System

Abstract

A system is provided.SOLUTION: A system including means for receiving a question of a user input by an information terminal, means for analyzing the received question by natural language processing and grasping an intention of the user, means for integrating past data of the user and information acquired from an external information service and generating optimal information, means for displaying the generated information on the information terminal, and means for collecting feedback of the user input by the information terminal and causing a generation model to learn.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, 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] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] When users search for information on the Web, they often find it difficult to efficiently obtain information because they do not know what specific keywords to use to obtain the desired results. Many users also find it difficult to organize and obtain information that requires specialized knowledge, and information overload and the difficulty of obtaining specialized information contribute to a decline in quality of life (QOL). Furthermore, conventional search systems are unable to provide personalized, optimized advice based on the user's past hobbies and interests, limiting their ability to improve the user experience. [Means for solving the problem]

[0005] The present invention provides a means for receiving a user's question entered via an information terminal, analyzing the received question using natural language processing, and understanding the user's intent. Furthermore, the present invention provides a system including a means for integrating the user's past data and information obtained from external information services to generate optimal information and display it on the information terminal. Additionally, by providing a means for collecting user feedback and continuously training a generative model, it becomes possible to provide individually optimized information. This system stores user registration information in a database, enabling the provision of information tailored to each user. It also includes a function for evaluating the suitability of the generated information and selecting the most appropriate information from information obtained from multiple external information services. This allows users to efficiently obtain the information they need, eliminating the difficulty of obtaining information and expected to improve their quality of life.

[0006] An "information terminal" is a device that allows a user to input information, search, or view information, and includes, for example, a smartphone, tablet, or PC.

[0007] "Natural language processing" is a technology that allows computers to understand and analyze human language, and is used to analyze input sentences and grasp their meaning.

[0008] "User intent" refers to inferring and understanding what a user wants to know or what they are looking for from the questions or search terms they enter.

[0009] "Past data" refers to all records related to a user, such as the user's past search history, operation history, and registration information.

[0010] An "external information service" is an information source provided by an external database or API used by the system, such as medical information, traffic information, or cooking recipes.

[0011] "Optimal information" refers to information that is judged to be most useful to the user based on the user's questions, past data, and information obtained from external information services.

[0012] "Feedback" refers to the evaluations, comments, and operation results that users make on the information provided, and is used to improve the system.

[0013] A "generative model" refers to an algorithm or AI model that generates optimal information based on natural language processing and data analysis.

[0014] A "database" refers to an information management system for storing and managing user registration information, search history, etc.

[0015] "Relevance" refers to the criteria for evaluating whether the generated information matches the user's intentions and requirements. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0017] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0018] First, the terms used in the following description will be explained.

[0019] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0020] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0021] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0022] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0024] [First embodiment]

[0025] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0026] 1, a 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.

[0027] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0029] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the 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.

[0030] 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 of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0032] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 2, in the data processing device 12, a specific process 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" according to the technology of the present 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 process 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.

[0034] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the 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 process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0036] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0037] The present invention is a system that uses an information terminal, a server, and a database, and provides optimal answers to questions asked by users in information searches. A specific embodiment of this system will be described below.

[0038] User registration and information collection

[0039] First, a user accesses the service using an information terminal and enters basic information such as name, age, occupation, hobbies and interests into the registration form. The terminal sends this information to the server, which then stores it in a database.

[0040] Entering and parsing questions

[0041] The user inputs a specific question, such as "Tell me an easy pasta recipe," into the device and sends it. This input data is sent to the server. The server analyzes the received question using natural language processing (NLP) technology, extracts keywords such as "easy," "pasta," and "recipe," and understands the user's intent.

[0042] Integrating historical data and external information

[0043] Based on the analysis results, the server retrieves the user's past data from the database. For example, if the user has previously searched for Italian food, that information will be used as a reference. The server also connects with external information services (such as cooking recipe providers) via API to retrieve the latest recipe information.

[0044] Generating and providing optimal information

[0045] The server combines the user's hobby and interest data with external information to generate the most appropriate answer. In this case, it identifies "an easy recipe for spaghetti with tomatoes and basil" and generates an answer in natural language format. The generated answer is sent to the device and displayed to the user.

[0046] Gathering feedback and learning

[0047] The user reviews the provided information and sends feedback such as "This recipe was very helpful" via their device. This feedback is received by the server and stored in a database. The server uses this feedback to continuously train the generative model, improving the accuracy of answers from the next time onwards.

[0048] Specific examples

[0049] For example, suppose a user inputs a question about "easy pasta recipes." The server analyzes this question and references the user's past search history and hobbies and preferences from a database. If the user has previously liked "Italian food," the server connects with an external information service to retrieve the latest recipe for "spaghetti with tomato and basil." The server formats this information in natural language and sends it to the device. The device displays it to the user as "an easy recipe for spaghetti with tomato and basil." If the user then rates this recipe as useful, that feedback is sent back to the server and used as training data for the model.

[0050] In this way, the present invention is a system that continues to provide optimal information in response to user questions, thereby improving the efficiency of information acquisition and user satisfaction.

[0051] The processing flow will be explained below.

[0052] Step 1:

[0053] The user accesses the service using an information terminal and enters basic information such as name, age, occupation, hobbies and interests into the registration form.

[0054] The terminal transmits this information to the server.

[0055] Step 2:

[0056] The server receives the basic information sent by the user and stores it in a database.

[0057] The server also generates a unique user ID for each user and creates a corresponding record.

[0058] Step 3:

[0059] The user enters a specific question, such as "Tell me an easy pasta recipe," into the terminal and sends it.

[0060] The terminal transmits the entered question to the server.

[0061] Step 4:

[0062] The server analyzes the received question using natural language processing (NLP) techniques.

[0063] The server extracts key keywords from the question ("easy," "pasta," "recipe") and understands the user's intent.

[0064] Step 5:

[0065] The server retrieves the user's past data from the database.

[0066] For example, if the user has previously searched for "Italian cuisine," that information can be used as a reference.

[0067] Step 6:

[0068] The server initiates API integration with an external information service to obtain the latest recipe information.

[0069] For example, recipe data for "spaghetti with tomato and basil" is received from an external recipe service.

[0070] Step 7:

[0071] The server integrates the acquired external information with the user's past data to generate the optimal answer.

[0072] For example, create an "easy tomato and basil spaghetti recipe" in natural language format.

[0073] Step 8:

[0074] The server sends the generated response to the terminal.

[0075] The device receives the response and displays it to the user as an "easy tomato and basil spaghetti recipe."

[0076] Step 9:

[0077] The user reviews the information provided and sends feedback via the device, such as "This recipe was helpful."

[0078] The terminal transmits the feedback information to the server.

[0079] Step 10:

[0080] The server stores the user feedback in a database.

[0081] The server uses this feedback to train a generative model and uses it to improve the accuracy of answers from next time onwards.

[0082] Example 1

[0083] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0084] Conventional information provision systems have difficulty providing accurate answers to user questions, and in order to improve user satisfaction, it is necessary to effectively integrate the user's past behavioral history and external information. Furthermore, due to a lack of means to collect user feedback and improve the accuracy of the system, continuous learning is not possible, raising concerns that the quality of the information provided may decline.

[0085] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0086] In this invention, the server includes means for receiving a user's question input via an information terminal, means for analyzing the received question using natural language processing to understand the user's intention, means for integrating the user's past data and information acquired from an external database to generate optimal information, means for transmitting the generated information to the information terminal and displaying it, and means for collecting user feedback input via the information terminal and training the generation AI model. This makes it possible to continue providing optimal information in response to a user's question, thereby improving the efficiency of information acquisition and user satisfaction.

[0087] An "information terminal" is a device that allows a user to access the system via the Internet and input and display data, and includes smartphones, personal computers, etc.

[0088] A "user question" is a textual inquiry that a user inputs into the system using an information terminal, requesting specific information.

[0089] "Natural language processing" is a technology that analyzes input text data in a way that is easy for a computer to understand, and grasps its meaning and intent.

[0090] "User intent" refers to the specific information or answer that the user wants to obtain through the question.

[0091] "Past data" refers to information such as a user's past behavioral history, search history, and interests that is stored in a database.

[0092] An "external database" is a data source that exists outside the system, and includes various information providing services that can be accessed via API.

[0093] "Optimal information" refers to the most appropriate and useful answer to a user's question, and is generated by integrating information from past data and external databases.

[0094] A "generative AI model" is an algorithm that continuously learns from data and feedback provided by users, improving the accuracy of its answers.

[0095] "Feedback" refers to the evaluations and opinions that users give in response to the information provided, and is used to help the system learn.

[0096] "Display means" refers to a function for visually presenting optimal information on an information terminal, and includes a screen or display.

[0097] "Integration means" is a process that combines the user's past data and information obtained from external databases to generate information that is optimal for the user.

[0098] The present invention is a system that uses an information terminal, a server, and a database, and provides optimal answers to questions asked by users in information searches. An embodiment of this system will be described in detail below.

[0099] First, the user accesses the system's website using an information terminal (smartphone or PC). The user enters basic information such as name, age, occupation, hobbies and interests into the registration form on the website. When the user presses the "Submit" button, the information terminal sends the entered data to the server using the HTTPS protocol. The server stores the received data in a database server (e.g., MySQL, PostgreSQL).

[0100] Next, the user enters a specific question. For example, "Tell me an easy pasta recipe" and clicks the submit button. The input data is again sent to the server via HTTPS protocol. The server analyzes the received question using natural language processing (NLP) technology. At this time, NLP libraries such as spaCy and NLTK are used to extract keywords such as "easy," "pasta," and "recipe." This allows the server to understand the user's intent.

[0101] The server then retrieves the user's past behavioral history and preference information from the database. This information is obtained using an SQL query. The server also connects to external databases (e.g., recipe service) via API to retrieve the latest recipe information. The acquired internal data is then integrated with external data to generate the optimal recipe information for the user. A generative AI model is used to generate a natural language answer (e.g., "Recipe for spaghetti with tomato and basil").

[0102] The generated answer is sent from the server to the device, which displays it to the user. The user checks the displayed information, enters feedback (e.g., "This recipe was very helpful"), and sends it to the server via the device. The server stores the feedback in a database and uses this information to continuously train the generative AI model. This improves the accuracy of answers in future queries.

[0103] Specific examples

[0104] For example, if a user inputs a question such as "Tell me an easy pasta recipe," the system operates as follows: The server first analyzes the question using NLP technology to extract keywords. It then combines the user's past data with the latest recipe information obtained from an external API to generate an "easy tomato and basil spaghetti recipe." This generated information is sent to the device and displayed to the user. Finally, the user sends feedback to the server on whether the provided recipe was helpful, and this feedback is used to train the generative AI model.

[0105] In this way, the present invention is a system that continues to provide optimal information in response to user questions, thereby improving the efficiency of information acquisition and user satisfaction.

[0106] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0107] Step 1:

[0108] A user accesses the service and enters registration information.

[0109] Specific operation: The user uses an information terminal (smartphone or PC) to access the system's website.

[0110] Input: A user fills in a registration form on a website with information such as name, age, occupation, hobbies and interests.

[0111] Data processing: The device temporarily stores the entered information.

[0112] Output: The persisted data is ready to be sent to the server in the next step.

[0113] Step 2:

[0114] The terminal transmits the input user information to the server.

[0115] What happens: The user clicks a button to submit the registration form.

[0116] Input: Stored user information (name, age, occupation, hobbies and interests).

[0117] Data calculation: The information terminal uses the HTTPS protocol to send the input information to the server.

[0118] Output: The server receives the user information.

[0119] Step 3:

[0120] The server stores the received user information in a database.

[0121] Specific operation: The server analyzes the received registration data and stores it in a database.

[0122] Input: User registration information.

[0123] Data manipulation: Creating new records in the database using SQL queries.

[0124] Output: The saved user information.

[0125] Step 4:

[0126] The user types a question and submits it.

[0127] Specific operation: The user enters a question on the information terminal and clicks the send button.

[0128] Input: Question text (e.g. "What's an easy pasta recipe?").

[0129] Data calculation: The information terminal sends the entered query data to the server.

[0130] Output: The server receives the query data.

[0131] Step 5:

[0132] The server analyzes the question using natural language processing.

[0133] What happens: The server parses the question using an NLP library (e.g., spaCy, NLTK).

[0134] Input: The question text submitted by the user.

[0135] Data processing: Extract keywords from the question text to understand the user's intent.

[0136] Output: Extracted keywords and user intent.

[0137] Step 6:

[0138] The server retrieves the user's past data from the database.

[0139] Specific operation: The server retrieves past behavioral history and interest data from the database based on the user's ID.

[0140] Input: User ID and parsed question information.

[0141] Data operations: Retrieving relevant data from a database using SQL queries.

[0142] Output: User's historical data.

[0143] Step 7:

[0144] The server retrieves the latest information from an external database.

[0145] Specific operation: The server calls an external API to obtain the latest recipe information.

[0146] Input: Extracted keywords (e.g., pasta, recipe).

[0147] Data Calculation: Sending requests to external APIs and parsing the responses.

[0148] Output: Latest recipe information.

[0149] Step 8:

[0150] The server generates the most appropriate information and provides it to the user.

[0151] Specific operation: The server integrates the user's past data with external data and generates answers in natural language format using a generative AI model.

[0152] Input: User's past data, latest recipe information, parsed question information.

[0153] Data processing: Generate optimal answers from integrated data.

[0154] Output: A natural language answer (e.g., "Easy tomato and basil spaghetti recipe").

[0155] Step 9:

[0156] The terminal displays the generated answer to the user.

[0157] Specific operation: The device displays the answer received from the server to the user.

[0158] Input: The answer sent by the server in natural language format.

[0159] Data processing: Formatting text for visual display on the device.

[0160] Output: The answer displayed to the user.

[0161] Step 10:

[0162] The user enters and submits feedback.

[0163] Specific actions: The user enters feedback on the device and clicks the submit button.

[0164] Input: Feedback text (e.g., "This recipe was very helpful").

[0165] Data calculation: The information terminal sends the input feedback to the server.

[0166] Output: The server receives the feedback data.

[0167] Step 11:

[0168] The server stores the feedback in a database and trains a generative AI model.

[0169] Specific operation: The server stores the feedback data in a database and updates the generative AI model.

[0170] Input: Received feedback data.

[0171] Data processing: Store the feedback in a database and use it as new data for training the AI ​​model.

[0172] Output: An updated generative AI model.

[0173] This allows the system to continue providing optimal information in response to user questions, improving the efficiency of information acquisition and user satisfaction.

[0174] (Application example 1)

[0175] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0176] Conventional product recommendation systems in virtual stores were unable to fully reflect the individual preferences and past purchase history of users, making it difficult to provide optimal products. Furthermore, there was no mechanism for effectively collecting and utilizing user feedback, making it difficult to improve the quality of service.

[0177] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0178] In this invention, the server includes means for receiving a user question input via an information terminal, means for analyzing the received question using natural language processing to understand the user's intention, means for integrating the user's past data and information acquired from external information services to recommend optimal products based on the user's preferences, means for displaying the generated product information on the information terminal, and means for collecting user feedback input via the information terminal and training a generative model. This makes it possible to propose optimal products based on the individual preferences of users and improve the quality of services by utilizing feedback.

[0179] An "information terminal" refers to an electronic device, such as a smartphone or tablet, that allows a user to input questions and receive answers from a server.

[0180] "Natural language processing" refers to the technology that allows computers to understand and analyze human language, allowing them to understand the intent of users' questions.

[0181] "User's past data" refers to information such as the user's past search history, purchase history, and feedback.

[0182] "External Information Services" refers to information or databases provided by third parties, such as product catalogs or review sites.

[0183] "Means for recommending optimal products" refers to algorithms or programs that integrate the user's past data with information obtained from external information services and select the most suitable product based on the user's preferences.

[0184] The "generated product information" refers to detailed product information selected by the optimal product recommendation means, including, for example, the product name, price, and features.

[0185] "Feedback" refers to a user's evaluation or opinion of the information provided or the recommended products.

[0186] A "generative model" refers to a machine learning model that generates optimal information based on user questions and feedback.

[0187] The present invention provides a personalized shopping assistant system for a virtual store that recommends optimal products based on a user's preferences and past purchase history. Specific embodiments of this system are described below.

[0188] Hardware and software used

[0189] This system consists of information terminals (e.g., smartphones, tablets, smart glasses) accessed by users and a server that processes data. The software includes the following elements:

[0190] 1. Natural Language Processing (NLP) engine: used to analyze the user's question and extract keywords, for example using Python and the transformers library.

[0191] 2. Database: Stores user registration information, past data, and product information obtained from external information services. For example, a database system such as MySQL is used.

[0192] 3. API integration module: Connects to external information services (e.g., product catalogs, review sites, etc.) via API to obtain the latest product information.

[0193] 4. Generative AI model: A model for recommending optimal products based on user preference data and external information. As an example, we will use a machine learning model.

[0194] System Operation

[0195] The server operates as follows:

[0196] 1. Accepting a user question: The user's question (e.g., "Please recommend a new smartphone") is received via the information terminal.

[0197] 2. Natural language processing of questions: The content of the question is analyzed using natural language processing technology to extract important keywords.

[0198] 3. Information integration: Integrates the user's past data (e.g., past purchase history) and the latest product information obtained from external information services to select the optimal product based on the user's preferences.

[0199] 4. Product information generation and display: The selected product information is formatted in natural language and displayed on the information terminal.

[0200] 5. Feedback collection: Collect user feedback entered via the information terminal and continuously train the generative AI model.

[0201] Specific examples

[0202] For example, a user might use an information terminal to ask, "Please recommend a new smartphone." This input is sent to the server and analyzed by a natural language processing engine. As a result of the analysis, important keywords such as "smartphone" and "recommendation" are extracted. The server then retrieves the user's past purchase history and preference data from a database, and obtains the latest smartphone information through the API of an external information service.

[0203] The generative AI model integrates this information and recommends the most suitable smartphone for the user (e.g., "latest model smartphone"). This information is formatted in natural language and displayed on the information terminal. When the user reviews the recommendation and sends feedback saying "this suggestion was helpful," that feedback is stored on the server and used as training data for the generative AI model.

[0204] Examples of prompt statements

[0205] User: Recommend me a new smartphone.

[0206] System: Based on your past search and purchase history, we estimate that a new smartphone model would be suitable. We recommend a 'new smartphone model' featuring a high-performance camera and a long-lasting battery. The price is 50,000 yen. How about this product?

[0207] As described above, the present invention makes it possible to provide appropriate and personalized product suggestions based on the individual preferences and past data of the user.

[0208] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0209] Step 1:

[0210] The user enters a question.

[0211] The user uses an information terminal (e.g., a smartphone) to input a question. Specifically, the user inputs text such as "Please recommend a new smartphone" and presses the send button. The input question (text data) is sent to the server.

[0212] Step 2:

[0213] The server accepts the query.

[0214] The server receives questions sent by users, stores them as text data, and then passes the text data to a natural language processing engine.

[0215] Step 3:

[0216] The server performs natural language processing.

[0217] The server uses a natural language processing engine (e.g., the transformers library) to analyze the received text data, split the question, and extract important keywords (e.g., "smartphone," "recommended"). The results with the extracted keywords are then returned.

[0218] Step 4:

[0219] The server obtains the user's past data and external information.

[0220] The server retrieves the user's past data (past purchase history, search history, etc.) from the database based on the keywords. At the same time, it calls an external information service API to retrieve the latest product information related to the keywords. The retrieved data is then integrated.

[0221] Step 5:

[0222] The server recommends the best products.

[0223] The server uses the generative AI model to analyze and integrate the user's past data and product data obtained from external information, thereby selecting the product (e.g., the latest model smartphone) that best suits the user's preferences.

