System

Abstract

To provide a system which enables a user to obtain quick and proper answers to various problems and questions and can greatly contribute to the problem solution of daily life.SOLUTION: A system includes means for receiving input information from a user terminal, means for selecting an appropriate expert model on the basis of the received information, means for generating an appropriate answer on the basis of the selected expert model, and means for transmitting the generated answer to the user terminal.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] The problem this invention aims to solve is the difficulty of obtaining quick and appropriate answers to the wide variety of problems and questions users face in their daily lives. Traditional web searches and Q&A sites can make it difficult to find the information needed, or to directly consult with someone with specialized knowledge. While generative AI technology has emerged, users often hesitate to use it because they do not know how to use it properly. [Means for solving the problem]

[0005] The present invention solves the above problems by the following means.

[0006] 1. Provide a means to receive input information from user devices, allowing users to easily input and send their inquiries.

[0007] 2. Provide a means to automatically select the appropriate expert model based on the information received, so that the answer best suited to the user's needs can be provided.

[0008] 3. Provide a means for transmitting the answers generated by the selected expert model to the user terminal, allowing the user to quickly obtain useful information.

[0009] 4. Provide a means for users to mark individual expert models as favorites, allowing quick access to specific experts.

[0010] 5. Provide a means to search for the anonymous consultation content of other users, so that information can be obtained from similar consultation cases in the past.

[0011] This will enable users to receive prompt and appropriate answers to a wide range of problems and questions, providing a system that can make a significant contribution to solving problems in everyday life.

[0012] "User terminal" means an electronic device used by a user to input or receive information, including a smartphone or computer.

[0013] "Input Information" refers to text or other data that a User sends to the System via a User Terminal.

[0014] "Means for receiving" refers to a device or software that has the function of taking in input information sent from a user terminal and processing it within the system.

[0015] An "expert model" refers to an artificial intelligence algorithm that generates answers based on specialized knowledge in a specific field, such as medicine, law, or business.

[0016] "Means for selecting" refers to a device or software that has the function of automatically selecting the most suitable expert model based on input information.

[0017] "Means for generating an answer" refers to a device or software that has the function of allowing the expert model to analyze input information and generate an appropriate answer.

[0018] "Transmitting means" refers to a device or software that has the function of transferring the generated response to a user terminal.

[0019] "Favorite Registration Means" means a device or software that allows a user to save a particular Expert Model for easy access.

[0020] "Anonymous search means" refers to a device or software that has the function of making the content of a consultation searchable while excluding the personal information of other users.

[0021] "Means for displaying search results in a list" refers to a device or software that has the function of displaying the searched consultation contents in a list format on the user's terminal. [Brief explanation of the drawings]

[0022] [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

[0023] 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.

[0024] First, the terms used in the following description will be explained.

[0025] 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).

[0026] 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.

[0027] 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.

[0028] 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.

[0029] 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."

[0030] [First embodiment]

[0031] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0032] 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.

[0033] 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).

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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.

[0038] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0039] 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.

[0040] 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.

[0041] 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.

[0042] 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."

[0043] The present invention uses a system consisting of a user terminal, a server, and an expert model to provide prompt and appropriate answers to various inquiries in the user's daily life. A specific embodiment of this system will be described below.

[0044] User Registration and Login

[0045] User Registration:

[0046] When a user uses the app for the first time, they enter their personal information (name, email address, password) to register. This information is sent from the user's device to the server and stored in a database. Once registration is complete, a confirmation message is returned to the user's device.

[0047] Login:

[0048] Existing users log in by entering their email address and password. The entered login information is sent from the user's device to the server and checked against the database. If authentication is successful, a session ID is generated and returned to the user's device, allowing the user to access the home screen.

[0049] Individual chat consultation

[0050] Enter and submit your inquiry:

[0051] When a user wants to ask a specific question, they enter the question in text format on the chat screen of their device and press the send button. The question and session ID are then sent to the server.

[0052] Analyzing the inquiry and generating an answer:

[0053] The server analyzes the received consultation content and selects the most appropriate expert model (medical, legal, business, etc.). The selected expert model analyzes the input consultation content and generates an appropriate answer.

[0054] Submit and view your answers:

[0055] The generated answers are sent from the server to the user's device and displayed on the chat screen, allowing the user to quickly obtain the information they need.

[0056] Search and browse functions

[0057] Search for consultation content:

[0058] Users can search for past consultations with other users by entering specific keywords in the search bar. This search request is sent from the user's device to the server, and the consultations in the database are anonymized and returned as search results.

[0059] Displaying search results:

[0060] The search results sent from the server are displayed as a list on the user's device, and the user can view past similar consultations. This allows users to refer to information on cases similar to their own and use it as a reference for self-solving.

[0061] Add to Favorites

[0062] Favorite settings:

[0063] If a user likes a particular expert model, they can register it by pressing the "Favorite" button. The favorite registration request is sent from the user's terminal to the server and stored in the database.

[0064] Use Favorites:

[0065] Registered favorite expert models can be easily accessed on the user's device, and can be quickly recalled if the user wishes to access them again, improving convenience.

[0066] Specific examples

[0067] For example, consider the case of a medical consultation. When a user inputs and sends the question, "I've been having terrible headaches lately. What should I do?", the server passes this information to the medical expert model. The medical expert model generates a response such as, "Possible causes of headaches include stress, eye strain, and lack of sleep. We recommend that you first get plenty of rest. If this does not improve, please consult a doctor." This response is sent to the user's device, where the user can view it on the chat screen.

[0068] As described above, the present invention provides a system that enables users to obtain prompt and appropriate answers to a variety of problems and questions.

[0069] The processing flow will be explained below.

[0070] User Registration and Login

[0071] New Registration

[0072] Step 1:

[0073] The user launches the app and sees the signup form.

[0074] Step 2:

[0075] The user's device prompts them to enter their name, email address, and password, then presses the "Send" button.

[0076] Step 3:

[0077] The user terminal sends the input information to the server.

[0078] Step 4:

[0079] The server stores the received information in a database.

[0080] Step 5:

[0081] The server returns a registration success message to the user terminal.

[0082] Log in

[0083] Step 1:

[0084] The user enters their email address and password on the login screen and clicks the "Login" button.

[0085] Step 2:

[0086] The user terminal sends the input information to the server.

[0087] Step 3:

[0088] The server checks the received information against a database.

[0089] Step 4:

[0090] If the server matches, it generates a session ID and sends it to the user's device.

[0091] Step 5:

[0092] The user device saves the session ID and displays the home screen.

[0093] Individual chat consultation

[0094] Step 1:

[0095] The user enters the consultation content in text on the app's consultation screen and presses the send button.

[0096] Step 2:

[0097] The user device sends the consultation content and session ID to the server.

[0098] Step 3:

[0099] The server analyzes the received consultation content and selects an appropriate expert model.

[0100] Step 4:

[0101] The expert model analyzes the consultation content and generates an appropriate answer.

[0102] Step 5:

[0103] The expert model returns the generated answer to the server.

[0104] Step 6:

[0105] The server sends the response to the user's device.

[0106] Step 7:

[0107] The user's device will display the response on the chat screen.

[0108] Search and browse functions

[0109] Step 1:

[0110] A user enters keywords into the search bar and presses the search button.

[0111] Step 2:

[0112] The user terminal sends the keyword to the server.

[0113] Step 3:

[0114] The server searches the database for anonymized consultation content based on the received keywords.

[0115] Step 4:

[0116] The server sends the search results to the user's device.

[0117] Step 5:

[0118] The user's device displays a list of search results.

[0119] Add to Favorites

[0120] Step 1:

[0121] The user presses the "favorite" button for a particular expert model.

[0122] Step 2:

[0123] The user terminal sends a favorite registration request to the server along with the session ID.

[0124] Step 3:

[0125] The server stores the favorites information in a database.

[0126] Step 4:

[0127] The server sends a confirmation message to the user terminal.

[0128] Step 5:

[0129] The user terminal displays a confirmation message.

[0130] Example 1

[0131] 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."

[0132] Conventional online consultation systems have faced challenges in that it is difficult for users to quickly obtain appropriate answers to their inquiries. In addition, searching for consultation content and accessing favorite expert models is cumbersome, resulting in a poor user experience.

[0133] 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.

[0134] In this invention, the server includes means for receiving input information from a user terminal, means for storing the received information in a database, means for verifying and authenticating login information of existing users, means for analyzing the received consultation content using a natural language processing algorithm, means for selecting an appropriate expert model based on the analyzed information, means for generating an appropriate answer based on the selected expert model, means for transmitting the generated answer to the user terminal, means for anonymizing and searching for the consultation content of other users, and means for displaying a list of search results. This allows users to quickly obtain appropriate answers and also makes it easier to search for consultation content and access their favorite expert models.

[0135] A "user terminal" is an electronic device such as a computer, smartphone, or tablet that is directly operated by a user.

[0136] "Input information" refers to data that a user provides to the system via their terminal, such as name, email address, password, and consultation details.

[0137] A "server" is a computer system that processes information received from user devices and manages and analyzes information using databases and algorithms.

[0138] A "session ID" is a unique identifier used to identify and manage a user's login status.

[0139] A "natural language processing algorithm" is a technology that analyzes the content of the consultation received, understands the text data, and generates an appropriate response.

[0140] An "expert model" is a program or system that has expert knowledge in a particular field (e.g., medicine, law, business) and generates appropriate answers to inquiries from users.

[0141] A "database" is a system for systematically storing digital information and for efficiently searching and managing it.

[0142] "Authentication" is the process of verifying that the user information provided is valid.

[0143] "Analysis" is the process of breaking down received data, interpreting its meaning, and extracting relevant information.

[0144] "Anonymization" is the process of removing or transforming personally identifiable information to protect privacy.

[0145] "Search results" are lists of relevant data extracted from a database based on a specific keyword or query.

[0146] The present invention provides a system that includes a user terminal, a server, and an expert model to quickly and appropriately answer various inquiries from users. Specific embodiments of the system will be described below.

[0147] User Registration and Login

[0148] When a user first uses the system, they enter information such as their name, email address, and password from their device and send it to the server. The server stores the received information in a database and sends a registration completion confirmation message to the user's device. Existing users enter their email address and password on the login screen and send it to the server. After the server verifies the information in the database, if authentication is successful, it generates a session ID and sends it to the user's device. This allows the user to access the home screen.

[0149] Individual chat consultation

[0150] When a user wants to ask a specific question, they enter the question content in text format on the chat screen of their device and press the send button. The question content and session ID are then sent to the server. The server analyzes the received question content using a natural language processing algorithm and selects the most appropriate expert model. The selected expert model generates an answer based on the received question content, and the server sends that answer to the user's device. This allows the user to quickly check the answer on the chat screen.

[0151] Search and browse functions

[0152] Users can search for past consultations by entering keywords in the search bar. This search request is sent from the user's device to the server, which then searches the database for anonymized consultations. The search results are displayed as a list on the user's device, allowing the user to check similar past consultations.

[0153] Add to Favorites

[0154] Users can register specific expert models as favorites. A favorite registration request is sent from the user terminal to the server, and the server stores it in the database. Users can easily access the registered expert models from the favorites list.

[0155] Specific examples

[0156] For example, a user might type a question on the chat screen, such as "I've been having really bad headaches lately. What should I do?" and press the send button. The device then sends this question and the session ID to the server. The server analyzes the consultation and passes it on to the medical expert model. The medical expert model generates an answer: "Possible causes of headaches include stress, eye strain, and lack of sleep. We recommend that you first get plenty of rest. If this does not improve, consult a doctor." and sends this answer to the user's device. The user can then check this answer on the chat screen.

[0157] As described above, the present invention provides a system that enables users to obtain prompt and appropriate answers to a variety of problems and questions.

[0158] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0159] Step 1:

[0160] User Registration

[0161] The user enters information such as name, email address, and password into the user device.

[0162] Input: Name, Email Address, Password

[0163] Output: Input information

[0164] The terminal transmits the input information to the server.

[0165] Input: Input information

[0166] Output: Data sent to the server

[0167] The server stores the received information in a database.

[0168] Input: Send data

[0169] Output: Information stored in the database

[0170] The server generates a confirmation message for completing the registration and sends it to the user terminal.

[0171] Input: Information stored in a database

[0172] Output: Confirmation message sent

[0173] Step 2:

[0174] Log in

[0175] The user enters their email address and password into the user terminal.

[0176] Input: Email address, password

[0177] Output: Input information

[0178] The terminal transmits the entered login information to the server.

[0179] Input: Input information

[0180] Output: Data sent to the server

[0181] The server checks the information against a database and authenticates it.

[0182] Input: Send data

[0183] Output: Authentication result

[0184] If the authentication is successful, the server generates a session ID and sends it to the user terminal.

[0185] Input: Authentication result

[0186] Output: Sending the session ID

[0187] The user device displays the home screen.

[0188] Input: Session ID

[0189] Output: Home screen display

[0190] Step 3:

[0191] Start individual chat consultation

[0192] The user enters the content of the consultation in text format on the chat screen and presses the send button.

[0193] Input: Consultation details

[0194] Output: Input consultation content

[0195] The terminal sends the consultation content and the session ID to the server.

[0196] Input: Consultation details, session ID

[0197] Output: Data sent to the server

[0198] Step 4:

[0199] Analysis of consultation content and generation of answers

[0200] The server analyzes the received consultation content using a natural language processing algorithm.

[0201] Input: Send data

[0202] Output: Analyzed consultation content

[0203] The server selects an appropriate expert model based on the analyzed consultation content.

[0204] Input: Analyzed consultation content

[0205] Output: Selected expert model

[0206] The expert model analyzes the content of the consultation and generates an appropriate answer.

[0207] Input: Analyzed consultation content

[0208] Output: The generated answer

[0209] Step 5:

[0210] Submitting and viewing answers

[0211] The server sends the generated response to the user terminal.

[0212] Input: Generated Answer

[0213] Output: Sending the answer

[0214] The user's device will display the response on the chat screen.

[0215] Input: Submit answer

[0216] Output: Display reply on chat screen

[0217] Step 6:

[0218] Search for consultation content

[0219] Users enter keywords into the search bar.

[0220] Input: Search keyword

[0221] Output: Input search keywords

[0222] The terminal sends a search request to the server.

[0223] Input: Enter search keywords

[0224] output: Search request to the server

[0225] The server searches the database for anonymized consultation content.

[0226] Input: Search request

[0227] Output: Search results

[0228] The server sends the search results to the user's terminal.

[0229] Input: Search results

[0230] output: sent to the user's terminal

[0231] The user's device displays a list of search results.

[0232] input:Submitted search results

[0233] output: Search results display

[0234] Step 7:

[0235] Registering and using favorites

[0236] Users can register specific expert models as "favorites."

[0237] Input: Favorite registration request

[0238] output: The input registration request

[0239] The terminal sends a favorites registration request to the server.

[0240] input:The input registration request

[0241] output: Send to server

[0242] The server stores the received requests in a database.

[0243] input:Submitted registration request

[0244] output: Save to database

[0245] Users access registered expert models from their favorites list.

[0246] input:Retrieve from database

[0247] output: Access to favorite models

[0248] (Application example 1)

[0249] 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."

[0250] Conventional food delivery systems have difficulty suggesting menus that take into account the user's health condition and nutritional balance, or providing expert advice in real time to individual questions. Furthermore, they lack the functionality to refer to frequently used menus or the health-related inquiries of other users, making them ineffective for self-resolving. There is a need to improve this situation.

[0251] 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.

[0252] In this invention, the server includes means for receiving input information from a user terminal, means for selecting an appropriate expert model based on the received information, means for generating an appropriate answer based on the selected expert model, means for transmitting the generated answer to the user terminal, means for analyzing the content of the consultation from the user terminal and selecting a food and nutrition expert model, means for generating a health menu based on the food and nutrition expert model, and means for presenting the generated menu to the user terminal. This allows users to receive real-time menu suggestions that take into account their health status and nutritional balance. Furthermore, users can search for other users' consultations anonymized and refer to menus generated based on similar past consultations, which can be used as a reference for self-solving.

[0253] "User terminal" refers to the communication device used by the user, including smartphones, tablets, and personal computers.

[0254] "Input Information" refers to text and selection information sent by a user through a user terminal.

[0255] "Server" refers to a computer system that receives, analyzes, and processes information from user terminals.

[0256] An "expert model" refers to an AI model with a knowledge base specialized in a particular field (e.g., medicine, law, nutrition, etc.).

[0257] A "food and nutrition expert model" refers to an AI model with specialized knowledge about food and nutrition.

[0258] "Menu" means the food options and suggestions provided to Users.

[0259] "Health menu" refers to meal plans and food suggestions generated taking into account the user's health status and nutritional balance.

[0260] "Anonymization" refers to the process of processing information so that a specific user cannot be identified.

[0261] "Consultation content" refers to questions or matters of consultation that users send to the system.

[0262] "Generated answers" refers to answers generated by the server based on an expert model.

[0263] "Session ID" refers to a unique identifier generated by the server after a user logs in.

[0264] "Display" refers to the process of visually presenting information on the screen of a user device.

[0265] "Search results" refers to the list of relevant information returned by the server in response to a user's search request.

[0266] "Favorites" refers to a feature that allows users to register specific information for easy access later.

[0267] The present invention is a system that uses a user terminal, a server, and an expert model to suggest menus that take into account the user's health condition and nutritional balance. A specific embodiment of this system is described below.

[0268] User Registration and Login

[0269] User Registration:

[0270] When a user first uses the system, they enter detailed information such as their name, email address, password, and dietary and allergy information from their device and send it to the server. This information is stored in the server's database. Once registration is complete, a confirmation message is returned to the user's device.

[0271] Login:

[0272] Existing users log in by entering their email address and password. The login information is sent to the server and checked against a database. If authentication is successful, a session ID is generated and returned to the user's device, allowing the user to access the home screen.

[0273] Individual chat consultation

[0274] Enter and submit your inquiry:

[0275] If a user wants to ask for advice about a specific dietary issue, they enter the content of their inquiry in text format on the chat screen of their device and press the send button. The content of their inquiry and the session ID are then sent to the server.

[0276] Analyzing the inquiry and generating an answer:

[0277] The server analyzes the received consultation content and selects the most appropriate food and nutrition expert model. The selected food and nutrition expert model analyzes the input consultation content and generates a health menu.

[0278] Submit and view your answers:

[0279] The generated menu is sent from the server to the user's device and displayed on the chat screen, allowing the user to quickly obtain a healthy menu.

[0280] For example, when a user inputs and sends a question such as "What healthy menu do you recommend for dinner tonight?", the server passes this information to the food and nutrition expert model. The food and nutrition expert model generates a menu that reads, "For a healthy dinner, we recommend steamed vegetables and chicken breast." This menu is sent to the user's device, where the user can view it on the chat screen.

[0281] Search and browse functions

[0282] Search for consultation content:

[0283] Users can search for past consultations with other users by entering specific keywords in the search bar. This search request is sent to the server, which returns the consultations in the database in anonymized form.

[0284] Displaying search results:

[0285] The search results sent from the server are displayed as a list on the user's device, and the user can view past similar consultations. This allows users to refer to information on cases similar to their own and use it as a reference for self-solving.

[0286] Add to Favorites

[0287] Favorite settings:

[0288] If a user likes a particular food nutrition expert model or generated menu, they can register it by pressing the "Favorite" button. The request to register as a favorite is sent to the server and stored in the database.

[0289] Use Favorites:

[0290] Registered favorite models and menus can be easily accessed on the user's device, and if the user wants to access them again, they can be quickly recalled, improving convenience.

[0291] This system mainly uses the following hardware and software.

[0292] Hardware:

[0293] Server: Cloud server (e.g. AWS, Google Cloud, etc.)

[0294] User device: Smartphone (iOS / Android)

[0295] software:

[0296] Framework: Flask (Python)

[0297] Database: SQLite

[0298] AI model: Language model (e.g., OpenAI's GPT)

[0299] For example, if a user inputs "What low-calorie menu would you recommend for today's lunch?", the food and nutrition expert model will generate a menu that reads, "Steamed chicken salad and smoothies are recommended as low-calorie menu items." This menu is sent to the user's device, and the user can view it on the chat screen. Examples of prompt sentences are as follows:

[0300] User: "What's a low-calorie option for lunch today?"

[0301] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0302] Step 1:

[0303] The user sends input information such as name, email address, password, and dietary and allergy information from the user device. The server receives this information and stores it in a database. Input: User information. Output: User information stored in the database.

[0304] Step 2:

[0305] An existing user logs in with their email address and password. The login information sent from the user's device to the server is checked against the database. If authentication is successful, the server generates a session ID and returns it to the user's device. Input: Login information. Output: Session ID.

[0306] Step 3:

[0307] The user enters the consultation content in text format on the chat screen of the user device and presses the send button. This sends the consultation content and session ID to the server. Input: Consultation content and session ID. Output: Data sent to the server.

[0308] Step 4:

[0309] The server analyzes the received consultation content and selects the most appropriate food and nutrition expert model. The selected model performs analysis based on the input consultation content. Input: Consultation content. Output: Selected food and nutrition expert model.

[0310] Step 5:

[0311] The selected food and nutrition expert model generates a health-related menu based on the consultation. For example, if a user inputs "What healthy menu would you recommend for dinner today?", the model generates a menu that says "For a healthy dinner, I recommend steamed vegetables and chicken breast." Input: Consultation. Output: Generated menu.

[0312] Step 6:

[0313] The generated menu is sent from the server to the user's device and displayed on the chat screen. Input: Generated menu. Output: Menu sent to the user's device.

[0314] Step 7:

[0315] If the user likes a particular food nutrition expert model or generated menu, they can register it by pressing the "Favorite" button. This favorite registration request is sent to the server and stored in the database. Input: Favorite registration request. Output: Favorite information stored in the database.

[0316] Step 8:

[0317] When the user wants to access a favorite item again, the server extracts the item from the database based on the request from the user's device and sends it to the user's device. Input: Request from the user's device. Output: Extracted item.

