in this way, it becomes possible to make more accurate product recommendations and increase user satisfaction.

[1282] The following describes the processing flow.

[1283] Step 1:

[1284] During system initialization, the server establishes connections to the product database and the question database and loads the necessary data. This includes the initial setup of the emotion engine.

[1285] Step 2:

[1286] The device displays an initial screen that allows the user to enter basic information. It also simultaneously displays interfaces for voice input and a facial recognition camera to acquire emotional data.

[1287] Step 3:

[1288] On the initial screen displayed, the user enters basic information (e.g., age, gender, categories of interest). Simultaneously, facial expressions and voice tone are captured through the camera and microphone built into the device.

[1289] Step 4:

[1290] The device receives basic information, facial expression data, and voice data, and sends them to the server.

[1291] Step 5:

[1292] The server analyzes the received basic information and emotional data, and prepares to generate initial questions from the question database. This includes a process that evaluates the user's psychological state based on their facial expressions and tone of voice.

[1293] Step 6:

[1294] The server generates appropriate initial questions based on basic information and analysis results, and sends them to the terminal. Example: "What size television are you interested in?"

[1295] Step 7:

[1296] The terminal displays the questions received from the server to the user.

[1297] Step 8:

[1298] The user answers questions. During this process, the device's camera captures the user's facial expressions, and the microphone records their voice tone.

[1299] Step 9:

[1300] The device sends the user's response along with newly acquired emotional data (facial expressions and voice) to the server.

[1301] Step 10:

[1302] The server receives user responses and sentiment data and stores them in a database. Simultaneously, it generates the next question or begins analysis to specifically infer the user's purpose for visiting the store.

[1303] Step 11:

[1304] The server uses natural language processing (NLP) techniques and an emotion engine to analyze the user's responses and emotional data. For example, if a user responds "A 50-inch TV is interesting" while smiling, the server infers their intention as "They have a strong interest in 50-inch TVs."

[1305] Step 12:

[1306] Based on the analysis results, the server generates a list of products suitable for the user from the product database. The generated list also takes into account the user's emotional state.

[1307] Step 13:

[1308] The server sends a list of the suspected purpose of visit and recommended products to the terminal.

[1309] Step 14:

[1310] The terminal displays the user's estimated purpose for visiting the store and a list of recommended products. Example: "Your purpose for visiting the store is to purchase a 50-inch TV. Please see the recommended products below."

[1311] Step 15:

[1312] The user reviews the displayed list of recommended products and selects items that interest them. At this point, the device continues to monitor the user's facial expressions and voice, collecting emotional data.

[1313] Step 16:

[1314] Based on the user's choices, the server analyzes further sentiment data to generate the next question or proceed with the purchase process.

[1315] Step 17:

[1316] Depending on the user's choice, the device will either display the question again or show a screen for the purchase process. If the user chooses to proceed with the purchase, the necessary options and input fields will be displayed.

[1317] Step 18:

[1318] The user proceeds with the purchase process, and the purchase is completed. During the purchase process, the device collects emotional data, allowing for monitoring of the user's satisfaction level.

[1319] This enables more accurate product recommendations that take user emotions into account, resulting in a more personalized and smoother purchasing experience throughout the entire process.

[1320] (Example 2)

[1321] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1322] Conventional user information collection systems generate questions and perform intent analysis based on basic user information and response history, but this alone makes it difficult to accurately grasp the user's true intentions and emotions. As a result, appropriate product recommendations may not be made, potentially lowering user satisfaction.

[1323] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a question generation means that generates an initial question based on the user's basic information and emotional data, an answer receiving means that receives the user's answer and emotional data, and an intent analysis means that analyzes the answer and emotional data to infer the user's purpose for visiting the store. By incorporating the user's emotional data into the analysis, it becomes possible to accurately grasp the user's intentions and emotions and make appropriate product suggestions.

[1324] "Basic information" refers to identifying information such as the user's age, gender, and categories of interest.

[1325] "Emotional data" refers to information about a user's emotions, obtained from their facial expressions, voice tone, and other sources.

[1326] A "question generation means" is a means for generating initial and subsequent questions based on the user's basic information and sentiment data.

[1327] "Display means" refers to an interface for displaying questions generated for the user and recommended products.

[1328] "Response receiving means" refers to a means of receiving user responses and sentiment data.

[1329] "Intention analysis methods" are techniques that analyze received responses and emotional data to infer the user's purpose for visiting the store.

[1330] A "product recommendation method" is a means of recommending appropriate products based on the purpose of the customer's visit, which is inferred by intent analysis methods.

[1331] "Natural language processing technology" is a technology that analyzes a user's language response and understands its meaning.

[1332] An "emotion engine" is software or a system used to analyze a user's emotional data.

[1333] This invention relates to a system that uses AI and an emotion engine to analyze the purpose of a user's visit to a consumer electronics store and recommend appropriate products. Based on the basic information and emotion data entered by the user, this system generates questions, receives the answers, analyzes the answers and emotion data to clarify the user's intentions, and makes optimal product suggestions.

[1334] System Configuration

[1335] 1. The server establishes connections to the product database and the question database during system initialization and loads the necessary data. Specifically, it uses MySQL as the database management system and IBM Watson Emotion Recognition as the sentiment analysis engine. This establishes the initial engine settings and database connections.

[1336] 2. The device displays a screen for the user to enter basic information. This screen also includes interfaces that utilize voice input (Google Speech-to-Text API) and a facial recognition camera (Azure Face API). This makes it possible to obtain emotional data from the user's facial expressions and voice.

[1337] 3. The user enters basic information (age, gender, categories of interest, etc.) on the initial screen that is displayed. The entered basic information is sent to the server by the device.

[1338] 4. The device transmits not only basic information but also emotional data obtained from voice input and facial recognition to the server.

[1339] 5. The server analyzes the received basic information and sentiment data to generate an initial question from the question database. For example, a specific question such as "What size television are you interested in?" is generated.

[1340] 6. The terminal displays questions sent from the server to the user. The user answers the displayed questions. When answering, facial expressions and voice are also captured through the terminal and simultaneously sent to the server as emotion data.

[1341] 7. The server stores the received responses in a database and uses an emotion engine to analyze the user's emotions at the time of the response. This allows the server to infer the user's specific intentions and purpose for visiting the store.

[1342] 8. Based on the analysis results, the server generates an appropriate product list from the product database. For example, if the intention "I want a 50-inch TV" is extracted, a product list specifically for 50-inch TVs will be generated.

[1343] 9. The terminal displays a list of recommended products sent from the server to the user. Sentimental data continues to be collected even after the display, making it possible to understand more precisely which products the user is interested in.

[1344] 10. The user reviews the displayed list of recommended products and selects items of interest. In some cases, they may answer further questions or proceed with the purchase.

[1345] Specific example

[1346] For example, if a user enters basic information such as "30 years old, male, interested in home appliances" and answers the question "I want a 50-inch TV," the server uses an emotion engine to analyze the user's facial expressions and tone of voice during the response. Based on this analysis, if the server determines that the user is genuinely very interested in a 50-inch TV, a 50-inch TV will be strongly reflected in the recommendation list. If the user shows little interest, a TV of a different size or with different characteristics will be recommended instead.

[1347] Example of a prompt

[1348] An example of a prompt to input into a generative AI model is: "A user showed interest in a 50-inch TV when they visited the store. Please explain the process for recommending products that match this user's interests."

[1349] By incorporating user emotional data into the analysis in this way, it becomes possible to provide a system that makes more accurate and appropriate product recommendations.

[1350] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1351] Step 1:

[1352] The server establishes connections to the product database and the question database during system initialization. Specifically, it uses a database management system (e.g., MySQL) to load the necessary data (e.g., product characteristics, question templates). It also performs the initial setup of the sentiment analysis engine (e.g., IBM Watson Emotion Recognition). The input is the database connection information and the sentiment engine configuration information, and the output is the connected database and the initialized sentiment engine.

