System

Abstract

A system is provided.SOLUTION: A system, comprising: means for receiving a query input by a user; means for using an artificial intelligence model to analyze the query and generate an optimal answer; means for returning the generated answer to the user; means for receiving feedback from the user on the answer; and means for sending the feedback to the artificial intelligence model to improve the accuracy of the model.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] In today's information society, it is extremely important for users to quickly and accurately obtain the information they need. However, traditional bulletin boards and similar information sharing platforms have problems with the accuracy and reliability of the information they provide. Another issue is that when a large number of answers are posted, it can take a long time for users to find useful information. Furthermore, the quality of the answers is inconsistent, making it difficult to maintain user satisfaction. Therefore, there is a need for a system that can provide appropriate, high-quality answers to users' questions and quickly obtain the information they need. [Means for solving the problem]

[0005] The present invention provides a system that receives a user's question, analyzes it using an artificial intelligence model, and generates an appropriate answer. Specifically, the system includes a means for receiving a user's question, a means for using an artificial intelligence model to analyze the question and generate an optimal answer, and a means for returning the generated answer to the user. The system continuously improves the quality of the answer by including a means for receiving feedback on the answer from the user and a means for sending the feedback to the artificial intelligence model to improve the accuracy of the model. Furthermore, by further adding a means for storing the question content and user evaluation data in a database, the quality of the knowledge of the entire system can be improved by utilizing past questions, answers, and feedback. In this way, users can quickly and accurately obtain the information they need.

[0006] "User" refers to any individual or organization that uses the System to post inquiries and receive responses.

[0007] "Input inquiry" refers to the act of a user sending a question or request for information to the system in text format.

[0008] "Means for receiving" refers to the method or technology by which the system receives inquiries from users.

[0009] "Artificial intelligence model that analyzes and generates optimal answers" refers to an algorithm or program that implements machine learning and natural language processing techniques used to understand the content of a user's inquiry and generate an appropriate answer.

[0010] "Generated answer" refers to the answer that the artificial intelligence model analyzes and creates in response to the user's inquiry.

[0011] "Means for returning" refers to the method or technique for transmitting and displaying the generated answer to the user.

[0012] "Feedback" refers to the act of a user providing an evaluation or opinion regarding the quality or usefulness of a response they have received.

[0013] "Means for receiving feedback" refers to the methods and techniques by which the system receives feedback from the user.

[0014] "Means for sending to an AI model to improve the model's accuracy" refers to methods or techniques for providing received feedback to an AI model so that the model can learn from that information and improve the accuracy or quality of its answers.

[0015] "Database" refers to a structured data storage system for storing and managing inquiry details, user evaluation data, generated answers, etc. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0024] [First embodiment]

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

[0026] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

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

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

[0029] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

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

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

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

[0033] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

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

[0035] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

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

[0037] The system of the present invention has a process for receiving a query entered by a user, analyzing the query, generating an optimal answer, and returning the generated answer to the user. Furthermore, the system receives feedback from the user and transmits the feedback to an artificial intelligence model to improve the accuracy of the model. Specific embodiments for implementing this system are described below.

[0038] What the program does

[0039] A user posts a question

[0040] A user uses the bulletin board interface to enter their inquiry and presses the "Post" button, which sends the user's question to the server.

[0041] The server receives the query

[0042] The server receives user queries and stores them in a database for later analysis and answer generation.

[0043] The server sends the question to the generative AI model

[0044] The server sends the received question to the generative AI model, where the question is passed in text format and the model prepares it for analysis.

[0045] Generative AI model performs analysis and generates answers

[0046] The generative AI model analyzes the user's question and understands their intent. For example, if a user asks, "What is a list comprehension in Python?", the model generates an answer about the basic concept and usage of list comprehension.

[0047] Example answer: "List comprehensions are a concise way to create lists in Python. For example, \[x for x in range(10)\] creates a list containing the numbers 0 through 9."

[0048] The server sends the answer back to the user

[0049] The server receives the answer generated by the generative AI model and sends it back to the user, which then displays the appropriate answer on the user's screen.

[0050] Users rate the answers

[0051] Users provide feedback on the answers they receive, such as rating whether the answer was helpful or if there is room for improvement. This feedback information is sent to the server via rating buttons and comments.

[0052] The server feeds the evaluation back to the AI ​​model.

[0053] The server sends the feedback received from the user to the generative AI model, which then uses this feedback to improve itself and its ability to generate higher quality answers in the future.

[0054] Specific examples

[0055] For example, if a user posts a question such as "I don't really understand Python decorators," the system will act as follows:

[0056] 1. A user enters a question and posts it.

[0057] 2. The server receives the question and stores it in a database.

[0058] 3. The server sends the question to the generative AI model.

[0059] 4. A generative AI model analyzes the question and generates an answer, such as "A decorator is a mechanism for modifying functions to provide additional functionality."

[0060] 5. The server generates the answer and sends it back to the user.

[0061] 6. Users view the answer and rate whether it was helpful or not.

[0062] 7. The server feeds the evaluation back into the generative AI model, improving its accuracy.

[0063] In this way, the system of the present invention can provide fast and accurate answers to user questions, and can leverage user feedback to continually improve the quality of knowledge throughout the system.

[0064] The processing flow will be explained below.

[0065] Step 1:

[0066] A user opens the message board interface, enters the content of their inquiry, and presses the "Post" button. This action sends the question entered by the user from the terminal to the server.

[0067] Step 2:

[0068] The device converts the user's input into JSON format and sends it to the server as an HTTP request. The specific data format is as follows:

[0069] json

[0070] {

[0071] "question": "What is a list comprehension in Python?"

[0072] }

[0073] Step 3:

[0074] The server analyzes the request received from the terminal and extracts the inquiry content, while also validating the user's question to ensure it is in the correct format.

[0075] Step 4:

[0076] The server stores the user's question in a database, allowing you to later refer to the answers and feedback for this question.

[0077] python

[0078] db.save_question(question)

[0079] Step 5:

[0080] The server passes the saved question to the generative AI model, which then sends the question as an API request to the generative AI model.

[0081] python

[0082] ai_response = generate_answer(question)

[0083] Step 6:

[0084] The generative AI model analyzes the questions it receives and generates appropriate answers based on the questions, using natural language processing and machine learning algorithms in the analysis process.

[0085] Example: "List comprehensions are a concise way to create lists in Python. For example, \[x for x in range(10)\] creates a list containing the numbers 0 through 9."

[0086] Step 7:

[0087] The generative AI model sends the generated answer back to the server, which receives it and converts it back into JSON format to send back to the user.

[0088] Step 8:

[0089] The server generates an answer and sends it to the user, where it is returned as an HTTP response and displayed on the device.

[0090] json

[0091] {

[0092] "answer": "List comprehensions are a concise way to create lists in Python. For example, [x for x in range(10)] creates a list containing the numbers 0 through 9."

[0093] }

[0094] Step 9:

[0095] The user views the presented answer on the device and evaluates whether the answer meets their expectations.

[0096] Step 10:

[0097] Users can use the provided rating feature to provide feedback on answers, such as "helpful" or "more details needed."

[0098] Step 11:

[0099] The device converts the user's feedback into JSON format and sends it to the server as an HTTP request.

[0100] json

[0101] {

[0102] "question_id": "12345",

[0103] "feedback": "This answer was helpful"

[0104] }

[0105] Step 12:

[0106] The server receives feedback from the device and sends it to a database and generative AI model for storage and model training.

[0107] python

[0108] update_model_with_feedback(question_id, feedback)

[0109] Step 13:

[0110] The generative AI model adjusts its algorithm based on the feedback it receives, which improves the accuracy of responses to future inquiries.

[0111] Step 14:

[0112] The server stores user feedback information in a database for future reference, allowing past evaluation data to be used to improve the system as a whole.

[0113] python

[0114] db.save_feedback(question_id, feedback)

[0115] Example 1

[0116] 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."

[0117] Conventional inquiry response systems have had difficulty in providing prompt and appropriate answers to user inquiries. They also have been unable to effectively utilize user feedback to improve system performance. Furthermore, the processes required to manage inquiry content and generate appropriate answers are cumbersome, creating a need for efficient system operation.

[0118] 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.

[0119] In this invention, the server includes means for receiving queries entered by users, means for saving the query content in a database, means for analyzing the query content and using a generative AI model to generate an optimal answer, means for returning the generated answer to the user, means for receiving feedback on the answer from the user, and means for sending the feedback to the generative AI model to improve the accuracy of the model. This enables queries to be processed quickly and appropriately, and system performance can be improved by effectively utilizing user feedback.

[0120] A "query" is a question or request for information made by a user to the system.

[0121] A "database" is a system for efficiently managing and storing inquiry details, evaluation data, etc.

[0122] A "generative AI model" is a program or algorithm that uses artificial intelligence technology to analyze a user's question and generate the optimal answer.

[0123] "Feedback" refers to the evaluation or opinion of the answer provided by the user.

[0124] A "prompt" is a text-based instruction used to input a question to the generative AI model.

[0125] A "server" is a computer system that receives inquiries from users, performs the necessary processing, and works with the generative AI model to generate and return answers.

[0126] "User" means an individual or organization that uses the system to make an inquiry.

[0127] "Evaluation data" refers to information used to manage and store the content of feedback provided by users.

[0128] The "system" is an integrated mechanism that performs a series of steps: accepting inquiries, managing a database, generating answers using a generative AI model, returning answers to users, processing feedback, and improving the model.

[0129] The system of the present invention receives a query from a user, generates an optimal answer using a generative AI model, and returns it to the user. It also receives feedback from the user and sends it to the generative AI model to improve the accuracy of the model. A specific method for implementing the present invention is described below.

[0130] System Configuration

[0131] This system mainly consists of a server, a terminal, a generative AI model, and a database.

[0132] Hardware and software used

[0133] Server: A cloud server can be used to process user queries and generate answers in conjunction with the generative AI model.

[0134] Devices: A variety of devices are available, including PCs, smartphones, and tablets, and users use these to make inquiries.

[0135] Generative AI models: Use large language models (e.g., OpenAI GPT-3 or GPT-4) to analyze queries and generate optimal answers.

[0136] Database: Use a relational database such as PostgreSQL or MySQL to store inquiries and user feedback.

[0137] Operational Overview

[0138] 1. A user posts a query

[0139] A user accesses the bulletin board interface through a web browser, enters their inquiry, and presses the "Post" button. For example, a user enters the following question:

[0140] "Please explain list comprehensions in Python."

[0141] 2. The server receives the query

[0142] The server receives user queries as HTTP requests and stores them in a database for later analysis and answer generation.

[0143] 3. The server sends a query to the generative AI model

[0144] The server takes the query stored in the database and converts it into a prompt like this:

[0145] "A user has posted the following question: 'What is a list comprehension in Python?' Please generate a suitable answer."

[0146] This prompt is then sent to the API of the generative AI model.

[0147] 4. Generative AI models generate answers

[0148] The generative AI model analyzes the prompt and generates the best answer for the question. For example, it might generate an answer like this:

[0149] "List comprehensions are a concise way to create lists in Python. For example, [x for x in range(10)] creates a list containing the numbers 0 through 9."

[0150] This response is sent back to the server.

[0151] 5. The server generates the answer and sends it to the user

[0152] The server receives the generated answer, converts it into HTML format for display on the user's terminal, and sends it to the user. During this process, the answer to the question is displayed on the user's screen.

[0153] 6. Users rate answers

[0154] The user can rate the displayed answer and indicate whether it was helpful or whether it needs improvement. If the user rates it as "helpful," the rating is sent to the server.

[0155] 7. The server feeds the evaluation back to the generative AI model

[0156] The server feeds user evaluation data back to the generative AI model, which allows the model to improve itself and increase its ability to generate more accurate answers.

[0157] Specific examples

[0158] If a user posts a question such as "I don't really understand Python decorators," the process goes as follows:

[0159] 1. The user types, "I don't really understand Python decorators," and presses the submit button.

[0160] 2. The server receives this question and stores it in a database.

[0161] 3. The server parses the question and sends a prompt to the generative AI model, such as:

[0162] "A user has posted the following question: 'I don't really understand Python decorators.' Please generate a suitable answer for this."

[0163] 4. The generative AI model analyzes the prompt and generates an answer like this:

[0164] "Decorators are a mechanism for decorating functions to provide additional functionality."

[0165] 5. The server returns the generated answer to the user and displays it on the user's terminal.

[0166] 6. The user views the answer and rates it as helpful.

[0167] 7. The server feeds the evaluation back into the generative AI model, improving its accuracy.

[0168] This allows the system to provide fast and accurate answers to user queries, while also leveraging user feedback to continuously improve overall system performance.

[0169] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0170] Program processing flow

[0171] Step 1:

[0172] A user uses the bulletin board interface to enter their inquiry and presses the "Post" button. Specifically, the user enters the following text into the inquiry form displayed in their web browser:

[0173] "Please explain list comprehensions in Python."

[0174] Based on this input, the query content is sent to the server.

[0175] Step 2:

[0176] The server receives the query sent by the user as an HTTP request and executes a SQL query like the following to store the query in the database:

[0177] sql

[0178] INSERT INTO questions (content, user_id, created_at) VALUES ('What is a list comprehension in Python?', 123, NOW());

[0179] Input: User's inquiry and user ID

[0180] Output: Query data stored in a database

[0181] Step 3:

[0182] The server takes the query stored in the database and converts it into a prompt to be passed to the generative AI model. Specifically, it generates a text prompt like this:

[0183] "A user has posted the following question: 'What is a list comprehension in Python?' Please generate a suitable answer."

[0184] Input: Query content retrieved from the database

[0185] Output: Generated prompt statement

[0186] Step 4:

[0187] The server sends a prompt to the API of the generative AI model, using an HTTP request to send the prompt to the model.

[0188] Input: Generated prompt text

[0189] Output: The request data sent to the generative AI model

[0190] Step 5:

[0191] The generative AI model analyzes the received prompt and generates the best answer to the user's question. Specifically, it generates the following answer:

[0192] "List comprehensions are a concise way to create lists in Python. For example, [x for x in range(10)] creates a list containing the numbers 0 through 9."

[0193] Input: prompt statement

[0194] Output: Generated answer text

[0195] Step 6:

[0196] The server receives the answer returned by the generative AI model and converts it into HTML format for display to the user. Specifically, it generates HTML code like this:

[0197] html

[0198]

[0199] List comprehensions are a concise way to create lists in Python, for example [x for x in range(10)] will create a list containing the numbers 0 to 9.

[0200]

[0201] Input: Answer text returned by the generative AI model

[0202] Output: Generated HTML formatted response data

[0203] Step 7:

[0204] The server returns the generated HTML answer to the user's terminal, which displays the answer to the question on the user's screen.

[0205] Input: Generated HTML formatted answer data

[0206] Output: Answer displayed on the user's terminal

[0207] Step 8:

[0208] Users rate the displayed answers, providing feedback such as "helpful" or "needs improvement." Specifically, the following data is generated:

[0209] "Rating: Helpful, Question ID: 456"

[0210] Input: User rating

[0211] Output: Rating data sent to the server

[0212] Step 9:

[0213] The server receives the evaluation data sent by the user and feeds it back to the generative AI model. Specifically, it sends it to the feedback API in the following format:

[0214] "Feedback: 'Helpful', Question ID: 456"

[0215] Input: User rating data

[0216] Output: Feedback data sent to the generative AI model

[0217] This allows the entire system to respond quickly and accurately to user queries and use feedback to improve the accuracy of the model.

[0218] (Application example 1)

[0219] 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."

[0220] Traditionally, customer guidance in brick-and-mortar stores relied on sales staff, which created efficiency issues. Searching for products and checking inventory in the store also took time and effort, often resulting in lower customer satisfaction. Especially during busy times, sales staff were often unable to respond to each individual customer, making it difficult to provide appropriate guidance.

[0221] 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.

[0222] In this invention, the server includes means for receiving an inquiry entered by a user, means for analyzing the content of the inquiry and using an artificial intelligence model to generate an optimal answer, means for returning the generated answer to the user, means for receiving feedback on the answer from the user, means for sending the feedback to the artificial intelligence model to improve the accuracy of the model, means for converting the user's voice inquiry into text using speech recognition technology, means for displaying the generated answer on a display device, and means for the user to provide feedback via a head-mounted display. This makes it possible to efficiently and quickly provide appropriate answers to customer questions in physical stores, thereby improving customer satisfaction.

[0223] The "means for receiving user-input inquiries" is an interface for obtaining questions or requests from users in digital form.

[0224] "Means using an artificial intelligence model to analyze the content of the inquiry and generate an optimal response" refers to an artificial intelligence mechanism used to analyze the input inquiry and automatically generate an appropriate response.

[0225] The "means for returning the generated answer to the user" refers to a communication means or display means for conveying the answer generated by the artificial intelligence model to the user.

[0226] The "means for receiving feedback on answers from users" is a system component for collecting ratings and opinions of answers provided by users.

[0227] "Means for sending feedback to an AI model to improve the model's accuracy" refers to the process of inputting collected feedback information into an AI model to improve its performance and the quality of its answers.

[0228] "Means for converting a user's spoken query into text using speech recognition technology" is the process of utilizing speech recognition software to convert a user's verbal query into text form.

[0229] The "means for displaying the generated answer on a display device" is a mechanism for displaying the answer generated by the artificial intelligence model on a display device so that the user can visually confirm it.

[0230] The "means for the user to provide feedback through the head-mounted display" is an interface that allows the user to provide visual or audio feedback using the head-mounted display.

[0231] The system of the present invention is a process that receives a user-input query, analyzes the query, generates an optimal answer, and returns it to the user. It also receives feedback from the user and sends it to an artificial intelligence model to improve the model's accuracy. Specific embodiments for implementing this system are described below.

[0232] Hardware and software used

[0233] Head-mounted display (HMD)

[0234] microphone

[0235] server

[0236] Display device

[0237] Speech recognition software: Google Speech-to-Text

[0238] Generative AI model: OpenAI GPT-4

[0239] Programming language: Python

[0240] Data processing and calculation flow

[0241] Acquiring voice input

[0242] The user wears a head-mounted display (HMD) and issues a voice query through a microphone. The HMD's voice recognition function is used to capture the user's voice.

[0243] Converting audio data to text

[0244] The server uses voice recognition software (Google Speech-to-Text) to convert the voice data into text, which is then processed as the query.

[0245] Inquiry analysis and answer generation

[0246] The server sends the text of the query to a generative AI model (OpenAI GPT-4), which analyzes the query and generates the most appropriate answer.

[0247] Show Answers

[0248] The generated answers are sent to the server and displayed on the HMD display, allowing the user to see the answers in real time.

[0249] Gathering user feedback

[0250] The user provides feedback on the generated answers through the HMD, either by voice or touch. The feedback data is sent to the server and used to improve the generative AI model.

[0251] Specific examples

[0252] For example, if a user asks the question "Where is the tomato sauce?", the system will act as follows:

[0253] 1. The user uses the HMD and microphone to ask, "Where is the tomato sauce?"

[0254] 2. The server converts the speech into text and sends the text data, "Where is the tomato sauce?" to the generative AI model.

[0255] 3. The generative AI model generates the answer, "It's on the top shelf in the grocery section."

[0256] 4. The answer is displayed on the HMD display and the user confirms it.

[0257] 5. The user rates whether they were satisfied with the answer and provides feedback.

[0258] 6. Feedback is sent to the generative AI model through the server, improving the model's accuracy.

[0259] Prompt Sentence Examples

[0260] User asks: "Where is the tomato sauce?"

[0261] Answer: "Tomato sauce is on the top shelf in the grocery section."

[0262] In this way, the system of the present invention can efficiently and effectively guide customers in physical stores, and can improve the accuracy of the model based on user feedback, allowing for even higher quality service in the future.

[0263] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0264] Step 1:

[0265] A user wears a head-mounted display (HMD) and asks a question by voice. The user's voice input is captured through a microphone. The input is the user's voice data, and this voice data is the target for the next step. Specifically, the user asks, "Where is the tomato sauce?"

[0266] Step 2:

[0267] The user's voice data is sent to the server, and the server converts the voice data into text using speech recognition software (Google Speech-to-Text). The input is voice data, and the output is text data. Specifically, the text generated is "Where is the tomato sauce?"

[0268] Step 3:

[0269] The converted text data is sent to the server's generative AI model (OpenAI GPT-4). The server uses the generative AI model to analyze the text and generate the optimal answer. The input is text data, and the output is the answer text. Specifically, the model generates the answer, "Tomato sauce is on the top shelf in the grocery section."

[0270] Step 4:

[0271] The generated answer text is sent from the server to the HMD. The answer is displayed on the HMD display. The input is the answer text, and the output is the text displayed on the HMD display. Specifically, the answer "Tomato sauce is on the top shelf in the food section" is displayed on the HMD display.

[0272] Step 5:

[0273] The user checks the answer displayed through the HMD and provides feedback. The feedback is input by voice or touch. The input is the user's feedback data, which is used in the next step. Specifically, the user evaluates the answer as "helpful."