[0318] Step 9:

[0319] A user enters a specific keyword in the search bar to search for past consultations from other users. This search request is sent to the server, and the consultations in the database are anonymized and returned as search results. Input: Keyword. Output: Anonymized search results.

[0320] Step 10:

[0321] The search results sent from the server are displayed as a list on the user's device. The user can view past similar consultations. Input: Search results. Output: List displayed on the user's device.

[0322] Example prompt sentence:

[0323] User: "What's a low-calorie option for lunch today?"

[0324] 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.

[0325] The present invention uses a system that combines a user terminal, a server, an expert model, and an emotion engine to provide quick and appropriate answers to various inquiries users have in their daily lives. A specific embodiment of this system will be described below.

[0326] User Registration and Login

[0327] User Registration:

[0328] When a user uses the app for the first time, they enter their personal information (name, email address, password) to register. This information is sent from the user's device to the server and stored in a database. Once registration is complete, a confirmation message is returned to the user's device.

[0329] Login:

[0330] Existing users log in by entering their email address and password. The entered login information is sent from the user's device to the server and checked against the database. If authentication is successful, a session ID is generated and returned to the user's device, allowing the user to access the home screen.

[0331] Individual chat consultation

[0332] Enter and submit your inquiry:

[0333] When a user wants to ask a specific question, they enter the question in text format on the chat screen of their device and press the send button. The question and session ID are then sent to the server.

[0334] Emotion Recognition with Emotion Engine:

[0335] When the server receives the consultation content, it analyzes the user's emotions using an emotion engine. Based on the analysis results, it selects an appropriate expert model.

[0336] Analyzing the inquiry and generating an answer:

[0337] The selected expert model analyzes the content of the consultation and the emotional information from the emotion engine to generate an appropriate answer. This analysis allows for more personalized answers based on the user's emotions.

[0338] Submit and view your answers:

[0339] The generated answers are sent from the server to the user's device and displayed on the chat screen, allowing the user to quickly obtain the information they need.

[0340] Search and browse functions

[0341] Search for consultation content:

[0342] Users can search for past consultations with other users by entering specific keywords in the search bar. This search request is sent from the user's device to the server, and the consultations in the database are anonymized and returned as search results.

[0343] Displaying search results:

[0344] The search results sent from the server are displayed as a list on the user's device, and the user can view past similar consultations. This allows users to refer to information on cases similar to their own and use it as a reference for self-solving.

[0345] Add to Favorites

[0346] Favorite settings:

[0347] If a user likes a particular expert model, they can register it by pressing the "Favorite" button. The favorite registration request is sent from the user's terminal to the server and stored in the database.

[0348] Use Favorites:

[0349] Registered favorite expert models can be easily accessed on the user's device, and can be quickly recalled if the user wishes to access them again, improving convenience.

[0350] Specific examples

[0351] For example, consider the case of a medical consultation. When a user inputs and sends the question, "I've been having really bad headaches lately. What should I do?", the server passes this content to the emotion engine. The emotion engine recognizes the user's emotion as "anxiety" and passes this information along with the medical expert model. The medical expert model takes "anxiety" into consideration and generates a response such as, "Possible causes of headaches include stress, eye strain, and lack of sleep. We recommend that you first get plenty of rest. If your symptoms persist, consult a doctor." This response is sent to the user's device and can be viewed by the user on the chat screen.

[0352] As described above, the present invention realizes a system that provides more accurate and human-like answers by taking the user's emotions into consideration.

[0353] The processing flow will be explained below.

[0354] User Registration and Login

[0355] New Registration

[0356] Step 1:

[0357] The user launches the app and is presented with a sign-up form.

[0358] Step 2:

[0359] The user's device prompts them to enter their name, email address, and password, and then the user presses the "Send" button.

[0360] Step 3:

[0361] The user terminal sends the input information to the server.

[0362] Step 4:

[0363] The server stores the received information in a database.

[0364] Step 5:

[0365] The server generates a registration success message and sends it to the user terminal.

[0366] Log in

[0367] Step 1:

[0368] The user enters their email address and password on the login screen and clicks the "Login" button.

[0369] Step 2:

[0370] The user terminal sends the input information to the server.

[0371] Step 3:

[0372] The server checks the received login information against its database.

[0373] Step 4:

[0374] If the server is successful in the authentication, it generates a session ID and sends it to the user's terminal.

[0375] Step 5:

[0376] The user device saves the session ID and displays the home screen.

[0377] Individual chat consultation

[0378] Step 1:

[0379] The user accesses the consultation screen, enters the consultation content in text format, and presses the send button.

[0380] Step 2:

[0381] The user device sends the consultation content and session ID to the server.

[0382] Step 3:

[0383] The server receives the consultation content and transfers it to the emotion engine.

[0384] Step 4:

[0385] The emotion engine analyzes the content of the consultation and recognizes the user's emotions (e.g., "anxiety," "impatience," etc.).

[0386] Step 5:

[0387] The server receives the analysis results of the emotion engine and selects an appropriate expert model.

[0388] Step 6:

[0389] The expert model generates an appropriate answer based on the consultation content and emotional information.

[0390] Step 7:

[0391] The expert model returns the generated answer to the server.

[0392] Step 8:

[0393] The server sends the response to the user's device.

[0394] Step 9:

[0395] The user's device will display the response on the chat screen.

[0396] Search and browse functions

[0397] Step 1:

[0398] A user enters keywords into the search bar and presses the search button.

[0399] Step 2:

[0400] The user terminal sends the keyword to the server.

[0401] Step 3:

[0402] The server searches the database for anonymized consultation content based on the received keywords.

[0403] Step 4:

[0404] The server sends the search results to the user's device.

[0405] Step 5:

[0406] The user's device displays a list of search results.

[0407] Add to Favorites

[0408] Step 1:

[0409] The user presses the "favorite" button for a particular expert model.

[0410] Step 2:

[0411] The user terminal sends a "favorite" registration request to the server along with the session ID.

[0412] Step 3:

[0413] The server stores the "favorites" information in a database.

[0414] Step 4:

[0415] The server sends a confirmation message to the user terminal.

[0416] Step 5:

[0417] The user terminal displays a confirmation message.

[0418] Example: Medical consultation

[0419] Step 1:

[0420] The user logs in, accesses the home screen, and selects "Medical Consultation."

[0421] Step 2:

[0422] The user enters the question they want to ask, such as "I've been having terrible headaches lately. What should I do?" and presses the send button.

[0423] Step 3:

[0424] The user device sends the consultation content and session ID to the server.

[0425] Step 4:

[0426] The server transfers the consultation content to the emotion engine, which then recognizes the emotion "anxiety."

[0427] Step 5:

[0428] The server receives the analysis results of the emotion engine ("anxiety") and selects a medical expert model.

[0429] Step 6:

[0430] The medical expert model analyzes the consultation content while taking into account "anxiety" and generates the following answer: "Possible causes of headaches include stress, eye strain, and lack of sleep. We recommend that you first get plenty of rest. If your symptoms do not improve, please consult a doctor."

[0431] Step 7:

[0432] The medical expert model returns the generated answer to the server.

[0433] Step 8:

[0434] The server sends the response to the user's device.

[0435] Step 9:

[0436] The user's device will display the response on the chat screen.

[0437] By going through these processing steps, users can quickly receive an appropriate answer that corresponds to their feelings.

[0438] Example 2

[0439] 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."

[0440] Conventional expert systems provide answers without considering the user's feelings, resulting in insufficient appropriate responses. Furthermore, they lacked the functionality to efficiently search and view inquiries from multiple users, making it difficult for users to access similar past cases to resolve their own issues. Furthermore, it was difficult for users to call their favorite experts again.

[0441] 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.

[0442] In this invention, the server includes a means for using an emotion analysis engine that analyzes emotions based on user input information, a means for personalizing answers based on the emotion analysis results, and a means for anonymizing and searching for consultation details from other users. This enables the provision of more accurate and personalized answers that take the user's emotions into consideration. Furthermore, by efficiently searching and browsing similar past consultation details, users can easily obtain the information they need to resolve their issues. Furthermore, by providing a function that allows users to register their favorite experts as favorites and easily call them up, user convenience can be improved.

[0443] "User terminal" means a device through which a user accesses the system and enters or receives information.

[0444] "Input information" refers to questions, inquiries, or other data provided by users to the system.

[0445] A "server" is a central computer system that processes information sent from user terminals and performs the necessary analysis and responses.

[0446] An "expert model" is an algorithm or program with specific expertise that provides appropriate answers to user questions.

[0447] The "means for generating an answer" is a mechanism for constructing an appropriate answer to the user's input information using an expert model.

[0448] An "emotion analysis engine" is an analysis system for reading emotions from user input information.

[0449] "Anonymization" is the process of processing information so that individuals cannot be identified.

[0450] A "database" is a collection of information that stores user information, consultation details, etc. in an organized manner and can be searched or retrieved as needed.

[0451] "Favorites" is a feature that allows users to save specific expert models for easy access again.

[0452] "Personalization" means optimizing services and information according to the individual characteristics and emotions of each user.

[0453] The present invention uses a system that combines a user terminal, a server, an expert model, and an emotion analysis engine to provide quick and appropriate answers to various inquiries users have in their daily lives. Specific embodiments of this system are described below.

[0454] User Registration and Login

[0455] User Registration

[0456] When a user uses the system for the first time, they enter their name, email address, and password to register. This information is sent from the user's device to the server, which stores the received information in a database. Once registration is complete, the server sends a confirmation message to the user's device.

[0457] Log in

[0458] Existing users log in by entering their email address and password. This login information is sent from the user's device to the server, which checks the information against a database. If authentication is successful, the server generates a session ID and sends it to the user's device. The user's device retains the received session ID and displays the home screen.

[0459] Individual chat consultation

[0460] Enter and submit your consultation details

[0461] When a user wants to ask a specific question, they enter the question in text format on the chat screen of their device and press the send button. The question and session ID are then sent from the user device to the server.

[0462] Emotion recognition by emotion engine

[0463] The server passes the received consultation content to the emotion analysis engine, which analyzes the user's emotions and returns the results to the server, which selects an appropriate expert model based on the emotion analysis results.

[0464] Analysis of consultation content and generation of answers

[0465] The selected expert model analyzes the consultation content and emotional information and generates an appropriate answer that reflects the results of the emotional analysis, making it more personalized for the user.

[0466] Submitting and viewing answers

[0467] The generated answers are sent from the server to the user's device and displayed on the chat screen, allowing the user to quickly obtain the information they need.

[0468] Search and browse functions

[0469] Search for consultation content

[0470] Users can search for past consultations with other users by entering specific keywords in the search bar. This search request is sent from the user's device to the server, which then anonymizes the consultations in the database and returns them as search results.

[0471] Displaying search results

[0472] The search results are sent from the server to the user's device and displayed as a list. Users can then browse similar past consultations and find information on cases similar to their own.

[0473] Add to Favorites

[0474] Favorites Settings

[0475] If a user likes a particular expert model, he or she can register it by pressing the "Favorite" button. This favorite registration request is sent from the user terminal to the server, and the server stores it in a database.

[0476] Using Favorites

[0477] Saved favorite expert models can be easily accessed on the user's device, and can be quickly recalled when the user wants to access them again, improving convenience.

[0478] Specific examples

[0479] For example, consider the case where a user seeks medical advice. When the user enters and submits the question, "I've been having really bad headaches lately. What should I do?", the server passes this information to the emotion analysis engine. The emotion analysis engine recognizes the user's emotion as "anxiety." The server then sends this information to the medical expert model. Taking "anxiety" into consideration, the medical expert model generates the following response: "Possible causes of headaches include stress, eye strain, and lack of sleep. We recommend that you first get plenty of rest. If your symptoms persist, consult a doctor." This response is then sent from the server to the user's device, where the user can view it on the chat screen.

[0480] Prompt Sentence Examples

[0481] User: I've been having terrible headaches lately, what should I do?

[0482] AI: Headaches can be caused by stress, eye strain, lack of sleep, etc. First of all, we recommend getting plenty of rest. If your symptoms persist, consult your doctor.

[0483] The system also takes the user's emotions into account to provide more accurate and human-like answers.

[0484] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0485] Program processing flow

[0486] User Registration and Login

[0487] Step 1:

[0488] The user enters new registration information.

[0489] Enter your name, email address, and password.

[0490] How it works: The user enters the required information into the application's registration screen.

[0491] Output: Input data (name, email address, password).

[0492] Step 2:

[0493] The terminal sends the registration information to the server.

[0494] Input: Input data (name, email address, password).

[0495] Operation: The user terminal sends input data to the server.

[0496] Output: User registration information sent to the server.

[0497] Step 3:

[0498] The server stores the user information in a database.

[0499] Input: User registration information sent to the server.

[0500] How it works: The server stores the received information in a database.

[0501] Output: Status of user information saved to database.

[0502] Step 4:

[0503] The server sends a registration completion message to the terminal.

[0504] Input: Status of saving user information to database.

[0505] Operation: The server generates and sends a confirmation message to the user device indicating that the registration process is complete.

[0506] Output: Sending a registration completion message to the user's terminal.

[0507] Step 5:

[0508] The user enters their login information.

[0509] Enter your email address and password.

[0510] How it works: A user enters their email address and password on a login screen.

[0511] Output: Input data (email address, password).

[0512] Step 6:

[0513] The device sends the login information to the server.

[0514] Input: Input data (email address, password).

[0515] How it works: The user device sends login information to the server.

[0516] Output: The login information sent to the server.

[0517] Step 7:

[0518] The server checks the information against a database.

[0519] Input: The login information sent to the server.

[0520] How it works: The server checks the login information it receives against its database.

[0521] Output: Authentication success / failure result.

[0522] Step 8:

[0523] The server sends the authentication result and session ID to the terminal.

[0524] Input: The result of successful authentication.

[0525] How it works: If authentication is successful, the server generates a session ID and sends it to the user's device.

[0526] Output: A session ID is generated and sent to the user's device.

[0527] Step 9:

[0528] The device displays the home screen.

[0529] Input: Session ID.

[0530] Operation: The user device retains the received session ID and displays the home screen.

[0531] Output: Home screen display.

[0532] Individual chat consultation

[0533] Step 1:

[0534] The user inputs the consultation content.

[0535] Input: Consultation details.

[0536] How it works: The user enters the content of their inquiry in text format on the chat screen.

[0537] Output: The consultation content entered.

[0538] Step 2:

[0539] The device sends the consultation content and session ID to the server.

[0540] Input: Session ID, consultation details.

[0541] Operation: The user terminal sends the consultation content and session ID to the server.

[0542] Output: The consultation details and session ID sent to the server.

[0543] Step 3:

[0544] The server passes the consultation content to the emotion engine.

[0545] Input: The consultation content sent to the server.

[0546] Operation: The server passes the received consultation content to the emotion analysis engine.

[0547] Output: Transferring the consultation content to the emotion engine.

[0548] Step 4:

[0549] The emotion engine analyzes the user's emotions.

[0550] Input: Consultation details.

[0551] How it works: The emotion engine analyzes the content of the consultation and recognizes the user's emotions.

[0552] Output: Emotion analysis results.

[0553] Step 5:

[0554] The server obtains the emotion analysis results.

[0555] Input: Sentiment analysis results.

[0556] Operation: The server obtains the analysis results from the emotion engine.

[0557] Output: The server receives the sentiment analysis results.

[0558] Step 6:

[0559] The server selects an expert model.

[0560] Input: Sentiment analysis results.

[0561] Operation: The server selects an appropriate expert model based on the sentiment analysis results.

[0562] Output: Selected expert model.

[0563] Step 7:

[0564] The server passes the consultation content to the expert model.

[0565] Input: Consultation content, emotional information.

[0566] Operation: The server passes the consultation content and emotion information to the expert model.

[0567] Output: Transfer of input data to the expert model.

[0568] Step 8:

[0569] The expert model generates the answer.

[0570] Input: Consultation content, emotional information.

[0571] How it works: The expert model analyzes the consultation content and emotional information and generates an appropriate answer.

[0572] Output: The generated answer.

[0573] Step 9:

[0574] The server sends the generated response to the terminal.

[0575] Input: The generated answer.

[0576] Operation: The server generates a response and sends it to the user's device.

[0577] Output: Sending the answer to the user's terminal.

[0578] Step 10:

[0579] The device will display the response on the chat screen.

[0580] Input:Answer.

[0581] Operation: The user's device displays the received reply on the chat screen.

[0582] Output: The response displayed in the chat window.

[0583] Add to Favorites

[0584] Step 1:

[0585] The user presses the "Favorite" button.

[0586] Input: None.

[0587] How it works: A user who likes a particular expert model presses the "favorite" button.

[0588] Output: Favorite button click event.

[0589] Step 2:

[0590] The terminal sends a favorites registration request to the server.

[0591] Input: Favorite button click event.

[0592] Operation: The device sends a favorite registration request to the server.

[0593] Output: The favorite request sent to the server.

[0594] Step 3:

[0595] The server stores the favorites information in a database.

[0596] Input: The favorite request sent to the server.

[0597] How it works: The server stores information in a database based on the favorites request.

[0598] Output: Status of saving favorite information to database.

[0599] Step 4:

[0600] The device displays registered expert models.

[0601] Input: None.

[0602] How it works: Saved favorite expert models are easily accessible on the user's device.

[0603] Output: Display of favorite expert model on user's device.

[0604] (Application example 2)

[0605] 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."

[0606] Conventional customer support systems in virtual stores respond uniformly without considering the individual emotional state of each user, making it impossible to provide personalized, appropriate support to users. As a result, users often do not receive appropriate support, leaving them frustrated and confused. Furthermore, conventional systems lacked functionality to easily search for past user inquiries, making it difficult to quickly obtain information for self-resolving.

[0607] 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.

[0608] In this invention, the server includes means for receiving input information from a user terminal, means for selecting an appropriate expert model based on the received information, means for analyzing the user's emotions using emotion analysis means, means for selecting an expert model based on the analyzed emotion information, means for generating an appropriate answer based on the selected expert model, and means for transmitting the generated answer to the user terminal, thereby enabling personalized and appropriate support according to the user's emotional state.

[0609] "User terminal" means an electronic device used by a user, including a smartphone, tablet, or PC.

[0610] "Input information" refers to various data sent by a user via a user terminal, including text, images, audio, etc.

[0611] An "expert model" is an algorithm or program that uses knowledge in a specific field to generate appropriate answers to users' questions or inquiries.

[0612] "Emotion analysis means" refers to a system or algorithm that analyzes input information from a user and recognizes their emotional state.

[0613] "Emotion information" refers to data relating to the user's emotional state obtained by emotion analysis means.

[0614] The "selection means" refers to a method or mechanism for selecting the most appropriate model from among multiple expert models based on the user's input information and emotion information.

[0615] An "answer generator" is an algorithm or program that generates an appropriate answer to a user's question or inquiry based on a selected expert model.

[0616] "Transmitting means" refers to a system or method for sending the generated answer to the user terminal.

[0617] An embodiment of the present invention will be described.

[0618] As an application example of the invention, we implement a customer support system in a virtual store. This system mainly uses a user terminal, a server, a sentiment analysis means, and an expert model. The specific hardware used includes a smartphone, and the software used includes AWS Lambda, Amazon DynamoDB, and the OpenAI GPT-3 model.

[0619] First, a user accesses the application using a user device such as a smartphone, registers as a user if necessary, and logs in. When registering, the user enters their name, email address, and password, which are sent to the server. The server receives this information and stores it in Amazon DynamoDB. When logging in, the email address and password are verified, and if authentication is successful, a session ID is generated and returned to the user.

[0620] While browsing products in a virtual store, a user inputs and submits a specific inquiry in text format. The user's device sends this input information to the server. When the server receives the input information, it activates the emotion analysis means to recognize the user's emotional state. This emotion analysis uses the OpenAI GPT-3 model. For example, if the input inquiry is a question such as "Is this refrigerator energy efficient?", the emotion analysis means recognizes this as an emotion that indicates a desire for information.

[0621] Next, based on the analyzed emotional information, the server selects the most appropriate expert model. The selected expert model generates an appropriate answer based on the user's emotional state and the content of the inquiry. For example, if the question is about energy efficiency, a technical expert model will be selected and generate an answer that explains the details of energy efficiency. This generated answer is then sent back to the user's device via the server.

[0622] A specific example is sentiment analysis and answer generation using the following prompt sentence:

[0623] Example of a prompt for sentiment analysis:

[0624] Choose an expert model by analyzing the sentiment of the message: 'Is this refrigerator energy efficient?'

[0625] Example prompts for answer generation:

[0626] General Breakdown Technician's Answer: Provide a good answer to 'Is this refrigerator energy efficient?'

[0627] This system allows users to receive personalized support in real time based on their emotions. It also allows users to anonymously search for other inquiries and gain insights from similar past inquiries. This results in higher user satisfaction and more effective self-resolution than traditional customer support systems.

[0628] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0629] Step 1:

[0630] Receive input information from the user's device

[0631] The user uses a dedicated application in the virtual store to input and send the details of their inquiry in text format. The user's device sends this input information to the server. The input information includes the user ID and session ID.

[0632] Step 2:

[0633] Emotion analysis using emotion analysis tools

[0634] The server analyzes the user's emotions using the OpenAI GPT-3 model based on the received input information. The server sends the user's consultation details as input data to GPT-3 and receives analyzed emotional information. This emotional information indicates emotional states such as "anxiety," "doubt," and "joy."

[0635] Step 3:

[0636] Expert model selection

[0637] The server selects the most appropriate expert model based on the emotion information obtained from the emotion analysis means. For example, if the user's emotion is analyzed as "question," an expert model that is good at answering technical questions will be selected. In this selection process, an algorithm determines the most appropriate model from multiple expert models.

[0638] Step 4:

[0639] Generate answers

[0640] The server uses the selected expert model to generate an answer based on the input information. A generative AI model is used to create a specific answer based on the consultation content. This model generates an answer that is appropriate for the user's question based on the prompt text.

[0641] Step 5:

[0642] Submit your answer

[0643] The generated answers are sent from the server to the user's device, and are displayed in the user's application, allowing the user to obtain appropriate information in real time.

[0644] Specifically, the reply text is sent to the user's terminal and displayed on the chat screen of the application.