[1353] Step 2:

[1354] The device displays an initial screen for the user to enter basic information. This screen also includes voice input (e.g., Google Speech-to-Text API) and a facial recognition camera (e.g., Azure Face API), providing an interface for acquiring emotional data from the user's facial expressions and voice. Input consists of basic information provided by the user (e.g., age, gender, categories of interest), and output consists of the user's basic information and the display of the initial screen.

[1355] Step 3:

[1356] The user enters basic information on the initial screen. This information includes age, gender, and categories of interest. The input is basic information, and the output is basic information sent to the device.

[1357] Step 4:

[1358] The terminal receives basic information entered by the user and sends it to the server. It also simultaneously collects emotional data obtained from voice input and facial recognition and sends this data to the server as well. The input consists of the user's basic information and emotional data, while the output is the data sent to the server.

[1359] Step 5:

[1360] The server analyzes the received basic information and sentiment data and generates an initial question from the question database. Specifically, it selects an appropriate question based on the received basic information and sentiment data, and generates a concrete question such as, "What size television are you interested in?" The input is basic information and sentiment data, and the output is the generated question.

[1361] Step 6:

[1362] The terminal displays the initial question sent from the server to the user. The input is the initial question from the server, and the output is the display of the question to the user.

[1363] Step 7:

[1364] The user answers the displayed questions. During the answering process, facial expressions and voice are also collected through a facial recognition camera and voice input function. Input consists of the questions, user answers, facial expressions, and voice data, while output consists of the answers and emotion data sent to the device.

[1365] Step 8:

[1366] The terminal receives user responses and sentiment data and sends them to the server. The input is the user's responses and sentiment data, and the output is the data sent to the server.

[1367] Step 9:

[1368] The server stores user responses in a database and uses an emotion engine to analyze the user's emotions at the time of their response. The input consists of the user's response and emotion data, while the output consists of the stored data and the analysis results.

[1369] Step 10:

[1370] The server analyzes the user's intent based on the response content and sentiment data using natural language processing technology (e.g., Google Natural Language API) and a sentiment engine. For example, it extracts a specific objective, such as "I want a 50-inch TV," along with sentiment data. The input is the user's response and sentiment data, and the output is the analyzed user intent.

[1371] Step 11:

[1372] The server generates a list of appropriate products from the product database based on the analysis results. The input is the analysis results, and the output is the generated product list.

[1373] Step 12:

[1374] The server sends the generated product list to the terminal. The input is the generated product list, and the output is the data sent to the terminal.

[1375] Step 13:

[1376] The terminal displays a list of recommended products to the user. For example, it might display, "Your purpose for visiting is to purchase a 50-inch television. Please see the recommended products below." The input is a list of recommended products from the server, and the output is the displayed product list.

[1377] Step 14:

[1378] Even while displaying recommended products, the device utilizes an emotion engine to continuously collect emotional data obtained from the user's facial expressions and voice, and transmits it to the server. The input is the user's facial expressions and voice data, and the output is the data transmitted to the server.

[1379] Step 15:

[1380] The server provides users with optimal additional suggestions based on collected sentiment data. The input is continuously collected sentiment data, and the output is data for additional suggestions.

[1381] Step 16:

[1382] The user can choose to view the recommended products and decide whether to purchase them, or to answer another question. The input is a list of recommended products or a new question, and the output is the user's choice or answer.

[1383] Step 17:

[1384] The terminal either generates new questions or guides the user through the purchase process based on their selection. The input is the user's selection, and the output is a new question or a purchase process screen.

[1385] (Application Example 2)

[1386] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1387] Existing consumer electronics retailers face challenges in accurately understanding what customers want to buy, hindering efficient product recommendations. Furthermore, recommending products without understanding user emotions or specific intentions can lead to decreased customer satisfaction. Moreover, with the rise of online shopping, there's a need to enhance the unique customer experience offered by physical stores. To address these challenges, a system is needed that utilizes not only basic customer information but also emotional data to accurately predict purchasing intent and provide precise product recommendations.

[1388] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes an input means for inputting the user's basic information, a question generation means for generating initial questions based on the basic information, a display means for displaying the generated questions to the user, an answer receiving means for receiving the user's answers and emotional data, an intent analysis means for analyzing the answers and emotional data to infer the purpose of the visit, a product recommendation means for recommending appropriate products based on the inferred purpose of the visit, and a display means for displaying the recommended products. This makes it possible to accurately grasp the user's emotions and intentions and make product suggestions that match the purpose of the visit, thereby improving customer satisfaction.

[1389] A "user" is a customer who visits an electronics retail store and uses the system.

[1390] "Basic information" refers to data that users enter, such as age, gender, and categories of interest.

[1391] "Input method" refers to an interface that allows users to input basic information, and includes devices such as smartphones and smart glasses.

[1392] "Question generation method" refers to the process of creating appropriate questions based on the basic information entered by the user.

[1393] "Display means" refers to displays or screens used to present generated questions and recommended products to the user.

[1394] A "response receiving method" refers to an interface that receives responses and sentiment data entered by users in response to questions.

[1395] "Emotional data" refers to emotional information obtained from the user's facial expressions, voice, and other sources.

[1396] "Intention analysis means" refers to a function that analyzes user responses and emotional data to infer the user's purpose for visiting the store.

[1397] "Natural language processing technology" refers to artificial intelligence technology used to understand and interpret user text data.

[1398] An "emotion engine" refers to an algorithm and software that analyzes a user's emotions from their facial expressions and voice.

[1399] "Product recommendation methods" refer to the process of selecting and recommending appropriate products based on the user's purpose for visiting the store.

[1400] A "server" is a central device that manages the entire system and performs tasks such as data analysis, storage, and query generation.

[1401] To implement the present invention, terminals such as smartphones and smart glasses, and servers connected to them, are used in consumer electronics stores. The main components of the system include input means for inputting basic user information, question generation means for generating questions, display means for displaying the generated questions, answer receiving means for receiving user answers and sentiment data, intent analysis means for analyzing user answers and sentiment data, product recommendation means for recommending appropriate products, and display means for displaying recommended products.

[1402] The server connects to the product database and question database during system initialization and performs the initial setup of the emotion engine. It uses "EmotionEngine" as the emotion engine and "NlpProcessor" as the natural language processing technology. The server generates questions based on the user's basic information and past answer history and sends them to the terminal.

[1403] The device takes the form of a smartphone or smart glasses and displays an initial screen for the user to input basic information. The device also includes features such as voice input and a facial recognition camera, which are used to collect emotional data. The basic information entered by the user is sent to a server, which then generates appropriate initial questions based on this information.

[1404] Users respond to questions displayed on their device using voice or touch input. The device sends these responses to a server, which analyzes the emotional data acquired along with the responses. By using an emotion engine to read the user's emotions from their facial expressions and voice, and by analyzing the content of their responses using NLP technology, the server infers the user's purpose for visiting the store.

[1405] The server selects appropriate products from its product database based on the inferred purpose of the visit, generates a list of recommended products, and sends it to the terminal. The terminal displays this list to the user, who then selects products that interest them. Throughout this process, the server continuously collects sentiment data and provides recommendations tailored to the user's preferences.

[1406] For example, when a user responds that they are "looking for a large refrigerator," the system analyzes the user's facial expressions and tone of voice to estimate whether their interest is genuine. Based on these results, "refrigerators of 500 liters or more" are strongly reflected in the recommendation list.

[1407] Examples of prompt statements used as input to a generative AI model are as follows:

[1408] Please generate questions to infer the user's purpose of purchase. The basic user information is as follows:

[1409] Age: 35

[1410] Gender: Male

[1411] Categories of interest: Home appliances

[1412] Please generate questions based on the following data:

[1413] Emotional data: Joy (high)

[1414] Example questions to output:

[1415] "What size refrigerator in liters are you interested in?"

[1416] This makes it possible to accurately understand the user's emotions and intentions, and by suggesting products that match their purpose for visiting the store, customer satisfaction can be improved.