[0274] Step 6:

[0275] The server receives the user's feedback data and sends it to the generative AI model. The model uses the feedback to self-train and improve the accuracy of future answers. The input is the feedback data, and the output is updated model parameters. Specifically, the model learns to "generate better answers to similar questions in the future."

[0276] 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.

[0277] The system of the present invention has a process for receiving a query entered by a user, analyzing the query, generating an optimal answer, and returning the generated answer to the user. Furthermore, the system receives feedback from the user and sends the feedback to an artificial intelligence model to improve the accuracy of the model. Furthermore, by combining an emotion engine that recognizes the user's emotions, the system adjusts the tone of the answer to provide an answer that is more in line with the user's intention. Specific embodiments for implementing this system are described below.

[0278] What the program does

[0279] A user posts a question

[0280] A user uses the bulletin board interface to enter a question and presses the "Post" button, which sends the user's question to the server.

[0281] The server receives the query

[0282] The server receives the user's query and stores it in a database, allowing it to be referenced later.

[0283] The server sends the question to the emotion engine and generative AI model.

[0284] The server first sends the question received to the emotion engine to recognize the user's emotion. The emotion engine extracts emotions such as "joy," "sadness," and "anger" from the user's input. At the same time, the server sends the question to the generative AI model for analysis and answer generation.

[0285] Example: In response to the question "Tell me about list comprehensions in Python," the emotion engine recognizes "curiosity."

[0286] A generative AI model generates answers, reflecting the results of the emotion engine.

[0287] The generative AI model analyzes the user's question and generates an appropriate answer based on its content. It also adjusts the tone of the answer, taking into account the user's emotions as recognized by the emotion engine. For example, if the user expresses "sadness," the answer will have a more friendly and polite tone.

[0288] Example answer: "List comprehensions are a concise way to create lists in Python. For example, \[x for x in range(10)\] creates a list containing the numbers 0 through 9."

[0289] The server sends the answer and emotion data back to the user.

[0290] The server returns the answer generated by the generative AI model and the emotion data recognized by the emotion engine to the user, and through this process, the appropriate answer is displayed on the user's screen.

[0291] Users rate the answers

[0292] Users can provide feedback on the answers they see, using the rating buttons and comments section to rate whether the answer was helpful or if there are areas for improvement. Users can also provide emotional feedback.

[0293] The server receives the ratings and emotion data.

[0294] The server receives user feedback and emotion data and sends it to a database and generative AI model for storage and model training.

[0295] Reflecting feedback from rating and sentiment data

[0296] The generative AI model adjusts its algorithm based on the feedback it receives, and emotional data is also used as training data for the model to improve the accuracy of responses to future inquiries.

[0297] Specific examples

[0298] For example, if a user posts a question such as "I don't really understand Python decorators," the system will act as follows:

[0299] 1. A user enters a question and posts it.

[0300] 2. The server receives the question and stores it in a database.

[0301] 3. The server sends the question to the emotion engine, which recognizes the user's emotion as "confusion," and simultaneously sends the question to the generative AI model.

[0302] 4. The generative AI model analyzes the question and generates an answer, such as "Decorators are a mechanism for modifying functions to provide additional functionality," and adjusts the tone to be more helpful based on the results of the emotion engine.

[0303] 5. The server returns the generated answer to the user along with the emotion data.

[0304] 6. Users view the answers and rate their quality.

[0305] 7. The server receives the feedback and emotion data and feeds it back into the database and generative AI model.

[0306] In this way, the system of the present invention can not only provide fast and accurate answers to user questions, but also respond in a way that takes into account the user's emotions, and utilize user feedback to continuously improve the quality of the system's overall knowledge and user satisfaction.

[0307] The processing flow will be explained below.

[0308] Step 1:

[0309] A user opens the message board interface, enters the content of their inquiry, and presses the "Post" button. This action sends the user's question from their device to the server.

[0310] Step 2:

[0311] The device converts the user's input into JSON format and sends it to the server as an HTTP request. The specific data format is as follows:

[0312] json

[0313] {

[0314] "question": "What is a list comprehension in Python?"

[0315] }

[0316] Step 3:

[0317] The server analyzes the request received from the terminal and extracts the inquiry content, while also validating the user's question to ensure it is in the correct format.

[0318] Step 4:

[0319] The server stores the user's question in a database, allowing you to later refer to the answers and feedback for this question.

[0320] python

[0321] db.save_question(question)

[0322] Step 5:

[0323] The server sends the received question to the emotion engine, which analyzes the emotion from the user's input.

[0324] Example: In response to the question "Tell me about list comprehensions in Python," the emotion engine recognizes "curiosity."

[0325] Step 6:

[0326] The server sends the user's question to the generative AI model, along with the emotion data obtained by the emotion engine.

[0327] python

[0328] ai_response, user_emotion = generate_answer(question, emotion)

[0329] Step 7:

[0330] The generative AI model analyzes the user's question and generates an appropriate answer based on the question's content. This analysis process uses natural language processing and machine learning algorithms. When generating the answer, it takes into account emotional data obtained by the emotion engine and adjusts the tone of the answer.

[0331] Example answer: "List comprehensions are a concise way to create lists in Python. For example, \[x for x in range(10)\] creates a list containing the numbers 0 through 9."

[0332] Step 8:

[0333] The generative AI model sends the generated answer back to the server, which receives it and converts it back into JSON format to send back to the user.

[0334] json

[0335] {

[0336] "answer": ai_response,

[0337] "emotion": user_emotion

[0338] }

[0339] Step 9:

[0340] The server sends the generated answer and emotion data to the user, which is returned as an HTTP response and displayed on the device.

[0341] json

[0342] {

[0343] "answer": "List comprehensions are a concise way to create lists in Python. For example, [x for x in range(10)] creates a list containing the numbers 0 through 9.",

[0344] "emotion": "curiosity"

[0345] }

[0346] Step 10:

[0347] The user views the presented answer on the device and evaluates whether the answer meets their expectations.

[0348] Step 11:

[0349] Users can use the provided rating feature to provide feedback on answers, such as "helpful" or "more details needed."

[0350] Step 12:

[0351] The device converts the user's feedback into JSON format and sends it to the server as an HTTP request.

[0352] json

[0353] {

[0354] "question_id": "12345",

[0355] "feedback": "This answer was helpful",

[0356] "emotion": "satisfaction"

[0357] }

[0358] Step 13:

[0359] The server receives feedback from the device and sends it to a database and generative AI model for storage and model training.

[0360] python

[0361] update_model_with_feedback(question_id, feedback, emotion)

[0362] Step 14:

[0363] The generative AI model adjusts its algorithm based on the feedback it receives, which improves the accuracy of responses to future inquiries. Emotional data is also used as training data for the model.

[0364] python

[0365] model.learn_from_feedback(feedback, emotion)

[0366] Step 15:

[0367] The server stores the user's feedback information and emotion data in a database for future reference, allowing the system to utilize past evaluation data and emotion data to improve the system as a whole.

[0368] python

[0369] db.save_feedback(question_id, feedback, emotion)

[0370] Example 2

[0371] 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."

[0372] Conventional conversational AI systems generate responses uniformly without considering the user's emotions, making it difficult to communicate optimally according to the user's intentions and the situation. Furthermore, while mechanisms exist for receiving feedback, it is difficult to effectively utilize that feedback to improve the model. Furthermore, the lack of a function to adjust the tone of the generated responses often leads to a poor user experience.

[0373] 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.

[0374] In this invention, the server includes means for receiving a query entered by a user, means for analyzing the content of the query and using an AI model to generate an optimal answer, means for returning the generated answer to the user, means for receiving feedback on the answer from the user, means for sending the feedback to the AI ​​model to improve the accuracy of the model, and means for adjusting the tone of the answer using an emotion engine that recognizes the user's emotions. This makes it possible to provide an appropriate answer that reflects the user's emotions and to continuously improve the accuracy of the model by effectively utilizing the feedback.

[0375] "User" refers to the entity that uses the system to make an inquiry.

[0376] An "inquiry" refers to a question or request that a user inputs to the system.

[0377] "Analysis" refers to the process of understanding the user's inquiry and grasping its meaning.

[0378] An "artificial intelligence model" refers to a program that uses machine learning technology to generate optimal answers.

[0379] "Generated answer" refers to a response generated by an artificial intelligence model based on the analysis results.

[0380] "Feedback" refers to the ratings and opinions of answers provided by users.

[0381] "Accuracy" refers to the degree to which the AI ​​model's answers match the user's intentions and requirements.

[0382] An "emotion engine" refers to software that recognizes and analyzes emotions from user input.

[0383] "Tone" refers to the expressive and phrasing characteristics of the responses generated.

[0384] A "database" refers to a system that systematically stores information such as inquiries and feedback.

[0385] The system of the present invention has a process for receiving a user's query, analyzing it, generating an optimal answer, and returning the generated answer to the user. It also receives feedback from the user and sends the feedback to an artificial intelligence model to improve the accuracy of the model. It also combines an emotion engine that recognizes the user's emotions to adjust the tone of the answer and provide an answer that more closely matches the user's intentions.

[0386] System Embodiments

[0387] A user posts a question

[0388] The user uses the bulletin board interface on the terminal, enters the content of the inquiry, and presses the "Post" button. This operation sends the user's question to the server. The terminal can be an information device such as a personal computer or smartphone.

[0389] The server receives the query

[0390] The server receives queries from users and stores the details in a database. The database stores information such as the query details, user ID, and timestamp. This database can be a relational database (RDB) or a NoSQL database.

[0391] The server sends the question to the emotion engine and generative AI model.

[0392] The server first sends the received question to an emotion engine to recognize the user's emotions. The emotion engine uses software that uses natural language processing (NLP) technology (e.g., IBM Watson's emotion analysis API). At the same time, the server sends the question to a generative AI model, which analyzes it and generates an answer. The generative AI model used is, for example, GPT-3 (OpenAI).

[0393] A generative AI model generates answers, reflecting the results of the emotion engine.

[0394] The generative AI model analyzes the user's question and generates an appropriate answer based on its content. It also adjusts the tone of the answer, taking into account the user's emotions as recognized by the emotion engine. For example, if the user expresses "sadness," the answer will have a more friendly and polite tone.

[0395] The server sends the answer and emotion data back to the user.

[0396] The server returns the answer generated by the generative AI model and the emotion data recognized by the emotion engine to the user, and through this process, the appropriate answer is displayed on the user's screen.

[0397] Users rate the answers

[0398] Users can provide feedback on the displayed answers by rating them using the rating buttons and comments section, indicating whether the answer was helpful or if there are areas for improvement. Users can also provide emotional feedback.

[0399] The server receives the ratings and emotion data.

[0400] The server receives user feedback and emotion data and sends it to a database and generative AI model for storage and model training.

[0401] Reflecting feedback from rating and sentiment data

[0402] The generative AI model adjusts its algorithm based on the feedback it receives, and emotional data is also used as training data for the model to improve the accuracy of responses to future inquiries.

[0403] Specific examples

[0404] For example, if a user posts a question such as "I don't really understand Python decorators," the system will act as follows:

[0405] 1. A user enters a question and posts it.

[0406] 2. The server receives the question and stores it in a database.

[0407] 3. The server sends the question to the emotion engine, which recognizes the user's emotion as "confusion," and simultaneously sends the question to the generative AI model.

[0408] 4. The generative AI model analyzes the question and generates an answer, such as "A decorator is a mechanism for modifying functions to provide additional functionality," and adjusts the tone to be more helpful, reflecting the results of the emotion engine.

[0409] 5. The server returns the generated answer to the user along with the emotion data.

[0410] 6. Users view the answers and provide ratings and feedback.

[0411] 7. The server receives the feedback and emotion data and stores it in a database.

[0412] 8. The generative AI model learns from the feedback data and improves the accuracy of the answers next time.

[0413] Examples of prompt statements

[0414] Using the prompt "Tell me about Python decorators. What are they and how are they used?", the emotion engine recognizes the "confusion" and adjusts the generative AI model to return a clear and polite explanation.

[0415] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0416] Step 1: User posts a question

[0417] A user uses the bulletin board interface to enter a question and press the "Post" button. For example, they might enter "Please tell me about list comprehensions in Python." This action converts the entered question into JSON format on the terminal and sends it to the server. The input is the user's question text, and the output is the JSON-formatted data sent to the server.

[0418] Step 2: The server receives the query

[0419] The server waits for user queries on a receiving port and parses the received JSON data. Specifically, the server stores the query content in a database, which stores meta information such as the query content, user ID, and timestamp. The input is the query data sent by the user, and the output is a record stored in the database.

[0420] Step 3: The server sends the question to the emotion engine and generative AI model

[0421] The server first sends the received question to an emotion engine to recognize the user's emotions. The emotion engine uses NLP technology to extract emotions from the input text. For example, it uses IBM Watson's emotion analysis API to extract emotions such as "joy," "sadness," and "anger." At the same time, the server sends the question to a generative AI model (e.g., GPT-3), which analyzes the question and generates the optimal answer. The input is the question text from the user, and the output is the emotion data extracted by the emotion engine and the answer generated by the generative AI model.

[0422] Step 4: The generative AI model generates an answer, reflecting the results of the emotion engine.

[0423] The generative AI model analyzes the question and generates an appropriate answer based on its content. At this time, the tone is adjusted to reflect the results of the emotion engine. For example, if the emotion engine recognizes "confused," the explanation will have a more friendly and specific tone. For example, it generates the answer "List comprehension is a concise way to create lists in Python. For example, \[x for x in range(10)\] creates a list containing the numbers 0 to 9." The input is emotion data and the question text, and the output is the tone-adjusted answer text.

[0424] Step 5: The server sends the answer and emotion data back to the user

[0425] The server compiles the answer generated by the generative AI model and the emotion data recognized by the emotion engine in JSON format and returns it to the user. This is sent to the device as an HTTP response, and the appropriate answer is displayed on the user's screen. The input is the data obtained from the generative AI model and the emotion engine, and the output is displayed on the user's device.

[0426] Step 6: Users rate the answers

[0427] Users provide feedback on the displayed answers using rating buttons (e.g., "helpful" or "not helpful") or a comment field. Users may also provide emotional feedback. The input is the user's feedback, and the output is the rating data sent to the server.

[0428] Step 7: The server receives the ratings and sentiment data

[0429] The server receives feedback and emotion data from users and stores it in a database. Feedback information, such as answer ratings and areas for improvement, is stored in the "feedback" table. The input is the user's rating data, and the output is the feedback information stored in the database.

[0430] Step 8: Incorporate feedback from rating and sentiment data

[0431] The generative AI model adjusts its algorithm based on the feedback it receives. The server periodically runs a program that incorporates the feedback data into the model, and also uses the emotional data as training data. This improves the accuracy of responses to subsequent inquiries. The input is feedback and emotional data, and the output is an improved generative AI model.

[0432] (Application example 2)

[0433] 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."

[0434] Conventional content distribution services face the challenges of tedious processes for providing appropriate answers to user inquiries, and difficulty in adjusting the tone of the answer based on the user's emotions. Answers that do not take the user's emotions into consideration can ruin the user experience and reduce satisfaction. Furthermore, there is a lack of mechanisms for efficiently utilizing user feedback to improve the accuracy of AI models. Furthermore, the functionality for recommending content based on user emotional data is limited. To solve these challenges, a system incorporating tone adjustment using an emotion analysis engine and content recommendation functions is needed.

[0435] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving an inquiry entered by a user, means for analyzing the inquiry content and using a generative AI model to generate an optimal answer, means for returning the generated answer to the user, means for receiving feedback on the answer from the user, means for sending the feedback to the generative AI model to improve the accuracy of the model, means for using an emotion analysis engine that recognizes the user's emotions and adds an appropriate tone, and means for recommending content based on the user's emotion data. This makes it possible to adjust the tone according to the user's emotions and provide a more appropriate answer. Furthermore, it is possible to provide more accurate answers and a more satisfying user experience than conventional systems.

[0436] A "means for receiving user-input queries" refers to an interface or software component that receives questions or requests made by users to the system.

[0437] "Means of using a generative AI model that analyzes the content of the inquiry and generates the optimal response" refers to an artificial intelligence model that analyzes the user's inquiry using technologies such as natural language processing and generates an appropriate response.

[0438] "Means for returning the generated answer to the user" refers to an interface or communication means for presenting the answer created by the generative AI model to the user.

[0439] The "means for receiving feedback on answers from users" refers to an interface or mechanism for users to provide ratings and comments on the answers they receive.

[0440] "Means for sending feedback to the generative AI model to improve the model's accuracy" refers to a system for collecting feedback information from users and using it to train and tune the generative AI model.

[0441] "Means using an emotion analysis engine that recognizes a user's emotion and applies an appropriate tone" refers to a software engine that analyzes the emotion from a user's input and adjusts the tone of the response according to that emotion.

[0442] "Means for recommending content based on user emotional data" refers to algorithms or systems for recommending optimal content based on the user's emotions.

[0443] "Means for storing data in a database" refers to storage devices or software for storing collected inquiry details, feedback, emotional data, etc.

[0444] The system of the present invention is a system that provides appropriate answers to user inquiries in a content distribution service, recognizes the user's emotions and adjusts the tone of the answers, and recommends content taking the user's emotions into consideration. Specific embodiments for implementing this system are described below.

[0445] Program Overview

[0446] This system consists of a terminal that receives user inquiries, a server that analyzes and processes the inquiry content, and a mechanism for returning the results to the user.

[0447] Hardware and software used

[0448] The following hardware and software are utilized to implement the system.

[0449] Smartphone: Used as a user interface.

[0450] Server: Performs data processing and analysis. Specifically, it includes database servers and application servers.

[0451] Sentiment Analysis Engine: A software engine for analyzing emotions from user input (e.g., IBM Watson, Google Cloud Natural Language).

[0452] Generative AI model: An artificial intelligence model that analyzes questions and generates answers (e.g., OpenAI GPT-4).

[0453] Processing flow and data calculation

[0454] User enters and submits inquiry

[0455] Users use a dedicated smartphone app to enter their inquiries or requests in text format and press the send button, which then sends the inquiry from the device to the server.

[0456] The server receives and stores the query

[0457] The server stores the received queries in a database for future reference and analysis.

[0458] Sentiment analysis and answer generation

[0459] The server sends the received query to the sentiment analysis engine to analyze the user's sentiment. At the same time, the query is sent to the generative AI model to generate an appropriate answer. In this step, the results of the sentiment analysis are used to adjust the tone of the answer.

[0460] Returning answers and emotion data

[0461] The generated answers and the results of the sentiment analysis are sent back to the user from the server, who then views the received answers on their smartphone screen.

[0462] Collecting user feedback

[0463] The user can rate and comment on the answers they receive, and this feedback is sent back to the server, which stores it in a database and uses it as training data for the generative AI model.

[0464] Accuracy Improvement and Content Recommendation

[0465] The server applies the collected feedback to the generative AI model to improve its accuracy, and also applies an algorithm to recommend optimal content based on the user's emotional data.

[0466] Specific examples

[0467] If a user inputs and sends the message, "I'm feeling down, so please recommend some uplifting movies," the emotion analysis engine will analyze "sadness," and the generative AI model will present a list of uplifting movies. An example of an actual prompt is as follows:

[0468] Prompt Sentence Examples

[0469] If a user asks, "I'm feeling down, please recommend some uplifting movies," the generative AI model should take that emotion into consideration and suggest the best movie list. The suggested movie list should include movies with uplifting scenes and positive storylines.

[0470] In this way, the system of the present invention realizes appropriate answers and content recommendations that take into account the user's emotions.

[0471] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0472] Step 1:

[0473] The user uses a smartphone to input the details of their inquiry and presses the "Send" button. The input is sent as text data from the device to the server, which receives this text data.

[0474] Step 2:

[0475] The server stores the received text data in a database. This process stores the query for future reference and analysis. The input is text data, and the output is stored in the database.

[0476] Step 3:

[0477] The server sends the query text data to the emotion analysis engine. The emotion analysis engine analyzes the user's emotions from the text data and extracts emotion tags such as "sadness," "joy," and "anger." The input is the query text data, and the output is emotion tag data.

[0478] Step 4:

[0479] The server sends the same query text data to the generative AI model, which analyzes the text data and generates an optimal answer. At this time, it adjusts the tone of the answer based on the emotion tag obtained from the emotion analysis engine. The input is the query text data and emotion tag, and the output is the adjusted answer text.

[0480] Step 5:

[0481] The server returns the generated answer and emotion tag to the user, along with the answer text and emotion tag generated by the generative AI model. The input is the answer text and emotion tag, and the output is the answer displayed on the user's screen.

[0482] Step 6:

[0483] The user checks the returned answers and rates them. The user provides feedback through the rating buttons and comment fields. The input is the user's rating and comment text, and the output is the feedback data.

[0484] Step 7:

[0485] The server stores the feedback data received from the user in a database. The stored feedback data is used as training data for the generative AI model. The input is the feedback data, and the output is saved in the database.