[0645] Step 6:

[0646] Search function for past consultation content (optional)

[0647] Users can search for past consultations by entering specific keywords in the search bar within the application. The search request sent from the user's device is processed by the server, and related consultations are searched for in an anonymized form. The search results are sent back to the user's device and displayed in a list within the application.

[0648] Step 7:

[0649] Register your favorite expert models (optional)

[0650] Users can register their favorite expert models as favorites. A request to add to favorites is sent from the user's device to the server, and the server stores the information in a database, allowing users to instantly access specific expert models.

[0651] 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.

[0652] 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.

[0653] 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.

[0654] [Second embodiment]

[0655] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0656] 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.

[0657] 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).

[0658] 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.

[0659] 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.

[0660] 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).

[0661] 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.

[0662] 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.

[0663] 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.

[0664] 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.

[0665] 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.

[0666] 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."

[0667] The present invention uses a system consisting of a user terminal, a server, and an expert model to provide prompt and appropriate answers to various inquiries in the user's daily life. A specific embodiment of this system will be described below.

[0668] User Registration and Login

[0669] User Registration:

[0670] When a user uses the app for the first time, they enter their personal information (name, email address, password) to register. This information is sent from the user's device to the server and stored in a database. Once registration is complete, a confirmation message is returned to the user's device.

[0671] Login:

[0672] Existing users log in by entering their email address and password. The entered login information is sent from the user's device to the server and checked against the database. If authentication is successful, a session ID is generated and returned to the user's device, allowing the user to access the home screen.

[0673] Individual chat consultation

[0674] Enter and submit your inquiry:

[0675] When a user wants to ask a specific question, they enter the question in text format on the chat screen of their device and press the send button. The question and session ID are then sent to the server.

[0676] Analyzing the inquiry and generating an answer:

[0677] The server analyzes the received consultation content and selects the most appropriate expert model (medical, legal, business, etc.). The selected expert model analyzes the input consultation content and generates an appropriate answer.

[0678] Submit and view your answers:

[0679] The generated answers are sent from the server to the user's device and displayed on the chat screen, allowing the user to quickly obtain the information they need.

[0680] Search and browse functions

[0681] Search for consultation content:

[0682] Users can search for past consultations with other users by entering specific keywords in the search bar. This search request is sent from the user's device to the server, and the consultations in the database are anonymized and returned as search results.

[0683] Displaying search results:

[0684] The search results sent from the server are displayed as a list on the user's device, and the user can view past similar consultations. This allows users to refer to information on cases similar to their own and use it as a reference for self-solving.

[0685] Add to Favorites

[0686] Favorite settings:

[0687] If a user likes a particular expert model, they can register it by pressing the "Favorite" button. The favorite registration request is sent from the user's terminal to the server and stored in the database.

[0688] Use Favorites:

[0689] Registered favorite expert models can be easily accessed on the user's device, and can be quickly recalled if the user wishes to access them again, improving convenience.

[0690] Specific examples

[0691] For example, consider the case of a medical consultation. When a user inputs and sends the question, "I've been having terrible headaches lately. What should I do?", the server passes this information to the medical expert model. The medical expert model generates a response such as, "Possible causes of headaches include stress, eye strain, and lack of sleep. We recommend that you first get plenty of rest. If this does not improve, please consult a doctor." This response is sent to the user's device, where the user can view it on the chat screen.

[0692] As described above, the present invention provides a system that enables users to obtain prompt and appropriate answers to a variety of problems and questions.

[0693] The processing flow will be explained below.

[0694] User Registration and Login

[0695] New Registration

[0696] Step 1:

[0697] The user launches the app and sees the signup form.

[0698] Step 2:

[0699] The user's device prompts them to enter their name, email address, and password, then presses the "Send" button.

[0700] Step 3:

[0701] The user terminal sends the input information to the server.

[0702] Step 4:

[0703] The server stores the received information in a database.

[0704] Step 5:

[0705] The server returns a registration success message to the user terminal.

[0706] Log in

[0707] Step 1:

[0708] The user enters their email address and password on the login screen and clicks the "Login" button.

[0709] Step 2:

[0710] The user terminal sends the input information to the server.

[0711] Step 3:

[0712] The server checks the received information against a database.

[0713] Step 4:

[0714] If the server matches, it generates a session ID and sends it to the user's device.

[0715] Step 5:

[0716] The user device saves the session ID and displays the home screen.

[0717] Individual chat consultation

[0718] Step 1:

[0719] The user enters the consultation content in text on the app's consultation screen and presses the send button.

[0720] Step 2:

[0721] The user device sends the consultation content and session ID to the server.

[0722] Step 3:

[0723] The server analyzes the received consultation content and selects an appropriate expert model.

[0724] Step 4:

[0725] The expert model analyzes the consultation content and generates an appropriate answer.

[0726] Step 5:

[0727] The expert model returns the generated answer to the server.

[0728] Step 6:

[0729] The server sends the response to the user's device.

[0730] Step 7:

[0731] The user's device will display the response on the chat screen.

[0732] Search and browse functions

[0733] Step 1:

[0734] A user enters keywords into the search bar and presses the search button.

[0735] Step 2:

[0736] The user terminal sends the keyword to the server.

[0737] Step 3:

[0738] The server searches the database for anonymized consultation content based on the received keywords.

[0739] Step 4:

[0740] The server sends the search results to the user's device.

[0741] Step 5:

[0742] The user's device displays a list of search results.

[0743] Add to Favorites

[0744] Step 1:

[0745] The user presses the "favorite" button for a particular expert model.

[0746] Step 2:

[0747] The user terminal sends a favorite registration request to the server along with the session ID.

[0748] Step 3:

[0749] The server stores the favorites information in a database.

[0750] Step 4:

[0751] The server sends a confirmation message to the user terminal.

[0752] Step 5:

[0753] The user terminal displays a confirmation message.

[0754] Example 1

[0755] 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."

[0756] Conventional online consultation systems have faced challenges in that it is difficult for users to quickly obtain appropriate answers to their inquiries. In addition, searching for consultation content and accessing favorite expert models is cumbersome, resulting in a poor user experience.

[0757] 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.

[0758] In this invention, the server includes means for receiving input information from a user terminal, means for storing the received information in a database, means for verifying and authenticating login information of existing users, means for analyzing the received consultation content using a natural language processing algorithm, means for selecting an appropriate expert model based on the analyzed information, means for generating an appropriate answer based on the selected expert model, means for transmitting the generated answer to the user terminal, means for anonymizing and searching for the consultation content of other users, and means for displaying a list of search results. This allows users to quickly obtain appropriate answers and also makes it easier to search for consultation content and access their favorite expert models.

[0759] A "user terminal" is an electronic device such as a computer, smartphone, or tablet that is directly operated by a user.

[0760] "Input information" refers to data that a user provides to the system via their terminal, such as name, email address, password, and consultation details.

[0761] A "server" is a computer system that processes information received from user devices and manages and analyzes information using databases and algorithms.

[0762] A "session ID" is a unique identifier used to identify and manage a user's login status.

[0763] A "natural language processing algorithm" is a technology that analyzes the content of the consultation received, understands the text data, and generates an appropriate response.

[0764] An "expert model" is a program or system that has expert knowledge in a particular field (e.g., medicine, law, business) and generates appropriate answers to inquiries from users.

[0765] A "database" is a system for systematically storing digital information and for efficiently searching and managing it.

[0766] "Authentication" is the process of verifying that the user information provided is valid.

[0767] "Analysis" is the process of breaking down received data, interpreting its meaning, and extracting relevant information.

[0768] "Anonymization" is the process of removing or transforming personally identifiable information to protect privacy.

[0769] "Search results" are lists of relevant data extracted from a database based on a specific keyword or query.

[0770] The present invention provides a system that includes a user terminal, a server, and an expert model to quickly and appropriately answer various inquiries from users. Specific embodiments of the system will be described below.

[0771] User Registration and Login

[0772] When a user first uses the system, they enter information such as their name, email address, and password from their device and send it to the server. The server stores the received information in a database and sends a registration completion confirmation message to the user's device. Existing users enter their email address and password on the login screen and send it to the server. After the server verifies the information in the database, if authentication is successful, it generates a session ID and sends it to the user's device. This allows the user to access the home screen.

[0773] Individual chat consultation

[0774] When a user wants to ask a specific question, they enter the question content in text format on the chat screen of their device and press the send button. The question content and session ID are then sent to the server. The server analyzes the received question content using a natural language processing algorithm and selects the most appropriate expert model. The selected expert model generates an answer based on the received question content, and the server sends that answer to the user's device. This allows the user to quickly check the answer on the chat screen.

[0775] Search and browse functions

[0776] Users can search for past consultations by entering keywords in the search bar. This search request is sent from the user's device to the server, which then searches the database for anonymized consultations. The search results are displayed as a list on the user's device, allowing the user to check similar past consultations.

[0777] Add to Favorites

[0778] Users can register specific expert models as favorites. A favorite registration request is sent from the user terminal to the server, and the server stores it in the database. Users can easily access the registered expert models from the favorites list.

[0779] Specific examples

[0780] For example, a user might type a question on the chat screen, such as "I've been having really bad headaches lately. What should I do?" and press the send button. The device then sends this question and the session ID to the server. The server analyzes the consultation and passes it on to the medical expert model. The medical expert model generates an answer: "Possible causes of headaches include stress, eye strain, and lack of sleep. We recommend that you first get plenty of rest. If this does not improve, consult a doctor." and sends this answer to the user's device. The user can then check this answer on the chat screen.

[0781] As described above, the present invention provides a system that enables users to obtain prompt and appropriate answers to a variety of problems and questions.

[0782] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0783] Step 1:

[0784] User Registration

[0785] The user enters information such as name, email address, and password into the user device.

[0786] Input: Name, Email Address, Password

[0787] Output: Input information

[0788] The terminal transmits the input information to the server.

[0789] Input: Input information

[0790] Output: Data sent to the server

[0791] The server stores the received information in a database.

[0792] Input: Send data

[0793] Output: Information stored in the database

[0794] The server generates a confirmation message for completing the registration and sends it to the user terminal.

[0795] Input: Information stored in a database

[0796] Output: Confirmation message sent

[0797] Step 2:

[0798] Log in

[0799] The user enters their email address and password into the user terminal.

[0800] Input: Email address, password

[0801] Output: Input information

[0802] The terminal transmits the entered login information to the server.

[0803] Input: Input information

[0804] Output: Data sent to the server

[0805] The server checks the information against a database and authenticates it.

[0806] Input: Send data

[0807] Output: Authentication result

[0808] If the authentication is successful, the server generates a session ID and sends it to the user terminal.

[0809] Input: Authentication result

[0810] Output: Sending the session ID

[0811] The user device displays the home screen.

[0812] Input: Session ID

[0813] Output: Home screen display

[0814] Step 3:

[0815] Start individual chat consultation

[0816] The user enters the content of the consultation in text format on the chat screen and presses the send button.

[0817] Input: Consultation details

[0818] Output: Input consultation content

[0819] The terminal sends the consultation content and the session ID to the server.

[0820] Input: Consultation details, session ID

[0821] Output: Data sent to the server

[0822] Step 4:

[0823] Analysis of consultation content and generation of answers

[0824] The server analyzes the received consultation content using a natural language processing algorithm.

[0825] Input: Send data

[0826] Output: Analyzed consultation content

[0827] The server selects an appropriate expert model based on the analyzed consultation content.

[0828] Input: Analyzed consultation content

[0829] Output: Selected expert model

[0830] The expert model analyzes the content of the consultation and generates an appropriate answer.

[0831] Input: Analyzed consultation content

[0832] Output: The generated answer

[0833] Step 5:

[0834] Submitting and viewing answers

[0835] The server sends the generated response to the user terminal.

[0836] Input: Generated Answer

[0837] Output: Sending the answer

[0838] The user's device will display the response on the chat screen.

[0839] Input: Submit answer

[0840] Output: Display reply on chat screen

[0841] Step 6:

[0842] Search for consultation content

[0843] Users enter keywords into the search bar.

[0844] Input: Search keyword

[0845] Output: Input search keywords

[0846] The terminal sends a search request to the server.

[0847] Input: Enter search keywords

[0848] output: Search request to the server

[0849] The server searches the database for anonymized consultation content.

[0850] Input: Search request

[0851] Output: Search results

[0852] The server sends the search results to the user's terminal.

[0853] Input: Search results

[0854] output: sent to the user's terminal

[0855] The user's device displays a list of search results.

[0856] input:Submitted search results

[0857] output: Search results display

[0858] Step 7:

[0859] Registering and using favorites

[0860] Users can register specific expert models as "favorites."

[0861] Input: Favorite registration request

[0862] output: The input registration request

[0863] The terminal sends a favorites registration request to the server.

[0864] input:The input registration request

[0865] output: Send to server

[0866] The server stores the received requests in a database.

[0867] input:Submitted registration request

[0868] output: Save to database

[0869] Users access registered expert models from their favorites list.

[0870] input:Retrieve from database

[0871] output: Access to favorite models

[0872] (Application example 1)

[0873] 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."

[0874] Conventional food delivery systems have difficulty suggesting menus that take into account the user's health condition and nutritional balance, or providing expert advice in real time to individual questions. Furthermore, they lack the functionality to refer to frequently used menus or the health-related inquiries of other users, making them ineffective for self-resolving. There is a need to improve this situation.

[0875] 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.

[0876] In this invention, the server includes means for receiving input information from a user terminal, means for selecting an appropriate expert model based on the received information, means for generating an appropriate answer based on the selected expert model, means for transmitting the generated answer to the user terminal, means for analyzing the content of the consultation from the user terminal and selecting a food and nutrition expert model, means for generating a health menu based on the food and nutrition expert model, and means for presenting the generated menu to the user terminal. This allows users to receive real-time menu suggestions that take into account their health status and nutritional balance. Furthermore, users can search for other users' consultations anonymized and refer to menus generated based on similar past consultations, which can be used as a reference for self-solving.

[0877] "User terminal" refers to the communication device used by the user, including smartphones, tablets, and personal computers.

[0878] "Input Information" refers to text and selection information sent by a user through a user terminal.

[0879] "Server" refers to a computer system that receives, analyzes, and processes information from user terminals.

[0880] An "expert model" refers to an AI model with a knowledge base specialized in a particular field (e.g., medicine, law, nutrition, etc.).

[0881] A "food and nutrition expert model" refers to an AI model with specialized knowledge about food and nutrition.

[0882] "Menu" means the food options and suggestions provided to Users.

[0883] "Health menu" refers to meal plans and food suggestions generated taking into account the user's health status and nutritional balance.

[0884] "Anonymization" refers to the process of processing information so that a specific user cannot be identified.

[0885] "Consultation content" refers to questions or matters of consultation that users send to the system.

[0886] "Generated answers" refers to answers generated by the server based on an expert model.

[0887] "Session ID" refers to a unique identifier generated by the server after a user logs in.

[0888] "Display" refers to the process of visually presenting information on the screen of a user device.

[0889] "Search results" refers to the list of relevant information returned by the server in response to a user's search request.

[0890] "Favorites" refers to a feature that allows users to register specific information for easy access later.

[0891] The present invention is a system that uses a user terminal, a server, and an expert model to suggest menus that take into account the user's health condition and nutritional balance. A specific embodiment of this system is described below.

[0892] User Registration and Login

[0893] User Registration:

[0894] When a user first uses the system, they enter detailed information such as their name, email address, password, and dietary and allergy information from their device and send it to the server. This information is stored in the server's database. Once registration is complete, a confirmation message is returned to the user's device.

[0895] Login:

[0896] Existing users log in by entering their email address and password. The login information is sent to the server and checked against a database. If authentication is successful, a session ID is generated and returned to the user's device, allowing the user to access the home screen.

[0897] Individual chat consultation

[0898] Enter and submit your inquiry:

[0899] If a user wants to ask for advice about a specific dietary issue, they enter the content of their inquiry in text format on the chat screen of their device and press the send button. The content of their inquiry and the session ID are then sent to the server.

[0900] Analyzing the inquiry and generating an answer:

[0901] The server analyzes the received consultation content and selects the most appropriate food and nutrition expert model. The selected food and nutrition expert model analyzes the input consultation content and generates a health menu.

[0902] Submit and view your answers:

[0903] The generated menu is sent from the server to the user's device and displayed on the chat screen, allowing the user to quickly obtain a healthy menu.

[0904] For example, when a user inputs and sends a question such as "What healthy menu do you recommend for dinner tonight?", the server passes this information to the food and nutrition expert model. The food and nutrition expert model generates a menu that reads, "For a healthy dinner, we recommend steamed vegetables and chicken breast." This menu is sent to the user's device, where the user can view it on the chat screen.

[0905] Search and browse functions

[0906] Search for consultation content:

[0907] Users can search for past consultations with other users by entering specific keywords in the search bar. This search request is sent to the server, which returns the consultations in the database in anonymized form.

[0908] Displaying search results:

[0909] The search results sent from the server are displayed as a list on the user's device, and the user can view past similar consultations. This allows users to refer to information on cases similar to their own and use it as a reference for self-solving.

[0910] Add to Favorites

[0911] Favorite settings:

[0912] If a user likes a particular food nutrition expert model or generated menu, they can register it by pressing the "Favorite" button. The request to register as a favorite is sent to the server and stored in the database.

[0913] Use Favorites:

[0914] Registered favorite models and menus can be easily accessed on the user's device, and if the user wants to access them again, they can be quickly recalled, improving convenience.

[0915] This system mainly uses the following hardware and software.

[0916] Hardware:

[0917] Server: Cloud server (e.g. AWS, Google Cloud, etc.)

[0918] User device: Smartphone (iOS / Android)

[0919] software:

[0920] Framework: Flask (Python)

[0921] Database: SQLite

[0922] AI model: Language model (e.g., OpenAI's GPT)

[0923] For example, if a user inputs "What low-calorie menu would you recommend for today's lunch?", the food and nutrition expert model will generate a menu that reads, "Steamed chicken salad and smoothies are recommended as low-calorie menu items." This menu is sent to the user's device, and the user can view it on the chat screen. Examples of prompt sentences are as follows:

[0924] User: "What's a low-calorie option for lunch today?"

[0925] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0926] Step 1:

[0927] The user sends input information such as name, email address, password, and dietary and allergy information from the user device. The server receives this information and stores it in a database. Input: User information. Output: User information stored in the database.

[0928] Step 2:

[0929] An existing user logs in with their email address and password. The login information sent from the user's device to the server is checked against the database. If authentication is successful, the server generates a session ID and returns it to the user's device. Input: Login information. Output: Session ID.

[0930] Step 3:

[0931] The user enters the consultation content in text format on the chat screen of the user device and presses the send button. This sends the consultation content and session ID to the server. Input: Consultation content and session ID. Output: Data sent to the server.

[0932] Step 4:

[0933] The server analyzes the received consultation content and selects the most appropriate food and nutrition expert model. The selected model performs analysis based on the input consultation content. Input: Consultation content. Output: Selected food and nutrition expert model.

[0934] Step 5:

[0935] The selected food and nutrition expert model generates a health-related menu based on the consultation. For example, if a user inputs "What healthy menu would you recommend for dinner today?", the model generates a menu that says "For a healthy dinner, I recommend steamed vegetables and chicken breast." Input: Consultation. Output: Generated menu.

[0936] Step 6:

[0937] The generated menu is sent from the server to the user's device and displayed on the chat screen. Input: Generated menu. Output: Menu sent to the user's device.

[0938] Step 7:

[0939] If the user likes a particular food nutrition expert model or generated menu, they can register it by pressing the "Favorite" button. This favorite registration request is sent to the server and stored in the database. Input: Favorite registration request. Output: Favorite information stored in the database.

[0940] Step 8:

[0941] When the user wants to access a favorite item again, the server extracts the item from the database based on the request from the user's device and sends it to the user's device. Input: Request from the user's device. Output: Extracted item.

[0942] Step 9:

[0943] A user enters a specific keyword in the search bar to search for past consultations from other users. This search request is sent to the server, and the consultations in the database are anonymized and returned as search results. Input: Keyword. Output: Anonymized search results.

[0944] Step 10:

[0945] The search results sent from the server are displayed as a list on the user's device. The user can view past similar consultations. Input: Search results. Output: List displayed on the user's device.

[0946] Example prompt sentence:

[0947] User: "What's a low-calorie option for lunch today?"

[0948] 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.

[0949] The present invention uses a system that combines a user terminal, a server, an expert model, and an emotion engine to provide quick and appropriate answers to various inquiries users have in their daily lives. A specific embodiment of this system will be described below.

[0950] User Registration and Login

[0951] User Registration:

[0952] When a user uses the app for the first time, they enter their personal information (name, email address, password) to register. This information is sent from the user's device to the server and stored in a database. Once registration is complete, a confirmation message is returned to the user's device.

[0953] Login:

[0954] Existing users log in by entering their email address and password. The entered login information is sent from the user's device to the server and checked against the database. If authentication is successful, a session ID is generated and returned to the user's device, allowing the user to access the home screen.

[0955] Individual chat consultation

[0956] Enter and submit your inquiry:

[0957] When a user wants to ask a specific question, they enter the question in text format on the chat screen of their device and press the send button. The question and session ID are then sent to the server.

[0958] Emotion Recognition with Emotion Engine:

[0959] When the server receives the consultation content, it analyzes the user's emotions using an emotion engine. Based on the analysis results, it selects an appropriate expert model.

[0960] Analyzing the inquiry and generating an answer:

[0961] The selected expert model analyzes the content of the consultation and the emotional information from the emotion engine to generate an appropriate answer. This analysis allows for more personalized answers based on the user's emotions.

[0962] Submit and view your answers:

[0963] The generated answers are sent from the server to the user's device and displayed on the chat screen, allowing the user to quickly obtain the information they need.

[0964] Search and browse functions

[0965] Search for consultation content:

[0966] Users can search for past consultations with other users by entering specific keywords in the search bar. This search request is sent from the user's device to the server, and the consultations in the database are anonymized and returned as search results.