[0224] Step 6:

[0225] The server generates and displays product information.

[0226] The server formats the selected product details (product name, price, features, etc.) into a natural language format. This generated product information is sent to the information terminal, which then displays the information to the user.

[0227] Step 7:

[0228] The user enters feedback.

[0229] The user checks the presented product information and inputs feedback such as "This suggestion was helpful." The input feedback is then sent back to the server.

[0230] Step 8:

[0231] The server collects feedback and learns.

[0232] The server receives feedback sent by the user, stores it in a database, and uses it as training data for the generative AI model, thereby improving the accuracy of recommendations in the future.

[0233] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0234] The present invention is a system that uses an information terminal, a server, a database, and an emotion engine to provide optimal answers to questions asked by users in information searches, and further adjusts the content of the answers taking into account the user's emotions. A specific embodiment of this system is described below.

[0235] User registration and information collection

[0236] First, a user accesses the service using an information terminal and enters basic information such as name, age, occupation, hobbies and interests into the registration form. The terminal sends this information to the server, which stores it in a database. The emotion engine also performs initial settings to understand the user's basic emotional patterns.

[0237] Entering and parsing questions

[0238] The user inputs a specific question, such as "Tell me an easy pasta recipe," into the device and sends it. This input data is sent to the server. The server analyzes the received question using natural language processing (NLP) technology, extracts keywords such as "easy," "pasta," and "recipe," and understands the user's intent.

[0239] Acquiring user emotion data

[0240] When a question is entered, the emotion engine collects the user's emotion data and uses, for example, text analysis and facial expression recognition technology to understand the user's current emotional state (such as joy, sadness, or anger).

[0241] Integrating historical data and external information

[0242] Based on the analysis results, the server retrieves the user's past data from the database. For example, if the user has previously searched for Italian food, that information will be used as a reference. The server also connects with external information services (such as cooking recipe providers) via API to retrieve the latest recipe information.

[0243] Optimal information generation and display

[0244] The server combines the acquired external information, the user's past data, and emotional data to generate the optimal answer. For example, it identifies an "easy tomato and basil spaghetti recipe" and generates an answer tailored to the user's emotional state, such as "You seem to be in a low mood today. Try this easy pasta recipe to cheer you up." The generated answer is sent to the device and displayed to the user.

[0245] Gathering feedback and learning

[0246] The user reviews the provided information and sends feedback via their device, such as "This recipe was very helpful." This feedback is received by the server and stored in a database. The server uses this feedback to continuously train the generative model and emotion engine, and uses it to improve the accuracy of answers in future searches.

[0247] Specific examples

[0248] For example, if a user enters "easy pasta recipes," the server analyzes the question and references the user's past search history and hobbies and preferences from a database. If the emotion engine recognizes the emotion "a little tired" from the user's input, the server connects with an external information service to retrieve a recipe for "spaghetti with tomato and basil." Based on this, the server makes adjustments such as "You seem a little tired today. Try refreshing yourself with this easy pasta recipe," and sends and displays it on the device. If the user rates the recipe as "helpful," that feedback is sent back to the server and used as learning data for the model. The emotion engine also improves its accuracy based on feedback.

[0249] In this way, the present invention is a system that provides optimal information in response to a user's question while enabling flexible responses according to the user's emotional state, thereby improving the efficiency of information acquisition and user satisfaction.

[0250] The processing flow will be explained below.

[0251] Step 1:

[0252] The user accesses the service using an information terminal and enters basic information such as name, age, occupation, hobbies and interests into the registration form.

[0253] The terminal transmits this information to the server.

[0254] Step 2:

[0255] The server receives the basic information sent by the user and stores it in a database.

[0256] The server also generates a unique user ID for each user and creates a corresponding record.

[0257] Step 3:

[0258] The user enters a specific question, such as "Tell me an easy pasta recipe," into the terminal and sends it.

[0259] The terminal transmits the entered question to the server.

[0260] Step 4:

[0261] The emotion engine analyzes the user's input text and recognizes emotional states (e.g., joy, sadness, fatigue, etc.).

[0262] The server acquires the emotion data and processes it together with the question data.

[0263] Step 5:

[0264] The server analyzes the received question using natural language processing (NLP) techniques.

[0265] The server extracts key keywords from the question ("easy," "pasta," "recipe") and understands the user's intent.

[0266] Step 6:

[0267] The server retrieves the user's past data from the database.

[0268] For example, if the user has previously searched for "Italian cuisine," that information can be used as a reference.

[0269] Step 7:

[0270] The server initiates API integration with an external information service to obtain the latest recipe information.

[0271] For example, recipe data for "spaghetti with tomato and basil" is received from an external recipe service.

[0272] Step 8:

[0273] The server integrates the acquired external information, past data, and emotional data to generate the optimal answer.

[0274] For example, if the user's emotional state is recognized as "tired," the system adjusts the response by saying, "You seem a little tired today. Try refreshing yourself with this simple pasta recipe."

[0275] Step 9:

[0276] The server sends the generated response to the terminal.

[0277] The device receives the response and displays it to the user as an "easy tomato and basil spaghetti recipe."

[0278] Step 10:

[0279] The user reviews the information provided and sends feedback via the device, such as "This recipe was very helpful."

[0280] The terminal transmits the feedback information to the server.

[0281] Step 11:

[0282] The server stores the user feedback in a database.

[0283] The server uses this feedback to continuously train the generative model and emotion engine, and uses it to improve the accuracy of answers from next time onwards.

[0284] Example 2

[0285] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0286] Conventional information retrieval systems provide answers only to questions entered by the user and do not take the user's emotional state into account, which means that the answers may not necessarily meet the user's needs. Furthermore, the means for collecting feedback are limited, which leaves the user experience unsatisfactory. Therefore, the present invention aims to provide optimal information that takes into account the user's emotional state and continuously improve the system's accuracy by utilizing feedback.

[0287] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0288] In this invention, the server includes means for receiving a user question input via an information terminal, means for analyzing the received question using natural language processing to understand the user's intention, means for analyzing the user's current emotional state, means for integrating the user's past data and information obtained from external information services to generate optimal information, means for adjusting the generated information according to the user's emotional state and displaying it on the information terminal, and means for collecting user feedback input via the information terminal and training the generative model and emotion engine. This makes it possible to provide optimal information in response to the user's question while also responding flexibly to the user's emotional state.

[0289] An "information terminal" is a device used by a user to input a question, and includes a computer, a smartphone, a tablet, and the like.

[0290] A "server" is a central system that receives data from information terminals via a network and analyzes and processes the data.

[0291] "Natural Language Processing (NLP)" is a technology that allows computers to understand and analyze human language and grasp the user's intent.

[0292] An "emotion engine" is an algorithm and software that analyzes and identifies a user's emotional state from their text and facial expressions.

[0293] A "database" is a system for managing and storing user registration information and past data.

[0294] "External information services" are third-party services that provide information via the Internet.

[0295] A "generative model" is an algorithm and software that generates optimal information in response to a user's question.

[0296] "Feedback" refers to the evaluation or opinion that a user enters regarding the information provided.

[0297] "Synthesis" is the process of combining data obtained from multiple different sources into a single, consistent piece of information.

[0298] User registration and information collection

[0299] First, the user accesses the service using an information terminal (smartphone, computer, tablet, etc.) and enters basic information such as name, age, occupation, hobbies and interests into the registration form. This information is then sent from the terminal to the server. The server stores the received information in a database and performs initial settings to understand the user's basic emotional patterns using an emotion engine. At this point, the server stores the user's individual information in a database, making it possible to provide individualized information.

[0300] Entering and parsing questions

[0301] The user inputs a specific question into an information terminal. For example, they can type in the text "Tell me an easy pasta recipe" and send it. This input data is sent from the terminal to the server. The server analyzes the received question using natural language processing (NLP) technology, extracts keywords such as "easy," "pasta," and "recipe," and understands the user's intent. This analysis can be performed using open-source natural language processing libraries or cloud-based NLP services.

[0302] Acquiring user emotion data

[0303] When a user enters a question, the emotion engine collects the user's emotion data. It can use a camera-equipped information device to understand the user's current emotional state (e.g., joy, sadness, anger) using text analysis and facial expression recognition technology. For emotion recognition, it can use, for example, an open-source emotion analysis library or a cloud-based emotion recognition API.

[0304] Integrating historical data and external information

[0305] Based on the analysis results, the server retrieves the user's past data (e.g., past search history, hobbies, and preferences) from the database. It also connects with external information services (e.g., cooking recipe services) via API to retrieve the latest recipe information. During this process, the server communicates with external services using an enterprise DBMS, cloud database service, or REST API.

[0306] Optimal information generation and display

[0307] The server combines the acquired external information, the user's past data, and emotional data to generate the optimal answer. For example, it identifies an "easy tomato and basil spaghetti recipe" and generates an answer tailored to the user's emotional state, such as "You seem to be in a low mood today. Try this easy pasta recipe to cheer you up." The generated answer is sent to the device and displayed to the user.

[0308] Gathering feedback and learning

[0309] The user reviews the provided information and sends feedback via their device, such as "This recipe was very helpful." This feedback is received by the server and stored in a database. The server uses this feedback to continuously train the generative model and emotion engine, and uses it to improve the accuracy of answers in the future.

[0310] Specific examples

[0311] For example, if a user enters "easy pasta recipes," the server analyzes the question and references the user's past search history and hobbies and preferences from a database. If the emotion engine recognizes the emotion "a little tired" from the user's input, the server connects with an external information service to retrieve a recipe for "spaghetti with tomato and basil." Based on this, the server makes adjustments such as "You seem a little tired today. Try refreshing yourself with this easy pasta recipe," and sends and displays it on the device. If the user rates the recipe as "helpful," that feedback is sent back to the server and used as learning data for the model. The emotion engine also improves its accuracy based on feedback.

[0312] Example prompts to input to the generative AI model

[0313] "When a user is looking for an easy pasta recipe, give them the perfect recipe based on their emotional state and why. For example, 'You seem a little tired today. Try refreshing yourself with this easy pasta recipe.'"

[0314] In this way, the present invention is a system that provides optimal information in response to a user's question while enabling flexible responses according to the user's emotional state, thereby improving the efficiency of information acquisition and user satisfaction.

[0315] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0316] Step 1:

[0317] User registration and information collection

[0318] Input: The user uses an information terminal to input basic information such as name, age, occupation, hobbies and interests.

[0319] Processing: The device sends the input information to the server. The server stores the received information in a database and sets the user's initial emotional pattern using the emotion engine.

[0320] Output: User information is saved in the database and the emotion engine completes its initialization.

[0321] How it works: The user accesses the registration page in a web browser or mobile app, enters the required information in the form, and then presses the "Register" button. The device sends this data to the server as an HTTP request. The server saves the information in a database and calls the emotion recognition API to set the initial emotion pattern.

[0322] Step 2:

[0323] Entering and parsing questions

[0324] Input: The user types a specific question into the device and sends it. For example, the text "Tell me a simple pasta recipe."

[0325] Processing: The device sends the input data to the server, which uses natural language processing (NLP) techniques to analyze the question and extract keywords.

[0326] Output: Keyword data is generated to understand the intent of the question.

[0327] Specific operation: The user enters a question into the text input form and presses the "Submit" button. The device sends the text data as an HTTP POST request to the server. The server uses the NLP module to extract keywords such as "easy," "pasta," and "recipe."

[0328] Step 3:

[0329] Acquiring user emotion data

[0330] Input: User text and facial expression data when entering a question.

[0331] Processing: The server uses the emotion engine to analyze the user's current emotional state (happiness, sadness, anger, etc.).

[0332] Output: Data of the current user's emotional state.

[0333] Specific operation: The server inputs text data into the emotion engine and calls the emotion recognition API. The emotion engine analyzes emotions such as "joy," "sadness," and "anger." The analysis results are recorded for use in the next step.

[0334] Step 4:

[0335] Integrating historical data and external information

[0336] Input: Question analysis results, user history, data from external information services.

[0337] Processing: The server retrieves the user's past data from the database, connects to external information services via API, and retrieves the latest information. This data is then integrated.

[0338] Output: A consolidated dataset.

[0339] What it does: The server queries the database using the user ID to retrieve past search history and hobbies and preferences. It sends a request to an external API to retrieve the latest recipe information, for example, from a cooking recipe service. It then integrates all the data and creates a dataset to generate optimal information.

[0340] Step 5:

[0341] Optimal information generation and display

[0342] Input: Unified dataset, user emotional state.

[0343] Processing: The server generates the optimal answer based on the integrated data, adjusts the content according to the user's emotional state, and sends the generated answer to the device.

[0344] Output: The adjusted answer is displayed on the terminal.

[0345] How it works: The server uses its internal algorithm to identify an "easy recipe for spaghetti with tomato and basil." The emotion engine adds a comment based on the user's emotional state. The server sends the answer in JSON format to the device, which then displays it on the screen.

[0346] Step 6:

[0347] Gathering feedback and learning

[0348] Input: User feedback data.

[0349] Processing: The device sends feedback data to the server. The server receives the feedback and stores it in a database. This information is used to train the generative model and emotion engine.

[0350] Output: A trained generative model and an emotion engine.

[0351] Specific operation: The user checks the provided information, enters a rating, and presses the "Submit" button. The device sends the feedback data to the server. The server stores the feedback in a database, and the learning module uses it to update the parameters of the generative model and emotion engine.

[0352] (Application example 2)

[0353] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0354] In modern brick-and-mortar stores, improving customer satisfaction and providing personalized service is a major challenge. It is particularly difficult to grasp a customer's emotional state in real time and provide appropriate product recommendations and services accordingly. Furthermore, conventional information provision systems using information terminals lack the ability to respond flexibly and consider the user's emotions. Furthermore, they lack an effective mechanism for accumulating feedback and reflecting it in future customer service.

[0355] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0356] In this invention, the server includes means for receiving a user's question input via an information terminal, means for analyzing the received question using natural language processing and grasping the user's intention, means for integrating the user's past data and information acquired from an external information service to generate optimal information, means for displaying the generated information on the information terminal, means for collecting user feedback input via the information terminal and training a generative model, means for recognizing the user's facial expression and grasping their emotional state, means for generating and displaying information taking the user's emotional state into consideration, and means for visually providing the displayed information via smart glasses. This enables personalized responses adapted to customer emotions and improves customer satisfaction in real time.

[0357] An "information terminal" is an electronic device that allows a user to input or receive information.

[0358] The "receiving means" is a device or program that has the function of receiving data transmitted from an information terminal.

[0359] "Natural language processing" is a technology in which a computer analyzes human language and understands its meaning.

[0360] "User intent" is the purpose or intention behind the question or request entered by the user.

[0361] "Past data" refers to data based on information previously generated by the user and their behavioral history.

[0362] "External information services" are additional sources of information provided by external servers or databases.

[0363] "Best information" is the answer or suggestion that best suits the user's question or requirements.

[0364] "Generating means" refers to a device or program that has the function of combining information based on collected data and creating the necessary information.

[0365] The "display means" is a device or program that has the function of presenting the generated information in a form that can be visually recognized by the user.

[0366] "Feedback" refers to a user's evaluation or opinion of the information or service provided.

[0367] The "collection means" is a device or program that has the function of collecting feedback from users.

[0368] A "generative model" is a computational model that generates information based on large amounts of data.

[0369] A "training means" is a device or program that has the function of improving the generative model based on collected feedback.

[0370] "Means for recognizing facial expressions" refers to a device or program that has the function of recognizing and analyzing a user's facial expressions using a camera or sensor.

[0371] "Emotional state" refers to a user's current feelings or mood.

[0372] A "means for taking into account" is a device or program that has the ability to adjust information based on the user's emotional state.

[0373] "Smart glasses" are a wearable eyeglass-type device that has the function of visually displaying information.

[0374] The present invention is a system that uses an information terminal, a server, a database, and an emotion engine to provide optimal answers to user questions. This system adjusts the content of the answers taking into account the user's emotions, thereby providing more appropriate information.

[0375] System Overview

[0376] The system consists of the following main components:

[0377] Information terminal: A device (e.g., smartphone, tablet, PC, etc.) through which a user enters questions and receives answers.

[0378] Server: The primary computer system for information analysis, data integration, and answer generation.

[0379] Database: Storage for saving user history and registration information.

[0380] Emotion engine: Software for recognizing a user's facial expressions and understanding their emotional state.

[0381] What the program does

[0382] The server receives the user's question sent from the information terminal and analyzes it using natural language processing (NLP) technology. Specifically, it uses the Python libraries TextBlob and NLTK. As a result of the analysis, keywords are extracted from the question to understand the user's intent.

[0383] The system then integrates the user's past data (e.g., previously searched information and ratings) with information obtained from external information services (e.g., the latest recipe service or product information service). This process uses API integration technology, for example, to obtain data from external information services via RESTful APIs.

[0384] Furthermore, an emotion engine (e.g., an OpenCV library for facial expression recognition) is used to analyze the user's facial expressions captured by the device's camera and understand their emotional state. Emotional data can be used to adjust the content of responses.

[0385] For example, if a user asks "What's an easy pasta recipe?", the system generates an answer in the following order:

[0386] 1. The server receives the question and uses NLP technology to extract the keywords "easy," "pasta," and "recipe."

[0387] 2. The server references the user's past search history in the database and uses data on Italian cuisine that the user previously searched for as a reference.

[0388] 3. The emotion engine analyzes facial expressions and recognizes the user's emotional state as "a little tired."

[0389] 4. The server retrieves the latest "Tomato and Basil Spaghetti" recipe from an external information service and generates a message saying, "You seem a little tired today. Try refreshing yourself with this easy pasta recipe."

[0390] 5. The generated answer is sent to the information terminal and visually displayed to the user.

[0391] Specific examples

[0392] Here is an example where a user enters "easy pasta recipes." The server analyzes the question and references the user's past search history from the database. If the emotion engine recognizes the emotion "a little tired" from the user's input, the server connects with an external information service and retrieves a recipe for "spaghetti with tomato and basil." Based on this, the server makes adjustments such as "You seem a little tired today. Try refreshing yourself with this easy pasta recipe," generates an answer, and sends it to the information terminal for display.

[0393] Prompt Sentence Examples

[0394] 1. "Tell me an easy pasta recipe."

[0395] 2. "Do you have any drinks to cheer me up?"

[0396] 3. "I'm looking for some new shoes. What do you recommend?"

[0397] 4. "Do you have any teas that can help relieve stress?"

[0398] In this way, the present invention is a system that enables flexible responses according to the user's emotions, thereby realizing more appropriate information provision and improving customer satisfaction.

[0399] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0400] Step 1:

[0401] The server receives the user's question entered via an information terminal. The input data is a question, specifically text information such as "Tell me an easy pasta recipe." The received question is passed directly to the next analysis step.

[0402] Step 2:

[0403] The server analyzes the received question using natural language processing (NLP) techniques. Specifically, it uses Python's TextBlob and NLTK libraries to parse the question and extract keywords. The input is the question, and the output is a list of keywords such as "easy," "pasta," and "recipe."

[0404] Step 3:

[0405] The server retrieves the user's search history and evaluation data from the user's past database. At the same time, it also retrieves the latest related information through the API of an external information service (e.g., a recipe provision service). The input is a list of keywords, and the output is a list of relevant information.

[0406] Step 4:

[0407] The emotion engine captures the user's facial expressions in real time through the information terminal's camera and analyzes their emotional state using facial expression recognition technology (such as the OpenCV library).The input is real-time facial image data, and the output is emotional states such as "happiness," "sadness," and "tiredness."

[0408] Step 5:

[0409] The server combines the user's emotional state, past data, and external information to generate the optimal information response. Using a generative AI model, it combines these data and creates a response with wording appropriate to the user's emotions. The input is the emotional state and a list of information, and the output is the text of the optimal response.