[0486] Step 8:

[0487] The server applies the saved feedback data to the generative AI model to improve its accuracy. This process is performed to update the algorithm of the generative AI model and improve the accuracy of responses to subsequent inquiries. The input is the feedback data, and the output is the updated generative AI model.

[0488] Step 9:

[0489] Based on the user's emotional data, the server recommends the most suitable content to the user. The emotional data is analyzed, a list of content desired by the user is compiled, and this is provided to the user as a recommendation list. The input is emotional data, and the output is a list of recommended content.

[0490] By describing the specific actions at each step in detail, it becomes clear how the system responds to user inquiries and provides answers that take emotions into account. As part of this process, prompts are appropriately set and a generative AI model and sentiment analysis engine are effectively used to achieve high user satisfaction.

[0491] 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.

[0492] 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.

[0493] 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.

[0494] [Second embodiment]

[0495] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0496] 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.

[0497] 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).

[0498] 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.

[0499] 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.

[0500] 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).

[0501] 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.

[0502] 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.

[0503] 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.

[0504] 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.

[0505] In the smart glasses 214, 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.

[0506] 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."

[0507] The system of the present invention has a process for receiving a query entered by a user, analyzing the query, generating an optimal answer, and returning the generated answer to the user. Furthermore, the system receives feedback from the user and transmits the feedback to an artificial intelligence model to improve the accuracy of the model. Specific embodiments for implementing this system are described below.

[0508] What the program does

[0509] A user posts a question

[0510] A user uses the bulletin board interface to enter their inquiry and presses the "Post" button, which sends the user's question to the server.

[0511] The server receives the query

[0512] The server receives user queries and stores them in a database for later analysis and answer generation.

[0513] The server sends the question to the generative AI model

[0514] The server sends the received question to the generative AI model, where the question is passed in text format and the model prepares it for analysis.

[0515] Generative AI model performs analysis and generates answers

[0516] The generative AI model analyzes the user's question and understands their intent. For example, if a user asks, "What is a list comprehension in Python?", the model generates an answer about the basic concept and usage of list comprehension.

[0517] Example answer: "List comprehensions are a concise way to create lists in Python. For example, \[x for x in range(10)\] creates a list containing the numbers 0 through 9."

[0518] The server sends the answer back to the user

[0519] The server receives the answer generated by the generative AI model and sends it back to the user, which then displays the appropriate answer on the user's screen.

[0520] Users rate the answers

[0521] Users provide feedback on the answers they receive, such as rating whether the answer was helpful or if there is room for improvement. This feedback information is sent to the server via rating buttons and comments.

[0522] The server feeds the evaluation back to the AI ​​model.

[0523] The server sends the feedback received from the user to the generative AI model, which then uses this feedback to improve itself and its ability to generate higher quality answers in the future.

[0524] Specific examples

[0525] For example, if a user posts a question such as "I don't really understand Python decorators," the system will act as follows:

[0526] 1. A user enters a question and posts it.

[0527] 2. The server receives the question and stores it in a database.

[0528] 3. The server sends the question to the generative AI model.

[0529] 4. A generative AI model analyzes the question and generates an answer, such as "A decorator is a mechanism for modifying functions to provide additional functionality."

[0530] 5. The server generates the answer and sends it back to the user.

[0531] 6. Users view the answer and rate whether it was helpful or not.

[0532] 7. The server feeds the evaluation back into the generative AI model, improving its accuracy.

[0533] In this way, the system of the present invention can provide fast and accurate answers to user questions, and can leverage user feedback to continually improve the quality of knowledge throughout the system.

[0534] The processing flow will be explained below.

[0535] Step 1:

[0536] A user opens the message board interface, enters the content of their inquiry, and presses the "Post" button. This action sends the question entered by the user from the terminal to the server.

[0537] Step 2:

[0538] The device converts the user's input into JSON format and sends it to the server as an HTTP request. The specific data format is as follows:

[0539] json

[0540] {

[0541] "question": "What is a list comprehension in Python?"

[0542] }

[0543] Step 3:

[0544] The server analyzes the request received from the terminal and extracts the inquiry content, while also validating the user's question to ensure it is in the correct format.

[0545] Step 4:

[0546] The server stores the user's question in a database, allowing you to later refer to the answers and feedback for this question.

[0547] python

[0548] db.save_question(question)

[0549] Step 5:

[0550] The server passes the saved question to the generative AI model, which then sends the question as an API request to the generative AI model.

[0551] python

[0552] ai_response = generate_answer(question)

[0553] Step 6:

[0554] The generative AI model analyzes the questions it receives and generates appropriate answers based on the questions, using natural language processing and machine learning algorithms in the analysis process.

[0555] Example: "List comprehensions are a concise way to create lists in Python. For example, \[x for x in range(10)\] creates a list containing the numbers 0 through 9."

[0556] Step 7:

[0557] The generative AI model sends the generated answer back to the server, which receives it and converts it back into JSON format to send back to the user.

[0558] Step 8:

[0559] The server generates an answer and sends it to the user, where it is returned as an HTTP response and displayed on the device.

[0560] json

[0561] {

[0562] "answer": "List comprehensions are a concise way to create lists in Python. For example, [x for x in range(10)] creates a list containing the numbers 0 through 9."

[0563] }

[0564] Step 9:

[0565] The user views the presented answer on the device and evaluates whether the answer meets their expectations.

[0566] Step 10:

[0567] Users can use the provided rating feature to provide feedback on answers, such as "helpful" or "more details needed."

[0568] Step 11:

[0569] The device converts the user's feedback into JSON format and sends it to the server as an HTTP request.

[0570] json

[0571] {

[0572] "question_id": "12345",

[0573] "feedback": "This answer was helpful"

[0574] }

[0575] Step 12:

[0576] The server receives feedback from the device and sends it to a database and generative AI model for storage and model training.

[0577] python

[0578] update_model_with_feedback(question_id, feedback)

[0579] Step 13:

[0580] The generative AI model adjusts its algorithm based on the feedback it receives, which improves the accuracy of responses to future inquiries.

[0581] Step 14:

[0582] The server stores user feedback information in a database for future reference, allowing past evaluation data to be used to improve the system as a whole.

[0583] python

[0584] db.save_feedback(question_id, feedback)

[0585] Example 1

[0586] 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."

[0587] Conventional inquiry response systems have had difficulty in providing prompt and appropriate answers to user inquiries. They also have been unable to effectively utilize user feedback to improve system performance. Furthermore, the processes required to manage inquiry content and generate appropriate answers are cumbersome, creating a need for efficient system operation.

[0588] 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.

[0589] In this invention, the server includes means for receiving queries entered by users, means for saving the query content in a database, means for analyzing the query content and using a generative AI model to generate an optimal answer, means for returning the generated answer to the user, means for receiving feedback on the answer from the user, and means for sending the feedback to the generative AI model to improve the accuracy of the model. This enables queries to be processed quickly and appropriately, and system performance can be improved by effectively utilizing user feedback.

[0590] A "query" is a question or request for information made by a user to the system.

[0591] A "database" is a system for efficiently managing and storing inquiry details, evaluation data, etc.

[0592] A "generative AI model" is a program or algorithm that uses artificial intelligence technology to analyze a user's question and generate the optimal answer.

[0593] "Feedback" refers to the evaluation or opinion of the answer provided by the user.

[0594] A "prompt" is a text-based instruction used to input a question to the generative AI model.

[0595] A "server" is a computer system that receives inquiries from users, performs the necessary processing, and works with the generative AI model to generate and return answers.

[0596] "User" means an individual or organization that uses the system to make an inquiry.

[0597] "Evaluation data" refers to information used to manage and store the content of feedback provided by users.

[0598] The "system" is an integrated mechanism that performs a series of steps: accepting inquiries, managing a database, generating answers using a generative AI model, returning answers to users, processing feedback, and improving the model.

[0599] The system of the present invention receives a query from a user, generates an optimal answer using a generative AI model, and returns it to the user. It also receives feedback from the user and sends it to the generative AI model to improve the accuracy of the model. A specific method for implementing the present invention is described below.

[0600] System Configuration

[0601] This system mainly consists of a server, a terminal, a generative AI model, and a database.

[0602] Hardware and software used

[0603] Server: A cloud server can be used to process user queries and generate answers in conjunction with the generative AI model.

[0604] Devices: A variety of devices are available, including PCs, smartphones, and tablets, and users use these to make inquiries.

[0605] Generative AI models: Use large language models (e.g., OpenAI GPT-3 or GPT-4) to analyze queries and generate optimal answers.

[0606] Database: Use a relational database such as PostgreSQL or MySQL to store inquiries and user feedback.

[0607] Operational Overview

[0608] 1. A user posts a query

[0609] A user accesses the bulletin board interface through a web browser, enters their inquiry, and presses the "Post" button. For example, a user enters the following question:

[0610] "Please explain list comprehensions in Python."

[0611] 2. The server receives the query

[0612] The server receives user queries as HTTP requests and stores them in a database for later analysis and answer generation.

[0613] 3. The server sends a query to the generative AI model

[0614] The server takes the query stored in the database and converts it into a prompt like this:

[0615] "A user has posted the following question: 'What is a list comprehension in Python?' Please generate a suitable answer."

[0616] This prompt is then sent to the API of the generative AI model.

[0617] 4. Generative AI models generate answers

[0618] The generative AI model analyzes the prompt and generates the best answer for the question. For example, it might generate an answer like this:

[0619] "List comprehensions are a concise way to create lists in Python. For example, [x for x in range(10)] creates a list containing the numbers 0 through 9."

[0620] This response is sent back to the server.

[0621] 5. The server generates the answer and sends it to the user

[0622] The server receives the generated answer, converts it into HTML format for display on the user's terminal, and sends it to the user. During this process, the answer to the question is displayed on the user's screen.

[0623] 6. Users rate answers

[0624] The user can rate the displayed answer and indicate whether it was helpful or whether it needs improvement. If the user rates it as "helpful," the rating is sent to the server.

[0625] 7. The server feeds the evaluation back to the generative AI model

[0626] The server feeds user evaluation data back to the generative AI model, which allows the model to improve itself and increase its ability to generate more accurate answers.

[0627] Specific examples

[0628] If a user posts a question such as "I don't really understand Python decorators," the process goes as follows:

[0629] 1. The user types, "I don't really understand Python decorators," and presses the submit button.

[0630] 2. The server receives this question and stores it in a database.

[0631] 3. The server parses the question and sends a prompt to the generative AI model, such as:

[0632] "A user has posted the following question: 'I don't really understand Python decorators.' Please generate a suitable answer for this."

[0633] 4. The generative AI model analyzes the prompt and generates an answer like this:

[0634] "Decorators are a mechanism for decorating functions to provide additional functionality."

[0635] 5. The server returns the generated answer to the user and displays it on the user's terminal.

[0636] 6. The user views the answer and rates it as helpful.

[0637] 7. The server feeds the evaluation back into the generative AI model, improving its accuracy.

[0638] This allows the system to provide fast and accurate answers to user queries, while also leveraging user feedback to continuously improve overall system performance.

[0639] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0640] Program processing flow

[0641] Step 1:

[0642] A user uses the bulletin board interface to enter their inquiry and presses the "Post" button. Specifically, the user enters the following text into the inquiry form displayed in their web browser:

[0643] "Please explain list comprehensions in Python."

[0644] Based on this input, the query content is sent to the server.

[0645] Step 2:

[0646] The server receives the query sent by the user as an HTTP request and executes a SQL query like the following to store the query in the database:

[0647] sql

[0648] INSERT INTO questions (content, user_id, created_at) VALUES ('What is a list comprehension in Python?', 123, NOW());

[0649] Input: User's inquiry and user ID

[0650] Output: Query data stored in a database

[0651] Step 3:

[0652] The server takes the query stored in the database and converts it into a prompt to be passed to the generative AI model. Specifically, it generates a text prompt like this:

[0653] "A user has posted the following question: 'What is a list comprehension in Python?' Please generate a suitable answer."

[0654] Input: Query content retrieved from the database

[0655] Output: Generated prompt statement

[0656] Step 4:

[0657] The server sends a prompt to the API of the generative AI model, using an HTTP request to send the prompt to the model.

[0658] Input: Generated prompt text

[0659] Output: The request data sent to the generative AI model

[0660] Step 5:

[0661] The generative AI model analyzes the received prompt and generates the best answer to the user's question. Specifically, it generates the following answer:

[0662] "List comprehensions are a concise way to create lists in Python. For example, [x for x in range(10)] creates a list containing the numbers 0 through 9."

[0663] Input: prompt statement

[0664] Output: Generated answer text

[0665] Step 6:

[0666] The server receives the answer returned by the generative AI model and converts it into HTML format for display to the user. Specifically, it generates HTML code like this:

[0667] html

[0668]

[0669] List comprehensions are a concise way to create lists in Python, for example [x for x in range(10)] will create a list containing the numbers 0 to 9.

[0670]

[0671] Input: Answer text returned by the generative AI model

[0672] Output: Generated HTML formatted response data

[0673] Step 7:

[0674] The server returns the generated HTML answer to the user's terminal, which displays the answer to the question on the user's screen.

[0675] Input: Generated HTML formatted answer data

[0676] Output: Answer displayed on the user's terminal

[0677] Step 8:

[0678] Users rate the displayed answers, providing feedback such as "helpful" or "needs improvement." Specifically, the following data is generated:

[0679] "Rating: Helpful, Question ID: 456"

[0680] Input: User rating

[0681] Output: Rating data sent to the server

[0682] Step 9:

[0683] The server receives the evaluation data sent by the user and feeds it back to the generative AI model. Specifically, it sends it to the feedback API in the following format:

[0684] "Feedback: 'Helpful', Question ID: 456"

[0685] Input: User rating data

[0686] Output: Feedback data sent to the generative AI model

[0687] This allows the entire system to respond quickly and accurately to user queries and use feedback to improve the accuracy of the model.

[0688] (Application example 1)

[0689] 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."

[0690] Traditionally, customer guidance in brick-and-mortar stores relied on sales staff, which created efficiency issues. Searching for products and checking inventory in the store also took time and effort, often resulting in lower customer satisfaction. Especially during busy times, sales staff were often unable to respond to each individual customer, making it difficult to provide appropriate guidance.

[0691] 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.

[0692] In this invention, the server includes means for receiving an inquiry entered by a user, means for analyzing the content of the inquiry and using an artificial intelligence model to generate an optimal answer, means for returning the generated answer to the user, means for receiving feedback on the answer from the user, means for sending the feedback to the artificial intelligence model to improve the accuracy of the model, means for converting the user's voice inquiry into text using speech recognition technology, means for displaying the generated answer on a display device, and means for the user to provide feedback via a head-mounted display. This makes it possible to efficiently and quickly provide appropriate answers to customer questions in physical stores, thereby improving customer satisfaction.

[0693] The "means for receiving user-input inquiries" is an interface for obtaining questions or requests from users in digital form.

[0694] "Means using an artificial intelligence model to analyze the content of the inquiry and generate an optimal response" refers to an artificial intelligence mechanism used to analyze the input inquiry and automatically generate an appropriate response.

[0695] The "means for returning the generated answer to the user" refers to a communication means or display means for conveying the answer generated by the artificial intelligence model to the user.

[0696] The "means for receiving feedback on answers from users" is a system component for collecting ratings and opinions of answers provided by users.

[0697] "Means for sending feedback to an AI model to improve the model's accuracy" refers to the process of inputting collected feedback information into an AI model to improve its performance and the quality of its answers.

[0698] "Means for converting a user's spoken query into text using speech recognition technology" is the process of utilizing speech recognition software to convert a user's verbal query into text form.

[0699] The "means for displaying the generated answer on a display device" is a mechanism for displaying the answer generated by the artificial intelligence model on a display device so that the user can visually confirm it.

[0700] The "means for the user to provide feedback through the head-mounted display" is an interface that allows the user to provide visual or audio feedback using the head-mounted display.

[0701] The system of the present invention is a process that receives a user-input query, analyzes the query, generates an optimal answer, and returns it to the user. It also receives feedback from the user and sends it to an artificial intelligence model to improve the model's accuracy. Specific embodiments for implementing this system are described below.

[0702] Hardware and software used

[0703] Head-mounted display (HMD)

[0704] microphone

[0705] server

[0706] Display device

[0707] Speech recognition software: Google Speech-to-Text

[0708] Generative AI model: OpenAI GPT-4

[0709] Programming language: Python

[0710] Data processing and calculation flow

[0711] Acquiring voice input

[0712] The user wears a head-mounted display (HMD) and issues a voice query through a microphone. The HMD's voice recognition function is used to capture the user's voice.

[0713] Converting audio data to text

[0714] The server uses voice recognition software (Google Speech-to-Text) to convert the voice data into text, which is then processed as the query.

[0715] Inquiry analysis and answer generation

[0716] The server sends the text of the query to a generative AI model (OpenAI GPT-4), which analyzes the query and generates the most appropriate answer.

[0717] Show Answers

[0718] The generated answers are sent to the server and displayed on the HMD display, allowing the user to see the answers in real time.

[0719] Gathering user feedback

[0720] The user provides feedback on the generated answers through the HMD, either by voice or touch. The feedback data is sent to the server and used to improve the generative AI model.

[0721] Specific examples

[0722] For example, if a user asks the question "Where is the tomato sauce?", the system will act as follows:

[0723] 1. The user uses the HMD and microphone to ask, "Where is the tomato sauce?"

[0724] 2. The server converts the speech into text and sends the text data, "Where is the tomato sauce?" to the generative AI model.

[0725] 3. The generative AI model generates the answer, "It's on the top shelf in the grocery section."

[0726] 4. The answer is displayed on the HMD display and the user confirms it.

[0727] 5. The user rates whether they were satisfied with the answer and provides feedback.

[0728] 6. Feedback is sent to the generative AI model through the server, improving the model's accuracy.

[0729] Prompt Sentence Examples

[0730] User asks: "Where is the tomato sauce?"

[0731] Answer: "Tomato sauce is on the top shelf in the grocery section."

[0732] In this way, the system of the present invention can efficiently and effectively guide customers in physical stores, and can improve the accuracy of the model based on user feedback, allowing for even higher quality service in the future.

[0733] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0734] Step 1:

[0735] A user wears a head-mounted display (HMD) and asks a question by voice. The user's voice input is captured through a microphone. The input is the user's voice data, and this voice data is the target for the next step. Specifically, the user asks, "Where is the tomato sauce?"

[0736] Step 2:

[0737] The user's voice data is sent to the server, and the server converts the voice data into text using speech recognition software (Google Speech-to-Text). The input is voice data, and the output is text data. Specifically, the text generated is "Where is the tomato sauce?"

[0738] Step 3:

[0739] The converted text data is sent to the server's generative AI model (OpenAI GPT-4). The server uses the generative AI model to analyze the text and generate the optimal answer. The input is text data, and the output is the answer text. Specifically, the model generates the answer, "Tomato sauce is on the top shelf in the grocery section."

[0740] Step 4:

[0741] The generated answer text is sent from the server to the HMD. The answer is displayed on the HMD display. The input is the answer text, and the output is the text displayed on the HMD display. Specifically, the answer "Tomato sauce is on the top shelf in the food section" is displayed on the HMD display.

[0742] Step 5:

[0743] The user checks the answer displayed through the HMD and provides feedback. The feedback is input by voice or touch. The input is the user's feedback data, which is used in the next step. Specifically, the user evaluates the answer as "helpful."

[0744] Step 6:

[0745] The server receives the user's feedback data and sends it to the generative AI model. The model uses the feedback to self-train and improve the accuracy of future answers. The input is the feedback data, and the output is updated model parameters. Specifically, the model learns to "generate better answers to similar questions in the future."

[0746] 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.

[0747] The system of the present invention has a process for receiving a query entered by a user, analyzing the query, generating an optimal answer, and returning the generated answer to the user. Furthermore, the system receives feedback from the user and sends the feedback to an artificial intelligence model to improve the accuracy of the model. Furthermore, by combining an emotion engine that recognizes the user's emotions, the system adjusts the tone of the answer to provide an answer that is more in line with the user's intention. Specific embodiments for implementing this system are described below.

[0748] What the program does

[0749] A user posts a question

[0750] A user uses the bulletin board interface to enter a question and presses the "Post" button, which sends the user's question to the server.

[0751] The server receives the query

[0752] The server receives the user's query and stores it in a database, allowing it to be referenced later.

[0753] The server sends the question to the emotion engine and generative AI model.

[0754] The server first sends the question received to the emotion engine to recognize the user's emotion. The emotion engine extracts emotions such as "joy," "sadness," and "anger" from the user's input. At the same time, the server sends the question to the generative AI model for analysis and answer generation.

[0755] Example: In response to the question "Tell me about list comprehensions in Python," the emotion engine recognizes "curiosity."

[0756] A generative AI model generates answers, reflecting the results of the emotion engine.

[0757] The generative AI model analyzes the user's question and generates an appropriate answer based on its content. It also adjusts the tone of the answer, taking into account the user's emotions as recognized by the emotion engine. For example, if the user expresses "sadness," the answer will have a more friendly and polite tone.