[1417] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1418] Step 1:

[1419] During system initialization, the server establishes connections to the product database and question database and performs initial setup of the sentiment engine. At this time, the server reads product information and question templates and sets initial parameters for sentiment analysis.

[1420] Input: Product database, question database, emotion engine configuration information

[1421] Output: System state after initial setup is complete

[1422] Specific actions: Connecting to the database, loading configuration files.

[1423] Step 2:

[1424] The device displays an initial screen where the user can enter basic information. The device is also equipped with voice input and a facial recognition camera, providing an interface to acquire user emotion data.

[1425] Input: System initial settings

[1426] Output: Basic Information Input Screen

[1427] Specific actions: rendering the initial screen, voice input, and camera activation.

[1428] Step 3:

[1429] The user enters basic information (age, gender, and categories of interest) using the initial screen displayed. The device also acquires emotional data through voice input and facial recognition.

[1430] Input: User's basic information, voice and facial recognition data

[1431] Output: User basic information, sentiment data

[1432] Specific operations: Collection of user input data, real-time acquisition of sentiment data.

[1433] Step 4:

[1434] The terminal analyzes the collected basic information and emotional data and sends it to the server. The server receives this data and prepares it for analysis.

[1435] Input: User's basic information, sentiment data

[1436] Output: Data ready for analysis

[1437] Specific actions: Packing input data and sending it to the server.

[1438] Step 5:

[1439] The server generates an initial question from the question database based on the user's basic information and sentiment data.

[1440] Input: User's basic information, sentiment data

[1441] Output: Initial Question

[1442] Specific actions: Execute database queries, apply query templates.

[1443] Step 6:

[1444] The device displays the generated questions to the user. The user enters answers to the questions and expresses emotions through voice and facial expressions.

[1445] Input: Initial Question

[1446] Output: User responses, sentiment data

[1447] Specific actions: Displaying questions, providing an interface for entering answers.

[1448] Step 7:

[1449] The device sends the user's responses and sentiment data to the server, which receives, stores, and then analyzes this data.

[1450] Input: User responses, sentiment data

[1451] Output: Analysis results

[1452] Specific actions: Packing data, sending it to the server, and saving it to the database.

[1453] Step 8:

[1454] The server uses natural language processing technology and an emotion engine to analyze the user's responses and emotion data, and to infer the user's purpose for visiting the store.

[1455] Input: User responses, sentiment data

[1456] Output: Estimated results of the purpose of visit

[1457] Specific actions: Application of NLP algorithms, performance of sentiment analysis.

[1458] Step 9:

[1459] Based on the inferred purpose of the customer's visit, the server selects appropriate products from the product database and generates a list of recommended products.

[1460] Input: Estimated result of the purpose of visit

[1461] Output: Recommended product list

[1462] Specific actions: Execute database queries, select recommended products.

[1463] Step 10:

[1464] The terminal displays a list of recommended products to the user, who then selects products of interest. The server continuously collects sentiment data and provides optimal additional suggestions.

[1465] Input: Recommended product list, user selection

[1466] Output: Optimal product proposal

[1467] Specific actions: Rendering of recommended products, real-time analysis of sentiment data.

[1468] Step 11:

[1469] When a user selects a product or answers additional questions, the device sends data back to the server, which then asks the next most appropriate question or suggests a product as needed.

[1470] Input: User selection, answers to additional questions

[1471] Output: Next question or product suggestion

[1472] Specific actions: data retransmission, reapplication of analysis algorithms.

[1473] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1474] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1475] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1476] [Fourth Embodiment]

[1477] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1478] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1479] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1480] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1481] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1482] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1483] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1484] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1485] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1486] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1487] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1488] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1489] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1490] As an embodiment of the present invention, a system that uses AI to predict the purchase purpose of a customer visiting a consumer electronics store will be described. This system has a process of taking the user's basic information as input, generating questions based on that information, and analyzing the user's answers to infer the purpose of the visit.

[1491] Initialization process

[1492] The server reads the product database and the question database, and prepares the necessary information in advance when the system starts up.

[1493] The device displays an initial screen that allows the user to enter basic information.

[1494] The user enters basic information (e.g., age, gender, categories of interest) on the displayed screen.

[1495] User information entry

[1496] The terminal receives basic information entered by the user and sends it to the server.

[1497] The server analyzes the received basic information and prepares to generate appropriate initial questions.

[1498] question generation

[1499] Based on the basic information, the server selects a suitable question from the question database and generates a question to display to the user.

[1500] Example: "What size TV screen are you interested in?"

[1501] The terminal displays the generated question to the user.

[1502] The user answers the question and sends the answer to the server via their device.

[1503] Receiving and analyzing responses

[1504] The server receives the user's response and saves it to its internal database.

[1505] The server analyzes the user's responses and processes them to generate the next question or to specifically infer the purpose of their visit.

[1506] Intent Analysis

[1507] The server uses natural language processing (NLP) techniques to analyze user responses and infer the user's purpose for visiting the store.

[1508] Example: Analyze the intention behind the statement, "I want a 50-inch TV."

[1509] Based on the analysis results, the server selects appropriate products from the product database and generates a list of recommended products.

[1510] Displaying recommended results

[1511] The server sends a list of recommended products along with an estimated reason for the customer's visit to the terminal.

[1512] The terminal displays to the user a list of suggested reasons for visiting the store and recommended products.

[1513] Example: "Your purpose for visiting our store is to purchase a 50-inch television. Please see our recommended products below."

[1514] The user reviews the recommended product list and selects the products they are interested in.

[1515] Ending Phase

[1516] The user can choose to either view the recommended products and decide to purchase them, or answer the questions again.

[1517] Based on the user's selection, the device will either generate further questions or guide them through the purchase process.

[1518] As described above, the system can start with user input, go through AI analysis, and ultimately present recommended products. This system allows customers to efficiently reach their intended purpose for visiting the store, providing a smooth shopping experience.

[1519] The following describes the processing flow.

[1520] Step 1:

[1521] During system initialization, the server establishes connections between the product database and the question database and loads the necessary data. This prepares the information required for system operation.

[1522] Step 2:

[1523] The device displays an initial screen to the user and provides an input form for basic information. This is for entering information such as age, gender, and categories of interest.

[1524] Step 3:

[1525] The user enters their basic information into the input form displayed on this initial screen and clicks the submit button.

[1526] Step 4:

[1527] The terminal receives basic information entered by the user and sends that information to the server.

[1528] Step 5:

[1529] The server analyzes the received basic information and prepares to generate initial questions from the question database. Based on this analysis, it selects appropriate questions related to the user's interests.

[1530] Step 6:

[1531] The server sends the generated initial question to the terminal and displays it to the user.

[1532] Step 7:

[1533] The terminal displays an initial question received from the server to the user. For example, it might display, "What size television are you interested in?"

[1534] Step 8:

[1535] The user answers the displayed questions and sends those answers to the server via their device.

[1536] Step 9:

[1537] The server receives the user's response and stores it in the database. At the same time, it either performs analysis to generate the next question or starts processing to specifically infer the purpose of the visit.

[1538] Step 10:

[1539] The server uses natural language processing (NLP) techniques to analyze user responses and infer user-specific intentions and reasons for visiting the store. For example, it might extract specific objectives such as "I want a 50-inch TV."

[1540] Step 11:

[1541] Based on the analysis results, the server generates a list of products suitable for the user from the product database. This list also includes detailed information about recommended products.

[1542] Step 12:

[1543] The server sends a list of the suspected purpose of the visit and recommended products to the terminal.

[1544] Step 13:

[1545] The terminal displays the user's estimated purpose for visiting the store and a list of recommended products. For example, it might say, "Your purpose for visiting the store is to purchase a 50-inch TV. Please see the recommended products below."

[1546] Step 14:

[1547] The user reviews the displayed list of recommended products and selects the products they are interested in.

[1548] Step 15:

[1549] The user clicks on the selected product to view details. They then choose to proceed with the purchase or answer further questions.