[0967] Displaying search results:

[0968] The search results sent from the server are displayed as a list on the user's device, and the user can view past similar consultations. This allows users to refer to information on cases similar to their own and use it as a reference for self-solving.

[0969] Add to Favorites

[0970] Favorite settings:

[0971] If a user likes a particular expert model, they can register it by pressing the "Favorite" button. The favorite registration request is sent from the user's terminal to the server and stored in the database.

[0972] Use Favorites:

[0973] Registered favorite expert models can be easily accessed on the user's device, and can be quickly recalled if the user wishes to access them again, improving convenience.

[0974] Specific examples

[0975] For example, consider the case of a medical consultation. When a user inputs and sends the question, "I've been having really bad headaches lately. What should I do?", the server passes this content to the emotion engine. The emotion engine recognizes the user's emotion as "anxiety" and passes this information along with the medical expert model. The medical expert model takes "anxiety" into consideration and generates a response such as, "Possible causes of headaches include stress, eye strain, and lack of sleep. We recommend that you first get plenty of rest. If your symptoms persist, consult a doctor." This response is sent to the user's device and can be viewed by the user on the chat screen.

[0976] As described above, the present invention realizes a system that provides more accurate and human-like answers by taking the user's emotions into consideration.

[0977] The processing flow will be explained below.

[0978] User Registration and Login

[0979] New Registration

[0980] Step 1:

[0981] The user launches the app and is presented with a sign-up form.

[0982] Step 2:

[0983] The user's device prompts them to enter their name, email address, and password, and then the user presses the "Send" button.

[0984] Step 3:

[0985] The user terminal sends the input information to the server.

[0986] Step 4:

[0987] The server stores the received information in a database.

[0988] Step 5:

[0989] The server generates a registration success message and sends it to the user terminal.

[0990] Log in

[0991] Step 1:

[0992] The user enters their email address and password on the login screen and clicks the "Login" button.

[0993] Step 2:

[0994] The user terminal sends the input information to the server.

[0995] Step 3:

[0996] The server checks the received login information against its database.

[0997] Step 4:

[0998] If the server is successful in the authentication, it generates a session ID and sends it to the user's terminal.

[0999] Step 5:

[1000] The user device saves the session ID and displays the home screen.

[1001] Individual chat consultation

[1002] Step 1:

[1003] The user accesses the consultation screen, enters the consultation content in text format, and presses the send button.

[1004] Step 2:

[1005] The user device sends the consultation content and session ID to the server.

[1006] Step 3:

[1007] The server receives the consultation content and transfers it to the emotion engine.

[1008] Step 4:

[1009] The emotion engine analyzes the content of the consultation and recognizes the user's emotions (e.g., "anxiety," "impatience," etc.).

[1010] Step 5:

[1011] The server receives the analysis results of the emotion engine and selects an appropriate expert model.

[1012] Step 6:

[1013] The expert model generates an appropriate answer based on the consultation content and emotional information.

[1014] Step 7:

[1015] The expert model returns the generated answer to the server.

[1016] Step 8:

[1017] The server sends the response to the user's device.

[1018] Step 9:

[1019] The user's device will display the response on the chat screen.

[1020] Search and browse functions

[1021] Step 1:

[1022] A user enters keywords into the search bar and presses the search button.

[1023] Step 2:

[1024] The user terminal sends the keyword to the server.

[1025] Step 3:

[1026] The server searches the database for anonymized consultation content based on the received keywords.

[1027] Step 4:

[1028] The server sends the search results to the user's device.

[1029] Step 5:

[1030] The user's device displays a list of search results.

[1031] Add to Favorites

[1032] Step 1:

[1033] The user presses the "favorite" button for a particular expert model.

[1034] Step 2:

[1035] The user terminal sends a "favorite" registration request to the server along with the session ID.

[1036] Step 3:

[1037] The server stores the "favorites" information in a database.

[1038] Step 4:

[1039] The server sends a confirmation message to the user terminal.

[1040] Step 5:

[1041] The user terminal displays a confirmation message.

[1042] Example: Medical consultation

[1043] Step 1:

[1044] The user logs in, accesses the home screen, and selects "Medical Consultation."

[1045] Step 2:

[1046] The user enters the question they want to ask, such as "I've been having terrible headaches lately. What should I do?" and presses the send button.

[1047] Step 3:

[1048] The user device sends the consultation content and session ID to the server.

[1049] Step 4:

[1050] The server transfers the consultation content to the emotion engine, which then recognizes the emotion "anxiety."

[1051] Step 5:

[1052] The server receives the analysis results of the emotion engine ("anxiety") and selects a medical expert model.

[1053] Step 6:

[1054] The medical expert model analyzes the consultation content while taking into account "anxiety" and generates the following answer: "Possible causes of headaches include stress, eye strain, and lack of sleep. We recommend that you first get plenty of rest. If your symptoms do not improve, please consult a doctor."

[1055] Step 7:

[1056] The medical expert model returns the generated answer to the server.

[1057] Step 8:

[1058] The server sends the response to the user's device.

[1059] Step 9:

[1060] The user's device will display the response on the chat screen.

[1061] By going through these processing steps, users can quickly receive an appropriate answer that corresponds to their feelings.

[1062] Example 2

[1063] 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."

[1064] Conventional expert systems provide answers without considering the user's feelings, resulting in insufficient appropriate responses. Furthermore, they lacked the functionality to efficiently search and view inquiries from multiple users, making it difficult for users to access similar past cases to resolve their own issues. Furthermore, it was difficult for users to call their favorite experts again.

[1065] 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.

[1066] In this invention, the server includes a means for using an emotion analysis engine that analyzes emotions based on user input information, a means for personalizing answers based on the emotion analysis results, and a means for anonymizing and searching for consultation details from other users. This enables the provision of more accurate and personalized answers that take the user's emotions into consideration. Furthermore, by efficiently searching and browsing similar past consultation details, users can easily obtain the information they need to resolve their issues. Furthermore, by providing a function that allows users to register their favorite experts as favorites and easily call them up, user convenience can be improved.

[1067] "User terminal" means a device through which a user accesses the system and enters or receives information.

[1068] "Input information" refers to questions, inquiries, or other data provided by users to the system.

[1069] A "server" is a central computer system that processes information sent from user terminals and performs the necessary analysis and responses.

[1070] An "expert model" is an algorithm or program with specific expertise that provides appropriate answers to user questions.

[1071] The "means for generating an answer" is a mechanism for constructing an appropriate answer to the user's input information using an expert model.

[1072] An "emotion analysis engine" is an analysis system for reading emotions from user input information.

[1073] "Anonymization" is the process of processing information so that individuals cannot be identified.

[1074] A "database" is a collection of information that stores user information, consultation details, etc. in an organized manner and can be searched or retrieved as needed.

[1075] "Favorites" is a feature that allows users to save specific expert models for easy access again.

[1076] "Personalization" means optimizing services and information according to the individual characteristics and emotions of each user.

[1077] The present invention uses a system that combines a user terminal, a server, an expert model, and an emotion analysis engine to provide quick and appropriate answers to various inquiries users have in their daily lives. Specific embodiments of this system are described below.

[1078] User Registration and Login

[1079] User Registration

[1080] When a user uses the system for the first time, they enter their name, email address, and password to register. This information is sent from the user's device to the server, which stores the received information in a database. Once registration is complete, the server sends a confirmation message to the user's device.

[1081] Log in

[1082] Existing users log in by entering their email address and password. This login information is sent from the user's device to the server, which checks the information against a database. If authentication is successful, the server generates a session ID and sends it to the user's device. The user's device retains the received session ID and displays the home screen.

[1083] Individual chat consultation

[1084] Enter and submit your consultation details

[1085] When a user wants to ask a specific question, they enter the question in text format on the chat screen of their device and press the send button. The question and session ID are then sent from the user device to the server.

[1086] Emotion recognition by emotion engine

[1087] The server passes the received consultation content to the emotion analysis engine, which analyzes the user's emotions and returns the results to the server, which selects an appropriate expert model based on the emotion analysis results.

[1088] Analysis of consultation content and generation of answers

[1089] The selected expert model analyzes the consultation content and emotional information and generates an appropriate answer that reflects the results of the emotional analysis, making it more personalized for the user.

[1090] Submitting and viewing answers

[1091] The generated answers are sent from the server to the user's device and displayed on the chat screen, allowing the user to quickly obtain the information they need.

[1092] Search and browse functions

[1093] Search for consultation content

[1094] Users can search for past consultations with other users by entering specific keywords in the search bar. This search request is sent from the user's device to the server, which then anonymizes the consultations in the database and returns them as search results.

[1095] Displaying search results

[1096] The search results are sent from the server to the user's device and displayed as a list. Users can then browse similar past consultations and find information on cases similar to their own.

[1097] Add to Favorites

[1098] Favorites Settings

[1099] If a user likes a particular expert model, he or she can register it by pressing the "Favorite" button. This favorite registration request is sent from the user terminal to the server, and the server stores it in a database.

[1100] Using Favorites

[1101] Saved favorite expert models can be easily accessed on the user's device, and can be quickly recalled when the user wants to access them again, improving convenience.

[1102] Specific examples

[1103] For example, consider the case where a user seeks medical advice. When the user enters and submits the question, "I've been having really bad headaches lately. What should I do?", the server passes this information to the emotion analysis engine. The emotion analysis engine recognizes the user's emotion as "anxiety." The server then sends this information to the medical expert model. Taking "anxiety" into consideration, the medical expert model generates the following response: "Possible causes of headaches include stress, eye strain, and lack of sleep. We recommend that you first get plenty of rest. If your symptoms persist, consult a doctor." This response is then sent from the server to the user's device, where the user can view it on the chat screen.

[1104] Prompt Sentence Examples

[1105] User: I've been having terrible headaches lately, what should I do?

[1106] AI: Headaches can be caused by stress, eye strain, lack of sleep, etc. First of all, we recommend getting plenty of rest. If your symptoms persist, consult your doctor.

[1107] The system also takes the user's emotions into account to provide more accurate and human-like answers.

[1108] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1109] Program processing flow

[1110] User Registration and Login

[1111] Step 1:

[1112] The user enters new registration information.

[1113] Enter your name, email address, and password.

[1114] How it works: The user enters the required information into the application's registration screen.

[1115] Output: Input data (name, email address, password).

[1116] Step 2:

[1117] The terminal sends the registration information to the server.

[1118] Input: Input data (name, email address, password).

[1119] Operation: The user terminal sends input data to the server.

[1120] Output: User registration information sent to the server.

[1121] Step 3:

[1122] The server stores the user information in a database.

[1123] Input: User registration information sent to the server.

[1124] How it works: The server stores the received information in a database.

[1125] Output: Status of user information saved to database.

[1126] Step 4:

[1127] The server sends a registration completion message to the terminal.

[1128] Input: Status of saving user information to database.

[1129] Operation: The server generates and sends a confirmation message to the user device indicating that the registration process is complete.

[1130] Output: Sending a registration completion message to the user's terminal.

[1131] Step 5:

[1132] The user enters their login information.

[1133] Enter your email address and password.

[1134] How it works: A user enters their email address and password on a login screen.

[1135] Output: Input data (email address, password).

[1136] Step 6:

[1137] The device sends the login information to the server.

[1138] Input: Input data (email address, password).

[1139] How it works: The user device sends login information to the server.

[1140] Output: The login information sent to the server.

[1141] Step 7:

[1142] The server checks the information against a database.

[1143] Input: The login information sent to the server.

[1144] How it works: The server checks the login information it receives against its database.

[1145] Output: Authentication success / failure result.

[1146] Step 8:

[1147] The server sends the authentication result and session ID to the terminal.

[1148] Input: The result of successful authentication.

[1149] How it works: If authentication is successful, the server generates a session ID and sends it to the user's device.

[1150] Output: A session ID is generated and sent to the user's device.

[1151] Step 9:

[1152] The device displays the home screen.

[1153] Input: Session ID.

[1154] Operation: The user device retains the received session ID and displays the home screen.

[1155] Output: Home screen display.

[1156] Individual chat consultation

[1157] Step 1:

[1158] The user inputs the consultation content.

[1159] Input: Consultation details.

[1160] How it works: The user enters the content of their inquiry in text format on the chat screen.

[1161] Output: The consultation content entered.

[1162] Step 2:

[1163] The device sends the consultation content and session ID to the server.

[1164] Input: Session ID, consultation details.

[1165] Operation: The user terminal sends the consultation content and session ID to the server.

[1166] Output: The consultation details and session ID sent to the server.

[1167] Step 3:

[1168] The server passes the consultation content to the emotion engine.

[1169] Input: The consultation content sent to the server.

[1170] Operation: The server passes the received consultation content to the emotion analysis engine.

[1171] Output: Transferring the consultation content to the emotion engine.

[1172] Step 4:

[1173] The emotion engine analyzes the user's emotions.

[1174] Input: Consultation details.

[1175] How it works: The emotion engine analyzes the content of the consultation and recognizes the user's emotions.

[1176] Output: Emotion analysis results.

[1177] Step 5:

[1178] The server obtains the emotion analysis results.

[1179] Input: Sentiment analysis results.

[1180] Operation: The server obtains the analysis results from the emotion engine.

[1181] Output: The server receives the sentiment analysis results.

[1182] Step 6:

[1183] The server selects an expert model.

[1184] Input: Sentiment analysis results.

[1185] Operation: The server selects an appropriate expert model based on the sentiment analysis results.

[1186] Output: Selected expert model.

[1187] Step 7:

[1188] The server passes the consultation content to the expert model.

[1189] Input: Consultation content, emotional information.

[1190] Operation: The server passes the consultation content and emotion information to the expert model.

[1191] Output: Transfer of input data to the expert model.

[1192] Step 8:

[1193] The expert model generates the answer.

[1194] Input: Consultation content, emotional information.

[1195] How it works: The expert model analyzes the consultation content and emotional information and generates an appropriate answer.

[1196] Output: The generated answer.

[1197] Step 9:

[1198] The server sends the generated response to the terminal.

[1199] Input: The generated answer.

[1200] Operation: The server generates a response and sends it to the user's device.

[1201] Output: Sending the answer to the user's terminal.

[1202] Step 10:

[1203] The device will display the response on the chat screen.

[1204] Input:Answer.

[1205] Operation: The user's device displays the received reply on the chat screen.

[1206] Output: The response displayed in the chat window.

[1207] Add to Favorites

[1208] Step 1:

[1209] The user presses the "Favorite" button.

[1210] Input: None.

[1211] How it works: A user who likes a particular expert model presses the "favorite" button.

[1212] Output: Favorite button click event.

[1213] Step 2:

[1214] The terminal sends a favorites registration request to the server.

[1215] Input: Favorite button click event.

[1216] Operation: The device sends a favorite registration request to the server.

[1217] Output: The favorite request sent to the server.

[1218] Step 3:

[1219] The server stores the favorites information in a database.

[1220] Input: The favorite request sent to the server.

[1221] How it works: The server stores information in a database based on the favorites request.

[1222] Output: Status of saving favorite information to database.

[1223] Step 4:

[1224] The device displays registered expert models.

[1225] Input: None.

[1226] How it works: Saved favorite expert models are easily accessible on the user's device.

[1227] Output: Display of favorite expert model on user's device.

[1228] (Application example 2)

[1229] 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."

[1230] Conventional customer support systems in virtual stores respond uniformly without considering the individual emotional state of each user, making it impossible to provide personalized, appropriate support to users. As a result, users often do not receive appropriate support, leaving them frustrated and confused. Furthermore, conventional systems lacked functionality to easily search for past user inquiries, making it difficult to quickly obtain information for self-resolving.

[1231] 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.

[1232] In this invention, the server includes means for receiving input information from a user terminal, means for selecting an appropriate expert model based on the received information, means for analyzing the user's emotions using emotion analysis means, means for selecting an expert model based on the analyzed emotion information, means for generating an appropriate answer based on the selected expert model, and means for transmitting the generated answer to the user terminal, thereby enabling personalized and appropriate support according to the user's emotional state.

[1233] "User terminal" means an electronic device used by a user, including a smartphone, tablet, or PC.

[1234] "Input information" refers to various data sent by a user via a user terminal, including text, images, audio, etc.

[1235] An "expert model" is an algorithm or program that uses knowledge in a specific field to generate appropriate answers to users' questions or inquiries.

[1236] "Emotion analysis means" refers to a system or algorithm that analyzes input information from a user and recognizes their emotional state.

[1237] "Emotion information" refers to data relating to the user's emotional state obtained by emotion analysis means.

[1238] The "selection means" refers to a method or mechanism for selecting the most appropriate model from among multiple expert models based on the user's input information and emotion information.

[1239] An "answer generator" is an algorithm or program that generates an appropriate answer to a user's question or inquiry based on a selected expert model.

[1240] "Transmitting means" refers to a system or method for sending the generated answer to the user terminal.

[1241] An embodiment of the present invention will be described.

[1242] As an application example of the invention, we implement a customer support system in a virtual store. This system mainly uses a user terminal, a server, a sentiment analysis means, and an expert model. The specific hardware used includes a smartphone, and the software used includes AWS Lambda, Amazon DynamoDB, and the OpenAI GPT-3 model.

[1243] First, a user accesses the application using a user device such as a smartphone, registers as a user if necessary, and logs in. When registering, the user enters their name, email address, and password, which are sent to the server. The server receives this information and stores it in Amazon DynamoDB. When logging in, the email address and password are verified, and if authentication is successful, a session ID is generated and returned to the user.

[1244] While browsing products in a virtual store, a user inputs and submits a specific inquiry in text format. The user's device sends this input information to the server. When the server receives the input information, it activates the emotion analysis means to recognize the user's emotional state. This emotion analysis uses the OpenAI GPT-3 model. For example, if the input inquiry is a question such as "Is this refrigerator energy efficient?", the emotion analysis means recognizes this as an emotion that indicates a desire for information.

[1245] Next, based on the analyzed emotional information, the server selects the most appropriate expert model. The selected expert model generates an appropriate answer based on the user's emotional state and the content of the inquiry. For example, if the question is about energy efficiency, a technical expert model will be selected and generate an answer that explains the details of energy efficiency. This generated answer is then sent back to the user's device via the server.

[1246] A specific example is sentiment analysis and answer generation using the following prompt sentence:

[1247] Example of a prompt for sentiment analysis:

[1248] Choose an expert model by analyzing the sentiment of the message: 'Is this refrigerator energy efficient?'

[1249] Example prompts for answer generation:

[1250] General Breakdown Technician's Answer: Provide a good answer to 'Is this refrigerator energy efficient?'

[1251] This system allows users to receive personalized support in real time based on their emotions. It also allows users to anonymously search for other inquiries and gain insights from similar past inquiries. This results in higher user satisfaction and more effective self-resolution than traditional customer support systems.

[1252] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1253] Step 1:

[1254] Receive input information from the user's device

[1255] The user uses a dedicated application in the virtual store to input and send the details of their inquiry in text format. The user's device sends this input information to the server. The input information includes the user ID and session ID.

[1256] Step 2:

[1257] Emotion analysis using emotion analysis tools

[1258] The server analyzes the user's emotions using the OpenAI GPT-3 model based on the received input information. The server sends the user's consultation details as input data to GPT-3 and receives analyzed emotional information. This emotional information indicates emotional states such as "anxiety," "doubt," and "joy."

[1259] Step 3:

[1260] Expert model selection

[1261] The server selects the most appropriate expert model based on the emotion information obtained from the emotion analysis means. For example, if the user's emotion is analyzed as "question," an expert model that is good at answering technical questions will be selected. In this selection process, an algorithm determines the most appropriate model from multiple expert models.

[1262] Step 4:

[1263] Generate answers

[1264] The server uses the selected expert model to generate an answer based on the input information. A generative AI model is used to create a specific answer based on the consultation content. This model generates an answer that is appropriate for the user's question based on the prompt text.

[1265] Step 5:

[1266] Submit your answer

[1267] The generated answers are sent from the server to the user's device, and are displayed in the user's application, allowing the user to obtain appropriate information in real time.

[1268] Specifically, the reply text is sent to the user's terminal and displayed on the chat screen of the application.

[1269] Step 6:

[1270] Search function for past consultation content (optional)

[1271] Users can search for past consultations by entering specific keywords in the search bar within the application. The search request sent from the user's device is processed by the server, and related consultations are searched for in an anonymized form. The search results are sent back to the user's device and displayed in a list within the application.

[1272] Step 7:

[1273] Register your favorite expert models (optional)

[1274] Users can register their favorite expert models as favorites. A request to add to favorites is sent from the user's device to the server, and the server stores the information in a database, allowing users to instantly access specific expert models.

[1275] 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.

[1276] 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.

[1277] 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.

[1278] [Third embodiment]

[1279] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1280] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1281] 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).

[1282] 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.

[1283] 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.

[1284] 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).

[1285] 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.

[1286] 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.

[1287] 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.

[1288] 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.

[1289] 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.

[1290] 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."

[1291] The present invention uses a system consisting of a user terminal, a server, and an expert model to provide prompt and appropriate answers to various inquiries in the user's daily life. A specific embodiment of this system will be described below.

[1292] User Registration and Login

[1293] User Registration:

[1294] When a user uses the app for the first time, they enter their personal information (name, email address, password) to register. This information is sent from the user's device to the server and stored in a database. Once registration is complete, a confirmation message is returned to the user's device.

[1295] Login:

[1296] Existing users log in by entering their email address and password. The entered login information is sent from the user's device to the server and checked against the database. If authentication is successful, a session ID is generated and returned to the user's device, allowing the user to access the home screen.

[1297] Individual chat consultation

[1298] Enter and submit your inquiry:

[1299] When a user wants to ask a specific question, they enter the question in text format on the chat screen of their device and press the send button. The question and session ID are then sent to the server.

[1300] Analyzing the inquiry and generating an answer:

[1301] The server analyzes the received consultation content and selects the most appropriate expert model (medical, legal, business, etc.). The selected expert model analyzes the input consultation content and generates an appropriate answer.

[1302] Submit and view your answers:

[1303] The generated answers are sent from the server to the user's device and displayed on the chat screen, allowing the user to quickly obtain the information they need.