[0758] Example answer: "List comprehensions are a concise way to create lists in Python. For example, \[x for x in range(10)\] creates a list containing the numbers 0 through 9."

[0759] The server sends the answer and emotion data back to the user.

[0760] The server returns the answer generated by the generative AI model and the emotion data recognized by the emotion engine to the user, and through this process, the appropriate answer is displayed on the user's screen.

[0761] Users rate the answers

[0762] Users can provide feedback on the answers they see, using the rating buttons and comments section to rate whether the answer was helpful or if there are areas for improvement. Users can also provide emotional feedback.

[0763] The server receives the ratings and emotion data.

[0764] The server receives user feedback and emotion data and sends it to a database and generative AI model for storage and model training.

[0765] Reflecting feedback from rating and sentiment data

[0766] The generative AI model adjusts its algorithm based on the feedback it receives, and emotional data is also used as training data for the model to improve the accuracy of responses to future inquiries.

[0767] Specific examples

[0768] For example, if a user posts a question such as "I don't really understand Python decorators," the system will act as follows:

[0769] 1. A user enters a question and posts it.

[0770] 2. The server receives the question and stores it in a database.

[0771] 3. The server sends the question to the emotion engine, which recognizes the user's emotion as "confusion," and simultaneously sends the question to the generative AI model.

[0772] 4. The generative AI model analyzes the question and generates an answer, such as "Decorators are a mechanism for modifying functions to provide additional functionality," and adjusts the tone to be more helpful based on the results of the emotion engine.

[0773] 5. The server returns the generated answer to the user along with the emotion data.

[0774] 6. Users view the answers and rate their quality.

[0775] 7. The server receives the feedback and emotion data and feeds it back into the database and generative AI model.

[0776] In this way, the system of the present invention can not only provide fast and accurate answers to user questions, but also respond in a way that takes into account the user's emotions, and utilize user feedback to continuously improve the quality of the system's overall knowledge and user satisfaction.

[0777] The processing flow will be explained below.

[0778] Step 1:

[0779] A user opens the message board interface, enters the content of their inquiry, and presses the "Post" button. This action sends the user's question from their device to the server.

[0780] Step 2:

[0781] The device converts the user's input into JSON format and sends it to the server as an HTTP request. The specific data format is as follows:

[0782] json

[0783] {

[0784] "question": "What is a list comprehension in Python?"

[0785] }

[0786] Step 3:

[0787] The server analyzes the request received from the terminal and extracts the inquiry content, while also validating the user's question to ensure it is in the correct format.

[0788] Step 4:

[0789] The server stores the user's question in a database, allowing you to later refer to the answers and feedback for this question.

[0790] python

[0791] db.save_question(question)

[0792] Step 5:

[0793] The server sends the received question to the emotion engine, which analyzes the emotion from the user's input.

[0794] Example: In response to the question "Tell me about list comprehensions in Python," the emotion engine recognizes "curiosity."

[0795] Step 6:

[0796] The server sends the user's question to the generative AI model, along with the emotion data obtained by the emotion engine.

[0797] python

[0798] ai_response, user_emotion = generate_answer(question, emotion)

[0799] Step 7:

[0800] The generative AI model analyzes the user's question and generates an appropriate answer based on the question's content. This analysis process uses natural language processing and machine learning algorithms. When generating the answer, it takes into account emotional data obtained by the emotion engine and adjusts the tone of the answer.

[0801] Example answer: "List comprehensions are a concise way to create lists in Python. For example, \[x for x in range(10)\] creates a list containing the numbers 0 through 9."

[0802] Step 8:

[0803] The generative AI model sends the generated answer back to the server, which receives it and converts it back into JSON format to send back to the user.

[0804] json

[0805] {

[0806] "answer": ai_response,

[0807] "emotion": user_emotion

[0808] }

[0809] Step 9:

[0810] The server sends the generated answer and emotion data to the user, which is returned as an HTTP response and displayed on the device.

[0811] json

[0812] {

[0813] "answer": "List comprehensions are a concise way to create lists in Python. For example, [x for x in range(10)] creates a list containing the numbers 0 through 9.",

[0814] "emotion": "curiosity"

[0815] }

[0816] Step 10:

[0817] The user views the presented answer on the device and evaluates whether the answer meets their expectations.

[0818] Step 11:

[0819] Users can use the provided rating feature to provide feedback on answers, such as "helpful" or "more details needed."

[0820] Step 12:

[0821] The device converts the user's feedback into JSON format and sends it to the server as an HTTP request.

[0822] json

[0823] {

[0824] "question_id": "12345",

[0825] "feedback": "This answer was helpful",

[0826] "emotion": "satisfaction"

[0827] }

[0828] Step 13:

[0829] The server receives feedback from the device and sends it to a database and generative AI model for storage and model training.

[0830] python

[0831] update_model_with_feedback(question_id, feedback, emotion)

[0832] Step 14:

[0833] The generative AI model adjusts its algorithm based on the feedback it receives, which improves the accuracy of responses to future inquiries. Emotional data is also used as training data for the model.

[0834] python

[0835] model.learn_from_feedback(feedback, emotion)

[0836] Step 15:

[0837] The server stores the user's feedback information and emotion data in a database for future reference, allowing the system to utilize past evaluation data and emotion data to improve the system as a whole.

[0838] python

[0839] db.save_feedback(question_id, feedback, emotion)

[0840] Example 2

[0841] 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."

[0842] Conventional conversational AI systems generate responses uniformly without considering the user's emotions, making it difficult to communicate optimally according to the user's intentions and the situation. Furthermore, while mechanisms exist for receiving feedback, it is difficult to effectively utilize that feedback to improve the model. Furthermore, the lack of a function to adjust the tone of the generated responses often leads to a poor user experience.

[0843] 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.

[0844] In this invention, the server includes means for receiving a query entered by a user, means for analyzing the content of the query and using an AI model to generate an optimal answer, means for returning the generated answer to the user, means for receiving feedback on the answer from the user, means for sending the feedback to the AI ​​model to improve the accuracy of the model, and means for adjusting the tone of the answer using an emotion engine that recognizes the user's emotions. This makes it possible to provide an appropriate answer that reflects the user's emotions and to continuously improve the accuracy of the model by effectively utilizing the feedback.

[0845] "User" refers to the entity that uses the system to make an inquiry.

[0846] An "inquiry" refers to a question or request that a user inputs to the system.

[0847] "Analysis" refers to the process of understanding the user's inquiry and grasping its meaning.

[0848] An "artificial intelligence model" refers to a program that uses machine learning technology to generate optimal answers.

[0849] "Generated answer" refers to a response generated by an artificial intelligence model based on the analysis results.

[0850] "Feedback" refers to the ratings and opinions of answers provided by users.

[0851] "Accuracy" refers to the degree to which the AI ​​model's answers match the user's intentions and requirements.

[0852] An "emotion engine" refers to software that recognizes and analyzes emotions from user input.

[0853] "Tone" refers to the expressive and phrasing characteristics of the responses generated.

[0854] A "database" refers to a system that systematically stores information such as inquiries and feedback.

[0855] The system of the present invention has a process for receiving a user's query, analyzing it, generating an optimal answer, and returning the generated answer to the user. It also receives feedback from the user and sends the feedback to an artificial intelligence model to improve the accuracy of the model. It also combines an emotion engine that recognizes the user's emotions to adjust the tone of the answer and provide an answer that more closely matches the user's intentions.

[0856] System Embodiments

[0857] A user posts a question

[0858] The user uses the bulletin board interface on the terminal, enters the content of the inquiry, and presses the "Post" button. This operation sends the user's question to the server. The terminal can be an information device such as a personal computer or smartphone.

[0859] The server receives the query

[0860] The server receives queries from users and stores the details in a database. The database stores information such as the query details, user ID, and timestamp. This database can be a relational database (RDB) or a NoSQL database.

[0861] The server sends the question to the emotion engine and generative AI model.

[0862] The server first sends the received question to an emotion engine to recognize the user's emotions. The emotion engine uses software that uses natural language processing (NLP) technology (e.g., IBM Watson's emotion analysis API). At the same time, the server sends the question to a generative AI model, which analyzes it and generates an answer. The generative AI model used is, for example, GPT-3 (OpenAI).

[0863] A generative AI model generates answers, reflecting the results of the emotion engine.

[0864] The generative AI model analyzes the user's question and generates an appropriate answer based on its content. It also adjusts the tone of the answer, taking into account the user's emotions as recognized by the emotion engine. For example, if the user expresses "sadness," the answer will have a more friendly and polite tone.

[0865] The server sends the answer and emotion data back to the user.

[0866] The server returns the answer generated by the generative AI model and the emotion data recognized by the emotion engine to the user, and through this process, the appropriate answer is displayed on the user's screen.

[0867] Users rate the answers

[0868] Users can provide feedback on the displayed answers by rating them using the rating buttons and comments section, indicating whether the answer was helpful or if there are areas for improvement. Users can also provide emotional feedback.

[0869] The server receives the ratings and emotion data.

[0870] The server receives user feedback and emotion data and sends it to a database and generative AI model for storage and model training.

[0871] Reflecting feedback from rating and sentiment data

[0872] The generative AI model adjusts its algorithm based on the feedback it receives, and emotional data is also used as training data for the model to improve the accuracy of responses to future inquiries.

[0873] Specific examples

[0874] For example, if a user posts a question such as "I don't really understand Python decorators," the system will act as follows:

[0875] 1. A user enters a question and posts it.

[0876] 2. The server receives the question and stores it in a database.

[0877] 3. The server sends the question to the emotion engine, which recognizes the user's emotion as "confusion," and simultaneously sends the question to the generative AI model.

[0878] 4. The generative AI model analyzes the question and generates an answer, such as "A decorator is a mechanism for modifying functions to provide additional functionality," and adjusts the tone to be more helpful, reflecting the results of the emotion engine.

[0879] 5. The server returns the generated answer to the user along with the emotion data.

[0880] 6. Users view the answers and provide ratings and feedback.

[0881] 7. The server receives the feedback and emotion data and stores it in a database.

[0882] 8. The generative AI model learns from the feedback data and improves the accuracy of the answers next time.

[0883] Examples of prompt statements

[0884] Using the prompt "Tell me about Python decorators. What are they and how are they used?", the emotion engine recognizes the "confusion" and adjusts the generative AI model to return a clear and polite explanation.

[0885] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0886] Step 1: User posts a question

[0887] A user uses the bulletin board interface to enter a question and press the "Post" button. For example, they might enter "Please tell me about list comprehensions in Python." This action converts the entered question into JSON format on the terminal and sends it to the server. The input is the user's question text, and the output is the JSON-formatted data sent to the server.

[0888] Step 2: The server receives the query

[0889] The server waits for user queries on a receiving port and parses the received JSON data. Specifically, the server stores the query content in a database, which stores meta information such as the query content, user ID, and timestamp. The input is the query data sent by the user, and the output is a record stored in the database.

[0890] Step 3: The server sends the question to the emotion engine and generative AI model

[0891] The server first sends the received question to an emotion engine to recognize the user's emotions. The emotion engine uses NLP technology to extract emotions from the input text. For example, it uses IBM Watson's emotion analysis API to extract emotions such as "joy," "sadness," and "anger." At the same time, the server sends the question to a generative AI model (e.g., GPT-3), which analyzes the question and generates the optimal answer. The input is the question text from the user, and the output is the emotion data extracted by the emotion engine and the answer generated by the generative AI model.

[0892] Step 4: The generative AI model generates an answer, reflecting the results of the emotion engine.

[0893] The generative AI model analyzes the question and generates an appropriate answer based on its content. At this time, the tone is adjusted to reflect the results of the emotion engine. For example, if the emotion engine recognizes "confused," the explanation will have a more friendly and specific tone. For example, it generates the answer "List comprehension is a concise way to create lists in Python. For example, \[x for x in range(10)\] creates a list containing the numbers 0 to 9." The input is emotion data and the question text, and the output is the tone-adjusted answer text.

[0894] Step 5: The server sends the answer and emotion data back to the user

[0895] The server compiles the answer generated by the generative AI model and the emotion data recognized by the emotion engine in JSON format and returns it to the user. This is sent to the device as an HTTP response, and the appropriate answer is displayed on the user's screen. The input is the data obtained from the generative AI model and the emotion engine, and the output is displayed on the user's device.

[0896] Step 6: Users rate the answers

[0897] Users provide feedback on the displayed answers using rating buttons (e.g., "helpful" or "not helpful") or a comment field. Users may also provide emotional feedback. The input is the user's feedback, and the output is the rating data sent to the server.

[0898] Step 7: The server receives the ratings and sentiment data

[0899] The server receives feedback and emotion data from users and stores it in a database. Feedback information, such as answer ratings and areas for improvement, is stored in the "feedback" table. The input is the user's rating data, and the output is the feedback information stored in the database.

[0900] Step 8: Incorporate feedback from rating and sentiment data

[0901] The generative AI model adjusts its algorithm based on the feedback it receives. The server periodically runs a program that incorporates the feedback data into the model, and also uses the emotional data as training data. This improves the accuracy of responses to subsequent inquiries. The input is feedback and emotional data, and the output is an improved generative AI model.

[0902] (Application example 2)

[0903] 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."

[0904] Conventional content distribution services face the challenges of tedious processes for providing appropriate answers to user inquiries, and difficulty in adjusting the tone of the answer based on the user's emotions. Answers that do not take the user's emotions into consideration can ruin the user experience and reduce satisfaction. Furthermore, there is a lack of mechanisms for efficiently utilizing user feedback to improve the accuracy of AI models. Furthermore, the functionality for recommending content based on user emotional data is limited. To solve these challenges, a system incorporating tone adjustment using an emotion analysis engine and content recommendation functions is needed.

[0905] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving an inquiry entered by a user, means for analyzing the inquiry content and using a generative AI model to generate an optimal answer, means for returning the generated answer to the user, means for receiving feedback on the answer from the user, means for sending the feedback to the generative AI model to improve the accuracy of the model, means for using an emotion analysis engine that recognizes the user's emotions and adds an appropriate tone, and means for recommending content based on the user's emotion data. This makes it possible to adjust the tone according to the user's emotions and provide a more appropriate answer. Furthermore, it is possible to provide more accurate answers and a more satisfying user experience than conventional systems.

[0906] A "means for receiving user-input queries" refers to an interface or software component that receives questions or requests made by users to the system.

[0907] "Means of using a generative AI model that analyzes the content of the inquiry and generates the optimal response" refers to an artificial intelligence model that analyzes the user's inquiry using technologies such as natural language processing and generates an appropriate response.

[0908] "Means for returning the generated answer to the user" refers to an interface or communication means for presenting the answer created by the generative AI model to the user.

[0909] The "means for receiving feedback on answers from users" refers to an interface or mechanism for users to provide ratings and comments on the answers they receive.

[0910] "Means for sending feedback to the generative AI model to improve the model's accuracy" refers to a system for collecting feedback information from users and using it to train and tune the generative AI model.

[0911] "Means using an emotion analysis engine that recognizes a user's emotion and applies an appropriate tone" refers to a software engine that analyzes the emotion from a user's input and adjusts the tone of the response according to that emotion.

[0912] "Means for recommending content based on user emotional data" refers to algorithms or systems for recommending optimal content based on the user's emotions.

[0913] "Means for storing data in a database" refers to storage devices or software for storing collected inquiry details, feedback, emotional data, etc.

[0914] The system of the present invention is a system that provides appropriate answers to user inquiries in a content distribution service, recognizes the user's emotions and adjusts the tone of the answers, and recommends content taking the user's emotions into consideration. Specific embodiments for implementing this system are described below.

[0915] Program Overview

[0916] This system consists of a terminal that receives user inquiries, a server that analyzes and processes the inquiry content, and a mechanism for returning the results to the user.

[0917] Hardware and software used

[0918] The following hardware and software are utilized to implement the system.

[0919] Smartphone: Used as a user interface.

[0920] Server: Performs data processing and analysis. Specifically, it includes database servers and application servers.

[0921] Sentiment Analysis Engine: A software engine for analyzing emotions from user input (e.g., IBM Watson, Google Cloud Natural Language).

[0922] Generative AI model: An artificial intelligence model that analyzes questions and generates answers (e.g., OpenAI GPT-4).

[0923] Processing flow and data calculation

[0924] User enters and submits inquiry

[0925] Users use a dedicated smartphone app to enter their inquiries or requests in text format and press the send button, which then sends the inquiry from the device to the server.

[0926] The server receives and stores the query

[0927] The server stores the received queries in a database for future reference and analysis.

[0928] Sentiment analysis and answer generation

[0929] The server sends the received query to the sentiment analysis engine to analyze the user's sentiment. At the same time, the query is sent to the generative AI model to generate an appropriate answer. In this step, the results of the sentiment analysis are used to adjust the tone of the answer.

[0930] Returning answers and emotion data

[0931] The generated answers and the results of the sentiment analysis are sent back to the user from the server, who then views the received answers on their smartphone screen.

[0932] Collecting user feedback

[0933] The user can rate and comment on the answers they receive, and this feedback is sent back to the server, which stores it in a database and uses it as training data for the generative AI model.

[0934] Accuracy Improvement and Content Recommendation

[0935] The server applies the collected feedback to the generative AI model to improve its accuracy, and also applies an algorithm to recommend optimal content based on the user's emotional data.

[0936] Specific examples

[0937] If a user inputs and sends the message, "I'm feeling down, so please recommend some uplifting movies," the emotion analysis engine will analyze "sadness," and the generative AI model will present a list of uplifting movies. An example of an actual prompt is as follows:

[0938] Prompt Sentence Examples

[0939] If a user asks, "I'm feeling down, please recommend some uplifting movies," the generative AI model should take that emotion into consideration and suggest the best movie list. The suggested movie list should include movies with uplifting scenes and positive storylines.

[0940] In this way, the system of the present invention realizes appropriate answers and content recommendations that take into account the user's emotions.

[0941] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0942] Step 1:

[0943] The user uses a smartphone to input the details of their inquiry and presses the "Send" button. The input is sent as text data from the device to the server, which receives this text data.

[0944] Step 2:

[0945] The server stores the received text data in a database. This process stores the query for future reference and analysis. The input is text data, and the output is stored in the database.

[0946] Step 3:

[0947] The server sends the query text data to the emotion analysis engine. The emotion analysis engine analyzes the user's emotions from the text data and extracts emotion tags such as "sadness," "joy," and "anger." The input is the query text data, and the output is emotion tag data.

[0948] Step 4:

[0949] The server sends the same query text data to the generative AI model, which analyzes the text data and generates an optimal answer. At this time, it adjusts the tone of the answer based on the emotion tag obtained from the emotion analysis engine. The input is the query text data and emotion tag, and the output is the adjusted answer text.

[0950] Step 5:

[0951] The server returns the generated answer and emotion tag to the user, along with the answer text and emotion tag generated by the generative AI model. The input is the answer text and emotion tag, and the output is the answer displayed on the user's screen.

[0952] Step 6:

[0953] The user checks the returned answers and rates them. The user provides feedback through the rating buttons and comment fields. The input is the user's rating and comment text, and the output is the feedback data.

[0954] Step 7:

[0955] The server stores the feedback data received from the user in a database. The stored feedback data is used as training data for the generative AI model. The input is the feedback data, and the output is saved in the database.

[0956] Step 8:

[0957] The server applies the saved feedback data to the generative AI model to improve its accuracy. This process is performed to update the algorithm of the generative AI model and improve the accuracy of responses to subsequent inquiries. The input is the feedback data, and the output is the updated generative AI model.

[0958] Step 9:

[0959] Based on the user's emotional data, the server recommends the most suitable content to the user. The emotional data is analyzed, a list of content desired by the user is compiled, and this is provided to the user as a recommendation list. The input is emotional data, and the output is a list of recommended content.

[0960] By describing the specific operations at each step in detail, it becomes clear how the system responds to user inquiries and provides answers that take emotions into account. As part of this process, prompts are appropriately set and a generative AI model and sentiment analysis engine are effectively used to achieve high user satisfaction.

[0961] 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.

[0962] 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.

[0963] 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.

[0964] [Third embodiment]

[0965] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0966] 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.

[0967] 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).

[0968] 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.

[0969] 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.

[0970] 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).

[0971] 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.

[0972] 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.

[0973] 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.

[0974] 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.

[0975] 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.

[0976] 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."

[0977] The system of the present invention has a process for receiving a query entered by a user, analyzing the query, generating an optimal answer, and returning the generated answer to the user. Furthermore, the system receives feedback from the user and transmits the feedback to an artificial intelligence model to improve the accuracy of the model. Specific embodiments for implementing this system are described below.

[0978] What the program does

[0979] A user posts a question

[0980] A user uses the bulletin board interface to enter their inquiry and presses the "Post" button, which sends the user's question to the server.

[0981] The server receives the query

[0982] The server receives user queries and stores them in a database for later analysis and answer generation.

[0983] The server sends the question to the generative AI model

[0984] The server sends the received question to the generative AI model, where the question is passed in text format and the model prepares it for analysis.