[1550] Step 16:

[1551] The device will determine the next action based on the user's selection. If the user chooses to answer the question again, it will send a request to the server and restart the question generation process. If the user chooses to proceed with the purchase, it will display the appropriate purchase process screen.

[1552] Step 17:

[1553] The server generates new questions and performs processes related to the purchase procedure, returning appropriate feedback based on the user's choices. This completes the entire process.

[1554] Through these steps, the system can efficiently understand the user's purpose for visiting the store and suggest appropriate products.

[1555] (Example 1)

[1556] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1557] In modern consumer electronics stores, quickly and accurately understanding a customer's purpose for visiting is crucial for providing a smooth shopping experience. However, existing systems struggle to efficiently carry out the entire process from user input to question generation, response analysis, intent inference, and product recommendation. This invention aims to solve these problems and provide a system that presents users with appropriate questions, quickly and accurately infers their purpose for visiting, and recommends appropriate products.

[1558] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1559] In this invention, the server includes an input means for inputting the user's basic information, a question generation means for generating an initial question based on the basic information received from the input means, a display means for displaying the generated question to the user, an answer receiving means for receiving the user's answer, an intent analysis means for analyzing the answer received from the answer receiving means and inferring the user's purpose for visiting the store, a product recommendation means for recommending appropriate products based on the purpose of visiting the store inferred by the intent analysis means, a display means for displaying the recommended products to the user, a communication means for performing question generation, answer reception, intent analysis, product recommendation, and display via a terminal and the server, and a next question generation means for generating the next question that the user is likely to be interested in based on the generated list of recommended products. This makes it possible to efficiently infer the purpose of visiting the store from the user's basic information and answers and to quickly recommend appropriate products.

[1560] "User basic information" refers to personal information that users enter, such as age, gender, and categories of interest.

[1561] An "input means" is a device or interface for a user to input basic information.

[1562] A "question generation means" is an algorithm or system that automatically generates appropriate questions based on the user's basic information.

[1563] A "display means" is an interface that visually presents generated questions and recommended products to the user.

[1564] A "response receiving system" is a system that receives responses entered by users and saves them in an appropriate data format.

[1565] A "means of intent analysis" is a system that uses natural language processing technology to analyze user responses and infer their purpose for visiting the store.

[1566] A "product recommendation system" is a system that selects and recommends appropriate products to users based on their purpose for visiting the store, as inferred by intent analysis.

[1567] "Communication methods" refer to network infrastructure and protocols used to send and receive data between a server and a terminal.

[1568] "Next question generation means" refers to an algorithm or system that generates the next appropriate question based on the user's basic information and previous answers.

[1569] A "recommended product list" is a list of products suitable for the user's purpose of visiting the store, and represents a group of product options presented to the user.

[1570] To implement the present invention, a system is needed that takes basic user information as input, uses AI to infer the purpose of the visit based on that information, and recommends appropriate products. This system involves cooperation between a server, a terminal, and the user, and includes the following main components and processes.

[1571] Hardware and software

[1572] Hardware used: A typical server device (e.g., Dell PowerEdge R740) and a terminal device for user operation (e.g., iPad Pro)

[1573] Software used: Database management system (e.g., MySQL), Web framework (e.g., Flask for Python), Natural Language Processing library (e.g., NLTK for Python), Frontend framework (e.g., React)

[1574] Specific description of the system's operation

[1575] The server first loads the product database and the question database, preparing the information needed when the system starts up. This includes executing SQL queries such as "SELECT FROM products" and "SELECT FROM questions," for example.

[1576] The device displays an initial screen where the user can enter basic information (age, gender, categories of interest, etc.). This initial screen is built using React and designed to be easy for the user to use.

[1577] The user enters their basic information into the form displayed on the initial screen. For example, the user might enter information such as age "30 years old," gender "female," and category of interest "home appliances."

[1578] The terminal receives the basic information entered by the user and sends it to the server via an HTTP POST request using the Axios library. Specifically, the terminal sends {"age": 30, "gender": "female", "category": "electronics"} in JSON format to "POST / api / userinfo".

[1579] The server analyzes the received basic information and selects an appropriate question from the question database. The technology used here is the Python natural language processing library NLTK. Based on the user's age (30 years old), gender (female), and category (home appliances), the server generates the question, "What size TV are you interested in?"

[1580] The generated question is sent from the server to the terminal, which then displays it to the user. The user then enters an answer to the displayed question. For example, the user might answer "50 inches".

[1581] The terminal receives the user's response and sends it back to the server using an HTTP POST request. The server stores the received response in its internal database and analyzes it using NLTK. This allows the server to infer the user's purpose for visiting the store, such as "the user is looking for a 50-inch television."

[1582] Based on the inferred purpose of the visit, the server selects appropriate products from the product database and generates a list of recommended products. This list is then sent back to the terminal and displayed to the user. Based on the displayed list of recommended products, the user can select products.

[1583] Example prompt statements

[1584] Prompt example: "The user is 30 years old, female, and interested in home electronics. Please generate the first question to infer their purpose for visiting."

[1585] As demonstrated here, the present invention can provide an efficient and smooth purchasing experience through a series of processes that begin with user input, go through server processing, and display the results on the terminal. This makes it possible to quickly identify the customer's purchase purpose and support them in selecting appropriate products.

[1586] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1587] System program processing flow

[1588] Step 1:

[1589] Initialization process

[1590] The server reads the product database and the question database, preparing the information needed when the system starts up. This includes executing SQL queries such as "SELECT FROM products" and "SELECT FROM questions" from the MySQL database to load all the data into memory.

[1591] Input: Database connection information

[1592] Output: Product data and question data loaded into memory

[1593] Step 2:

[1594] Initial screen display

[1595] The device displays an initial screen for the user to enter basic information. This screen is designed using JavaScript and the React framework.

[1596] Input: UI design information

[1597] Output: Screen displaying an input form to the user

[1598] Step 3:

[1599] Entering basic information

[1600] The user enters their basic information (e.g., age "30", gender "female", category of interest "home appliances") into the form displayed on the initial screen.

[1601] Input: User's basic information

[1602] Output: Basic information data sent to the terminal

[1603] Step 4:

[1604] Receiving and sending basic information

[1605] The terminal receives the basic information entered by the user and sends it to the server using an HTTP POST request. The Axios library is used to send the data in JSON format to "POST / api / userinfo".

[1606] Input: User's basic information

[1607] Output: Basic information sent to the server

[1608] Step 5:

[1609] question generation

[1610] The server analyzes the received basic information and selects a suitable question from the question database. Using Python's NLTK, it selects a question based on information such as "the user is 30 years old, female, and in the home appliance category."

[1611] Input: User's basic information

[1612] Output: Generated question (e.g., "What size TV screen are you interested in?")

[1613] Step 6:

[1614] Submitting and displaying questions

[1615] The server sends the generated question to the terminal.

[1616] The terminal displays the received question to the user.

[1617] Input: Generated Question

[1618] Output: Questions displayed to the user

[1619] Step 7:

[1620] User response input

[1621] The user enters their answers to the questions displayed on the device.

[1622] Input: User's response (e.g., "50 inches")

[1623] Output: Response data sent to the terminal

[1624] Step 8:

[1625] Receiving and sending responses

[1626] The terminal receives the user's response and sends it to the server using an HTTP POST request.

[1627] Input: User's response

[1628] Output: Response sent to the server

[1629] Step 9:

[1630] Saving and analyzing responses

[1631] The server stores the received responses in an internal database and then uses NLTK again to analyze and infer the user's intent.

[1632] Input: User's response

[1633] Output: Estimated purpose of visit (e.g., "The user is looking for a 50-inch TV")

[1634] Step 10:

[1635] Product Recommendation

[1636] The server selects appropriate products from the product database based on the analysis results and generates a list of recommended products. The SQL query "SELECT FROM products WHERE category='TV' AND size='50 inches'" is executed.

