system

Abstract

Provide a system. 【Solution means】 Means for inputting questions, Means for receiving the input questions, Means for generating answers and explanations for the received questions, Means for providing the generated answers and explanations to the user, Means for evaluating the generated answers, Means for analyzing the user's tendency based on the evaluation results, Means for generating the next question based on the user's tendency, Means for providing the generated question to the user, Means for interacting with the user in a chat format, A system including the above.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a persona chatbot control method performed by at least one processor, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] Conventionally, learning methods that rely on individual tutors or learning schools have problems such as being restricted by limited time periods and locations, making efficient learning difficult. In particular, it is difficult to obtain a quick answer when a question arises, and it often takes a long time to reinforce weaknesses. Also, it is difficult to maintain motivation in solo learning, and the continuity of learning may be lacking. The present invention aims to solve these problems and provide a system that can provide learning support at any time, 24 hours a day.

Means for Solving the Problems

[0005] The present invention is a learning support system that includes means for inputting questions, means for receiving the inputted questions, means for generating answers and explanations for the received questions, means for providing the generated answers and explanations to the user, means for evaluating the generated answers, means for analyzing the user's tendencies based on the evaluation results, means for generating the next question based on the user's tendencies, means for providing the generated question to the user, and means for interacting with the user in a chat format. The system further includes means for identifying the user's weaknesses based on the evaluation results and generating questions that focus on the identified weaknesses, and means for providing the generated answers and explanations to the user in real time, thereby improving the user's learning efficiency.

[0006] "Questions" refer to parts that users find difficult to understand during the learning process, or problems they don't know how to solve.

[0007] "Means of input" refers to an interface that allows users to input questions and answers in text format.

[0008] "Means of receiving" refers to a mechanism that sends input questions and answers from a terminal to a server and receives them accurately.

[0009] "Generating means" refers to a mechanism that uses AI technology to automatically create answers and explanations to input questions.

[0010] "Means of providing to the user" refers to the interface for notifying and displaying the generated answers, explanations, and feedback to the user.

[0011] "Means of evaluation" refers to a mechanism that uses AI technology to analyze user-entered responses and determine their accuracy and quality.

[0012] "Means of analysis" refers to a mechanism that uses evaluation results to analyze user learning trends and identify individual weaknesses and strengths.

[0013] "Means for generating the next problem" refers to a mechanism that creates new learning problems that correspond to the user's learning level and weaknesses based on the analysis results.

[0014] "A means of communication in a chat format" refers to a communication method in which users can input questions and inquiries 24 hours a day and receive responses in real time. [Brief explanation of the drawing]

[0015] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13]It is a sequence diagram showing the processing flow of the data processing system in Example 2 when combined with an emotion engine. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when combined with an emotion engine.

Embodiments for Carrying Out the Invention

[0016] Hereinafter, an example of an embodiment of the system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0017] First, the terms used in the following description will be explained.

[0018] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0019] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0020] In the following embodiments, the numbered storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, and the like.

[0021] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0022] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0023] [First Embodiment]

[0024] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0025] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0026] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0027] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0028] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0029] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0030] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0031] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0033] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0034] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0035] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0036] The learning support system of the present invention includes generating answers and explanations to questions, evaluating answers, generating problems, and a chat function. The following describes specific embodiments for implementing the present invention.

[0037] System Configuration

[0038] This system consists of user terminals and servers. Users use individual terminals (PCs, smartphones, tablets, etc.), and the servers are located in a cloud environment.

[0039] User question input and answer generation

[0040] 1. User input questions:

[0041] User: Enter any questions or doubts that arise during the learning process into the text box and click the submit button.

[0042] Terminal: Sends the entered question to the server as an HTTP request.

[0043] 2. Generating the solution:

[0044] Server: Inputs the received questions into an AI model (for example, a natural language processing model such as BERT or GPT).

[0045] Server: The AI ​​model analyzes the question and generates the answer and explanation.

[0046] Server: Sends the generated answer and explanation to the terminal as an HTTP response.

[0047] 3. Display to the user:

[0048] Terminal: Receives responses from the server and displays the answers and explanations on the screen.

[0049] Evaluation and feedback on the answers

[0050] 1. Enter your answer:

[0051] User: Enter your answer to the submitted question and click the submit button.

[0052] Terminal: Sends the entered answer to the server as an HTTP request.

[0053] 2. Rating:

[0054] Server: Inputs received answers into the AI ​​model and performs evaluation. Evaluation criteria include accuracy, logic, and clarity.

[0055] Server: Generates detailed feedback and explanations based on the evaluation results.

[0056] Server: Sends feedback to the terminal as an HTTP response.

[0057] 3. Display to the user:

[0058] Terminal: Receives feedback from the server and displays it on the screen.

[0059] Problem generation and learning cycle

[0060] 1. User trend analysis:

[0061] Server: Analyzes user learning trends based on evaluation results. This is done by analyzing past answer data using data mining techniques.

[0062] 2. Problem generation:

[0063] Server: Based on the analysis results, it identifies the user's weaknesses and uses an AI model to generate the next problem. The generated problem corresponds to the user's learning level and weaknesses.

[0064] Server: Sends a new problem to the terminal as an HTTP response.

[0065] 3. Display to the user:

[0066] Terminal: Receives new issues from the server and displays them on the screen.

[0067] 24-hour chat function

[0068] 1. User question input:

[0069] User: Type your question in the chat box and click the send button.

[0070] Terminal: Sends the entered questions to the server in real time.

[0071] 2. Real-time response:

[0072] Server: Inputs received questions into the AI ​​chatbot model. Generates appropriate responses.

[0073] Server: Sends the generated response to the terminal as an HTTP response in real time.

[0074] 3. Display to the user:

[0075] Terminal: Receives responses from the server and displays them in the chat box.

[0076] Specific example

[0077] For example, if a user asks, "Please explain the basics of differentiation," the server's AI model will generate the answer, "Differentiation is a method for measuring the rate of change of a function." Next, if the user delves deeper into the generated answer and asks, "What is the answer to this problem?", the AI ​​model will provide a specific answer to that problem. Finally, when the user submits their own answer, the system evaluates it and provides appropriate feedback, helping the user to learn efficiently.

[0078] The system of this invention dramatically improves the user's learning efficiency by acting as a 24-hour available private tutor.

[0079] The following describes the processing flow.

[0080] Explanation of questions and evaluation of answers

[0081] Step 1: Enter the user's question.

[0082] User: Enter any questions or doubts that arise during the learning process into the text box and click the submit button.

[0083] Terminal: Sends the entered text to the server as an HTTP request.

[0084] Step 2: Server-side solution generation

[0085] Server: Inputs the received questions into a natural language processing model (e.g., BERT or GPT model).

[0086] Server: The AI ​​model analyzes the questions and generates answers and explanations.

[0087] Step 3: Providing answers and explanations

[0088] Server: Sends the generated answer and explanation to the terminal as an HTTP response.

[0089] Terminal: Receives responses from the server and displays the answers and explanations on the screen.

[0090] Problem generation and learning cycle

[0091] Step 1: User input

[0092] User: Enter your answer to the provided question and click the submit button.

[0093] Terminal: Sends the entered answer to the server as an HTTP request.

[0094] Step 2: Evaluation of the answer

[0095] Server: Inputs received answers into the AI ​​model and performs evaluation. Evaluation criteria include accuracy, logic, and clarity.

[0096] Server: Generates detailed feedback and explanations based on the evaluation results.

[0097] Step 3: Provide feedback

[0098] Server: Sends feedback to the terminal as an HTTP response.

[0099] Terminal: Receives feedback from the server and displays it on the screen.

[0100] Step 4: Analyzing User Trends

[0101] Server: Analyzes user learning trends based on evaluation results. This uses data mining techniques and machine learning models.

[0102] Step 5: Generating a new problem

[0103] Server: Based on the analysis results, it identifies user weaknesses and generates new problems.

[0104] Server: Sends a new problem to the terminal as an HTTP response.

[0105] Terminal: Receives new issues from the server and displays them on the screen.

[0106] Chat function

[0107] Step 1: Enter the question

[0108] User: Type your question in the chat box and click the send button.

[0109] Terminal: Sends the entered questions to the server in real time.

[0110] Step 2: Generating a real-time response

[0111] Server: Inputs received questions into the AI ​​chatbot model.

[0112] Server: The AI ​​chatbot analyzes the question and generates a response in real time.

[0113] Step 3: Providing a response

[0114] Server: Sends the generated response to the terminal as an HTTP response in real time.

[0115] Terminal: Receives responses from the server and displays them in the chat box.

[0116] (Example 1)

[0117] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0118] Conventional learning support systems struggled to quickly generate appropriate answers and explanations to users' questions. Furthermore, a lack of evaluation and feedback on the generated answers, and the failure to provide problems tailored to users' learning tendencies, resulted in decreased learning efficiency. Additionally, there was a lack of systems that provided 24 / 7 responses to questions and concerns.

[0119] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0120] In this invention, the server includes means for inputting questions, means for receiving the input questions, means for generating answers and explanations for the received questions using a generative AI model, means for sending and displaying the generated answers and explanations as an HTTP response to the user's terminal, means for evaluating the generated answers, means for analyzing the user's tendencies based on the evaluation results using data mining technology, means for generating the next question using the generative AI model based on the user's tendencies, means for sending and displaying the generated question to the user's terminal as an HTTP response, and means for interacting with the user in real time in a chat format. As a result, the user can obtain quick and appropriate answers and explanations to their questions, receive evaluations and feedback on the answers, improve learning efficiency, and enable learning support that can answer questions anytime, 24 hours a day.

[0121] "Means for entering questions" refers to an interface that allows users to enter questions or doubts that arise during their learning process into a text box.

[0122] "Means for receiving submitted questions" refers to the network connection and protocol used by the server to receive questions sent from the user's terminal.

[0123] "Means for generating answers and explanations using a generative AI model" refers to a system that utilizes an AI model (for example, a natural language processing model) to create answers and explanations based on received questions.

[0124] "Means of sending and displaying as an HTTP response to the user's device" refers to a method of sending the generated answer and explanation as an HTTP response to the user's device and displaying its contents.

[0125] "Means for evaluating generated answers" refers to a system that uses AI models or algorithms to evaluate the accuracy, logic, and other aspects of the answers generated.

[0126] "Methods of analysis using data mining techniques" refers to data mining methods used to analyze users' past answer data and identify learning trends.

[0127] "A means of generating the next problem using an AI model based on user tendencies" refers to a system that uses an AI model to generate new problems tailored to the user's learning tendencies and weaknesses.

[0128] "Means of sending and displaying as an HTTP response to the user's device" refers to a method of sending the generated problem as an HTTP response to the user's device and displaying its contents.

[0129] "A means of interacting with users in real time via chat" refers to a function where users can input questions in real time using a chat box, and the AI ​​responds immediately.

[0130] The learning support system of the present invention includes functions such as generating answers and explanations to user questions, evaluating answers, generating problems, and a chat function. This system consists of the user's terminal (PC, smartphone, tablet, etc.) and a server located in a cloud environment. The following describes specific embodiments for implementing the present invention.

[0131] 1. System Configuration

[0132] The system operates by users interacting with the server using individual terminals. The server is located in a cloud environment and performs data analysis and generation using AI models (e.g., GPT-4®).

[0133] 2. User input of questions and generation of answers

[0134] The user enters their questions into a text box and clicks the submit button. The device sends the entered question data to the server as an HTTP request. The server analyzes the received question and generates an answer and explanation using a generative AI model. The generated answer and explanation are then sent to the device as an HTTP response and displayed on the user's screen.

[0135] For example, if a user inputs "Please explain the basics of differentiation," the server's AI model will generate the answer "Differentiation is a method for measuring the rate of change of a function" and send it to the terminal.

[0136] 3. Evaluation and feedback on the answers

[0137] When a user enters an answer to a presented problem and clicks the submit button, the device sends the entered answer data to the server as an HTTP request. The server inputs the received answer into an AI model and evaluates it based on criteria such as accuracy, logic, and clarity. Based on the evaluation results, it generates detailed feedback, sends it to the device as an HTTP response, and displays it on the user's screen.

[0138] For example, when a user submits an answer to a math problem, the server evaluates the answer and sends back feedback such as, "It's correct, but the explanation is insufficient."

[0139] 4. Problem generation and learning cycle

[0140] The server analyzes the user's past answer data using data mining techniques to identify the user's learning tendencies. Based on this analysis, it uses a generative AI model to generate the next problem that addresses the user's weaknesses and sends it to the terminal as an HTTP response. The terminal then displays the new problem to the user.

[0141] For example, if the server determines that a user does not understand "differentiation of a function," it will generate and send a problem such as "Find the derivative of the following function f(x)."

[0142] 5. 24-hour chat function

[0143] The user types a question into the chat box and clicks the send button. The device sends the entered question to the server in real time. The server inputs the received question into an AI chatbot model and generates an appropriate response. The generated response is sent to the device in real time as an HTTP response and displayed in the chat box.

[0144] For example, if a user asks in the chat, "Please explain how to calculate a definite integral," the server's AI chatbot will generate a response such as, "A definite integral is a method for finding the area of ​​a function within an interval," and immediately provide it to the user.

[0145] Example of a prompt

[0146] "Could you explain the basics of differential calculus?"

[0147] "What is differentiation?"

[0148] "Please solve these differential calculus problems."

[0149] "Please evaluate my answer."

[0150] The system of this invention acts as a 24-hour available tutor, dramatically improving the user's learning efficiency.

[0151] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0152] Step 1:

[0153] User's question input

[0154] Users enter any questions or doubts they have during their learning process into a text box and click the submit button.

[0155] Input: Text information entered by the user.

[0156] Specific operation: The browser on the user's device sends the entered question to the server via an asynchronous request (such as Ajax).

[0157] The terminal sends the entered question data to the server as an HTTP request.

[0158] Output: Query data sent to the server as an HTTP request.

[0159] Step 2:

[0160] Receiving questions on the server

[0161] The server receives HTTP requests sent from the terminal.

[0162] Input: An HTTP request containing question data sent from the terminal.

[0163] Specific operation: The server parses the content of the HTTP request and extracts the query data.

[0164] The server analyzes the received question data for processing.

[0165] Output: Analysis results of the questionnaire data.

[0166] Step 3:

[0167] Generating answers and explanations

[0168] The server inputs the analyzed question data into a generating AI model (e.g., GPT-4).

[0169] Input: Analyzed question data.

[0170] Specific operation: The server calls the API of the generation AI model and feeds the question data into the model. The model performs natural language processing and generates appropriate answers and explanations.

[0171] The server receives the output from the generated AI model.

[0172] Output: Generated solution and explanation.

[0173] Step 4:

[0174] Submit your answer and explanation.

[0175] The server sends the generated answer and explanation to the user's terminal as an HTTP response.

[0176] Input: Generated answer and explanation.

[0177] Specific operation: The server formats the answer and explanation into an HTTP response and sends it to the user's terminal.

[0178] The device displays the received answers and explanations on its screen.

[0179] Output: The answer and explanation displayed on the user's device.

[0180] Step 5:

[0181] User's answer input

[0182] The user enters their answer to the presented question and clicks the submit button.

[0183] Input: User-submitted answer data.

[0184] Specific operation: The device's browser sends the answer data to the server as an asynchronous request.

[0185] The terminal sends the entered answer data to the server as an HTTP request.

[0186] Output: HTTP request containing the answer data.

[0187] Step 6:

[0188] Evaluation of the answers

[0189] The server inputs the received answer data into the AI ​​model and performs evaluation.

[0190] Input: Received answer data.

[0191] Specific operation: The server inputs the answer data into an AI model for evaluation and performs evaluation based on criteria such as accuracy and logic. It then generates the evaluation results.

[0192] The server generates feedback based on the evaluation results.

[0193] Output: Evaluation results and generated feedback.

[0194] Step 7:

[0195] Send feedback

[0196] The server sends the feedback to the user's terminal as an HTTP response.

[0197] Input: Evaluation results and feedback.

[0198] Specific operation: The server formats the feedback into an HTTP response and sends it to the user's terminal.

[0199] The device displays the received feedback to the user.

[0200] Output: Feedback displayed on the user's device.

[0201] Step 8:

[0202] User learning trend analysis

[0203] The server analyzes past answer data using data mining techniques.

[0204] Input: User's past answer data.

[0205] Specific operation: The server extracts the user's past answer data from the cloud database and runs data mining algorithms to identify the user's learning tendencies and weaknesses.

[0206] The server saves the analysis results.

[0207] Output: User learning trend data.

[0208] Step 9:

[0209] Generating the next problem

[0210] The server generates the next problem using an AI model based on the user's learning tendencies.

[0211] Input: User learning trend data.

[0212] Specific operation: The server feeds learning trend data into the generating AI model and generates problems tailored to the user's level and weaknesses.

[0213] The server sends the generated problem to the user's terminal as an HTTP response.

[0214] Output: HTTP response containing the generated problem.

[0215] Step 10:

[0216] Display new problems

[0217] The device displays any new issues received to the user.

[0218] Input: HTTP response containing the generated problem.

[0219] Specific operation: The device's browser parses the HTTP response and displays the new problem on the screen.

[0220] Output: New issues displayed on the user's device.

[0221] Step 11:

[0222] Real-time chat function

[0223] The user types their question into the chat box and clicks the send button.

[0224] Input: The question entered in the chat box.

[0225] Specific operation: The device's browser sends chat messages to the server in real time (using WebSocket, etc.).

[0226] The server inputs the received question into the AI ​​chatbot model and generates an appropriate response.

[0227] Output: The generated response.

[0228] Specific operation: The server calls the AI ​​chatbot model's API in real time, analyzes the question, and generates a response.

[0229] The server sends the generated response to the terminal in real time as an HTTP response.

[0230] The device displays the received response in the chat box.

[0231] Output: The response displayed in the chat box.

[0232] (Application Example 1)

[0233] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0234] Modern food delivery services have limited capabilities to respond quickly and accurately to user questions and feedback, making it difficult for users to have a satisfying experience. Furthermore, there is a lack of information regarding ingredients and dishes, as well as appropriate recipe suggestions, forcing users to do their own research. Additionally, recipe suggestions and improvements based on user preferences and habits are not efficiently implemented. As a result, this leads to decreased user satisfaction and lower repeat customer rates.

[0235] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0236] In this invention, the server includes means for the user to input questions about ingredients or dishes, means for receiving the input questions, means for generating answers and explanations for the received questions, means for providing the generated answers and explanations to the user, means for the user to input feedback on delivered products, means for evaluating the input feedback, means for analyzing user trends based on the evaluation results, means for generating the next order or recipe based on the user trends, means for providing the generated recipe to the user, and means for interacting with the user in a chat format. This enables quick and accurate responses to user questions and feedback, and allows for the suggestion and improvement of recipes based on trends.

[0237] A "user" is an individual or legal entity that uses this system.

[0238] "Ingredients" refer to edible foods such as fresh produce and processed foods used for cooking or consumption.

[0239] "Cooking" refers to food that has been prepared by cooking ingredients in an appropriate manner and is ready for consumption.

[0240] "Questions" refer to unresolved questions or points of confusion that users have regarding ingredients or cooking.

[0241] "Feedback" refers to the evaluations and opinions that users provide regarding the delivered goods or services provided.

[0242] "Answer" refers to information that provides answers or explanations to users' questions.

[0243] A "recipe" is a set of instructions provided to the user that lists the ingredients and how to prepare a dish.

[0244] A "server" is a computer system that processes user questions and feedback and generates the necessary answers and recipes.

[0245] A "system" is a collection of means and devices for responding to user questions and feedback, and for performing trend analysis and suggesting recipes.

[0246] "Explanation" refers to a detailed explanation provided to address a problem or question in a way that makes it easy for the user to understand.

[0247] "Evaluation" involves analyzing feedback and making judgments based on appropriate criteria.

[0248] A "tendency" is a specific pattern or characteristic based on user behavior and preferences.

[0249] "Chat format" refers to a method of communication between the user and the system using text or voice.

[0250] The following describes embodiments for specifically implementing the present invention. The following is an example of a food delivery support application.

[0251] System Configuration

[0252] This system consists of user terminals and servers. Users use individual terminals (smartphones, tablets, etc.), and the servers are located in a cloud environment.

[0253] User question input and answer generation

[0254] 1. User input of questions: Users enter their questions or concerns about ingredients and cooking into the app's text box and click the submit button.

[0255] 2. Server-side answer generation: The server inputs the received question into a generating AI model (e.g., a natural language processing model such as GPT-2). The model analyzes the question and generates an answer and explanation. This answer is then provided to the user in real time.

[0256] User feedback and ratings

[0257] 1. Feedback Input: Users enter their feedback about the delivered product into the app's text box and click the submit button.

[0258] 2. Server Evaluation: The server analyzes and evaluates the received feedback. Criteria used for this evaluation include accuracy, appropriateness, and clarity. Based on the evaluation results, detailed feedback and explanations are generated.

[0259] 3. Providing evaluations: Provide users with generated explanations in real time.

[0260] Problem generation and learning cycle

[0261] 1. User Trend Analysis: The server analyzes user trends based on the evaluation results. This includes analyzing past response data.

[0262] 2. Suggestions for the next order and recipe: Based on the analysis results, the server generates and provides orders and recipes that match the user's preferences and tendencies.

[0263] 24-hour chat function

[0264] 1. User Question Input: The user enters their question in the chat box and clicks the send button. The entered question is sent to the server in real time.

[0265] 2. Real-time response: The server inputs the received question into the AI ​​chatbot model and generates an appropriate response. The generated response is provided to the user in real time.

[0266] Hardware and software

[0267] Hardware: Cloud servers, smartphones, tablets

[0268] Software: Flask (web framework), Transformers library, GPT-2 model

[0269] Specific example

[0270] For example, if a user asks, "What's the perfect recipe for a special family dinner on Wednesday?", the server's AI model will generate an answer such as, "Parmesan chicken with garlic mashed potatoes is perfect for a family dinner on Wednesday. Add some leafy greens and lemon dressing, and you've got the perfect course."

[0271] In this way, users can easily resolve their questions and provide feedback on the quality and taste of the delivered products, which can then be used to improve their next order.

[0272] Example of a prompt:

[0273] "What are some great recipes for a special dinner to bring the family together on Wednesday?"

[0274] The system works by having the AI ​​model generate appropriate answers based on user input and providing them to the user via the server.

[0275] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0276] Step 1:

[0277] The user launches the smartphone app, enters their questions about ingredients or cooking into the text box, and clicks the submit button.

[0278] Input: User's question (e.g., "What side dishes would go well with this dish?")

[0279] Output: User input data (HTTP request format)

[0280] Step 2:

[0281] The terminal receives the input user question and sends it to the server as an HTTP request.

[0282] Input: User input data

[0283] Output: HTTP request to the server

[0284] Step 3:

[0285] The server inputs the received user question into the generative AI model, analyzes the data, and generates an answer and an explanation.

[0286] Input: User question included in the HTTP request

[0287] Output: Generated answer and explanation (text data)

[0288] Step 4:

[0289] The server sends the generated answer and explanation to the terminal as an HTTP response.

[0290] Input: Generated answer and explanation

[0291] Output: HTTP response (including the answer and explanation)

[0292] Step 5:

[0293] The terminal receives the HTTP response from the server and displays the answer and explanation on the user interface.

[0294] Input: HTTP response

[0295] Output: Answer and explanation displayed on the user screen

[0296] Step 6:

[0297] The user inputs feedback on the delivered product into the text box and clicks the send button.

[0298] Input: User feedback (e.g., "The taste of this dish was very good, but it was a bit cold.")

[0299] Output: User input data (in the form of an HTTP request)

[0300] Step 7:

[0301] The terminal receives the input user feedback and sends it to the server as an HTTP request.

[0302] Input: User input data

[0303] Output: HTTP request to the server

[0304] Step 8:

[0305] The server analyzes the received feedback, conducts an evaluation, and generates detailed feedback and explanations.

[0306] Input: User feedback

[0307] Output: Evaluation results and explanations (text data)

[0308] Step 9:

[0309] The server sends the generated evaluation results and explanations to the terminal as an HTTP response.

[0310] Input: Evaluation results and explanations

[0311] Output: HTTP response (including evaluation results and explanations)

[0312] Step 10:

[0313] The terminal receives an HTTP response from the server and displays the evaluation results and explanations on the user interface.