[0985] Generative AI model performs analysis and generates answers

[0986] The generative AI model analyzes the user's question and understands their intent. For example, if a user asks, "What is a list comprehension in Python?", the model generates an answer about the basic concept and usage of list comprehension.

[0987] Example answer: "List comprehensions are a concise way to create lists in Python. For example, \[x for x in range(10)\] creates a list containing the numbers 0 through 9."

[0988] The server sends the answer back to the user

[0989] The server receives the answer generated by the generative AI model and sends it back to the user, which then displays the appropriate answer on the user's screen.

[0990] Users rate the answers

[0991] Users provide feedback on the answers they receive, such as rating whether the answer was helpful or if there is room for improvement. This feedback information is sent to the server via rating buttons and comments.

[0992] The server feeds the evaluation back to the AI ​​model.

[0993] The server sends the feedback received from the user to the generative AI model, which then uses this feedback to improve itself and its ability to generate higher quality answers in the future.

[0994] Specific examples

[0995] For example, if a user posts a question such as "I don't really understand Python decorators," the system will act as follows:

[0996] 1. A user enters a question and posts it.

[0997] 2. The server receives the question and stores it in a database.

[0998] 3. The server sends the question to the generative AI model.

[0999] 4. A generative AI model analyzes the question and generates an answer, such as "A decorator is a mechanism for modifying functions to provide additional functionality."

[1000] 5. The server generates the answer and sends it back to the user.

[1001] 6. Users view the answer and rate whether it was helpful or not.

[1002] 7. The server feeds the evaluation back into the generative AI model, improving its accuracy.

[1003] In this way, the system of the present invention can provide fast and accurate answers to user questions, and can leverage user feedback to continually improve the quality of knowledge throughout the system.

[1004] The processing flow will be explained below.

[1005] Step 1:

[1006] A user opens the message board interface, enters the content of their inquiry, and presses the "Post" button. This action sends the question entered by the user from the terminal to the server.

[1007] Step 2:

[1008] The device converts the user's input into JSON format and sends it to the server as an HTTP request. The specific data format is as follows:

[1009] json

[1010] {

[1011] "question": "What is a list comprehension in Python?"

[1012] }

[1013] Step 3:

[1014] The server analyzes the request received from the terminal and extracts the inquiry content, while also validating the user's question to ensure it is in the correct format.

[1015] Step 4:

[1016] The server stores the user's question in a database, allowing you to later refer to the answers and feedback for this question.

[1017] python

[1018] db.save_question(question)

[1019] Step 5:

[1020] The server passes the saved question to the generative AI model, which then sends the question as an API request to the generative AI model.

[1021] python

[1022] ai_response = generate_answer(question)

[1023] Step 6:

[1024] The generative AI model analyzes the questions it receives and generates appropriate answers based on the questions, using natural language processing and machine learning algorithms in the analysis process.

[1025] Example: "List comprehensions are a concise way to create lists in Python. For example, \[x for x in range(10)\] creates a list containing the numbers 0 through 9."

[1026] Step 7:

[1027] The generative AI model sends the generated answer back to the server, which receives it and converts it back into JSON format to send back to the user.

[1028] Step 8:

[1029] The server generates an answer and sends it to the user, where it is returned as an HTTP response and displayed on the device.

[1030] json

[1031] {

[1032] "answer": "List comprehensions are a concise way to create lists in Python. For example, [x for x in range(10)] creates a list containing the numbers 0 through 9."

[1033] }

[1034] Step 9:

[1035] The user views the presented answer on the device and evaluates whether the answer meets their expectations.

[1036] Step 10:

[1037] Users can use the provided rating feature to provide feedback on answers, such as "helpful" or "more details needed."

[1038] Step 11:

[1039] The device converts the user's feedback into JSON format and sends it to the server as an HTTP request.

[1040] json

[1041] {

[1042] "question_id": "12345",

[1043] "feedback": "This answer was helpful"

[1044] }

[1045] Step 12:

[1046] The server receives feedback from the device and sends it to a database and generative AI model for storage and model training.

[1047] python

[1048] update_model_with_feedback(question_id, feedback)

[1049] Step 13:

[1050] The generative AI model adjusts its algorithm based on the feedback it receives, which improves the accuracy of responses to future inquiries.

[1051] Step 14:

[1052] The server stores user feedback information in a database for future reference, allowing past evaluation data to be used to improve the system as a whole.

[1053] python

[1054] db.save_feedback(question_id, feedback)

[1055] Example 1

[1056] 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."

[1057] Conventional inquiry response systems have had difficulty in providing prompt and appropriate answers to user inquiries. They also have been unable to effectively utilize user feedback to improve system performance. Furthermore, the processes required to manage inquiry content and generate appropriate answers are cumbersome, creating a need for efficient system operation.

[1058] 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.

[1059] In this invention, the server includes means for receiving queries entered by users, means for saving the query content in a database, means for analyzing the query content and using a generative AI model to generate an optimal answer, means for returning the generated answer to the user, means for receiving feedback on the answer from the user, and means for sending the feedback to the generative AI model to improve the accuracy of the model. This enables queries to be processed quickly and appropriately, and system performance can be improved by effectively utilizing user feedback.

[1060] A "query" is a question or request for information made by a user to the system.

[1061] A "database" is a system for efficiently managing and storing inquiry details, evaluation data, etc.

[1062] A "generative AI model" is a program or algorithm that uses artificial intelligence technology to analyze a user's question and generate the optimal answer.

[1063] "Feedback" refers to the evaluation or opinion of the answer provided by the user.

[1064] A "prompt" is a text-based instruction used to input a question to the generative AI model.

[1065] A "server" is a computer system that receives inquiries from users, performs the necessary processing, and works with the generative AI model to generate and return answers.

[1066] "User" means an individual or organization that uses the system to make an inquiry.

[1067] "Evaluation data" refers to information used to manage and store the content of feedback provided by users.

[1068] The "system" is an integrated mechanism that performs a series of steps: accepting inquiries, managing a database, generating answers using a generative AI model, returning answers to users, processing feedback, and improving the model.

[1069] The system of the present invention receives a query from a user, generates an optimal answer using a generative AI model, and returns it to the user. It also receives feedback from the user and sends it to the generative AI model to improve the accuracy of the model. A specific method for implementing the present invention is described below.

[1070] System Configuration

[1071] This system mainly consists of a server, a terminal, a generative AI model, and a database.

[1072] Hardware and software used

[1073] Server: A cloud server can be used to process user queries and generate answers in conjunction with the generative AI model.

[1074] Devices: A variety of devices are available, including PCs, smartphones, and tablets, and users use these to make inquiries.

[1075] Generative AI models: Use large language models (e.g., OpenAI GPT-3 or GPT-4) to analyze queries and generate optimal answers.

[1076] Database: Use a relational database such as PostgreSQL or MySQL to store inquiries and user feedback.

[1077] Operational Overview

[1078] 1. A user posts a query

[1079] A user accesses the bulletin board interface through a web browser, enters their inquiry, and presses the "Post" button. For example, a user enters the following question:

[1080] "Please explain list comprehensions in Python."

[1081] 2. The server receives the query

[1082] The server receives user queries as HTTP requests and stores them in a database for later analysis and answer generation.

[1083] 3. The server sends a query to the generative AI model

[1084] The server takes the query stored in the database and converts it into a prompt like this:

[1085] "A user has posted the following question: 'What is a list comprehension in Python?' Please generate a suitable answer."

[1086] This prompt is then sent to the API of the generative AI model.

[1087] 4. Generative AI models generate answers

[1088] The generative AI model analyzes the prompt and generates the best answer for the question. For example, it might generate an answer like this:

[1089] "List comprehensions are a concise way to create lists in Python. For example, [x for x in range(10)] creates a list containing the numbers 0 through 9."

[1090] This response is sent back to the server.

[1091] 5. The server generates the answer and sends it to the user

[1092] The server receives the generated answer, converts it into HTML format for display on the user's terminal, and sends it to the user. During this process, the answer to the question is displayed on the user's screen.

[1093] 6. Users rate answers

[1094] The user can rate the displayed answer and indicate whether it was helpful or whether it needs improvement. If the user rates it as "helpful," the rating is sent to the server.

[1095] 7. The server feeds the evaluation back to the generative AI model

[1096] The server feeds user evaluation data back to the generative AI model, which allows the model to improve itself and increase its ability to generate more accurate answers.

[1097] Specific examples

[1098] If a user posts a question such as "I don't really understand Python decorators," the process goes as follows:

[1099] 1. The user types, "I don't really understand Python decorators," and presses the submit button.

[1100] 2. The server receives this question and stores it in a database.

[1101] 3. The server parses the question and sends a prompt to the generative AI model, such as:

[1102] "A user has posted the following question: 'I don't really understand Python decorators.' Please generate a suitable answer for this."

[1103] 4. The generative AI model analyzes the prompt and generates an answer like this:

[1104] "Decorators are a mechanism for decorating functions to provide additional functionality."

[1105] 5. The server returns the generated answer to the user and displays it on the user's terminal.

[1106] 6. The user views the answer and rates it as helpful.

[1107] 7. The server feeds the evaluation back into the generative AI model, improving its accuracy.

[1108] This allows the system to provide fast and accurate answers to user queries, while also leveraging user feedback to continuously improve overall system performance.

[1109] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1110] Program processing flow

[1111] Step 1:

[1112] A user uses the bulletin board interface to enter their inquiry and presses the "Post" button. Specifically, the user enters the following text into the inquiry form displayed in their web browser:

[1113] "Please explain list comprehensions in Python."

[1114] Based on this input, the query content is sent to the server.

[1115] Step 2:

[1116] The server receives the query sent by the user as an HTTP request and executes a SQL query like the following to store the query in the database:

[1117] sql

[1118] INSERT INTO questions (content, user_id, created_at) VALUES ('What is a list comprehension in Python?', 123, NOW());

[1119] Input: User's inquiry and user ID

[1120] Output: Query data stored in a database

[1121] Step 3:

[1122] The server takes the query stored in the database and converts it into a prompt to be passed to the generative AI model. Specifically, it generates a text prompt like this:

[1123] "A user has posted the following question: 'What is a list comprehension in Python?' Please generate a suitable answer."

[1124] Input: Query content retrieved from the database

[1125] Output: Generated prompt statement

[1126] Step 4:

[1127] The server sends a prompt to the API of the generative AI model, using an HTTP request to send the prompt to the model.

[1128] Input: Generated prompt text

[1129] Output: The request data sent to the generative AI model

[1130] Step 5:

[1131] The generative AI model analyzes the received prompt and generates the best answer to the user's question. Specifically, it generates the following answer:

[1132] "List comprehensions are a concise way to create lists in Python. For example, [x for x in range(10)] creates a list containing the numbers 0 through 9."

[1133] Input: prompt statement

[1134] Output: Generated answer text

[1135] Step 6:

[1136] The server receives the answer returned by the generative AI model and converts it into HTML format for display to the user. Specifically, it generates HTML code like this:

[1137] html

[1138]

[1139] List comprehensions are a concise way to create lists in Python, for example [x for x in range(10)] will create a list containing the numbers 0 to 9.

[1140]

[1141] Input: Answer text returned by the generative AI model

[1142] Output: Generated HTML formatted response data

[1143] Step 7:

[1144] The server returns the generated HTML answer to the user's terminal, which displays the answer to the question on the user's screen.

[1145] Input: Generated HTML formatted answer data

[1146] Output: Answer displayed on the user's terminal

[1147] Step 8:

[1148] Users rate the displayed answers, providing feedback such as "helpful" or "needs improvement." Specifically, the following data is generated:

[1149] "Rating: Helpful, Question ID: 456"

[1150] Input: User rating

[1151] Output: Rating data sent to the server

[1152] Step 9:

[1153] The server receives the evaluation data sent by the user and feeds it back to the generative AI model. Specifically, it sends it to the feedback API in the following format:

[1154] "Feedback: 'Helpful', Question ID: 456"

[1155] Input: User rating data

[1156] Output: Feedback data sent to the generative AI model

[1157] This allows the entire system to respond quickly and accurately to user queries and use feedback to improve the accuracy of the model.

[1158] (Application example 1)

[1159] 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."

[1160] Traditionally, customer guidance in brick-and-mortar stores relied on sales staff, which created efficiency issues. Searching for products and checking inventory in the store also took time and effort, often resulting in lower customer satisfaction. Especially during busy times, sales staff were often unable to respond to each individual customer, making it difficult to provide appropriate guidance.

[1161] 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.

[1162] In this invention, the server includes means for receiving an inquiry entered by a user, means for analyzing the content of the inquiry and using an artificial intelligence model to generate an optimal answer, means for returning the generated answer to the user, means for receiving feedback on the answer from the user, means for sending the feedback to the artificial intelligence model to improve the accuracy of the model, means for converting the user's voice inquiry into text using speech recognition technology, means for displaying the generated answer on a display device, and means for the user to provide feedback via a head-mounted display. This makes it possible to efficiently and quickly provide appropriate answers to customer questions in physical stores, thereby improving customer satisfaction.

[1163] The "means for receiving user-input inquiries" is an interface for obtaining questions or requests from users in digital form.

[1164] "Means using an artificial intelligence model to analyze the content of the inquiry and generate an optimal response" refers to an artificial intelligence mechanism used to analyze the input inquiry and automatically generate an appropriate response.

[1165] The "means for returning the generated answer to the user" refers to a communication means or display means for conveying the answer generated by the artificial intelligence model to the user.

[1166] The "means for receiving feedback on answers from users" is a system component for collecting ratings and opinions of answers provided by users.

[1167] "Means for sending feedback to an AI model to improve the model's accuracy" refers to the process of inputting collected feedback information into an AI model to improve its performance and the quality of its answers.

[1168] "Means for converting a user's spoken query into text using speech recognition technology" is the process of utilizing speech recognition software to convert a user's verbal query into text form.

[1169] The "means for displaying the generated answer on a display device" is a mechanism for displaying the answer generated by the artificial intelligence model on a display device so that the user can visually confirm it.

[1170] The "means for the user to provide feedback through the head-mounted display" is an interface that allows the user to provide visual or audio feedback using the head-mounted display.

[1171] The system of the present invention is a process that receives a user-input query, analyzes the query, generates an optimal answer, and returns it to the user. It also receives feedback from the user and sends it to an artificial intelligence model to improve the model's accuracy. Specific embodiments for implementing this system are described below.

[1172] Hardware and software used

[1173] Head-mounted display (HMD)

[1174] microphone

[1175] server

[1176] Display device

[1177] Speech recognition software: Google Speech-to-Text

[1178] Generative AI model: OpenAI GPT-4

[1179] Programming language: Python

[1180] Data processing and calculation flow

[1181] Acquiring voice input

[1182] The user wears a head-mounted display (HMD) and issues a voice query through a microphone. The HMD's voice recognition function is used to capture the user's voice.

[1183] Converting audio data to text

[1184] The server uses voice recognition software (Google Speech-to-Text) to convert the voice data into text, which is then processed as the query.

[1185] Inquiry analysis and answer generation

[1186] The server sends the text of the query to a generative AI model (OpenAI GPT-4), which analyzes the query and generates the most appropriate answer.

[1187] Show Answers

[1188] The generated answers are sent to the server and displayed on the HMD display, allowing the user to see the answers in real time.

[1189] Gathering user feedback

[1190] The user provides feedback on the generated answers through the HMD, either by voice or touch. The feedback data is sent to the server and used to improve the generative AI model.

[1191] Specific examples

[1192] For example, if a user asks the question "Where is the tomato sauce?", the system will act as follows:

[1193] 1. The user uses the HMD and microphone to ask, "Where is the tomato sauce?"

[1194] 2. The server converts the speech into text and sends the text data, "Where is the tomato sauce?" to the generative AI model.

[1195] 3. The generative AI model generates the answer, "It's on the top shelf in the grocery section."

[1196] 4. The answer is displayed on the HMD display and the user confirms it.

[1197] 5. The user rates whether they were satisfied with the answer and provides feedback.

[1198] 6. Feedback is sent to the generative AI model through the server, improving the model's accuracy.

[1199] Prompt Sentence Examples

[1200] User asks: "Where is the tomato sauce?"

[1201] Answer: "Tomato sauce is on the top shelf in the grocery section."

[1202] In this way, the system of the present invention can efficiently and effectively guide customers in physical stores, and can improve the accuracy of the model based on user feedback, allowing for even higher quality service in the future.

[1203] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1204] Step 1:

[1205] A user wears a head-mounted display (HMD) and asks a question by voice. The user's voice input is captured through a microphone. The input is the user's voice data, and this voice data is the target for the next step. Specifically, the user asks, "Where is the tomato sauce?"

[1206] Step 2:

[1207] The user's voice data is sent to the server, and the server converts the voice data into text using speech recognition software (Google Speech-to-Text). The input is voice data, and the output is text data. Specifically, the text generated is "Where is the tomato sauce?"

[1208] Step 3:

[1209] The converted text data is sent to the server's generative AI model (OpenAI GPT-4). The server uses the generative AI model to analyze the text and generate the optimal answer. The input is text data, and the output is the answer text. Specifically, the model generates the answer, "Tomato sauce is on the top shelf in the grocery section."

[1210] Step 4:

[1211] The generated answer text is sent from the server to the HMD. The answer is displayed on the HMD display. The input is the answer text, and the output is the text displayed on the HMD display. Specifically, the answer "Tomato sauce is on the top shelf in the food section" is displayed on the HMD display.

[1212] Step 5:

[1213] The user checks the answer displayed through the HMD and provides feedback. The feedback is input by voice or touch. The input is the user's feedback data, which is used in the next step. Specifically, the user evaluates the answer as "helpful."

[1214] Step 6:

[1215] The server receives the user's feedback data and sends it to the generative AI model. The model uses the feedback to self-train and improve the accuracy of future answers. The input is the feedback data, and the output is updated model parameters. Specifically, the model learns to "generate better answers to similar questions in the future."

[1216] 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.

[1217] The system of the present invention has a process for receiving a query entered by a user, analyzing the query, generating an optimal answer, and returning the generated answer to the user. Furthermore, the system receives feedback from the user and sends the feedback to an artificial intelligence model to improve the accuracy of the model. Furthermore, by combining an emotion engine that recognizes the user's emotions, the system adjusts the tone of the answer to provide an answer that is more in line with the user's intention. Specific embodiments for implementing this system are described below.

[1218] What the program does

[1219] A user posts a question

[1220] A user uses the bulletin board interface to enter a question and presses the "Post" button, which sends the user's question to the server.

[1221] The server receives the query

[1222] The server receives the user's query and stores it in a database, allowing it to be referenced later.

[1223] The server sends the question to the emotion engine and generative AI model.

[1224] The server first sends the question received to the emotion engine to recognize the user's emotion. The emotion engine extracts emotions such as "joy," "sadness," and "anger" from the user's input. At the same time, the server sends the question to the generative AI model for analysis and answer generation.

[1225] Example: In response to the question "Tell me about list comprehensions in Python," the emotion engine recognizes "curiosity."

[1226] A generative AI model generates answers, reflecting the results of the emotion engine.

[1227] The generative AI model analyzes the user's question and generates an appropriate answer based on its content. It also adjusts the tone of the answer, taking into account the user's emotions as recognized by the emotion engine. For example, if the user expresses "sadness," the answer will have a more friendly and polite tone.

[1228] Example answer: "List comprehensions are a concise way to create lists in Python. For example, \[x for x in range(10)\] creates a list containing the numbers 0 through 9."

[1229] The server sends the answer and emotion data back to the user.

[1230] The server returns the answer generated by the generative AI model and the emotion data recognized by the emotion engine to the user, and through this process, the appropriate answer is displayed on the user's screen.

[1231] Users rate the answers

[1232] Users can provide feedback on the answers they see, using the rating buttons and comments section to rate whether the answer was helpful or if there are areas for improvement. Users can also provide emotional feedback.

[1233] The server receives the ratings and emotion data.

[1234] The server receives user feedback and emotion data and sends it to a database and generative AI model for storage and model training.

[1235] Reflecting feedback from rating and sentiment data

[1236] The generative AI model adjusts its algorithm based on the feedback it receives, and emotional data is also used as training data for the model to improve the accuracy of responses to future inquiries.

[1237] Specific examples

[1238] For example, if a user posts a question such as "I don't really understand Python decorators," the system will act as follows:

[1239] 1. A user enters a question and posts it.

[1240] 2. The server receives the question and stores it in a database.

[1241] 3. The server sends the question to the emotion engine, which recognizes the user's emotion as "confusion," and simultaneously sends the question to the generative AI model.

[1242] 4. The generative AI model analyzes the question and generates an answer, such as "Decorators are a mechanism for modifying functions to provide additional functionality," and adjusts the tone to be more helpful based on the results of the emotion engine.

[1243] 5. The server returns the generated answer to the user along with the emotion data.

[1244] 6. Users view the answers and rate their quality.

[1245] 7. The server receives the feedback and emotion data and feeds it back into the database and generative AI model.