[1304] Search and browse functions

[1305] Search for consultation content:

[1306] Users can search for past consultations with other users by entering specific keywords in the search bar. This search request is sent from the user's device to the server, and the consultations in the database are anonymized and returned as search results.

[1307] Displaying search results:

[1308] The search results sent from the server are displayed as a list on the user's device, and the user can view past similar consultations. This allows users to refer to information on cases similar to their own and use it as a reference for self-solving.

[1309] Add to Favorites

[1310] Favorite settings:

[1311] If a user likes a particular expert model, they can register it by pressing the "Favorite" button. The favorite registration request is sent from the user's terminal to the server and stored in the database.

[1312] Use Favorites:

[1313] Registered favorite expert models can be easily accessed on the user's device, and can be quickly recalled if the user wishes to access them again, improving convenience.

[1314] Specific examples

[1315] For example, consider the case of a medical consultation. When a user inputs and sends the question, "I've been having terrible headaches lately. What should I do?", the server passes this information to the medical expert model. The medical expert model generates a response such as, "Possible causes of headaches include stress, eye strain, and lack of sleep. We recommend that you first get plenty of rest. If this does not improve, please consult a doctor." This response is sent to the user's device, where the user can view it on the chat screen.

[1316] As described above, the present invention provides a system that enables users to obtain prompt and appropriate answers to a variety of problems and questions.

[1317] The processing flow will be explained below.

[1318] User Registration and Login

[1319] New Registration

[1320] Step 1:

[1321] The user launches the app and sees the signup form.

[1322] Step 2:

[1323] The user's device prompts them to enter their name, email address, and password, then presses the "Send" button.

[1324] Step 3:

[1325] The user terminal sends the input information to the server.

[1326] Step 4:

[1327] The server stores the received information in a database.

[1328] Step 5:

[1329] The server returns a registration success message to the user terminal.

[1330] Log in

[1331] Step 1:

[1332] The user enters their email address and password on the login screen and clicks the "Login" button.

[1333] Step 2:

[1334] The user terminal sends the input information to the server.

[1335] Step 3:

[1336] The server checks the received information against a database.

[1337] Step 4:

[1338] If the server matches, it generates a session ID and sends it to the user's device.

[1339] Step 5:

[1340] The user device saves the session ID and displays the home screen.

[1341] Individual chat consultation

[1342] Step 1:

[1343] The user enters the consultation content in text on the app's consultation screen and presses the send button.

[1344] Step 2:

[1345] The user device sends the consultation content and session ID to the server.

[1346] Step 3:

[1347] The server analyzes the received consultation content and selects an appropriate expert model.

[1348] Step 4:

[1349] The expert model analyzes the consultation content and generates an appropriate answer.

[1350] Step 5:

[1351] The expert model returns the generated answer to the server.

[1352] Step 6:

[1353] The server sends the response to the user's device.

[1354] Step 7:

[1355] The user's device will display the response on the chat screen.

[1356] Search and browse functions

[1357] Step 1:

[1358] A user enters keywords into the search bar and presses the search button.

[1359] Step 2:

[1360] The user terminal sends the keyword to the server.

[1361] Step 3:

[1362] The server searches the database for anonymized consultation content based on the received keywords.

[1363] Step 4:

[1364] The server sends the search results to the user's device.

[1365] Step 5:

[1366] The user's device displays a list of search results.

[1367] Add to Favorites

[1368] Step 1:

[1369] The user presses the "favorite" button for a particular expert model.

[1370] Step 2:

[1371] The user terminal sends a favorite registration request to the server along with the session ID.

[1372] Step 3:

[1373] The server stores the favorites information in a database.

[1374] Step 4:

[1375] The server sends a confirmation message to the user terminal.

[1376] Step 5:

[1377] The user terminal displays a confirmation message.

[1378] Example 1

[1379] 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."

[1380] Conventional online consultation systems have faced challenges in that it is difficult for users to quickly obtain appropriate answers to their inquiries. In addition, searching for consultation content and accessing favorite expert models is cumbersome, resulting in a poor user experience.

[1381] 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.

[1382] In this invention, the server includes means for receiving input information from a user terminal, means for storing the received information in a database, means for verifying and authenticating login information of existing users, means for analyzing the received consultation content using a natural language processing algorithm, means for selecting an appropriate expert model based on the analyzed information, means for generating an appropriate answer based on the selected expert model, means for transmitting the generated answer to the user terminal, means for anonymizing and searching for the consultation content of other users, and means for displaying a list of search results. This allows users to quickly obtain appropriate answers and also makes it easier to search for consultation content and access their favorite expert models.

[1383] A "user terminal" is an electronic device such as a computer, smartphone, or tablet that is directly operated by a user.

[1384] "Input information" refers to data that a user provides to the system via their terminal, such as name, email address, password, and consultation details.

[1385] A "server" is a computer system that processes information received from user devices and manages and analyzes information using databases and algorithms.

[1386] A "session ID" is a unique identifier used to identify and manage a user's login status.

[1387] A "natural language processing algorithm" is a technology that analyzes the content of the consultation received, understands the text data, and generates an appropriate response.

[1388] An "expert model" is a program or system that has expert knowledge in a particular field (e.g., medicine, law, business) and generates appropriate answers to inquiries from users.

[1389] A "database" is a system for systematically storing digital information and for efficiently searching and managing it.

[1390] "Authentication" is the process of verifying that the user information provided is valid.

[1391] "Analysis" is the process of breaking down received data, interpreting its meaning, and extracting relevant information.

[1392] "Anonymization" is the process of removing or transforming personally identifiable information to protect privacy.

[1393] "Search results" are lists of relevant data extracted from a database based on a specific keyword or query.

[1394] The present invention provides a system that includes a user terminal, a server, and an expert model to quickly and appropriately answer various inquiries from users. Specific embodiments of the system will be described below.

[1395] User Registration and Login

[1396] When a user first uses the system, they enter information such as their name, email address, and password from their device and send it to the server. The server stores the received information in a database and sends a registration completion confirmation message to the user's device. Existing users enter their email address and password on the login screen and send it to the server. After the server verifies the information in the database, if authentication is successful, it generates a session ID and sends it to the user's device. This allows the user to access the home screen.

[1397] Individual chat consultation

[1398] When a user wants to ask a specific question, they enter the question content in text format on the chat screen of their device and press the send button. The question content and session ID are then sent to the server. The server analyzes the received question content using a natural language processing algorithm and selects the most appropriate expert model. The selected expert model generates an answer based on the received question content, and the server sends that answer to the user's device. This allows the user to quickly check the answer on the chat screen.

[1399] Search and browse functions

[1400] Users can search for past consultations by entering keywords in the search bar. This search request is sent from the user's device to the server, which then searches the database for anonymized consultations. The search results are displayed as a list on the user's device, allowing the user to check similar past consultations.

[1401] Add to Favorites

[1402] Users can register specific expert models as favorites. A favorite registration request is sent from the user terminal to the server, and the server stores it in the database. Users can easily access the registered expert models from the favorites list.

[1403] Specific examples

[1404] For example, a user might type a question on the chat screen, such as "I've been having really bad headaches lately. What should I do?" and press the send button. The device then sends this question and the session ID to the server. The server analyzes the consultation and passes it on to the medical expert model. The medical expert model generates an answer: "Possible causes of headaches include stress, eye strain, and lack of sleep. We recommend that you first get plenty of rest. If this does not improve, consult a doctor." and sends this answer to the user's device. The user can then check this answer on the chat screen.

[1405] As described above, the present invention provides a system that enables users to obtain prompt and appropriate answers to a variety of problems and questions.

[1406] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1407] Step 1:

[1408] User Registration

[1409] The user enters information such as name, email address, and password into the user device.

[1410] Input: Name, Email Address, Password

[1411] Output: Input information

[1412] The terminal transmits the input information to the server.

[1413] Input: Input information

[1414] Output: Data sent to the server

[1415] The server stores the received information in a database.

[1416] Input: Send data

[1417] Output: Information stored in the database

[1418] The server generates a confirmation message for completing the registration and sends it to the user terminal.

[1419] Input: Information stored in a database

[1420] Output: Confirmation message sent

[1421] Step 2:

[1422] Log in

[1423] The user enters their email address and password into the user terminal.

[1424] Input: Email address, password

[1425] Output: Input information

[1426] The terminal transmits the entered login information to the server.

[1427] Input: Input information

[1428] Output: Data sent to the server

[1429] The server checks the information against a database and authenticates it.

[1430] Input: Send data

[1431] Output: Authentication result

[1432] If the authentication is successful, the server generates a session ID and sends it to the user terminal.

[1433] Input: Authentication result

[1434] Output: Sending the session ID

[1435] The user device displays the home screen.

[1436] Input: Session ID

[1437] Output: Home screen display

[1438] Step 3:

[1439] Start individual chat consultation

[1440] The user enters the content of the consultation in text format on the chat screen and presses the send button.

[1441] Input: Consultation details

[1442] Output: Input consultation content

[1443] The terminal sends the consultation content and the session ID to the server.

[1444] Input: Consultation details, session ID

[1445] Output: Data sent to the server

[1446] Step 4:

[1447] Analysis of consultation content and generation of answers

[1448] The server analyzes the received consultation content using a natural language processing algorithm.

[1449] Input: Send data

[1450] Output: Analyzed consultation content

[1451] The server selects an appropriate expert model based on the analyzed consultation content.

[1452] Input: Analyzed consultation content

[1453] Output: Selected expert model

[1454] The expert model analyzes the content of the consultation and generates an appropriate answer.

[1455] Input: Analyzed consultation content

[1456] Output: The generated answer

[1457] Step 5:

[1458] Submitting and viewing answers

[1459] The server sends the generated response to the user terminal.

[1460] Input: Generated Answer

[1461] Output: Sending the answer

[1462] The user's device will display the response on the chat screen.

[1463] Input: Submit answer

[1464] Output: Display reply on chat screen

[1465] Step 6:

[1466] Search for consultation content

[1467] Users enter keywords into the search bar.

[1468] Input: Search keyword

[1469] Output: Input search keywords

[1470] The terminal sends a search request to the server.

[1471] Input: Enter search keywords

[1472] output: Search request to the server

[1473] The server searches the database for anonymized consultation content.

[1474] Input: Search request

[1475] Output: Search results

[1476] The server sends the search results to the user's terminal.

[1477] Input: Search results

[1478] output: sent to the user's terminal

[1479] The user's device displays a list of search results.

[1480] input:Submitted search results

[1481] output: Search results display

[1482] Step 7:

[1483] Registering and using favorites

[1484] Users can register specific expert models as "favorites."

[1485] Input: Favorite registration request

[1486] output: The input registration request

[1487] The terminal sends a favorites registration request to the server.

[1488] input:The input registration request

[1489] output: Send to server

[1490] The server stores the received requests in a database.

[1491] input:Submitted registration request

[1492] output: Save to database

[1493] Users access registered expert models from their favorites list.

[1494] input:Retrieve from database

[1495] output: Access to favorite models

[1496] (Application example 1)

[1497] 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."

[1498] Conventional food delivery systems have difficulty suggesting menus that take into account the user's health condition and nutritional balance, or providing expert advice in real time to individual questions. Furthermore, they lack the functionality to refer to frequently used menus or the health-related inquiries of other users, making them ineffective for self-resolving. There is a need to improve this situation.

[1499] 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.

[1500] In this invention, the server includes means for receiving input information from a user terminal, means for selecting an appropriate expert model based on the received information, means for generating an appropriate answer based on the selected expert model, means for transmitting the generated answer to the user terminal, means for analyzing the content of the consultation from the user terminal and selecting a food and nutrition expert model, means for generating a health menu based on the food and nutrition expert model, and means for presenting the generated menu to the user terminal. This allows users to receive real-time menu suggestions that take into account their health status and nutritional balance. Furthermore, users can search for other users' consultations anonymized and refer to menus generated based on similar past consultations, which can be used as a reference for self-solving.

[1501] "User terminal" refers to the communication device used by the user, including smartphones, tablets, and personal computers.

[1502] "Input Information" refers to text and selection information sent by a user through a user terminal.

[1503] "Server" refers to a computer system that receives, analyzes, and processes information from user terminals.

[1504] An "expert model" refers to an AI model with a knowledge base specialized in a particular field (e.g., medicine, law, nutrition, etc.).

[1505] A "food and nutrition expert model" refers to an AI model with specialized knowledge about food and nutrition.

[1506] "Menu" means the food options and suggestions provided to Users.

[1507] "Health menu" refers to meal plans and food suggestions generated taking into account the user's health status and nutritional balance.

[1508] "Anonymization" refers to the process of processing information so that a specific user cannot be identified.

[1509] "Consultation content" refers to questions or matters of consultation that users send to the system.

[1510] "Generated answers" refers to answers generated by the server based on an expert model.

[1511] "Session ID" refers to a unique identifier generated by the server after a user logs in.

[1512] "Display" refers to the process of visually presenting information on the screen of a user device.

[1513] "Search results" refers to the list of relevant information returned by the server in response to a user's search request.

[1514] "Favorites" refers to a feature that allows users to register specific information for easy access later.

[1515] The present invention is a system that uses a user terminal, a server, and an expert model to suggest menus that take into account the user's health condition and nutritional balance. A specific embodiment of this system is described below.

[1516] User Registration and Login

[1517] User Registration:

[1518] When a user first uses the system, they enter detailed information such as their name, email address, password, and dietary and allergy information from their device and send it to the server. This information is stored in the server's database. Once registration is complete, a confirmation message is returned to the user's device.

[1519] Login:

[1520] Existing users log in by entering their email address and password. The login information is sent to the server and checked against a database. If authentication is successful, a session ID is generated and returned to the user's device, allowing the user to access the home screen.

[1521] Individual chat consultation

[1522] Enter and submit your inquiry:

[1523] If a user wants to ask for advice about a specific dietary issue, they enter the content of their inquiry in text format on the chat screen of their device and press the send button. The content of their inquiry and the session ID are then sent to the server.

[1524] Analyzing the inquiry and generating an answer:

[1525] The server analyzes the received consultation content and selects the most appropriate food and nutrition expert model. The selected food and nutrition expert model analyzes the input consultation content and generates a health menu.

[1526] Submit and view your answers:

[1527] The generated menu is sent from the server to the user's device and displayed on the chat screen, allowing the user to quickly obtain a healthy menu.

[1528] For example, when a user inputs and sends a question such as "What healthy menu do you recommend for dinner tonight?", the server passes this information to the food and nutrition expert model. The food and nutrition expert model generates a menu that reads, "For a healthy dinner, we recommend steamed vegetables and chicken breast." This menu is sent to the user's device, where the user can view it on the chat screen.

[1529] Search and browse functions

[1530] Search for consultation content:

[1531] Users can search for past consultations with other users by entering specific keywords in the search bar. This search request is sent to the server, which returns the consultations in the database in anonymized form.

[1532] Displaying search results:

[1533] The search results sent from the server are displayed as a list on the user's device, and the user can view past similar consultations. This allows users to refer to information on cases similar to their own and use it as a reference for self-solving.

[1534] Add to Favorites

[1535] Favorite settings:

[1536] If a user likes a particular food nutrition expert model or generated menu, they can register it by pressing the "Favorite" button. The request to register as a favorite is sent to the server and stored in the database.

[1537] Use Favorites:

[1538] Registered favorite models and menus can be easily accessed on the user's device, and if the user wants to access them again, they can be quickly recalled, improving convenience.

[1539] This system mainly uses the following hardware and software.

[1540] Hardware:

[1541] Server: Cloud server (e.g. AWS, Google Cloud, etc.)

[1542] User device: Smartphone (iOS / Android)

[1543] software:

[1544] Framework: Flask (Python)

[1545] Database: SQLite

[1546] AI model: Language model (e.g., OpenAI's GPT)

[1547] For example, if a user inputs "What low-calorie menu would you recommend for today's lunch?", the food and nutrition expert model will generate a menu that reads, "Steamed chicken salad and smoothies are recommended as low-calorie menu items." This menu is sent to the user's device, and the user can view it on the chat screen. Examples of prompt sentences are as follows:

[1548] User: "What's a low-calorie option for lunch today?"

[1549] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1550] Step 1:

[1551] The user sends input information such as name, email address, password, and dietary and allergy information from the user device. The server receives this information and stores it in a database. Input: User information. Output: User information stored in the database.

[1552] Step 2:

[1553] An existing user logs in with their email address and password. The login information sent from the user's device to the server is checked against the database. If authentication is successful, the server generates a session ID and returns it to the user's device. Input: Login information. Output: Session ID.

[1554] Step 3:

[1555] The user enters the consultation content in text format on the chat screen of the user device and presses the send button. This sends the consultation content and session ID to the server. Input: Consultation content and session ID. Output: Data sent to the server.

[1556] Step 4:

[1557] The server analyzes the received consultation content and selects the most appropriate food and nutrition expert model. The selected model performs analysis based on the input consultation content. Input: Consultation content. Output: Selected food and nutrition expert model.

[1558] Step 5:

[1559] The selected food and nutrition expert model generates a health-related menu based on the consultation. For example, if a user inputs "What healthy menu would you recommend for dinner today?", the model generates a menu that says "For a healthy dinner, I recommend steamed vegetables and chicken breast." Input: Consultation. Output: Generated menu.

[1560] Step 6:

[1561] The generated menu is sent from the server to the user's device and displayed on the chat screen. Input: Generated menu. Output: Menu sent to the user's device.

[1562] Step 7:

[1563] If the user likes a particular food nutrition expert model or generated menu, they can register it by pressing the "Favorite" button. This favorite registration request is sent to the server and stored in the database. Input: Favorite registration request. Output: Favorite information stored in the database.

[1564] Step 8:

[1565] When the user wants to access a favorite item again, the server extracts the item from the database based on the request from the user's device and sends it to the user's device. Input: Request from the user's device. Output: Extracted item.

[1566] Step 9:

[1567] A user enters a specific keyword in the search bar to search for past consultations from other users. This search request is sent to the server, and the consultations in the database are anonymized and returned as search results. Input: Keyword. Output: Anonymized search results.

[1568] Step 10:

[1569] The search results sent from the server are displayed as a list on the user's device. The user can view past similar consultations. Input: Search results. Output: List displayed on the user's device.

[1570] Example prompt sentence:

[1571] User: "What's a low-calorie option for lunch today?"

[1572] 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.

[1573] The present invention uses a system that combines a user terminal, a server, an expert model, and an emotion engine to provide quick and appropriate answers to various inquiries users have in their daily lives. A specific embodiment of this system will be described below.

[1574] User Registration and Login

[1575] User Registration:

[1576] When a user uses the app for the first time, they enter their personal information (name, email address, password) to register. This information is sent from the user's device to the server and stored in a database. Once registration is complete, a confirmation message is returned to the user's device.

[1577] Login:

[1578] Existing users log in by entering their email address and password. The entered login information is sent from the user's device to the server and checked against the database. If authentication is successful, a session ID is generated and returned to the user's device, allowing the user to access the home screen.

[1579] Individual chat consultation

[1580] Enter and submit your inquiry:

[1581] When a user wants to ask a specific question, they enter the question in text format on the chat screen of their device and press the send button. The question and session ID are then sent to the server.

[1582] Emotion Recognition with Emotion Engine:

[1583] When the server receives the consultation content, it analyzes the user's emotions using an emotion engine. Based on the analysis results, it selects an appropriate expert model.

[1584] Analyzing the inquiry and generating an answer:

[1585] The selected expert model analyzes the content of the consultation and the emotional information from the emotion engine to generate an appropriate answer. This analysis allows for more personalized answers based on the user's emotions.

[1586] Submit and view your answers:

[1587] The generated answers are sent from the server to the user's device and displayed on the chat screen, allowing the user to quickly obtain the information they need.

[1588] Search and browse functions

[1589] Search for consultation content:

[1590] Users can search for past consultations with other users by entering specific keywords in the search bar. This search request is sent from the user's device to the server, and the consultations in the database are anonymized and returned as search results.

[1591] Displaying search results:

[1592] The search results sent from the server are displayed as a list on the user's device, and the user can view past similar consultations. This allows users to refer to information on cases similar to their own and use it as a reference for self-solving.

[1593] Add to Favorites

[1594] Favorite settings:

[1595] If a user likes a particular expert model, they can register it by pressing the "Favorite" button. The favorite registration request is sent from the user's terminal to the server and stored in the database.

[1596] Use Favorites:

[1597] Registered favorite expert models can be easily accessed on the user's device, and can be quickly recalled if the user wishes to access them again, improving convenience.

[1598] Specific examples

[1599] For example, consider the case of a medical consultation. When a user inputs and sends the question, "I've been having really bad headaches lately. What should I do?", the server passes this content to the emotion engine. The emotion engine recognizes the user's emotion as "anxiety" and passes this information along with the medical expert model. The medical expert model takes "anxiety" into consideration and generates a response such as, "Possible causes of headaches include stress, eye strain, and lack of sleep. We recommend that you first get plenty of rest. If your symptoms persist, consult a doctor." This response is sent to the user's device and can be viewed by the user on the chat screen.

[1600] As described above, the present invention realizes a system that provides more accurate and human-like answers by taking the user's emotions into consideration.

[1601] The processing flow will be explained below.

[1602] User Registration and Login

[1603] New Registration

[1604] Step 1:

[1605] The user launches the app and is presented with a sign-up form.

[1606] Step 2:

[1607] The user's device prompts them to enter their name, email address, and password, and then the user presses the "Send" button.

[1608] Step 3:

[1609] The user terminal sends the input information to the server.

[1610] Step 4:

[1611] The server stores the received information in a database.

[1612] Step 5:

[1613] The server generates a registration success message and sends it to the user terminal.

[1614] Log in

[1615] Step 1:

[1616] The user enters their email address and password on the login screen and clicks the "Login" button.

[1617] Step 2:

[1618] The user terminal sends the input information to the server.

[1619] Step 3:

[1620] The server checks the received login information against its database.

[1621] Step 4:

[1622] If the server is successful in the authentication, it generates a session ID and sends it to the user's terminal.

[1623] Step 5:

[1624] The user device saves the session ID and displays the home screen.