[0314] Input: HTTP response

[0315] Output: Evaluation results and explanations displayed on the user screen

[0316] Step 11:

[0317] The server analyzes user trends based on evaluation results and generates subsequent orders and recipes.

[0318] Input: Evaluation result data

[0319] Output: New order suggestions and recipes (text data)

[0320] Step 12:

[0321] The server sends the generated new order suggestions and recipes to the terminal as an HTTP response.

[0322] Input: New order suggestions or recipes

[0323] Output: HTTP response (including new order suggestions and recipes)

[0324] Step 13:

[0325] The terminal receives an HTTP response from the server and displays new order suggestions and recipes on the user interface.

[0326] Input: HTTP response

[0327] Output: New order suggestions and recipes displayed on the user screen

[0328] The above outlines the specific processing steps. In each step, the user, terminal, and server work together to process the data, creating a system that enables the resolution of user questions and the utilization of user feedback.

[0329] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0330] The learning support system of the present invention features the generation of answers and explanations to questions, evaluation of answers, generation of questions, a chat function, and an emotion engine that recognizes the user's emotions. The following describes specific embodiments for implementing the present invention.

[0331] System Configuration

[0332] This system consists of a user's device, a server, and an emotion engine. Users use individual devices (PCs, smartphones, tablets, etc.), and the server is located in a cloud environment. The emotion engine is equipped with AI technology that analyzes emotions from the user's voice, facial expressions, text, etc.

[0333] User question input and answer generation

[0334] 1. User input questions:

[0335] User: Enter any questions or doubts that arise during the learning process into the text box and click the submit button.

[0336] Terminal: Sends the entered text to the server as an HTTP request.

[0337] 2. Generating the solution:

[0338] Server: Inputs the received questions into a natural language processing model (e.g., BERT or GPT model).

[0339] Server: The AI ​​model analyzes the questions and generates answers and explanations.

[0340] 3. Emotion recognition:

[0341] Device: Uses an emotion engine to recognize the user's emotional state. This includes facial expression analysis and voice analysis.

[0342] 4. Adjustments based on answers and emotions:

[0343] Server: Adjusts the content and tone of the answers and explanations based on the recognized emotional state.

[0344] Server: Sends the generated answer and explanation to the terminal as an HTTP response.

[0345] 5. Display to the user:

[0346] Terminal: Receives responses from the server and displays the answers and explanations on the screen.

[0347] Evaluation and feedback on the answers

[0348] 1. Enter your answer:

[0349] User: Enter your answer to the provided question and click the submit button.

[0350] Terminal: Sends the entered answer to the server as an HTTP request.

[0351] 2. Rating:

[0352] Server: Inputs received answers into the AI ​​model and performs evaluation. Evaluation criteria include accuracy, logic, and clarity.

[0353] Server: Generates detailed feedback and explanations based on the evaluation results.

[0354] 3. Emotion recognition:

[0355] Device: Uses an emotion engine to recognize the user's emotional state.

[0356] 4. Feedback and emotional adjustments:

[0357] Server: Adjusts the content and tone of feedback based on the recognized emotional state.

[0358] Server: Sends feedback to the terminal as an HTTP response.

[0359] 5. Display to the user:

[0360] Terminal: Receives feedback from the server and displays it on the screen.

[0361] Problem generation and learning cycle

[0362] 1. User trend analysis:

[0363] Server: Analyzes user learning trends based on evaluation results. This uses data mining techniques and machine learning models.

[0364] 2. Generating new problems:

[0365] Server: Based on the analysis results, it identifies user weaknesses and generates new problems.

[0366] Server: Sends a new problem to the terminal as an HTTP response.

[0367] 3. Emotion recognition:

[0368] Terminal: Uses an emotion engine to recognize the user's emotional state. This allows for adjustments to the difficulty and content of the generated problems.

[0369] 4. Display to the user:

[0370] Terminal: Receives new issues from the server and displays them on the screen.

[0371] Chat function

[0372] 1. Enter your question:

[0373] User: Type your question in the chat box and click the send button.

[0374] Terminal: Sends the entered questions to the server in real time.

[0375] 2. Real-time response generation:

[0376] Server: Inputs received questions into the AI ​​chatbot model.

[0377] Server: The AI ​​chatbot analyzes the question and generates a response in real time.

[0378] 3. Emotion recognition:

[0379] Device: Uses an emotion engine to recognize the user's emotional state, thereby adjusting the response accordingly.

[0380] 4. Providing a response:

[0381] Server: Sends the generated response to the terminal as an HTTP response in real time.

[0382] Terminal: Receives responses from the server and displays them in the chat box.

[0383] Specific example

[0384] For example, if a user asks, "Could you explain the basics of differentiation?", the server's AI model will generate the answer, "Differentiation is a method for measuring the rate of change of a function." Furthermore, if the emotion engine detects that the user is feeling fatigued, the answer will be adjusted to something like, "Differentiation is a method for measuring the rate of change of a function, and it may seem difficult, but let's work through it together."

[0385] The system of this invention optimizes learning by taking into account the user's emotional state, thereby providing more effective learning support. This dramatically improves the user's motivation and learning efficiency.

[0386] The following describes the processing flow.

[0387] User question input and answer generation

[0388] Step 1:

[0389] User: Enter any questions or doubts that arise during the learning process into the text box and click the submit button.

[0390] Step 2:

[0391] Terminal: Sends the entered text to the server as an HTTP request.

[0392] Step 3:

[0393] Server: Inputs the received questions into a natural language processing model (e.g., BERT or GPT model).

[0394] Step 4:

[0395] Server: The AI ​​model analyzes the questions and generates answers and explanations.

[0396] Step 5:

[0397] Device: Captures the user's facial expressions and voice and sends them to the emotion engine.

[0398] Step 6:

[0399] Emotion Engine: Analyzes the user's emotional state based on captured data.

[0400] Step 7:

[0401] Server: Adjusts the content and tone of the answers and explanations based on the recognized emotional state.

[0402] Step 8:

[0403] Server: Sends the adjusted answer and explanation to the terminal as an HTTP response.

[0404] Step 9:

[0405] Terminal: Receives responses from the server and displays the answers and explanations on the screen.

[0406] Evaluation and feedback on the answers

[0407] Step 1:

[0408] User: Enter your answer to the provided question and click the submit button.

[0409] Step 2:

[0410] Terminal: Sends the entered answer to the server as an HTTP request.

[0411] Step 3:

[0412] Server: Inputs received answers into the AI ​​model and performs evaluation.

[0413] Step 4:

[0414] Server: Generates detailed feedback and explanations based on the evaluation results.

[0415] Step 5:

[0416] Device: Captures the user's facial expressions and voice and sends them to the emotion engine.

[0417] Step 6:

[0418] Emotion Engine: Analyzes the user's emotional state based on captured data.

[0419] Step 7:

[0420] Server: Adjusts the content and tone of feedback based on the recognized emotional state.

[0421] Step 8:

[0422] Server: Sends the adjusted feedback to the terminal as an HTTP response.

[0423] Step 9:

[0424] Terminal: Receives feedback from the server and displays it on the screen.

[0425] Problem generation and learning cycle

[0426] Step 1:

[0427] Server: Analyzes user learning trends based on evaluation results.

[0428] Step 2:

[0429] Server: Based on the analysis results, it identifies user weaknesses and generates new problems.

[0430] Step 3:

[0431] Device: Captures the user's facial expressions and voice and sends them to the emotion engine.

[0432] Step 4:

[0433] Emotion Engine: Analyzes the user's emotional state based on captured data.

[0434] Step 5:

[0435] Server: Adjusts the difficulty and content of generated problems based on the recognized emotional state.

[0436] Step 6:

[0437] Server: Sends a new problem to the terminal as an HTTP response.

[0438] Step 7:

[0439] Terminal: Receives new issues from the server and displays them on the screen.

[0440] Chat function

[0441] Step 1:

[0442] User: Type your question in the chat box and click the send button.

[0443] Step 2:

[0444] Terminal: Sends the entered questions to the server in real time.

[0445] Step 3:

[0446] Server: Inputs received questions into the AI ​​chatbot model.

[0447] Step 4:

[0448] Server: The AI ​​chatbot analyzes the question and generates a response in real time.

[0449] Step 5:

[0450] Device: Captures the user's facial expressions and voice and sends them to the emotion engine.

[0451] Step 6:

[0452] Emotion Engine: Analyzes the user's emotional state based on captured data.

[0453] Step 7:

[0454] Server: Adjusts the response based on the recognized emotional state.

[0455] Step 8:

[0456] Server: Sends the adjusted response to the terminal as an HTTP response in real time.

[0457] Step 9:

[0458] Terminal: Receives responses from the server and displays them in the chat box.

[0459] (Example 2)

[0460] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0461] Traditional learning support systems can provide answers and explanations to users' questions, but they lack the ability to consider the user's emotional state. Therefore, they cannot adequately recognize the fatigue or frustration a user might experience during learning and adjust the content of responses and feedback accordingly, resulting in a failure to maximize learning effectiveness. Furthermore, the lack of real-time emotion recognition and feedback based on those results made it difficult to maintain user motivation.

[0462] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0463] In this invention, the server includes means for the user to input questions, means for receiving the input questions, means for generating answers and explanations for the received questions using a generative AI model, means for providing the generated answers and explanations to the user, means for evaluating the generated answers, means for analyzing the user's tendencies based on the evaluation results, means for generating the next question based on the user's tendencies, means for providing the generated question to the user, means for interacting with the user in a chat format, means for recognizing the user's emotional state, and means for adjusting the content of answers and feedback based on the emotional state. This makes it possible to recognize the user's emotional state in real time and adjust the response content and feedback according to that state, thereby increasing the user's motivation to learn and enabling more effective learning support.

[0464] A "user" is an individual or group who uses a learning support system to seek answers to their questions or receive feedback.

[0465] "Questions" refer to the questions or things that users need to understand during the learning process.

[0466] "Input means" refers to interface means for users to input questions and answers into the system, and includes keyboards and touch panels.

[0467] "Receiving means" refers to the means for receiving data transmitted by a user. This includes data receiving systems via a network.

[0468] A "generative AI model" is an artificial intelligence model that uses natural language processing to analyze user input and generate appropriate answers and explanations.

[0469] "Answer" refers to the content of the response to the user's question.

[0470] "Explanation" refers to a detailed explanation provided alongside the answer, serving as supplementary information to help users gain a deeper understanding of their questions.

[0471] "Means of provision" refers to means of displaying or communicating the generated answers and explanations to the user.

[0472] "Evaluation means" refers to methods for evaluating the accuracy, logic, clarity, etc., of answers entered by users, and includes AI models.

[0473] "Evaluation results" refer to data calculated using evaluation methods for user responses.

[0474] A "trend analysis method" is a means of analyzing a user's learning tendencies based on evaluation results.

[0475] The "problem generation method" is a means of generating new learning problems based on the user's learning tendencies and weaknesses, and it uses a generation AI model.

[0476] "Chat dialogue methods" refer to means by which users and systems can interact in real time, and include chatbots that use natural language processing.

[0477] "Emotion recognition means" refers to methods for recognizing a user's emotional state by analyzing their voice, facial expressions, text, etc.

[0478] "Adjustment means" are means for adjusting the content and tone of responses and feedback based on the emotional state obtained by the emotion recognition means.

[0479] The learning support system of the present invention combines the generation of answers and explanations to user questions, evaluation of answers, generation of problems, a chat function, and an emotion recognition function. The following describes in detail the embodiments for specifically implementing the present invention.

[0480] System Configuration

[0481] This system consists of a user's device, a server, and an emotion engine. Users utilize devices such as PCs, smartphones, and tablets. The server is located in a cloud environment and performs the necessary computational processing. The emotion engine incorporates AI technology that analyzes emotions from the user's voice, facial expressions, and text. Specifically, the device collects the user's facial expressions and voice through its camera and microphone and sends them to the server for analysis.

[0482] Hardware and software usage

[0483] The server implements generative AI models (such as natural language processing models like BERT and GPT) to analyze questions and answers submitted by users. It also uses data mining techniques and machine learning models to analyze users' learning tendencies. The terminal is equipped with an emotion recognition engine that analyzes the user's voice and facial expressions in real time to recognize their emotional state. Emotion recognition utilizes facial recognition and voice analysis technologies.

[0484] User question input and answer generation

[0485] The user enters any questions or doubts that arise during the learning process into a text box and clicks the submit button. For example, they might enter, "Please explain the basics of differentiation." The device sends the entered text to the server as an HTTP request. The server inputs the received question into a generating AI model and generates an answer and explanation. An example of a generated answer is, "Differentiation is a method for measuring the rate of change of a function."

[0486] Emotion recognition and regulation

[0487] The emotion engine built into the device analyzes the user's facial expressions, voice, and text to convey emotions. For example, it might recognize "fatigue" from the user's facial expression. Based on the recognized emotional state, the server adjusts the content and tone of the answers and explanations. For instance, it might add an additional message such as, "This might seem difficult, but let's work hard together."

[0488] Evaluation and feedback on the answers

[0489] The user enters an answer to a given question and clicks the submit button. The device sends the entered answer to the server as an HTTP request. The server inputs the received answer into an AI model and evaluates it based on accuracy, logic, and clarity. Based on the evaluation results, it generates detailed feedback and explanations. For example, if the user answers the question "What are the fundamental principles of electromagnetism?" with "The motion of electrons and their field interactions," the server provides feedback such as "This answer is superficially correct, but it would be good to explain it in more detail."

[0490] Problem generation and learning cycle

[0491] The server analyzes the user's learning tendencies based on the evaluation results and identifies the user's weaknesses using data mining techniques and machine learning models. It generates new problems related to the identified weaknesses and sends them to the terminal as an HTTP response. The terminal displays the new problems on its screen, allowing the user to continue learning.

[0492] Chat function

[0493] When a user types a question into the chat box and clicks the send button, the device sends the entered question to the server in real time. The server inputs the received question into an AI chatbot model and generates a response in real time. If emotion recognition determines that the user's emotional state is "frustrated," the response is adjusted to something like, "This may be difficult to understand, but I will explain it carefully." The server then sends the generated response to the device as an HTTP response in real time, and the device displays it in the chat box.

[0494] As described above, the present invention provides a more effective learning experience by providing real-time responses to user questions and answers, adjusting feedback content based on emotion recognition, and generating new problems by analyzing the user's learning tendencies.

[0495] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0496] Step 1:

[0497] The user enters their question into the text box and clicks the submit button.

[0498] Input: The user enters "Please explain the basics of differentiation."

[0499] Operation: The terminal retrieves the entered text.

[0500] Output: The entered question is displayed on the terminal.

[0501] Step 2:

[0502] The terminal sends the entered question to the server as an HTTP request.

[0503] Input: The question entered in the text box.

[0504] Operation: The terminal converts the question into packets and sends them to the server via the network.

[0505] Output: The server receives the question.

[0506] Step 3:

[0507] The server inputs the received questions into a natural language processing model.

[0508] Input: Text data of the question sent from the device.

[0509] Operation: The server processes the question into an appropriate format and inputs it as a prompt to a generating AI model (e.g., GPT-3®).

[0510] Output: The generative AI model generates the answer and explanation.

[0511] Step 4:

[0512] The server retrieves the answer and explanation as output from the generated AI model and generates a response to provide to the user.

[0513] Input: Text data of answers and explanations from a generative AI model.

[0514] Operation: Converts the answers and explanations obtained by the server into an HTTP response format.

[0515] Output: Response data is generated.

[0516] Step 5:

[0517] The device uses an emotion engine to recognize the user's emotional state.

[0518] Input: User's facial image and voice data acquired from the camera and microphone.

[0519] Operation: The device uses an emotion engine to analyze the user's facial expressions and voice to detect their emotional state.

[0520] Output: The user's emotional state (e.g., fatigue, irritation) is identified by the emotion engine.

[0521] Step 6:

[0522] The server adjusts the content and tone of the answers and explanations based on the recognized emotional state.

[0523] Input: Emotional state data sent from the device, along with the generated answer and explanation.

[0524] Operation: The server analyzes the emotional state and makes adjustments such as, "Differentiation is a method for measuring the rate of change of a function. It may seem difficult, but let's work through it together."

[0525] Output: Text data of the adjusted answers and explanations.

[0526] Step 7:

[0527] The server sends the adjusted answer to the terminal as an HTTP response.

[0528] Input: Text data of the adjusted answers and explanations.

[0529] Operation: The server converts the data into an HTTP response format and sends it to the terminal over the network.

[0530] Output: The terminal receives a response.

[0531] Step 8:

[0532] The terminal displays the response from the server on the screen.

[0533] Input: Adjusted answer and explanation data received from the server.

[0534] Operation: The device displays the answer and explanation on the screen.

[0535] Output: The user can view the answer and explanation on the screen.

[0536] Step 9:

[0537] The user enters their answer to the provided question and clicks the submit button.

[0538] Input: User's answer (e.g., "The motion of electrons and their field interactions").

[0539] Action: The device retrieves the answer and prepares to send it to the server as an HTTP request.

[0540] Output: The entered answer is ready to be sent by the terminal.

[0541] Step 10:

[0542] The device sends the answer to the server as an HTTP request.

[0543] Input: Text data of the answer entered by the user.

[0544] Operation: The terminal converts the answer into packets and sends them to the server over the network.

[0545] Output: The server receives the answer.

[0546] Step 11:

[0547] The server inputs the answer into the AI ​​model and performs evaluation.

[0548] Input: Text data of the answer sent from the device.

[0549] Operation: The server inputs the answer into the AI ​​model and performs analysis based on evaluation criteria.

[0550] Output: Evaluation results are generated.

[0551] Step 12:

[0552] The server generates detailed feedback and explanations based on the evaluation results.

[0553] Input: Evaluation result data generated by an AI model.

[0554] Operation: The server generates feedback and explanations for the user based on the evaluation results.

[0555] Output: Text data of feedback and commentary.

[0556] Step 13:

[0557] The device recognizes the user's emotional state.

[0558] Input: User's facial image and voice data acquired from the camera and microphone.

[0559] Operation: The device uses an emotion engine to analyze the user's facial expressions and voice.

[0560] Output: The user's emotional state (e.g., satisfied, dissatisfied) is identified.

[0561] Step 14:

[0562] The server adjusts the content and tone of the feedback based on the emotional state.

[0563] Input: Emotional state data sent from the device, along with generated feedback and commentary.

[0564] Operation: The server adjusts the tone of feedback and commentary based on the emotional state.

[0565] Output: Text data of adjusted feedback and commentary.

[0566] Step 15:

[0567] The server sends the adjusted feedback to the terminal as an HTTP response.

[0568] Input: Text data of adjusted feedback and commentary.

[0569] Operation: The server converts the data into an HTTP response format and sends it to the terminal over the network.

[0570] Output: The terminal receives a response.

[0571] Step 16:

[0572] The terminal displays feedback and explanations from the server on the screen.

[0573] Input: Feedback and commentary data received from the server.

[0574] Action: The device displays feedback and explanations on the screen.

[0575] Output: The user will be able to view feedback and explanations on the screen.

[0576] The detailed processing flow for each step described above allows users to input questions, receive answers, get feedback on those answers, and even have their feedback adjusted based on their emotions, thereby maximizing the effectiveness of the learning support system.

[0577] (Application Example 2)

[0578] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0579] Conventional learning support systems provide functions for answering user questions, evaluating their performance, and generating new problems. However, they fail to take into account the user's emotional state, resulting in insufficient improvement in learning effectiveness. As a result, users may not receive appropriate support when they feel fatigued or stressed, potentially leading to a decrease in their motivation to learn. Furthermore, because answers and feedback are provided regardless of the user's emotions, there is a need to address the issue of poor satisfaction and comprehension during the learning process.

[0580] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for inputting questions, means for receiving the input questions, means for generating answers and explanations for the received questions, means for providing the generated answers and explanations to the user, means for recognizing the user's emotions, means for adjusting the tone and content of the answers and explanations based on the recognized emotions, means for evaluating the generated answers, means for analyzing the user's tendencies based on the evaluation results, means for generating the next question based on the user's tendencies, means for adjusting the difficulty and content of the questions considering the user's emotional state before providing the questions, means for providing the questions to the user, means for interacting with the user in a chat format, and means for recognizing the user's emotional state during the chat and adjusting the response content based on the recognized emotional state. This enables appropriate learning support that takes into account the user's emotional state, and is expected to improve learning effectiveness and user satisfaction.

[0581] "Means for inputting questions" refers to an interface for users to input questions or doubts that arise during their learning process.

[0582] "Means for receiving entered questions" refers to a mechanism for the system to receive questions entered by the user.

[0583] "Means for generating answers and explanations to received questions" refers to a set of algorithms and programs for generating appropriate answers and additional explanations based on received questions.

[0584] "Means of providing users with generated answers and explanations" refers to a function for displaying generated answers and explanations to users.

[0585] "Means of recognizing user emotions" refers to technologies and devices that analyze a user's voice, facial expressions, text, etc., to recognize their emotional state.

[0586] "Means of adjusting the tone and content of answers and explanations based on recognized emotions" refers to algorithms and functions that adjust the wording and level of detail of answers and explanations according to the recognized emotional state of the user.

[0587] "Means for evaluating generated answers" refers to a mechanism for evaluating the accuracy and validity of generated answers.

[0588] "Means for analyzing user trends based on evaluation results" refers to algorithms for analyzing users' learning patterns and proficiency levels based on the evaluation results of their answers.

[0589] "Means for generating the next problem based on user trends" refers to a system that analyzes the user's learning trends and generates the next problem to be provided based on those trends.

[0590] "Means of adjusting the difficulty and content of generated problems based on the user's emotional state before providing them" refers to a function that appropriately adjusts the difficulty and content of generated problems based on the user's current emotional state.

[0591] "Means of providing generated problems to users" refers to a function for displaying adjusted problems to users.

[0592] "Means of interacting with users in a chat format" refers to chat interfaces and related technologies that allow users to interact with the system in real time.

[0593] "Means for recognizing a user's emotional state during a chat and adjusting the response based on that emotional state" refers to algorithms and technologies for recognizing a user's emotions during a chat and dynamically adjusting the content of the response accordingly.

[0594] The learning support system of the present invention, in addition to generating answers to user questions, providing evaluations, generating problems, and offering a chat function, improves learning effectiveness by recognizing the user's emotions and adjusting learning content and responses based on those emotions. This system consists of the user's terminal, a server, and an emotion engine.

[0595] System Configuration

[0596] A terminal is an information processing device (smartphone, tablet, or personal computer) that receives user input and communicates with a server. The terminal is equipped with a camera and microphone to collect data for emotion recognition.

[0597] Server: Located in a cloud environment, it runs various AI models (e.g., natural language processing models and emotion recognition models). Its main roles are generating answers and explanations, evaluation, problem generation, and adjustments based on the user's emotional state.

[0598] Emotion Engine: Analyzes the user's emotional state from their voice, facial expressions, text, etc., and adjusts responses and answers accordingly.

[0599] User question input and answer generation

[0600] 1. Users enter any questions or doubts that arise during the learning process into the text box on their device and submit them.

[0601] 2. The terminal sends the entered text to the server.

[0602] 3. The server inputs the received questions into a natural language processing model (e.g., BERT or GPT model) and generates answers and explanations.

[0603] 4. The device analyzes the user's emotions using their facial expressions and voice, and sends that data to the server.

[0604] 5. The server adjusts the content and tone of the answers and explanations based on the recognized emotional state and sends them to the terminal.

[0605] 6. The device displays the adjusted answers and explanations on the screen.

[0606] Evaluation and feedback on the answers

[0607] 1. The user enters and submits their answer to the provided question.

[0608] 2. The terminal sends the entered answer to the server.

[0609] 3. The server inputs the answer into the AI ​​model and performs an evaluation. Based on the evaluation results, it generates detailed feedback and explanations.

[0610] 4. The device uses an emotion engine to recognize the user's emotional state and sends that data to the server.

[0611] 5. The server adjusts the content and tone of the feedback based on the emotional state and sends it to the device.

[0612] 6. The device displays the adjusted feedback on the screen.

[0613] Problem generation and learning cycle

[0614] 1. Based on the evaluation results, the server analyzes the user's learning tendencies and generates the next problem using data mining techniques and machine learning models.

[0615] 2. The device recognizes the user's emotional state before providing a problem and sends that data to the server.