[1246] In this way, the system of the present invention can not only provide fast and accurate answers to user questions, but also respond in a way that takes into account the user's emotions, and utilize user feedback to continuously improve the quality of the system's overall knowledge and user satisfaction.

[1247] The processing flow will be explained below.

[1248] Step 1:

[1249] A user opens the message board interface, enters the content of their inquiry, and presses the "Post" button. This action sends the user's question from their device to the server.

[1250] Step 2:

[1251] The device converts the user's input into JSON format and sends it to the server as an HTTP request. The specific data format is as follows:

[1252] json

[1253] {

[1254] "question": "What is a list comprehension in Python?"

[1255] }

[1256] Step 3:

[1257] The server analyzes the request received from the terminal and extracts the inquiry content, while also validating the user's question to ensure it is in the correct format.

[1258] Step 4:

[1259] The server stores the user's question in a database, allowing you to later refer to the answers and feedback for this question.

[1260] python

[1261] db.save_question(question)

[1262] Step 5:

[1263] The server sends the received question to the emotion engine, which analyzes the emotion from the user's input.

[1264] Example: In response to the question "Tell me about list comprehensions in Python," the emotion engine recognizes "curiosity."

[1265] Step 6:

[1266] The server sends the user's question to the generative AI model, along with the emotion data obtained by the emotion engine.

[1267] python

[1268] ai_response, user_emotion = generate_answer(question, emotion)

[1269] Step 7:

[1270] The generative AI model analyzes the user's question and generates an appropriate answer based on the question's content. This analysis process uses natural language processing and machine learning algorithms. When generating the answer, it takes into account emotional data obtained by the emotion engine and adjusts the tone of the answer.

[1271] Example answer: "List comprehensions are a concise way to create lists in Python. For example, \[x for x in range(10)\] creates a list containing the numbers 0 through 9."

[1272] Step 8:

[1273] The generative AI model sends the generated answer back to the server, which receives it and converts it back into JSON format to send back to the user.

[1274] json

[1275] {

[1276] "answer": ai_response,

[1277] "emotion": user_emotion

[1278] }

[1279] Step 9:

[1280] The server sends the generated answer and emotion data to the user, which is returned as an HTTP response and displayed on the device.

[1281] json

[1282] {

[1283] "answer": "List comprehensions are a concise way to create lists in Python. For example, [x for x in range(10)] creates a list containing the numbers 0 through 9.",

[1284] "emotion": "curiosity"

[1285] }

[1286] Step 10:

[1287] The user views the presented answer on the device and evaluates whether the answer meets their expectations.

[1288] Step 11:

[1289] Users can use the provided rating feature to provide feedback on answers, such as "helpful" or "more details needed."

[1290] Step 12:

[1291] The device converts the user's feedback into JSON format and sends it to the server as an HTTP request.

[1292] json

[1293] {

[1294] "question_id": "12345",

[1295] "feedback": "This answer was helpful",

[1296] "emotion": "satisfaction"

[1297] }

[1298] Step 13:

[1299] The server receives feedback from the device and sends it to a database and generative AI model for storage and model training.

[1300] python

[1301] update_model_with_feedback(question_id, feedback, emotion)

[1302] Step 14:

[1303] The generative AI model adjusts its algorithm based on the feedback it receives, which improves the accuracy of responses to future inquiries. Emotional data is also used as training data for the model.

[1304] python

[1305] model.learn_from_feedback(feedback, emotion)

[1306] Step 15:

[1307] The server stores the user's feedback information and emotion data in a database for future reference, allowing the system to utilize past evaluation data and emotion data to improve the system as a whole.

[1308] python

[1309] db.save_feedback(question_id, feedback, emotion)

[1310] Example 2

[1311] 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."

[1312] Conventional conversational AI systems generate responses uniformly without considering the user's emotions, making it difficult to communicate optimally according to the user's intentions and the situation. Furthermore, while mechanisms exist for receiving feedback, it is difficult to effectively utilize that feedback to improve the model. Furthermore, the lack of a function to adjust the tone of the generated responses often leads to a poor user experience.

[1313] 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.

[1314] In this invention, the server includes means for receiving a query entered by a user, means for analyzing the content of the query and using an AI model to generate an optimal answer, means for returning the generated answer to the user, means for receiving feedback on the answer from the user, means for sending the feedback to the AI ​​model to improve the accuracy of the model, and means for adjusting the tone of the answer using an emotion engine that recognizes the user's emotions. This makes it possible to provide an appropriate answer that reflects the user's emotions and to continuously improve the accuracy of the model by effectively utilizing the feedback.

[1315] "User" refers to the entity that uses the system to make an inquiry.

[1316] An "inquiry" refers to a question or request that a user inputs to the system.

[1317] "Analysis" refers to the process of understanding the user's inquiry and grasping its meaning.

[1318] An "artificial intelligence model" refers to a program that uses machine learning technology to generate optimal answers.

[1319] "Generated answer" refers to a response generated by an artificial intelligence model based on the analysis results.

[1320] "Feedback" refers to the ratings and opinions of answers provided by users.

[1321] "Accuracy" refers to the degree to which the AI ​​model's answers match the user's intentions and requirements.

[1322] An "emotion engine" refers to software that recognizes and analyzes emotions from user input.

[1323] "Tone" refers to the expressive and phrasing characteristics of the responses generated.

[1324] A "database" refers to a system that systematically stores information such as inquiries and feedback.

[1325] The system of the present invention has a process for receiving a user's query, analyzing it, generating an optimal answer, and returning the generated answer to the user. It also receives feedback from the user and sends the feedback to an artificial intelligence model to improve the accuracy of the model. It also combines an emotion engine that recognizes the user's emotions to adjust the tone of the answer and provide an answer that more closely matches the user's intentions.

[1326] System Embodiments

[1327] A user posts a question

[1328] The user uses the bulletin board interface on the terminal, enters the content of the inquiry, and presses the "Post" button. This operation sends the user's question to the server. The terminal can be an information device such as a personal computer or smartphone.

[1329] The server receives the query

[1330] The server receives queries from users and stores the details in a database. The database stores information such as the query details, user ID, and timestamp. This database can be a relational database (RDB) or a NoSQL database.

[1331] The server sends the question to the emotion engine and generative AI model.

[1332] The server first sends the received question to an emotion engine to recognize the user's emotions. The emotion engine uses software that uses natural language processing (NLP) technology (e.g., IBM Watson's emotion analysis API). At the same time, the server sends the question to a generative AI model, which analyzes it and generates an answer. The generative AI model used is, for example, GPT-3 (OpenAI).

[1333] A generative AI model generates answers, reflecting the results of the emotion engine.

[1334] The generative AI model analyzes the user's question and generates an appropriate answer based on its content. It also adjusts the tone of the answer, taking into account the user's emotions as recognized by the emotion engine. For example, if the user expresses "sadness," the answer will have a more friendly and polite tone.

[1335] The server sends the answer and emotion data back to the user.

[1336] The server returns the answer generated by the generative AI model and the emotion data recognized by the emotion engine to the user, and through this process, the appropriate answer is displayed on the user's screen.

[1337] Users rate the answers

[1338] Users can provide feedback on the displayed answers by rating them using the rating buttons and comments section, indicating whether the answer was helpful or if there are areas for improvement. Users can also provide emotional feedback.

[1339] The server receives the ratings and emotion data.

[1340] The server receives user feedback and emotion data and sends it to a database and generative AI model for storage and model training.

[1341] Reflecting feedback from rating and sentiment data

[1342] The generative AI model adjusts its algorithm based on the feedback it receives, and emotional data is also used as training data for the model to improve the accuracy of responses to future inquiries.

[1343] Specific examples

[1344] For example, if a user posts a question such as "I don't really understand Python decorators," the system will act as follows:

[1345] 1. A user enters a question and posts it.

[1346] 2. The server receives the question and stores it in a database.

[1347] 3. The server sends the question to the emotion engine, which recognizes the user's emotion as "confusion," and simultaneously sends the question to the generative AI model.

[1348] 4. The generative AI model analyzes the question and generates an answer, such as "A decorator is a mechanism for modifying functions to provide additional functionality," and adjusts the tone to be more helpful, reflecting the results of the emotion engine.

[1349] 5. The server returns the generated answer to the user along with the emotion data.

[1350] 6. Users view the answers and provide ratings and feedback.

[1351] 7. The server receives the feedback and emotion data and stores it in a database.

[1352] 8. The generative AI model learns from the feedback data and improves the accuracy of the answers next time.

[1353] Examples of prompt statements

[1354] Using the prompt "Tell me about Python decorators. What are they and how are they used?", the emotion engine recognizes the "confusion" and adjusts the generative AI model to return a clear and polite explanation.

[1355] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1356] Step 1: User posts a question

[1357] A user uses the bulletin board interface to enter a question and press the "Post" button. For example, they might enter "Please tell me about list comprehensions in Python." This action converts the entered question into JSON format on the terminal and sends it to the server. The input is the user's question text, and the output is the JSON-formatted data sent to the server.

[1358] Step 2: The server receives the query

[1359] The server waits for user queries on a receiving port and parses the received JSON data. Specifically, the server stores the query content in a database, which stores meta information such as the query content, user ID, and timestamp. The input is the query data sent by the user, and the output is a record stored in the database.

[1360] Step 3: The server sends the question to the emotion engine and generative AI model

[1361] The server first sends the received question to an emotion engine to recognize the user's emotions. The emotion engine uses NLP technology to extract emotions from the input text. For example, it uses IBM Watson's emotion analysis API to extract emotions such as "joy," "sadness," and "anger." At the same time, the server sends the question to a generative AI model (e.g., GPT-3), which analyzes the question and generates the optimal answer. The input is the question text from the user, and the output is the emotion data extracted by the emotion engine and the answer generated by the generative AI model.

[1362] Step 4: The generative AI model generates an answer, reflecting the results of the emotion engine.

[1363] The generative AI model analyzes the question and generates an appropriate answer based on its content. At this time, the tone is adjusted to reflect the results of the emotion engine. For example, if the emotion engine recognizes "confused," the explanation will have a more friendly and specific tone. For example, it generates the answer "List comprehension is a concise way to create lists in Python. For example, \[x for x in range(10)\] creates a list containing the numbers 0 to 9." The input is emotion data and the question text, and the output is the tone-adjusted answer text.

[1364] Step 5: The server sends the answer and emotion data back to the user

[1365] The server compiles the answer generated by the generative AI model and the emotion data recognized by the emotion engine in JSON format and returns it to the user. This is sent to the device as an HTTP response, and the appropriate answer is displayed on the user's screen. The input is the data obtained from the generative AI model and the emotion engine, and the output is displayed on the user's device.

[1366] Step 6: Users rate the answers

[1367] Users provide feedback on the displayed answers using rating buttons (e.g., "helpful" or "not helpful") or a comment field. Users may also provide emotional feedback. The input is the user's feedback, and the output is the rating data sent to the server.

[1368] Step 7: The server receives the ratings and sentiment data

[1369] The server receives feedback and emotion data from users and stores it in a database. Feedback information, such as answer ratings and areas for improvement, is stored in the "feedback" table. The input is the user's rating data, and the output is the feedback information stored in the database.

[1370] Step 8: Incorporate feedback from rating and sentiment data

[1371] The generative AI model adjusts its algorithm based on the feedback it receives. The server periodically runs a program that incorporates the feedback data into the model, and also uses the emotional data as training data. This improves the accuracy of responses to subsequent inquiries. The input is feedback and emotional data, and the output is an improved generative AI model.

[1372] (Application example 2)

[1373] 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."

[1374] Conventional content distribution services face the challenges of tedious processes for providing appropriate answers to user inquiries, and difficulty in adjusting the tone of the answer based on the user's emotions. Answers that do not take the user's emotions into consideration can ruin the user experience and reduce satisfaction. Furthermore, there is a lack of mechanisms for efficiently utilizing user feedback to improve the accuracy of AI models. Furthermore, the functionality for recommending content based on user emotional data is limited. To solve these challenges, a system incorporating tone adjustment using an emotion analysis engine and content recommendation functions is needed.

[1375] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving an inquiry entered by a user, means for analyzing the inquiry content and using a generative AI model to generate an optimal answer, means for returning the generated answer to the user, means for receiving feedback on the answer from the user, means for sending the feedback to the generative AI model to improve the accuracy of the model, means for using an emotion analysis engine that recognizes the user's emotions and adds an appropriate tone, and means for recommending content based on the user's emotion data. This makes it possible to adjust the tone according to the user's emotions and provide a more appropriate answer. Furthermore, it is possible to provide more accurate answers and a more satisfying user experience than conventional systems.

[1376] A "means for receiving user-input queries" refers to an interface or software component that receives questions or requests made by users to the system.

[1377] "Means of using a generative AI model that analyzes the content of the inquiry and generates the optimal response" refers to an artificial intelligence model that analyzes the user's inquiry using technologies such as natural language processing and generates an appropriate response.

[1378] "Means for returning the generated answer to the user" refers to an interface or communication means for presenting the answer created by the generative AI model to the user.

[1379] The "means for receiving feedback on answers from users" refers to an interface or mechanism for users to provide ratings and comments on the answers they receive.

[1380] "Means for sending feedback to the generative AI model to improve the model's accuracy" refers to a system for collecting feedback information from users and using it to train and tune the generative AI model.

[1381] "Means using an emotion analysis engine that recognizes a user's emotion and applies an appropriate tone" refers to a software engine that analyzes the emotion from a user's input and adjusts the tone of the response according to that emotion.

[1382] "Means for recommending content based on user emotional data" refers to algorithms or systems for recommending optimal content based on the user's emotions.

[1383] "Means for storing data in a database" refers to storage devices or software for storing collected inquiry details, feedback, emotional data, etc.

[1384] The system of the present invention is a system that provides appropriate answers to user inquiries in a content distribution service, recognizes the user's emotions and adjusts the tone of the answers, and recommends content taking the user's emotions into consideration. Specific embodiments for implementing this system are described below.

[1385] Program Overview

[1386] This system consists of a terminal that receives user inquiries, a server that analyzes and processes the inquiry content, and a mechanism for returning the results to the user.

[1387] Hardware and software used

[1388] The following hardware and software are utilized to implement the system.

[1389] Smartphone: Used as a user interface.

[1390] Server: Performs data processing and analysis. Specifically, it includes database servers and application servers.

[1391] Sentiment Analysis Engine: A software engine for analyzing emotions from user input (e.g., IBM Watson, Google Cloud Natural Language).

[1392] Generative AI model: An artificial intelligence model that analyzes questions and generates answers (e.g., OpenAI GPT-4).

[1393] Processing flow and data calculation

[1394] User enters and submits inquiry

[1395] Users use a dedicated smartphone app to enter their inquiries or requests in text format and press the send button, which then sends the inquiry from the device to the server.

[1396] The server receives and stores the query

[1397] The server stores the received queries in a database for future reference and analysis.

[1398] Sentiment analysis and answer generation

[1399] The server sends the received query to the sentiment analysis engine to analyze the user's sentiment. At the same time, the query is sent to the generative AI model to generate an appropriate answer. In this step, the results of the sentiment analysis are used to adjust the tone of the answer.

[1400] Returning answers and emotion data

[1401] The generated answers and the results of the sentiment analysis are sent back to the user from the server, who then views the received answers on their smartphone screen.

[1402] Collecting user feedback

[1403] The user can rate and comment on the answers they receive, and this feedback is sent back to the server, which stores it in a database and uses it as training data for the generative AI model.

[1404] Accuracy Improvement and Content Recommendation

[1405] The server applies the collected feedback to the generative AI model to improve its accuracy, and also applies an algorithm to recommend optimal content based on the user's emotional data.

[1406] Specific examples

[1407] If a user inputs and sends the message, "I'm feeling down, so please recommend some uplifting movies," the emotion analysis engine will analyze "sadness," and the generative AI model will present a list of uplifting movies. An example of an actual prompt is as follows:

[1408] Prompt Sentence Examples

[1409] If a user asks, "I'm feeling down, please recommend some uplifting movies," the generative AI model should take that emotion into consideration and suggest the best movie list. The suggested movie list should include movies with uplifting scenes and positive storylines.

[1410] In this way, the system of the present invention realizes appropriate answers and content recommendations that take into account the user's emotions.

[1411] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1412] Step 1:

[1413] The user uses a smartphone to input the details of their inquiry and presses the "Send" button. The input is sent as text data from the device to the server, which receives this text data.

[1414] Step 2:

[1415] The server stores the received text data in a database. This process stores the query for future reference and analysis. The input is text data, and the output is stored in the database.

[1416] Step 3:

[1417] The server sends the query text data to the emotion analysis engine. The emotion analysis engine analyzes the user's emotions from the text data and extracts emotion tags such as "sadness," "joy," and "anger." The input is the query text data, and the output is emotion tag data.

[1418] Step 4:

[1419] The server sends the same query text data to the generative AI model, which analyzes the text data and generates an optimal answer. At this time, it adjusts the tone of the answer based on the emotion tag obtained from the emotion analysis engine. The input is the query text data and emotion tag, and the output is the adjusted answer text.

[1420] Step 5:

[1421] The server returns the generated answer and emotion tag to the user, along with the answer text and emotion tag generated by the generative AI model. The input is the answer text and emotion tag, and the output is the answer displayed on the user's screen.

[1422] Step 6:

[1423] The user checks the returned answers and rates them. The user provides feedback through the rating buttons and comment fields. The input is the user's rating and comment text, and the output is the feedback data.

[1424] Step 7:

[1425] The server stores the feedback data received from the user in a database. The stored feedback data is used as training data for the generative AI model. The input is the feedback data, and the output is saved in the database.

[1426] Step 8:

[1427] The server applies the saved feedback data to the generative AI model to improve its accuracy. This process is performed to update the algorithm of the generative AI model and improve the accuracy of responses to subsequent inquiries. The input is the feedback data, and the output is the updated generative AI model.

[1428] Step 9:

[1429] Based on the user's emotional data, the server recommends the most suitable content to the user. The emotional data is analyzed, a list of content desired by the user is compiled, and this is provided to the user as a recommendation list. The input is emotional data, and the output is a list of recommended content.

[1430] By describing the specific actions at each step in detail, it becomes clear how the system responds to user inquiries and provides answers that take emotions into account. As part of this process, prompts are appropriately set and a generative AI model and sentiment analysis engine are effectively used to achieve high user satisfaction.

[1431] 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.

[1432] 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.

[1433] 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.

[1434] [Fourth embodiment]

[1435] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1436] 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.

[1437] 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).

[1438] 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.

[1439] 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.

[1440] 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).

[1441] 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.

[1442] 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.

[1443] 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.

[1444] 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.

[1445] 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.

[1446] 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.

[1447] 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."

[1448] The system of the present invention has a process for receiving a query entered by a user, analyzing the query, generating an optimal answer, and returning the generated answer to the user. Furthermore, the system receives feedback from the user and transmits the feedback to an artificial intelligence model to improve the accuracy of the model. Specific embodiments for implementing this system are described below.

[1449] What the program does

[1450] A user posts a question

[1451] A user uses the bulletin board interface to enter their inquiry and presses the "Post" button, which sends the user's question to the server.

[1452] The server receives the query

[1453] The server receives user queries and stores them in a database for later analysis and answer generation.

[1454] The server sends the question to the generative AI model

[1455] The server sends the received question to the generative AI model, where the question is passed in text format and the model prepares it for analysis.

[1456] Generative AI model performs analysis and generates answers

[1457] The generative AI model analyzes the user's question and understands their intent. For example, if a user asks, "What is a list comprehension in Python?", the model generates an answer about the basic concept and usage of list comprehension.

[1458] Example answer: "List comprehensions are a concise way to create lists in Python. For example, \[x for x in range(10)\] creates a list containing the numbers 0 through 9."

[1459] The server sends the answer back to the user

[1460] The server receives the answer generated by the generative AI model and sends it back to the user, which then displays the appropriate answer on the user's screen.

[1461] Users rate the answers

[1462] Users provide feedback on the answers they receive, such as rating whether the answer was helpful or if there is room for improvement. This feedback information is sent to the server via rating buttons and comments.

[1463] The server feeds the evaluation back to the AI ​​model.

[1464] The server sends the feedback received from the user to the generative AI model, which then uses this feedback to improve itself and its ability to generate higher quality answers in the future.

[1465] Specific examples

[1466] For example, if a user posts a question such as "I don't really understand Python decorators," the system will act as follows:

[1467] 1. A user enters a question and posts it.

[1468] 2. The server receives the question and stores it in a database.

[1469] 3. The server sends the question to the generative AI model.

[1470] 4. A generative AI model analyzes the question and generates an answer, such as "A decorator is a mechanism for modifying functions to provide additional functionality."

[1471] 5. The server generates the answer and sends it back to the user.

[1472] 6. Users view the answer and rate whether it was helpful or not.

[1473] 7. The server feeds the evaluation back into the generative AI model, improving its accuracy.

[1474] In this way, the system of the present invention can provide fast and accurate answers to user questions, and can leverage user feedback to continually improve the quality of knowledge throughout the system.