[0410] Step 6:

[0411] The server sends the generated answer to the information terminal and displays it on the user's smart glasses or other information terminal. The user can visually confirm the displayed answer. The input is the text of the best answer, and the output is the answer displayed on the information terminal.

[0412] Step 7:

[0413] The user inputs feedback on the provided information through the information terminal. The feedback includes specific evaluations such as "This recipe was very useful." The input is the user's feedback text, and the output is the feedback data.

[0414] Step 8:

[0415] The server stores the collected feedback in a database and uses it as training data for the generative model and emotion engine. This improves the accuracy of answers from the next time onwards. The input is the feedback data, and the output is the model reflected in the training.

[0416] 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 a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the 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.

[0417] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0418] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0419] [Second embodiment]

[0420] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0421] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0422] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0423] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0424] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0425] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0426] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0427] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0428] The specific processing program 56 is an example of a "program" according to the technology of the present 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.

[0429] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0430] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. 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 process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0431] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. 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."

[0432] The present invention is a system that uses an information terminal, a server, and a database, and provides optimal answers to questions asked by users in information searches. A specific embodiment of this system will be described below.

[0433] User registration and information collection

[0434] First, a user accesses the service using an information terminal and enters basic information such as name, age, occupation, hobbies and interests into the registration form. The terminal sends this information to the server, which then stores it in a database.

[0435] Entering and parsing questions

[0436] The user inputs a specific question, such as "Tell me an easy pasta recipe," into the device and sends it. This input data is sent to the server. The server analyzes the received question using natural language processing (NLP) technology, extracts keywords such as "easy," "pasta," and "recipe," and understands the user's intent.

[0437] Integrating historical data and external information

[0438] Based on the analysis results, the server retrieves the user's past data from the database. For example, if the user has previously searched for Italian food, that information will be used as a reference. The server also connects with external information services (such as cooking recipe providers) via API to retrieve the latest recipe information.

[0439] Generating and providing optimal information

[0440] The server combines the user's hobby and interest data with external information to generate the most appropriate answer. In this case, it identifies "an easy recipe for spaghetti with tomatoes and basil" and generates an answer in natural language format. The generated answer is sent to the device and displayed to the user.

[0441] Gathering feedback and learning

[0442] The user reviews the provided information and sends feedback such as "This recipe was very helpful" via their device. This feedback is received by the server and stored in a database. The server uses this feedback to continuously train the generative model, improving the accuracy of answers from the next time onwards.

[0443] Specific examples

[0444] For example, suppose a user inputs a question about "easy pasta recipes." The server analyzes this question and references the user's past search history and hobbies and preferences from a database. If the user has previously liked "Italian food," the server connects with an external information service to retrieve the latest recipe for "spaghetti with tomato and basil." The server formats this information in natural language and sends it to the device. The device displays it to the user as "an easy recipe for spaghetti with tomato and basil." If the user then rates this recipe as useful, that feedback is sent back to the server and used as training data for the model.

[0445] In this way, the present invention is a system that continues to provide optimal information in response to user questions, thereby improving the efficiency of information acquisition and user satisfaction.

[0446] The processing flow will be explained below.

[0447] Step 1:

[0448] The user accesses the service using an information terminal and enters basic information such as name, age, occupation, hobbies and interests into the registration form.

[0449] The terminal transmits this information to the server.

[0450] Step 2:

[0451] The server receives the basic information sent by the user and stores it in a database.

[0452] The server also generates a unique user ID for each user and creates a corresponding record.

[0453] Step 3:

[0454] The user enters a specific question, such as "Tell me an easy pasta recipe," into the terminal and sends it.

[0455] The terminal transmits the entered question to the server.

[0456] Step 4:

[0457] The server analyzes the received question using natural language processing (NLP) techniques.

[0458] The server extracts key keywords from the question ("easy," "pasta," "recipe") and understands the user's intent.

[0459] Step 5:

[0460] The server retrieves the user's past data from the database.

[0461] For example, if the user has previously searched for "Italian cuisine," that information can be used as a reference.

[0462] Step 6:

[0463] The server initiates API integration with an external information service to obtain the latest recipe information.

[0464] For example, recipe data for "spaghetti with tomato and basil" is received from an external recipe service.

[0465] Step 7:

[0466] The server integrates the acquired external information with the user's past data to generate the optimal answer.

[0467] For example, create an "easy tomato and basil spaghetti recipe" in natural language format.

[0468] Step 8:

[0469] The server sends the generated response to the terminal.

[0470] The device receives the response and displays it to the user as an "easy tomato and basil spaghetti recipe."

[0471] Step 9:

[0472] The user reviews the information provided and sends feedback via the device, such as "This recipe was helpful."

[0473] The terminal transmits the feedback information to the server.

[0474] Step 10:

[0475] The server stores the user feedback in a database.

[0476] The server uses this feedback to train a generative model and uses it to improve the accuracy of answers from next time onwards.

[0477] Example 1

[0478] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0479] Conventional information provision systems have difficulty providing accurate answers to user questions, and in order to improve user satisfaction, it is necessary to effectively integrate the user's past behavioral history and external information. Furthermore, due to a lack of means to collect user feedback and improve the accuracy of the system, continuous learning is not possible, raising concerns that the quality of the information provided may decline.

[0480] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0481] In this invention, the server includes means for receiving a user's question input via an information terminal, means for analyzing the received question using natural language processing to understand the user's intention, means for integrating the user's past data and information acquired from an external database to generate optimal information, means for transmitting the generated information to the information terminal and displaying it, and means for collecting user feedback input via the information terminal and training the generation AI model. This makes it possible to continue providing optimal information in response to a user's question, thereby improving the efficiency of information acquisition and user satisfaction.

[0482] An "information terminal" is a device that allows a user to access the system via the Internet and input and display data, and includes smartphones, personal computers, etc.

[0483] A "user question" is a textual inquiry that a user inputs into the system using an information terminal, requesting specific information.

[0484] "Natural language processing" is a technology that analyzes input text data in a way that is easy for a computer to understand, and grasps its meaning and intent.

[0485] "User intent" refers to the specific information or answer that the user wants to obtain through the question.

[0486] "Past data" refers to information such as a user's past behavioral history, search history, and interests that is stored in a database.

[0487] An "external database" is a data source that exists outside the system, and includes various information providing services that can be accessed via API.

[0488] "Optimal information" refers to the most appropriate and useful answer to a user's question, and is generated by integrating information from past data and external databases.

[0489] A "generative AI model" is an algorithm that continuously learns from data and feedback provided by users, improving the accuracy of its answers.

[0490] "Feedback" refers to the evaluations and opinions that users give in response to the information provided, and is used to help the system learn.

[0491] "Display means" refers to a function for visually presenting optimal information on an information terminal, and includes a screen or display.

[0492] "Integration means" is a process that combines the user's past data and information obtained from external databases to generate information that is optimal for the user.

[0493] The present invention is a system that uses an information terminal, a server, and a database, and provides optimal answers to questions asked by users in information searches. An embodiment of this system will be described in detail below.

[0494] First, the user accesses the system's website using an information terminal (smartphone or PC). The user enters basic information such as name, age, occupation, hobbies and interests into the registration form on the website. When the user presses the "Submit" button, the information terminal sends the entered data to the server using the HTTPS protocol. The server stores the received data in a database server (e.g., MySQL, PostgreSQL).

[0495] Next, the user enters a specific question. For example, "Tell me an easy pasta recipe" and clicks the submit button. The input data is again sent to the server via HTTPS protocol. The server analyzes the received question using natural language processing (NLP) technology. At this time, NLP libraries such as spaCy and NLTK are used to extract keywords such as "easy," "pasta," and "recipe." This allows the server to understand the user's intent.

[0496] The server then retrieves the user's past behavioral history and preference information from the database. This information is obtained using an SQL query. The server also connects to external databases (e.g., recipe service) via API to retrieve the latest recipe information. The acquired internal data is then integrated with external data to generate the optimal recipe information for the user. A generative AI model is used to generate a natural language answer (e.g., "Recipe for spaghetti with tomato and basil").

[0497] The generated answer is sent from the server to the device, which displays it to the user. The user checks the displayed information, enters feedback (e.g., "This recipe was very helpful"), and sends it to the server via the device. The server stores the feedback in a database and uses this information to continuously train the generative AI model. This improves the accuracy of answers in future queries.

[0498] Specific examples

[0499] For example, if a user inputs a question such as "Tell me an easy pasta recipe," the system operates as follows: The server first analyzes the question using NLP technology to extract keywords. It then combines the user's past data with the latest recipe information obtained from an external API to generate an "easy tomato and basil spaghetti recipe." This generated information is sent to the device and displayed to the user. Finally, the user sends feedback to the server on whether the provided recipe was helpful, and this feedback is used to train the generative AI model.

[0500] In this way, the present invention is a system that continues to provide optimal information in response to user questions, thereby improving the efficiency of information acquisition and user satisfaction.

[0501] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0502] Step 1:

[0503] A user accesses the service and enters registration information.

[0504] Specific operation: The user uses an information terminal (smartphone or PC) to access the system's website.

[0505] Input: A user fills in a registration form on a website with information such as name, age, occupation, hobbies and interests.

[0506] Data processing: The device temporarily stores the entered information.

[0507] Output: The persisted data is ready to be sent to the server in the next step.

[0508] Step 2:

[0509] The terminal transmits the input user information to the server.

[0510] What happens: The user clicks a button to submit the registration form.

[0511] Input: Stored user information (name, age, occupation, hobbies and interests).

[0512] Data calculation: The information terminal uses the HTTPS protocol to send the input information to the server.

[0513] Output: The server receives the user information.

[0514] Step 3:

[0515] The server stores the received user information in a database.

[0516] Specific operation: The server analyzes the received registration data and stores it in a database.

[0517] Input: User registration information.

[0518] Data manipulation: Creating new records in the database using SQL queries.

[0519] Output: The saved user information.

[0520] Step 4:

[0521] The user types a question and submits it.

[0522] Specific operation: The user enters a question on the information terminal and clicks the send button.

[0523] Input: Question text (e.g. "What's an easy pasta recipe?").

[0524] Data calculation: The information terminal sends the entered query data to the server.

[0525] Output: The server receives the query data.

[0526] Step 5:

[0527] The server analyzes the question using natural language processing.

[0528] What happens: The server parses the question using an NLP library (e.g., spaCy, NLTK).

[0529] Input: The question text submitted by the user.

[0530] Data processing: Extract keywords from the question text to understand the user's intent.

[0531] Output: Extracted keywords and user intent.

[0532] Step 6:

[0533] The server retrieves the user's past data from the database.

[0534] Specific operation: The server retrieves past behavioral history and interest data from the database based on the user's ID.

[0535] Input: User ID and parsed question information.

[0536] Data operations: Retrieving relevant data from a database using SQL queries.

[0537] Output: User's historical data.

[0538] Step 7:

[0539] The server retrieves the latest information from an external database.

[0540] Specific operation: The server calls an external API to obtain the latest recipe information.

[0541] Input: Extracted keywords (e.g., pasta, recipe).

[0542] Data Calculation: Sending requests to external APIs and parsing the responses.

[0543] Output: Latest recipe information.

[0544] Step 8:

[0545] The server generates the most appropriate information and provides it to the user.

[0546] Specific operation: The server integrates the user's past data with external data and generates answers in natural language format using a generative AI model.

[0547] Input: User's past data, latest recipe information, parsed question information.

[0548] Data processing: Generate optimal answers from integrated data.

[0549] Output: A natural language answer (e.g., "Easy tomato and basil spaghetti recipe").

[0550] Step 9:

[0551] The terminal displays the generated answer to the user.

[0552] Specific operation: The device displays the answer received from the server to the user.

[0553] Input: The answer sent by the server in natural language format.

[0554] Data processing: Formatting text for visual display on the device.

[0555] Output: The answer displayed to the user.

[0556] Step 10:

[0557] The user enters and submits feedback.

[0558] Specific actions: The user enters feedback on the device and clicks the submit button.

[0559] Input: Feedback text (e.g., "This recipe was very helpful").

[0560] Data calculation: The information terminal sends the input feedback to the server.

[0561] Output: The server receives the feedback data.

[0562] Step 11:

[0563] The server stores the feedback in a database and trains a generative AI model.

[0564] Specific operation: The server stores the feedback data in a database and updates the generative AI model.

[0565] Input: Received feedback data.

[0566] Data processing: Store the feedback in a database and use it as new data for training the AI ​​model.

[0567] Output: An updated generative AI model.

[0568] This allows the system to continue providing optimal information in response to user questions, improving the efficiency of information acquisition and user satisfaction.

[0569] (Application example 1)

[0570] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0571] Conventional product recommendation systems in virtual stores were unable to fully reflect the individual preferences and past purchase history of users, making it difficult to provide optimal products. Furthermore, there was no mechanism for effectively collecting and utilizing user feedback, making it difficult to improve the quality of service.

[0572] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0573] In this invention, the server includes means for receiving a user question input via an information terminal, means for analyzing the received question using natural language processing to understand the user's intention, means for integrating the user's past data and information acquired from external information services to recommend optimal products based on the user's preferences, means for displaying the generated product information on the information terminal, and means for collecting user feedback input via the information terminal and training a generative model. This makes it possible to propose optimal products based on the individual preferences of users and improve the quality of services by utilizing feedback.

[0574] An "information terminal" refers to an electronic device, such as a smartphone or tablet, that allows a user to input questions and receive answers from a server.

[0575] "Natural language processing" refers to the technology that allows computers to understand and analyze human language, allowing them to understand the intent of users' questions.

[0576] "User's past data" refers to information such as the user's past search history, purchase history, and feedback.

[0577] "External Information Services" refers to information or databases provided by third parties, such as product catalogs or review sites.

[0578] "Means for recommending optimal products" refers to algorithms or programs that integrate the user's past data with information obtained from external information services and select the most suitable product based on the user's preferences.

[0579] The "generated product information" refers to detailed product information selected by the optimal product recommendation means, including, for example, the product name, price, and features.

[0580] "Feedback" refers to a user's evaluation or opinion of the information provided or the recommended products.

[0581] A "generative model" refers to a machine learning model that generates optimal information based on user questions and feedback.

[0582] The present invention provides a personalized shopping assistant system for a virtual store that recommends optimal products based on a user's preferences and past purchase history. Specific embodiments of this system are described below.

[0583] Hardware and software used

[0584] This system consists of information terminals (e.g., smartphones, tablets, smart glasses) accessed by users and a server that processes data. The software includes the following elements:

[0585] 1. Natural Language Processing (NLP) engine: used to analyze the user's question and extract keywords, for example using Python and the transformers library.

[0586] 2. Database: Stores user registration information, past data, and product information obtained from external information services. For example, a database system such as MySQL is used.

[0587] 3. API integration module: Connects to external information services (e.g., product catalogs, review sites, etc.) via API to obtain the latest product information.

[0588] 4. Generative AI model: A model for recommending optimal products based on user preference data and external information. As an example, we will use a machine learning model.

[0589] System Operation

[0590] The server operates as follows:

[0591] 1. Accepting a user question: The user's question (e.g., "Please recommend a new smartphone") is received via the information terminal.

[0592] 2. Natural language processing of questions: The content of the question is analyzed using natural language processing technology to extract important keywords.

[0593] 3. Information integration: Integrates the user's past data (e.g., past purchase history) and the latest product information obtained from external information services to select the optimal product based on the user's preferences.

[0594] 4. Product information generation and display: The selected product information is formatted in natural language and displayed on the information terminal.

[0595] 5. Feedback collection: Collect user feedback entered via the information terminal and continuously train the generative AI model.

[0596] Specific examples

[0597] For example, a user might use an information terminal to ask, "Please recommend a new smartphone." This input is sent to the server and analyzed by a natural language processing engine. As a result of the analysis, important keywords such as "smartphone" and "recommendation" are extracted. The server then retrieves the user's past purchase history and preference data from a database, and obtains the latest smartphone information through the API of an external information service.

[0598] The generative AI model integrates this information and recommends the most suitable smartphone for the user (e.g., "latest model smartphone"). This information is formatted in natural language and displayed on the information terminal. When the user reviews the recommendation and sends feedback saying "this suggestion was helpful," that feedback is stored on the server and used as training data for the generative AI model.

[0599] Examples of prompt statements

[0600] User: Recommend me a new smartphone.

[0601] System: Based on your past search and purchase history, we estimate that a new smartphone model would be suitable. We recommend a 'new smartphone model' featuring a high-performance camera and a long-lasting battery. The price is 50,000 yen. How about this product?

[0602] As described above, the present invention makes it possible to provide appropriate and personalized product suggestions based on the individual preferences and past data of the user.

[0603] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0604] Step 1:

[0605] The user enters a question.

[0606] The user uses an information terminal (e.g., a smartphone) to input a question. Specifically, the user inputs text such as "Please recommend a new smartphone" and presses the send button. The input question (text data) is sent to the server.

[0607] Step 2:

[0608] The server accepts the query.

[0609] The server receives questions sent by users, stores them as text data, and then passes the text data to a natural language processing engine.

[0610] Step 3:

[0611] The server performs natural language processing.

[0612] The server uses a natural language processing engine (e.g., the transformers library) to analyze the received text data, split the question, and extract important keywords (e.g., "smartphone," "recommended"). The results with the extracted keywords are then returned.

[0613] Step 4:

[0614] The server obtains the user's past data and external information.

[0615] The server retrieves the user's past data (past purchase history, search history, etc.) from the database based on the keywords. At the same time, it calls an external information service API to retrieve the latest product information related to the keywords. The retrieved data is then integrated.

[0616] Step 5:

[0617] The server recommends the best products.

[0618] The server uses the generative AI model to analyze and integrate the user's past data and product data obtained from external information, thereby selecting the product (e.g., the latest model smartphone) that best suits the user's preferences.

[0619] Step 6:

[0620] The server generates and displays product information.

[0621] The server formats the selected product details (product name, price, features, etc.) into a natural language format. This generated product information is sent to the information terminal, which then displays the information to the user.

[0622] Step 7:

[0623] The user enters feedback.

[0624] The user checks the presented product information and inputs feedback such as "This suggestion was helpful." The input feedback is then sent back to the server.

[0625] Step 8:

[0626] The server collects feedback and learns.

[0627] The server receives feedback sent by the user, stores it in a database, and uses it as training data for the generative AI model, thereby improving the accuracy of recommendations in the future.

[0628] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0629] The present invention is a system that uses an information terminal, a server, a database, and an emotion engine to provide optimal answers to questions asked by users in information searches, and further adjusts the content of the answers taking into account the user's emotions. A specific embodiment of this system is described below.

[0630] User registration and information collection

[0631] First, a user accesses the service using an information terminal and enters basic information such as name, age, occupation, hobbies and interests into the registration form. The terminal sends this information to the server, which stores it in a database. The emotion engine also performs initial settings to understand the user's basic emotional patterns.

[0632] Entering and parsing questions

[0633] The user inputs a specific question, such as "Tell me an easy pasta recipe," into the device and sends it. This input data is sent to the server. The server analyzes the received question using natural language processing (NLP) technology, extracts keywords such as "easy," "pasta," and "recipe," and understands the user's intent.

[0634] Acquiring user emotion data

[0635] When a question is entered, the emotion engine collects the user's emotion data and uses, for example, text analysis and facial expression recognition technology to understand the user's current emotional state (such as joy, sadness, or anger).

[0636] Integrating historical data and external information

[0637] Based on the analysis results, the server retrieves the user's past data from the database. For example, if the user has previously searched for Italian food, that information will be used as a reference. The server also connects with external information services (such as cooking recipe providers) via API to retrieve the latest recipe information.