[0616] 3. The server adjusts the difficulty and content of the problem, taking into account the emotional state, and sends it to the terminal.

[0617] 4. The device displays the adjusted issues on the screen.

[0618] Chat function

[0619] 1. The user types their question into the chat box and sends it.

[0620] 2. The terminal sends the question to the server in real time.

[0621] 3. The server uses an AI chatbot model to analyze the question and generate a response in real time.

[0622] 4. The device uses an emotion engine to recognize the user's emotional state and sends that data to the server.

[0623] 5. The server adjusts the response based on the emotional state and sends it to the terminal.

[0624] 6. The device displays the adjusted response in the chat box.

[0625] Specific examples and prompt statements

[0626] For example, if a user asks, "Could you explain the basics of differentiation?", the server will generate the answer, "Differentiation is a method for measuring the rate of change of a function." If the emotion engine detects that the user is fatigued, the answer will be adjusted to, "Differentiation is a method for measuring the rate of change of a function, and it may seem difficult, but let's work through it together."

[0627] Specific examples of prompt statements are as follows:

[0628] "Please explain the basics of differentiation. The user seems a little tired. Please provide the answer in a gentle tone."

[0629] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0630] Step 1:

[0631] Users enter any questions or doubts that arise during the learning process into the text box on their device and submit them.

[0632] Input: Text entered by the user

[0633] Output: HTTP request from terminal to server

[0634] Specific operation: When the user enters a question and clicks the "Send" button, the terminal sends the entered text data to the server.

[0635] Step 2:

[0636] The terminal sends the entered text to the server.

[0637] Input: Text data of the question

[0638] Output: Text data sent to the server

[0639] Specific action: The device sends text data containing the user's question to the server as an HTTP request.

[0640] Step 3:

[0641] The server inputs the received questions into a natural language processing model (e.g., BERT or GPT model) and generates answers and explanations.

[0642] Input: Text data of the question

[0643] Output: Generated solution and explanation data

[0644] Specific operation: The server inputs the text data of the question into a natural language processing model, and uses a generative AI model to generate the answer and explanation.

[0645] Step 4:

[0646] The device analyzes the user's emotions using their facial expressions and voice, and sends that data to a server.

[0647] Input: User's facial expression data, voice data

[0648] Output: Emotional state data

[0649] Specific operation: The device uses facial expressions and voice data collected by the camera and microphone to allow the emotion engine to recognize the emotional state and send that data to the server.

[0650] Step 5:

[0651] The server adjusts the content and tone of the answers and explanations based on the recognized emotional state, and then sends them to the terminal.

[0652] Input: Emotional state data, generated answers and explanatory data

[0653] Output: Adjusted answer and explanation data

[0654] Specific operation: The server uses the emotion data recognized by the emotion engine to apply an algorithm that adjusts the tone and level of detail of the answers and explanations, generates the adjusted answer and explanation data, and sends it to the terminal.

[0655] Step 6:

[0656] The device displays the adjusted answers and explanations on the screen.

[0657] Input: Adjusted answer and explanation data

[0658] Output: Adjusted answers and explanations displayed on the screen

[0659] Specific operation: The device displays the adjusted answer and explanation data it has received on the screen in a user-friendly format.

[0660] Step 7:

[0661] Users enter and submit their answers to the provided questions.

[0662] Input: User-entered answer data

[0663] Output: Answer data sent as an HTTP request from the terminal to the server.

[0664] Specific operation: When the user enters the answer to the problem and clicks the "Submit" button, the device sends the answer data to the server.

[0665] Step 8:

[0666] The terminal sends the entered answer to the server.

[0667] Input: Answer data

[0668] Output: Answer data sent to the server

[0669] Specific operation: The device sends the user's answer data to the server as an HTTP request.

[0670] Step 9:

[0671] The server inputs the answer into the AI ​​model and performs an evaluation. Based on the evaluation results, it generates detailed feedback and explanations.

[0672] Input: Answer data

[0673] Output: Evaluation results, feedback, and explanatory data

[0674] Specific operation: The server inputs the answer data into the AI ​​model, applies an evaluation algorithm to generate evaluation results, and then generates feedback and explanatory data based on those results.

[0675] Step 10:

[0676] The device uses an emotion engine to recognize the user's emotional state and sends that data to the server.

[0677] Input: User's facial expression data, voice data

[0678] Output: Emotional state data

[0679] Specific operation: The device uses facial expressions and voice data collected by the camera and microphone to allow the emotion engine to recognize the emotional state and send that data to the server.

[0680] Step 11:

[0681] The server adjusts the content and tone of the feedback based on the emotional state and sends it to the device.

[0682] Input: Emotional state data, evaluation results, and feedback data

[0683] Output: Adjusted feedback data

[0684] Specific operation: The server uses the emotion data recognized by the emotion engine to apply an algorithm that adjusts the tone and level of detail of the feedback, generates the adjusted feedback data, and sends it to the terminal.

[0685] Step 12:

[0686] The device displays adjusted feedback on the screen.

[0687] Input: Adjusted feedback data

[0688] Output: Adjusted feedback displayed on the screen

[0689] Specific operation: The device displays the adjusted feedback data it has received on the screen in a user-friendly format.

[0690] Step 13:

[0691] Based on the evaluation results, the server analyzes the user's learning tendencies and generates the next problem using data mining techniques and machine learning models.

[0692] Input: Evaluation result data

[0693] Output: Newly generated problem data

[0694] Specific operation: The server uses the evaluation result data to analyze the user's learning tendencies using data mining techniques and machine learning models, and then generates the next problem to be presented.

[0695] Step 14:

[0696] The device recognizes the user's emotional state before providing a problem and sends that data to the server.

[0697] Input: User's facial expression data, voice data

[0698] Output: Emotional state data

[0699] Specific operation: The device uses facial expressions and voice data collected by the camera and microphone to allow the emotion engine to recognize the emotional state and send that data to the server.

[0700] Step 15:

[0701] The server adjusts the difficulty and content of the problem, taking into account the emotional state, and then sends it to the terminal.

[0702] Input: Emotional state data, new problem data

[0703] Output: Adjusted new problem data

[0704] Specific operation: The server uses the emotion data recognized by the emotion engine to apply an algorithm that adjusts the difficulty and content of the problem, generates new adjusted problem data, and sends it to the terminal.

[0705] Step 16:

[0706] The device displays the adjusted issues on the screen.

[0707] Input: Adjusted new problem data

[0708] Output: Adjusted issues displayed on the screen

[0709] Specific operation: The device displays the adjusted problem data it has received on the screen in a user-friendly format.

[0710] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0711] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0712] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0713] [Second Embodiment]

[0714] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0715] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0716] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0717] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0718] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0719] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0720] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0721] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0722] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0723] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0724] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0725] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0726] The learning support system of the present invention includes generating answers and explanations to questions, evaluating answers, generating problems, and a chat function. The following describes specific embodiments for implementing the present invention.

[0727] System Configuration

[0728] This system consists of user terminals and servers. Users use individual terminals (PCs, smartphones, tablets, etc.), and the servers are located in a cloud environment.

[0729] User question input and answer generation

[0730] 1. User input questions:

[0731] User: Enter any questions or doubts that arise during the learning process into the text box and click the submit button.

[0732] Terminal: Sends the entered question to the server as an HTTP request.

[0733] 2. Generating the solution:

[0734] Server: Inputs the received questions into an AI model (for example, a natural language processing model such as BERT or GPT).

[0735] Server: The AI ​​model analyzes the question and generates the answer and explanation.

[0736] Server: Sends the generated answer and explanation to the terminal as an HTTP response.

[0737] 3. Display to the user:

[0738] Terminal: Receives responses from the server and displays the answers and explanations on the screen.

[0739] Evaluation and feedback on the answers

[0740] 1. Enter your answer:

[0741] User: Enter your answer to the submitted question and click the submit button.

[0742] Terminal: Sends the entered answer to the server as an HTTP request.

[0743] 2. Rating:

[0744] Server: Inputs received answers into the AI ​​model and performs evaluation. Evaluation criteria include accuracy, logic, and clarity.

[0745] Server: Generates detailed feedback and explanations based on the evaluation results.

[0746] Server: Sends feedback to the terminal as an HTTP response.

[0747] 3. Display to the user:

[0748] Terminal: Receives feedback from the server and displays it on the screen.

[0749] Problem generation and learning cycle

[0750] 1. User trend analysis:

[0751] Server: Analyzes user learning trends based on evaluation results. This is done by analyzing past answer data using data mining techniques.

[0752] 2. Problem generation:

[0753] Server: Based on the analysis results, it identifies the user's weaknesses and uses an AI model to generate the next problem. The generated problem corresponds to the user's learning level and weaknesses.

[0754] Server: Sends a new problem to the terminal as an HTTP response.

[0755] 3. Display to the user:

[0756] Terminal: Receives new issues from the server and displays them on the screen.

[0757] 24-hour chat function

[0758] 1. User question input:

[0759] User: Type your question in the chat box and click the send button.

[0760] Terminal: Sends the entered questions to the server in real time.

[0761] 2. Real-time response:

[0762] Server: Inputs received questions into the AI ​​chatbot model. Generates appropriate responses.

[0763] Server: Sends the generated response to the terminal as an HTTP response in real time.

[0764] 3. Display to the user:

[0765] Terminal: Receives responses from the server and displays them in the chat box.

[0766] Specific example

[0767] For example, if a user asks, "Please explain the basics of differentiation," the server's AI model will generate the answer, "Differentiation is a method for measuring the rate of change of a function." Next, if the user delves deeper into the generated answer and asks, "What is the answer to this problem?", the AI ​​model will provide a specific answer to that problem. Finally, when the user submits their own answer, the system evaluates it and provides appropriate feedback, helping the user to learn efficiently.

[0768] The system of this invention dramatically improves the user's learning efficiency by acting as a 24-hour available private tutor.

[0769] The following describes the processing flow.

[0770] Explanation of questions and evaluation of answers

[0771] Step 1: Enter the user's question.

[0772] User: Enter any questions or doubts that arise during the learning process into the text box and click the submit button.

[0773] Terminal: Sends the entered text to the server as an HTTP request.

[0774] Step 2: Server-side solution generation

[0775] Server: Inputs the received questions into a natural language processing model (e.g., BERT or GPT model).

[0776] Server: The AI ​​model analyzes the questions and generates answers and explanations.

[0777] Step 3: Providing answers and explanations

[0778] Server: Sends the generated answer and explanation to the terminal as an HTTP response.

[0779] Terminal: Receives responses from the server and displays the answers and explanations on the screen.

[0780] Problem generation and learning cycle

[0781] Step 1: User input

[0782] User: Enter your answer to the provided question and click the submit button.

[0783] Terminal: Sends the entered answer to the server as an HTTP request.

[0784] Step 2: Evaluation of the answer

[0785] Server: Inputs received answers into the AI ​​model and performs evaluation. Evaluation criteria include accuracy, logic, and clarity.

[0786] Server: Generates detailed feedback and explanations based on the evaluation results.

[0787] Step 3: Provide feedback

[0788] Server: Sends feedback to the terminal as an HTTP response.

[0789] Terminal: Receives feedback from the server and displays it on the screen.

[0790] Step 4: Analyzing User Trends

[0791] Server: Analyzes user learning trends based on evaluation results. This uses data mining techniques and machine learning models.

[0792] Step 5: Generating a new problem

[0793] Server: Based on the analysis results, it identifies user weaknesses and generates new problems.

[0794] Server: Sends a new problem to the terminal as an HTTP response.

[0795] Terminal: Receives new issues from the server and displays them on the screen.

[0796] Chat function

[0797] Step 1: Enter the question

[0798] User: Type your question in the chat box and click the send button.

[0799] Terminal: Sends the entered questions to the server in real time.

[0800] Step 2: Generating a real-time response

[0801] Server: Inputs received questions into the AI ​​chatbot model.

[0802] Server: The AI ​​chatbot analyzes the question and generates a response in real time.

[0803] Step 3: Providing a response

[0804] Server: Sends the generated response to the terminal as an HTTP response in real time.

[0805] Terminal: Receives responses from the server and displays them in the chat box.

[0806] (Example 1)

[0807] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0808] Conventional learning support systems struggled to quickly generate appropriate answers and explanations to users' questions. Furthermore, a lack of evaluation and feedback on the generated answers, and the failure to provide problems tailored to users' learning tendencies, resulted in decreased learning efficiency. Additionally, there was a lack of systems that provided 24 / 7 responses to questions and concerns.

[0809] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0810] In this invention, the server includes means for inputting questions, means for receiving the input questions, means for generating answers and explanations for the received questions using a generative AI model, means for sending and displaying the generated answers and explanations as an HTTP response to the user's terminal, means for evaluating the generated answers, means for analyzing the user's tendencies based on the evaluation results using data mining technology, means for generating the next question using the generative AI model based on the user's tendencies, means for sending and displaying the generated question to the user's terminal as an HTTP response, and means for interacting with the user in real time in a chat format. As a result, the user can obtain quick and appropriate answers and explanations to their questions, receive evaluations and feedback on the answers, improve learning efficiency, and enable learning support that can answer questions anytime, 24 hours a day.

[0811] "Means for entering questions" refers to an interface that allows users to enter questions or doubts that arise during their learning process into a text box.

[0812] "Means for receiving submitted questions" refers to the network connection and protocol used by the server to receive questions sent from the user's terminal.

[0813] "Means for generating answers and explanations using a generative AI model" refers to a system that utilizes an AI model (for example, a natural language processing model) to create answers and explanations based on received questions.

[0814] "Means of sending and displaying as an HTTP response to the user's device" refers to a method of sending the generated answer and explanation as an HTTP response to the user's device and displaying its contents.

[0815] "Means for evaluating generated answers" refers to a system that uses AI models or algorithms to evaluate the accuracy, logic, and other aspects of the answers generated.

[0816] "Methods of analysis using data mining techniques" refers to data mining methods used to analyze users' past answer data and identify learning trends.

[0817] "A means of generating the next problem using an AI model based on user tendencies" refers to a system that uses an AI model to generate new problems tailored to the user's learning tendencies and weaknesses.

[0818] "Means of sending and displaying as an HTTP response to the user's device" refers to a method of sending the generated problem as an HTTP response to the user's device and displaying its contents.

[0819] "A means of interacting with users in real time via chat" refers to a function where users can input questions in real time using a chat box, and the AI ​​responds immediately.

[0820] The learning support system of the present invention includes functions such as generating answers and explanations to user questions, evaluating answers, generating problems, and a chat function. This system consists of the user's terminal (PC, smartphone, tablet, etc.) and a server located in a cloud environment. The following describes specific embodiments for implementing the present invention.

[0821] 1. System Configuration

[0822] The system operates by users interacting with the server using individual terminals. The server is located in a cloud environment and performs data analysis and generation using AI models (e.g., GPT-4).

[0823] 2. User input of questions and generation of answers

[0824] The user enters their questions into a text box and clicks the submit button. The device sends the entered question data to the server as an HTTP request. The server analyzes the received question and generates an answer and explanation using a generative AI model. The generated answer and explanation are then sent to the device as an HTTP response and displayed on the user's screen.

[0825] For example, if a user inputs "Please explain the basics of differentiation," the server's AI model will generate the answer "Differentiation is a method for measuring the rate of change of a function" and send it to the terminal.

[0826] 3. Evaluation and feedback on the answers

[0827] When a user enters an answer to a presented problem and clicks the submit button, the device sends the entered answer data to the server as an HTTP request. The server inputs the received answer into an AI model and evaluates it based on criteria such as accuracy, logic, and clarity. Based on the evaluation results, it generates detailed feedback, sends it to the device as an HTTP response, and displays it on the user's screen.

[0828] For example, when a user submits an answer to a math problem, the server evaluates the answer and sends back feedback such as, "It's correct, but the explanation is insufficient."

[0829] 4. Problem generation and learning cycle

[0830] The server analyzes the user's past answer data using data mining techniques to identify the user's learning tendencies. Based on this analysis, it uses a generative AI model to generate the next problem that addresses the user's weaknesses and sends it to the terminal as an HTTP response. The terminal then displays the new problem to the user.

[0831] For example, if the server determines that a user does not understand "differentiation of a function," it will generate and send a problem such as "Find the derivative of the following function f(x)."

[0832] 5. 24-hour chat function

[0833] The user types a question into the chat box and clicks the send button. The device sends the entered question to the server in real time. The server inputs the received question into an AI chatbot model and generates an appropriate response. The generated response is sent to the device in real time as an HTTP response and displayed in the chat box.

[0834] For example, if a user asks in the chat, "Please explain how to calculate a definite integral," the server's AI chatbot will generate a response such as, "A definite integral is a method for finding the area of ​​a function within an interval," and immediately provide it to the user.

[0835] Example of a prompt

[0836] "Could you explain the basics of differential calculus?"

[0837] "What is differentiation?"

[0838] "Please solve these differential calculus problems."

[0839] "Please evaluate my answer."

[0840] The system of this invention acts as a 24-hour available tutor, dramatically improving the user's learning efficiency.

[0841] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0842] Step 1:

[0843] User's question input

[0844] Users enter any questions or doubts they have during their learning process into a text box and click the submit button.

[0845] Input: Text information entered by the user.

[0846] Specific operation: The browser on the user's device sends the entered question to the server via an asynchronous request (such as Ajax).

[0847] The terminal sends the entered question data to the server as an HTTP request.

[0848] Output: Query data sent to the server as an HTTP request.

[0849] Step 2:

[0850] Receiving questions on the server

[0851] The server receives HTTP requests sent from the terminal.

[0852] Input: An HTTP request containing question data sent from the terminal.

[0853] Specific operation: The server parses the content of the HTTP request and extracts the query data.

[0854] The server analyzes the received question data for processing.

[0855] Output: Analysis results of the questionnaire data.

[0856] Step 3:

[0857] Generating answers and explanations

[0858] The server inputs the analyzed question data into a generating AI model (e.g., GPT-4).

[0859] Input: Analyzed question data.

[0860] Specific operation: The server calls the API of the generation AI model and feeds the question data into the model. The model performs natural language processing and generates appropriate answers and explanations.

[0861] The server receives the output from the generated AI model.

[0862] Output: Generated solution and explanation.

[0863] Step 4:

[0864] Submit your answer and explanation.

[0865] The server sends the generated answer and explanation to the user's terminal as an HTTP response.

[0866] Input: Generated answer and explanation.

[0867] Specific operation: The server formats the answer and explanation into an HTTP response and sends it to the user's terminal.

[0868] The device displays the received answers and explanations on its screen.

[0869] Output: The answer and explanation displayed on the user's device.

[0870] Step 5:

[0871] User's answer input

[0872] The user enters their answer to the presented question and clicks the submit button.

[0873] Input: User-submitted answer data.

[0874] Specific operation: The device's browser sends the answer data to the server as an asynchronous request.

[0875] The terminal sends the entered answer data to the server as an HTTP request.

[0876] Output: HTTP request containing the answer data.

[0877] Step 6:

[0878] Evaluation of the answers

[0879] The server inputs the received answer data into the AI ​​model and performs evaluation.

[0880] Input: Received answer data.

[0881] Specific operation: The server inputs the answer data into an AI model for evaluation and performs evaluation based on criteria such as accuracy and logic. It then generates the evaluation results.

[0882] The server generates feedback based on the evaluation results.

[0883] Output: Evaluation results and generated feedback.

[0884] Step 7:

[0885] Send feedback

[0886] The server sends the feedback to the user's terminal as an HTTP response.

[0887] Input: Evaluation results and feedback.

[0888] Specific operation: The server formats the feedback into an HTTP response and sends it to the user's terminal.

[0889] The device displays the received feedback to the user.

[0890] Output: Feedback displayed on the user's device.

[0891] Step 8:

[0892] User learning trend analysis

[0893] The server analyzes past answer data using data mining techniques.

[0894] Input: User's past answer data.

[0895] Specific operation: The server extracts the user's past answer data from the cloud database and runs data mining algorithms to identify the user's learning tendencies and weaknesses.

[0896] The server saves the analysis results.

[0897] Output: User learning trend data.

[0898] Step 9:

[0899] Generating the next problem

[0900] The server generates the next problem using an AI model based on the user's learning tendencies.

[0901] Input: User learning trend data.

[0902] Specific operation: The server feeds learning trend data into the generating AI model and generates problems tailored to the user's level and weaknesses.

[0903] The server sends the generated problem to the user's terminal as an HTTP response.

[0904] Output: HTTP response containing the generated problem.

[0905] Step 10:

[0906] Display new problems

[0907] The device displays any new issues received to the user.

[0908] Input: HTTP response containing the generated problem.

[0909] Specific operation: The device's browser parses the HTTP response and displays the new problem on the screen.

[0910] Output: New issues displayed on the user's device.

[0911] Step 11:

[0912] Real-time chat function

[0913] The user types their question into the chat box and clicks the send button.

[0914] Input: The question entered in the chat box.

[0915] Specific operation: The device's browser sends chat messages to the server in real time (using WebSocket, etc.).

[0916] The server inputs the received question into the AI ​​chatbot model and generates an appropriate response.

[0917] Output: The generated response.

[0918] Specific operation: The server calls the AI ​​chatbot model's API in real time, analyzes the question, and generates a response.

[0919] The server sends the generated response to the terminal in real time as an HTTP response.

[0920] The device displays the received response in the chat box.

[0921] Output: The response displayed in the chat box.

[0922] (Application Example 1)

[0923] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0924] Modern food delivery services have limited capabilities to respond quickly and accurately to user questions and feedback, making it difficult for users to have a satisfying experience. Furthermore, there is a lack of information regarding ingredients and dishes, as well as appropriate recipe suggestions, forcing users to do their own research. Additionally, recipe suggestions and improvements based on user preferences and habits are not efficiently implemented. As a result, this leads to decreased user satisfaction and lower repeat customer rates.

[0925] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0926] In this invention, the server includes means for the user to input questions about ingredients or dishes, means for receiving the input questions, means for generating answers and explanations for the received questions, means for providing the generated answers and explanations to the user, means for the user to input feedback on delivered products, means for evaluating the input feedback, means for analyzing user trends based on the evaluation results, means for generating the next order or recipe based on the user trends, means for providing the generated recipe to the user, and means for interacting with the user in a chat format. This enables quick and accurate responses to user questions and feedback, and allows for the suggestion and improvement of recipes based on trends.

[0927] A "user" is an individual or legal entity that uses this system.

[0928] "Ingredients" refer to edible foods such as fresh produce and processed foods used for cooking or consumption.

[0929] "Cooking" refers to food that has been prepared by cooking ingredients in an appropriate manner and is ready for consumption.

[0930] "Questions" refer to unresolved questions or points of confusion that users have regarding ingredients or cooking.

[0931] "Feedback" refers to the evaluations and opinions that users provide regarding the delivered goods or services provided.

[0932] "Answer" refers to information that provides answers or explanations to users' questions.

[0933] A "recipe" is a set of instructions provided to the user that lists the ingredients and how to prepare a dish.

[0934] A "server" is a computer system that processes user questions and feedback and generates the necessary answers and recipes.

[0935] A "system" is a collection of means and devices for responding to user questions and feedback, and for performing trend analysis and suggesting recipes.

[0936] "Explanation" refers to a detailed explanation provided to address a problem or question in a way that makes it easy for the user to understand.

[0937] "Evaluation" involves analyzing feedback and making judgments based on appropriate criteria.

[0938] A "tendency" is a specific pattern or characteristic based on user behavior and preferences.

[0939] "Chat format" refers to a method of communication between the user and the system using text or voice.

[0940] The following describes embodiments for specifically implementing the present invention. The following is an example of a food delivery support application.

[0941] System Configuration

[0942] This system consists of user terminals and servers. Users use individual terminals (smartphones, tablets, etc.), and the servers are located in a cloud environment.

[0943] User question input and answer generation

[0944] 1. User input of questions: Users enter their questions or concerns about ingredients and cooking into the app's text box and click the submit button.

[0945] 2. Server-side answer generation: The server inputs the received question into a generating AI model (e.g., a natural language processing model such as GPT-2). The model analyzes the question and generates an answer and explanation. This answer is then provided to the user in real time.

[0946] User feedback and ratings

[0947] 1. Feedback Input: Users enter their feedback about the delivered product into the app's text box and click the submit button.