[1475] The processing flow will be explained below.

[1476] Step 1:

[1477] A user opens the message board interface, enters the content of their inquiry, and presses the "Post" button. This action sends the question entered by the user from the terminal to the server.

[1478] Step 2:

[1479] The device converts the user's input into JSON format and sends it to the server as an HTTP request. The specific data format is as follows:

[1480] json

[1481] {

[1482] "question": "What is a list comprehension in Python?"

[1483] }

[1484] Step 3:

[1485] The server analyzes the request received from the terminal and extracts the inquiry content, while also validating the user's question to ensure it is in the correct format.

[1486] Step 4:

[1487] The server stores the user's question in a database, allowing you to later refer to the answers and feedback for this question.

[1488] python

[1489] db.save_question(question)

[1490] Step 5:

[1491] The server passes the saved question to the generative AI model, which then sends the question as an API request to the generative AI model.

[1492] python

[1493] ai_response = generate_answer(question)

[1494] Step 6:

[1495] The generative AI model analyzes the questions it receives and generates appropriate answers based on the questions, using natural language processing and machine learning algorithms in the analysis process.

[1496] Example: "List comprehensions are a concise way to create lists in Python. For example, \[x for x in range(10)\] creates a list containing the numbers 0 through 9."

[1497] Step 7:

[1498] The generative AI model sends the generated answer back to the server, which receives it and converts it back into JSON format to send back to the user.

[1499] Step 8:

[1500] The server generates an answer and sends it to the user, where it is returned as an HTTP response and displayed on the device.

[1501] json

[1502] {

[1503] "answer": "List comprehensions are a concise way to create lists in Python. For example, [x for x in range(10)] creates a list containing the numbers 0 through 9."

[1504] }

[1505] Step 9:

[1506] The user views the presented answer on the device and evaluates whether the answer meets their expectations.

[1507] Step 10:

[1508] Users can use the provided rating feature to provide feedback on answers, such as "helpful" or "more details needed."

[1509] Step 11:

[1510] The device converts the user's feedback into JSON format and sends it to the server as an HTTP request.

[1511] json

[1512] {

[1513] "question_id": "12345",

[1514] "feedback": "This answer was helpful"

[1515] }

[1516] Step 12:

[1517] The server receives feedback from the device and sends it to a database and generative AI model for storage and model training.

[1518] python

[1519] update_model_with_feedback(question_id, feedback)

[1520] Step 13:

[1521] The generative AI model adjusts its algorithm based on the feedback it receives, which improves the accuracy of responses to future inquiries.

[1522] Step 14:

[1523] The server stores user feedback information in a database for future reference, allowing past evaluation data to be used to improve the system as a whole.

[1524] python

[1525] db.save_feedback(question_id, feedback)

[1526] Example 1

[1527] 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."

[1528] Conventional inquiry response systems have had difficulty in providing prompt and appropriate answers to user inquiries. They also have been unable to effectively utilize user feedback to improve system performance. Furthermore, the processes required to manage inquiry content and generate appropriate answers are cumbersome, creating a need for efficient system operation.

[1529] 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.

[1530] In this invention, the server includes means for receiving queries entered by users, means for saving the query content in a database, means for analyzing the query content and using a generative AI model to generate an optimal answer, means for returning the generated answer to the user, means for receiving feedback on the answer from the user, and means for sending the feedback to the generative AI model to improve the accuracy of the model. This enables queries to be processed quickly and appropriately, and system performance can be improved by effectively utilizing user feedback.

[1531] A "query" is a question or request for information made by a user to the system.

[1532] A "database" is a system for efficiently managing and storing inquiry details, evaluation data, etc.

[1533] A "generative AI model" is a program or algorithm that uses artificial intelligence technology to analyze a user's question and generate the optimal answer.

[1534] "Feedback" refers to the evaluation or opinion of the answer provided by the user.

[1535] A "prompt" is a text-based instruction used to input a question to the generative AI model.

[1536] A "server" is a computer system that receives inquiries from users, performs the necessary processing, and works with the generative AI model to generate and return answers.

[1537] "User" means an individual or organization that uses the system to make an inquiry.

[1538] "Evaluation data" refers to information used to manage and store the content of feedback provided by users.

[1539] The "system" is an integrated mechanism that performs a series of steps: accepting inquiries, managing a database, generating answers using a generative AI model, returning answers to users, processing feedback, and improving the model.

[1540] The system of the present invention receives a query from a user, generates an optimal answer using a generative AI model, and returns it to the user. It also receives feedback from the user and sends it to the generative AI model to improve the accuracy of the model. A specific method for implementing the present invention is described below.

[1541] System Configuration

[1542] This system mainly consists of a server, a terminal, a generative AI model, and a database.

[1543] Hardware and software used

[1544] Server: A cloud server can be used to process user queries and generate answers in conjunction with the generative AI model.

[1545] Devices: A variety of devices are available, including PCs, smartphones, and tablets, and users use these to make inquiries.

[1546] Generative AI models: Use large language models (e.g., OpenAI GPT-3 or GPT-4) to analyze queries and generate optimal answers.

[1547] Database: Use a relational database such as PostgreSQL or MySQL to store inquiries and user feedback.

[1548] Operational Overview

[1549] 1. A user posts a query

[1550] A user accesses the bulletin board interface through a web browser, enters their inquiry, and presses the "Post" button. For example, a user enters the following question:

[1551] "Please explain list comprehensions in Python."

[1552] 2. The server receives the query

[1553] The server receives user queries as HTTP requests and stores them in a database for later analysis and answer generation.

[1554] 3. The server sends a query to the generative AI model

[1555] The server takes the query stored in the database and converts it into a prompt like this:

[1556] "A user has posted the following question: 'What is a list comprehension in Python?' Please generate a suitable answer."

[1557] This prompt is then sent to the API of the generative AI model.

[1558] 4. Generative AI models generate answers

[1559] The generative AI model analyzes the prompt and generates the best answer for the question. For example, it might generate an answer like this:

[1560] "List comprehensions are a concise way to create lists in Python. For example, [x for x in range(10)] creates a list containing the numbers 0 through 9."

[1561] This response is sent back to the server.

[1562] 5. The server generates the answer and sends it to the user

[1563] The server receives the generated answer, converts it into HTML format for display on the user's terminal, and sends it to the user. During this process, the answer to the question is displayed on the user's screen.

[1564] 6. Users rate answers

[1565] The user can rate the displayed answer and indicate whether it was helpful or whether it needs improvement. If the user rates it as "helpful," the rating is sent to the server.

[1566] 7. The server feeds the evaluation back to the generative AI model

[1567] The server feeds user evaluation data back to the generative AI model, which allows the model to improve itself and increase its ability to generate more accurate answers.

[1568] Specific examples

[1569] If a user posts a question such as "I don't really understand Python decorators," the process goes as follows:

[1570] 1. The user types, "I don't really understand Python decorators," and presses the submit button.

[1571] 2. The server receives this question and stores it in a database.

[1572] 3. The server parses the question and sends a prompt to the generative AI model, such as:

[1573] "A user has posted the following question: 'I don't really understand Python decorators.' Please generate a suitable answer for this."

[1574] 4. The generative AI model analyzes the prompt and generates an answer like this:

[1575] "Decorators are a mechanism for decorating functions to provide additional functionality."

[1576] 5. The server returns the generated answer to the user and displays it on the user's terminal.

[1577] 6. The user views the answer and rates it as helpful.

[1578] 7. The server feeds the evaluation back into the generative AI model, improving its accuracy.

[1579] This allows the system to provide fast and accurate answers to user queries, while also leveraging user feedback to continuously improve overall system performance.

[1580] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1581] Program processing flow

[1582] Step 1:

[1583] A user uses the bulletin board interface to enter their inquiry and presses the "Post" button. Specifically, the user enters the following text into the inquiry form displayed in their web browser:

[1584] "Please explain list comprehensions in Python."

[1585] Based on this input, the query content is sent to the server.

[1586] Step 2:

[1587] The server receives the query sent by the user as an HTTP request and executes a SQL query like the following to store the query in the database:

[1588] sql

[1589] INSERT INTO questions (content, user_id, created_at) VALUES ('What is a list comprehension in Python?', 123, NOW());

[1590] Input: User's inquiry and user ID

[1591] Output: Query data stored in a database

[1592] Step 3:

[1593] The server takes the query stored in the database and converts it into a prompt to be passed to the generative AI model. Specifically, it generates a text prompt like this:

[1594] "A user has posted the following question: 'What is a list comprehension in Python?' Please generate a suitable answer."

[1595] Input: Query content retrieved from the database

[1596] Output: Generated prompt statement

[1597] Step 4:

[1598] The server sends a prompt to the API of the generative AI model, using an HTTP request to send the prompt to the model.

[1599] Input: Generated prompt text

[1600] Output: The request data sent to the generative AI model

[1601] Step 5:

[1602] The generative AI model analyzes the received prompt and generates the best answer to the user's question. Specifically, it generates the following answer:

[1603] "List comprehensions are a concise way to create lists in Python. For example, [x for x in range(10)] creates a list containing the numbers 0 through 9."

[1604] Input: prompt statement

[1605] Output: Generated answer text

[1606] Step 6:

[1607] The server receives the answer returned by the generative AI model and converts it into HTML format for display to the user. Specifically, it generates HTML code like this:

[1608] html

[1609]

[1610] List comprehensions are a concise way to create lists in Python, for example [x for x in range(10)] will create a list containing the numbers 0 to 9.

[1611]

[1612] Input: Answer text returned by the generative AI model

[1613] Output: Generated HTML formatted response data

[1614] Step 7:

[1615] The server returns the generated HTML answer to the user's terminal, which displays the answer to the question on the user's screen.

[1616] Input: Generated HTML formatted answer data

[1617] Output: Answer displayed on the user's terminal

[1618] Step 8:

[1619] Users rate the displayed answers, providing feedback such as "helpful" or "needs improvement." Specifically, the following data is generated:

[1620] "Rating: Helpful, Question ID: 456"

[1621] Input: User rating

[1622] Output: Rating data sent to the server

[1623] Step 9:

[1624] The server receives the evaluation data sent by the user and feeds it back to the generative AI model. Specifically, it sends it to the feedback API in the following format:

[1625] "Feedback: 'Helpful', Question ID: 456"

[1626] Input: User rating data

[1627] Output: Feedback data sent to the generative AI model

[1628] This allows the entire system to respond quickly and accurately to user queries and use feedback to improve the accuracy of the model.

[1629] (Application example 1)

[1630] 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."

[1631] Traditionally, customer guidance in brick-and-mortar stores relied on sales staff, which created efficiency issues. Searching for products and checking inventory in the store also took time and effort, often resulting in lower customer satisfaction. Especially during busy times, sales staff were often unable to respond to each individual customer, making it difficult to provide appropriate guidance.

[1632] 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.

[1633] In this invention, the server includes means for receiving an inquiry entered by a user, means for analyzing the content of the inquiry and using an artificial intelligence model to generate an optimal answer, means for returning the generated answer to the user, means for receiving feedback on the answer from the user, means for sending the feedback to the artificial intelligence model to improve the accuracy of the model, means for converting the user's voice inquiry into text using speech recognition technology, means for displaying the generated answer on a display device, and means for the user to provide feedback via a head-mounted display. This makes it possible to efficiently and quickly provide appropriate answers to customer questions in physical stores, thereby improving customer satisfaction.

[1634] The "means for receiving user-input inquiries" is an interface for obtaining questions or requests from users in digital form.

[1635] "Means using an artificial intelligence model to analyze the content of the inquiry and generate an optimal response" refers to an artificial intelligence mechanism used to analyze the input inquiry and automatically generate an appropriate response.

[1636] The "means for returning the generated answer to the user" refers to a communication means or display means for conveying the answer generated by the artificial intelligence model to the user.

[1637] The "means for receiving feedback on answers from users" is a system component for collecting ratings and opinions of answers provided by users.

[1638] "Means for sending feedback to an AI model to improve the model's accuracy" refers to the process of inputting collected feedback information into an AI model to improve its performance and the quality of its answers.

[1639] "Means for converting a user's spoken query into text using speech recognition technology" is the process of utilizing speech recognition software to convert a user's verbal query into text form.

[1640] The "means for displaying the generated answer on a display device" is a mechanism for displaying the answer generated by the artificial intelligence model on a display device so that the user can visually confirm it.

[1641] The "means for the user to provide feedback through the head-mounted display" is an interface that allows the user to provide visual or audio feedback using the head-mounted display.

[1642] The system of the present invention is a process that receives a user-input query, analyzes the query, generates an optimal answer, and returns it to the user. It also receives feedback from the user and sends it to an artificial intelligence model to improve the model's accuracy. Specific embodiments for implementing this system are described below.

[1643] Hardware and software used

[1644] Head-mounted display (HMD)

[1645] microphone

[1646] server

[1647] Display device

[1648] Speech recognition software: Google Speech-to-Text

[1649] Generative AI model: OpenAI GPT-4

[1650] Programming language: Python

[1651] Data processing and calculation flow

[1652] Acquiring voice input

[1653] The user wears a head-mounted display (HMD) and issues a voice query through a microphone. The HMD's voice recognition function is used to capture the user's voice.

[1654] Converting audio data to text

[1655] The server uses voice recognition software (Google Speech-to-Text) to convert the voice data into text, which is then processed as the query.

[1656] Inquiry analysis and answer generation

[1657] The server sends the text of the query to a generative AI model (OpenAI GPT-4), which analyzes the query and generates the most appropriate answer.

[1658] Show Answers

[1659] The generated answers are sent to the server and displayed on the HMD display, allowing the user to see the answers in real time.

[1660] Gathering user feedback

[1661] The user provides feedback on the generated answers through the HMD, either by voice or touch. The feedback data is sent to the server and used to improve the generative AI model.

[1662] Specific examples

[1663] For example, if a user asks the question "Where is the tomato sauce?", the system will act as follows:

[1664] 1. The user uses the HMD and microphone to ask, "Where is the tomato sauce?"

[1665] 2. The server converts the speech into text and sends the text data, "Where is the tomato sauce?" to the generative AI model.

[1666] 3. The generative AI model generates the answer, "It's on the top shelf in the grocery section."

[1667] 4. The answer is displayed on the HMD display and the user confirms it.

[1668] 5. The user rates whether they were satisfied with the answer and provides feedback.

[1669] 6. Feedback is sent to the generative AI model through the server, improving the model's accuracy.

[1670] Prompt Sentence Examples

[1671] User asks: "Where is the tomato sauce?"

[1672] Answer: "Tomato sauce is on the top shelf in the grocery section."

[1673] In this way, the system of the present invention can efficiently and effectively guide customers in physical stores, and can improve the accuracy of the model based on user feedback, allowing for even higher quality service in the future.

[1674] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1675] Step 1:

[1676] A user wears a head-mounted display (HMD) and asks a question by voice. The user's voice input is captured through a microphone. The input is the user's voice data, and this voice data is the target for the next step. Specifically, the user asks, "Where is the tomato sauce?"

[1677] Step 2:

[1678] The user's voice data is sent to the server, and the server converts the voice data into text using speech recognition software (Google Speech-to-Text). The input is voice data, and the output is text data. Specifically, the text generated is "Where is the tomato sauce?"

[1679] Step 3:

[1680] The converted text data is sent to the server's generative AI model (OpenAI GPT-4). The server uses the generative AI model to analyze the text and generate the optimal answer. The input is text data, and the output is the answer text. Specifically, the model generates the answer, "Tomato sauce is on the top shelf in the grocery section."

[1681] Step 4:

[1682] The generated answer text is sent from the server to the HMD. The answer is displayed on the HMD display. The input is the answer text, and the output is the text displayed on the HMD display. Specifically, the answer "Tomato sauce is on the top shelf in the food section" is displayed on the HMD display.

[1683] Step 5:

[1684] The user checks the answer displayed through the HMD and provides feedback. The feedback is input by voice or touch. The input is the user's feedback data, which is used in the next step. Specifically, the user evaluates the answer as "helpful."

[1685] Step 6:

[1686] The server receives the user's feedback data and sends it to the generative AI model. The model uses the feedback to self-train and improve the accuracy of future answers. The input is the feedback data, and the output is updated model parameters. Specifically, the model learns to "generate better answers to similar questions in the future."

[1687] 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.

[1688] The system of the present invention has a process for receiving a query entered by a user, analyzing the query, generating an optimal answer, and returning the generated answer to the user. Furthermore, the system receives feedback from the user and sends the feedback to an artificial intelligence model to improve the accuracy of the model. Furthermore, by combining an emotion engine that recognizes the user's emotions, the system adjusts the tone of the answer to provide an answer that is more in line with the user's intention. Specific embodiments for implementing this system are described below.

[1689] What the program does

[1690] A user posts a question

[1691] A user uses the bulletin board interface to enter a question and presses the "Post" button, which sends the user's question to the server.

[1692] The server receives the query

[1693] The server receives the user's query and stores it in a database, allowing it to be referenced later.

[1694] The server sends the question to the emotion engine and generative AI model.

[1695] The server first sends the question received to the emotion engine to recognize the user's emotion. The emotion engine extracts emotions such as "joy," "sadness," and "anger" from the user's input. At the same time, the server sends the question to the generative AI model for analysis and answer generation.

[1696] Example: In response to the question "Tell me about list comprehensions in Python," the emotion engine recognizes "curiosity."

[1697] A generative AI model generates answers, reflecting the results of the emotion engine.

[1698] The generative AI model analyzes the user's question and generates an appropriate answer based on its content. It also adjusts the tone of the answer, taking into account the user's emotions as recognized by the emotion engine. For example, if the user expresses "sadness," the answer will have a more friendly and polite tone.

[1699] Example answer: "List comprehensions are a concise way to create lists in Python. For example, \[x for x in range(10)\] creates a list containing the numbers 0 through 9."

[1700] The server sends the answer and emotion data back to the user.

[1701] The server returns the answer generated by the generative AI model and the emotion data recognized by the emotion engine to the user, and through this process, the appropriate answer is displayed on the user's screen.

[1702] Users rate the answers

[1703] Users can provide feedback on the answers they see, using the rating buttons and comments section to rate whether the answer was helpful or if there are areas for improvement. Users can also provide emotional feedback.

[1704] The server receives the ratings and emotion data.

[1705] The server receives user feedback and emotion data and sends it to a database and generative AI model for storage and model training.

[1706] Reflecting feedback from rating and sentiment data

[1707] The generative AI model adjusts its algorithm based on the feedback it receives, and emotional data is also used as training data for the model to improve the accuracy of responses to future inquiries.

[1708] Specific examples

[1709] For example, if a user posts a question such as "I don't really understand Python decorators," the system will act as follows:

[1710] 1. A user enters a question and posts it.

[1711] 2. The server receives the question and stores it in a database.

[1712] 3. The server sends the question to the emotion engine, which recognizes the user's emotion as "confusion," and simultaneously sends the question to the generative AI model.

[1713] 4. The generative AI model analyzes the question and generates an answer, such as "Decorators are a mechanism for modifying functions to provide additional functionality," and adjusts the tone to be more helpful based on the results of the emotion engine.

[1714] 5. The server returns the generated answer to the user along with the emotion data.

[1715] 6. Users view the answers and rate their quality.

[1716] 7. The server receives the feedback and emotion data and feeds it back into the database and generative AI model.

[1717] In this way, the system of the present invention can not only provide fast and accurate answers to user questions, but also respond in a way that takes into account the user's emotions, and utilize user feedback to continuously improve the quality of the system's overall knowledge and user satisfaction.

[1718] The processing flow will be explained below.

[1719] Step 1:

[1720] A user opens the message board interface, enters the content of their inquiry, and presses the "Post" button. This action sends the user's question from their device to the server.

[1721] Step 2:

[1722] The device converts the user's input into JSON format and sends it to the server as an HTTP request. The specific data format is as follows:

[1723] json

[1724] {

[1725] "question": "What is a list comprehension in Python?"

[1726] }

[1727] Step 3:

[1728] The server analyzes the request received from the terminal and extracts the inquiry content, while also validating the user's question to ensure it is in the correct format.

[1729] Step 4:

[1730] The server stores the user's question in a database, allowing you to later refer to the answers and feedback for this question.

[1731] python

[1732] db.save_question(question)

[1733] Step 5:

[1734] The server sends the received question to the emotion engine, which analyzes the emotion from the user's input.

[1735] Example: In response to the question "Tell me about list comprehensions in Python," the emotion engine recognizes "curiosity."

[1736] Step 6:

[1737] The server sends the user's question to the generative AI model, along with the emotion data obtained by the emotion engine.

[1738] python

[1739] ai_response, user_emotion = generate_answer(question, emotion)

[1740] Step 7:

[1741] The generative AI model analyzes the user's question and generates an appropriate answer based on the question's content. This analysis process uses natural language processing and machine learning algorithms. When generating the answer, it takes into account emotional data obtained by the emotion engine and adjusts the tone of the answer.