[1625] Individual chat consultation

[1626] Step 1:

[1627] The user accesses the consultation screen, enters the consultation content in text format, and presses the send button.

[1628] Step 2:

[1629] The user device sends the consultation content and session ID to the server.

[1630] Step 3:

[1631] The server receives the consultation content and transfers it to the emotion engine.

[1632] Step 4:

[1633] The emotion engine analyzes the content of the consultation and recognizes the user's emotions (e.g., "anxiety," "impatience," etc.).

[1634] Step 5:

[1635] The server receives the analysis results of the emotion engine and selects an appropriate expert model.

[1636] Step 6:

[1637] The expert model generates an appropriate answer based on the consultation content and emotional information.

[1638] Step 7:

[1639] The expert model returns the generated answer to the server.

[1640] Step 8:

[1641] The server sends the response to the user's device.

[1642] Step 9:

[1643] The user's device will display the response on the chat screen.

[1644] Search and browse functions

[1645] Step 1:

[1646] A user enters keywords into the search bar and presses the search button.

[1647] Step 2:

[1648] The user terminal sends the keyword to the server.

[1649] Step 3:

[1650] The server searches the database for anonymized consultation content based on the received keywords.

[1651] Step 4:

[1652] The server sends the search results to the user's device.

[1653] Step 5:

[1654] The user's device displays a list of search results.

[1655] Add to Favorites

[1656] Step 1:

[1657] The user presses the "favorite" button for a particular expert model.

[1658] Step 2:

[1659] The user terminal sends a "favorite" registration request to the server along with the session ID.

[1660] Step 3:

[1661] The server stores the "favorites" information in a database.

[1662] Step 4:

[1663] The server sends a confirmation message to the user terminal.

[1664] Step 5:

[1665] The user terminal displays a confirmation message.

[1666] Example: Medical consultation

[1667] Step 1:

[1668] The user logs in, accesses the home screen, and selects "Medical Consultation."

[1669] Step 2:

[1670] The user enters the question they want to ask, such as "I've been having terrible headaches lately. What should I do?" and presses the send button.

[1671] Step 3:

[1672] The user device sends the consultation content and session ID to the server.

[1673] Step 4:

[1674] The server transfers the consultation content to the emotion engine, which then recognizes the emotion "anxiety."

[1675] Step 5:

[1676] The server receives the analysis results of the emotion engine ("anxiety") and selects a medical expert model.

[1677] Step 6:

[1678] The medical expert model analyzes the consultation content while taking into account "anxiety" and generates the following answer: "Possible causes of headaches include stress, eye strain, and lack of sleep. We recommend that you first get plenty of rest. If your symptoms do not improve, please consult a doctor."

[1679] Step 7:

[1680] The medical expert model returns the generated answer to the server.

[1681] Step 8:

[1682] The server sends the response to the user's device.

[1683] Step 9:

[1684] The user's device will display the response on the chat screen.

[1685] By going through these processing steps, users can quickly receive an appropriate answer that corresponds to their feelings.

[1686] Example 2

[1687] 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."

[1688] Conventional expert systems provide answers without considering the user's feelings, resulting in insufficient appropriate responses. Furthermore, they lacked the functionality to efficiently search and view inquiries from multiple users, making it difficult for users to access similar past cases to resolve their own issues. Furthermore, it was difficult for users to call their favorite experts again.

[1689] 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.

[1690] In this invention, the server includes a means for using an emotion analysis engine that analyzes emotions based on user input information, a means for personalizing answers based on the emotion analysis results, and a means for anonymizing and searching for consultation details from other users. This enables the provision of more accurate and personalized answers that take the user's emotions into consideration. Furthermore, by efficiently searching and browsing similar past consultation details, users can easily obtain the information they need to resolve their issues. Furthermore, by providing a function that allows users to register their favorite experts as favorites and easily call them up, user convenience can be improved.

[1691] "User terminal" means a device through which a user accesses the system and enters or receives information.

[1692] "Input information" refers to questions, inquiries, or other data provided by users to the system.

[1693] A "server" is a central computer system that processes information sent from user terminals and performs the necessary analysis and responses.

[1694] An "expert model" is an algorithm or program with specific expertise that provides appropriate answers to user questions.

[1695] The "means for generating an answer" is a mechanism for constructing an appropriate answer to the user's input information using an expert model.

[1696] An "emotion analysis engine" is an analysis system for reading emotions from user input information.

[1697] "Anonymization" is the process of processing information so that individuals cannot be identified.

[1698] A "database" is a collection of information that stores user information, consultation details, etc. in an organized manner and can be searched or retrieved as needed.

[1699] "Favorites" is a feature that allows users to save specific expert models for easy access again.

[1700] "Personalization" means optimizing services and information according to the individual characteristics and emotions of each user.

[1701] The present invention uses a system that combines a user terminal, a server, an expert model, and an emotion analysis engine to provide quick and appropriate answers to various inquiries users have in their daily lives. Specific embodiments of this system are described below.

[1702] User Registration and Login

[1703] User Registration

[1704] When a user uses the system for the first time, they enter their name, email address, and password to register. This information is sent from the user's device to the server, which stores the received information in a database. Once registration is complete, the server sends a confirmation message to the user's device.

[1705] Log in

[1706] Existing users log in by entering their email address and password. This login information is sent from the user's device to the server, which checks the information against a database. If authentication is successful, the server generates a session ID and sends it to the user's device. The user's device retains the received session ID and displays the home screen.

[1707] Individual chat consultation

[1708] Enter and submit your consultation details

[1709] When a user wants to ask a specific question, they enter the question in text format on the chat screen of their device and press the send button. The question and session ID are then sent from the user device to the server.

[1710] Emotion recognition by emotion engine

[1711] The server passes the received consultation content to the emotion analysis engine, which analyzes the user's emotions and returns the results to the server, which selects an appropriate expert model based on the emotion analysis results.

[1712] Analysis of consultation content and generation of answers

[1713] The selected expert model analyzes the consultation content and emotional information and generates an appropriate answer that reflects the results of the emotional analysis, making it more personalized for the user.

[1714] Submitting and viewing answers

[1715] The generated answers are sent from the server to the user's device and displayed on the chat screen, allowing the user to quickly obtain the information they need.

[1716] Search and browse functions

[1717] Search for consultation content

[1718] Users can search for past consultations with other users by entering specific keywords in the search bar. This search request is sent from the user's device to the server, which then anonymizes the consultations in the database and returns them as search results.

[1719] Displaying search results

[1720] The search results are sent from the server to the user's device and displayed as a list. Users can then browse similar past consultations and find information on cases similar to their own.

[1721] Add to Favorites

[1722] Favorites Settings

[1723] If a user likes a particular expert model, he or she can register it by pressing the "Favorite" button. This favorite registration request is sent from the user terminal to the server, and the server stores it in a database.

[1724] Using Favorites

[1725] Saved favorite expert models can be easily accessed on the user's device, and can be quickly recalled when the user wants to access them again, improving convenience.

[1726] Specific examples

[1727] For example, consider the case where a user seeks medical advice. When the user enters and submits the question, "I've been having really bad headaches lately. What should I do?", the server passes this information to the emotion analysis engine. The emotion analysis engine recognizes the user's emotion as "anxiety." The server then sends this information to the medical expert model. Taking "anxiety" into consideration, the medical expert model generates the following response: "Possible causes of headaches include stress, eye strain, and lack of sleep. We recommend that you first get plenty of rest. If your symptoms persist, consult a doctor." This response is then sent from the server to the user's device, where the user can view it on the chat screen.

[1728] Prompt Sentence Examples

[1729] User: I've been having terrible headaches lately, what should I do?

[1730] AI: Headaches can be caused by stress, eye strain, lack of sleep, etc. First of all, we recommend getting plenty of rest. If your symptoms persist, consult your doctor.

[1731] The system also takes the user's emotions into account to provide more accurate and human-like answers.

[1732] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1733] Program processing flow

[1734] User Registration and Login

[1735] Step 1:

[1736] The user enters new registration information.

[1737] Enter your name, email address, and password.

[1738] How it works: The user enters the required information into the application's registration screen.

[1739] Output: Input data (name, email address, password).

[1740] Step 2:

[1741] The terminal sends the registration information to the server.

[1742] Input: Input data (name, email address, password).

[1743] Operation: The user terminal sends input data to the server.

[1744] Output: User registration information sent to the server.

[1745] Step 3:

[1746] The server stores the user information in a database.

[1747] Input: User registration information sent to the server.

[1748] How it works: The server stores the received information in a database.

[1749] Output: Status of user information saved to database.

[1750] Step 4:

[1751] The server sends a registration completion message to the terminal.

[1752] Input: Status of saving user information to database.

[1753] Operation: The server generates and sends a confirmation message to the user device indicating that the registration process is complete.

[1754] Output: Sending a registration completion message to the user's terminal.

[1755] Step 5:

[1756] The user enters their login information.

[1757] Enter your email address and password.

[1758] How it works: A user enters their email address and password on a login screen.

[1759] Output: Input data (email address, password).

[1760] Step 6:

[1761] The device sends the login information to the server.

[1762] Input: Input data (email address, password).

[1763] How it works: The user device sends login information to the server.

[1764] Output: The login information sent to the server.

[1765] Step 7:

[1766] The server checks the information against a database.

[1767] Input: The login information sent to the server.

[1768] How it works: The server checks the login information it receives against its database.

[1769] Output: Authentication success / failure result.

[1770] Step 8:

[1771] The server sends the authentication result and session ID to the terminal.

[1772] Input: The result of successful authentication.

[1773] How it works: If authentication is successful, the server generates a session ID and sends it to the user's device.

[1774] Output: A session ID is generated and sent to the user's device.

[1775] Step 9:

[1776] The device displays the home screen.

[1777] Input: Session ID.

[1778] Operation: The user device retains the received session ID and displays the home screen.

[1779] Output: Home screen display.

[1780] Individual chat consultation

[1781] Step 1:

[1782] The user inputs the consultation content.

[1783] Input: Consultation details.

[1784] How it works: The user enters the content of their inquiry in text format on the chat screen.

[1785] Output: The consultation content entered.

[1786] Step 2:

[1787] The device sends the consultation content and session ID to the server.

[1788] Input: Session ID, consultation details.

[1789] Operation: The user terminal sends the consultation content and session ID to the server.

[1790] Output: The consultation details and session ID sent to the server.

[1791] Step 3:

[1792] The server passes the consultation content to the emotion engine.

[1793] Input: The consultation content sent to the server.

[1794] Operation: The server passes the received consultation content to the emotion analysis engine.

[1795] Output: Transferring the consultation content to the emotion engine.

[1796] Step 4:

[1797] The emotion engine analyzes the user's emotions.

[1798] Input: Consultation details.

[1799] How it works: The emotion engine analyzes the content of the consultation and recognizes the user's emotions.

[1800] Output: Emotion analysis results.

[1801] Step 5:

[1802] The server obtains the emotion analysis results.

[1803] Input: Sentiment analysis results.

[1804] Operation: The server obtains the analysis results from the emotion engine.

[1805] Output: The server receives the sentiment analysis results.

[1806] Step 6:

[1807] The server selects an expert model.

[1808] Input: Sentiment analysis results.

[1809] Operation: The server selects an appropriate expert model based on the sentiment analysis results.

[1810] Output: Selected expert model.

[1811] Step 7:

[1812] The server passes the consultation content to the expert model.

[1813] Input: Consultation content, emotional information.

[1814] Operation: The server passes the consultation content and emotion information to the expert model.

[1815] Output: Transfer of input data to the expert model.

[1816] Step 8:

[1817] The expert model generates the answer.

[1818] Input: Consultation content, emotional information.

[1819] How it works: The expert model analyzes the consultation content and emotional information and generates an appropriate answer.

[1820] Output: The generated answer.

[1821] Step 9:

[1822] The server sends the generated response to the terminal.

[1823] Input: The generated answer.

[1824] Operation: The server generates a response and sends it to the user's device.

[1825] Output: Sending the answer to the user's terminal.

[1826] Step 10:

[1827] The device will display the response on the chat screen.

[1828] Input:Answer.

[1829] Operation: The user's device displays the received reply on the chat screen.

[1830] Output: The response displayed in the chat window.

[1831] Add to Favorites

[1832] Step 1:

[1833] The user presses the "Favorite" button.

[1834] Input: None.

[1835] How it works: A user who likes a particular expert model presses the "favorite" button.

[1836] Output: Favorite button click event.

[1837] Step 2:

[1838] The terminal sends a favorites registration request to the server.

[1839] Input: Favorite button click event.

[1840] Operation: The device sends a favorite registration request to the server.

[1841] Output: The favorite request sent to the server.

[1842] Step 3:

[1843] The server stores the favorites information in a database.

[1844] Input: The favorite request sent to the server.

[1845] How it works: The server stores information in a database based on the favorites request.

[1846] Output: Status of saving favorite information to database.

[1847] Step 4:

[1848] The device displays registered expert models.

[1849] Input: None.

[1850] How it works: Saved favorite expert models are easily accessible on the user's device.

[1851] Output: Display of favorite expert model on user's device.

[1852] (Application example 2)

[1853] 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."

[1854] Conventional customer support systems in virtual stores respond uniformly without considering the individual emotional state of each user, making it impossible to provide personalized, appropriate support to users. As a result, users often do not receive appropriate support, leaving them frustrated and confused. Furthermore, conventional systems lacked functionality to easily search for past user inquiries, making it difficult to quickly obtain information for self-resolving.

[1855] 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.

[1856] In this invention, the server includes means for receiving input information from a user terminal, means for selecting an appropriate expert model based on the received information, means for analyzing the user's emotions using emotion analysis means, means for selecting an expert model based on the analyzed emotion information, means for generating an appropriate answer based on the selected expert model, and means for transmitting the generated answer to the user terminal, thereby enabling personalized and appropriate support according to the user's emotional state.

[1857] "User terminal" means an electronic device used by a user, including a smartphone, tablet, or PC.

[1858] "Input information" refers to various data sent by a user via a user terminal, including text, images, audio, etc.

[1859] An "expert model" is an algorithm or program that uses knowledge in a specific field to generate appropriate answers to users' questions or inquiries.

[1860] "Emotion analysis means" refers to a system or algorithm that analyzes input information from a user and recognizes their emotional state.

[1861] "Emotion information" refers to data relating to the user's emotional state obtained by emotion analysis means.

[1862] The "selection means" refers to a method or mechanism for selecting the most appropriate model from among multiple expert models based on the user's input information and emotion information.

[1863] An "answer generator" is an algorithm or program that generates an appropriate answer to a user's question or inquiry based on a selected expert model.

[1864] "Transmitting means" refers to a system or method for sending the generated answer to the user terminal.

[1865] An embodiment of the present invention will be described.

[1866] As an application example of the invention, we implement a customer support system in a virtual store. This system mainly uses a user terminal, a server, a sentiment analysis means, and an expert model. The specific hardware used includes a smartphone, and the software used includes AWS Lambda, Amazon DynamoDB, and the OpenAI GPT-3 model.

[1867] First, a user accesses the application using a user device such as a smartphone, registers as a user if necessary, and logs in. When registering, the user enters their name, email address, and password, which are sent to the server. The server receives this information and stores it in Amazon DynamoDB. When logging in, the email address and password are verified, and if authentication is successful, a session ID is generated and returned to the user.

[1868] While browsing products in a virtual store, a user inputs and submits a specific inquiry in text format. The user's device sends this input information to the server. When the server receives the input information, it activates the emotion analysis means to recognize the user's emotional state. This emotion analysis uses the OpenAI GPT-3 model. For example, if the input inquiry is a question such as "Is this refrigerator energy efficient?", the emotion analysis means recognizes this as an emotion that indicates a desire for information.

[1869] Next, based on the analyzed emotional information, the server selects the most appropriate expert model. The selected expert model generates an appropriate answer based on the user's emotional state and the content of the inquiry. For example, if the question is about energy efficiency, a technical expert model will be selected and generate an answer that explains the details of energy efficiency. This generated answer is then sent back to the user's device via the server.

[1870] A specific example is sentiment analysis and answer generation using the following prompt sentence:

[1871] Example of a prompt for sentiment analysis:

[1872] Choose an expert model by analyzing the sentiment of the message: 'Is this refrigerator energy efficient?'

[1873] Example prompts for answer generation:

[1874] General Breakdown Technician's Answer: Provide a good answer to 'Is this refrigerator energy efficient?'

[1875] This system allows users to receive personalized support in real time based on their emotions. It also allows users to anonymously search for other inquiries and gain insights from similar past inquiries. This results in higher user satisfaction and more effective self-resolution than traditional customer support systems.

[1876] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1877] Step 1:

[1878] Receive input information from the user's device

[1879] The user uses a dedicated application in the virtual store to input and send the details of their inquiry in text format. The user's device sends this input information to the server. The input information includes the user ID and session ID.

[1880] Step 2:

[1881] Emotion analysis using emotion analysis tools

[1882] The server analyzes the user's emotions using the OpenAI GPT-3 model based on the received input information. The server sends the user's consultation details as input data to GPT-3 and receives analyzed emotional information. This emotional information indicates emotional states such as "anxiety," "doubt," and "joy."

[1883] Step 3:

[1884] Expert model selection

[1885] The server selects the most appropriate expert model based on the emotion information obtained from the emotion analysis means. For example, if the user's emotion is analyzed as "question," an expert model that is good at answering technical questions will be selected. In this selection process, an algorithm determines the most appropriate model from multiple expert models.

[1886] Step 4:

[1887] Generate answers

[1888] The server uses the selected expert model to generate an answer based on the input information. A generative AI model is used to create a specific answer based on the consultation content. This model generates an answer that is appropriate for the user's question based on the prompt text.

[1889] Step 5:

[1890] Submit your answer

[1891] The generated answers are sent from the server to the user's device, and are displayed in the user's application, allowing the user to obtain appropriate information in real time.

[1892] Specifically, the reply text is sent to the user's terminal and displayed on the chat screen of the application.

[1893] Step 6:

[1894] Search function for past consultation content (optional)

[1895] Users can search for past consultations by entering specific keywords in the search bar within the application. The search request sent from the user's device is processed by the server, and related consultations are searched for in an anonymized form. The search results are sent back to the user's device and displayed in a list within the application.

[1896] Step 7:

[1897] Register your favorite expert models (optional)

[1898] Users can register their favorite expert models as favorites. A request to add to favorites is sent from the user's device to the server, and the server stores the information in a database, allowing users to instantly access specific expert models.

[1899] 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.

[1900] 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.

[1901] 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.

[1902] [Fourth embodiment]

[1903] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1904] 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.

[1905] 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).

[1906] 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.

[1907] 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.

[1908] 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).

[1909] 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.

[1910] 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.

[1911] 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.

[1912] 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.

[1913] 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.

[1914] 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.

[1915] 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."

[1916] The present invention uses a system consisting of a user terminal, a server, and an expert model to provide prompt and appropriate answers to various inquiries in the user's daily life. A specific embodiment of this system will be described below.

[1917] User Registration and Login

[1918] User Registration:

[1919] When a user uses the app for the first time, they enter their personal information (name, email address, password) to register. This information is sent from the user's device to the server and stored in a database. Once registration is complete, a confirmation message is returned to the user's device.

[1920] Login:

[1921] Existing users log in by entering their email address and password. The entered login information is sent from the user's device to the server and checked against the database. If authentication is successful, a session ID is generated and returned to the user's device, allowing the user to access the home screen.

[1922] Individual chat consultation

[1923] Enter and submit your inquiry:

[1924] When a user wants to ask a specific question, they enter the question in text format on the chat screen of their device and press the send button. The question and session ID are then sent to the server.

[1925] Analyzing the inquiry and generating an answer:

[1926] The server analyzes the received consultation content and selects the most appropriate expert model (medical, legal, business, etc.). The selected expert model analyzes the input consultation content and generates an appropriate answer.

[1927] Submit and view your answers:

[1928] The generated answers are sent from the server to the user's device and displayed on the chat screen, allowing the user to quickly obtain the information they need.

[1929] Search and browse functions

[1930] Search for consultation content:

[1931] Users can search for past consultations with other users by entering specific keywords in the search bar. This search request is sent from the user's device to the server, and the consultations in the database are anonymized and returned as search results.

[1932] Displaying search results:

[1933] The search results sent from the server are displayed as a list on the user's device, and the user can view past similar consultations. This allows users to refer to information on cases similar to their own and use it as a reference for self-solving.

[1934] Add to Favorites

[1935] Favorite settings:

[1936] If a user likes a particular expert model, they can register it by pressing the "Favorite" button. The favorite registration request is sent from the user's terminal to the server and stored in the database.

[1937] Use Favorites:

[1938] Registered favorite expert models can be easily accessed on the user's device, and can be quickly recalled if the user wishes to access them again, improving convenience.

[1939] Specific examples

[1940] For example, consider the case of a medical consultation. When a user inputs and sends the question, "I've been having terrible headaches lately. What should I do?", the server passes this information to the medical expert model. The medical expert model generates a response such as, "Possible causes of headaches include stress, eye strain, and lack of sleep. We recommend that you first get plenty of rest. If this does not improve, please consult a doctor." This response is sent to the user's device, where the user can view it on the chat screen.

[1941] As described above, the present invention provides a system that enables users to obtain prompt and appropriate answers to a variety of problems and questions.

[1942] The processing flow will be explained below.

[1943] User Registration and Login

[1944] New Registration

[1945] Step 1:

[1946] The user launches the app and sees the signup form.

[1947] Step 2:

[1948] The user's device prompts them to enter their name, email address, and password, then presses the "Send" button.

[1949] Step 3:

[1950] The user terminal sends the input information to the server.

[1951] Step 4:

[1952] The server stores the received information in a database.

[1953] Step 5:

[1954] The server returns a registration success message to the user terminal.

[1955] Log in

[1956] Step 1:

[1957] The user enters their email address and password on the login screen and clicks the "Login" button.

[1958] Step 2:

[1959] The user terminal sends the input information to the server.

[1960] Step 3:

[1961] The server checks the received information against a database.