[0948] 2. Server Evaluation: The server analyzes and evaluates the received feedback. Criteria used for this evaluation include accuracy, appropriateness, and clarity. Based on the evaluation results, detailed feedback and explanations are generated.

[0949] 3. Providing evaluations: Provide users with generated explanations in real time.

[0950] Problem generation and learning cycle

[0951] 1. User Trend Analysis: The server analyzes user trends based on the evaluation results. This includes analyzing past response data.

[0952] 2. Suggestions for the next order and recipe: Based on the analysis results, the server generates and provides orders and recipes that match the user's preferences and tendencies.

[0953] 24-hour chat function

[0954] 1. User Question Input: The user enters their question in the chat box and clicks the send button. The entered question is sent to the server in real time.

[0955] 2. Real-time response: The server inputs the received question into the AI ​​chatbot model and generates an appropriate response. The generated response is provided to the user in real time.

[0956] Hardware and software

[0957] Hardware: Cloud servers, smartphones, tablets

[0958] Software: Flask (web framework), Transformers library, GPT-2 model

[0959] Specific example

[0960] For example, if a user asks, "What's the perfect recipe for a special family dinner on Wednesday?", the server's AI model will generate an answer such as, "Parmesan chicken with garlic mashed potatoes is perfect for a family dinner on Wednesday. Add some leafy greens and lemon dressing, and you've got the perfect course."

[0961] In this way, users can easily resolve their questions and provide feedback on the quality and taste of the delivered products, which can then be used to improve their next order.

[0962] Example of a prompt:

[0963] "What are some great recipes for a special dinner to bring the family together on Wednesday?"

[0964] The system works by having the AI ​​model generate appropriate answers based on user input and providing them to the user via the server.

[0965] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0966] Step 1:

[0967] The user launches the smartphone app, enters their questions about ingredients or cooking into the text box, and clicks the submit button.

[0968] Input: User's question (e.g., "What side dishes would go well with this dish?")

[0969] Output: User input data (HTTP request format)

[0970] Step 2:

[0971] The terminal receives the user's question and sends it to the server as an HTTP request.

[0972] Input: User input data

[0973] Output: HTTP request to the server

[0974] Step 3:

[0975] The server inputs the user's question into a generating AI model, analyzes the data, and generates an answer and explanation.

[0976] Input: User's question included in the HTTP request

[0977] Output: Generated answer and explanation (text data)

[0978] Step 4:

[0979] The server sends the generated answer and explanation to the terminal as an HTTP response.

[0980] Input: Generated answer and explanation

[0981] Output: HTTP response (including answer and explanation)

[0982] Step 5:

[0983] The terminal receives an HTTP response from the server and displays the answer and explanation on the user interface.

[0984] Input: HTTP response

[0985] Output: Answers and explanations displayed on the user screen

[0986] Step 6:

[0987] The user enters feedback about the delivered product in the text box and clicks the submit button.

[0988] Input: User feedback (e.g., "The food tasted very good, but it was a little cold.")

[0989] Output: User input data (HTTP request format)

[0990] Step 7:

[0991] The terminal receives the user's input and sends it to the server as an HTTP request.

[0992] Input: User input data

[0993] Output: HTTP request to the server

[0994] Step 8:

[0995] The server analyzes and evaluates the received feedback, and generates detailed feedback and explanations.

[0996] Input: User feedback

[0997] Output: Evaluation results and explanations (text data)

[0998] Step 9:

[0999] The server sends the generated evaluation results and explanations to the terminal as an HTTP response.

[1000] Input: Evaluation results and explanation

[1001] Output: HTTP response (including evaluation results and explanations)

[1002] Step 10:

[1003] The terminal receives an HTTP response from the server and displays the evaluation results and explanations on the user interface.

[1004] Input: HTTP response

[1005] Output: Evaluation results and explanations displayed on the user screen

[1006] Step 11:

[1007] The server analyzes user trends based on evaluation results and generates subsequent orders and recipes.

[1008] Input: Evaluation result data

[1009] Output: New order suggestions and recipes (text data)

[1010] Step 12:

[1011] The server sends the generated new order suggestions and recipes to the terminal as an HTTP response.

[1012] Input: New order suggestions or recipes

[1013] Output: HTTP response (including new order suggestions and recipes)

[1014] Step 13:

[1015] The terminal receives an HTTP response from the server and displays new order suggestions and recipes on the user interface.

[1016] Input: HTTP response

[1017] Output: New order suggestions and recipes displayed on the user screen

[1018] The above outlines the specific processing steps. In each step, the user, terminal, and server work together to process the data, creating a system that enables the resolution of user questions and the utilization of user feedback.

[1019] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1020] The learning support system of the present invention features the generation of answers and explanations to questions, evaluation of answers, generation of questions, a chat function, and an emotion engine that recognizes the user's emotions. The following describes specific embodiments for implementing the present invention.

[1021] System Configuration

[1022] This system consists of a user's device, a server, and an emotion engine. Users use individual devices (PCs, smartphones, tablets, etc.), and the server is located in a cloud environment. The emotion engine is equipped with AI technology that analyzes emotions from the user's voice, facial expressions, text, etc.

[1023] User question input and answer generation

[1024] 1. User input questions:

[1025] User: Enter any questions or doubts that arise during the learning process into the text box and click the submit button.

[1026] Terminal: Sends the entered text to the server as an HTTP request.

[1027] 2. Generating the solution:

[1028] Server: Inputs the received questions into a natural language processing model (e.g., BERT or GPT model).

[1029] Server: The AI ​​model analyzes the questions and generates answers and explanations.

[1030] 3. Emotion recognition:

[1031] Device: Uses an emotion engine to recognize the user's emotional state. This includes facial expression analysis and voice analysis.

[1032] 4. Adjustments based on answers and emotions:

[1033] Server: Adjusts the content and tone of the answers and explanations based on the recognized emotional state.

[1034] Server: Sends the generated answer and explanation to the terminal as an HTTP response.

[1035] 5. Display to the user:

[1036] Terminal: Receives responses from the server and displays the answers and explanations on the screen.

[1037] Evaluation and feedback on the answers

[1038] 1. Enter your answer:

[1039] User: Enter your answer to the provided question and click the submit button.

[1040] Terminal: Sends the entered answer to the server as an HTTP request.

[1041] 2. Rating:

[1042] Server: Inputs received answers into the AI ​​model and performs evaluation. Evaluation criteria include accuracy, logic, and clarity.

[1043] Server: Generates detailed feedback and explanations based on the evaluation results.

[1044] 3. Emotion recognition:

[1045] Device: Uses an emotion engine to recognize the user's emotional state.

[1046] 4. Feedback and emotional adjustments:

[1047] Server: Adjusts the content and tone of feedback based on the recognized emotional state.

[1048] Server: Sends feedback to the terminal as an HTTP response.

[1049] 5. Display to the user:

[1050] Terminal: Receives feedback from the server and displays it on the screen.

[1051] Problem generation and learning cycle

[1052] 1. User trend analysis:

[1053] Server: Analyzes user learning trends based on evaluation results. This uses data mining techniques and machine learning models.

[1054] 2. Generating new problems:

[1055] Server: Based on the analysis results, it identifies user weaknesses and generates new problems.

[1056] Server: Sends a new problem to the terminal as an HTTP response.

[1057] 3. Emotion recognition:

[1058] Terminal: Uses an emotion engine to recognize the user's emotional state. This allows for adjustments to the difficulty and content of the generated problems.

[1059] 4. Display to the user:

[1060] Terminal: Receives new issues from the server and displays them on the screen.

[1061] Chat function

[1062] 1. Enter your question:

[1063] User: Type your question in the chat box and click the send button.

[1064] Terminal: Sends the entered questions to the server in real time.

[1065] 2. Real-time response generation:

[1066] Server: Inputs received questions into the AI ​​chatbot model.

[1067] Server: The AI ​​chatbot analyzes the question and generates a response in real time.

[1068] 3. Emotion recognition:

[1069] Device: Uses an emotion engine to recognize the user's emotional state, thereby adjusting the response accordingly.

[1070] 4. Providing a response:

[1071] Server: Sends the generated response to the terminal as an HTTP response in real time.

[1072] Terminal: Receives responses from the server and displays them in the chat box.

[1073] Specific example

[1074] For example, if a user asks, "Could you explain the basics of differentiation?", the server's AI model will generate the answer, "Differentiation is a method for measuring the rate of change of a function." Furthermore, if the emotion engine detects that the user is feeling fatigued, the answer will be adjusted to something like, "Differentiation is a method for measuring the rate of change of a function, and it may seem difficult, but let's work through it together."

[1075] The system of this invention optimizes learning by taking into account the user's emotional state, thereby providing more effective learning support. This dramatically improves the user's motivation and learning efficiency.

[1076] The following describes the processing flow.

[1077] User question input and answer generation

[1078] Step 1:

[1079] User: Enter any questions or doubts that arise during the learning process into the text box and click the submit button.

[1080] Step 2:

[1081] Terminal: Sends the entered text to the server as an HTTP request.

[1082] Step 3:

[1083] Server: Inputs the received questions into a natural language processing model (e.g., BERT or GPT model).

[1084] Step 4:

[1085] Server: The AI ​​model analyzes the questions and generates answers and explanations.

[1086] Step 5:

[1087] Device: Captures the user's facial expressions and voice and sends them to the emotion engine.

[1088] Step 6:

[1089] Emotion Engine: Analyzes the user's emotional state based on captured data.

[1090] Step 7:

[1091] Server: Adjusts the content and tone of the answers and explanations based on the recognized emotional state.

[1092] Step 8:

[1093] Server: Sends the adjusted answer and explanation to the terminal as an HTTP response.

[1094] Step 9:

[1095] Terminal: Receives responses from the server and displays the answers and explanations on the screen.

[1096] Evaluation and feedback on the answers

[1097] Step 1:

[1098] User: Enter your answer to the provided question and click the submit button.

[1099] Step 2:

[1100] Terminal: Sends the entered answer to the server as an HTTP request.

[1101] Step 3:

[1102] Server: Inputs received answers into the AI ​​model and performs evaluation.

[1103] Step 4:

[1104] Server: Generates detailed feedback and explanations based on the evaluation results.

[1105] Step 5:

[1106] Device: Captures the user's facial expressions and voice and sends them to the emotion engine.

[1107] Step 6:

[1108] Emotion Engine: Analyzes the user's emotional state based on captured data.

[1109] Step 7:

[1110] Server: Adjusts the content and tone of feedback based on the recognized emotional state.

[1111] Step 8:

[1112] Server: Sends the adjusted feedback to the terminal as an HTTP response.

[1113] Step 9:

[1114] Terminal: Receives feedback from the server and displays it on the screen.

[1115] Problem generation and learning cycle

[1116] Step 1:

[1117] Server: Analyzes user learning trends based on evaluation results.

[1118] Step 2:

[1119] Server: Based on the analysis results, it identifies user weaknesses and generates new problems.

[1120] Step 3:

[1121] Device: Captures the user's facial expressions and voice and sends them to the emotion engine.

[1122] Step 4:

[1123] Emotion Engine: Analyzes the user's emotional state based on captured data.

[1124] Step 5:

[1125] Server: Adjusts the difficulty and content of generated problems based on the recognized emotional state.

[1126] Step 6:

[1127] Server: Sends a new problem to the terminal as an HTTP response.

[1128] Step 7:

[1129] Terminal: Receives new issues from the server and displays them on the screen.

[1130] Chat function

[1131] Step 1:

[1132] User: Type your question in the chat box and click the send button.

[1133] Step 2:

[1134] Terminal: Sends the entered questions to the server in real time.

[1135] Step 3:

[1136] Server: Inputs received questions into the AI ​​chatbot model.

[1137] Step 4:

[1138] Server: The AI ​​chatbot analyzes the question and generates a response in real time.

[1139] Step 5:

[1140] Device: Captures the user's facial expressions and voice and sends them to the emotion engine.

[1141] Step 6:

[1142] Emotion Engine: Analyzes the user's emotional state based on captured data.

[1143] Step 7:

[1144] Server: Adjusts the response based on the recognized emotional state.

[1145] Step 8:

[1146] Server: Sends the adjusted response to the terminal as an HTTP response in real time.

[1147] Step 9:

[1148] Terminal: Receives responses from the server and displays them in the chat box.

[1149] (Example 2)

[1150] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[1151] Traditional learning support systems can provide answers and explanations to users' questions, but they lack the ability to consider the user's emotional state. Therefore, they cannot adequately recognize the fatigue or frustration a user might experience during learning and adjust the content of responses and feedback accordingly, resulting in a failure to maximize learning effectiveness. Furthermore, the lack of real-time emotion recognition and feedback based on those results made it difficult to maintain user motivation.

[1152] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1153] In this invention, the server includes means for the user to input questions, means for receiving the input questions, means for generating answers and explanations for the received questions using a generative AI model, means for providing the generated answers and explanations to the user, means for evaluating the generated answers, means for analyzing the user's tendencies based on the evaluation results, means for generating the next question based on the user's tendencies, means for providing the generated question to the user, means for interacting with the user in a chat format, means for recognizing the user's emotional state, and means for adjusting the content of answers and feedback based on the emotional state. This makes it possible to recognize the user's emotional state in real time and adjust the response content and feedback according to that state, thereby increasing the user's motivation to learn and enabling more effective learning support.

[1154] A "user" is an individual or group who uses a learning support system to seek answers to their questions or receive feedback.

[1155] "Questions" refer to the questions or things that users need to understand during the learning process.

[1156] "Input means" refers to interface means for users to input questions and answers into the system, and includes keyboards and touch panels.

[1157] "Receiving means" refers to the means for receiving data transmitted by a user. This includes data receiving systems via a network.

[1158] A "generative AI model" is an artificial intelligence model that uses natural language processing to analyze user input and generate appropriate answers and explanations.

[1159] "Answer" refers to the content of the response to the user's question.

[1160] "Explanation" refers to a detailed explanation provided alongside the answer, serving as supplementary information to help users gain a deeper understanding of their questions.

[1161] "Means of provision" refers to means of displaying or communicating the generated answers and explanations to the user.

[1162] "Evaluation means" refers to methods for evaluating the accuracy, logic, clarity, etc., of answers entered by users, and includes AI models.

[1163] "Evaluation results" refer to data calculated using evaluation methods for user responses.

[1164] A "trend analysis method" is a means of analyzing a user's learning tendencies based on evaluation results.

[1165] The "problem generation method" is a means of generating new learning problems based on the user's learning tendencies and weaknesses, and it uses a generation AI model.

[1166] "Chat dialogue methods" refer to means by which users and systems can interact in real time, and include chatbots that use natural language processing.

[1167] "Emotion recognition means" refers to methods for recognizing a user's emotional state by analyzing their voice, facial expressions, text, etc.

[1168] "Adjustment means" are means for adjusting the content and tone of responses and feedback based on the emotional state obtained by the emotion recognition means.

[1169] The learning support system of the present invention combines the generation of answers and explanations to user questions, evaluation of answers, generation of problems, a chat function, and an emotion recognition function. The following describes in detail the embodiments for specifically implementing the present invention.

[1170] System Configuration

[1171] This system consists of a user's device, a server, and an emotion engine. Users utilize devices such as PCs, smartphones, and tablets. The server is located in a cloud environment and performs the necessary computational processing. The emotion engine incorporates AI technology that analyzes emotions from the user's voice, facial expressions, and text. Specifically, the device collects the user's facial expressions and voice through its camera and microphone and sends them to the server for analysis.

[1172] Hardware and software usage

[1173] The server implements generative AI models (such as natural language processing models like BERT and GPT) to analyze questions and answers submitted by users. It also uses data mining techniques and machine learning models to analyze users' learning tendencies. The terminal is equipped with an emotion recognition engine that analyzes the user's voice and facial expressions in real time to recognize their emotional state. Emotion recognition utilizes facial recognition and voice analysis technologies.

[1174] User question input and answer generation

[1175] The user enters any questions or doubts that arise during the learning process into a text box and clicks the submit button. For example, they might enter, "Please explain the basics of differentiation." The device sends the entered text to the server as an HTTP request. The server inputs the received question into a generating AI model and generates an answer and explanation. An example of a generated answer is, "Differentiation is a method for measuring the rate of change of a function."

[1176] Emotion recognition and regulation

[1177] The emotion engine built into the device analyzes the user's facial expressions, voice, and text to convey emotions. For example, it might recognize "fatigue" from the user's facial expression. Based on the recognized emotional state, the server adjusts the content and tone of the answers and explanations. For instance, it might add an additional message such as, "This might seem difficult, but let's work hard together."

[1178] Evaluation and feedback on the answers

[1179] The user enters an answer to a given question and clicks the submit button. The device sends the entered answer to the server as an HTTP request. The server inputs the received answer into an AI model and evaluates it based on accuracy, logic, and clarity. Based on the evaluation results, it generates detailed feedback and explanations. For example, if the user answers the question "What are the fundamental principles of electromagnetism?" with "The motion of electrons and their field interactions," the server provides feedback such as "This answer is superficially correct, but it would be good to explain it in more detail."

[1180] Problem generation and learning cycle

[1181] The server analyzes the user's learning tendencies based on the evaluation results and identifies the user's weaknesses using data mining techniques and machine learning models. It generates new problems related to the identified weaknesses and sends them to the terminal as an HTTP response. The terminal displays the new problems on its screen, allowing the user to continue learning.

[1182] Chat function

[1183] When a user types a question into the chat box and clicks the send button, the device sends the entered question to the server in real time. The server inputs the received question into an AI chatbot model and generates a response in real time. If emotion recognition determines that the user's emotional state is "frustrated," the response is adjusted to something like, "This may be difficult to understand, but I will explain it carefully." The server then sends the generated response to the device as an HTTP response in real time, and the device displays it in the chat box.

[1184] As described above, the present invention provides a more effective learning experience by providing real-time responses to user questions and answers, adjusting feedback content based on emotion recognition, and generating new problems by analyzing the user's learning tendencies.

[1185] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1186] Step 1:

[1187] The user enters their question into the text box and clicks the submit button.

[1188] Input: The user enters "Please explain the basics of differentiation."

[1189] Operation: The terminal retrieves the entered text.

[1190] Output: The entered question is displayed on the terminal.

[1191] Step 2:

[1192] The terminal sends the entered question to the server as an HTTP request.

[1193] Input: The question entered in the text box.

[1194] Operation: The terminal converts the question into packets and sends them to the server via the network.

[1195] Output: The server receives the question.

[1196] Step 3:

[1197] The server inputs the received questions into a natural language processing model.

[1198] Input: Text data of the question sent from the device.

[1199] Operation: The server processes the question into an appropriate format and inputs it as a prompt to the generating AI model (e.g., GPT-3).

[1200] Output: The generative AI model generates the answer and explanation.

[1201] Step 4:

[1202] The server retrieves the answer and explanation as output from the generated AI model and generates a response to provide to the user.

[1203] Input: Text data of answers and explanations from a generative AI model.

[1204] Operation: Converts the answers and explanations obtained by the server into an HTTP response format.

[1205] Output: Response data is generated.

[1206] Step 5:

[1207] The device uses an emotion engine to recognize the user's emotional state.

[1208] Input: User's facial image and voice data acquired from the camera and microphone.

[1209] Operation: The device uses an emotion engine to analyze the user's facial expressions and voice to detect their emotional state.

[1210] Output: The user's emotional state (e.g., fatigue, irritation) is identified by the emotion engine.

[1211] Step 6:

[1212] The server adjusts the content and tone of the answers and explanations based on the recognized emotional state.

[1213] Input: Emotional state data sent from the device, along with the generated answer and explanation.

[1214] Operation: The server analyzes the emotional state and makes adjustments such as, "Differentiation is a method for measuring the rate of change of a function. It may seem difficult, but let's work through it together."

[1215] Output: Text data of the adjusted answers and explanations.

[1216] Step 7:

[1217] The server sends the adjusted answer to the terminal as an HTTP response.

[1218] Input: Text data of the adjusted answers and explanations.

[1219] Operation: The server converts the data into an HTTP response format and sends it to the terminal over the network.

[1220] Output: The terminal receives a response.

[1221] Step 8:

[1222] The terminal displays the response from the server on the screen.

[1223] Input: Adjusted answer and explanation data received from the server.

[1224] Operation: The device displays the answer and explanation on the screen.

[1225] Output: The user can view the answer and explanation on the screen.

[1226] Step 9:

[1227] The user enters their answer to the provided question and clicks the submit button.

[1228] Input: User's answer (e.g., "The motion of electrons and their field interactions").

[1229] Action: The device retrieves the answer and prepares to send it to the server as an HTTP request.

[1230] Output: The entered answer is ready to be sent by the terminal.

[1231] Step 10:

[1232] The device sends the answer to the server as an HTTP request.

[1233] Input: Text data of the answer entered by the user.

[1234] Operation: The terminal converts the answer into packets and sends them to the server over the network.

[1235] Output: The server receives the answer.

[1236] Step 11:

[1237] The server inputs the answer into the AI ​​model and performs evaluation.

[1238] Input: Text data of the answer sent from the device.

[1239] Operation: The server inputs the answer into the AI ​​model and performs analysis based on evaluation criteria.

[1240] Output: Evaluation results are generated.

[1241] Step 12:

[1242] The server generates detailed feedback and explanations based on the evaluation results.

[1243] Input: Evaluation result data generated by an AI model.

[1244] Operation: The server generates feedback and explanations for the user based on the evaluation results.

[1245] Output: Text data of feedback and commentary.

[1246] Step 13:

[1247] The device recognizes the user's emotional state.

[1248] Input: User's facial image and voice data acquired from the camera and microphone.

[1249] Operation: The device uses an emotion engine to analyze the user's facial expressions and voice.

[1250] Output: The user's emotional state (e.g., satisfied, dissatisfied) is identified.

[1251] Step 14:

[1252] The server adjusts the content and tone of the feedback based on the emotional state.

[1253] Input: Emotional state data sent from the device, along with generated feedback and commentary.

[1254] Operation: The server adjusts the tone of feedback and commentary based on the emotional state.

[1255] Output: Text data of adjusted feedback and commentary.

[1256] Step 15:

[1257] The server sends the adjusted feedback to the terminal as an HTTP response.

[1258] Input: Text data of adjusted feedback and commentary.

[1259] Operation: The server converts the data into an HTTP response format and sends it to the terminal over the network.

[1260] Output: The terminal receives a response.

[1261] Step 16:

[1262] The terminal displays feedback and explanations from the server on the screen.

[1263] Input: Feedback and commentary data received from the server.

[1264] Action: The device displays feedback and explanations on the screen.

[1265] Output: The user will be able to view feedback and explanations on the screen.

[1266] The detailed processing flow for each step described above allows users to input questions, receive answers, get feedback on those answers, and even have their feedback adjusted based on their emotions, thereby maximizing the effectiveness of the learning support system.

[1267] (Application Example 2)

[1268] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[1269] Conventional learning support systems provide functions for answering user questions, evaluating their performance, and generating new problems. However, they fail to take into account the user's emotional state, resulting in insufficient improvement in learning effectiveness. As a result, users may not receive appropriate support when they feel fatigued or stressed, potentially leading to a decrease in their motivation to learn. Furthermore, because answers and feedback are provided regardless of the user's emotions, there is a need to address the issue of poor satisfaction and comprehension during the learning process.

[1270] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for inputting questions, means for receiving the input questions, means for generating answers and explanations for the received questions, means for providing the generated answers and explanations to the user, means for recognizing the user's emotions, means for adjusting the tone and content of the answers and explanations based on the recognized emotions, means for evaluating the generated answers, means for analyzing the user's tendencies based on the evaluation results, means for generating the next question based on the user's tendencies, means for adjusting the difficulty and content of the questions considering the user's emotional state before providing the questions, means for providing the questions to the user, means for interacting with the user in a chat format, and means for recognizing the user's emotional state during the chat and adjusting the response content based on the recognized emotional state. This enables appropriate learning support that takes into account the user's emotional state, and is expected to improve learning effectiveness and user satisfaction.

[1271] "Means for inputting questions" refers to an interface for users to input questions or doubts that arise during their learning process.

[1272] "Means for receiving entered questions" refers to a mechanism for the system to receive questions entered by the user.

[1273] "Means for generating answers and explanations to received questions" refers to a set of algorithms and programs for generating appropriate answers and additional explanations based on received questions.