[0638] Optimal information generation and display

[0639] The server combines the acquired external information, the user's past data, and emotional data to generate the optimal answer. For example, it identifies an "easy tomato and basil spaghetti recipe" and generates an answer tailored to the user's emotional state, such as "You seem to be in a low mood today. Try this easy pasta recipe to cheer you up." The generated answer is sent to the device and displayed to the user.

[0640] Gathering feedback and learning

[0641] The user reviews the provided information and sends feedback via their device, such as "This recipe was very helpful." This feedback is received by the server and stored in a database. The server uses this feedback to continuously train the generative model and emotion engine, and uses it to improve the accuracy of answers in future searches.

[0642] Specific examples

[0643] For example, if a user enters "easy pasta recipes," the server analyzes the question and references the user's past search history and hobbies and preferences from a database. If the emotion engine recognizes the emotion "a little tired" from the user's input, the server connects with an external information service to retrieve a recipe for "spaghetti with tomato and basil." Based on this, the server makes adjustments such as "You seem a little tired today. Try refreshing yourself with this easy pasta recipe," and sends and displays it on the device. If the user rates the recipe as "helpful," that feedback is sent back to the server and used as learning data for the model. The emotion engine also improves its accuracy based on feedback.

[0644] In this way, the present invention is a system that provides optimal information in response to a user's question while enabling flexible responses according to the user's emotional state, thereby improving the efficiency of information acquisition and user satisfaction.

[0645] The processing flow will be explained below.

[0646] Step 1:

[0647] The user accesses the service using an information terminal and enters basic information such as name, age, occupation, hobbies and interests into the registration form.

[0648] The terminal transmits this information to the server.

[0649] Step 2:

[0650] The server receives the basic information sent by the user and stores it in a database.

[0651] The server also generates a unique user ID for each user and creates a corresponding record.

[0652] Step 3:

[0653] The user enters a specific question, such as "Tell me an easy pasta recipe," into the terminal and sends it.

[0654] The terminal transmits the entered question to the server.

[0655] Step 4:

[0656] The emotion engine analyzes the user's input text and recognizes emotional states (e.g., joy, sadness, fatigue, etc.).

[0657] The server acquires the emotion data and processes it together with the question data.

[0658] Step 5:

[0659] The server analyzes the received question using natural language processing (NLP) techniques.

[0660] The server extracts key keywords from the question ("easy," "pasta," "recipe") and understands the user's intent.

[0661] Step 6:

[0662] The server retrieves the user's past data from the database.

[0663] For example, if the user has previously searched for "Italian cuisine," that information can be used as a reference.

[0664] Step 7:

[0665] The server initiates API integration with an external information service to obtain the latest recipe information.

[0666] For example, recipe data for "spaghetti with tomato and basil" is received from an external recipe service.

[0667] Step 8:

[0668] The server integrates the acquired external information, past data, and emotional data to generate the optimal answer.

[0669] For example, if the user's emotional state is recognized as "tired," the system adjusts the response by saying, "You seem a little tired today. Try refreshing yourself with this simple pasta recipe."

[0670] Step 9:

[0671] The server sends the generated response to the terminal.

[0672] The device receives the response and displays it to the user as an "easy tomato and basil spaghetti recipe."

[0673] Step 10:

[0674] The user reviews the information provided and sends feedback via the device, such as "This recipe was very helpful."

[0675] The terminal transmits the feedback information to the server.

[0676] Step 11:

[0677] The server stores the user feedback in a database.

[0678] The server uses this feedback to continuously train the generative model and emotion engine, and uses it to improve the accuracy of answers from next time onwards.

[0679] Example 2

[0680] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0681] Conventional information retrieval systems provide answers only to questions entered by the user and do not take the user's emotional state into account, which means that the answers may not necessarily meet the user's needs. Furthermore, the means for collecting feedback are limited, which leaves the user experience unsatisfactory. Therefore, the present invention aims to provide optimal information that takes into account the user's emotional state and continuously improve the system's accuracy by utilizing feedback.

[0682] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0683] In this invention, the server includes means for receiving a user question input via an information terminal, means for analyzing the received question using natural language processing to understand the user's intention, means for analyzing the user's current emotional state, means for integrating the user's past data and information obtained from external information services to generate optimal information, means for adjusting the generated information according to the user's emotional state and displaying it on the information terminal, and means for collecting user feedback input via the information terminal and training the generative model and emotion engine. This makes it possible to provide optimal information in response to the user's question while also responding flexibly to the user's emotional state.

[0684] An "information terminal" is a device used by a user to input a question, and includes a computer, a smartphone, a tablet, and the like.

[0685] A "server" is a central system that receives data from information terminals via a network and analyzes and processes the data.

[0686] "Natural Language Processing (NLP)" is a technology that allows computers to understand and analyze human language and grasp the user's intent.

[0687] An "emotion engine" is an algorithm and software that analyzes and identifies a user's emotional state from their text and facial expressions.

[0688] A "database" is a system for managing and storing user registration information and past data.

[0689] "External information services" are third-party services that provide information via the Internet.

[0690] A "generative model" is an algorithm and software that generates optimal information in response to a user's question.

[0691] "Feedback" refers to the evaluation or opinion that a user enters regarding the information provided.

[0692] "Synthesis" is the process of combining data obtained from multiple different sources into a single, consistent piece of information.

[0693] User registration and information collection

[0694] First, the user accesses the service using an information terminal (smartphone, computer, tablet, etc.) and enters basic information such as name, age, occupation, hobbies and interests into the registration form. This information is then sent from the terminal to the server. The server stores the received information in a database and performs initial settings to understand the user's basic emotional patterns using an emotion engine. At this point, the server stores the user's individual information in a database, making it possible to provide individualized information.

[0695] Entering and parsing questions

[0696] The user inputs a specific question into an information terminal. For example, they can type in the text "Tell me an easy pasta recipe" and send it. This input data is sent from the terminal to the server. The server analyzes the received question using natural language processing (NLP) technology, extracts keywords such as "easy," "pasta," and "recipe," and understands the user's intent. This analysis can be performed using open-source natural language processing libraries or cloud-based NLP services.

[0697] Acquiring user emotion data

[0698] When a user enters a question, the emotion engine collects the user's emotion data. It can use a camera-equipped information device to understand the user's current emotional state (e.g., joy, sadness, anger) using text analysis and facial expression recognition technology. For emotion recognition, it can use, for example, an open-source emotion analysis library or a cloud-based emotion recognition API.

[0699] Integrating historical data and external information

[0700] Based on the analysis results, the server retrieves the user's past data (e.g., past search history, hobbies, and preferences) from the database. It also connects with external information services (e.g., cooking recipe services) via API to retrieve the latest recipe information. During this process, the server communicates with external services using an enterprise DBMS, cloud database service, or REST API.

[0701] Optimal information generation and display

[0702] The server combines the acquired external information, the user's past data, and emotional data to generate the optimal answer. For example, it identifies an "easy tomato and basil spaghetti recipe" and generates an answer tailored to the user's emotional state, such as "You seem to be in a low mood today. Try this easy pasta recipe to cheer you up." The generated answer is sent to the device and displayed to the user.

[0703] Gathering feedback and learning

[0704] The user reviews the provided information and sends feedback via their device, such as "This recipe was very helpful." This feedback is received by the server and stored in a database. The server uses this feedback to continuously train the generative model and emotion engine, and uses it to improve the accuracy of answers in the future.

[0705] Specific examples

[0706] For example, if a user enters "easy pasta recipes," the server analyzes the question and references the user's past search history and hobbies and preferences from a database. If the emotion engine recognizes the emotion "a little tired" from the user's input, the server connects with an external information service to retrieve a recipe for "spaghetti with tomato and basil." Based on this, the server makes adjustments such as "You seem a little tired today. Try refreshing yourself with this easy pasta recipe," and sends and displays it on the device. If the user rates the recipe as "helpful," that feedback is sent back to the server and used as learning data for the model. The emotion engine also improves its accuracy based on feedback.

[0707] Example prompts to input to the generative AI model

[0708] "When a user is looking for an easy pasta recipe, give them the perfect recipe based on their emotional state and why. For example, 'You seem a little tired today. Try refreshing yourself with this easy pasta recipe.'"

[0709] In this way, the present invention is a system that provides optimal information in response to a user's question while enabling flexible responses according to the user's emotional state, thereby improving the efficiency of information acquisition and user satisfaction.

[0710] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0711] Step 1:

[0712] User registration and information collection

[0713] Input: The user uses an information terminal to input basic information such as name, age, occupation, hobbies and interests.

[0714] Processing: The device sends the input information to the server. The server stores the received information in a database and sets the user's initial emotional pattern using the emotion engine.

[0715] Output: User information is saved in the database and the emotion engine completes its initialization.

[0716] How it works: The user accesses the registration page in a web browser or mobile app, enters the required information in the form, and then presses the "Register" button. The device sends this data to the server as an HTTP request. The server saves the information in a database and calls the emotion recognition API to set the initial emotion pattern.

[0717] Step 2:

[0718] Entering and parsing questions

[0719] Input: The user types a specific question into the device and sends it. For example, the text "Tell me a simple pasta recipe."

[0720] Processing: The device sends the input data to the server, which uses natural language processing (NLP) techniques to analyze the question and extract keywords.

[0721] Output: Keyword data is generated to understand the intent of the question.

[0722] Specific operation: The user enters a question into the text input form and presses the "Submit" button. The device sends the text data as an HTTP POST request to the server. The server uses the NLP module to extract keywords such as "easy," "pasta," and "recipe."

[0723] Step 3:

[0724] Acquiring user emotion data

[0725] Input: User text and facial expression data when entering a question.

[0726] Processing: The server uses the emotion engine to analyze the user's current emotional state (happiness, sadness, anger, etc.).

[0727] Output: Data of the current user's emotional state.

[0728] Specific operation: The server inputs text data into the emotion engine and calls the emotion recognition API. The emotion engine analyzes emotions such as "joy," "sadness," and "anger." The analysis results are recorded for use in the next step.

[0729] Step 4:

[0730] Integrating historical data and external information

[0731] Input: Question analysis results, user history, data from external information services.

[0732] Processing: The server retrieves the user's past data from the database, connects to external information services via API, and retrieves the latest information. This data is then integrated.

[0733] Output: A consolidated dataset.

[0734] What it does: The server queries the database using the user ID to retrieve past search history and hobbies and preferences. It sends a request to an external API to retrieve the latest recipe information, for example, from a cooking recipe service. It then integrates all the data and creates a dataset to generate optimal information.

[0735] Step 5:

[0736] Optimal information generation and display

[0737] Input: Unified dataset, user emotional state.

[0738] Processing: The server generates the optimal answer based on the integrated data, adjusts the content according to the user's emotional state, and sends the generated answer to the device.

[0739] Output: The adjusted answer is displayed on the terminal.

[0740] How it works: The server uses its internal algorithm to identify an "easy recipe for spaghetti with tomato and basil." The emotion engine adds a comment based on the user's emotional state. The server sends the answer in JSON format to the device, which then displays it on the screen.

[0741] Step 6:

[0742] Gathering feedback and learning

[0743] Input: User feedback data.

[0744] Processing: The device sends feedback data to the server. The server receives the feedback and stores it in a database. This information is used to train the generative model and emotion engine.

[0745] Output: A trained generative model and an emotion engine.

[0746] Specific operation: The user checks the provided information, enters a rating, and presses the "Submit" button. The device sends the feedback data to the server. The server stores the feedback in a database, and the learning module uses it to update the parameters of the generative model and emotion engine.

[0747] (Application example 2)

[0748] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0749] In modern brick-and-mortar stores, improving customer satisfaction and providing personalized service is a major challenge. It is particularly difficult to grasp a customer's emotional state in real time and provide appropriate product recommendations and services accordingly. Furthermore, conventional information provision systems using information terminals lack the ability to respond flexibly and consider the user's emotions. Furthermore, they lack an effective mechanism for accumulating feedback and reflecting it in future customer service.

[0750] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0751] In this invention, the server includes means for receiving a user's question input via an information terminal, means for analyzing the received question using natural language processing and grasping the user's intention, means for integrating the user's past data and information acquired from an external information service to generate optimal information, means for displaying the generated information on the information terminal, means for collecting user feedback input via the information terminal and training a generative model, means for recognizing the user's facial expression and grasping their emotional state, means for generating and displaying information taking the user's emotional state into consideration, and means for visually providing the displayed information via smart glasses. This enables personalized responses adapted to customer emotions and improves customer satisfaction in real time.

[0752] An "information terminal" is an electronic device that allows a user to input or receive information.

[0753] The "receiving means" is a device or program that has the function of receiving data transmitted from an information terminal.

[0754] "Natural language processing" is a technology in which a computer analyzes human language and understands its meaning.

[0755] "User intent" is the purpose or intention behind the question or request entered by the user.

[0756] "Past data" refers to data based on information previously generated by the user and their behavioral history.

[0757] "External information services" are additional sources of information provided by external servers or databases.

[0758] "Best information" is the answer or suggestion that best suits the user's question or requirements.

[0759] "Generating means" refers to a device or program that has the function of combining information based on collected data and creating the necessary information.

[0760] The "display means" is a device or program that has the function of presenting the generated information in a form that can be visually recognized by the user.

[0761] "Feedback" refers to a user's evaluation or opinion of the information or service provided.

[0762] The "collection means" is a device or program that has the function of collecting feedback from users.

[0763] A "generative model" is a computational model that generates information based on large amounts of data.

[0764] A "training means" is a device or program that has the function of improving the generative model based on collected feedback.

[0765] "Means for recognizing facial expressions" refers to a device or program that has the function of recognizing and analyzing a user's facial expressions using a camera or sensor.

[0766] "Emotional state" refers to a user's current feelings or mood.

[0767] A "means for taking into account" is a device or program that has the ability to adjust information based on the user's emotional state.

[0768] "Smart glasses" are a wearable eyeglass-type device that has the function of visually displaying information.

[0769] The present invention is a system that uses an information terminal, a server, a database, and an emotion engine to provide optimal answers to user questions. This system adjusts the content of the answers taking into account the user's emotions, thereby providing more appropriate information.

[0770] System Overview

[0771] The system consists of the following main components:

[0772] Information terminal: A device (e.g., smartphone, tablet, PC, etc.) through which a user enters questions and receives answers.

[0773] Server: The primary computer system for information analysis, data integration, and answer generation.

[0774] Database: Storage for saving user history and registration information.

[0775] Emotion engine: Software for recognizing a user's facial expressions and understanding their emotional state.

[0776] What the program does

[0777] The server receives the user's question sent from the information terminal and analyzes it using natural language processing (NLP) technology. Specifically, it uses the Python libraries TextBlob and NLTK. As a result of the analysis, keywords are extracted from the question to understand the user's intent.

[0778] The system then integrates the user's past data (e.g., previously searched information and ratings) with information obtained from external information services (e.g., the latest recipe service or product information service). This process uses API integration technology, for example, to obtain data from external information services via RESTful APIs.

[0779] Furthermore, an emotion engine (e.g., an OpenCV library for facial expression recognition) is used to analyze the user's facial expressions captured by the device's camera and understand their emotional state. Emotional data can be used to adjust the content of responses.

[0780] For example, if a user asks "What's an easy pasta recipe?", the system generates an answer in the following order:

[0781] 1. The server receives the question and uses NLP technology to extract the keywords "easy," "pasta," and "recipe."

[0782] 2. The server references the user's past search history in the database and uses data on Italian cuisine that the user previously searched for as a reference.

[0783] 3. The emotion engine analyzes facial expressions and recognizes the user's emotional state as "a little tired."

[0784] 4. The server retrieves the latest "Tomato and Basil Spaghetti" recipe from an external information service and generates a message saying, "You seem a little tired today. Try refreshing yourself with this easy pasta recipe."

[0785] 5. The generated answer is sent to the information terminal and visually displayed to the user.

[0786] Specific examples

[0787] Here is an example where a user enters "easy pasta recipes." The server analyzes the question and references the user's past search history from the database. If the emotion engine recognizes the emotion "a little tired" from the user's input, the server connects with an external information service and retrieves a recipe for "spaghetti with tomato and basil." Based on this, the server makes adjustments such as "You seem a little tired today. Try refreshing yourself with this easy pasta recipe," generates an answer, and sends it to the information terminal for display.

[0788] Prompt Sentence Examples

[0789] 1. "Tell me an easy pasta recipe."

[0790] 2. "Do you have any drinks to cheer me up?"

[0791] 3. "I'm looking for some new shoes. What do you recommend?"

[0792] 4. "Do you have any teas that can help relieve stress?"

[0793] In this way, the present invention is a system that enables flexible responses according to the user's emotions, thereby realizing more appropriate information provision and improving customer satisfaction.

[0794] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0795] Step 1:

[0796] The server receives the user's question entered via an information terminal. The input data is a question, specifically text information such as "Tell me an easy pasta recipe." The received question is passed directly to the next analysis step.

[0797] Step 2:

[0798] The server analyzes the received question using natural language processing (NLP) techniques. Specifically, it uses Python's TextBlob and NLTK libraries to parse the question and extract keywords. The input is the question, and the output is a list of keywords such as "easy," "pasta," and "recipe."

[0799] Step 3:

[0800] The server retrieves the user's search history and evaluation data from the user's past database. At the same time, it also retrieves the latest related information through the API of an external information service (e.g., a recipe provision service). The input is a list of keywords, and the output is a list of relevant information.

[0801] Step 4:

[0802] The emotion engine captures the user's facial expressions in real time through the information terminal's camera and analyzes their emotional state using facial expression recognition technology (such as the OpenCV library).The input is real-time facial image data, and the output is emotional states such as "happiness," "sadness," and "tiredness."

[0803] Step 5:

[0804] The server combines the user's emotional state, past data, and external information to generate the optimal information response. Using a generative AI model, it combines these data and creates a response with wording appropriate to the user's emotions. The input is the emotional state and a list of information, and the output is the text of the optimal response.

[0805] Step 6:

[0806] The server sends the generated answer to the information terminal and displays it on the user's smart glasses or other information terminal. The user can visually confirm the displayed answer. The input is the text of the best answer, and the output is the answer displayed on the information terminal.

[0807] Step 7:

[0808] The user inputs feedback on the provided information through the information terminal. The feedback includes specific evaluations such as "This recipe was very useful." The input is the user's feedback text, and the output is the feedback data.

[0809] Step 8:

[0810] The server stores the collected feedback in a database and uses it as training data for the generative model and emotion engine. This improves the accuracy of answers from the next time onwards. The input is the feedback data, and the output is the model reflected in the training.

[0811] 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 a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0812] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0813] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0814] [Third embodiment]

[0815] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0816] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0817] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0818] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0819] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0820] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0821] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0822] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0823] The specific processing program 56 is an example of a "program" according to the technology of the present 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.

[0824] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0825] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. 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 process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0826] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. 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."

[0827] The present invention is a system that uses an information terminal, a server, and a database, and provides optimal answers to questions asked by users in information searches. A specific embodiment of this system will be described below.

[0828] User registration and information collection

[0829] First, a user accesses the service using an information terminal and enters basic information such as name, age, occupation, hobbies and interests into the registration form. The terminal sends this information to the server, which then stores it in a database.

[0830] Entering and parsing questions

[0831] The user inputs a specific question, such as "Tell me an easy pasta recipe," into the device and sends it. This input data is sent to the server. The server analyzes the received question using natural language processing (NLP) technology, extracts keywords such as "easy," "pasta," and "recipe," and understands the user's intent.

[0832] Integrating historical data and external information

[0833] Based on the analysis results, the server retrieves the user's past data from the database. For example, if the user has previously searched for Italian food, that information will be used as a reference. The server also connects with external information services (such as cooking recipe providers) via API to retrieve the latest recipe information.