[1742] Example answer: "List comprehensions are a concise way to create lists in Python. For example, \[x for x in range(10)\] creates a list containing the numbers 0 through 9."

[1743] Step 8:

[1744] The generative AI model sends the generated answer back to the server, which receives it and converts it back into JSON format to send back to the user.

[1745] json

[1746] {

[1747] "answer": ai_response,

[1748] "emotion": user_emotion

[1749] }

[1750] Step 9:

[1751] The server sends the generated answer and emotion data to the user, which is returned as an HTTP response and displayed on the device.

[1752] json

[1753] {

[1754] "answer": "List comprehensions are a concise way to create lists in Python. For example, [x for x in range(10)] creates a list containing the numbers 0 through 9.",

[1755] "emotion": "curiosity"

[1756] }

[1757] Step 10:

[1758] The user views the presented answer on the device and evaluates whether the answer meets their expectations.

[1759] Step 11:

[1760] Users can use the provided rating feature to provide feedback on answers, such as "helpful" or "more details needed."

[1761] Step 12:

[1762] The device converts the user's feedback into JSON format and sends it to the server as an HTTP request.

[1763] json

[1764] {

[1765] "question_id": "12345",

[1766] "feedback": "This answer was helpful",

[1767] "emotion": "satisfaction"

[1768] }

[1769] Step 13:

[1770] The server receives feedback from the device and sends it to a database and generative AI model for storage and model training.

[1771] python

[1772] update_model_with_feedback(question_id, feedback, emotion)

[1773] Step 14:

[1774] The generative AI model adjusts its algorithm based on the feedback it receives, which improves the accuracy of responses to future inquiries. Emotional data is also used as training data for the model.

[1775] python

[1776] model.learn_from_feedback(feedback, emotion)

[1777] Step 15:

[1778] The server stores the user's feedback information and emotion data in a database for future reference, allowing the system to utilize past evaluation data and emotion data to improve the system as a whole.

[1779] python

[1780] db.save_feedback(question_id, feedback, emotion)

[1781] Example 2

[1782] 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."

[1783] Conventional conversational AI systems generate responses uniformly without considering the user's emotions, making it difficult to communicate optimally according to the user's intentions and the situation. Furthermore, while mechanisms exist for receiving feedback, it is difficult to effectively utilize that feedback to improve the model. Furthermore, the lack of a function to adjust the tone of the generated responses often leads to a poor user experience.

[1784] 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.

[1785] In this invention, the server includes means for receiving a query entered by a user, means for analyzing the content of the query and using an AI model to generate an optimal answer, means for returning the generated answer to the user, means for receiving feedback on the answer from the user, means for sending the feedback to the AI ​​model to improve the accuracy of the model, and means for adjusting the tone of the answer using an emotion engine that recognizes the user's emotions. This makes it possible to provide an appropriate answer that reflects the user's emotions and to continuously improve the accuracy of the model by effectively utilizing the feedback.

[1786] "User" refers to the entity that uses the system to make an inquiry.

[1787] An "inquiry" refers to a question or request that a user inputs to the system.

[1788] "Analysis" refers to the process of understanding the user's inquiry and grasping its meaning.

[1789] An "artificial intelligence model" refers to a program that uses machine learning technology to generate optimal answers.

[1790] "Generated answer" refers to a response generated by an artificial intelligence model based on the analysis results.

[1791] "Feedback" refers to the ratings and opinions of answers provided by users.

[1792] "Accuracy" refers to the degree to which the AI ​​model's answers match the user's intentions and requirements.

[1793] An "emotion engine" refers to software that recognizes and analyzes emotions from user input.

[1794] "Tone" refers to the expressive and phrasing characteristics of the responses generated.

[1795] A "database" refers to a system that systematically stores information such as inquiries and feedback.

[1796] The system of the present invention has a process for receiving a user's query, analyzing it, generating an optimal answer, and returning the generated answer to the user. It also receives feedback from the user and sends the feedback to an artificial intelligence model to improve the accuracy of the model. It also combines an emotion engine that recognizes the user's emotions to adjust the tone of the answer and provide an answer that more closely matches the user's intentions.

[1797] System Embodiments

[1798] A user posts a question

[1799] The user uses the bulletin board interface on the terminal, enters the content of the inquiry, and presses the "Post" button. This operation sends the user's question to the server. The terminal can be an information device such as a personal computer or smartphone.

[1800] The server receives the query

[1801] The server receives queries from users and stores the details in a database. The database stores information such as the query details, user ID, and timestamp. This database can be a relational database (RDB) or a NoSQL database.

[1802] The server sends the question to the emotion engine and generative AI model.

[1803] The server first sends the received question to an emotion engine to recognize the user's emotions. The emotion engine uses software that uses natural language processing (NLP) technology (e.g., IBM Watson's emotion analysis API). At the same time, the server sends the question to a generative AI model, which analyzes it and generates an answer. The generative AI model used is, for example, GPT-3 (OpenAI).

[1804] A generative AI model generates answers, reflecting the results of the emotion engine.

[1805] The generative AI model analyzes the user's question and generates an appropriate answer based on its content. It also adjusts the tone of the answer, taking into account the user's emotions as recognized by the emotion engine. For example, if the user expresses "sadness," the answer will have a more friendly and polite tone.

[1806] The server sends the answer and emotion data back to the user.

[1807] The server returns the answer generated by the generative AI model and the emotion data recognized by the emotion engine to the user, and through this process, the appropriate answer is displayed on the user's screen.

[1808] Users rate the answers

[1809] Users can provide feedback on the displayed answers by rating them using the rating buttons and comments section, indicating whether the answer was helpful or if there are areas for improvement. Users can also provide emotional feedback.

[1810] The server receives the ratings and emotion data.

[1811] The server receives user feedback and emotion data and sends it to a database and generative AI model for storage and model training.

[1812] Reflecting feedback from rating and sentiment data

[1813] The generative AI model adjusts its algorithm based on the feedback it receives, and emotional data is also used as training data for the model to improve the accuracy of responses to future inquiries.

[1814] Specific examples

[1815] For example, if a user posts a question such as "I don't really understand Python decorators," the system will act as follows:

[1816] 1. A user enters a question and posts it.

[1817] 2. The server receives the question and stores it in a database.

[1818] 3. The server sends the question to the emotion engine, which recognizes the user's emotion as "confusion," and simultaneously sends the question to the generative AI model.

[1819] 4. The generative AI model analyzes the question and generates an answer, such as "A decorator is a mechanism for modifying functions to provide additional functionality," and adjusts the tone to be more helpful, reflecting the results of the emotion engine.

[1820] 5. The server returns the generated answer to the user along with the emotion data.

[1821] 6. Users view the answers and provide ratings and feedback.

[1822] 7. The server receives the feedback and emotion data and stores it in a database.

[1823] 8. The generative AI model learns from the feedback data and improves the accuracy of the answers next time.

[1824] Examples of prompt statements

[1825] Using the prompt "Tell me about Python decorators. What are they and how are they used?", the emotion engine recognizes the "confusion" and adjusts the generative AI model to return a clear and polite explanation.

[1826] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1827] Step 1: User posts a question

[1828] A user uses the bulletin board interface to enter a question and press the "Post" button. For example, they might enter "Please tell me about list comprehensions in Python." This action converts the entered question into JSON format on the terminal and sends it to the server. The input is the user's question text, and the output is the JSON-formatted data sent to the server.

[1829] Step 2: The server receives the query

[1830] The server waits for user queries on a receiving port and parses the received JSON data. Specifically, the server stores the query content in a database, which stores meta information such as the query content, user ID, and timestamp. The input is the query data sent by the user, and the output is a record stored in the database.

[1831] Step 3: The server sends the question to the emotion engine and generative AI model

[1832] The server first sends the received question to an emotion engine to recognize the user's emotions. The emotion engine uses NLP technology to extract emotions from the input text. For example, it uses IBM Watson's emotion analysis API to extract emotions such as "joy," "sadness," and "anger." At the same time, the server sends the question to a generative AI model (e.g., GPT-3), which analyzes the question and generates the optimal answer. The input is the question text from the user, and the output is the emotion data extracted by the emotion engine and the answer generated by the generative AI model.

[1833] Step 4: The generative AI model generates an answer, reflecting the results of the emotion engine.

[1834] The generative AI model analyzes the question and generates an appropriate answer based on its content. At this time, the tone is adjusted to reflect the results of the emotion engine. For example, if the emotion engine recognizes "confused," the explanation will have a more friendly and specific tone. For example, it generates the answer "List comprehension is a concise way to create lists in Python. For example, \[x for x in range(10)\] creates a list containing the numbers 0 to 9." The input is emotion data and the question text, and the output is the tone-adjusted answer text.

[1835] Step 5: The server sends the answer and emotion data back to the user

[1836] The server compiles the answer generated by the generative AI model and the emotion data recognized by the emotion engine in JSON format and returns it to the user. This is sent to the device as an HTTP response, and the appropriate answer is displayed on the user's screen. The input is the data obtained from the generative AI model and the emotion engine, and the output is displayed on the user's device.

[1837] Step 6: Users rate the answers

[1838] Users provide feedback on the displayed answers using rating buttons (e.g., "helpful" or "not helpful") or a comment field. Users may also provide emotional feedback. The input is the user's feedback, and the output is the rating data sent to the server.

[1839] Step 7: The server receives the ratings and sentiment data

[1840] The server receives feedback and emotion data from users and stores it in a database. Feedback information, such as answer ratings and areas for improvement, is stored in the "feedback" table. The input is the user's rating data, and the output is the feedback information stored in the database.

[1841] Step 8: Incorporate feedback from rating and sentiment data

[1842] The generative AI model adjusts its algorithm based on the feedback it receives. The server periodically runs a program that incorporates the feedback data into the model, and also uses the emotional data as training data. This improves the accuracy of responses to subsequent inquiries. The input is feedback and emotional data, and the output is an improved generative AI model.

[1843] (Application example 2)

[1844] 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."

[1845] Conventional content distribution services face the challenges of tedious processes for providing appropriate answers to user inquiries, and difficulty in adjusting the tone of the answer based on the user's emotions. Answers that do not take the user's emotions into consideration can ruin the user experience and reduce satisfaction. Furthermore, there is a lack of mechanisms for efficiently utilizing user feedback to improve the accuracy of AI models. Furthermore, the functionality for recommending content based on user emotional data is limited. To solve these challenges, a system incorporating tone adjustment using an emotion analysis engine and content recommendation functions is needed.

[1846] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving an inquiry entered by a user, means for analyzing the inquiry content and using a generative AI model to generate an optimal answer, means for returning the generated answer to the user, means for receiving feedback on the answer from the user, means for sending the feedback to the generative AI model to improve the accuracy of the model, means for using an emotion analysis engine that recognizes the user's emotions and adds an appropriate tone, and means for recommending content based on the user's emotion data. This makes it possible to adjust the tone according to the user's emotions and provide a more appropriate answer. Furthermore, it is possible to provide more accurate answers and a more satisfying user experience than conventional systems.

[1847] A "means for receiving user-input queries" refers to an interface or software component that receives questions or requests made by users to the system.

[1848] "Means of using a generative AI model that analyzes the content of the inquiry and generates the optimal response" refers to an artificial intelligence model that analyzes the user's inquiry using technologies such as natural language processing and generates an appropriate response.

[1849] "Means for returning the generated answer to the user" refers to an interface or communication means for presenting the answer created by the generative AI model to the user.

[1850] The "means for receiving feedback on answers from users" refers to an interface or mechanism for users to provide ratings and comments on the answers they receive.

[1851] "Means for sending feedback to the generative AI model to improve the model's accuracy" refers to a system for collecting feedback information from users and using it to train and tune the generative AI model.

[1852] "Means using an emotion analysis engine that recognizes a user's emotion and applies an appropriate tone" refers to a software engine that analyzes the emotion from a user's input and adjusts the tone of the response according to that emotion.

[1853] "Means for recommending content based on user emotional data" refers to algorithms or systems for recommending optimal content based on the user's emotions.

[1854] "Means for storing data in a database" refers to storage devices or software for storing collected inquiry details, feedback, emotional data, etc.

[1855] The system of the present invention is a system that provides appropriate answers to user inquiries in a content distribution service, recognizes the user's emotions and adjusts the tone of the answers, and recommends content taking the user's emotions into consideration. Specific embodiments for implementing this system are described below.

[1856] Program Overview

[1857] This system consists of a terminal that receives user inquiries, a server that analyzes and processes the inquiry content, and a mechanism for returning the results to the user.

[1858] Hardware and software used

[1859] The following hardware and software are utilized to implement the system.

[1860] Smartphone: Used as a user interface.

[1861] Server: Performs data processing and analysis. Specifically, it includes database servers and application servers.

[1862] Sentiment Analysis Engine: A software engine for analyzing emotions from user input (e.g., IBM Watson, Google Cloud Natural Language).

[1863] Generative AI model: An artificial intelligence model that analyzes questions and generates answers (e.g., OpenAI GPT-4).

[1864] Processing flow and data calculation

[1865] User enters and submits inquiry

[1866] Users use a dedicated smartphone app to enter their inquiries or requests in text format and press the send button, which then sends the inquiry from the device to the server.

[1867] The server receives and stores the query

[1868] The server stores the received queries in a database for future reference and analysis.

[1869] Sentiment analysis and answer generation

[1870] The server sends the received query to the sentiment analysis engine to analyze the user's sentiment. At the same time, the query is sent to the generative AI model to generate an appropriate answer. In this step, the results of the sentiment analysis are used to adjust the tone of the answer.

[1871] Returning answers and emotion data

[1872] The generated answers and the results of the sentiment analysis are sent back to the user from the server, who then views the received answers on their smartphone screen.

[1873] Collecting user feedback

[1874] The user can rate and comment on the answers they receive, and this feedback is sent back to the server, which stores it in a database and uses it as training data for the generative AI model.

[1875] Accuracy Improvement and Content Recommendation

[1876] The server applies the collected feedback to the generative AI model to improve its accuracy, and also applies an algorithm to recommend optimal content based on the user's emotional data.

[1877] Specific examples

[1878] If a user inputs and sends the message, "I'm feeling down, so please recommend some uplifting movies," the emotion analysis engine will analyze "sadness," and the generative AI model will present a list of uplifting movies. An example of an actual prompt is as follows:

[1879] Prompt Sentence Examples

[1880] If a user asks, "I'm feeling down, please recommend some uplifting movies," the generative AI model should take that emotion into consideration and suggest the best movie list. The suggested movie list should include movies with uplifting scenes and positive storylines.

[1881] In this way, the system of the present invention realizes appropriate answers and content recommendations that take into account the user's emotions.

[1882] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1883] Step 1:

[1884] The user uses a smartphone to input the details of their inquiry and presses the "Send" button. The input is sent as text data from the device to the server, which receives this text data.

[1885] Step 2:

[1886] The server stores the received text data in a database. This process stores the query for future reference and analysis. The input is text data, and the output is stored in the database.

[1887] Step 3:

[1888] The server sends the query text data to the emotion analysis engine. The emotion analysis engine analyzes the user's emotions from the text data and extracts emotion tags such as "sadness," "joy," and "anger." The input is the query text data, and the output is emotion tag data.

[1889] Step 4:

[1890] The server sends the same query text data to the generative AI model, which analyzes the text data and generates an optimal answer. At this time, it adjusts the tone of the answer based on the emotion tag obtained from the emotion analysis engine. The input is the query text data and emotion tag, and the output is the adjusted answer text.

[1891] Step 5:

[1892] The server returns the generated answer and emotion tag to the user, along with the answer text and emotion tag generated by the generative AI model. The input is the answer text and emotion tag, and the output is the answer displayed on the user's screen.

[1893] Step 6:

[1894] The user checks the returned answers and rates them. The user provides feedback through the rating buttons and comment fields. The input is the user's rating and comment text, and the output is the feedback data.

[1895] Step 7:

[1896] The server stores the feedback data received from the user in a database. The stored feedback data is used as training data for the generative AI model. The input is the feedback data, and the output is saved in the database.

[1897] Step 8:

[1898] The server applies the saved feedback data to the generative AI model to improve its accuracy. This process is performed to update the algorithm of the generative AI model and improve the accuracy of responses to subsequent inquiries. The input is the feedback data, and the output is the updated generative AI model.

[1899] Step 9:

[1900] Based on the user's emotional data, the server recommends the most suitable content to the user. The emotional data is analyzed, a list of content desired by the user is compiled, and this is provided to the user as a recommendation list. The input is emotional data, and the output is a list of recommended content.

[1901] By describing the specific actions at each step in detail, it becomes clear how the system responds to user inquiries and provides answers that take emotions into account. As part of this process, prompts are appropriately set and a generative AI model and sentiment analysis engine are effectively used to achieve high user satisfaction.

[1902] 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.

[1903] 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.

[1904] 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.

[1905] 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.

[1906] 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.

[1907] 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.

[1908] 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).

[1909] 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.

[1910] 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."

[1911] 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.

[1912] 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).

[1913] 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.

[1914] 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.

[1915] 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.

[1916] 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.

[1917] 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.

[1918] 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.

[1919] 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.

[1920] 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.

[1921] 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.

[1922] 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.

[1923] The following is further disclosed regarding the above embodiment.

[1924] (Claim 1)

[1925] means for receiving a user-entered query;

[1926] Using an artificial intelligence model to analyze the inquiry and generate an optimal response;

[1927] means for returning the generated answer to the user;

[1928] a means for receiving feedback from the user on the response;

[1929] a means for providing feedback to the artificial intelligence model to improve its accuracy; and

[1930] A system including:

[1931] (Claim 2)

[1932] 10. The system of claim 1, further comprising means for storing the query in a database.

[1933] (Claim 3)

[1934] 10. The system of claim 1, further comprising means for storing user rating data in a database.

[1935] "Example 1"

[1936] (Claim 1)

[1937] means for receiving a user-entered query;

[1938] a means for storing the inquiry in a database;

[1939] A means of using a generative AI model that analyzes the content of inquiries and generates optimal answers;

[1940] means for returning the generated answer to the user;

[1941] a means for receiving feedback from the user on the response;

[1942] A means of sending feedback to the generative AI model to improve its accuracy; and

[1943] A system including:

[1944] (Claim 2)

[1945] 10. The system of claim 1, further comprising means for storing user rating data in a database.

[1946] (Claim 3)

[1947] 10. The system of claim 1, further comprising: means for converting query content into prompt sentences and sending them to the generative AI model.

[1948] "Application Example 1"

[1949] (Claim 1)

[1950] means for receiving a user-entered query;

[1951] Using an artificial intelligence model to analyze the inquiry and generate an optimal response;

[1952] means for returning the generated answer to the user;

[1953] a means for receiving feedback from the user on the response;

[1954] a means for providing feedback to the artificial intelligence model to improve its accuracy; and

[1955] means for converting a user's voice query into text using speech recognition technology;

[1956] means for displaying the generated answer on a display device;

[1957] a means for the user to provide feedback via the head mounted display;

[1958] A system including:

[1959] (Claim 2)

[1960] 10. The system of claim 1, further comprising means for storing the query in a database.

[1961] (Claim 3)

[1962] 10. The system of claim 1, further comprising means for storing user rating data in a database.

[1963] "Example 2: Combining Emotion Engines"

[1964] (Claim 1)

[1965] means for receiving a user-entered query;

[1966] Using an artificial intelligence model to analyze the inquiry and generate an optimal response;

[1967] means for returning the generated answer to the user;

[1968] a means for receiving feedback from the user on the response;

[1969] a means for providing feedback to the artificial intelligence model to improve its accuracy; and

[1970] a means for adjusting the tone of the response using an emotion engine that recognizes the user's emotions;

[1971] A system including:

[1972] (Claim 2)

[1973] 10. The system of claim 1, further comprising means for storing the query in a database.

[1974] (Claim 3)

[1975] 10. The system of claim 1, further comprising means for storing user rating data in a database.

[1976] "Application example 2 when combining emotion engines"

[1977] (Claim 1)

[1978] means for receiving a user-entered query;

[1979] A means of using a generative AI model that analyzes the content of inquiries and generates optimal answers;

[1980] means for returning the generated answer to the user;

[1981] a means for receiving feedback from the user on the response;

[1982] A means of sending feedback to the generative AI model to improve its accuracy; and

[1983] using a sentiment analysis engine to recognize the user's emotions and add an appropriate tone;

[1984] A means for recommending content based on user emotion data;

[1985] A system including:

[1986] (Claim 2)

[1987] 10. The system of claim 1, further comprising means for storing the query in a database.

[1988] (Claim 3)

[1989] 10. The system of claim 1, further comprising means for storing user rating data and emotion data in a database. [Explanation of symbols]

[1990] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for receiving a user-entered query; Using an artificial intelligence model to analyze the inquiry and generate an optimal response; means for returning the generated answer to the user; a means for receiving feedback from the user on the response; a means for providing feedback to the artificial intelligence model to improve its accuracy; and A system including:

2. 10. The system of claim 1, further comprising means for storing query content in a database.

3. The system of claim 1 further comprising means for storing user rating data in a database.

Images