[1274] "Means of providing users with generated answers and explanations" refers to a function for displaying generated answers and explanations to users.

[1275] "Means of recognizing user emotions" refers to technologies and devices that analyze a user's voice, facial expressions, text, etc., to recognize their emotional state.

[1276] "Means of adjusting the tone and content of answers and explanations based on recognized emotions" refers to algorithms and functions that adjust the wording and level of detail of answers and explanations according to the recognized emotional state of the user.

[1277] "Means for evaluating generated answers" refers to a mechanism for evaluating the accuracy and validity of generated answers.

[1278] "Means for analyzing user trends based on evaluation results" refers to algorithms for analyzing users' learning patterns and proficiency levels based on the evaluation results of their answers.

[1279] "Means for generating the next problem based on user trends" refers to a system that analyzes the user's learning trends and generates the next problem to be provided based on those trends.

[1280] "Means of adjusting the difficulty and content of generated problems based on the user's emotional state before providing them" refers to a function that appropriately adjusts the difficulty and content of generated problems based on the user's current emotional state.

[1281] "Means of providing generated problems to users" refers to a function for displaying adjusted problems to users.

[1282] "Means of interacting with users in a chat format" refers to chat interfaces and related technologies that allow users to interact with the system in real time.

[1283] "Means for recognizing a user's emotional state during a chat and adjusting the response based on that emotional state" refers to algorithms and technologies for recognizing a user's emotions during a chat and dynamically adjusting the content of the response accordingly.

[1284] The learning support system of the present invention, in addition to generating answers to user questions, providing evaluations, generating problems, and offering a chat function, improves learning effectiveness by recognizing the user's emotions and adjusting learning content and responses based on those emotions. This system consists of the user's terminal, a server, and an emotion engine.

[1285] System Configuration

[1286] A terminal is an information processing device (smartphone, tablet, or personal computer) that receives user input and communicates with a server. The terminal is equipped with a camera and microphone to collect data for emotion recognition.

[1287] Server: Located in a cloud environment, it runs various AI models (e.g., natural language processing models and emotion recognition models). Its main roles are generating answers and explanations, evaluation, problem generation, and adjustments based on the user's emotional state.

[1288] Emotion Engine: Analyzes the user's emotional state from their voice, facial expressions, text, etc., and adjusts responses and answers accordingly.

[1289] User question input and answer generation

[1290] 1. Users enter any questions or doubts that arise during the learning process into the text box on their device and submit them.

[1291] 2. The terminal sends the entered text to the server.

[1292] 3. The server inputs the received questions into a natural language processing model (e.g., BERT or GPT model) and generates answers and explanations.

[1293] 4. The device analyzes the user's emotions using their facial expressions and voice, and sends that data to the server.

[1294] 5. The server adjusts the content and tone of the answers and explanations based on the recognized emotional state and sends them to the terminal.

[1295] 6. The device displays the adjusted answers and explanations on the screen.

[1296] Evaluation and feedback on the answers

[1297] 1. The user enters and submits their answer to the provided question.

[1298] 2. The terminal sends the entered answer to the server.

[1299] 3. The server inputs the answer into the AI ​​model and performs an evaluation. Based on the evaluation results, it generates detailed feedback and explanations.

[1300] 4. The device uses an emotion engine to recognize the user's emotional state and sends that data to the server.

[1301] 5. The server adjusts the content and tone of the feedback based on the emotional state and sends it to the device.

[1302] 6. The device displays the adjusted feedback on the screen.

[1303] Problem generation and learning cycle

[1304] 1. Based on the evaluation results, the server analyzes the user's learning tendencies and generates the next problem using data mining techniques and machine learning models.

[1305] 2. The device recognizes the user's emotional state before providing a problem and sends that data to the server.

[1306] 3. The server adjusts the difficulty and content of the problem, taking into account the emotional state, and sends it to the terminal.

[1307] 4. The device displays the adjusted issues on the screen.

[1308] Chat function

[1309] 1. The user types their question into the chat box and sends it.

[1310] 2. The terminal sends the question to the server in real time.

[1311] 3. The server uses an AI chatbot model to analyze the question and generate a response in real time.

[1312] 4. The device uses an emotion engine to recognize the user's emotional state and sends that data to the server.

[1313] 5. The server adjusts the response based on the emotional state and sends it to the terminal.

[1314] 6. The device displays the adjusted response in the chat box.

[1315] Specific examples and prompt statements

[1316] For example, if a user asks, "Could you explain the basics of differentiation?", the server will generate the answer, "Differentiation is a method for measuring the rate of change of a function." If the emotion engine detects that the user is fatigued, the answer will be adjusted to, "Differentiation is a method for measuring the rate of change of a function, and it may seem difficult, but let's work through it together."

[1317] Specific examples of prompt statements are as follows:

[1318] "Please explain the basics of differentiation. The user seems a little tired. Please provide the answer in a gentle tone."

[1319] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1320] Step 1:

[1321] Users enter any questions or doubts that arise during the learning process into the text box on their device and submit them.

[1322] Input: Text entered by the user

[1323] Output: HTTP request from terminal to server

[1324] Specific operation: When the user enters a question and clicks the "Send" button, the terminal sends the entered text data to the server.

[1325] Step 2:

[1326] The terminal sends the entered text to the server.

[1327] Input: Text data of the question

[1328] Output: Text data sent to the server

[1329] Specific action: The device sends text data containing the user's question to the server as an HTTP request.

[1330] Step 3:

[1331] The server inputs the received questions into a natural language processing model (e.g., BERT or GPT model) and generates answers and explanations.

[1332] Input: Text data of the question

[1333] Output: Generated solution and explanation data

[1334] Specific operation: The server inputs the text data of the question into a natural language processing model, and uses a generative AI model to generate the answer and explanation.

[1335] Step 4:

[1336] The device analyzes the user's emotions using their facial expressions and voice, and sends that data to a server.

[1337] Input: User's facial expression data, voice data

[1338] Output: Emotional state data

[1339] Specific operation: The device uses facial expressions and voice data collected by the camera and microphone to allow the emotion engine to recognize the emotional state and send that data to the server.

[1340] Step 5:

[1341] The server adjusts the content and tone of the answers and explanations based on the recognized emotional state, and then sends them to the terminal.

[1342] Input: Emotional state data, generated answers and explanatory data

[1343] Output: Adjusted answer and explanation data

[1344] Specific operation: The server uses the emotion data recognized by the emotion engine to apply an algorithm that adjusts the tone and level of detail of the answers and explanations, generates the adjusted answer and explanation data, and sends it to the terminal.

[1345] Step 6:

[1346] The device displays the adjusted answers and explanations on the screen.

[1347] Input: Adjusted answer and explanation data

[1348] Output: Adjusted answers and explanations displayed on the screen

[1349] Specific operation: The device displays the adjusted answer and explanation data it has received on the screen in a user-friendly format.

[1350] Step 7:

[1351] Users enter and submit their answers to the provided questions.

[1352] Input: User-entered answer data

[1353] Output: Answer data sent as an HTTP request from the terminal to the server.

[1354] Specific operation: When the user enters the answer to the problem and clicks the "Submit" button, the device sends the answer data to the server.

[1355] Step 8:

[1356] The terminal sends the entered answer to the server.

[1357] Input: Answer data

[1358] Output: Answer data sent to the server

[1359] Specific operation: The device sends the user's answer data to the server as an HTTP request.

[1360] Step 9:

[1361] The server inputs the answer into the AI ​​model and performs an evaluation. Based on the evaluation results, it generates detailed feedback and explanations.

[1362] Input: Answer data

[1363] Output: Evaluation results, feedback, and explanatory data

[1364] Specific operation: The server inputs the answer data into the AI ​​model, applies an evaluation algorithm to generate evaluation results, and then generates feedback and explanatory data based on those results.

[1365] Step 10:

[1366] The device uses an emotion engine to recognize the user's emotional state and sends that data to the server.

[1367] Input: User's facial expression data, voice data

[1368] Output: Emotional state data

[1369] Specific operation: The device uses facial expressions and voice data collected by the camera and microphone to allow the emotion engine to recognize the emotional state and send that data to the server.

[1370] Step 11:

[1371] The server adjusts the content and tone of the feedback based on the emotional state and sends it to the device.

[1372] Input: Emotional state data, evaluation results, and feedback data

[1373] Output: Adjusted feedback data

[1374] Specific operation: The server uses the emotion data recognized by the emotion engine to apply an algorithm that adjusts the tone and level of detail of the feedback, generates the adjusted feedback data, and sends it to the terminal.

[1375] Step 12:

[1376] The device displays adjusted feedback on the screen.

[1377] Input: Adjusted feedback data

[1378] Output: Adjusted feedback displayed on the screen

[1379] Specific operation: The device displays the adjusted feedback data it has received on the screen in a user-friendly format.

[1380] Step 13:

[1381] Based on the evaluation results, the server analyzes the user's learning tendencies and generates the next problem using data mining techniques and machine learning models.

[1382] Input: Evaluation result data

[1383] Output: Newly generated problem data

[1384] Specific operation: The server uses the evaluation result data to analyze the user's learning tendencies using data mining techniques and machine learning models, and then generates the next problem to be presented.

[1385] Step 14:

[1386] The device recognizes the user's emotional state before providing a problem and sends that data to the server.

[1387] Input: User's facial expression data, voice data

[1388] Output: Emotional state data

[1389] Specific operation: The device uses facial expressions and voice data collected by the camera and microphone to allow the emotion engine to recognize the emotional state and send that data to the server.

[1390] Step 15:

[1391] The server adjusts the difficulty and content of the problem, taking into account the emotional state, and then sends it to the terminal.

[1392] Input: Emotional state data, new problem data

[1393] Output: Adjusted new problem data

[1394] Specific operation: The server uses the emotion data recognized by the emotion engine to apply an algorithm that adjusts the difficulty and content of the problem, generates new adjusted problem data, and sends it to the terminal.

[1395] Step 16:

[1396] The device displays the adjusted issues on the screen.

[1397] Input: Adjusted new problem data

[1398] Output: Adjusted issues displayed on the screen

[1399] Specific operation: The device displays the adjusted problem data it has received on the screen in a user-friendly format.

[1400] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1401] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1402] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[1403] [Third Embodiment]

[1404] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[1405] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1406] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1407] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[1408] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1409] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1410] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1411] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1412] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1413] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1414] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1415] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[1416] The learning support system of the present invention includes generating answers and explanations to questions, evaluating answers, generating problems, and a chat function. The following describes specific embodiments for implementing the present invention.

[1417] System Configuration

[1418] This system consists of user terminals and servers. Users use individual terminals (PCs, smartphones, tablets, etc.), and the servers are located in a cloud environment.

[1419] User question input and answer generation

[1420] 1. User input questions:

[1421] User: Enter any questions or doubts that arise during the learning process into the text box and click the submit button.

[1422] Terminal: Sends the entered question to the server as an HTTP request.

[1423] 2. Generating the solution:

[1424] Server: Inputs the received questions into an AI model (for example, a natural language processing model such as BERT or GPT).

[1425] Server: The AI ​​model analyzes the question and generates the answer and explanation.

[1426] Server: Sends the generated answer and explanation to the terminal as an HTTP response.

[1427] 3. Display to the user:

[1428] Terminal: Receives responses from the server and displays the answers and explanations on the screen.

[1429] Evaluation and feedback on the answers

[1430] 1. Enter your answer:

[1431] User: Enter your answer to the submitted question and click the submit button.

[1432] Terminal: Sends the entered answer to the server as an HTTP request.

[1433] 2. Rating:

[1434] Server: Inputs received answers into the AI ​​model and performs evaluation. Evaluation criteria include accuracy, logic, and clarity.

[1435] Server: Generates detailed feedback and explanations based on the evaluation results.

[1436] Server: Sends feedback to the terminal as an HTTP response.

[1437] 3. Display to the user:

[1438] Terminal: Receives feedback from the server and displays it on the screen.

[1439] Problem generation and learning cycle

[1440] 1. User trend analysis:

[1441] Server: Analyzes user learning trends based on evaluation results. This is done by analyzing past answer data using data mining techniques.

[1442] 2. Problem generation:

[1443] Server: Based on the analysis results, it identifies the user's weaknesses and uses an AI model to generate the next problem. The generated problem corresponds to the user's learning level and weaknesses.

[1444] Server: Sends a new problem to the terminal as an HTTP response.

[1445] 3. Display to the user:

[1446] Terminal: Receives new issues from the server and displays them on the screen.

[1447] 24-hour chat function

[1448] 1. User question input:

[1449] User: Type your question in the chat box and click the send button.

[1450] Terminal: Sends the entered questions to the server in real time.

[1451] 2. Real-time response:

[1452] Server: Inputs received questions into the AI ​​chatbot model. Generates appropriate responses.

[1453] Server: Sends the generated response to the terminal as an HTTP response in real time.

[1454] 3. Display to the user:

[1455] Terminal: Receives responses from the server and displays them in the chat box.

[1456] Specific example

[1457] For example, if a user asks, "Please explain the basics of differentiation," the server's AI model will generate the answer, "Differentiation is a method for measuring the rate of change of a function." Next, if the user delves deeper into the generated answer and asks, "What is the answer to this problem?", the AI ​​model will provide a specific answer to that problem. Finally, when the user submits their own answer, the system evaluates it and provides appropriate feedback, helping the user to learn efficiently.

[1458] The system of this invention dramatically improves the user's learning efficiency by acting as a 24-hour available private tutor.

[1459] The following describes the processing flow.

[1460] Explanation of questions and evaluation of answers

[1461] Step 1: Enter the user's question.

[1462] User: Enter any questions or doubts that arise during the learning process into the text box and click the submit button.

[1463] Terminal: Sends the entered text to the server as an HTTP request.

[1464] Step 2: Server-side solution generation

[1465] Server: Inputs the received questions into a natural language processing model (e.g., BERT or GPT model).

[1466] Server: The AI ​​model analyzes the questions and generates answers and explanations.

[1467] Step 3: Providing answers and explanations

[1468] Server: Sends the generated answer and explanation to the terminal as an HTTP response.

[1469] Terminal: Receives responses from the server and displays the answers and explanations on the screen.

[1470] Problem generation and learning cycle

[1471] Step 1: User input

[1472] User: Enter your answer to the provided question and click the submit button.

[1473] Terminal: Sends the entered answer to the server as an HTTP request.

[1474] Step 2: Evaluation of the answer

[1475] Server: Inputs received answers into the AI ​​model and performs evaluation. Evaluation criteria include accuracy, logic, and clarity.

[1476] Server: Generates detailed feedback and explanations based on the evaluation results.

[1477] Step 3: Provide feedback

[1478] Server: Sends feedback to the terminal as an HTTP response.

[1479] Terminal: Receives feedback from the server and displays it on the screen.

[1480] Step 4: Analyzing User Trends

[1481] Server: Analyzes user learning trends based on evaluation results. This uses data mining techniques and machine learning models.

[1482] Step 5: Generating a new problem

[1483] Server: Based on the analysis results, it identifies user weaknesses and generates new problems.

[1484] Server: Sends a new problem to the terminal as an HTTP response.

[1485] Terminal: Receives new issues from the server and displays them on the screen.

[1486] Chat function

[1487] Step 1: Enter the question

[1488] User: Type your question in the chat box and click the send button.

[1489] Terminal: Sends the entered questions to the server in real time.

[1490] Step 2: Generating a real-time response

[1491] Server: Inputs received questions into the AI ​​chatbot model.

[1492] Server: The AI ​​chatbot analyzes the question and generates a response in real time.

[1493] Step 3: Providing a response

[1494] Server: Sends the generated response to the terminal as an HTTP response in real time.

[1495] Terminal: Receives responses from the server and displays them in the chat box.

[1496] (Example 1)

[1497] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1498] Conventional learning support systems struggled to quickly generate appropriate answers and explanations to users' questions. Furthermore, a lack of evaluation and feedback on the generated answers, and the failure to provide problems tailored to users' learning tendencies, resulted in decreased learning efficiency. Additionally, there was a lack of systems that provided 24 / 7 responses to questions and concerns.

[1499] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1500] In this invention, the server includes means for inputting questions, means for receiving the input questions, means for generating answers and explanations for the received questions using a generative AI model, means for sending and displaying the generated answers and explanations as an HTTP response to the user's terminal, means for evaluating the generated answers, means for analyzing the user's tendencies based on the evaluation results using data mining technology, means for generating the next question using the generative AI model based on the user's tendencies, means for sending and displaying the generated question to the user's terminal as an HTTP response, and means for interacting with the user in real time in a chat format. As a result, the user can obtain quick and appropriate answers and explanations to their questions, receive evaluations and feedback on the answers, improve learning efficiency, and enable learning support that can answer questions anytime, 24 hours a day.

[1501] "Means for entering questions" refers to an interface that allows users to enter questions or doubts that arise during their learning process into a text box.

[1502] "Means for receiving submitted questions" refers to the network connection and protocol used by the server to receive questions sent from the user's terminal.

[1503] "Means for generating answers and explanations using a generative AI model" refers to a system that utilizes an AI model (for example, a natural language processing model) to create answers and explanations based on received questions.

[1504] "Means of sending and displaying as an HTTP response to the user's device" refers to a method of sending the generated answer and explanation as an HTTP response to the user's device and displaying its contents.

[1505] "Means for evaluating generated answers" refers to a system that uses AI models or algorithms to evaluate the accuracy, logic, and other aspects of the answers generated.

[1506] "Methods of analysis using data mining techniques" refers to data mining methods used to analyze users' past answer data and identify learning trends.

[1507] "A means of generating the next problem using an AI model based on user tendencies" refers to a system that uses an AI model to generate new problems tailored to the user's learning tendencies and weaknesses.

[1508] "Means of sending and displaying as an HTTP response to the user's device" refers to a method of sending the generated problem as an HTTP response to the user's device and displaying its contents.

[1509] "A means of interacting with users in real time via chat" refers to a function where users can input questions in real time using a chat box, and the AI ​​responds immediately.

[1510] The learning support system of the present invention includes functions such as generating answers and explanations to user questions, evaluating answers, generating problems, and a chat function. This system consists of the user's terminal (PC, smartphone, tablet, etc.) and a server located in a cloud environment. The following describes specific embodiments for implementing the present invention.

[1511] 1. System Configuration

[1512] The system operates by users interacting with the server using individual terminals. The server is located in a cloud environment and performs data analysis and generation using AI models (e.g., GPT-4).

[1513] 2. User input of questions and generation of answers

[1514] The user enters their questions into a text box and clicks the submit button. The device sends the entered question data to the server as an HTTP request. The server analyzes the received question and generates an answer and explanation using a generative AI model. The generated answer and explanation are then sent to the device as an HTTP response and displayed on the user's screen.

[1515] For example, if a user inputs "Please explain the basics of differentiation," the server's AI model will generate the answer "Differentiation is a method for measuring the rate of change of a function" and send it to the terminal.

[1516] 3. Evaluation and feedback on the answers

[1517] When a user enters an answer to a presented problem and clicks the submit button, the device sends the entered answer data to the server as an HTTP request. The server inputs the received answer into an AI model and evaluates it based on criteria such as accuracy, logic, and clarity. Based on the evaluation results, it generates detailed feedback, sends it to the device as an HTTP response, and displays it on the user's screen.

[1518] For example, when a user submits an answer to a math problem, the server evaluates the answer and sends back feedback such as, "It's correct, but the explanation is insufficient."

[1519] 4. Problem generation and learning cycle

[1520] The server analyzes the user's past answer data using data mining techniques to identify the user's learning tendencies. Based on this analysis, it uses a generative AI model to generate the next problem that addresses the user's weaknesses and sends it to the terminal as an HTTP response. The terminal then displays the new problem to the user.

[1521] For example, if the server determines that a user does not understand "differentiation of a function," it will generate and send a problem such as "Find the derivative of the following function f(x)."

[1522] 5. 24-hour chat function

[1523] The user types a question into the chat box and clicks the send button. The device sends the entered question to the server in real time. The server inputs the received question into an AI chatbot model and generates an appropriate response. The generated response is sent to the device in real time as an HTTP response and displayed in the chat box.

[1524] For example, if a user asks in the chat, "Please explain how to calculate a definite integral," the server's AI chatbot will generate a response such as, "A definite integral is a method for finding the area of ​​a function within an interval," and immediately provide it to the user.

[1525] Example of a prompt

[1526] "Could you explain the basics of differential calculus?"

[1527] "What is differentiation?"

[1528] "Please solve these differential calculus problems."

[1529] "Please evaluate my answer."

[1530] The system of this invention acts as a 24-hour available tutor, dramatically improving the user's learning efficiency.

[1531] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1532] Step 1:

[1533] User's question input

[1534] Users enter any questions or doubts they have during their learning process into a text box and click the submit button.

[1535] Input: Text information entered by the user.

[1536] Specific operation: The browser on the user's device sends the entered question to the server via an asynchronous request (such as Ajax).

[1537] The terminal sends the entered question data to the server as an HTTP request.

[1538] Output: Query data sent to the server as an HTTP request.

[1539] Step 2:

[1540] Receiving questions on the server

[1541] The server receives HTTP requests sent from the terminal.

[1542] Input: An HTTP request containing question data sent from the terminal.

[1543] Specific operation: The server parses the content of the HTTP request and extracts the query data.

[1544] The server analyzes the received question data for processing.

[1545] Output: Analysis results of the questionnaire data.

[1546] Step 3:

[1547] Generating answers and explanations

[1548] The server inputs the analyzed question data into a generating AI model (e.g., GPT-4).

[1549] Input: Analyzed question data.

[1550] Specific operation: The server calls the API of the generation AI model and feeds the question data into the model. The model performs natural language processing and generates appropriate answers and explanations.

[1551] The server receives the output from the generated AI model.

[1552] Output: Generated solution and explanation.

[1553] Step 4:

[1554] Submit your answer and explanation.

[1555] The server sends the generated answer and explanation to the user's terminal as an HTTP response.

[1556] Input: Generated answer and explanation.

[1557] Specific operation: The server formats the answer and explanation into an HTTP response and sends it to the user's terminal.

[1558] The device displays the received answers and explanations on its screen.

[1559] Output: The answer and explanation displayed on the user's device.

[1560] Step 5:

[1561] User's answer input

[1562] The user enters their answer to the presented question and clicks the submit button.

[1563] Input: User-submitted answer data.

[1564] Specific operation: The device's browser sends the answer data to the server as an asynchronous request.

[1565] The terminal sends the entered answer data to the server as an HTTP request.

[1566] Output: HTTP request containing the answer data.

[1567] Step 6:

[1568] Evaluation of the answers

[1569] The server inputs the received answer data into the AI ​​model and performs evaluation.

[1570] Input: Received answer data.

[1571] Specific operation: The server inputs the answer data into an AI model for evaluation and performs evaluation based on criteria such as accuracy and logic. It then generates the evaluation results.

[1572] The server generates feedback based on the evaluation results.

[1573] Output: Evaluation results and generated feedback.

[1574] Step 7:

[1575] Send feedback

[1576] The server sends the feedback to the user's terminal as an HTTP response.

[1577] Input: Evaluation results and feedback.

[1578] Specific operation: The server formats the feedback into an HTTP response and sends it to the user's terminal.

[1579] The device displays the received feedback to the user.

[1580] Output: Feedback displayed on the user's device.

[1581] Step 8:

[1582] User learning trend analysis

[1583] The server analyzes past answer data using data mining techniques.

[1584] Input: User's past answer data.

[1585] Specific operation: The server extracts the user's past answer data from the cloud database and runs data mining algorithms to identify the user's learning tendencies and weaknesses.

[1586] The server saves the analysis results.

[1587] Output: User learning trend data.

[1588] Step 9:

[1589] Generating the next problem

[1590] The server generates the next problem using an AI model based on the user's learning tendencies.

[1591] Input: User learning trend data.

[1592] Specific operation: The server feeds learning trend data into the generating AI model and generates problems tailored to the user's level and weaknesses.

[1593] The server sends the generated problem to the user's terminal as an HTTP response.

[1594] Output: HTTP response containing the generated problem.

[1595] Step 10:

[1596] Display new problems

[1597] The device displays any new issues received to the user.

[1598] Input: HTTP response containing the generated problem.

[1599] Specific operation: The device's browser parses the HTTP response and displays the new problem on the screen.

[1600] Output: New issues displayed on the user's device.