[1962] Step 4:

[1963] If the server matches, it generates a session ID and sends it to the user's device.

[1964] Step 5:

[1965] The user device saves the session ID and displays the home screen.

[1966] Individual chat consultation

[1967] Step 1:

[1968] The user enters the consultation content in text on the app's consultation screen and presses the send button.

[1969] Step 2:

[1970] The user device sends the consultation content and session ID to the server.

[1971] Step 3:

[1972] The server analyzes the received consultation content and selects an appropriate expert model.

[1973] Step 4:

[1974] The expert model analyzes the consultation content and generates an appropriate answer.

[1975] Step 5:

[1976] The expert model returns the generated answer to the server.

[1977] Step 6:

[1978] The server sends the response to the user's device.

[1979] Step 7:

[1980] The user's device will display the response on the chat screen.

[1981] Search and browse functions

[1982] Step 1:

[1983] A user enters keywords into the search bar and presses the search button.

[1984] Step 2:

[1985] The user terminal sends the keyword to the server.

[1986] Step 3:

[1987] The server searches the database for anonymized consultation content based on the received keywords.

[1988] Step 4:

[1989] The server sends the search results to the user's device.

[1990] Step 5:

[1991] The user's device displays a list of search results.

[1992] Add to Favorites

[1993] Step 1:

[1994] The user presses the "favorite" button for a particular expert model.

[1995] Step 2:

[1996] The user terminal sends a favorite registration request to the server along with the session ID.

[1997] Step 3:

[1998] The server stores the favorites information in a database.

[1999] Step 4:

[2000] The server sends a confirmation message to the user terminal.

[2001] Step 5:

[2002] The user terminal displays a confirmation message.

[2003] Example 1

[2004] 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."

[2005] Conventional online consultation systems have faced challenges in that it is difficult for users to quickly obtain appropriate answers to their inquiries. In addition, searching for consultation content and accessing favorite expert models is cumbersome, resulting in a poor user experience.

[2006] 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.

[2007] In this invention, the server includes means for receiving input information from a user terminal, means for storing the received information in a database, means for verifying and authenticating login information of existing users, means for analyzing the received consultation content using a natural language processing algorithm, means for selecting an appropriate expert model based on the analyzed information, means for generating an appropriate answer based on the selected expert model, means for transmitting the generated answer to the user terminal, means for anonymizing and searching for the consultation content of other users, and means for displaying a list of search results. This allows users to quickly obtain appropriate answers and also makes it easier to search for consultation content and access their favorite expert models.

[2008] A "user terminal" is an electronic device such as a computer, smartphone, or tablet that is directly operated by a user.

[2009] "Input information" refers to data that a user provides to the system via their terminal, such as name, email address, password, and consultation details.

[2010] A "server" is a computer system that processes information received from user devices and manages and analyzes information using databases and algorithms.

[2011] A "session ID" is a unique identifier used to identify and manage a user's login status.

[2012] A "natural language processing algorithm" is a technology that analyzes the content of the consultation received, understands the text data, and generates an appropriate response.

[2013] An "expert model" is a program or system that has expert knowledge in a particular field (e.g., medicine, law, business) and generates appropriate answers to inquiries from users.

[2014] A "database" is a system for systematically storing digital information and for efficiently searching and managing it.

[2015] "Authentication" is the process of verifying that the user information provided is valid.

[2016] "Analysis" is the process of breaking down received data, interpreting its meaning, and extracting relevant information.

[2017] "Anonymization" is the process of removing or transforming personally identifiable information to protect privacy.

[2018] "Search results" are lists of relevant data extracted from a database based on a specific keyword or query.

[2019] The present invention provides a system that includes a user terminal, a server, and an expert model to quickly and appropriately answer various inquiries from users. Specific embodiments of the system will be described below.

[2020] User Registration and Login

[2021] When a user first uses the system, they enter information such as their name, email address, and password from their device and send it to the server. The server stores the received information in a database and sends a registration completion confirmation message to the user's device. Existing users enter their email address and password on the login screen and send it to the server. After the server verifies the information in the database, if authentication is successful, it generates a session ID and sends it to the user's device. This allows the user to access the home screen.

[2022] Individual chat consultation

[2023] When a user wants to ask a specific question, they enter the question content in text format on the chat screen of their device and press the send button. The question content and session ID are then sent to the server. The server analyzes the received question content using a natural language processing algorithm and selects the most appropriate expert model. The selected expert model generates an answer based on the received question content, and the server sends that answer to the user's device. This allows the user to quickly check the answer on the chat screen.

[2024] Search and browse functions

[2025] Users can search for past consultations by entering keywords in the search bar. This search request is sent from the user's device to the server, which then searches the database for anonymized consultations. The search results are displayed as a list on the user's device, allowing the user to check similar past consultations.

[2026] Add to Favorites

[2027] Users can register specific expert models as favorites. A favorite registration request is sent from the user terminal to the server, and the server stores it in the database. Users can easily access the registered expert models from the favorites list.

[2028] Specific examples

[2029] For example, a user might type a question on the chat screen, such as "I've been having really bad headaches lately. What should I do?" and press the send button. The device then sends this question and the session ID to the server. The server analyzes the consultation and passes it on to the medical expert model. The medical expert model generates an answer: "Possible causes of headaches include stress, eye strain, and lack of sleep. We recommend that you first get plenty of rest. If this does not improve, consult a doctor." and sends this answer to the user's device. The user can then check this answer on the chat screen.

[2030] As described above, the present invention provides a system that enables users to obtain prompt and appropriate answers to a variety of problems and questions.

[2031] The flow of the identification process in the first embodiment will be described with reference to FIG.

[2032] Step 1:

[2033] User Registration

[2034] The user enters information such as name, email address, and password into the user device.

[2035] Input: Name, Email Address, Password

[2036] Output: Input information

[2037] The terminal transmits the input information to the server.

[2038] Input: Input information

[2039] Output: Data sent to the server

[2040] The server stores the received information in a database.

[2041] Input: Send data

[2042] Output: Information stored in the database

[2043] The server generates a confirmation message for completing the registration and sends it to the user terminal.

[2044] Input: Information stored in a database

[2045] Output: Confirmation message sent

[2046] Step 2:

[2047] Log in

[2048] The user enters their email address and password into the user terminal.

[2049] Input: Email address, password

[2050] Output: Input information

[2051] The terminal transmits the entered login information to the server.

[2052] Input: Input information

[2053] Output: Data sent to the server

[2054] The server checks the information against a database and authenticates it.

[2055] Input: Send data

[2056] Output: Authentication result

[2057] If the authentication is successful, the server generates a session ID and sends it to the user terminal.

[2058] Input: Authentication result

[2059] Output: Sending the session ID

[2060] The user device displays the home screen.

[2061] Input: Session ID

[2062] Output: Home screen display

[2063] Step 3:

[2064] Start individual chat consultation

[2065] The user enters the content of the consultation in text format on the chat screen and presses the send button.

[2066] Input: Consultation details

[2067] Output: Input consultation content

[2068] The terminal sends the consultation content and the session ID to the server.

[2069] Input: Consultation details, session ID

[2070] Output: Data sent to the server

[2071] Step 4:

[2072] Analysis of consultation content and generation of answers

[2073] The server analyzes the received consultation content using a natural language processing algorithm.

[2074] Input: Send data

[2075] Output: Analyzed consultation content

[2076] The server selects an appropriate expert model based on the analyzed consultation content.

[2077] Input: Analyzed consultation content

[2078] Output: Selected expert model

[2079] The expert model analyzes the content of the consultation and generates an appropriate answer.

[2080] Input: Analyzed consultation content

[2081] Output: The generated answer

[2082] Step 5:

[2083] Submitting and viewing answers

[2084] The server sends the generated response to the user terminal.

[2085] Input: Generated Answer

[2086] Output: Sending the answer

[2087] The user's device will display the response on the chat screen.

[2088] Input: Submit answer

[2089] Output: Display reply on chat screen

[2090] Step 6:

[2091] Search for consultation content

[2092] Users enter keywords into the search bar.

[2093] Input: Search keyword

[2094] Output: Input search keywords

[2095] The terminal sends a search request to the server.

[2096] Input: Enter search keywords

[2097] output: Search request to the server

[2098] The server searches the database for anonymized consultation content.

[2099] Input: Search request

[2100] Output: Search results

[2101] The server sends the search results to the user's terminal.

[2102] Input: Search results

[2103] output: sent to the user's terminal

[2104] The user's device displays a list of search results.

[2105] input:Submitted search results

[2106] output: Search results display

[2107] Step 7:

[2108] Registering and using favorites

[2109] Users can register specific expert models as "favorites."

[2110] Input: Favorite registration request

[2111] output: The input registration request

[2112] The terminal sends a favorites registration request to the server.

[2113] input:The input registration request

[2114] output: Send to server

[2115] The server stores the received requests in a database.

[2116] input:Submitted registration request

[2117] output: Save to database

[2118] Users access registered expert models from their favorites list.

[2119] input:Retrieve from database

[2120] output: Access to favorite models

[2121] (Application example 1)

[2122] 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."

[2123] Conventional food delivery systems have difficulty suggesting menus that take into account the user's health condition and nutritional balance, or providing expert advice in real time to individual questions. Furthermore, they lack the functionality to refer to frequently used menus or the health-related inquiries of other users, making them ineffective for self-resolving. There is a need to improve this situation.

[2124] 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.

[2125] In this invention, the server includes means for receiving input information from a user terminal, means for selecting an appropriate expert model based on the received information, means for generating an appropriate answer based on the selected expert model, means for transmitting the generated answer to the user terminal, means for analyzing the content of the consultation from the user terminal and selecting a food and nutrition expert model, means for generating a health menu based on the food and nutrition expert model, and means for presenting the generated menu to the user terminal. This allows users to receive real-time menu suggestions that take into account their health status and nutritional balance. Furthermore, users can search for other users' consultations anonymized and refer to menus generated based on similar past consultations, which can be used as a reference for self-solving.

[2126] "User terminal" refers to the communication device used by the user, including smartphones, tablets, and personal computers.

[2127] "Input Information" refers to text and selection information sent by a user through a user terminal.

[2128] "Server" refers to a computer system that receives, analyzes, and processes information from user terminals.

[2129] An "expert model" refers to an AI model with a knowledge base specialized in a particular field (e.g., medicine, law, nutrition, etc.).

[2130] A "food and nutrition expert model" refers to an AI model with specialized knowledge about food and nutrition.

[2131] "Menu" means the food options and suggestions provided to Users.

[2132] "Health menu" refers to meal plans and food suggestions generated taking into account the user's health status and nutritional balance.

[2133] "Anonymization" refers to the process of processing information so that a specific user cannot be identified.

[2134] "Consultation content" refers to questions or matters of consultation that users send to the system.

[2135] "Generated answers" refers to answers generated by the server based on an expert model.

[2136] "Session ID" refers to a unique identifier generated by the server after a user logs in.

[2137] "Display" refers to the process of visually presenting information on the screen of a user device.

[2138] "Search results" refers to the list of relevant information returned by the server in response to a user's search request.

[2139] "Favorites" refers to a feature that allows users to register specific information for easy access later.

[2140] The present invention is a system that uses a user terminal, a server, and an expert model to suggest menus that take into account the user's health condition and nutritional balance. A specific embodiment of this system is described below.

[2141] User Registration and Login

[2142] User Registration:

[2143] When a user first uses the system, they enter detailed information such as their name, email address, password, and dietary and allergy information from their device and send it to the server. This information is stored in the server's database. Once registration is complete, a confirmation message is returned to the user's device.

[2144] Login:

[2145] Existing users log in by entering their email address and password. The login information is sent to the server and checked against a database. If authentication is successful, a session ID is generated and returned to the user's device, allowing the user to access the home screen.

[2146] Individual chat consultation

[2147] Enter and submit your inquiry:

[2148] If a user wants to ask for advice about a specific dietary issue, they enter the content of their inquiry in text format on the chat screen of their device and press the send button. The content of their inquiry and the session ID are then sent to the server.

[2149] Analyzing the inquiry and generating an answer:

[2150] The server analyzes the received consultation content and selects the most appropriate food and nutrition expert model. The selected food and nutrition expert model analyzes the input consultation content and generates a health menu.

[2151] Submit and view your answers:

[2152] The generated menu is sent from the server to the user's device and displayed on the chat screen, allowing the user to quickly obtain a healthy menu.

[2153] For example, when a user inputs and sends a question such as "What healthy menu do you recommend for dinner tonight?", the server passes this information to the food and nutrition expert model. The food and nutrition expert model generates a menu that reads, "For a healthy dinner, we recommend steamed vegetables and chicken breast." This menu is sent to the user's device, where the user can view it on the chat screen.

[2154] Search and browse functions

[2155] Search for consultation content:

[2156] Users can search for past consultations with other users by entering specific keywords in the search bar. This search request is sent to the server, which returns the consultations in the database in anonymized form.

[2157] Displaying search results:

[2158] The search results sent from the server are displayed as a list on the user's device, and the user can view past similar consultations. This allows users to refer to information on cases similar to their own and use it as a reference for self-solving.

[2159] Add to Favorites

[2160] Favorite settings:

[2161] If a user likes a particular food nutrition expert model or generated menu, they can register it by pressing the "Favorite" button. The request to register as a favorite is sent to the server and stored in the database.

[2162] Use Favorites:

[2163] Registered favorite models and menus can be easily accessed on the user's device, and if the user wants to access them again, they can be quickly recalled, improving convenience.

[2164] This system mainly uses the following hardware and software.

[2165] Hardware:

[2166] Server: Cloud server (e.g. AWS, Google Cloud, etc.)

[2167] User device: Smartphone (iOS / Android)

[2168] software:

[2169] Framework: Flask (Python)

[2170] Database: SQLite

[2171] AI model: Language model (e.g., OpenAI's GPT)

[2172] For example, if a user inputs "What low-calorie menu would you recommend for today's lunch?", the food and nutrition expert model will generate a menu that reads, "Steamed chicken salad and smoothies are recommended as low-calorie menu items." This menu is sent to the user's device, and the user can view it on the chat screen. Examples of prompt sentences are as follows:

[2173] User: "What's a low-calorie option for lunch today?"

[2174] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[2175] Step 1:

[2176] The user sends input information such as name, email address, password, and dietary and allergy information from the user device. The server receives this information and stores it in a database. Input: User information. Output: User information stored in the database.

[2177] Step 2:

[2178] An existing user logs in with their email address and password. The login information sent from the user's device to the server is checked against the database. If authentication is successful, the server generates a session ID and returns it to the user's device. Input: Login information. Output: Session ID.

[2179] Step 3:

[2180] The user enters the consultation content in text format on the chat screen of the user device and presses the send button. This sends the consultation content and session ID to the server. Input: Consultation content and session ID. Output: Data sent to the server.

[2181] Step 4:

[2182] The server analyzes the received consultation content and selects the most appropriate food and nutrition expert model. The selected model performs analysis based on the input consultation content. Input: Consultation content. Output: Selected food and nutrition expert model.

[2183] Step 5:

[2184] The selected food and nutrition expert model generates a health-related menu based on the consultation. For example, if a user inputs "What healthy menu would you recommend for dinner today?", the model generates a menu that says "For a healthy dinner, I recommend steamed vegetables and chicken breast." Input: Consultation. Output: Generated menu.

[2185] Step 6:

[2186] The generated menu is sent from the server to the user's device and displayed on the chat screen. Input: Generated menu. Output: Menu sent to the user's device.

[2187] Step 7:

[2188] If the user likes a particular food nutrition expert model or generated menu, they can register it by pressing the "Favorite" button. This favorite registration request is sent to the server and stored in the database. Input: Favorite registration request. Output: Favorite information stored in the database.

[2189] Step 8:

[2190] When the user wants to access a favorite item again, the server extracts the item from the database based on the request from the user's device and sends it to the user's device. Input: Request from the user's device. Output: Extracted item.

[2191] Step 9:

[2192] A user enters a specific keyword in the search bar to search for past consultations from other users. This search request is sent to the server, and the consultations in the database are anonymized and returned as search results. Input: Keyword. Output: Anonymized search results.

[2193] Step 10:

[2194] The search results sent from the server are displayed as a list on the user's device. The user can view past similar consultations. Input: Search results. Output: List displayed on the user's device.

[2195] Example prompt sentence:

[2196] User: "What's a low-calorie option for lunch today?"

[2197] 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.

[2198] The present invention uses a system that combines a user terminal, a server, an expert model, and an emotion engine to provide quick and appropriate answers to various inquiries users have in their daily lives. A specific embodiment of this system will be described below.

[2199] User Registration and Login

[2200] User Registration:

[2201] When a user uses the app for the first time, they enter their personal information (name, email address, password) to register. This information is sent from the user's device to the server and stored in a database. Once registration is complete, a confirmation message is returned to the user's device.

[2202] Login:

[2203] Existing users log in by entering their email address and password. The entered login information is sent from the user's device to the server and checked against the database. If authentication is successful, a session ID is generated and returned to the user's device, allowing the user to access the home screen.

[2204] Individual chat consultation

[2205] Enter and submit your inquiry:

[2206] When a user wants to ask a specific question, they enter the question in text format on the chat screen of their device and press the send button. The question and session ID are then sent to the server.

[2207] Emotion Recognition with Emotion Engine:

[2208] When the server receives the consultation content, it analyzes the user's emotions using an emotion engine. Based on the analysis results, it selects an appropriate expert model.

[2209] Analyzing the inquiry and generating an answer:

[2210] The selected expert model analyzes the content of the consultation and the emotional information from the emotion engine to generate an appropriate answer. This analysis allows for more personalized answers based on the user's emotions.

[2211] Submit and view your answers:

[2212] The generated answers are sent from the server to the user's device and displayed on the chat screen, allowing the user to quickly obtain the information they need.

[2213] Search and browse functions

[2214] Search for consultation content:

[2215] Users can search for past consultations with other users by entering specific keywords in the search bar. This search request is sent from the user's device to the server, and the consultations in the database are anonymized and returned as search results.

[2216] Displaying search results:

[2217] The search results sent from the server are displayed as a list on the user's device, and the user can view past similar consultations. This allows users to refer to information on cases similar to their own and use it as a reference for self-solving.

[2218] Add to Favorites

[2219] Favorite settings:

[2220] If a user likes a particular expert model, they can register it by pressing the "Favorite" button. The favorite registration request is sent from the user's terminal to the server and stored in the database.

[2221] Use Favorites:

[2222] Registered favorite expert models can be easily accessed on the user's device, and can be quickly recalled if the user wishes to access them again, improving convenience.

[2223] Specific examples

[2224] For example, consider the case of a medical consultation. When a user inputs and sends the question, "I've been having really bad headaches lately. What should I do?", the server passes this content to the emotion engine. The emotion engine recognizes the user's emotion as "anxiety" and passes this information along with the medical expert model. The medical expert model takes "anxiety" into consideration and generates a response such as, "Possible causes of headaches include stress, eye strain, and lack of sleep. We recommend that you first get plenty of rest. If your symptoms persist, consult a doctor." This response is sent to the user's device and can be viewed by the user on the chat screen.

[2225] As described above, the present invention realizes a system that provides more accurate and human-like answers by taking the user's emotions into consideration.

[2226] The processing flow will be explained below.

[2227] User Registration and Login

[2228] New Registration

[2229] Step 1:

[2230] The user launches the app and is presented with a sign-up form.

[2231] Step 2:

[2232] The user's device prompts them to enter their name, email address, and password, and then the user presses the "Send" button.

[2233] Step 3:

[2234] The user terminal sends the input information to the server.

[2235] Step 4:

[2236] The server stores the received information in a database.

[2237] Step 5:

[2238] The server generates a registration success message and sends it to the user terminal.

[2239] Log in

[2240] Step 1:

[2241] The user enters their email address and password on the login screen and clicks the "Login" button.

[2242] Step 2:

[2243] The user terminal sends the input information to the server.

[2244] Step 3:

[2245] The server checks the received login information against its database.

[2246] Step 4:

[2247] If the server is successful in the authentication, it generates a session ID and sends it to the user's terminal.

[2248] Step 5:

[2249] The user device saves the session ID and displays the home screen.

[2250] Individual chat consultation

[2251] Step 1:

[2252] The user accesses the consultation screen, enters the consultation content in text format, and presses the send button.

[2253] Step 2:

[2254] The user device sends the consultation content and session ID to the server.

[2255] Step 3:

[2256] The server receives the consultation content and transfers it to the emotion engine.

[2257] Step 4:

[2258] The emotion engine analyzes the content of the consultation and recognizes the user's emotions (e.g., "anxiety," "impatience," etc.).

[2259] Step 5:

[2260] The server receives the analysis results of the emotion engine and selects an appropriate expert model.

[2261] Step 6:

[2262] The expert model generates an appropriate answer based on the consultation content and emotional information.

[2263] Step 7:

[2264] The expert model returns the generated answer to the server.

[2265] Step 8:

[2266] The server sends the response to the user's device.

[2267] Step 9:

[2268] The user's device will display the response on the chat screen.

[2269] Search and browse functions

[2270] Step 1:

[2271] A user enters keywords into the search bar and presses the search button.

[2272] Step 2:

[2273] The user terminal sends the keyword to the server.

[2274] Step 3:

[2275] The server searches the database for anonymized consultation content based on the received keywords.

[2276] Step 4:

[2277] The server sends the search results to the user's device.

[2278] Step 5:

[2279] The user's device displays a list of search results.

[2280] Add to Favorites

[2281] Step 1:

[2282] The user presses the "favorite" button for a particular expert model.

[2283] Step 2:

[2284] The user terminal sends a "favorite" registration request to the server along with the session ID.

[2285] Step 3:

[2286] The server stores the "favorites" information in a database.

[2287] Step 4:

[2288] The server sends a confirmation message to the user terminal.

[2289] Step 5:

[2290] The user terminal displays a confirmation message.

[2291] Example: Medical consultation

[2292] Step 1:

[2293] The user logs in, accesses the home screen, and selects "Medical Consultation."

[2294] Step 2:

[2295] The user enters the question they want to ask, such as "I've been having terrible headaches lately. What should I do?" and presses the send button.