[0834] Generating and providing optimal information

[0835] The server combines the user's hobby and interest data with external information to generate the most appropriate answer. In this case, it identifies "an easy recipe for spaghetti with tomatoes and basil" and generates an answer in natural language format. The generated answer is sent to the device and displayed to the user.

[0836] Gathering feedback and learning

[0837] The user reviews the provided information and sends feedback such as "This recipe was very helpful" via their device. This feedback is received by the server and stored in a database. The server uses this feedback to continuously train the generative model, improving the accuracy of answers from the next time onwards.

[0838] Specific examples

[0839] For example, suppose a user inputs a question about "easy pasta recipes." The server analyzes this question and references the user's past search history and hobbies and preferences from a database. If the user has previously liked "Italian food," the server connects with an external information service to retrieve the latest recipe for "spaghetti with tomato and basil." The server formats this information in natural language and sends it to the device. The device displays it to the user as "an easy recipe for spaghetti with tomato and basil." If the user then rates this recipe as useful, that feedback is sent back to the server and used as training data for the model.

[0840] In this way, the present invention is a system that continues to provide optimal information in response to user questions, thereby improving the efficiency of information acquisition and user satisfaction.

[0841] The processing flow will be explained below.

[0842] Step 1:

[0843] The user accesses the service using an information terminal and enters basic information such as name, age, occupation, hobbies and interests into the registration form.

[0844] The terminal transmits this information to the server.

[0845] Step 2:

[0846] The server receives the basic information sent by the user and stores it in a database.

[0847] The server also generates a unique user ID for each user and creates a corresponding record.

[0848] Step 3:

[0849] The user enters a specific question, such as "Tell me an easy pasta recipe," into the terminal and sends it.

[0850] The terminal transmits the entered question to the server.

[0851] Step 4:

[0852] The server analyzes the received question using natural language processing (NLP) techniques.

[0853] The server extracts key keywords from the question ("easy," "pasta," "recipe") and understands the user's intent.

[0854] Step 5:

[0855] The server retrieves the user's past data from the database.

[0856] For example, if the user has previously searched for "Italian cuisine," that information can be used as a reference.

[0857] Step 6:

[0858] The server initiates API integration with an external information service to obtain the latest recipe information.

[0859] For example, recipe data for "spaghetti with tomato and basil" is received from an external recipe service.

[0860] Step 7:

[0861] The server integrates the acquired external information with the user's past data to generate the optimal answer.

[0862] For example, create an "easy tomato and basil spaghetti recipe" in natural language format.

[0863] Step 8:

[0864] The server sends the generated response to the terminal.

[0865] The device receives the response and displays it to the user as an "easy tomato and basil spaghetti recipe."

[0866] Step 9:

[0867] The user reviews the information provided and sends feedback via the device, such as "This recipe was helpful."

[0868] The terminal transmits the feedback information to the server.

[0869] Step 10:

[0870] The server stores the user feedback in a database.

[0871] The server uses this feedback to train a generative model and uses it to improve the accuracy of answers from next time onwards.

[0872] Example 1

[0873] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0874] Conventional information provision systems have difficulty providing accurate answers to user questions, and in order to improve user satisfaction, it is necessary to effectively integrate the user's past behavioral history and external information. Furthermore, due to a lack of means to collect user feedback and improve the accuracy of the system, continuous learning is not possible, raising concerns that the quality of the information provided may decline.

[0875] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0876] In this invention, the server includes means for receiving a user's question input via an information terminal, means for analyzing the received question using natural language processing to understand the user's intention, means for integrating the user's past data and information acquired from an external database to generate optimal information, means for transmitting the generated information to the information terminal and displaying it, and means for collecting user feedback input via the information terminal and training the generation AI model. This makes it possible to continue providing optimal information in response to a user's question, thereby improving the efficiency of information acquisition and user satisfaction.

[0877] An "information terminal" is a device that allows a user to access the system via the Internet and input and display data, and includes smartphones, personal computers, etc.

[0878] A "user question" is a textual inquiry that a user inputs into the system using an information terminal, requesting specific information.

[0879] "Natural language processing" is a technology that analyzes input text data in a way that is easy for a computer to understand, and grasps its meaning and intent.

[0880] "User intent" refers to the specific information or answer that the user wants to obtain through the question.

[0881] "Past data" refers to information such as a user's past behavioral history, search history, and interests that is stored in a database.

[0882] An "external database" is a data source that exists outside the system, and includes various information providing services that can be accessed via API.

[0883] "Optimal information" refers to the most appropriate and useful answer to a user's question, and is generated by integrating information from past data and external databases.

[0884] A "generative AI model" is an algorithm that continuously learns from data and feedback provided by users, improving the accuracy of its answers.

[0885] "Feedback" refers to the evaluations and opinions that users give in response to the information provided, and is used to help the system learn.

[0886] "Display means" refers to a function for visually presenting optimal information on an information terminal, and includes a screen or display.

[0887] "Integration means" is a process that combines the user's past data and information obtained from external databases to generate information that is optimal for the user.

[0888] The present invention is a system that uses an information terminal, a server, and a database, and provides optimal answers to questions asked by users in information searches. An embodiment of this system will be described in detail below.

[0889] First, the user accesses the system's website using an information terminal (smartphone or PC). The user enters basic information such as name, age, occupation, hobbies and interests into the registration form on the website. When the user presses the "Submit" button, the information terminal sends the entered data to the server using the HTTPS protocol. The server stores the received data in a database server (e.g., MySQL, PostgreSQL).

[0890] Next, the user enters a specific question. For example, "Tell me an easy pasta recipe" and clicks the submit button. The input data is again sent to the server via HTTPS protocol. The server analyzes the received question using natural language processing (NLP) technology. At this time, NLP libraries such as spaCy and NLTK are used to extract keywords such as "easy," "pasta," and "recipe." This allows the server to understand the user's intent.

[0891] The server then retrieves the user's past behavioral history and preference information from the database. This information is obtained using an SQL query. The server also connects to external databases (e.g., recipe service) via API to retrieve the latest recipe information. The acquired internal data is then integrated with external data to generate the optimal recipe information for the user. A generative AI model is used to generate a natural language answer (e.g., "Recipe for spaghetti with tomato and basil").

[0892] The generated answer is sent from the server to the device, which displays it to the user. The user checks the displayed information, enters feedback (e.g., "This recipe was very helpful"), and sends it to the server via the device. The server stores the feedback in a database and uses this information to continuously train the generative AI model. This improves the accuracy of answers in future queries.

[0893] Specific examples

[0894] For example, if a user inputs a question such as "Tell me an easy pasta recipe," the system operates as follows: The server first analyzes the question using NLP technology to extract keywords. It then combines the user's past data with the latest recipe information obtained from an external API to generate an "easy tomato and basil spaghetti recipe." This generated information is sent to the device and displayed to the user. Finally, the user sends feedback to the server on whether the provided recipe was helpful, and this feedback is used to train the generative AI model.

[0895] In this way, the present invention is a system that continues to provide optimal information in response to user questions, thereby improving the efficiency of information acquisition and user satisfaction.

[0896] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0897] Step 1:

[0898] A user accesses the service and enters registration information.

[0899] Specific operation: The user uses an information terminal (smartphone or PC) to access the system's website.

[0900] Input: A user fills in a registration form on a website with information such as name, age, occupation, hobbies and interests.

[0901] Data processing: The device temporarily stores the entered information.

[0902] Output: The persisted data is ready to be sent to the server in the next step.

[0903] Step 2:

[0904] The terminal transmits the input user information to the server.

[0905] What happens: The user clicks a button to submit the registration form.

[0906] Input: Stored user information (name, age, occupation, hobbies and interests).

[0907] Data calculation: The information terminal uses the HTTPS protocol to send the input information to the server.

[0908] Output: The server receives the user information.

[0909] Step 3:

[0910] The server stores the received user information in a database.

[0911] Specific operation: The server analyzes the received registration data and stores it in a database.

[0912] Input: User registration information.

[0913] Data manipulation: Creating new records in the database using SQL queries.

[0914] Output: The saved user information.

[0915] Step 4:

[0916] The user types a question and submits it.

[0917] Specific operation: The user enters a question on the information terminal and clicks the send button.

[0918] Input: Question text (e.g. "What's an easy pasta recipe?").

[0919] Data calculation: The information terminal sends the entered query data to the server.

[0920] Output: The server receives the query data.

[0921] Step 5:

[0922] The server analyzes the question using natural language processing.

[0923] What happens: The server parses the question using an NLP library (e.g., spaCy, NLTK).

[0924] Input: The question text submitted by the user.

[0925] Data processing: Extract keywords from the question text to understand the user's intent.

[0926] Output: Extracted keywords and user intent.

[0927] Step 6:

[0928] The server retrieves the user's past data from the database.

[0929] Specific operation: The server retrieves past behavioral history and interest data from the database based on the user's ID.

[0930] Input: User ID and parsed question information.

[0931] Data operations: Retrieving relevant data from a database using SQL queries.

[0932] Output: User's historical data.

[0933] Step 7:

[0934] The server retrieves the latest information from an external database.

[0935] Specific operation: The server calls an external API to obtain the latest recipe information.

[0936] Input: Extracted keywords (e.g., pasta, recipe).

[0937] Data Calculation: Sending requests to external APIs and parsing the responses.

[0938] Output: Latest recipe information.

[0939] Step 8:

[0940] The server generates the most appropriate information and provides it to the user.

[0941] Specific operation: The server integrates the user's past data with external data and generates answers in natural language format using a generative AI model.

[0942] Input: User's past data, latest recipe information, parsed question information.

[0943] Data processing: Generate optimal answers from integrated data.

[0944] Output: A natural language answer (e.g., "Easy tomato and basil spaghetti recipe").

[0945] Step 9:

[0946] The terminal displays the generated answer to the user.

[0947] Specific operation: The device displays the answer received from the server to the user.

[0948] Input: The answer sent by the server in natural language format.

[0949] Data processing: Formatting text for visual display on the device.

[0950] Output: The answer displayed to the user.

[0951] Step 10:

[0952] The user enters and submits feedback.

[0953] Specific actions: The user enters feedback on the device and clicks the submit button.

[0954] Input: Feedback text (e.g., "This recipe was very helpful").

[0955] Data calculation: The information terminal sends the input feedback to the server.

[0956] Output: The server receives the feedback data.

[0957] Step 11:

[0958] The server stores the feedback in a database and trains a generative AI model.

[0959] Specific operation: The server stores the feedback data in a database and updates the generative AI model.

[0960] Input: Received feedback data.

[0961] Data processing: Store the feedback in a database and use it as new data for training the AI ​​model.

[0962] Output: An updated generative AI model.

[0963] This allows the system to continue providing optimal information in response to user questions, improving the efficiency of information acquisition and user satisfaction.

[0964] (Application example 1)

[0965] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0966] Conventional product recommendation systems in virtual stores were unable to fully reflect the individual preferences and past purchase history of users, making it difficult to provide optimal products. Furthermore, there was no mechanism for effectively collecting and utilizing user feedback, making it difficult to improve the quality of service.

[0967] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0968] In this invention, the server includes means for receiving a user question input via an information terminal, means for analyzing the received question using natural language processing to understand the user's intention, means for integrating the user's past data and information acquired from external information services to recommend optimal products based on the user's preferences, means for displaying the generated product information on the information terminal, and means for collecting user feedback input via the information terminal and training a generative model. This makes it possible to propose optimal products based on the individual preferences of users and improve the quality of services by utilizing feedback.

[0969] An "information terminal" refers to an electronic device, such as a smartphone or tablet, that allows a user to input questions and receive answers from a server.

[0970] "Natural language processing" refers to the technology that allows computers to understand and analyze human language, allowing them to understand the intent of users' questions.

[0971] "User's past data" refers to information such as the user's past search history, purchase history, and feedback.

[0972] "External Information Services" refers to information or databases provided by third parties, such as product catalogs or review sites.

[0973] "Means for recommending optimal products" refers to algorithms or programs that integrate the user's past data with information obtained from external information services and select the most suitable product based on the user's preferences.

[0974] The "generated product information" refers to detailed product information selected by the optimal product recommendation means, including, for example, the product name, price, and features.

[0975] "Feedback" refers to a user's evaluation or opinion of the information provided or the recommended products.

[0976] A "generative model" refers to a machine learning model that generates optimal information based on user questions and feedback.

[0977] The present invention provides a personalized shopping assistant system for a virtual store that recommends optimal products based on a user's preferences and past purchase history. Specific embodiments of this system are described below.

[0978] Hardware and software used

[0979] This system consists of information terminals (e.g., smartphones, tablets, smart glasses) accessed by users and a server that processes data. The software includes the following elements:

[0980] 1. Natural Language Processing (NLP) engine: used to analyze the user's question and extract keywords, for example using Python and the transformers library.

[0981] 2. Database: Stores user registration information, past data, and product information obtained from external information services. For example, a database system such as MySQL is used.

[0982] 3. API integration module: Connects to external information services (e.g., product catalogs, review sites, etc.) via API to obtain the latest product information.

[0983] 4. Generative AI model: A model for recommending optimal products based on user preference data and external information. As an example, we will use a machine learning model.

[0984] System Operation

[0985] The server operates as follows:

[0986] 1. Accepting a user question: The user's question (e.g., "Please recommend a new smartphone") is received via the information terminal.

[0987] 2. Natural language processing of questions: The content of the question is analyzed using natural language processing technology to extract important keywords.

[0988] 3. Information integration: Integrates the user's past data (e.g., past purchase history) and the latest product information obtained from external information services to select the optimal product based on the user's preferences.

[0989] 4. Product information generation and display: The selected product information is formatted in natural language and displayed on the information terminal.

[0990] 5. Feedback collection: Collect user feedback entered via the information terminal and continuously train the generative AI model.

[0991] Specific examples

[0992] For example, a user might use an information terminal to ask, "Please recommend a new smartphone." This input is sent to the server and analyzed by a natural language processing engine. As a result of the analysis, important keywords such as "smartphone" and "recommendation" are extracted. The server then retrieves the user's past purchase history and preference data from a database, and obtains the latest smartphone information through the API of an external information service.

[0993] The generative AI model integrates this information and recommends the most suitable smartphone for the user (e.g., "latest model smartphone"). This information is formatted in natural language and displayed on the information terminal. When the user reviews the recommendation and sends feedback saying "this suggestion was helpful," that feedback is stored on the server and used as training data for the generative AI model.

[0994] Examples of prompt statements

[0995] User: Recommend me a new smartphone.

[0996] System: Based on your past search and purchase history, we estimate that a new smartphone model would be suitable. We recommend a 'new smartphone model' featuring a high-performance camera and a long-lasting battery. The price is 50,000 yen. How about this product?

[0997] As described above, the present invention makes it possible to provide appropriate and personalized product suggestions based on the individual preferences and past data of the user.

[0998] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0999] Step 1:

[1000] The user enters a question.

[1001] The user uses an information terminal (e.g., a smartphone) to input a question. Specifically, the user inputs text such as "Please recommend a new smartphone" and presses the send button. The input question (text data) is sent to the server.

[1002] Step 2:

[1003] The server accepts the query.

[1004] The server receives questions sent by users, stores them as text data, and then passes the text data to a natural language processing engine.

[1005] Step 3:

[1006] The server performs natural language processing.

[1007] The server uses a natural language processing engine (e.g., the transformers library) to analyze the received text data, split the question, and extract important keywords (e.g., "smartphone," "recommended"). The results with the extracted keywords are then returned.

[1008] Step 4:

[1009] The server obtains the user's past data and external information.

[1010] The server retrieves the user's past data (past purchase history, search history, etc.) from the database based on the keywords. At the same time, it calls an external information service API to retrieve the latest product information related to the keywords. The retrieved data is then integrated.

[1011] Step 5:

[1012] The server recommends the best products.

[1013] The server uses the generative AI model to analyze and integrate the user's past data and product data obtained from external information, thereby selecting the product (e.g., the latest model smartphone) that best suits the user's preferences.

[1014] Step 6:

[1015] The server generates and displays product information.

[1016] The server formats the selected product details (product name, price, features, etc.) into a natural language format. This generated product information is sent to the information terminal, which then displays the information to the user.

[1017] Step 7:

[1018] The user enters feedback.

[1019] The user checks the presented product information and inputs feedback such as "This suggestion was helpful." The input feedback is then sent back to the server.

[1020] Step 8:

[1021] The server collects feedback and learns.

[1022] The server receives feedback sent by the user, stores it in a database, and uses it as training data for the generative AI model, thereby improving the accuracy of recommendations in the future.

[1023] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1024] The present invention is a system that uses an information terminal, a server, a database, and an emotion engine to provide optimal answers to questions asked by users in information searches, and further adjusts the content of the answers taking into account the user's emotions. A specific embodiment of this system is described below.

[1025] User registration and information collection

[1026] First, a user accesses the service using an information terminal and enters basic information such as name, age, occupation, hobbies and interests into the registration form. The terminal sends this information to the server, which stores it in a database. The emotion engine also performs initial settings to understand the user's basic emotional patterns.

[1027] Entering and parsing questions

[1028] The user inputs a specific question, such as "Tell me an easy pasta recipe," into the device and sends it. This input data is sent to the server. The server analyzes the received question using natural language processing (NLP) technology, extracts keywords such as "easy," "pasta," and "recipe," and understands the user's intent.

[1029] Acquiring user emotion data

[1030] When a question is entered, the emotion engine collects the user's emotion data and uses, for example, text analysis and facial expression recognition technology to understand the user's current emotional state (such as joy, sadness, or anger).

[1031] Integrating historical data and external information

[1032] Based on the analysis results, the server retrieves the user's past data from the database. For example, if the user has previously searched for Italian food, that information will be used as a reference. The server also connects with external information services (such as cooking recipe providers) via API to retrieve the latest recipe information.

[1033] Optimal information generation and display

[1034] The server combines the acquired external information, the user's past data, and emotional data to generate the optimal answer. For example, it identifies an "easy tomato and basil spaghetti recipe" and generates an answer tailored to the user's emotional state, such as "You seem to be in a low mood today. Try this easy pasta recipe to cheer you up." The generated answer is sent to the device and displayed to the user.

[1035] Gathering feedback and learning

[1036] The user reviews the provided information and sends feedback via their device, such as "This recipe was very helpful." This feedback is received by the server and stored in a database. The server uses this feedback to continuously train the generative model and emotion engine, and uses it to improve the accuracy of answers in future searches.

[1037] Specific examples

[1038] For example, if a user enters "easy pasta recipes," the server analyzes the question and references the user's past search history and hobbies and preferences from a database. If the emotion engine recognizes the emotion "a little tired" from the user's input, the server connects with an external information service to retrieve a recipe for "spaghetti with tomato and basil." Based on this, the server makes adjustments such as "You seem a little tired today. Try refreshing yourself with this easy pasta recipe," and sends and displays it on the device. If the user rates the recipe as "helpful," that feedback is sent back to the server and used as learning data for the model. The emotion engine also improves its accuracy based on feedback.

[1039] In this way, the present invention is a system that provides optimal information in response to a user's question while enabling flexible responses according to the user's emotional state, thereby improving the efficiency of information acquisition and user satisfaction.

[1040] The processing flow will be explained below.

[1041] Step 1:

[1042] The user accesses the service using an information terminal and enters basic information such as name, age, occupation, hobbies and interests into the registration form.

[1043] The terminal transmits this information to the server.

[1044] Step 2:

[1045] The server receives the basic information sent by the user and stores it in a database.

[1046] The server also generates a unique user ID for each user and creates a corresponding record.

[1047] Step 3:

[1048] The user enters a specific question, such as "Tell me an easy pasta recipe," into the terminal and sends it.

[1049] The terminal transmits the entered question to the server.

[1050] Step 4:

[1051] The emotion engine analyzes the user's input text and recognizes emotional states (e.g., joy, sadness, fatigue, etc.).