[1601] Step 11:

[1602] Real-time chat function

[1603] The user types their question into the chat box and clicks the send button.

[1604] Input: The question entered in the chat box.

[1605] Specific operation: The device's browser sends chat messages to the server in real time (using WebSocket, etc.).

[1606] The server inputs the received question into the AI ​​chatbot model and generates an appropriate response.

[1607] Output: The generated response.

[1608] Specific operation: The server calls the AI ​​chatbot model's API in real time, analyzes the question, and generates a response.

[1609] The server sends the generated response to the terminal in real time as an HTTP response.

[1610] The device displays the received response in the chat box.

[1611] Output: The response displayed in the chat box.

[1612] (Application Example 1)

[1613] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1614] Modern food delivery services have limited capabilities to respond quickly and accurately to user questions and feedback, making it difficult for users to have a satisfying experience. Furthermore, there is a lack of information regarding ingredients and dishes, as well as appropriate recipe suggestions, forcing users to do their own research. Additionally, recipe suggestions and improvements based on user preferences and habits are not efficiently implemented. As a result, this leads to decreased user satisfaction and lower repeat customer rates.

[1615] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1616] In this invention, the server includes means for the user to input questions about ingredients or dishes, means for receiving the input questions, means for generating answers and explanations for the received questions, means for providing the generated answers and explanations to the user, means for the user to input feedback on delivered products, means for evaluating the input feedback, means for analyzing user trends based on the evaluation results, means for generating the next order or recipe based on the user trends, means for providing the generated recipe to the user, and means for interacting with the user in a chat format. This enables quick and accurate responses to user questions and feedback, and allows for the suggestion and improvement of recipes based on trends.

[1617] A "user" is an individual or legal entity that uses this system.

[1618] "Ingredients" refer to edible foods such as fresh produce and processed foods used for cooking or consumption.

[1619] "Cooking" refers to food that has been prepared by cooking ingredients in an appropriate manner and is ready for consumption.

[1620] "Questions" refer to unresolved questions or points of confusion that users have regarding ingredients or cooking.

[1621] "Feedback" refers to the evaluations and opinions that users provide regarding the delivered goods or services provided.

[1622] "Answer" refers to information that provides answers or explanations to users' questions.

[1623] A "recipe" is a set of instructions provided to the user that lists the ingredients and how to prepare a dish.

[1624] A "server" is a computer system that processes user questions and feedback and generates the necessary answers and recipes.

[1625] A "system" is a collection of means and devices for responding to user questions and feedback, and for performing trend analysis and suggesting recipes.

[1626] "Explanation" refers to a detailed explanation provided to address a problem or question in a way that makes it easy for the user to understand.

[1627] "Evaluation" involves analyzing feedback and making judgments based on appropriate criteria.

[1628] A "tendency" is a specific pattern or characteristic based on user behavior and preferences.

[1629] "Chat format" refers to a method of communication between the user and the system using text or voice.

[1630] The following describes embodiments for specifically implementing the present invention. The following is an example of a food delivery support application.

[1631] System Configuration

[1632] This system consists of user terminals and servers. Users use individual terminals (smartphones, tablets, etc.), and the servers are located in a cloud environment.

[1633] User question input and answer generation

[1634] 1. User input of questions: Users enter their questions or concerns about ingredients and cooking into the app's text box and click the submit button.

[1635] 2. Server-side answer generation: The server inputs the received question into a generating AI model (e.g., a natural language processing model such as GPT-2). The model analyzes the question and generates an answer and explanation. This answer is then provided to the user in real time.

[1636] User feedback and ratings

[1637] 1. Feedback Input: Users enter their feedback about the delivered product into the app's text box and click the submit button.

[1638] 2. Server Evaluation: The server analyzes and evaluates the received feedback. Criteria used for this evaluation include accuracy, appropriateness, and clarity. Based on the evaluation results, detailed feedback and explanations are generated.

[1639] 3. Providing evaluations: Provide users with generated explanations in real time.

[1640] Problem generation and learning cycle

[1641] 1. User Trend Analysis: The server analyzes user trends based on the evaluation results. This includes analyzing past response data.

[1642] 2. Suggestions for the next order and recipe: Based on the analysis results, the server generates and provides orders and recipes that match the user's preferences and tendencies.

[1643] 24-hour chat function

[1644] 1. User Question Input: The user enters their question in the chat box and clicks the send button. The entered question is sent to the server in real time.

[1645] 2. Real-time response: The server inputs the received question into the AI ​​chatbot model and generates an appropriate response. The generated response is provided to the user in real time.

[1646] Hardware and software

[1647] Hardware: Cloud servers, smartphones, tablets

[1648] Software: Flask (web framework), Transformers library, GPT-2 model

[1649] Specific example

[1650] For example, if a user asks, "What's the perfect recipe for a special family dinner on Wednesday?", the server's AI model will generate an answer such as, "Parmesan chicken with garlic mashed potatoes is perfect for a family dinner on Wednesday. Add some leafy greens and lemon dressing, and you've got the perfect course."

[1651] In this way, users can easily resolve their questions and provide feedback on the quality and taste of the delivered products, which can then be used to improve their next order.

[1652] Example of a prompt:

[1653] "What are some great recipes for a special dinner to bring the family together on Wednesday?"

[1654] The system works by having the AI ​​model generate appropriate answers based on user input and providing them to the user via the server.

[1655] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1656] Step 1:

[1657] The user launches the smartphone app, enters their questions about ingredients or cooking into the text box, and clicks the submit button.

[1658] Input: User's question (e.g., "What side dishes would go well with this dish?")

[1659] Output: User input data (HTTP request format)

[1660] Step 2:

[1661] The terminal receives the user's question and sends it to the server as an HTTP request.

[1662] Input: User input data

[1663] Output: HTTP request to the server

[1664] Step 3:

[1665] The server inputs the user's question into a generating AI model, analyzes the data, and generates an answer and explanation.

[1666] Input: User's question included in the HTTP request

[1667] Output: Generated answer and explanation (text data)

[1668] Step 4:

[1669] The server sends the generated answer and explanation to the terminal as an HTTP response.

[1670] Input: Generated answer and explanation

[1671] Output: HTTP response (including answer and explanation)

[1672] Step 5:

[1673] The terminal receives an HTTP response from the server and displays the answer and explanation on the user interface.

[1674] Input: HTTP response

[1675] Output: Answers and explanations displayed on the user screen

[1676] Step 6:

[1677] The user enters feedback about the delivered product in the text box and clicks the submit button.

[1678] Input: User feedback (e.g., "The food tasted very good, but it was a little cold.")

[1679] Output: User input data (HTTP request format)

[1680] Step 7:

[1681] The terminal receives the user's input and sends it to the server as an HTTP request.

[1682] Input: User input data

[1683] Output: HTTP request to the server

[1684] Step 8:

[1685] The server analyzes and evaluates the received feedback, and generates detailed feedback and explanations.

[1686] Input: User feedback

[1687] Output: Evaluation results and explanations (text data)

[1688] Step 9:

[1689] The server sends the generated evaluation results and explanations to the terminal as an HTTP response.

[1690] Input: Evaluation results and explanation

[1691] Output: HTTP response (including evaluation results and explanations)

[1692] Step 10:

[1693] The terminal receives an HTTP response from the server and displays the evaluation results and explanations on the user interface.

[1694] Input: HTTP response

[1695] Output: Evaluation results and explanations displayed on the user screen

[1696] Step 11:

[1697] The server analyzes user trends based on evaluation results and generates subsequent orders and recipes.

[1698] Input: Evaluation result data

[1699] Output: New order suggestions and recipes (text data)

[1700] Step 12:

[1701] The server sends the generated new order suggestions and recipes to the terminal as an HTTP response.

[1702] Input: New order suggestions or recipes

[1703] Output: HTTP response (including new order suggestions and recipes)

[1704] Step 13:

[1705] The terminal receives an HTTP response from the server and displays new order suggestions and recipes on the user interface.

[1706] Input: HTTP response

[1707] Output: New order suggestions and recipes displayed on the user screen

[1708] The above outlines the specific processing steps. In each step, the user, terminal, and server work together to process the data, creating a system that enables the resolution of user questions and the utilization of user feedback.

[1709] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1710] The learning support system of the present invention features the generation of answers and explanations to questions, evaluation of answers, generation of questions, a chat function, and an emotion engine that recognizes the user's emotions. The following describes specific embodiments for implementing the present invention.

[1711] System Configuration

[1712] This system consists of a user's device, a server, and an emotion engine. Users use individual devices (PCs, smartphones, tablets, etc.), and the server is located in a cloud environment. The emotion engine is equipped with AI technology that analyzes emotions from the user's voice, facial expressions, text, etc.

[1713] User question input and answer generation

[1714] 1. User input questions:

[1715] User: Enter any questions or doubts that arise during the learning process into the text box and click the submit button.

[1716] Terminal: Sends the entered text to the server as an HTTP request.

[1717] 2. Generating the solution:

[1718] Server: Inputs the received questions into a natural language processing model (e.g., BERT or GPT model).

[1719] Server: The AI ​​model analyzes the questions and generates answers and explanations.

[1720] 3. Emotion recognition:

[1721] Device: Uses an emotion engine to recognize the user's emotional state. This includes facial expression analysis and voice analysis.

[1722] 4. Adjustments based on answers and emotions:

[1723] Server: Adjusts the content and tone of the answers and explanations based on the recognized emotional state.

[1724] Server: Sends the generated answer and explanation to the terminal as an HTTP response.

[1725] 5. Display to the user:

[1726] Terminal: Receives responses from the server and displays the answers and explanations on the screen.

[1727] Evaluation and feedback on the answers

[1728] 1. Enter your answer:

[1729] User: Enter your answer to the provided question and click the submit button.

[1730] Terminal: Sends the entered answer to the server as an HTTP request.

[1731] 2. Rating:

[1732] Server: Inputs received answers into the AI ​​model and performs evaluation. Evaluation criteria include accuracy, logic, and clarity.

[1733] Server: Generates detailed feedback and explanations based on the evaluation results.

[1734] 3. Emotion recognition:

[1735] Device: Uses an emotion engine to recognize the user's emotional state.

[1736] 4. Feedback and emotional adjustments:

[1737] Server: Adjusts the content and tone of feedback based on the recognized emotional state.

[1738] Server: Sends feedback to the terminal as an HTTP response.

[1739] 5. Display to the user:

[1740] Terminal: Receives feedback from the server and displays it on the screen.

[1741] Problem generation and learning cycle

[1742] 1. User trend analysis:

[1743] Server: Analyzes user learning trends based on evaluation results. This uses data mining techniques and machine learning models.

[1744] 2. Generating new problems:

[1745] Server: Based on the analysis results, it identifies user weaknesses and generates new problems.

[1746] Server: Sends a new problem to the terminal as an HTTP response.

[1747] 3. Emotion recognition:

[1748] Terminal: Uses an emotion engine to recognize the user's emotional state. This allows for adjustments to the difficulty and content of the generated problems.

[1749] 4. Display to the user:

[1750] Terminal: Receives new issues from the server and displays them on the screen.

[1751] Chat function

[1752] 1. Enter your question:

[1753] User: Type your question in the chat box and click the send button.

[1754] Terminal: Sends the entered questions to the server in real time.

[1755] 2. Real-time response generation:

[1756] Server: Inputs received questions into the AI ​​chatbot model.

[1757] Server: The AI ​​chatbot analyzes the question and generates a response in real time.

[1758] 3. Emotion recognition:

[1759] Device: Uses an emotion engine to recognize the user's emotional state, thereby adjusting the response accordingly.

[1760] 4. Providing a response:

[1761] Server: Sends the generated response to the terminal as an HTTP response in real time.

[1762] Terminal: Receives responses from the server and displays them in the chat box.

[1763] Specific example

[1764] For example, if a user asks, "Could you explain the basics of differentiation?", the server's AI model will generate the answer, "Differentiation is a method for measuring the rate of change of a function." Furthermore, if the emotion engine detects that the user is feeling fatigued, the answer will be adjusted to something like, "Differentiation is a method for measuring the rate of change of a function, and it may seem difficult, but let's work through it together."

[1765] The system of this invention optimizes learning by taking into account the user's emotional state, thereby providing more effective learning support. This dramatically improves the user's motivation and learning efficiency.

[1766] The following describes the processing flow.

[1767] User question input and answer generation

[1768] Step 1:

[1769] User: Enter any questions or doubts that arise during the learning process into the text box and click the submit button.

[1770] Step 2:

[1771] Terminal: Sends the entered text to the server as an HTTP request.

[1772] Step 3:

[1773] Server: Inputs the received questions into a natural language processing model (e.g., BERT or GPT model).

[1774] Step 4:

[1775] Server: The AI ​​model analyzes the questions and generates answers and explanations.

[1776] Step 5:

[1777] Device: Captures the user's facial expressions and voice and sends them to the emotion engine.

[1778] Step 6:

[1779] Emotion Engine: Analyzes the user's emotional state based on captured data.

[1780] Step 7:

[1781] Server: Adjusts the content and tone of the answers and explanations based on the recognized emotional state.

[1782] Step 8:

[1783] Server: Sends the adjusted answer and explanation to the terminal as an HTTP response.

[1784] Step 9:

[1785] Terminal: Receives responses from the server and displays the answers and explanations on the screen.

[1786] Evaluation and feedback on the answers

[1787] Step 1:

[1788] User: Enter your answer to the provided question and click the submit button.

[1789] Step 2:

[1790] Terminal: Sends the entered answer to the server as an HTTP request.

[1791] Step 3:

[1792] Server: Inputs received answers into the AI ​​model and performs evaluation.

[1793] Step 4:

[1794] Server: Generates detailed feedback and explanations based on the evaluation results.

[1795] Step 5:

[1796] Device: Captures the user's facial expressions and voice and sends them to the emotion engine.

[1797] Step 6:

[1798] Emotion Engine: Analyzes the user's emotional state based on captured data.

[1799] Step 7:

[1800] Server: Adjusts the content and tone of feedback based on the recognized emotional state.

[1801] Step 8:

[1802] Server: Sends the adjusted feedback to the terminal as an HTTP response.

[1803] Step 9:

[1804] Terminal: Receives feedback from the server and displays it on the screen.

[1805] Problem generation and learning cycle

[1806] Step 1:

[1807] Server: Analyzes user learning trends based on evaluation results.

[1808] Step 2:

[1809] Server: Based on the analysis results, it identifies user weaknesses and generates new problems.

[1810] Step 3:

[1811] Device: Captures the user's facial expressions and voice and sends them to the emotion engine.

[1812] Step 4:

[1813] Emotion Engine: Analyzes the user's emotional state based on captured data.

[1814] Step 5:

[1815] Server: Adjusts the difficulty and content of generated problems based on the recognized emotional state.

[1816] Step 6:

[1817] Server: Sends a new problem to the terminal as an HTTP response.

[1818] Step 7:

[1819] Terminal: Receives new issues from the server and displays them on the screen.

[1820] Chat function

[1821] Step 1:

[1822] User: Type your question in the chat box and click the send button.

[1823] Step 2:

[1824] Terminal: Sends the entered questions to the server in real time.

[1825] Step 3:

[1826] Server: Inputs received questions into the AI ​​chatbot model.

[1827] Step 4:

[1828] Server: The AI ​​chatbot analyzes the question and generates a response in real time.

[1829] Step 5:

[1830] Device: Captures the user's facial expressions and voice and sends them to the emotion engine.

[1831] Step 6:

[1832] Emotion Engine: Analyzes the user's emotional state based on captured data.

[1833] Step 7:

[1834] Server: Adjusts the response based on the recognized emotional state.

[1835] Step 8:

[1836] Server: Sends the adjusted response to the terminal as an HTTP response in real time.

[1837] Step 9:

[1838] Terminal: Receives responses from the server and displays them in the chat box.

[1839] (Example 2)

[1840] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1841] Traditional learning support systems can provide answers and explanations to users' questions, but they lack the ability to consider the user's emotional state. Therefore, they cannot adequately recognize the fatigue or frustration a user might experience during learning and adjust the content of responses and feedback accordingly, resulting in a failure to maximize learning effectiveness. Furthermore, the lack of real-time emotion recognition and feedback based on those results made it difficult to maintain user motivation.

[1842] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1843] In this invention, the server includes means for the user to input questions, means for receiving the input questions, means for generating answers and explanations for the received questions using a generative AI model, means for providing the generated answers and explanations to the user, means for evaluating the generated answers, means for analyzing the user's tendencies based on the evaluation results, means for generating the next question based on the user's tendencies, means for providing the generated question to the user, means for interacting with the user in a chat format, means for recognizing the user's emotional state, and means for adjusting the content of answers and feedback based on the emotional state. This makes it possible to recognize the user's emotional state in real time and adjust the response content and feedback according to that state, thereby increasing the user's motivation to learn and enabling more effective learning support.

[1844] A "user" is an individual or group who uses a learning support system to seek answers to their questions or receive feedback.

[1845] "Questions" refer to the questions or things that users need to understand during the learning process.

[1846] "Input means" refers to interface means for users to input questions and answers into the system, and includes keyboards and touch panels.

[1847] "Receiving means" refers to the means for receiving data transmitted by a user. This includes data receiving systems via a network.

[1848] A "generative AI model" is an artificial intelligence model that uses natural language processing to analyze user input and generate appropriate answers and explanations.

[1849] "Answer" refers to the content of the response to the user's question.

[1850] "Explanation" refers to a detailed explanation provided alongside the answer, serving as supplementary information to help users gain a deeper understanding of their questions.

[1851] "Means of provision" refers to means of displaying or communicating the generated answers and explanations to the user.

[1852] "Evaluation means" refers to methods for evaluating the accuracy, logic, clarity, etc., of answers entered by users, and includes AI models.

[1853] "Evaluation results" refer to data calculated using evaluation methods for user responses.

[1854] A "trend analysis method" is a means of analyzing a user's learning tendencies based on evaluation results.

[1855] The "problem generation method" is a means of generating new learning problems based on the user's learning tendencies and weaknesses, and it uses a generation AI model.

[1856] "Chat dialogue methods" refer to means by which users and systems can interact in real time, and include chatbots that use natural language processing.

[1857] "Emotion recognition means" refers to methods for recognizing a user's emotional state by analyzing their voice, facial expressions, text, etc.

[1858] "Adjustment means" are means for adjusting the content and tone of responses and feedback based on the emotional state obtained by the emotion recognition means.

[1859] The learning support system of the present invention combines the generation of answers and explanations to user questions, evaluation of answers, generation of problems, a chat function, and an emotion recognition function. The following describes in detail the embodiments for specifically implementing the present invention.

[1860] System Configuration

[1861] This system consists of a user's device, a server, and an emotion engine. Users utilize devices such as PCs, smartphones, and tablets. The server is located in a cloud environment and performs the necessary computational processing. The emotion engine incorporates AI technology that analyzes emotions from the user's voice, facial expressions, and text. Specifically, the device collects the user's facial expressions and voice through its camera and microphone and sends them to the server for analysis.

[1862] Hardware and software usage

[1863] The server implements generative AI models (such as natural language processing models like BERT and GPT) to analyze questions and answers submitted by users. It also uses data mining techniques and machine learning models to analyze users' learning tendencies. The terminal is equipped with an emotion recognition engine that analyzes the user's voice and facial expressions in real time to recognize their emotional state. Emotion recognition utilizes facial recognition and voice analysis technologies.

[1864] User question input and answer generation

[1865] The user enters any questions or doubts that arise during the learning process into a text box and clicks the submit button. For example, they might enter, "Please explain the basics of differentiation." The device sends the entered text to the server as an HTTP request. The server inputs the received question into a generating AI model and generates an answer and explanation. An example of a generated answer is, "Differentiation is a method for measuring the rate of change of a function."

[1866] Emotion recognition and regulation

[1867] The emotion engine built into the device analyzes the user's facial expressions, voice, and text to convey emotions. For example, it might recognize "fatigue" from the user's facial expression. Based on the recognized emotional state, the server adjusts the content and tone of the answers and explanations. For instance, it might add an additional message such as, "This might seem difficult, but let's work hard together."

[1868] Evaluation and feedback on the answers

[1869] The user enters an answer to a given question and clicks the submit button. The device sends the entered answer to the server as an HTTP request. The server inputs the received answer into an AI model and evaluates it based on accuracy, logic, and clarity. Based on the evaluation results, it generates detailed feedback and explanations. For example, if the user answers the question "What are the fundamental principles of electromagnetism?" with "The motion of electrons and their field interactions," the server provides feedback such as "This answer is superficially correct, but it would be good to explain it in more detail."

[1870] Problem generation and learning cycle

[1871] The server analyzes the user's learning tendencies based on the evaluation results and identifies the user's weaknesses using data mining techniques and machine learning models. It generates new problems related to the identified weaknesses and sends them to the terminal as an HTTP response. The terminal displays the new problems on its screen, allowing the user to continue learning.

[1872] Chat function

[1873] When a user types a question into the chat box and clicks the send button, the device sends the entered question to the server in real time. The server inputs the received question into an AI chatbot model and generates a response in real time. If emotion recognition determines that the user's emotional state is "frustrated," the response is adjusted to something like, "This may be difficult to understand, but I will explain it carefully." The server then sends the generated response to the device as an HTTP response in real time, and the device displays it in the chat box.

[1874] As described above, the present invention provides a more effective learning experience by providing real-time responses to user questions and answers, adjusting feedback content based on emotion recognition, and generating new problems by analyzing the user's learning tendencies.

[1875] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1876] Step 1:

[1877] The user enters their question into the text box and clicks the submit button.

[1878] Input: The user enters "Please explain the basics of differentiation."

[1879] Operation: The terminal retrieves the entered text.

[1880] Output: The entered question is displayed on the terminal.

[1881] Step 2:

[1882] The terminal sends the entered question to the server as an HTTP request.

[1883] Input: The question entered in the text box.

[1884] Operation: The terminal converts the question into packets and sends them to the server via the network.

[1885] Output: The server receives the question.

[1886] Step 3:

[1887] The server inputs the received questions into a natural language processing model.

[1888] Input: Text data of the question sent from the device.

[1889] Operation: The server processes the question into an appropriate format and inputs it as a prompt to the generating AI model (e.g., GPT-3).

[1890] Output: The generative AI model generates the answer and explanation.

[1891] Step 4:

[1892] The server retrieves the answer and explanation as output from the generated AI model and generates a response to provide to the user.

[1893] Input: Text data of answers and explanations from a generative AI model.

[1894] Operation: Converts the answers and explanations obtained by the server into an HTTP response format.

[1895] Output: Response data is generated.

[1896] Step 5:

[1897] The device uses an emotion engine to recognize the user's emotional state.

[1898] Input: User's facial image and voice data acquired from the camera and microphone.

[1899] Operation: The device uses an emotion engine to analyze the user's facial expressions and voice to detect their emotional state.

[1900] Output: The user's emotional state (e.g., fatigue, irritation) is identified by the emotion engine.

[1901] Step 6:

[1902] The server adjusts the content and tone of the answers and explanations based on the recognized emotional state.

[1903] Input: Emotional state data sent from the device, along with the generated answer and explanation.

[1904] Operation: The server analyzes the emotional state and makes adjustments such as, "Differentiation is a method for measuring the rate of change of a function. It may seem difficult, but let's work through it together."

[1905] Output: Text data of the adjusted answers and explanations.

[1906] Step 7:

[1907] The server sends the adjusted answer to the terminal as an HTTP response.

[1908] Input: Text data of the adjusted answers and explanations.

[1909] Operation: The server converts the data into an HTTP response format and sends it to the terminal over the network.

[1910] Output: The terminal receives a response.

[1911] Step 8:

[1912] The terminal displays the response from the server on the screen.

[1913] Input: Adjusted answer and explanation data received from the server.

[1914] Operation: The device displays the answer and explanation on the screen.

[1915] Output: The user can view the answer and explanation on the screen.

[1916] Step 9:

[1917] The user enters their answer to the provided question and clicks the submit button.

[1918] Input: User's answer (e.g., "The motion of electrons and their field interactions").

[1919] Action: The device retrieves the answer and prepares to send it to the server as an HTTP request.

[1920] Output: The entered answer is ready to be sent by the terminal.

[1921] Step 10:

[1922] The device sends the answer to the server as an HTTP request.

[1923] Input: Text data of the answer entered by the user.