[2296] Step 3:

[2297] The user device sends the consultation content and session ID to the server.

[2298] Step 4:

[2299] The server transfers the consultation content to the emotion engine, which then recognizes the emotion "anxiety."

[2300] Step 5:

[2301] The server receives the analysis results of the emotion engine ("anxiety") and selects a medical expert model.

[2302] Step 6:

[2303] The medical expert model analyzes the consultation content while taking into account "anxiety" and generates the following answer: "Possible causes of headaches include stress, eye strain, and lack of sleep. We recommend that you first get plenty of rest. If your symptoms do not improve, please consult a doctor."

[2304] Step 7:

[2305] The medical expert model returns the generated answer to the server.

[2306] Step 8:

[2307] The server sends the response to the user's device.

[2308] Step 9:

[2309] The user's device will display the response on the chat screen.

[2310] By going through these processing steps, users can quickly receive an appropriate answer that corresponds to their feelings.

[2311] Example 2

[2312] 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."

[2313] Conventional expert systems provide answers without considering the user's feelings, resulting in insufficient appropriate responses. Furthermore, they lacked the functionality to efficiently search and view inquiries from multiple users, making it difficult for users to access similar past cases to resolve their own issues. Furthermore, it was difficult for users to call their favorite experts again.

[2314] 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.

[2315] In this invention, the server includes a means for using an emotion analysis engine that analyzes emotions based on user input information, a means for personalizing answers based on the emotion analysis results, and a means for anonymizing and searching for consultation details from other users. This enables the provision of more accurate and personalized answers that take the user's emotions into consideration. Furthermore, by efficiently searching and browsing similar past consultation details, users can easily obtain the information they need to resolve their issues. Furthermore, by providing a function that allows users to register their favorite experts as favorites and easily call them up, user convenience can be improved.

[2316] "User terminal" means a device through which a user accesses the system and enters or receives information.

[2317] "Input information" refers to questions, inquiries, or other data provided by users to the system.

[2318] A "server" is a central computer system that processes information sent from user terminals and performs the necessary analysis and responses.

[2319] An "expert model" is an algorithm or program with specific expertise that provides appropriate answers to user questions.

[2320] The "means for generating an answer" is a mechanism for constructing an appropriate answer to the user's input information using an expert model.

[2321] An "emotion analysis engine" is an analysis system for reading emotions from user input information.

[2322] "Anonymization" is the process of processing information so that individuals cannot be identified.

[2323] A "database" is a collection of information that stores user information, consultation details, etc. in an organized manner and can be searched or retrieved as needed.

[2324] "Favorites" is a feature that allows users to save specific expert models for easy access again.

[2325] "Personalization" means optimizing services and information according to the individual characteristics and emotions of each user.

[2326] The present invention uses a system that combines a user terminal, a server, an expert model, and an emotion analysis engine to provide quick and appropriate answers to various inquiries users have in their daily lives. Specific embodiments of this system are described below.

[2327] User Registration and Login

[2328] User Registration

[2329] When a user uses the system for the first time, they enter their name, email address, and password to register. This information is sent from the user's device to the server, which stores the received information in a database. Once registration is complete, the server sends a confirmation message to the user's device.

[2330] Log in

[2331] Existing users log in by entering their email address and password. This login information is sent from the user's device to the server, which checks the information against a database. If authentication is successful, the server generates a session ID and sends it to the user's device. The user's device retains the received session ID and displays the home screen.

[2332] Individual chat consultation

[2333] Enter and submit your consultation details

[2334] When a user wants to ask a specific question, they enter the question in text format on the chat screen of their device and press the send button. The question and session ID are then sent from the user device to the server.

[2335] Emotion recognition by emotion engine

[2336] The server passes the received consultation content to the emotion analysis engine, which analyzes the user's emotions and returns the results to the server, which selects an appropriate expert model based on the emotion analysis results.

[2337] Analysis of consultation content and generation of answers

[2338] The selected expert model analyzes the consultation content and emotional information and generates an appropriate answer that reflects the results of the emotional analysis, making it more personalized for the user.

[2339] Submitting and viewing answers

[2340] The generated answers are sent from the server to the user's device and displayed on the chat screen, allowing the user to quickly obtain the information they need.

[2341] Search and browse functions

[2342] Search for consultation content

[2343] Users can search for past consultations with other users by entering specific keywords in the search bar. This search request is sent from the user's device to the server, which then anonymizes the consultations in the database and returns them as search results.

[2344] Displaying search results

[2345] The search results are sent from the server to the user's device and displayed as a list. Users can then browse similar past consultations and find information on cases similar to their own.

[2346] Add to Favorites

[2347] Favorites Settings

[2348] If a user likes a particular expert model, he or she can register it by pressing the "Favorite" button. This favorite registration request is sent from the user terminal to the server, and the server stores it in a database.

[2349] Using Favorites

[2350] Saved favorite expert models can be easily accessed on the user's device, and can be quickly recalled when the user wants to access them again, improving convenience.

[2351] Specific examples

[2352] For example, consider the case where a user seeks medical advice. When the user enters and submits the question, "I've been having really bad headaches lately. What should I do?", the server passes this information to the emotion analysis engine. The emotion analysis engine recognizes the user's emotion as "anxiety." The server then sends this information to the medical expert model. Taking "anxiety" into consideration, the medical expert model generates the following response: "Possible causes of headaches include stress, eye strain, and lack of sleep. We recommend that you first get plenty of rest. If your symptoms persist, consult a doctor." This response is then sent from the server to the user's device, where the user can view it on the chat screen.

[2353] Prompt Sentence Examples

[2354] User: I've been having terrible headaches lately, what should I do?

[2355] AI: Headaches can be caused by stress, eye strain, lack of sleep, etc. First of all, we recommend getting plenty of rest. If your symptoms persist, consult your doctor.

[2356] The system also takes the user's emotions into account to provide more accurate and human-like answers.

[2357] The flow of the identification process in the second embodiment will be described with reference to FIG.

[2358] Program processing flow

[2359] User Registration and Login

[2360] Step 1:

[2361] The user enters new registration information.

[2362] Enter your name, email address, and password.

[2363] How it works: The user enters the required information into the application's registration screen.

[2364] Output: Input data (name, email address, password).

[2365] Step 2:

[2366] The terminal sends the registration information to the server.

[2367] Input: Input data (name, email address, password).

[2368] Operation: The user terminal sends input data to the server.

[2369] Output: User registration information sent to the server.

[2370] Step 3:

[2371] The server stores the user information in a database.

[2372] Input: User registration information sent to the server.

[2373] How it works: The server stores the received information in a database.

[2374] Output: Status of user information saved to database.

[2375] Step 4:

[2376] The server sends a registration completion message to the terminal.

[2377] Input: Status of saving user information to database.

[2378] Operation: The server generates and sends a confirmation message to the user device indicating that the registration process is complete.

[2379] Output: Sending a registration completion message to the user's terminal.

[2380] Step 5:

[2381] The user enters their login information.

[2382] Enter your email address and password.

[2383] How it works: A user enters their email address and password on a login screen.

[2384] Output: Input data (email address, password).

[2385] Step 6:

[2386] The device sends the login information to the server.

[2387] Input: Input data (email address, password).

[2388] How it works: The user device sends login information to the server.

[2389] Output: The login information sent to the server.

[2390] Step 7:

[2391] The server checks the information against a database.

[2392] Input: The login information sent to the server.

[2393] How it works: The server checks the login information it receives against its database.

[2394] Output: Authentication success / failure result.

[2395] Step 8:

[2396] The server sends the authentication result and session ID to the terminal.

[2397] Input: The result of successful authentication.

[2398] How it works: If authentication is successful, the server generates a session ID and sends it to the user's device.

[2399] Output: A session ID is generated and sent to the user's device.

[2400] Step 9:

[2401] The device displays the home screen.

[2402] Input: Session ID.

[2403] Operation: The user device retains the received session ID and displays the home screen.

[2404] Output: Home screen display.

[2405] Individual chat consultation

[2406] Step 1:

[2407] The user inputs the consultation content.

[2408] Input: Consultation details.

[2409] How it works: The user enters the content of their inquiry in text format on the chat screen.

[2410] Output: The consultation content entered.

[2411] Step 2:

[2412] The device sends the consultation content and session ID to the server.

[2413] Input: Session ID, consultation details.

[2414] Operation: The user terminal sends the consultation content and session ID to the server.

[2415] Output: The consultation details and session ID sent to the server.

[2416] Step 3:

[2417] The server passes the consultation content to the emotion engine.

[2418] Input: The consultation content sent to the server.

[2419] Operation: The server passes the received consultation content to the emotion analysis engine.

[2420] Output: Transferring the consultation content to the emotion engine.

[2421] Step 4:

[2422] The emotion engine analyzes the user's emotions.

[2423] Input: Consultation details.

[2424] How it works: The emotion engine analyzes the content of the consultation and recognizes the user's emotions.

[2425] Output: Emotion analysis results.

[2426] Step 5:

[2427] The server obtains the emotion analysis results.

[2428] Input: Sentiment analysis results.

[2429] Operation: The server obtains the analysis results from the emotion engine.

[2430] Output: The server receives the sentiment analysis results.

[2431] Step 6:

[2432] The server selects an expert model.

[2433] Input: Sentiment analysis results.

[2434] Operation: The server selects an appropriate expert model based on the sentiment analysis results.

[2435] Output: Selected expert model.

[2436] Step 7:

[2437] The server passes the consultation content to the expert model.

[2438] Input: Consultation content, emotional information.

[2439] Operation: The server passes the consultation content and emotion information to the expert model.

[2440] Output: Transfer of input data to the expert model.

[2441] Step 8:

[2442] The expert model generates the answer.

[2443] Input: Consultation content, emotional information.

[2444] How it works: The expert model analyzes the consultation content and emotional information and generates an appropriate answer.

[2445] Output: The generated answer.

[2446] Step 9:

[2447] The server sends the generated response to the terminal.

[2448] Input: The generated answer.

[2449] Operation: The server generates a response and sends it to the user's device.

[2450] Output: Sending the answer to the user's terminal.

[2451] Step 10:

[2452] The device will display the response on the chat screen.

[2453] Input:Answer.

[2454] Operation: The user's device displays the received reply on the chat screen.

[2455] Output: The response displayed in the chat window.

[2456] Add to Favorites

[2457] Step 1:

[2458] The user presses the "Favorite" button.

[2459] Input: None.

[2460] How it works: A user who likes a particular expert model presses the "favorite" button.

[2461] Output: Favorite button click event.

[2462] Step 2:

[2463] The terminal sends a favorites registration request to the server.

[2464] Input: Favorite button click event.

[2465] Operation: The device sends a favorite registration request to the server.

[2466] Output: The favorite request sent to the server.

[2467] Step 3:

[2468] The server stores the favorites information in a database.

[2469] Input: The favorite request sent to the server.

[2470] How it works: The server stores information in a database based on the favorites request.

[2471] Output: Status of saving favorite information to database.

[2472] Step 4:

[2473] The device displays registered expert models.

[2474] Input: None.

[2475] How it works: Saved favorite expert models are easily accessible on the user's device.

[2476] Output: Display of favorite expert model on user's device.

[2477] (Application example 2)

[2478] 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."

[2479] Conventional customer support systems in virtual stores respond uniformly without considering the individual emotional state of each user, making it impossible to provide personalized, appropriate support to users. As a result, users often do not receive appropriate support, leaving them frustrated and confused. Furthermore, conventional systems lacked functionality to easily search for past user inquiries, making it difficult to quickly obtain information for self-resolving.

[2480] 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.

[2481] In this invention, the server includes means for receiving input information from a user terminal, means for selecting an appropriate expert model based on the received information, means for analyzing the user's emotions using emotion analysis means, means for selecting an expert model based on the analyzed emotion information, means for generating an appropriate answer based on the selected expert model, and means for transmitting the generated answer to the user terminal, thereby enabling personalized and appropriate support according to the user's emotional state.

[2482] "User terminal" means an electronic device used by a user, including a smartphone, tablet, or PC.

[2483] "Input information" refers to various data sent by a user via a user terminal, including text, images, audio, etc.

[2484] An "expert model" is an algorithm or program that uses knowledge in a specific field to generate appropriate answers to users' questions or inquiries.

[2485] "Emotion analysis means" refers to a system or algorithm that analyzes input information from a user and recognizes their emotional state.

[2486] "Emotion information" refers to data relating to the user's emotional state obtained by emotion analysis means.

[2487] The "selection means" refers to a method or mechanism for selecting the most appropriate model from among multiple expert models based on the user's input information and emotion information.

[2488] An "answer generator" is an algorithm or program that generates an appropriate answer to a user's question or inquiry based on a selected expert model.

[2489] "Transmitting means" refers to a system or method for sending the generated answer to the user terminal.

[2490] An embodiment of the present invention will be described.

[2491] As an application example of the invention, we implement a customer support system in a virtual store. This system mainly uses a user terminal, a server, a sentiment analysis means, and an expert model. The specific hardware used includes a smartphone, and the software used includes AWS Lambda, Amazon DynamoDB, and the OpenAI GPT-3 model.

[2492] First, a user accesses the application using a user device such as a smartphone, registers as a user if necessary, and logs in. When registering, the user enters their name, email address, and password, which are sent to the server. The server receives this information and stores it in Amazon DynamoDB. When logging in, the email address and password are verified, and if authentication is successful, a session ID is generated and returned to the user.

[2493] While browsing products in a virtual store, a user inputs and submits a specific inquiry in text format. The user's device sends this input information to the server. When the server receives the input information, it activates the emotion analysis means to recognize the user's emotional state. This emotion analysis uses the OpenAI GPT-3 model. For example, if the input inquiry is a question such as "Is this refrigerator energy efficient?", the emotion analysis means recognizes this as an emotion that indicates a desire for information.

[2494] Next, based on the analyzed emotional information, the server selects the most appropriate expert model. The selected expert model generates an appropriate answer based on the user's emotional state and the content of the inquiry. For example, if the question is about energy efficiency, a technical expert model will be selected and generate an answer that explains the details of energy efficiency. This generated answer is then sent back to the user's device via the server.

[2495] A specific example is sentiment analysis and answer generation using the following prompt sentence:

[2496] Example of a prompt for sentiment analysis:

[2497] Choose an expert model by analyzing the sentiment of the message: 'Is this refrigerator energy efficient?'

[2498] Example prompts for answer generation:

[2499] General Breakdown Technician's Answer: Provide a good answer to 'Is this refrigerator energy efficient?'

[2500] This system allows users to receive personalized support in real time based on their emotions. It also allows users to anonymously search for other inquiries and gain insights from similar past inquiries. This results in higher user satisfaction and more effective self-resolution than traditional customer support systems.

[2501] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2502] Step 1:

[2503] Receive input information from the user's device

[2504] The user uses a dedicated application in the virtual store to input and send the details of their inquiry in text format. The user's device sends this input information to the server. The input information includes the user ID and session ID.

[2505] Step 2:

[2506] Emotion analysis using emotion analysis tools

[2507] The server analyzes the user's emotions using the OpenAI GPT-3 model based on the received input information. The server sends the user's consultation details as input data to GPT-3 and receives analyzed emotional information. This emotional information indicates emotional states such as "anxiety," "doubt," and "joy."

[2508] Step 3:

[2509] Expert model selection

[2510] The server selects the most appropriate expert model based on the emotion information obtained from the emotion analysis means. For example, if the user's emotion is analyzed as "question," an expert model that is good at answering technical questions will be selected. In this selection process, an algorithm determines the most appropriate model from multiple expert models.

[2511] Step 4:

[2512] Generate answers

[2513] The server uses the selected expert model to generate an answer based on the input information. A generative AI model is used to create a specific answer based on the consultation content. This model generates an answer that is appropriate for the user's question based on the prompt text.

[2514] Step 5:

[2515] Submit your answer

[2516] The generated answers are sent from the server to the user's device, and are displayed in the user's application, allowing the user to obtain appropriate information in real time.

[2517] Specifically, the reply text is sent to the user's terminal and displayed on the chat screen of the application.

[2518] Step 6:

[2519] Search function for past consultation content (optional)

[2520] Users can search for past consultations by entering specific keywords in the search bar within the application. The search request sent from the user's device is processed by the server, and related consultations are searched for in an anonymized form. The search results are sent back to the user's device and displayed in a list within the application.

[2521] Step 7:

[2522] Register your favorite expert models (optional)

[2523] Users can register their favorite expert models as favorites. A request to add to favorites is sent from the user's device to the server, and the server stores the information in a database, allowing users to instantly access specific expert models.

[2524] 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.

[2525] 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.

[2526] 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.

[2527] 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.

[2528] 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.

[2529] 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.

[2530] 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).

[2531] 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.

[2532] 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."

[2533] 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.

[2534] 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).

[2535] 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.

[2536] 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.

[2537] 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.

[2538] 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.

[2539] 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.

[2540] 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.

[2541] 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.

[2542] 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.

[2543] 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.

[2544] 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.

[2545] The following is further disclosed regarding the above embodiment.

[2546] (Claim 1)

[2547] means for receiving input information from a user terminal;

[2548] a means for selecting an appropriate expert model based on the received information;

[2549] means for generating an appropriate answer based on the selected expert model;

[2550] means for transmitting the generated answer to a user terminal;

[2551] A system including:

[2552] (Claim 2)

[2553] A means for users to mark individual expert models as favorites;

[2554] A means for instantly calling up a favorited expert model;

[2555] The system of claim 1 further comprising:

[2556] (Claim 3)

[2557] A way to search for other users' consultations anonymously,

[2558] A means for displaying search results in a list on the user's device;

[2559] The system of claim 1 further comprising:

[2560] "Example 1"

[2561] (Claim 1)

[2562] means for receiving input information from a user terminal;

[2563] means for storing the received information in a database;

[2564] A means of verifying and authenticating existing user login information; and

[2565] A means for analyzing the received consultation content using a natural language processing algorithm;

[2566] A means for selecting an appropriate expert model based on the analyzed information;

[2567] means for generating an appropriate answer based on the selected expert model;

[2568] means for transmitting the generated answer to a user terminal;

[2569] A way to search for other users' consultations anonymously,

[2570] A means for displaying search results in a list;

[2571] A system including:

[2572] (Claim 2)

[2573] A means for users to mark individual expert models as favorites;

[2574] A means for instantly calling up a favorited expert model;

[2575] The system of claim 1 further comprising:

[2576] (Claim 3)

[2577] A means for receiving the consultation content and session ID entered from the user terminal;

[2578] A means for transmitting the generated session ID to the user terminal;

[2579] a means for displaying a home screen on the user device;

[2580] a means for authenticating a user using the received session ID;

[2581] The system of claim 1 further comprising:

[2582] "Application Example 1"

[2583] (Claim 1)

[2584] means for receiving input information from a user terminal;

[2585] a means for selecting an appropriate expert model based on the received information;

[2586] means for generating an appropriate answer based on the selected expert model;

[2587] means for transmitting the generated answer to a user terminal;

[2588] A means for analyzing the consultation content from the user's terminal and selecting a food nutrition expert model;

[2589] a means for generating health menus based on a food and nutrition expert model;

[2590] means for presenting the generated menu to a user terminal;

[2591] A system including:

[2592] (Claim 2)

[2593] A means for users to mark individual expert models as favorites;

[2594] A means for instantly calling up a favorited expert model;

[2595] a means for registering a menu generated based on the food and nutrition expert model as a favorite;

[2596] A means for instantly calling up a favorite menu;

[2597] 10. The system of claim 1, comprising:

[2598] (Claim 3)

[2599] A way to search for other users' consultations anonymously,

[2600] A means for displaying search results in a list on the user's device;

[2601] A means to search for similar health menus generated based on anonymized past consultations;

[2602] a means for displaying the menu generated as a result of the search on the user's device;

[2603] 10. The system of claim 1, comprising:

[2604] "Example 2: Combining Emotion Engines"

[2605] (Claim 1)

[2606] means for receiving input information from a user terminal;

[2607] a means for selecting an appropriate expert model based on the received information;

[2608] means for generating an appropriate answer based on the selected expert model;

[2609] means for transmitting the generated answer to a user terminal;

[2610] a means for using an emotion analysis engine to analyze emotions based on user input information;

[2611] A means to personalize responses based on sentiment analysis results,

[2612] A system including:

[2613] (Claim 2)

[2614] A means for users to mark individual expert models as favorites;

[2615] A means for instantly calling up a favorited expert model;

[2616] a means for storing the favorite information in a database and displaying the stored information on the user's device;

[2617] The system of claim 1 further comprising:

[2618] (Claim 3)

[2619] A way to search for other users' consultations anonymously,

[2620] A means for displaying search results in a list on the user's device;

[2621] A means for retrieving the searched anonymized user consultation content from the database;

[2622] The system of claim 1 further comprising:

[2623] "Application example 2 when combining emotion engines"

[2624] (Claim 1)

[2625] means for receiving input information from a user terminal;

[2626] a means for selecting an appropriate expert model based on the received information;

[2627] means for analyzing user sentiment using sentiment analysis means;

[2628] means for selecting an expert model based on the analyzed emotion information;

[2629] means for generating an appropriate answer based on the selected expert model;

[2630] means for transmitting the generated answer to a user terminal;

[2631] A system including:

[2632] (Claim 2)

[2633] A means for users to mark individual expert models as favorites;

[2634] A means for instantly calling up a favorited expert model;

[2635] The system of claim 1 further comprising:

[2636] (Claim 3)

[2637] A way to search for other users' consultations anonymously,

[2638] A means for displaying search results in a list on the user's device;

[2639] The system of claim 1 further comprising: [Explanation of symbols]

[2640] 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 input information from a user terminal; a means for selecting an appropriate expert model based on the received information; means for generating an appropriate answer based on the selected expert model; means for transmitting the generated answer to a user terminal; A system including:

2. A means for users to mark individual expert models as favorites; A means for instantly calling up a favorited expert model; The system of claim 1 further comprising:

3. A way to search for other users' consultations anonymously, A means for displaying search results in a list on the user's device; The system of claim 1 further comprising:

Images