[1052] The server acquires the emotion data and processes it together with the question data.

[1053] Step 5:

[1054] The server analyzes the received question using natural language processing (NLP) techniques.

[1055] The server extracts key keywords from the question ("easy," "pasta," "recipe") and understands the user's intent.

[1056] Step 6:

[1057] The server retrieves the user's past data from the database.

[1058] For example, if the user has previously searched for "Italian cuisine," that information can be used as a reference.

[1059] Step 7:

[1060] The server initiates API integration with an external information service to obtain the latest recipe information.

[1061] For example, recipe data for "spaghetti with tomato and basil" is received from an external recipe service.

[1062] Step 8:

[1063] The server integrates the acquired external information, past data, and emotional data to generate the optimal answer.

[1064] For example, if the user's emotional state is recognized as "tired," the system adjusts the response by saying, "You seem a little tired today. Try refreshing yourself with this simple pasta recipe."

[1065] Step 9:

[1066] The server sends the generated response to the terminal.

[1067] The device receives the response and displays it to the user as an "easy tomato and basil spaghetti recipe."

[1068] Step 10:

[1069] The user reviews the information provided and sends feedback via the device, such as "This recipe was very helpful."

[1070] The terminal transmits the feedback information to the server.

[1071] Step 11:

[1072] The server stores the user feedback in a database.

[1073] The server uses this feedback to continuously train the generative model and emotion engine, and uses it to improve the accuracy of answers from next time onwards.

[1074] Example 2

[1075] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1076] Conventional information retrieval systems provide answers only to questions entered by the user and do not take the user's emotional state into account, which means that the answers may not necessarily meet the user's needs. Furthermore, the means for collecting feedback are limited, which leaves the user experience unsatisfactory. Therefore, the present invention aims to provide optimal information that takes into account the user's emotional state and continuously improve the system's accuracy by utilizing feedback.

[1077] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1078] In this invention, the server includes means for receiving a user question input via an information terminal, means for analyzing the received question using natural language processing to understand the user's intention, means for analyzing the user's current emotional state, means for integrating the user's past data and information obtained from external information services to generate optimal information, means for adjusting the generated information according to the user's emotional state and displaying it on the information terminal, and means for collecting user feedback input via the information terminal and training the generative model and emotion engine. This makes it possible to provide optimal information in response to the user's question while also responding flexibly to the user's emotional state.

[1079] An "information terminal" is a device used by a user to input a question, and includes a computer, a smartphone, a tablet, and the like.

[1080] A "server" is a central system that receives data from information terminals via a network and analyzes and processes the data.

[1081] "Natural Language Processing (NLP)" is a technology that allows computers to understand and analyze human language and grasp the user's intent.

[1082] An "emotion engine" is an algorithm and software that analyzes and identifies a user's emotional state from their text and facial expressions.

[1083] A "database" is a system for managing and storing user registration information and past data.

[1084] "External information services" are third-party services that provide information via the Internet.

[1085] A "generative model" is an algorithm and software that generates optimal information in response to a user's question.

[1086] "Feedback" refers to the evaluation or opinion that a user enters regarding the information provided.

[1087] "Synthesis" is the process of combining data obtained from multiple different sources into a single, consistent piece of information.

[1088] User registration and information collection

[1089] First, the user accesses the service using an information terminal (smartphone, computer, tablet, etc.) and enters basic information such as name, age, occupation, hobbies and interests into the registration form. This information is then sent from the terminal to the server. The server stores the received information in a database and performs initial settings to understand the user's basic emotional patterns using an emotion engine. At this point, the server stores the user's individual information in a database, making it possible to provide individualized information.

[1090] Entering and parsing questions

[1091] The user inputs a specific question into an information terminal. For example, they can type in the text "Tell me an easy pasta recipe" and send it. This input data is sent from the terminal to the server. The server analyzes the received question using natural language processing (NLP) technology, extracts keywords such as "easy," "pasta," and "recipe," and understands the user's intent. This analysis can be performed using open-source natural language processing libraries or cloud-based NLP services.

[1092] Acquiring user emotion data

[1093] When a user enters a question, the emotion engine collects the user's emotion data. It can use a camera-equipped information device to understand the user's current emotional state (e.g., joy, sadness, anger) using text analysis and facial expression recognition technology. For emotion recognition, it can use, for example, an open-source emotion analysis library or a cloud-based emotion recognition API.

[1094] Integrating historical data and external information

[1095] Based on the analysis results, the server retrieves the user's past data (e.g., past search history, hobbies, and preferences) from the database. It also connects with external information services (e.g., cooking recipe services) via API to retrieve the latest recipe information. During this process, the server communicates with external services using an enterprise DBMS, cloud database service, or REST API.

[1096] Optimal information generation and display

[1097] The server combines the acquired external information, the user's past data, and emotional data to generate the optimal answer. For example, it identifies an "easy tomato and basil spaghetti recipe" and generates an answer tailored to the user's emotional state, such as "You seem to be in a low mood today. Try this easy pasta recipe to cheer you up." The generated answer is sent to the device and displayed to the user.

[1098] Gathering feedback and learning

[1099] The user reviews the provided information and sends feedback via their device, such as "This recipe was very helpful." This feedback is received by the server and stored in a database. The server uses this feedback to continuously train the generative model and emotion engine, and uses it to improve the accuracy of answers in the future.

[1100] Specific examples

[1101] For example, if a user enters "easy pasta recipes," the server analyzes the question and references the user's past search history and hobbies and preferences from a database. If the emotion engine recognizes the emotion "a little tired" from the user's input, the server connects with an external information service to retrieve a recipe for "spaghetti with tomato and basil." Based on this, the server makes adjustments such as "You seem a little tired today. Try refreshing yourself with this easy pasta recipe," and sends and displays it on the device. If the user rates the recipe as "helpful," that feedback is sent back to the server and used as learning data for the model. The emotion engine also improves its accuracy based on feedback.

[1102] Example prompts to input to the generative AI model

[1103] "When a user is looking for an easy pasta recipe, give them the perfect recipe based on their emotional state and why. For example, 'You seem a little tired today. Try refreshing yourself with this easy pasta recipe.'"

[1104] In this way, the present invention is a system that provides optimal information in response to a user's question while enabling flexible responses according to the user's emotional state, thereby improving the efficiency of information acquisition and user satisfaction.

[1105] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1106] Step 1:

[1107] User registration and information collection

[1108] Input: The user uses an information terminal to input basic information such as name, age, occupation, hobbies and interests.

[1109] Processing: The device sends the input information to the server. The server stores the received information in a database and sets the user's initial emotional pattern using the emotion engine.

[1110] Output: User information is saved in the database and the emotion engine completes its initialization.

[1111] How it works: The user accesses the registration page in a web browser or mobile app, enters the required information in the form, and then presses the "Register" button. The device sends this data to the server as an HTTP request. The server saves the information in a database and calls the emotion recognition API to set the initial emotion pattern.

[1112] Step 2:

[1113] Entering and parsing questions

[1114] Input: The user types a specific question into the device and sends it. For example, the text "Tell me a simple pasta recipe."

[1115] Processing: The device sends the input data to the server, which uses natural language processing (NLP) techniques to analyze the question and extract keywords.

[1116] Output: Keyword data is generated to understand the intent of the question.

[1117] Specific operation: The user enters a question into the text input form and presses the "Submit" button. The device sends the text data as an HTTP POST request to the server. The server uses the NLP module to extract keywords such as "easy," "pasta," and "recipe."

[1118] Step 3:

[1119] Acquiring user emotion data

[1120] Input: User text and facial expression data when entering a question.

[1121] Processing: The server uses the emotion engine to analyze the user's current emotional state (happiness, sadness, anger, etc.).

[1122] Output: Data of the current user's emotional state.

[1123] Specific operation: The server inputs text data into the emotion engine and calls the emotion recognition API. The emotion engine analyzes emotions such as "joy," "sadness," and "anger." The analysis results are recorded for use in the next step.

[1124] Step 4:

[1125] Integrating historical data and external information

[1126] Input: Question analysis results, user history, data from external information services.

[1127] Processing: The server retrieves the user's past data from the database, connects to external information services via API, and retrieves the latest information. This data is then integrated.

[1128] Output: A consolidated dataset.

[1129] What it does: The server queries the database using the user ID to retrieve past search history and hobbies and preferences. It sends a request to an external API to retrieve the latest recipe information, for example, from a cooking recipe service. It then integrates all the data and creates a dataset to generate optimal information.

[1130] Step 5:

[1131] Optimal information generation and display

[1132] Input: Unified dataset, user emotional state.

[1133] Processing: The server generates the optimal answer based on the integrated data, adjusts the content according to the user's emotional state, and sends the generated answer to the device.

[1134] Output: The adjusted answer is displayed on the terminal.

[1135] How it works: The server uses its internal algorithm to identify an "easy recipe for spaghetti with tomato and basil." The emotion engine adds a comment based on the user's emotional state. The server sends the answer in JSON format to the device, which then displays it on the screen.

[1136] Step 6:

[1137] Gathering feedback and learning

[1138] Input: User feedback data.

[1139] Processing: The device sends feedback data to the server. The server receives the feedback and stores it in a database. This information is used to train the generative model and emotion engine.

[1140] Output: A trained generative model and an emotion engine.

[1141] Specific operation: The user checks the provided information, enters a rating, and presses the "Submit" button. The device sends the feedback data to the server. The server stores the feedback in a database, and the learning module uses it to update the parameters of the generative model and emotion engine.

[1142] (Application example 2)

[1143] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1144] In modern brick-and-mortar stores, improving customer satisfaction and providing personalized service is a major challenge. It is particularly difficult to grasp a customer's emotional state in real time and provide appropriate product recommendations and services accordingly. Furthermore, conventional information provision systems using information terminals lack the ability to respond flexibly and consider the user's emotions. Furthermore, they lack an effective mechanism for accumulating feedback and reflecting it in future customer service.

[1145] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1146] In this invention, the server includes means for receiving a user's question input via an information terminal, means for analyzing the received question using natural language processing and grasping the user's intention, means for integrating the user's past data and information acquired from an external information service to generate optimal information, means for displaying the generated information on the information terminal, means for collecting user feedback input via the information terminal and training a generative model, means for recognizing the user's facial expression and grasping their emotional state, means for generating and displaying information taking the user's emotional state into consideration, and means for visually providing the displayed information via smart glasses. This enables personalized responses adapted to customer emotions and improves customer satisfaction in real time.

[1147] An "information terminal" is an electronic device that allows a user to input or receive information.

[1148] The "receiving means" is a device or program that has the function of receiving data transmitted from an information terminal.

[1149] "Natural language processing" is a technology in which a computer analyzes human language and understands its meaning.

[1150] "User intent" is the purpose or intention behind the question or request entered by the user.

[1151] "Past data" refers to data based on information previously generated by the user and their behavioral history.

[1152] "External information services" are additional sources of information provided by external servers or databases.

[1153] "Best information" is the answer or suggestion that best suits the user's question or requirements.

[1154] "Generating means" refers to a device or program that has the function of combining information based on collected data and creating the necessary information.

[1155] The "display means" is a device or program that has the function of presenting the generated information in a form that can be visually recognized by the user.

[1156] "Feedback" refers to a user's evaluation or opinion of the information or service provided.

[1157] The "collection means" is a device or program that has the function of collecting feedback from users.

[1158] A "generative model" is a computational model that generates information based on large amounts of data.

[1159] A "training means" is a device or program that has the function of improving the generative model based on collected feedback.

[1160] "Means for recognizing facial expressions" refers to a device or program that has the function of recognizing and analyzing a user's facial expressions using a camera or sensor.

[1161] "Emotional state" refers to a user's current feelings or mood.

[1162] A "means for taking into account" is a device or program that has the ability to adjust information based on the user's emotional state.

[1163] "Smart glasses" are a wearable eyeglass-type device that has the function of visually displaying information.

[1164] The present invention is a system that uses an information terminal, a server, a database, and an emotion engine to provide optimal answers to user questions. This system adjusts the content of the answers taking into account the user's emotions, thereby providing more appropriate information.

[1165] System Overview

[1166] The system consists of the following main components:

[1167] Information terminal: A device (e.g., smartphone, tablet, PC, etc.) through which a user enters questions and receives answers.

[1168] Server: The primary computer system for information analysis, data integration, and answer generation.

[1169] Database: Storage for saving user history and registration information.

[1170] Emotion engine: Software for recognizing a user's facial expressions and understanding their emotional state.

[1171] What the program does

[1172] The server receives the user's question sent from the information terminal and analyzes it using natural language processing (NLP) technology. Specifically, it uses the Python libraries TextBlob and NLTK. As a result of the analysis, keywords are extracted from the question to understand the user's intent.

[1173] The system then integrates the user's past data (e.g., previously searched information and ratings) with information obtained from external information services (e.g., the latest recipe service or product information service). This process uses API integration technology, for example, to obtain data from external information services via RESTful APIs.

[1174] Furthermore, an emotion engine (e.g., an OpenCV library for facial expression recognition) is used to analyze the user's facial expressions captured by the device's camera and understand their emotional state. Emotional data can be used to adjust the content of responses.

[1175] For example, if a user asks "What's an easy pasta recipe?", the system generates an answer in the following order:

[1176] 1. The server receives the question and uses NLP technology to extract the keywords "easy," "pasta," and "recipe."

[1177] 2. The server references the user's past search history in the database and uses data on Italian cuisine that the user previously searched for as a reference.

[1178] 3. The emotion engine analyzes facial expressions and recognizes the user's emotional state as "a little tired."

[1179] 4. The server retrieves the latest "Tomato and Basil Spaghetti" recipe from an external information service and generates a message saying, "You seem a little tired today. Try refreshing yourself with this easy pasta recipe."

[1180] 5. The generated answer is sent to the information terminal and visually displayed to the user.

[1181] Specific examples

[1182] Here is an example where a user enters "easy pasta recipes." The server analyzes the question and references the user's past search history from the database. If the emotion engine recognizes the emotion "a little tired" from the user's input, the server connects with an external information service and retrieves a recipe for "spaghetti with tomato and basil." Based on this, the server makes adjustments such as "You seem a little tired today. Try refreshing yourself with this easy pasta recipe," generates an answer, and sends it to the information terminal for display.

[1183] Prompt Sentence Examples

[1184] 1. "Tell me an easy pasta recipe."

[1185] 2. "Do you have any drinks to cheer me up?"

[1186] 3. "I'm looking for some new shoes. What do you recommend?"

[1187] 4. "Do you have any teas that can help relieve stress?"

[1188] In this way, the present invention is a system that enables flexible responses according to the user's emotions, thereby realizing more appropriate information provision and improving customer satisfaction.

[1189] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1190] Step 1:

[1191] The server receives the user's question entered via an information terminal. The input data is a question, specifically text information such as "Tell me an easy pasta recipe." The received question is passed directly to the next analysis step.

[1192] Step 2:

[1193] The server analyzes the received question using natural language processing (NLP) techniques. Specifically, it uses Python's TextBlob and NLTK libraries to parse the question and extract keywords. The input is the question, and the output is a list of keywords such as "easy," "pasta," and "recipe."

[1194] Step 3:

[1195] The server retrieves the user's search history and evaluation data from the user's past database. At the same time, it also retrieves the latest related information through the API of an external information service (e.g., a recipe provision service). The input is a list of keywords, and the output is a list of relevant information.

[1196] Step 4:

[1197] The emotion engine captures the user's facial expressions in real time through the information terminal's camera and analyzes their emotional state using facial expression recognition technology (such as the OpenCV library).The input is real-time facial image data, and the output is emotional states such as "happiness," "sadness," and "tiredness."

[1198] Step 5:

[1199] The server combines the user's emotional state, past data, and external information to generate the optimal information response. Using a generative AI model, it combines these data and creates a response with wording appropriate to the user's emotions. The input is the emotional state and a list of information, and the output is the text of the optimal response.

[1200] Step 6:

[1201] The server sends the generated answer to the information terminal and displays it on the user's smart glasses or other information terminal. The user can visually confirm the displayed answer. The input is the text of the best answer, and the output is the answer displayed on the information terminal.

[1202] Step 7:

[1203] The user inputs feedback on the provided information through the information terminal. The feedback includes specific evaluations such as "This recipe was very useful." The input is the user's feedback text, and the output is the feedback data.

[1204] Step 8:

[1205] The server stores the collected feedback in a database and uses it as training data for the generative model and emotion engine. This improves the accuracy of answers from the next time onwards. The input is the feedback data, and the output is the model reflected in the training.

[1206] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1207] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1208] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1209] [Fourth embodiment]

[1210] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1211] 7, a 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.

[1212] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1213] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1214] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1215] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1216] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1217] The control object 443 includes a display device, LEDs in the eyes, and motors for driving 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 emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1218] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1219] The specific processing program 56 is an example of a "program" according to the technology of the present 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.

[1220] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1221] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. 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 process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1222] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1223] The present invention is a system that uses an information terminal, a server, and a database, and provides optimal answers to questions asked by users in information searches. A specific embodiment of this system will be described below.

[1224] User registration and information collection

[1225] First, a user accesses the service using an information terminal and enters basic information such as name, age, occupation, hobbies and interests into the registration form. The terminal sends this information to the server, which then stores it in a database.

[1226] Entering and parsing questions

[1227] The user inputs a specific question, such as "Tell me an easy pasta recipe," into the device and sends it. This input data is sent to the server. The server analyzes the received question using natural language processing (NLP) technology, extracts keywords such as "easy," "pasta," and "recipe," and understands the user's intent.

[1228] Integrating historical data and external information

[1229] Based on the analysis results, the server retrieves the user's past data from the database. For example, if the user has previously searched for Italian food, that information will be used as a reference. The server also connects with external information services (such as cooking recipe providers) via API to retrieve the latest recipe information.

[1230] Generating and providing optimal information

[1231] The server combines the user's hobby and interest data with external information to generate the most appropriate answer. In this case, it identifies "an easy recipe for spaghetti with tomatoes and basil" and generates an answer in natural language format. The generated answer is sent to the device and displayed to the user.

[1232] Gathering feedback and learning

[1233] The user reviews the provided information and sends feedback such as "This recipe was very helpful" via their device. This feedback is received by the server and stored in a database. The server uses this feedback to continuously train the generative model, improving the accuracy of answers from the next time onwards.

[1234] Specific examples

[1235] For example, suppose a user inputs a question about "easy pasta recipes." The server analyzes this question and references the user's past search history and hobbies and preferences from a database. If the user has previously liked "Italian food," the server connects with an external information service to retrieve the latest recipe for "spaghetti with tomato and basil." The server formats this information in natural language and sends it to the device. The device displays it to the user as "an easy recipe for spaghetti with tomato and basil." If the user then rates this recipe as useful, that feedback is sent back to the server and used as training data for the model.

[1236] In this way, the present invention is a system that continues to provide optimal information in response to user questions, thereby improving the efficiency of information acquisition and user satisfaction.

[1237] The processing flow will be explained below.

[1238] Step 1:

[1239] The user accesses the service using an information terminal and enters basic information such as name, age, occupation, hobbies and interests into the registration form.

[1240] The terminal transmits this information to the server.

[1241] Step 2:

[1242] The server receives the basic information sent by the user and stores it in a database.

[1243] The server also generates a unique user ID for each user and creates a corresponding record.

[1244] Step 3:

[1245] The user enters a specific question, such as "Tell me an easy pasta recipe," into the terminal and sends it.

[1246] The terminal transmits the entered question to the server.

[1247] Step 4:

[1248] The server analyzes the received question using natural language processing (NLP) techniques.

[1249] The server extracts key keywords from the question ("easy," "pasta," "recipe") and understands the user's intent.