[1637] Input: Estimated purpose of visit

[1638] Output: Recommended product list

[1639] Step 11:

[1640] Sending and displaying recommended products

[1641] The server sends a list of recommended products and the presumed purpose of the visit to the terminal.

[1642] The device displays a list of recommended products to the user.

[1643] Input: Recommended Conlist

[1644] Output: Recommended product list displayed to the user

[1645] Step 12:

[1646] User choices and next actions

[1647] The user reviews the recommended product list and can choose to select products they are interested in, or choose to answer the questions again.

[1648] Input: User Selection

[1649] Output: Further questions or instructions for the purchase process

[1650] Step 13:

[1651] Continue with your question or purchase

[1652] Based on the user's selection, the device will either generate further questions or guide them through the purchase process.

[1653] Input: User Selection

[1654] Output: Display of the next question or guidance to the purchase page

[1655] This makes the entire process, starting with user input, going through server analysis, and displaying the results on the terminal, concretely clear.

[1656] (Application Example 1)

[1657] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1658] Traditional customer service systems in physical stores have struggled to accurately understand customers' purchase objectives and recommend appropriate products. This often resulted in an inefficient and smooth customer purchasing experience, potentially leading to decreased customer satisfaction. Furthermore, conventional systems sometimes failed to fully utilize basic customer information and past history, resulting in inappropriate question generation. To address these issues, this invention provides a system that uses AI to accurately predict customers' purchase objectives and recommend the most suitable products.

[1659] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1660] In this invention, the server includes an input means for inputting basic user information, a question generation means for generating initial questions based on the basic information received from the input means, a display means for displaying the generated questions to the user, an answer receiving means for receiving the user's answers, an intent analysis means for analyzing the answers received from the answer receiving means and inferring the user's purpose for visiting the store, a product recommendation means for recommending appropriate products based on the purpose of visiting the store inferred by the intent analysis means, a display means for displaying the recommended products to the user, an initialization means for obtaining a product database and a question database from the server via a specially designed computer program, and a recommendation means for analyzing the user's answers using natural language processing technology and recommending products based on the results. This makes it possible to accurately infer the customer's purchase purpose and efficiently recommend appropriate products.

[1661] "User" refers to consumers or users who utilize a system.

[1662] "Basic information" refers to information necessary for generating initial questions, such as the user's age, gender, and areas of interest.

[1663] "Input means" refers to devices or interfaces that users use to input basic information.

[1664] "Question generation means" refers to a system or algorithm that has the function of generating the next question to be displayed based on the user's basic information.

[1665] "Display means" refers to displays, screens, or interfaces used to show generated questions and recommended products to the user.

[1666] "Response receiving means" refers to the device or system used to receive responses entered by the user.

[1667] "Intention analysis methods" refer to natural language processing techniques and algorithms used to analyze user responses and infer their purchase intentions.

[1668] "Product recommendation methods" refer to systems and algorithms that select appropriate products based on the purpose of a customer's visit, as inferred by intent analysis methods, and recommend them to the user.

[1669] "Initialization means" refers to the function of retrieving the product database and question database from the server via a specially designed computer program.

[1670] "Recommendation methods" refer to a function that uses natural language processing technology to analyze user responses and recommends products based on the results.

[1671] A "server" refers to a central control unit that manages and processes data for the entire system.

[1672] A "product database" refers to a collection of data that stores information about products that are sold.

[1673] A "question database" refers to a collection of data that stores questions to be displayed to users.

[1674] This invention is a system that uses AI to predict customers' purchase intentions in physical stores and efficiently recommends appropriate products. Specific embodiments of this invention are shown below.

[1675] Hardware and software to be used

[1676] Server: A central control unit that manages and processes data for the entire system. It manages the product database and the inquiry database.

[1677] Device: Uses a smartphone to interact with users. This includes displaying questions, receiving answers, and showing recommended products.

[1678] software:

[1679] Python: Execute a script.

[1680] requests: Sends HTTP requests and communicates with the API.

[1681] scikit-learn: NLP analysis and database search.

[1682] Natural Language Processing (NLP): Used to analyze user responses and infer their purpose for visiting the store.

[1683] System configuration and operation

[1684] Initialization

[1685] The server retrieves and initializes the product database and question database from the server. This process utilizes the server API via a specially designed computer program.

[1686] Entering user information

[1687] The device provides an interface for users to input basic information (age, gender, and categories of interest). Users input this information through their smartphone's UI.

[1688] question generation

[1689] The server generates an initial question based on the basic information entered. This question is selected from a question database and displayed on the terminal. For example, a question such as "What size television are you interested in?" might be generated.

[1690] Receiving and analyzing responses

[1691] The user answers questions displayed on the device, and the device sends those answers to the server. The server analyzes the received answers and uses natural language processing techniques to infer their intent. For example, the answer "I want a 50-inch TV" is analyzed.

[1692] Product Recommendation

[1693] Based on the intent analysis results, the server selects appropriate products from the product database and generates a list of recommended products. This list of recommended products is displayed to the user via their terminal. For example, it might say, "Your purpose for visiting the store is to purchase a 50-inch television. Please see the recommended products below."

[1694] Display and select from the recommended product list

[1695] The terminal displays a list of recommended products to the user, who reviews them and selects products of interest. The server then generates additional questions as needed and makes further recommendations.

[1696] Examples of specific prompt messages

[1697] User's basic information: Age 30, Gender Male, Interests: Television

[1698] Initial question: What size television screen are you interested in?

[1699] User reply: I want a 50-inch TV.

[1700] AI analysis results: The recommended product is a 50-inch television, "BRAND A 50-inch".

[1701] Based on the above, the present invention provides a system that can accurately predict the purchase purpose of a store visitor and efficiently recommend appropriate products. This makes it possible to improve customer satisfaction and achieve a smooth purchasing experience.

[1702] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1703] Step 1:

[1704] The server loads and initializes the product database and the question database. Specifically, the server retrieves each database via the server API and loads it into memory. At this time, it sends an HTTP request using a specially designed computer program to download the database files. The database format is JSON or CSV, and the loaded data is stored in internal data storage.

[1705] Input: Database URL obtained from the server API

[1706] Output: Product database and question database

[1707] Step 2:

[1708] The device displays an interface for the user to enter basic information. Specifically, a basic information input form (e.g., age, gender, categories of interest) is displayed on the smartphone screen. The user enters their information into these fields and presses the submit button.

[1709] Input: Smartphone basic information input form

[1710] Output: Basic information entered by the user

[1711] Step 3:

[1712] The server receives basic user information sent from the terminal and generates an initial question. The server retrieves an appropriate question from the question database and customizes it based on the user information. Natural language generation technology may be used in this process. For example, if the user is "30 years old and their interest category is television," the server will generate the question, "What size television are you interested in?"

[1713] Input: Basic information sent from the device

[1714] Output: Initial Question

[1715] Step 4:

[1716] The device displays the generated initial question to the user. Specifically, the question text and answer input fields are displayed on the smartphone screen. The user enters their answer to the question and presses the submit button.

[1717] Input: Initial question received from the server

[1718] Output: User's response

[1719] Step 5:

[1720] The server receives user responses sent from the terminal and analyzes them. Natural language processing techniques are used for intent analysis, and the response content is vectorized for analysis. As a result of the analysis, the user's purpose for visiting the store is identified. For example, from the response "I want a 50-inch TV," the intention "to purchase a 50-inch TV" is inferred.

[1721] Input: User's response sent from the device

[1722] Output: Results of identifying the purpose of visit

[1723] Step 6:

[1724] The server selects appropriate products from the product database based on the intent analysis results. Machine learning algorithms are used to search the product database and find the best product for the user's needs. For example, a list of products suitable for a "50-inch television" is generated.

[1725] Input: Results of intent analysis

[1726] Output: Recommended product list

[1727] Step 7:

[1728] The device displays a list of recommended products to the user. Specifically, a list of recommended products is displayed on the smartphone screen. The user can select a product they are interested in from this list and view detailed information.