[1924] Operation: The terminal converts the answer into packets and sends them to the server over the network.

[1925] Output: The server receives the answer.

[1926] Step 11:

[1927] The server inputs the answer into the AI ​​model and performs evaluation.

[1928] Input: Text data of the answer sent from the device.

[1929] Operation: The server inputs the answer into the AI ​​model and performs analysis based on evaluation criteria.

[1930] Output: Evaluation results are generated.

[1931] Step 12:

[1932] The server generates detailed feedback and explanations based on the evaluation results.

[1933] Input: Evaluation result data generated by an AI model.

[1934] Operation: The server generates feedback and explanations for the user based on the evaluation results.

[1935] Output: Text data of feedback and commentary.

[1936] Step 13:

[1937] The device recognizes the user's emotional state.

[1938] Input: User's facial image and voice data acquired from the camera and microphone.

[1939] Operation: The device uses an emotion engine to analyze the user's facial expressions and voice.

[1940] Output: The user's emotional state (e.g., satisfied, dissatisfied) is identified.

[1941] Step 14:

[1942] The server adjusts the content and tone of the feedback based on the emotional state.

[1943] Input: Emotional state data sent from the device, along with generated feedback and commentary.

[1944] Operation: The server adjusts the tone of feedback and commentary based on the emotional state.

[1945] Output: Text data of adjusted feedback and commentary.

[1946] Step 15:

[1947] The server sends the adjusted feedback to the terminal as an HTTP response.

[1948] Input: Text data of adjusted feedback and commentary.

[1949] Operation: The server converts the data into an HTTP response format and sends it to the terminal over the network.

[1950] Output: The terminal receives a response.

[1951] Step 16:

[1952] The terminal displays feedback and explanations from the server on the screen.

[1953] Input: Feedback and commentary data received from the server.

[1954] Action: The device displays feedback and explanations on the screen.

[1955] Output: The user will be able to view feedback and explanations on the screen.

[1956] The detailed processing flow for each step described above allows users to input questions, receive answers, get feedback on those answers, and even have their feedback adjusted based on their emotions, thereby maximizing the effectiveness of the learning support system.

[1957] (Application Example 2)

[1958] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1959] Conventional learning support systems provide functions for answering user questions, evaluating their performance, and generating new problems. However, they fail to take into account the user's emotional state, resulting in insufficient improvement in learning effectiveness. As a result, users may not receive appropriate support when they feel fatigued or stressed, potentially leading to a decrease in their motivation to learn. Furthermore, because answers and feedback are provided regardless of the user's emotions, there is a need to address the issue of poor satisfaction and comprehension during the learning process.

[1960] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for inputting questions, means for receiving the input questions, means for generating answers and explanations for the received questions, means for providing the generated answers and explanations to the user, means for recognizing the user's emotions, means for adjusting the tone and content of the answers and explanations based on the recognized emotions, means for evaluating the generated answers, means for analyzing the user's tendencies based on the evaluation results, means for generating the next question based on the user's tendencies, means for adjusting the difficulty and content of the questions considering the user's emotional state before providing the questions, means for providing the questions to the user, means for interacting with the user in a chat format, and means for recognizing the user's emotional state during the chat and adjusting the response content based on the recognized emotional state. This enables appropriate learning support that takes into account the user's emotional state, and is expected to improve learning effectiveness and user satisfaction.

[1961] "Means for inputting questions" refers to an interface for users to input questions or doubts that arise during their learning process.

[1962] "Means for receiving entered questions" refers to a mechanism for the system to receive questions entered by the user.

[1963] "Means for generating answers and explanations to received questions" refers to a set of algorithms and programs for generating appropriate answers and additional explanations based on received questions.

[1964] "Means of providing users with generated answers and explanations" refers to a function for displaying generated answers and explanations to users.

[1965] "Means of recognizing user emotions" refers to technologies and devices that analyze a user's voice, facial expressions, text, etc., to recognize their emotional state.

[1966] "Means of adjusting the tone and content of answers and explanations based on recognized emotions" refers to algorithms and functions that adjust the wording and level of detail of answers and explanations according to the recognized emotional state of the user.

[1967] "Means for evaluating generated answers" refers to a mechanism for evaluating the accuracy and validity of generated answers.

[1968] "Means for analyzing user trends based on evaluation results" refers to algorithms for analyzing users' learning patterns and proficiency levels based on the evaluation results of their answers.

[1969] "Means for generating the next problem based on user trends" refers to a system that analyzes the user's learning trends and generates the next problem to be provided based on those trends.

[1970] "Means of adjusting the difficulty and content of generated problems based on the user's emotional state before providing them" refers to a function that appropriately adjusts the difficulty and content of generated problems based on the user's current emotional state.

[1971] "Means of providing generated problems to users" refers to a function for displaying adjusted problems to users.

[1972] "Means of interacting with users in a chat format" refers to chat interfaces and related technologies that allow users to interact with the system in real time.

[1973] "Means for recognizing a user's emotional state during a chat and adjusting the response based on that emotional state" refers to algorithms and technologies for recognizing a user's emotions during a chat and dynamically adjusting the content of the response accordingly.

[1974] The learning support system of the present invention, in addition to generating answers to user questions, providing evaluations, generating problems, and offering a chat function, improves learning effectiveness by recognizing the user's emotions and adjusting learning content and responses based on those emotions. This system consists of the user's terminal, a server, and an emotion engine.

[1975] System Configuration

[1976] A terminal is an information processing device (smartphone, tablet, or personal computer) that receives user input and communicates with a server. The terminal is equipped with a camera and microphone to collect data for emotion recognition.

[1977] Server: Located in a cloud environment, it runs various AI models (e.g., natural language processing models and emotion recognition models). Its main roles are generating answers and explanations, evaluation, problem generation, and adjustments based on the user's emotional state.

[1978] Emotion Engine: Analyzes the user's emotional state from their voice, facial expressions, text, etc., and adjusts responses and answers accordingly.

[1979] User question input and answer generation

[1980] 1. Users enter any questions or doubts that arise during the learning process into the text box on their device and submit them.

[1981] 2. The terminal sends the entered text to the server.

[1982] 3. The server inputs the received questions into a natural language processing model (e.g., BERT or GPT model) and generates answers and explanations.

[1983] 4. The device analyzes the user's emotions using their facial expressions and voice, and sends that data to the server.

[1984] 5. The server adjusts the content and tone of the answers and explanations based on the recognized emotional state and sends them to the terminal.

[1985] 6. The device displays the adjusted answers and explanations on the screen.

[1986] Evaluation and feedback on the answers

[1987] 1. The user enters and submits their answer to the provided question.

[1988] 2. The terminal sends the entered answer to the server.

[1989] 3. The server inputs the answer into the AI ​​model and performs an evaluation. Based on the evaluation results, it generates detailed feedback and explanations.

[1990] 4. The device uses an emotion engine to recognize the user's emotional state and sends that data to the server.

[1991] 5. The server adjusts the content and tone of the feedback based on the emotional state and sends it to the device.

[1992] 6. The device displays the adjusted feedback on the screen.

[1993] Problem generation and learning cycle

[1994] 1. Based on the evaluation results, the server analyzes the user's learning tendencies and generates the next problem using data mining techniques and machine learning models.

[1995] 2. The device recognizes the user's emotional state before providing a problem and sends that data to the server.

[1996] 3. The server adjusts the difficulty and content of the problem, taking into account the emotional state, and sends it to the terminal.

[1997] 4. The device displays the adjusted issues on the screen.

[1998] Chat function

[1999] 1. The user types their question into the chat box and sends it.

[2000] 2. The terminal sends the question to the server in real time.

[2001] 3. The server uses an AI chatbot model to analyze the question and generate a response in real time.

[2002] 4. The device uses an emotion engine to recognize the user's emotional state and sends that data to the server.

[2003] 5. The server adjusts the response based on the emotional state and sends it to the terminal.

[2004] 6. The device displays the adjusted response in the chat box.

[2005] Specific examples and prompt statements

[2006] For example, if a user asks, "Could you explain the basics of differentiation?", the server will generate the answer, "Differentiation is a method for measuring the rate of change of a function." If the emotion engine detects that the user is fatigued, the answer will be adjusted to, "Differentiation is a method for measuring the rate of change of a function, and it may seem difficult, but let's work through it together."

[2007] Specific examples of prompt statements are as follows:

[2008] "Please explain the basics of differentiation. The user seems a little tired. Please provide the answer in a gentle tone."

[2009] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[2010] Step 1:

[2011] Users enter any questions or doubts that arise during the learning process into the text box on their device and submit them.

[2012] Input: Text entered by the user

[2013] Output: HTTP request from terminal to server

[2014] Specific operation: When the user enters a question and clicks the "Send" button, the terminal sends the entered text data to the server.

[2015] Step 2:

[2016] The terminal sends the entered text to the server.

[2017] Input: Text data of the question

[2018] Output: Text data sent to the server

[2019] Specific action: The device sends text data containing the user's question to the server as an HTTP request.

[2020] Step 3:

[2021] The server inputs the received questions into a natural language processing model (e.g., BERT or GPT model) and generates answers and explanations.

[2022] Input: Text data of the question

[2023] Output: Generated solution and explanation data

[2024] Specific operation: The server inputs the text data of the question into a natural language processing model, and uses a generative AI model to generate the answer and explanation.

[2025] Step 4:

[2026] The device analyzes the user's emotions using their facial expressions and voice, and sends that data to a server.

[2027] Input: User's facial expression data, voice data

[2028] Output: Emotional state data

[2029] Specific operation: The device uses facial expressions and voice data collected by the camera and microphone to allow the emotion engine to recognize the emotional state and send that data to the server.

[2030] Step 5:

[2031] The server adjusts the content and tone of the answers and explanations based on the recognized emotional state, and then sends them to the terminal.

[2032] Input: Emotional state data, generated answers and explanatory data

[2033] Output: Adjusted answer and explanation data

[2034] Specific operation: The server uses the emotion data recognized by the emotion engine to apply an algorithm that adjusts the tone and level of detail of the answers and explanations, generates the adjusted answer and explanation data, and sends it to the terminal.

[2035] Step 6:

[2036] The device displays the adjusted answers and explanations on the screen.

[2037] Input: Adjusted answer and explanation data

[2038] Output: Adjusted answers and explanations displayed on the screen

[2039] Specific operation: The device displays the adjusted answer and explanation data it has received on the screen in a user-friendly format.

[2040] Step 7:

[2041] Users enter and submit their answers to the provided questions.

[2042] Input: User-entered answer data

[2043] Output: Answer data sent as an HTTP request from the terminal to the server.

[2044] Specific operation: When the user enters the answer to the problem and clicks the "Submit" button, the device sends the answer data to the server.

[2045] Step 8:

[2046] The terminal sends the entered answer to the server.

[2047] Input: Answer data

[2048] Output: Answer data sent to the server

[2049] Specific operation: The device sends the user's answer data to the server as an HTTP request.

[2050] Step 9:

[2051] The server inputs the answer into the AI ​​model and performs an evaluation. Based on the evaluation results, it generates detailed feedback and explanations.

[2052] Input: Answer data

[2053] Output: Evaluation results, feedback, and explanatory data

[2054] Specific operation: The server inputs the answer data into the AI ​​model, applies an evaluation algorithm to generate evaluation results, and then generates feedback and explanatory data based on those results.

[2055] Step 10:

[2056] The device uses an emotion engine to recognize the user's emotional state and sends that data to the server.

[2057] Input: User's facial expression data, voice data

[2058] Output: Emotional state data

[2059] Specific operation: The device uses facial expressions and voice data collected by the camera and microphone to allow the emotion engine to recognize the emotional state and send that data to the server.

[2060] Step 11:

[2061] The server adjusts the content and tone of the feedback based on the emotional state and sends it to the device.

[2062] Input: Emotional state data, evaluation results, and feedback data

[2063] Output: Adjusted feedback data

[2064] Specific operation: The server uses the emotion data recognized by the emotion engine to apply an algorithm that adjusts the tone and level of detail of the feedback, generates the adjusted feedback data, and sends it to the terminal.

[2065] Step 12:

[2066] The device displays adjusted feedback on the screen.

[2067] Input: Adjusted feedback data

[2068] Output: Adjusted feedback displayed on the screen

[2069] Specific operation: The device displays the adjusted feedback data it has received on the screen in a user-friendly format.

[2070] Step 13:

[2071] Based on the evaluation results, the server analyzes the user's learning tendencies and generates the next problem using data mining techniques and machine learning models.

[2072] Input: Evaluation result data

[2073] Output: Newly generated problem data

[2074] Specific operation: The server uses the evaluation result data to analyze the user's learning tendencies using data mining techniques and machine learning models, and then generates the next problem to be presented.

[2075] Step 14:

[2076] The device recognizes the user's emotional state before providing a problem and sends that data to the server.

[2077] Input: User's facial expression data, voice data

[2078] Output: Emotional state data

[2079] Specific operation: The device uses facial expressions and voice data collected by the camera and microphone to allow the emotion engine to recognize the emotional state and send that data to the server.

[2080] Step 15:

[2081] The server adjusts the difficulty and content of the problem, taking into account the emotional state, and then sends it to the terminal.

[2082] Input: Emotional state data, new problem data

[2083] Output: Adjusted new problem data

[2084] Specific operation: The server uses the emotion data recognized by the emotion engine to apply an algorithm that adjusts the difficulty and content of the problem, generates new adjusted problem data, and sends it to the terminal.

[2085] Step 16:

[2086] The device displays the adjusted issues on the screen.

[2087] Input: Adjusted new problem data

[2088] Output: Adjusted issues displayed on the screen

[2089] Specific operation: The device displays the adjusted problem data it has received on the screen in a user-friendly format.

[2090] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[2091] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[2092] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[2093] [Fourth Embodiment]

[2094] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[2095] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[2096] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[2097] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[2098] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[2099] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[2100] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[2101] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[2102] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[2103] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[2104] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[2105] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[2106] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[2107] The learning support system of the present invention includes generating answers and explanations to questions, evaluating answers, generating problems, and a chat function. The following describes specific embodiments for implementing the present invention.

[2108] System Configuration

[2109] This system consists of user terminals and servers. Users use individual terminals (PCs, smartphones, tablets, etc.), and the servers are located in a cloud environment.

[2110] User question input and answer generation

[2111] 1. User input questions:

[2112] User: Enter any questions or doubts that arise during the learning process into the text box and click the submit button.

[2113] Terminal: Sends the entered question to the server as an HTTP request.

[2114] 2. Generating the solution:

[2115] Server: Inputs the received questions into an AI model (for example, a natural language processing model such as BERT or GPT).

[2116] Server: The AI ​​model analyzes the question and generates the answer and explanation.

[2117] Server: Sends the generated answer and explanation to the terminal as an HTTP response.

[2118] 3. Display to the user:

[2119] Terminal: Receives responses from the server and displays the answers and explanations on the screen.

[2120] Evaluation and feedback on the answers

[2121] 1. Enter your answer:

[2122] User: Enter your answer to the submitted question and click the submit button.

[2123] Terminal: Sends the entered answer to the server as an HTTP request.

[2124] 2. Rating:

[2125] Server: Inputs received answers into the AI ​​model and performs evaluation. Evaluation criteria include accuracy, logic, and clarity.

[2126] Server: Generates detailed feedback and explanations based on the evaluation results.

[2127] Server: Sends feedback to the terminal as an HTTP response.

[2128] 3. Display to the user:

[2129] Terminal: Receives feedback from the server and displays it on the screen.

[2130] Problem generation and learning cycle

[2131] 1. User trend analysis:

[2132] Server: Analyzes user learning trends based on evaluation results. This is done by analyzing past answer data using data mining techniques.

[2133] 2. Problem generation:

[2134] Server: Based on the analysis results, it identifies the user's weaknesses and uses an AI model to generate the next problem. The generated problem corresponds to the user's learning level and weaknesses.

[2135] Server: Sends a new problem to the terminal as an HTTP response.

[2136] 3. Display to the user:

[2137] Terminal: Receives new issues from the server and displays them on the screen.

[2138] 24-hour chat function

[2139] 1. User question input:

[2140] User: Type your question in the chat box and click the send button.

[2141] Terminal: Sends the entered questions to the server in real time.

[2142] 2. Real-time response:

[2143] Server: Inputs received questions into the AI ​​chatbot model. Generates appropriate responses.

[2144] Server: Sends the generated response to the terminal as an HTTP response in real time.

[2145] 3. Display to the user:

[2146] Terminal: Receives responses from the server and displays them in the chat box.

[2147] Specific example

[2148] For example, if a user asks, "Please explain the basics of differentiation," the server's AI model will generate the answer, "Differentiation is a method for measuring the rate of change of a function." Next, if the user delves deeper into the generated answer and asks, "What is the answer to this problem?", the AI ​​model will provide a specific answer to that problem. Finally, when the user submits their own answer, the system evaluates it and provides appropriate feedback, helping the user to learn efficiently.

[2149] The system of this invention dramatically improves the user's learning efficiency by acting as a 24-hour available private tutor.

[2150] The following describes the processing flow.

[2151] Explanation of questions and evaluation of answers

[2152] Step 1: Enter the user's question.

[2153] User: Enter any questions or doubts that arise during the learning process into the text box and click the submit button.

[2154] Terminal: Sends the entered text to the server as an HTTP request.

[2155] Step 2: Server-side solution generation

[2156] Server: Inputs the received questions into a natural language processing model (e.g., BERT or GPT model).

[2157] Server: The AI ​​model analyzes the questions and generates answers and explanations.

[2158] Step 3: Providing answers and explanations

[2159] Server: Sends the generated answer and explanation to the terminal as an HTTP response.

[2160] Terminal: Receives responses from the server and displays the answers and explanations on the screen.

[2161] Problem generation and learning cycle

[2162] Step 1: User input

[2163] User: Enter your answer to the provided question and click the submit button.

[2164] Terminal: Sends the entered answer to the server as an HTTP request.

[2165] Step 2: Evaluation of the answer

[2166] Server: Inputs received answers into the AI ​​model and performs evaluation. Evaluation criteria include accuracy, logic, and clarity.

[2167] Server: Generates detailed feedback and explanations based on the evaluation results.

[2168] Step 3: Provide feedback

[2169] Server: Sends feedback to the terminal as an HTTP response.

[2170] Terminal: Receives feedback from the server and displays it on the screen.

[2171] Step 4: Analyzing User Trends

[2172] Server: Analyzes user learning trends based on evaluation results. This uses data mining techniques and machine learning models.

[2173] Step 5: Generating a new problem

[2174] Server: Based on the analysis results, it identifies user weaknesses and generates new problems.

[2175] Server: Sends a new problem to the terminal as an HTTP response.

[2176] Terminal: Receives new issues from the server and displays them on the screen.

[2177] Chat function

[2178] Step 1: Enter the question

[2179] User: Type your question in the chat box and click the send button.

[2180] Terminal: Sends the entered questions to the server in real time.

[2181] Step 2: Generating a real-time response

[2182] Server: Inputs received questions into the AI ​​chatbot model.

[2183] Server: The AI ​​chatbot analyzes the question and generates a response in real time.

[2184] Step 3: Providing a response

[2185] Server: Sends the generated response to the terminal as an HTTP response in real time.

[2186] Terminal: Receives responses from the server and displays them in the chat box.

[2187] (Example 1)

[2188] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[2189] Conventional learning support systems struggled to quickly generate appropriate answers and explanations to users' questions. Furthermore, a lack of evaluation and feedback on the generated answers, and the failure to provide problems tailored to users' learning tendencies, resulted in decreased learning efficiency. Additionally, there was a lack of systems that provided 24 / 7 responses to questions and concerns.

[2190] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[2191] In this invention, the server includes means for inputting questions, means for receiving the input questions, means for generating answers and explanations for the received questions using a generative AI model, means for sending and displaying the generated answers and explanations as an HTTP response to the user's terminal, means for evaluating the generated answers, means for analyzing the user's tendencies based on the evaluation results using data mining technology, means for generating the next question using the generative AI model based on the user's tendencies, means for sending and displaying the generated question to the user's terminal as an HTTP response, and means for interacting with the user in real time in a chat format. As a result, the user can obtain quick and appropriate answers and explanations to their questions, receive evaluations and feedback on the answers, improve learning efficiency, and enable learning support that can answer questions anytime, 24 hours a day.

[2192] "Means for entering questions" refers to an interface that allows users to enter questions or doubts that arise during their learning process into a text box.

[2193] "Means for receiving submitted questions" refers to the network connection and protocol used by the server to receive questions sent from the user's terminal.

[2194] "Means for generating answers and explanations using a generative AI model" refers to a system that utilizes an AI model (for example, a natural language processing model) to create answers and explanations based on received questions.

[2195] "Means of sending and displaying as an HTTP response to the user's device" refers to a method of sending the generated answer and explanation as an HTTP response to the user's device and displaying its contents.

[2196] "Means for evaluating generated answers" refers to a system that uses AI models or algorithms to evaluate the accuracy, logic, and other aspects of the answers generated.

[2197] "Methods of analysis using data mining techniques" refers to data mining methods used to analyze users' past answer data and identify learning trends.

[2198] "A means of generating the next problem using an AI model based on user tendencies" refers to a system that uses an AI model to generate new problems tailored to the user's learning tendencies and weaknesses.

[2199] "Means of sending and displaying as an HTTP response to the user's device" refers to a method of sending the generated problem as an HTTP response to the user's device and displaying its contents.

[2200] "A means of interacting with users in real time via chat" refers to a function where users can input questions in real time using a chat box, and the AI ​​responds immediately.

[2201] The learning support system of the present invention includes functions such as generating answers and explanations to user questions, evaluating answers, generating problems, and a chat function. This system consists of the user's terminal (PC, smartphone, tablet, etc.) and a server located in a cloud environment. The following describes specific embodiments for implementing the present invention.

[2202] 1. System Configuration

[2203] The system operates by users interacting with the server using individual terminals. The server is located in a cloud environment and performs data analysis and generation using AI models (e.g., GPT-4).

[2204] 2. User input of questions and generation of answers

[2205] The user enters their questions into a text box and clicks the submit button. The device sends the entered question data to the server as an HTTP request. The server analyzes the received question and generates an answer and explanation using a generative AI model. The generated answer and explanation are then sent to the device as an HTTP response and displayed on the user's screen.

[2206] For example, if a user inputs "Please explain the basics of differentiation," the server's AI model will generate the answer "Differentiation is a method for measuring the rate of change of a function" and send it to the terminal.

[2207] 3. Evaluation and feedback on the answers

[2208] When a user enters an answer to a presented problem and clicks the submit button, the device sends the entered answer data to the server as an HTTP request. The server inputs the received answer into an AI model and evaluates it based on criteria such as accuracy, logic, and clarity. Based on the evaluation results, it generates detailed feedback, sends it to the device as an HTTP response, and displays it on the user's screen.

[2209] For example, when a user submits an answer to a math problem, the server evaluates the answer and sends back feedback such as, "It's correct, but the explanation is insufficient."

[2210] 4. Problem generation and learning cycle

[2211] The server analyzes the user's past answer data using data mining techniques to identify the user's learning tendencies. Based on this analysis, it uses a generative AI model to generate the next problem that addresses the user's weaknesses and sends it to the terminal as an HTTP response. The terminal then displays the new problem to the user.

[2212] For example, if the server determines that a user does not understand "differentiation of a function," it will generate and send a problem such as "Find the derivative of the following function f(x)."

[2213] 5. 24-hour chat function

[2214] The user types a question into the chat box and clicks the send button. The device sends the entered question to the server in real time. The server inputs the received question into an AI chatbot model and generates an appropriate response. The generated response is sent to the device in real time as an HTTP response and displayed in the chat box.

[2215] For example, if a user asks in the chat, "Please explain how to calculate a definite integral," the server's AI chatbot will generate a response such as, "A definite integral is a method for finding the area of ​​a function within an interval," and immediately provide it to the user.

[2216] Example of a prompt

[2217] "Could you explain the basics of differential calculus?"

[2218] "What is differentiation?"

[2219] "Please solve these differential calculus problems."

[2220] "Please evaluate my answer."

[2221] The system of this invention acts as a 24-hour available tutor, dramatically improving the user's learning efficiency.

[2222] The flow of the specific processing in Example 1 will be explained using Figure 11.