[1250] Step 5:

[1251] The server retrieves the user's past data from the database.

[1252] For example, if the user has previously searched for "Italian cuisine," that information can be used as a reference.

[1253] Step 6:

[1254] The server initiates API integration with an external information service to obtain the latest recipe information.

[1255] For example, recipe data for "spaghetti with tomato and basil" is received from an external recipe service.

[1256] Step 7:

[1257] The server integrates the acquired external information with the user's past data to generate the optimal answer.

[1258] For example, create an "easy tomato and basil spaghetti recipe" in natural language format.

[1259] Step 8:

[1260] The server sends the generated response to the terminal.

[1261] The device receives the response and displays it to the user as an "easy tomato and basil spaghetti recipe."

[1262] Step 9:

[1263] The user reviews the information provided and sends feedback via the device, such as "This recipe was helpful."

[1264] The terminal transmits the feedback information to the server.

[1265] Step 10:

[1266] The server stores the user feedback in a database.

[1267] The server uses this feedback to train a generative model and uses it to improve the accuracy of answers from next time onwards.

[1268] Example 1

[1269] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1270] Conventional information provision systems have difficulty providing accurate answers to user questions, and in order to improve user satisfaction, it is necessary to effectively integrate the user's past behavioral history and external information. Furthermore, due to a lack of means to collect user feedback and improve the accuracy of the system, continuous learning is not possible, raising concerns that the quality of the information provided may decline.

[1271] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1272] In this invention, the server includes means for receiving a user's question input via an information terminal, means for analyzing the received question using natural language processing to understand the user's intention, means for integrating the user's past data and information acquired from an external database to generate optimal information, means for transmitting the generated information to the information terminal and displaying it, and means for collecting user feedback input via the information terminal and training the generation AI model. This makes it possible to continue providing optimal information in response to a user's question, thereby improving the efficiency of information acquisition and user satisfaction.

[1273] An "information terminal" is a device that allows a user to access the system via the Internet and input and display data, and includes smartphones, personal computers, etc.

[1274] A "user question" is a textual inquiry that a user inputs into the system using an information terminal, requesting specific information.

[1275] "Natural language processing" is a technology that analyzes input text data in a way that is easy for a computer to understand, and grasps its meaning and intent.

[1276] "User intent" refers to the specific information or answer that the user wants to obtain through the question.

[1277] "Past data" refers to information such as a user's past behavioral history, search history, and interests that is stored in a database.

[1278] An "external database" is a data source that exists outside the system, and includes various information providing services that can be accessed via API.

[1279] "Optimal information" refers to the most appropriate and useful answer to a user's question, and is generated by integrating information from past data and external databases.

[1280] A "generative AI model" is an algorithm that continuously learns from data and feedback provided by users, improving the accuracy of its answers.

[1281] "Feedback" refers to the evaluations and opinions that users give in response to the information provided, and is used to help the system learn.

[1282] "Display means" refers to a function for visually presenting optimal information on an information terminal, and includes a screen or display.

[1283] "Integration means" is a process that combines the user's past data and information obtained from external databases to generate information that is optimal for the user.

[1284] The present invention is a system that uses an information terminal, a server, and a database, and provides optimal answers to questions asked by users in information searches. An embodiment of this system will be described in detail below.

[1285] First, the user accesses the system's website using an information terminal (smartphone or PC). The user enters basic information such as name, age, occupation, hobbies and interests into the registration form on the website. When the user presses the "Submit" button, the information terminal sends the entered data to the server using the HTTPS protocol. The server stores the received data in a database server (e.g., MySQL, PostgreSQL).

[1286] Next, the user enters a specific question. For example, "Tell me an easy pasta recipe" and clicks the submit button. The input data is again sent to the server via HTTPS protocol. The server analyzes the received question using natural language processing (NLP) technology. At this time, NLP libraries such as spaCy and NLTK are used to extract keywords such as "easy," "pasta," and "recipe." This allows the server to understand the user's intent.

[1287] The server then retrieves the user's past behavioral history and preference information from the database. This information is obtained using an SQL query. The server also connects to external databases (e.g., recipe service) via API to retrieve the latest recipe information. The acquired internal data is then integrated with external data to generate the optimal recipe information for the user. A generative AI model is used to generate a natural language answer (e.g., "Recipe for spaghetti with tomato and basil").

[1288] The generated answer is sent from the server to the device, which displays it to the user. The user checks the displayed information, enters feedback (e.g., "This recipe was very helpful"), and sends it to the server via the device. The server stores the feedback in a database and uses this information to continuously train the generative AI model. This improves the accuracy of answers in future queries.

[1289] Specific examples

[1290] For example, if a user inputs a question such as "Tell me an easy pasta recipe," the system operates as follows: The server first analyzes the question using NLP technology to extract keywords. It then combines the user's past data with the latest recipe information obtained from an external API to generate an "easy tomato and basil spaghetti recipe." This generated information is sent to the device and displayed to the user. Finally, the user sends feedback to the server on whether the provided recipe was helpful, and this feedback is used to train the generative AI model.

[1291] In this way, the present invention is a system that continues to provide optimal information in response to user questions, thereby improving the efficiency of information acquisition and user satisfaction.

[1292] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1293] Step 1:

[1294] A user accesses the service and enters registration information.

[1295] Specific operation: The user uses an information terminal (smartphone or PC) to access the system's website.

[1296] Input: A user fills in a registration form on a website with information such as name, age, occupation, hobbies and interests.

[1297] Data processing: The device temporarily stores the entered information.

[1298] Output: The persisted data is ready to be sent to the server in the next step.

[1299] Step 2:

[1300] The terminal transmits the input user information to the server.

[1301] What happens: The user clicks a button to submit the registration form.

[1302] Input: Stored user information (name, age, occupation, hobbies and interests).

[1303] Data calculation: The information terminal uses the HTTPS protocol to send the input information to the server.

[1304] Output: The server receives the user information.

[1305] Step 3:

[1306] The server stores the received user information in a database.

[1307] Specific operation: The server analyzes the received registration data and stores it in a database.

[1308] Input: User registration information.

[1309] Data manipulation: Creating new records in the database using SQL queries.

[1310] Output: The saved user information.

[1311] Step 4:

[1312] The user types a question and submits it.

[1313] Specific operation: The user enters a question on the information terminal and clicks the send button.

[1314] Input: Question text (e.g. "What's an easy pasta recipe?").

[1315] Data calculation: The information terminal sends the entered query data to the server.

[1316] Output: The server receives the query data.

[1317] Step 5:

[1318] The server analyzes the question using natural language processing.

[1319] What happens: The server parses the question using an NLP library (e.g., spaCy, NLTK).

[1320] Input: The question text submitted by the user.

[1321] Data processing: Extract keywords from the question text to understand the user's intent.

[1322] Output: Extracted keywords and user intent.

[1323] Step 6:

[1324] The server retrieves the user's past data from the database.

[1325] Specific operation: The server retrieves past behavioral history and interest data from the database based on the user's ID.

[1326] Input: User ID and parsed question information.

[1327] Data operations: Retrieving relevant data from a database using SQL queries.

[1328] Output: User's historical data.

[1329] Step 7:

[1330] The server retrieves the latest information from an external database.

[1331] Specific operation: The server calls an external API to obtain the latest recipe information.

[1332] Input: Extracted keywords (e.g., pasta, recipe).

[1333] Data Calculation: Sending requests to external APIs and parsing the responses.

[1334] Output: Latest recipe information.

[1335] Step 8:

[1336] The server generates the most appropriate information and provides it to the user.

[1337] Specific operation: The server integrates the user's past data with external data and generates answers in natural language format using a generative AI model.

[1338] Input: User's past data, latest recipe information, parsed question information.

[1339] Data processing: Generate optimal answers from integrated data.

[1340] Output: A natural language answer (e.g., "Easy tomato and basil spaghetti recipe").

[1341] Step 9:

[1342] The terminal displays the generated answer to the user.

[1343] Specific operation: The device displays the answer received from the server to the user.

[1344] Input: The answer sent by the server in natural language format.

[1345] Data processing: Formatting text for visual display on the device.

[1346] Output: The answer displayed to the user.

[1347] Step 10:

[1348] The user enters and submits feedback.

[1349] Specific actions: The user enters feedback on the device and clicks the submit button.

[1350] Input: Feedback text (e.g., "This recipe was very helpful").

[1351] Data calculation: The information terminal sends the input feedback to the server.

[1352] Output: The server receives the feedback data.

[1353] Step 11:

[1354] The server stores the feedback in a database and trains a generative AI model.

[1355] Specific operation: The server stores the feedback data in a database and updates the generative AI model.

[1356] Input: Received feedback data.

[1357] Data processing: Store the feedback in a database and use it as new data for training the AI ​​model.

[1358] Output: An updated generative AI model.

[1359] This allows the system to continue providing optimal information in response to user questions, improving the efficiency of information acquisition and user satisfaction.

[1360] (Application example 1)

[1361] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1362] Conventional product recommendation systems in virtual stores were unable to fully reflect the individual preferences and past purchase history of users, making it difficult to provide optimal products. Furthermore, there was no mechanism for effectively collecting and utilizing user feedback, making it difficult to improve the quality of service.

[1363] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1364] In this invention, the server includes means for receiving a user question input via an information terminal, means for analyzing the received question using natural language processing to understand the user's intention, means for integrating the user's past data and information acquired from external information services to recommend optimal products based on the user's preferences, means for displaying the generated product information on the information terminal, and means for collecting user feedback input via the information terminal and training a generative model. This makes it possible to propose optimal products based on the individual preferences of users and improve the quality of services by utilizing feedback.

[1365] An "information terminal" refers to an electronic device, such as a smartphone or tablet, that allows a user to input questions and receive answers from a server.

[1366] "Natural language processing" refers to the technology that allows computers to understand and analyze human language, allowing them to understand the intent of users' questions.

[1367] "User's past data" refers to information such as the user's past search history, purchase history, and feedback.

[1368] "External Information Services" refers to information or databases provided by third parties, such as product catalogs or review sites.

[1369] "Means for recommending optimal products" refers to algorithms or programs that integrate the user's past data with information obtained from external information services and select the most suitable product based on the user's preferences.

[1370] The "generated product information" refers to detailed product information selected by the optimal product recommendation means, including, for example, the product name, price, and features.

[1371] "Feedback" refers to a user's evaluation or opinion of the information provided or the recommended products.

[1372] A "generative model" refers to a machine learning model that generates optimal information based on user questions and feedback.

[1373] The present invention provides a personalized shopping assistant system for a virtual store that recommends optimal products based on a user's preferences and past purchase history. Specific embodiments of this system are described below.

[1374] Hardware and software used

[1375] This system consists of information terminals (e.g., smartphones, tablets, smart glasses) accessed by users and a server that processes data. The software includes the following elements:

[1376] 1. Natural Language Processing (NLP) engine: used to analyze the user's question and extract keywords, for example using Python and the transformers library.

[1377] 2. Database: Stores user registration information, past data, and product information obtained from external information services. For example, a database system such as MySQL is used.

[1378] 3. API integration module: Connects to external information services (e.g., product catalogs, review sites, etc.) via API to obtain the latest product information.

[1379] 4. Generative AI model: A model for recommending optimal products based on user preference data and external information. As an example, we will use a machine learning model.

[1380] System Operation

[1381] The server operates as follows:

[1382] 1. Accepting a user question: The user's question (e.g., "Please recommend a new smartphone") is received via the information terminal.

[1383] 2. Natural language processing of questions: The content of the question is analyzed using natural language processing technology to extract important keywords.

[1384] 3. Information integration: Integrates the user's past data (e.g., past purchase history) and the latest product information obtained from external information services to select the optimal product based on the user's preferences.

[1385] 4. Product information generation and display: The selected product information is formatted in natural language and displayed on the information terminal.

[1386] 5. Feedback collection: Collect user feedback entered via the information terminal and continuously train the generative AI model.

[1387] Specific examples

[1388] For example, a user might use an information terminal to ask, "Please recommend a new smartphone." This input is sent to the server and analyzed by a natural language processing engine. As a result of the analysis, important keywords such as "smartphone" and "recommendation" are extracted. The server then retrieves the user's past purchase history and preference data from a database, and obtains the latest smartphone information through the API of an external information service.

[1389] The generative AI model integrates this information and recommends the most suitable smartphone for the user (e.g., "latest model smartphone"). This information is formatted in natural language and displayed on the information terminal. When the user reviews the recommendation and sends feedback saying "this suggestion was helpful," that feedback is stored on the server and used as training data for the generative AI model.

[1390] Examples of prompt statements

[1391] User: Recommend me a new smartphone.

[1392] System: Based on your past search and purchase history, we estimate that a new smartphone model would be suitable. We recommend a 'new smartphone model' featuring a high-performance camera and a long-lasting battery. The price is 50,000 yen. How about this product?

[1393] As described above, the present invention makes it possible to provide appropriate and personalized product suggestions based on the individual preferences and past data of the user.

[1394] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1395] Step 1:

[1396] The user enters a question.

[1397] The user uses an information terminal (e.g., a smartphone) to input a question. Specifically, the user inputs text such as "Please recommend a new smartphone" and presses the send button. The input question (text data) is sent to the server.

[1398] Step 2:

[1399] The server accepts the query.

[1400] The server receives questions sent by users, stores them as text data, and then passes the text data to a natural language processing engine.

[1401] Step 3:

[1402] The server performs natural language processing.

[1403] The server uses a natural language processing engine (e.g., the transformers library) to analyze the received text data, split the question, and extract important keywords (e.g., "smartphone," "recommended"). The results with the extracted keywords are then returned.

[1404] Step 4:

[1405] The server obtains the user's past data and external information.

[1406] The server retrieves the user's past data (past purchase history, search history, etc.) from the database based on the keywords. At the same time, it calls an external information service API to retrieve the latest product information related to the keywords. The retrieved data is then integrated.

[1407] Step 5:

[1408] The server recommends the best products.

[1409] The server uses the generative AI model to analyze and integrate the user's past data and product data obtained from external information, thereby selecting the product (e.g., the latest model smartphone) that best suits the user's preferences.

[1410] Step 6:

[1411] The server generates and displays product information.

[1412] The server formats the selected product details (product name, price, features, etc.) into a natural language format. This generated product information is sent to the information terminal, which then displays the information to the user.

[1413] Step 7:

[1414] The user enters feedback.

[1415] The user checks the presented product information and inputs feedback such as "This suggestion was helpful." The input feedback is then sent back to the server.

[1416] Step 8:

[1417] The server collects feedback and learns.

[1418] The server receives feedback sent by the user, stores it in a database, and uses it as training data for the generative AI model, thereby improving the accuracy of recommendations in the future.

[1419] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1420] The present invention is a system that uses an information terminal, a server, a database, and an emotion engine to provide optimal answers to questions asked by users in information searches, and further adjusts the content of the answers taking into account the user's emotions. A specific embodiment of this system is described below.

[1421] User registration and information collection

[1422] First, a user accesses the service using an information terminal and enters basic information such as name, age, occupation, hobbies and interests into the registration form. The terminal sends this information to the server, which stores it in a database. The emotion engine also performs initial settings to understand the user's basic emotional patterns.

[1423] Entering and parsing questions

[1424] The user inputs a specific question, such as "Tell me an easy pasta recipe," into the device and sends it. This input data is sent to the server. The server analyzes the received question using natural language processing (NLP) technology, extracts keywords such as "easy," "pasta," and "recipe," and understands the user's intent.

[1425] Acquiring user emotion data

[1426] When a question is entered, the emotion engine collects the user's emotion data and uses, for example, text analysis and facial expression recognition technology to understand the user's current emotional state (such as joy, sadness, or anger).

[1427] Integrating historical data and external information

[1428] Based on the analysis results, the server retrieves the user's past data from the database. For example, if the user has previously searched for Italian food, that information will be used as a reference. The server also connects with external information services (such as cooking recipe providers) via API to retrieve the latest recipe information.

[1429] Optimal information generation and display

[1430] The server combines the acquired external information, the user's past data, and emotional data to generate the optimal answer. For example, it identifies an "easy tomato and basil spaghetti recipe" and generates an answer tailored to the user's emotional state, such as "You seem to be in a low mood today. Try this easy pasta recipe to cheer you up." The generated answer is sent to the device and displayed to the user.

[1431] Gathering feedback and learning

[1432] The user reviews the provided information and sends feedback via their device, such as "This recipe was very helpful." This feedback is received by the server and stored in a database. The server uses this feedback to continuously train the generative model and emotion engine, and uses it to improve the accuracy of answers in future searches.

[1433] Specific examples

[1434] For example, if a user enters "easy pasta recipes," the server analyzes the question and references the user's past search history and hobbies and preferences from a database. If the emotion engine recognizes the emotion "a little tired" from the user's input, the server connects with an external information service to retrieve a recipe for "spaghetti with tomato and basil." Based on this, the server makes adjustments such as "You seem a little tired today. Try refreshing yourself with this easy pasta recipe," and sends and displays it on the device. If the user rates the recipe as "helpful," that feedback is sent back to the server and used as learning data for the model. The emotion engine also improves its accuracy based on feedback.

[1435] In this way, the present invention is a system that provides optimal information in response to a user's question while enabling flexible responses according to the user's emotional state, thereby improving the efficiency of information acquisition and user satisfaction.

[1436] The processing flow will be explained below.

[1437] Step 1:

[1438] The user accesses the service using an information terminal and enters basic information such as name, age, occupation, hobbies and interests into the registration form.

[1439] The terminal transmits this information to the server.

[1440] Step 2:

[1441] The server receives the basic information sent by the user and stores it in a database.

[1442] The server also generates a unique user ID for each user and creates a corresponding record.

[1443] Step 3:

[1444] The user enters a specific question, such as "Tell me an easy pasta recipe," into the terminal and sends it.

[1445] The terminal transmits the entered question to the server.

[1446] Step 4:

[1447] The emotion engine analyzes the user's input text and recognizes emotional states (e.g., joy, sadness, fatigue, etc.).

[1448] The server acquires the emotion data and processes it together with the question data.

[1449] Step 5:

[1450] The server analyzes the received question using natural language processing (NLP) techniques.

[1451] The server extracts key keywords from the question ("easy," "pasta," "recipe") and understands the user's intent.

[1452] Step 6:

[1453] The server retrieves the user's past data from the database.

[1454] For example, if the user has previously searched for "Italian cuisine," that information can be used as a reference.

[1455] Step 7:

[1456] The server initiates API integration with an external information service to obtain the latest recipe information.

[1457] For example, recipe data for "spaghetti with tomato and basil" is received from an external recipe service.

[1458] Step 8:

[1459] The server integrates the acquired external information, past data, and emotional data to generate the optimal answer.

[1460] For example, if the user's emotional state is recognized as "tired," the system adjusts the response by saying, "You seem a little tired today. Try refreshing yourself with this simple pasta recipe."

[1461] Step 9:

[1462] The server sends the generated response to the terminal.

[1463] The device receives the response and displays it to the user as an "easy tomato and basil spaghetti recipe."

[1464] Step 10:

[1465] The user reviews the information provided and sends feedback via the device, such as "This recipe was very helpful."

[1466] The terminal transmits the feedback information to the server.

[1467] Step 11:

[1468] The server stores the user feedback in a database.

[1469] The server uses this feedback to continuously train the generative model and emotion engine, and uses it to improve the accuracy of answers from next time onwards.

[1470] Example 2

[1471] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1472] Conventional information retrieval systems provide answers only to questions entered by the user and do not take the user's emotional state into account, which means that the answers may not necessarily meet the user's needs. Furthermore, the means for collecting feedback are limited, which leaves the user experience unsatisfactory. Therefore, the present invention aims to provide optimal information that takes into account the user's emotional state and continuously improve the system's accuracy by utilizing feedback.