[1729] Input: Recommended product list received from the server

[1730] Output: Recommended product list displayed on the smartphone screen

[1731] Step 8:

[1732] The user reviews the recommended product list and chooses whether to purchase the product or answer the questions again. The terminal then performs the following actions based on the user's choice. If the user answers the questions again, the process from step 3 onwards is repeated.

[1733] Input: User selection (purchase decision or follow-up question)

[1734] Output: Purchase procedure or generation of next question

[1735] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1736] As an embodiment of the present invention, a system that uses AI and an emotion engine to predict the purchase purpose of a customer visiting a consumer electronics store is described. This system has a process of taking the user's basic information as input, generating questions based on that information, and analyzing the user's answers and emotions to infer the purpose of the visit.

[1737] Initialization process

[1738] During system initialization, the server establishes connections to the product database and the question database and loads the necessary data. This includes the initial setup of the emotion engine.

[1739] The device displays an initial screen that allows the user to enter basic information. This screen also includes interfaces for capturing emotions, such as voice input and a facial recognition camera.

[1740] The user enters basic information (e.g., age, gender, categories of interest) on the initial screen that is displayed.

[1741] User information entry

[1742] The terminal receives basic information entered by the user and sends it to the server. It also obtains user emotion data from voice input and facial recognition and sends this to the server as well.

[1743] The server analyzes the received basic information and sentiment data, and prepares to generate initial questions from the question database.

[1744] question generation

[1745] Based on basic information and sentiment data, the server selects appropriate questions from the question database and generates questions to display to the user.

[1746] Example: "What size TV screen are you interested in?"

[1747] The terminal displays the generated question to the user.

[1748] The user answers questions and sends their answers to the server via their device. The user's facial expressions and voice during the answering process are also captured as emotion data.

[1749] Receiving and analyzing responses

[1750] The server receives the user's response and stores it in an internal database. Furthermore, it uses an emotion engine to analyze the user's emotions at the time of the response.

[1751] The server analyzes the responses and sentiment data, and then starts processing to generate the next question or to specifically infer the customer's purpose for visiting.

[1752] Intent Analysis

[1753] The server uses natural language processing (NLP) technology and an emotion engine to analyze user responses and emotion data, inferring the user's specific intentions and reasons for visiting the store. For example, it might extract a specific purpose, such as "I want a 50-inch TV," along with emotion data.

[1754] Based on the analysis results, the server generates a list of products suitable for the user from the product database. This list also includes detailed information about recommended products.

[1755] Displaying recommended results

[1756] The server sends a list of the suspected purpose of the visit and recommended products to the terminal.

[1757] The terminal displays the user's estimated purpose for visiting the store and a list of recommended products. For example, it might say, "Your purpose for visiting the store is to purchase a 50-inch TV. Please see the recommended products below."

[1758] The user reviews the displayed list of recommended products and selects the products they are interested in.

[1759] Utilization of emotional data

[1760] The server utilizes its emotion engine even while displaying recommended products, continuously collecting emotional data obtained from the user's facial expressions and voice. This allows for a more precise understanding of which products the user is most interested in.

[1761] The server can use emotional data to provide users with the most suitable additional suggestions.

[1762] Ending Phase

[1763] The user can choose to either view the recommended products and decide to purchase them, or answer the questions again.

[1764] Based on the user's selection, the device will either generate further questions or guide them through the purchase process.

[1765] Specific example

[1766] For example, if a user answers "50 inches" as the TV size they are interested in, the emotion engine can analyze the user's facial expressions and tone of voice during the response to determine if they are truly interested. As a result, if they show a high level of interest, "50-inch TVs" will be strongly reflected in the recommendation list; if they show a low level of interest, TVs of other sizes or with different characteristics can be recommended again.

[1767] By incorporating user sentiment data in this way, it becomes possible to make more accurate product recommendations and increase user satisfaction.

[1768] The following describes the processing flow.

[1769] Step 1:

[1770] During system initialization, the server establishes connections to the product database and the question database and loads the necessary data. This includes the initial setup of the emotion engine.

[1771] Step 2:

[1772] The device displays an initial screen that allows the user to enter basic information. It also simultaneously displays interfaces for voice input and a facial recognition camera to acquire emotional data.

[1773] Step 3:

[1774] On the initial screen displayed, the user enters basic information (e.g., age, gender, categories of interest). Simultaneously, facial expressions and voice tone are captured through the camera and microphone built into the device.

[1775] Step 4:

[1776] The device receives basic information, facial expression data, and voice data, and sends them to the server.

[1777] Step 5:

[1778] The server analyzes the received basic information and emotional data, and prepares to generate initial questions from the question database. This includes a process that evaluates the user's psychological state based on their facial expressions and tone of voice.

[1779] Step 6:

[1780] The server generates appropriate initial questions based on basic information and analysis results, and sends them to the terminal. Example: "What size television are you interested in?"

[1781] Step 7:

[1782] The terminal displays the questions received from the server to the user.

[1783] Step 8:

[1784] The user answers questions. During this process, the device's camera captures the user's facial expressions, and the microphone records their voice tone.

[1785] Step 9:

[1786] The device sends the user's response along with newly acquired emotional data (facial expressions and voice) to the server.

[1787] Step 10:

[1788] The server receives user responses and sentiment data and stores them in a database. Simultaneously, it generates the next question or begins analysis to specifically infer the user's purpose for visiting the store.

[1789] Step 11:

[1790] The server uses natural language processing (NLP) techniques and an emotion engine to analyze the user's responses and emotional data. For example, if a user responds "A 50-inch TV is interesting" while smiling, the server infers their intention as "They have a strong interest in 50-inch TVs."

[1791] Step 12:

[1792] Based on the analysis results, the server generates a list of products suitable for the user from the product database. The generated list also takes into account the user's emotional state.

[1793] Step 13:

[1794] The server sends a list of the suspected purpose of visit and recommended products to the terminal.

[1795] Step 14:

[1796] The terminal displays the user's estimated purpose for visiting the store and a list of recommended products. Example: "Your purpose for visiting the store is to purchase a 50-inch TV. Please see the recommended products below."

[1797] Step 15:

[1798] The user reviews the displayed list of recommended products and selects items that interest them. At this point, the device continues to monitor the user's facial expressions and voice, collecting emotional data.

[1799] Step 16:

[1800] Based on the user's choices, the server analyzes further sentiment data to generate the next question or proceed with the purchase process.

[1801] Step 17:

[1802] Depending on the user's choice, the device will either display the question again or show a screen for the purchase process. If the user chooses to proceed with the purchase, the necessary options and input fields will be displayed.

[1803] Step 18:

[1804] The user proceeds with the purchase process, and the purchase is completed. During the purchase process, the device collects emotional data, allowing for monitoring of the user's satisfaction level.

[1805] This enables more accurate product recommendations that take user emotions into account, resulting in a more personalized and smoother purchasing experience throughout the entire process.

[1806] (Example 2)

[1807] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1808] Conventional user information collection systems generate questions and perform intent analysis based on basic user information and response history, but this alone makes it difficult to accurately grasp the user's true intentions and emotions. As a result, appropriate product recommendations may not be made, potentially lowering user satisfaction.

[1809] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a question generation means that generates an initial question based on the user's basic information and emotional data, an answer receiving means that receives the user's answer and emotional data, and an intent analysis means that analyzes the answer and emotional data to infer the user's purpose for visiting the store. By incorporating the user's emotional data into the analysis, it becomes possible to accurately grasp the user's intentions and emotions and make appropriate product suggestions.

[1810] "Basic information" refers to identifying information such as the user's age, gender, and categories of interest.

[1811] "Emotional data" refers to information about a user's emotions, obtained from their facial expressions, voice tone, and other sources.

[1812] A "question generation means" is a means for generating initial and subsequent questions based on the user's basic information and sentiment data.

[1813] "Display means" refers to an interface for displaying questions generated for the user and recommended products.

[1814] "Response receiving means" refers to a means of receiving user responses and sentiment data.

[1815] "Intention analysis methods" are techniques that analyze received responses and emotional data to infer the user's purpose for visiting the store.

[1816] A "product recommendation method" is a means of recommending appropriate products based on the purpose of the customer's visit, which is inferred by intent analysis methods.

[1817] "Natural language processing technology" is a technology that analyzes a user's language response and understands its meaning.

[1818] An "emotion engine" is software or a system used to analyze a user's emotional data.

[1819] This invention relates to a system that uses AI and an emotion engine to analyze the purpose of a user's visit to a consumer electronics store and recommend appropriate products. Based on the basic information and emotion data entered by the user, this system generates questions, receives the answers, analyzes the answers and emotion data to clarify the user's intentions, and makes optimal product suggestions.

[1820] System Configuration

[1821] 1. The server establishes connections to the product database and the question database during system initialization and loads the necessary data. Specifically, it uses MySQL as the database management system and IBM Watson Emotion Recognition as the sentiment analysis engine. This establishes the initial engine settings and database connections.

[1822] 2. The device displays a screen for the user to enter basic information. This screen also includes interfaces that utilize voice input (Google Speech-to-Text API) and a facial recognition camera (Azure Face API). This makes it possible to obtain emotional data from the user's facial expressions and voice.

[1823] 3. The user enters basic information (age, gender, categories of interest, etc.) on the initial screen that is displayed. The entered basic information is sent to the server by the device.

[1824] 4. The device transmits not only basic information but also emotional data obtained from voice input and facial recognition to the server.

[1825] 5. The server analyzes the received basic information and sentiment data to generate an initial question from the question database. For example, a specific question such as "What size television are you interested in?" is generated.

[1826] 6. The terminal displays questions sent from the server to the user. The user answers the displayed questions. When answering, facial expressions and voice are also captured through the terminal and simultaneously sent to the server as emotion data.

[1827] 7. The server stores the received responses in a database and uses an emotion engine to analyze the user's emotions at the time of the response. This allows the server to infer the user's specific intentions and purpose for visiting the store.

[1828] 8. Based on the analysis results, the server generates an appropriate product list from the product database. For example, if the intention "I want a 50-inch TV" is extracted, a product list specifically for 50-inch TVs will be generated.

[1829] 9. The terminal displays a list of recommended products sent from the server to the user. Sentimental data continues to be collected even after the display, making it possible to understand more precisely which products the user is interested in.

[1830] 10. The user reviews the displayed list of recommended products and selects items of interest. In some cases, they may answer further questions or proceed with the purchase.

[1831] Specific example

[1832] For example, if a user enters basic information such as "30 years old, male, interested in home appliances" and answers the question "I want a 50-inch TV," the server uses an emotion engine to analyze the user's facial expressions and tone of voice during the response. Based on this analysis, if the server determines that the user is genuinely very interested in a 50-inch TV, a 50-inch TV will be strongly reflected in the recommendation list. If the user shows little interest, a TV of a different size or with different characteristics will be recommended instead.

[1833] Example of a prompt

[1834] An example of a prompt to input into a generative AI model is: "A user showed interest in a 50-inch TV when they visited the store. Please explain the process for recommending products that match this user's interests."

[1835] By incorporating user emotional data into the analysis in this way, it becomes possible to provide a system that makes more accurate and appropriate product recommendations.

[1836] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1837] Step 1:

[1838] The server establishes connections to the product database and the question database during system initialization. Specifically, it uses a database management system (e.g., MySQL) to load the necessary data (e.g., product characteristics, question templates). It also performs the initial setup of the sentiment analysis engine (e.g., IBM Watson Emotion Recognition). The input is the database connection information and the sentiment engine configuration information, and the output is the connected database and the initialized sentiment engine.

[1839] Step 2:

[1840] The device displays an initial screen for the user to enter basic information. This screen also includes voice input (e.g., Google Speech-to-Text API) and a facial recognition camera (e.g., Azure Face API), providing an interface for acquiring emotional data from the user's facial expressions and voice. Input consists of basic information provided by the user (e.g., age, gender, categories of interest), and output consists of the user's basic information and the display of the initial screen.

[1841] Step 3:

[1842] The user enters basic information on the initial screen. This information includes age, gender, and categories of interest. The input is basic information, and the output is basic information sent to the device.

[1843] Step 4:

[1844] The terminal receives basic information entered by the user and sends it to the server. It also simultaneously collects emotional data obtained from voice input and facial recognition and sends this data to the server as well. The input consists of the user's basic information and emotional data, while the output is the data sent to the server.

[1845] Step 5:

[1846] The server analyzes the received basic information and sentiment data and generates an initial question from the question database. Specifically, it selects an appropriate question based on the received basic information and sentiment data, and generates a concrete question such as, "What size television are you interested in?" The input is basic information and sentiment data, and the output is the generated question.

[1847] Step 6:

[1848] The terminal displays the initial question sent from the server to the user. The input is the initial question from the server, and the output is the display of the question to the user.

[1849] Step 7:

[1850] The user answers the displayed questions. During the answering process, facial expressions and voice are also collected through a facial recognition camera and voice input function. Input consists of the questions, user answers, facial expressions, and voice data, while output consists of the answers and emotion data sent to the device.

[1851] Step 8:

[1852] The terminal receives user responses and sentiment data and sends them to the server. The input is the user's responses and sentiment data, and the output is the data sent to the server.

[1853] Step 9:

[1854] The server stores user responses in a database and uses an emotion engine to analyze the user's emotions at the time of their response. The input consists of the user's response and emotion data, while the output consists of the stored data and the analysis results.

[1855] Step 10:

[1856] The server analyzes the user's intent based on the response content and sentiment data using natural language processing technology (e.g., Google Natural Language API) and a sentiment engine. For example, it extracts a specific objective, such as "I want a 50-inch TV," along with sentiment data. The input is the user's response and sentiment data, and the output is the analyzed user intent.

[1857] Step 11:

[1858] The server generates a list of appropriate products from the product database based on the analysis results. The input is the analysis results, and the output is the generated product list.

[1859] Step 12:

[1860] The server sends the generated product list to the terminal. The input is the generated product list, and the output is the data sent to the terminal.

[1861] Step 13:

[1862] The terminal displays a list of recommended products to the user. For example, it might display, "Your purpose for visiting is to purchase a 50-inch television. Please see the recommended products below." The input is a list of recommended products from the server, and the output is the displayed product list.

[1863] Step 14:

[1864] Even while displaying recommended products, the device utilizes an emotion engine to continuously collect emotional data obtained from the user's facial expressions and voice, and transmits it to the server. The input is the user's facial expressions and voice data, and the output is the data transmitted to the server.

[1865] Step 15:

[1866] The server provides users with optimal additional suggestions based on collected sentiment data. The input is continuously collected sentiment data, and the output is data for additional suggestions.

[1867] Step 16:

[1868] The user can choose to view the recommended products and decide whether to purchase them, or to answer another question. The input is a list of recommended products or a new question, and the output is the user's choice or answer.

[1869] Step 17:

[1870] The terminal either generates new questions or guides the user through the purchase process based on their selection. The input is the user's selection, and the output is a new question or a purchase process screen.

[1871] (Application Example 2)

[1872] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1873] Existing consumer electronics retailers face challenges in accurately understanding what customers want to buy, hindering efficient product recommendations. Furthermore, recommending products without understanding user emotions or specific intentions can lead to decreased customer satisfaction. Moreover, with the rise of online shopping, there's a need to enhance the unique customer experience offered by physical stores. To address these challenges, a system is needed that utilizes not only basic customer information but also emotional data to accurately predict purchasing intent and provide precise product recommendations.

[1874] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes an input means for inputting the user's basic information, a question generation means for generating initial questions based on the basic information, a display means for displaying the generated questions to the user, an answer receiving means for receiving the user's answers and emotional data, an intent analysis means for analyzing the answers and emotional data to infer the purpose of the visit, a product recommendation means for recommending appropriate products based on the inferred purpose of the visit, and a display means for displaying the recommended products. This makes it possible to accurately grasp the user's emotions and intentions and make product suggestions that match the purpose of the visit, thereby improving customer satisfaction.