[1473] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1474] In this invention, the server includes means for receiving a user question input via an information terminal, means for analyzing the received question using natural language processing to understand the user's intention, means for analyzing the user's current emotional state, means for integrating the user's past data and information obtained from external information services to generate optimal information, means for adjusting the generated information according to the user's emotional state and displaying it on the information terminal, and means for collecting user feedback input via the information terminal and training the generative model and emotion engine. This makes it possible to provide optimal information in response to the user's question while also responding flexibly to the user's emotional state.

[1475] An "information terminal" is a device used by a user to input a question, and includes a computer, a smartphone, a tablet, and the like.

[1476] A "server" is a central system that receives data from information terminals via a network and analyzes and processes the data.

[1477] "Natural Language Processing (NLP)" is a technology that allows computers to understand and analyze human language and grasp the user's intent.

[1478] An "emotion engine" is an algorithm and software that analyzes and identifies a user's emotional state from their text and facial expressions.

[1479] A "database" is a system for managing and storing user registration information and past data.

[1480] "External information services" are third-party services that provide information via the Internet.

[1481] A "generative model" is an algorithm and software that generates optimal information in response to a user's question.

[1482] "Feedback" refers to the evaluation or opinion that a user enters regarding the information provided.

[1483] "Synthesis" is the process of combining data obtained from multiple different sources into a single, consistent piece of information.

[1484] User registration and information collection

[1485] First, the user accesses the service using an information terminal (smartphone, computer, tablet, etc.) and enters basic information such as name, age, occupation, hobbies and interests into the registration form. This information is then sent from the terminal to the server. The server stores the received information in a database and performs initial settings to understand the user's basic emotional patterns using an emotion engine. At this point, the server stores the user's individual information in a database, making it possible to provide individualized information.

[1486] Entering and parsing questions

[1487] The user inputs a specific question into an information terminal. For example, they can type in the text "Tell me an easy pasta recipe" and send it. This input data is sent from the terminal to the server. The server analyzes the received question using natural language processing (NLP) technology, extracts keywords such as "easy," "pasta," and "recipe," and understands the user's intent. This analysis can be performed using open-source natural language processing libraries or cloud-based NLP services.

[1488] Acquiring user emotion data

[1489] When a user enters a question, the emotion engine collects the user's emotion data. It can use a camera-equipped information device to understand the user's current emotional state (e.g., joy, sadness, anger) using text analysis and facial expression recognition technology. For emotion recognition, it can use, for example, an open-source emotion analysis library or a cloud-based emotion recognition API.

[1490] Integrating historical data and external information

[1491] Based on the analysis results, the server retrieves the user's past data (e.g., past search history, hobbies, and preferences) from the database. It also connects with external information services (e.g., cooking recipe services) via API to retrieve the latest recipe information. During this process, the server communicates with external services using an enterprise DBMS, cloud database service, or REST API.

[1492] Optimal information generation and display

[1493] The server combines the acquired external information, the user's past data, and emotional data to generate the optimal answer. For example, it identifies an "easy tomato and basil spaghetti recipe" and generates an answer tailored to the user's emotional state, such as "You seem to be in a low mood today. Try this easy pasta recipe to cheer you up." The generated answer is sent to the device and displayed to the user.

[1494] Gathering feedback and learning

[1495] The user reviews the provided information and sends feedback via their device, such as "This recipe was very helpful." This feedback is received by the server and stored in a database. The server uses this feedback to continuously train the generative model and emotion engine, and uses it to improve the accuracy of answers in the future.

[1496] Specific examples

[1497] For example, if a user enters "easy pasta recipes," the server analyzes the question and references the user's past search history and hobbies and preferences from a database. If the emotion engine recognizes the emotion "a little tired" from the user's input, the server connects with an external information service to retrieve a recipe for "spaghetti with tomato and basil." Based on this, the server makes adjustments such as "You seem a little tired today. Try refreshing yourself with this easy pasta recipe," and sends and displays it on the device. If the user rates the recipe as "helpful," that feedback is sent back to the server and used as learning data for the model. The emotion engine also improves its accuracy based on feedback.

[1498] Example prompts to input to the generative AI model

[1499] "When a user is looking for an easy pasta recipe, give them the perfect recipe based on their emotional state and why. For example, 'You seem a little tired today. Try refreshing yourself with this easy pasta recipe.'"

[1500] In this way, the present invention is a system that provides optimal information in response to a user's question while enabling flexible responses according to the user's emotional state, thereby improving the efficiency of information acquisition and user satisfaction.

[1501] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1502] Step 1:

[1503] User registration and information collection

[1504] Input: The user uses an information terminal to input basic information such as name, age, occupation, hobbies and interests.

[1505] Processing: The device sends the input information to the server. The server stores the received information in a database and sets the user's initial emotional pattern using the emotion engine.

[1506] Output: User information is saved in the database and the emotion engine completes its initialization.

[1507] How it works: The user accesses the registration page in a web browser or mobile app, enters the required information in the form, and then presses the "Register" button. The device sends this data to the server as an HTTP request. The server saves the information in a database and calls the emotion recognition API to set the initial emotion pattern.

[1508] Step 2:

[1509] Entering and parsing questions

[1510] Input: The user types a specific question into the device and sends it. For example, the text "Tell me a simple pasta recipe."

[1511] Processing: The device sends the input data to the server, which uses natural language processing (NLP) techniques to analyze the question and extract keywords.

[1512] Output: Keyword data is generated to understand the intent of the question.

[1513] Specific operation: The user enters a question into the text input form and presses the "Submit" button. The device sends the text data as an HTTP POST request to the server. The server uses the NLP module to extract keywords such as "easy," "pasta," and "recipe."

[1514] Step 3:

[1515] Acquiring user emotion data

[1516] Input: User text and facial expression data when entering a question.

[1517] Processing: The server uses the emotion engine to analyze the user's current emotional state (happiness, sadness, anger, etc.).

[1518] Output: Data of the current user's emotional state.

[1519] Specific operation: The server inputs text data into the emotion engine and calls the emotion recognition API. The emotion engine analyzes emotions such as "joy," "sadness," and "anger." The analysis results are recorded for use in the next step.

[1520] Step 4:

[1521] Integrating historical data and external information

[1522] Input: Question analysis results, user history, data from external information services.

[1523] Processing: The server retrieves the user's past data from the database, connects to external information services via API, and retrieves the latest information. This data is then integrated.

[1524] Output: A consolidated dataset.

[1525] What it does: The server queries the database using the user ID to retrieve past search history and hobbies and preferences. It sends a request to an external API to retrieve the latest recipe information, for example, from a cooking recipe service. It then integrates all the data and creates a dataset to generate optimal information.

[1526] Step 5:

[1527] Optimal information generation and display

[1528] Input: Unified dataset, user emotional state.

[1529] Processing: The server generates the optimal answer based on the integrated data, adjusts the content according to the user's emotional state, and sends the generated answer to the device.

[1530] Output: The adjusted answer is displayed on the terminal.

[1531] How it works: The server uses its internal algorithm to identify an "easy recipe for spaghetti with tomato and basil." The emotion engine adds a comment based on the user's emotional state. The server sends the answer in JSON format to the device, which then displays it on the screen.

[1532] Step 6:

[1533] Gathering feedback and learning

[1534] Input: User feedback data.

[1535] Processing: The device sends feedback data to the server. The server receives the feedback and stores it in a database. This information is used to train the generative model and emotion engine.

[1536] Output: A trained generative model and an emotion engine.

[1537] Specific operation: The user checks the provided information, enters a rating, and presses the "Submit" button. The device sends the feedback data to the server. The server stores the feedback in a database, and the learning module uses it to update the parameters of the generative model and emotion engine.

[1538] (Application example 2)

[1539] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1540] In modern brick-and-mortar stores, improving customer satisfaction and providing personalized service is a major challenge. It is particularly difficult to grasp a customer's emotional state in real time and provide appropriate product recommendations and services accordingly. Furthermore, conventional information provision systems using information terminals lack the ability to respond flexibly and consider the user's emotions. Furthermore, they lack an effective mechanism for accumulating feedback and reflecting it in future customer service.

[1541] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1542] In this invention, the server includes means for receiving a user's question input via an information terminal, means for analyzing the received question using natural language processing and grasping the user's intention, means for integrating the user's past data and information acquired from an external information service to generate optimal information, means for displaying the generated information on the information terminal, means for collecting user feedback input via the information terminal and training a generative model, means for recognizing the user's facial expression and grasping their emotional state, means for generating and displaying information taking the user's emotional state into consideration, and means for visually providing the displayed information via smart glasses. This enables personalized responses adapted to customer emotions and improves customer satisfaction in real time.

[1543] An "information terminal" is an electronic device that allows a user to input or receive information.

[1544] The "receiving means" is a device or program that has the function of receiving data transmitted from an information terminal.

[1545] "Natural language processing" is a technology in which a computer analyzes human language and understands its meaning.

[1546] "User intent" is the purpose or intention behind the question or request entered by the user.

[1547] "Past data" refers to data based on information previously generated by the user and their behavioral history.

[1548] "External information services" are additional sources of information provided by external servers or databases.

[1549] "Best information" is the answer or suggestion that best suits the user's question or requirements.

[1550] "Generating means" refers to a device or program that has the function of combining information based on collected data and creating the necessary information.

[1551] The "display means" is a device or program that has the function of presenting the generated information in a form that can be visually recognized by the user.

[1552] "Feedback" refers to a user's evaluation or opinion of the information or service provided.

[1553] The "collection means" is a device or program that has the function of collecting feedback from users.

[1554] A "generative model" is a computational model that generates information based on large amounts of data.

[1555] A "training means" is a device or program that has the function of improving the generative model based on collected feedback.

[1556] "Means for recognizing facial expressions" refers to a device or program that has the function of recognizing and analyzing a user's facial expressions using a camera or sensor.

[1557] "Emotional state" refers to a user's current feelings or mood.

[1558] A "means for taking into account" is a device or program that has the ability to adjust information based on the user's emotional state.

[1559] "Smart glasses" are a wearable eyeglass-type device that has the function of visually displaying information.

[1560] The present invention is a system that uses an information terminal, a server, a database, and an emotion engine to provide optimal answers to user questions. This system adjusts the content of the answers taking into account the user's emotions, thereby providing more appropriate information.

[1561] System Overview

[1562] The system consists of the following main components:

[1563] Information terminal: A device (e.g., smartphone, tablet, PC, etc.) through which a user enters questions and receives answers.

[1564] Server: The primary computer system for information analysis, data integration, and answer generation.

[1565] Database: Storage for saving user history and registration information.

[1566] Emotion engine: Software for recognizing a user's facial expressions and understanding their emotional state.

[1567] What the program does

[1568] The server receives the user's question sent from the information terminal and analyzes it using natural language processing (NLP) technology. Specifically, it uses the Python libraries TextBlob and NLTK. As a result of the analysis, keywords are extracted from the question to understand the user's intent.

[1569] The system then integrates the user's past data (e.g., previously searched information and ratings) with information obtained from external information services (e.g., the latest recipe service or product information service). This process uses API integration technology, for example, to obtain data from external information services via RESTful APIs.

[1570] Furthermore, an emotion engine (e.g., an OpenCV library for facial expression recognition) is used to analyze the user's facial expressions captured by the device's camera and understand their emotional state. Emotional data can be used to adjust the content of responses.

[1571] For example, if a user asks "What's an easy pasta recipe?", the system generates an answer in the following order:

[1572] 1. The server receives the question and uses NLP technology to extract the keywords "easy," "pasta," and "recipe."

[1573] 2. The server references the user's past search history in the database and uses data on Italian cuisine that the user previously searched for as a reference.

[1574] 3. The emotion engine analyzes facial expressions and recognizes the user's emotional state as "a little tired."

[1575] 4. The server retrieves the latest "Tomato and Basil Spaghetti" recipe from an external information service and generates a message saying, "You seem a little tired today. Try refreshing yourself with this easy pasta recipe."

[1576] 5. The generated answer is sent to the information terminal and visually displayed to the user.

[1577] Specific examples

[1578] Here is an example where a user enters "easy pasta recipes." The server analyzes the question and references the user's past search history from the database. If the emotion engine recognizes the emotion "a little tired" from the user's input, the server connects with an external information service and retrieves a recipe for "spaghetti with tomato and basil." Based on this, the server makes adjustments such as "You seem a little tired today. Try refreshing yourself with this easy pasta recipe," generates an answer, and sends it to the information terminal for display.

[1579] Prompt Sentence Examples

[1580] 1. "Tell me an easy pasta recipe."

[1581] 2. "Do you have any drinks to cheer me up?"

[1582] 3. "I'm looking for some new shoes. What do you recommend?"

[1583] 4. "Do you have any teas that can help relieve stress?"

[1584] In this way, the present invention is a system that enables flexible responses according to the user's emotions, thereby realizing more appropriate information provision and improving customer satisfaction.

[1585] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1586] Step 1:

[1587] The server receives the user's question entered via an information terminal. The input data is a question, specifically text information such as "Tell me an easy pasta recipe." The received question is passed directly to the next analysis step.

[1588] Step 2:

[1589] The server analyzes the received question using natural language processing (NLP) techniques. Specifically, it uses Python's TextBlob and NLTK libraries to parse the question and extract keywords. The input is the question, and the output is a list of keywords such as "easy," "pasta," and "recipe."

[1590] Step 3:

[1591] The server retrieves the user's search history and evaluation data from the user's past database. At the same time, it also retrieves the latest related information through the API of an external information service (e.g., a recipe provision service). The input is a list of keywords, and the output is a list of relevant information.

[1592] Step 4:

[1593] The emotion engine captures the user's facial expressions in real time through the information terminal's camera and analyzes their emotional state using facial expression recognition technology (such as the OpenCV library).The input is real-time facial image data, and the output is emotional states such as "happiness," "sadness," and "tiredness."

[1594] Step 5:

[1595] The server combines the user's emotional state, past data, and external information to generate the optimal information response. Using a generative AI model, it combines these data and creates a response with wording appropriate to the user's emotions. The input is the emotional state and a list of information, and the output is the text of the optimal response.

[1596] Step 6:

[1597] The server sends the generated answer to the information terminal and displays it on the user's smart glasses or other information terminal. The user can visually confirm the displayed answer. The input is the text of the best answer, and the output is the answer displayed on the information terminal.

[1598] Step 7:

[1599] The user inputs feedback on the provided information through the information terminal. The feedback includes specific evaluations such as "This recipe was very useful." The input is the user's feedback text, and the output is the feedback data.

[1600] Step 8:

[1601] The server stores the collected feedback in a database and uses it as training data for the generative model and emotion engine. This improves the accuracy of answers from the next time onwards. The input is the feedback data, and the output is the model reflected in the training.

[1602] 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 control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1603] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1604] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1605] The emotion identification model 59 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 an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1606] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1607] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1608] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1609] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1610] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs 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 a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1611] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1612] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1613] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1614] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1615] 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.

[1616] It is not necessary to store all 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 all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1617] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1618] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with 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). Also, the hardware resource that executes the specific processing may be a single processor.

[1619] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1620] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1621] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1622] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1623] The following is further disclosed regarding the above embodiment.

[1624] (Claim 1)

[1625] means for receiving a user question inputted via an information terminal;

[1626] A means of analyzing received questions using natural language processing to understand the user's intent;

[1627] A means for integrating the user's past data and information acquired from external information services to generate optimal information;

[1628] means for displaying the generated information on an information terminal;

[1629] a means for collecting user feedback input via an information terminal and training a generative model;

[1630] A system including:

[1631] (Claim 2)

[1632] 2. The system according to claim 1, further comprising means for storing user registration information in a database and enabling provision of information individual to each user.

[1633] (Claim 3)

[1634] 2. The system according to claim 1, further comprising means for evaluating the suitability of the generated information and selecting the most suitable information from among information obtained from a plurality of external information services.

[1635] "Example 1"

[1636] (Claim 1)

[1637] means for receiving a user question inputted via an information terminal;

[1638] A means of analyzing received questions using natural language processing to understand the user's intent;

[1639] A means for integrating the user's past data and information acquired from an external database to generate optimal information;

[1640] means for transmitting the generated information to an information terminal and displaying it;

[1641] A means for collecting user feedback input via an information terminal and training a generative AI model;

[1642] A system including:

[1643] (Claim 2)

[1644] 2. The system according to claim 1, further comprising means for storing user registration information in a database and enabling provision of information individual to each user.

[1645] (Claim 3)

[1646] 10. The system of claim 1, further comprising means for evaluating the suitability of the generated information and selecting the most suitable information from among the information obtained from a plurality of external databases.

[1647] "Application Example 1"

[1648] (Claim 1)

[1649] means for receiving a user question inputted via an information terminal;

[1650] A means of analyzing received questions using natural language processing to understand the user's intent;

[1651] A means for integrating the user's past data and information acquired from external information services to recommend optimal products according to the user's preferences;

[1652] a means for displaying the generated product information on an information terminal;

[1653] a means for collecting user feedback input via an information terminal and training a generative model;

[1654] A system including:

[1655] (Claim 2)

[1656] 2. The system according to claim 1, further comprising means for storing user registration information in a database and enabling provision of information individual to each user.

[1657] (Claim 3)

[1658] 2. The system according to claim 1, further comprising means for evaluating the suitability of the generated information and selecting the most suitable information from among information obtained from a plurality of external information services.

[1659] "Example 2: Combining Emotion Engines"

[1660] (Claim 1)

[1661] means for receiving a user question inputted via an information terminal;

[1662] A means of analyzing received questions using natural language processing to understand the user's intent;

[1663] means for analyzing a user's current emotional state;

[1664] A means for integrating the user's past data and information acquired from external information services to generate optimal information;

[1665] a means for adjusting the generated information in accordance with the emotional state of the user and displaying the information on the information terminal;

[1666] a means for collecting user feedback input via an information terminal and training a generative model and an emotion engine;

[1667] A system including:

[1668] (Claim 2)

[1669] 2. The system according to claim 1, further comprising means for storing user registration information in a database and enabling provision of information individual to each user.

[1670] (Claim 3)

[1671] 2. The system according to claim 1, further comprising means for evaluating the suitability of the generated information and selecting the most suitable information from among information obtained from a plurality of external information services.

[1672] "Application example 2 when combining emotion engines"

[1673] (Claim 1)

[1674] means for receiving a user question inputted via an information terminal;

[1675] A means of analyzing received questions using natural language processing to understand the user's intent;

[1676] A means for integrating the user's past data and information acquired from external information services to generate optimal information;

[1677] means for displaying the generated information on an information terminal;

[1678] a means for collecting user feedback input via an information terminal and training a generative model;

[1679] means for recognizing a user's facial expression and understanding their emotional state;

[1680] means for generating and displaying information that takes into account the emotional state of a user;

[1681] a means for visually providing the displayed information through the smart glasses;

[1682] A system including:

[1683] (Claim 2)

[1684] 2. The system according to claim 1, further comprising means for storing user registration information in a database and enabling provision of information individual to each user.

[1685] (Claim 3)

[1686] 2. The system according to claim 1, further comprising means for evaluating the suitability of the generated information and selecting the most suitable information from among information obtained from a plurality of external information services. [Explanation of symbols]

[1687] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for receiving a user question inputted via an information terminal; A means of analyzing received questions using natural language processing to understand the user's intent; A means for integrating the user's past data and information acquired from external information services to generate optimal information; means for displaying the generated information on an information terminal; a means for collecting user feedback input via an information terminal and training a generative model; A system including:

2. 2. The system according to claim 1, further comprising means for storing user registration information in a database and enabling provision of information individual to each user.

3. 2. The system according to claim 1, further comprising means for evaluating the suitability of the generated information and selecting the most suitable information from among information obtained from a plurality of external information services.

Images