[1875] A "user" is a customer who visits an electronics retail store and uses the system.

[1876] "Basic information" refers to data that users enter, such as age, gender, and categories of interest.

[1877] "Input method" refers to an interface that allows users to input basic information, and includes devices such as smartphones and smart glasses.

[1878] "Question generation method" refers to the process of creating appropriate questions based on the basic information entered by the user.

[1879] "Display means" refers to displays or screens used to present generated questions and recommended products to the user.

[1880] A "response receiving method" refers to an interface that receives responses and sentiment data entered by users in response to questions.

[1881] "Emotional data" refers to emotional information obtained from the user's facial expressions, voice, and other sources.

[1882] "Intention analysis means" refers to a function that analyzes user responses and emotional data to infer the user's purpose for visiting the store.

[1883] "Natural language processing technology" refers to artificial intelligence technology used to understand and interpret user text data.

[1884] An "emotion engine" refers to an algorithm and software that analyzes a user's emotions from their facial expressions and voice.

[1885] "Product recommendation methods" refer to the process of selecting and recommending appropriate products based on the user's purpose for visiting the store.

[1886] A "server" is a central device that manages the entire system and performs tasks such as data analysis, storage, and query generation.

[1887] To implement the present invention, terminals such as smartphones and smart glasses, and servers connected to them, are used in consumer electronics stores. The main components of the system include input means for inputting basic user information, question generation means for generating questions, display means for displaying the generated questions, answer receiving means for receiving user answers and sentiment data, intent analysis means for analyzing user answers and sentiment data, product recommendation means for recommending appropriate products, and display means for displaying recommended products.

[1888] The server connects to the product database and question database during system initialization and performs the initial setup of the emotion engine. It uses "EmotionEngine" as the emotion engine and "NlpProcessor" as the natural language processing technology. The server generates questions based on the user's basic information and past answer history and sends them to the terminal.

[1889] The device takes the form of a smartphone or smart glasses and displays an initial screen for the user to input basic information. The device also includes features such as voice input and a facial recognition camera, which are used to collect emotional data. The basic information entered by the user is sent to a server, which then generates appropriate initial questions based on this information.

[1890] Users respond to questions displayed on their device using voice or touch input. The device sends these responses to a server, which analyzes the emotional data acquired along with the responses. By using an emotion engine to read the user's emotions from their facial expressions and voice, and by analyzing the content of their responses using NLP technology, the server infers the user's purpose for visiting the store.

[1891] The server selects appropriate products from its product database based on the inferred purpose of the visit, generates a list of recommended products, and sends it to the terminal. The terminal displays this list to the user, who then selects products that interest them. Throughout this process, the server continuously collects sentiment data and provides recommendations tailored to the user's preferences.

[1892] For example, when a user responds that they are "looking for a large refrigerator," the system analyzes the user's facial expressions and tone of voice to estimate whether their interest is genuine. Based on these results, "refrigerators of 500 liters or more" are strongly reflected in the recommendation list.

[1893] Examples of prompt statements used as input to a generative AI model are as follows:

[1894] Please generate questions to infer the user's purpose of purchase. The basic user information is as follows:

[1895] Age: 35

[1896] Gender: Male

[1897] Categories of interest: Home appliances

[1898] Please generate questions based on the following data:

[1899] Emotional data: Joy (high)

[1900] Example questions to output:

[1901] "What size refrigerator in liters are you interested in?"

[1902] This makes it possible to accurately understand the user's emotions and intentions, and by suggesting products that match their purpose for visiting the store, customer satisfaction can be improved.

[1903] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1904] Step 1:

[1905] During system initialization, the server establishes connections to the product database and question database and performs initial setup of the sentiment engine. At this time, the server reads product information and question templates and sets initial parameters for sentiment analysis.

[1906] Input: Product database, question database, emotion engine configuration information

[1907] Output: System state after initial setup is complete

[1908] Specific actions: Connecting to the database, loading configuration files.

[1909] Step 2:

[1910] The device displays an initial screen where the user can enter basic information. The device is also equipped with voice input and a facial recognition camera, providing an interface to acquire user emotion data.

[1911] Input: System initial settings

[1912] Output: Basic Information Input Screen

[1913] Specific actions: rendering the initial screen, voice input, and camera activation.

[1914] Step 3:

[1915] The user enters basic information (age, gender, and categories of interest) using the initial screen displayed. The device also acquires emotional data through voice input and facial recognition.

[1916] Input: User's basic information, voice and facial recognition data

[1917] Output: User basic information, sentiment data

[1918] Specific operations: Collection of user input data, real-time acquisition of sentiment data.

[1919] Step 4:

[1920] The terminal analyzes the collected basic information and emotional data and sends it to the server. The server receives this data and prepares it for analysis.

[1921] Input: User's basic information, sentiment data

[1922] Output: Data ready for analysis

[1923] Specific actions: Packing input data and sending it to the server.

[1924] Step 5:

[1925] The server generates an initial question from the question database based on the user's basic information and sentiment data.

[1926] Input: User's basic information, sentiment data

[1927] Output: Initial Question

[1928] Specific actions: Execute database queries, apply query templates.

[1929] Step 6:

[1930] The device displays the generated questions to the user. The user enters answers to the questions and expresses emotions through voice and facial expressions.

[1931] Input: Initial Question

[1932] Output: User responses, sentiment data

[1933] Specific actions: Displaying questions, providing an interface for entering answers.

[1934] Step 7:

[1935] The device sends the user's responses and sentiment data to the server, which receives, stores, and then analyzes this data.

[1936] Input: User responses, sentiment data

[1937] Output: Analysis results

[1938] Specific actions: Packing data, sending it to the server, and saving it to the database.

[1939] Step 8:

[1940] The server uses natural language processing technology and an emotion engine to analyze the user's responses and emotion data, and to infer the user's purpose for visiting the store.

[1941] Input: User responses, sentiment data

[1942] Output: Estimated results of the purpose of visit

[1943] Specific actions: Application of NLP algorithms, performance of sentiment analysis.

[1944] Step 9:

[1945] Based on the inferred purpose of the customer's visit, the server selects appropriate products from the product database and generates a list of recommended products.

[1946] Input: Estimated result of the purpose of visit

[1947] Output: Recommended product list

[1948] Specific actions: Execute database queries, select recommended products.

[1949] Step 10:

[1950] The terminal displays a list of recommended products to the user, who then selects products of interest. The server continuously collects sentiment data and provides optimal additional suggestions.

[1951] Input: Recommended product list, user selection

[1952] Output: Optimal product proposal

[1953] Specific actions: Rendering of recommended products, real-time analysis of sentiment data.

[1954] Step 11:

[1955] When a user selects a product or answers additional questions, the device sends data back to the server, which then asks the next most appropriate question or suggests a product as needed.

[1956] Input: User selection, answers to additional questions

[1957] Output: Next question or product suggestion

[1958] Specific actions: data retransmission, reapplication of analysis algorithms.

[1959] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1960] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1961] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1962] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1963] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1964] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1965] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1966] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1967] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1968] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1969] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1970] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1971] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1972] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1973] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1974] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1975] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1976] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1977] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1978] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1979] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical sta...

Claims

1. An input method for entering the user's basic information, A question generation means that generates an initial question based on the basic information received from the input means, A display means for displaying the generated question to the user, A means of receiving user responses, An intent analysis means analyzes the response received from the aforementioned response receiving means and infers the user's purpose for visiting the store, A product recommendation means that recommends appropriate products based on the purpose of visit inferred by the aforementioned intent analysis means, A display method for showing recommended products to the user, A system that includes this.

2. The system according to claim 1, characterized in that the question generation means generates the next question using the user's basic information and past answer history.

3. The system according to claim 1, characterized in that the intent analysis means analyzes the user's response using natural language processing technology.

Images