[2223] Step 1:

[2224] User's question input

[2225] Users enter any questions or doubts they have during their learning process into a text box and click the submit button.

[2226] Input: Text information entered by the user.

[2227] Specific operation: The browser on the user's device sends the entered question to the server via an asynchronous request (such as Ajax).

[2228] The terminal sends the entered question data to the server as an HTTP request.

[2229] Output: Query data sent to the server as an HTTP request.

[2230] Step 2:

[2231] Receiving questions on the server

[2232] The server receives HTTP requests sent from the terminal.

[2233] Input: An HTTP request containing question data sent from the terminal.

[2234] Specific operation: The server parses the content of the HTTP request and extracts the query data.

[2235] The server analyzes the received question data for processing.

[2236] Output: Analysis results of the questionnaire data.

[2237] Step 3:

[2238] Generating answers and explanations

[2239] The server inputs the analyzed question data into a generating AI model (e.g., GPT-4).

[2240] Input: Analyzed question data.

[2241] Specific operation: The server calls the API of the generation AI model and feeds the question data into the model. The model performs natural language processing and generates appropriate answers and explanations.

[2242] The server receives the output from the generated AI model.

[2243] Output: Generated solution and explanation.

[2244] Step 4:

[2245] Submit your answer and explanation.

[2246] The server sends the generated answer and explanation to the user's terminal as an HTTP response.

[2247] Input: Generated answer and explanation.

[2248] Specific operation: The server formats the answer and explanation into an HTTP response and sends it to the user's terminal.

[2249] The device displays the received answers and explanations on its screen.

[2250] Output: The answer and explanation displayed on the user's device.

[2251] Step 5:

[2252] User's answer input

[2253] The user enters their answer to the presented question and clicks the submit button.

[2254] Input: User-submitted answer data.

[2255] Specific operation: The device's browser sends the answer data to the server as an asynchronous request.

[2256] The terminal sends the entered answer data to the server as an HTTP request.

[2257] Output: HTTP request containing the answer data.

[2258] Step 6:

[2259] Evaluation of the answers

[2260] The server inputs the received answer data into the AI ​​model and performs evaluation.

[2261] Input: Received answer data.

[2262] Specific operation: The server inputs the answer data into an AI model for evaluation and performs evaluation based on criteria such as accuracy and logic. It then generates the evaluation results.

[2263] The server generates feedback based on the evaluation results.

[2264] Output: Evaluation results and generated feedback.

[2265] Step 7:

[2266] Send feedback

[2267] The server sends the feedback to the user's terminal as an HTTP response.

[2268] Input: Evaluation results and feedback.

[2269] Specific operation: The server formats the feedback into an HTTP response and sends it to the user's terminal.

[2270] The device displays the received feedback to the user.

[2271] Output: Feedback displayed on the user's device.

[2272] Step 8:

[2273] User learning trend analysis

[2274] The server analyzes past answer data using data mining techniques.

[2275] Input: User's past answer data.

[2276] Specific operation: The server extracts the user's past answer data from the cloud database and runs data mining algorithms to identify the user's learning tendencies and weaknesses.

[2277] The server saves the analysis results.

[2278] Output: User learning trend data.

[2279] Step 9:

[2280] Generating the next problem

[2281] The server generates the next problem using an AI model based on the user's learning tendencies.

[2282] Input: User learning trend data.

[2283] Specific operation: The server feeds learning trend data into the generating AI model and generates problems tailored to the user's level and weaknesses.

[2284] The server sends the generated problem to the user's terminal as an HTTP response.

[2285] Output: HTTP response containing the generated problem.

[2286] Step 10:

[2287] Display new problems

[2288] The device displays any new issues received to the user.

[2289] Input: HTTP response containing the generated problem.

[2290] Specific operation: The device's browser parses the HTTP response and displays the new problem on the screen.

[2291] Output: New issues displayed on the user's device.

[2292] Step 11:

[2293] Real-time chat function

[2294] The user types their question into the chat box and clicks the send button.

[2295] Input: The question entered in the chat box.

[2296] Specific operation: The device's browser sends chat messages to the server in real time (using WebSocket, etc.).

[2297] The server inputs the received question into the AI ​​chatbot model and generates an appropriate response.

[2298] Output: The generated response.

[2299] Specific operation: The server calls the AI ​​chatbot model's API in real time, analyzes the question, and generates a response.

[2300] The server sends the generated response to the terminal in real time as an HTTP response.

[2301] The device displays the received response in the chat box.

[2302] Output: The response displayed in the chat box.

[2303] (Application Example 1)

[2304] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[2305] Modern food delivery services have limited capabilities to respond quickly and accurately to user questions and feedback, making it difficult for users to have a satisfying experience. Furthermore, there is a lack of information regarding ingredients and dishes, as well as appropriate recipe suggestions, forcing users to do their own research. Additionally, recipe suggestions and improvements based on user preferences and habits are not efficiently implemented. As a result, this leads to decreased user satisfaction and lower repeat customer rates.

[2306] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[2307] In this invention, the server includes means for the user to input questions about ingredients or dishes, means for receiving the input questions, means for generating answers and explanations for the received questions, means for providing the generated answers and explanations to the user, means for the user to input feedback on delivered products, means for evaluating the input feedback, means for analyzing user trends based on the evaluation results, means for generating the next order or recipe based on the user trends, means for providing the generated recipe to the user, and means for interacting with the user in a chat format. This enables quick and accurate responses to user questions and feedback, and allows for the suggestion and improvement of recipes based on trends.

[2308] A "user" is an individual or legal entity that uses this system.

[2309] "Ingredients" refer to edible foods such as fresh produce and processed foods used for cooking or consumption.

[2310] "Cooking" refers to food that has been prepared by cooking ingredients in an appropriate manner and is ready for consumption.

[2311] "Questions" refer to unresolved questions or points of confusion that users have regarding ingredients or cooking.

[2312] "Feedback" refers to the evaluations and opinions that users provide regarding the delivered goods or services provided.

[2313] "Answer" refers to information that provides answers or explanations to users' questions.

[2314] A "recipe" is a set of instructions provided to the user that lists the ingredients and how to prepare a dish.

[2315] A "server" is a computer system that processes user questions and feedback and generates the necessary answers and recipes.

[2316] A "system" is a collection of means and devices for responding to user questions and feedback, and for performing trend analysis and suggesting recipes.

[2317] "Explanation" refers to a detailed explanation provided to address a problem or question in a way that makes it easy for the user to understand.

[2318] "Evaluation" involves analyzing feedback and making judgments based on appropriate criteria.

[2319] A "tendency" is a specific pattern or characteristic based on user behavior and preferences.

[2320] "Chat format" refers to a method of communication between the user and the system using text or voice.

[2321] The following describes embodiments for specifically implementing the present invention. The following is an example of a food delivery support application.

[2322] System Configuration

[2323] This system consists of user terminals and servers. Users use individual terminals (smartphones, tablets, etc.), and the servers are located in a cloud environment.

[2324] User question input and answer generation

[2325] 1. User input of questions: Users enter their questions or concerns about ingredients and cooking into the app's text box and click the submit button.

[2326] 2. Server-side answer generation: The server inputs the received question into a generating AI model (e.g., a natural language processing model such as GPT-2). The model analyzes the question and generates an answer and explanation. This answer is then provided to the user in real time.

[2327] User feedback and ratings

[2328] 1. Feedback Input: Users enter their feedback about the delivered product into the app's text box and click the submit button.

[2329] 2. Server Evaluation: The server analyzes and evaluates the received feedback. Criteria used for this evaluation include accuracy, appropriateness, and clarity. Based on the evaluation results, detailed feedback and explanations are generated.

[2330] 3. Providing evaluations: Provide users with generated explanations in real time.

[2331] Problem generation and learning cycle

[2332] 1. User Trend Analysis: The server analyzes user trends based on the evaluation results. This includes analyzing past response data.

[2333] 2. Suggestions for the next order and recipe: Based on the analysis results, the server generates and provides orders and recipes that match the user's preferences and tendencies.

[2334] 24-hour chat function

[2335] 1. User Question Input: The user enters their question in the chat box and clicks the send button. The entered question is sent to the server in real time.

[2336] 2. Real-time response: The server inputs the received question into the AI ​​chatbot model and generates an appropriate response. The generated response is provided to the user in real time.

[2337] Hardware and software

[2338] Hardware: Cloud servers, smartphones, tablets

[2339] Software: Flask (web framework), Transformers library, GPT-2 model

[2340] Specific example

[2341] For example, if a user asks, "What's the perfect recipe for a special family dinner on Wednesday?", the server's AI model will generate an answer such as, "Parmesan chicken with garlic mashed potatoes is perfect for a family dinner on Wednesday. Add some leafy greens and lemon dressing, and you've got the perfect course."

[2342] In this way, users can easily resolve their questions and provide feedback on the quality and taste of the delivered products, which can then be used to improve their next order.

[2343] Example of a prompt:

[2344] "What are some great recipes for a special dinner to bring the family together on Wednesday?"

[2345] The system works by having the AI ​​model generate appropriate answers based on user input and providing them to the user via the server.

[2346] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[2347] Step 1:

[2348] The user launches the smartphone app, enters their questions about ingredients or cooking into the text box, and clicks the submit button.

[2349] Input: User's question (e.g., "What side dishes would go well with this dish?")

[2350] Output: User input data (HTTP request format)

[2351] Step 2:

[2352] The terminal receives the user's question and sends it to the server as an HTTP request.

[2353] Input: User input data

[2354] Output: HTTP request to the server

[2355] Step 3:

[2356] The server inputs the user's question into a generating AI model, analyzes the data, and generates an answer and explanation.

[2357] Input: User's question included in the HTTP request

[2358] Output: Generated answer and explanation (text data)

[2359] Step 4:

[2360] The server sends the generated answer and explanation to the terminal as an HTTP response.

[2361] Input: Generated answer and explanation

[2362] Output: HTTP response (including answer and explanation)

[2363] Step 5:

[2364] The terminal receives an HTTP response from the server and displays the answer and explanation on the user interface.

[2365] Input: HTTP response

[2366] Output: Answers and explanations displayed on the user screen

[2367] Step 6:

[2368] The user enters feedback about the delivered product in the text box and clicks the submit button.

[2369] Input: User feedback (e.g., "The food tasted very good, but it was a little cold.")

[2370] Output: User input data (HTTP request format)

[2371] Step 7:

[2372] The terminal receives the user's input and sends it to the server as an HTTP request.

[2373] Input: User input data

[2374] Output: HTTP request to the server

[2375] Step 8:

[2376] The server analyzes and evaluates the received feedback, and generates detailed feedback and explanations.

[2377] Input: User feedback

[2378] Output: Evaluation results and explanations (text data)

[2379] Step 9:

[2380] The server sends the generated evaluation results and explanations to the terminal as an HTTP response.

[2381] Input: Evaluation results and explanation

[2382] Output: HTTP response (including evaluation results and explanations)

[2383] Step 10:

[2384] The terminal receives an HTTP response from the server and displays the evaluation results and explanations on the user interface.

[2385] Input: HTTP response

[2386] Output: Evaluation results and explanations displayed on the user screen

[2387] Step 11:

[2388] The server analyzes user trends based on evaluation results and generates subsequent orders and recipes.

[2389] Input: Evaluation result data

[2390] Output: New order suggestions and recipes (text data)

[2391] Step 12:

[2392] The server sends the generated new order suggestions and recipes to the terminal as an HTTP response.

[2393] Input: New order suggestions or recipes

[2394] Output: HTTP response (including new order suggestions and recipes)

[2395] Step 13:

[2396] The terminal receives an HTTP response from the server and displays new order suggestions and recipes on the user interface.

[2397] Input: HTTP response

[2398] Output: New order suggestions and recipes displayed on the user screen

[2399] The above outlines the specific processing steps. In each step, the user, terminal, and server work together to process the data, creating a system that enables the resolution of user questions and the utilization of user feedback.

[2400] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[2401] The learning support system of the present invention features the generation of answers and explanations to questions, evaluation of answers, generation of questions, a chat function, and an emotion engine that recognizes the user's emotions. The following describes specific embodiments for implementing the present invention.

[2402] System Configuration

[2403] This system consists of a user's device, a server, and an emotion engine. Users use individual devices (PCs, smartphones, tablets, etc.), and the server is located in a cloud environment. The emotion engine is equipped with AI technology that analyzes emotions from the user's voice, facial expressions, text, etc.

[2404] User question input and answer generation

[2405] 1. User input questions:

[2406] User: Enter any questions or doubts that arise during the learning process into the text box and click the submit button.

[2407] Terminal: Sends the entered text to the server as an HTTP request.

[2408] 2. Generating the solution:

[2409] Server: Inputs the received questions into a natural language processing model (e.g., BERT or GPT model).

[2410] Server: The AI ​​model analyzes the questions and generates answers and explanations.

[2411] 3. Emotion recognition:

[2412] Device: Uses an emotion engine to recognize the user's emotional state. This includes facial expression analysis and voice analysis.

[2413] 4. Adjustments based on answers and emotions:

[2414] Server: Adjusts the content and tone of the answers and explanations based on the recognized emotional state.

[2415] Server: Sends the generated answer and explanation to the terminal as an HTTP response.

[2416] 5. Display to the user:

[2417] Terminal: Receives responses from the server and displays the answers and explanations on the screen.

[2418] Evaluation and feedback on the answers

[2419] 1. Enter your answer:

[2420] User: Enter your answer to the provided question and click the submit button.

[2421] Terminal: Sends the entered answer to the server as an HTTP request.

[2422] 2. Rating:

[2423] Server: Inputs received answers into the AI ​​model and performs evaluation. Evaluation criteria include accuracy, logic, and clarity.

[2424] Server: Generates detailed feedback and explanations based on the evaluation results.

[2425] 3. Emotion recognition:

[2426] Device: Uses an emotion engine to recognize the user's emotional state.

[2427] 4. Feedback and emotional adjustments:

[2428] Server: Adjusts the content and tone of feedback based on the recognized emotional state.

[2429] Server: Sends feedback to the terminal as an HTTP response.

[2430] 5. Display to the user:

[2431] Terminal: Receives feedback from the server and displays it on the screen.

[2432] Problem generation and learning cycle

[2433] 1. User trend analysis:

[2434] Server: Analyzes user learning trends based on evaluation results. This uses data mining techniques and machine learning models.

[2435] 2. Generating new problems:

[2436] Server: Based on the analysis results, it identifies user weaknesses and generates new problems.

[2437] Server: Sends a new problem to the terminal as an HTTP response.

[2438] 3. Emotion recognition:

[2439] Terminal: Uses an emotion engine to recognize the user's emotional state. This allows for adjustments to the difficulty and content of the generated problems.

[2440] 4. Display to the user:

[2441] Terminal: Receives new issues from the server and displays them on the screen.

[2442] Chat function

[2443] 1. Enter your question:

[2444] User: Type your question in the chat box and click the send button.

[2445] Terminal: Sends the entered questions to the server in real time.

[2446] 2. Real-time response generation:

[2447] Server: Inputs received questions into the AI ​​chatbot model.

[2448] Server: The AI ​​chatbot analyzes the question and generates a response in real time.

[2449] 3. Emotion recognition:

[2450] Device: Uses an emotion engine to recognize the user's emotional state, thereby adjusting the response accordingly.

[2451] 4. Providing a response:

[2452] Server: Sends the generated response to the terminal as an HTTP response in real time.

[2453] Terminal: Receives responses from the server and displays them in the chat box.

[2454] Specific example

[2455] For example, if a user asks, "Could you explain the basics of differentiation?", the server's AI model will generate the answer, "Differentiation is a method for measuring the rate of change of a function." Furthermore, if the emotion engine detects that the user is feeling fatigued, the answer will be adjusted to something like, "Differentiation is a method for measuring the rate of change of a function, and it may seem difficult, but let's work through it together."

[2456] The system of this invention optimizes learning by taking into account the user's emotional state, thereby providing more effective learning support. This dramatically improves the user's motivation and learning efficiency.

[2457] The following describes the processing flow.

[2458] User question input and answer generation

[2459] Step 1:

[2460] User: Enter any questions or doubts that arise during the learning process into the text box and click the submit button.

[2461] Step 2:

[2462] Terminal: Sends the entered text to the server as an HTTP request.

[2463] Step 3:

[2464] Server: Inputs the received questions into a natural language processing model (e.g., BERT or GPT model).

[2465] Step 4:

[2466] Server: The AI ​​model analyzes the questions and generates answers and explanations.

[2467] Step 5:

[2468] Device: Captures the user's facial expressions and voice and sends them to the emotion engine.

[2469] Step 6:

[2470] Emotion Engine: Analyzes the user's emotional state based on captured data.

[2471] Step 7:

[2472] Server: Adjusts the content and tone of the answers and explanations based on the recognized emotional state.

[2473] Step 8:

[2474] Server: Sends the adjusted answer and explanation to the terminal as an HTTP response.

[2475] Step 9:

[2476] Terminal: Receives responses from the server and displays the answers and explanations on the screen.

[2477] Evaluation and feedback on the answers

[2478] Step 1:

[2479] User: Enter your answer to the provided question and click the submit button.

[2480] Step 2:

[2481] Terminal: Sends the entered answer to the server as an HTTP request.

[2482] Step 3:

[2483] Server: Inputs received answers into the AI ​​model and performs evaluation.

[2484] Step 4:

[2485] Server: Generates detailed feedback and explanations based on the evaluation results.

[2486] Step 5:

[2487] Device: Captures the user's facial expressions and voice and sends them to the emotion engine.

[2488] Step 6:

[2489] Emotion Engine: Analyzes the user's emotional state based on captured data.

[2490] Step 7:

[2491] Server: Adjusts the content and tone of feedback based on the recognized emotional state.

[2492] Step 8:

[2493] Server: Sends the adjusted feedback to the terminal as an HTTP response.

[2494] Step 9:

[2495] Terminal: Receives feedback from the server and displays it on the screen.

[2496] Problem generation and learning cycle

[2497] Step 1:

[2498] Server: Analyzes user learning trends based on evaluation results.

[2499] Step 2:

[2500] Server: Based on the analysis results, it identifies user weaknesses and generates new problems.

[2501] Step 3:

[2502] Device: Captures the user's facial expressions and voice and sends them to the emotion engine.

[2503] Step 4:

[2504] Emotion Engine: Analyzes the user's emotional state based on captured data.

[2505] Step 5:

[2506] Server: Adjusts the difficulty and content of generated problems based on the recognized emotional state.

[2507] Step 6:

[2508] Server: Sends a new problem to the terminal as an HTTP response.

[2509] Step 7:

[2510] Terminal: Receives new issues from the server and displays them on the screen.

[2511] Chat function

[2512] Step 1:

[2513] User: Type your question in the chat box and click the send button.

[2514] Step 2:

[2515] Terminal: Sends the entered questions to the server in real time.

[2516] Step 3:

[2517] Server: Inputs received questions into the AI ​​chatbot model.

[2518] Step 4:

[2519] Server: The AI ​​chatbot analyzes the question and generates a response in real time.

[2520] Step 5:

[2521] Device: Captures the user's facial expressions and voice and sends them to the emotion engine.

[2522] Step 6:

[2523] Emotion Engine: Analyzes the user's emotional state based on captured data.

[2524] Step 7:

[2525] Server: Adjusts the response based on the recognized emotional state.

[2526] Step 8:

[2527] Server: Sends the adjusted response to the terminal as an HTTP response in real time.

[2528] Step 9:

[2529] Terminal: Receives responses from the server and displays them in the chat box.

[2530] (Example 2)

[2531] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[2532] Traditional learning support systems can provide answers and explanations to users' questions, but they lack the ability to consider the user's emotional state. Therefore, they cannot adequately recognize the fatigue or frustration a user might experience during learning and adjust the content of responses and feedback accordingly, resulting in a failure to maximize learning effectiveness. Furthermore, the lack of real-time emotion recognition and feedback based on those results made it difficult to maintain user motivation.

[2533] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[2534] In this invention, the server includes means for the user to input questions, means for receiving the input questions, means for generating answers and explanations for the received questions using a generative AI model, means for providing the generated answers and explanations to the user, means for evaluating the generated answers, means for analyzing the user's tendencies based on the evaluation results, means for generating the next question based on the user's tendencies, means for providing the generated question to the user, means for interacting with the user in a chat format, means for recognizing the user's emotional state, and means for adjusting the content of answers and feedback based on the emotional state. This makes it possible to recognize the user's emotional state in real time and adjust the response content and feedback according to that state, thereby increasing the user's motivation to learn and enabling more effective learning support.

[2535] A "user" is an individual or group who uses a learning support system to seek answers to their questions or receive feedback.

[2536] "Questions" refer to the questions or things that users need to understand during the learning process.

[2537] "Input means" refers to interface means for users to input questions and answers into the system, and includes keyboards and touch panels.

[2538] "Receiving means" refers to the means for receiving data transmitted by a user. This includes data receiving systems via a network.

[2539] A "generative AI model" is an artificial intelligence model that uses natural language processing to analyze user input and generate appropriate answers and explanations.

[2540] "Answer" refers to the content of the response to the user's question.

[2541] "Explanation" refers to a detailed explanation provided alongside the answer, serving as supplementary information to help users gain a deeper understanding of their questions.

[2542] "Means of provision" refers to means of displaying or communicating the generated answers and explanations to the user.

[2543] "Evaluation means" refers to methods for evaluating the accuracy, logic, clarity, etc., of answers entered by users, and includes AI models.

[2544] "Evaluation results" refer to data calculated using evaluation methods for user responses.

[2545] A "trend analysis method" is a means of analyzing a user's learning tendencies based on evaluation results.

[2546] The "problem generation method" is a means of generating new learning problems based on the user's learning tendencies and weaknesses, and it uses a generation AI model.

[2547] "Chat dialogue methods" refer to means by which users and systems can interact in real time, and include chatbots that use natural language processing.

[2548] "Emotion recognition means" refers to methods for recognizing a user's emotional state by analyzing their voice, facial expressions, text, etc.

[2549] "Adjustment means" are means for adjusting the content and tone of responses and feedback based on the emotional state obtained by the emotion recognition means.

[2550] The learning support system of the present invention combines the generation of answers and explanations to user questions, evaluation of answers, generation of problems, a chat function, and an emotion recognition function. The following describes in detail the embodiments for specifically implementing the present invention.

[2551] System Configuration

[2552] This system consists of a user's device, a server, and an emotion engine. Users utilize devices such as PCs, smartphones, and tablets. The server is located in a cloud environment and performs the necessary computational processing. The emotion engine incorporates AI technology that analyzes emotions from the user's voice, facial expressions, and text. Specifically, the device collects the user's facial expressions and voice through its camera and microphone and sends them to the server for analysis.

[2553] Hardware and software usage

[2554] The server implements generative AI models (such as natural language processing models like BERT and GPT) to analyze questions and answers submitted by users. It also uses data mining techniques and machine learning models to analyze users' learning tendencies. The terminal is equipped with an emotion recognition engine that analyzes the user's voice and facial expressions in real time to recognize their emotional state. Emotion recognition utilizes facial recognition and voice analysis technologies.

[2555] User question input and answer generation

[2556] The user enters any questions or doubts that arise during the learning process into a text box and clicks the submit button. For example, they might enter, "Please explain the basics of differentiation." The device sends the entered text to the server as an HTTP request. The server inputs the received question into a generating AI model and generates an answer and explanation. An example of a generated answer is, "Differentiation is a method for measuring the rate of change of a function."

[2557] Emotion recognition and regulation

[2558] The emotion engine built into the device analyzes the user's facial expressions, voice, and text to convey emotions. For example, it might recognize "fatigue" from the user's facial expression. Based on the recognized emotional state, the server adjusts the content and tone of the answers and explanations. For instance, it might add an additional message such as, "This might seem difficult, but let's work hard together."

[2559] Evaluation and feedback on the answers

[2560] The user enters an answer to a given question and clicks the submit button. The device sends the entered answer to the server as an HTTP request. The server inputs the received answer into an AI model and evaluates it based on accuracy, logic, and clarity. Based on the evaluation results, it generates detailed feedback and explanations. For example, if the user answers the question "What are the fundamental principles of electromagnetism?" with "The motion of electrons and their field interactions," the server provides feedback such as "This answer is superficially correct, but it would be good to explain it in more detail."

[2561] Problem generation and learning cycle

[2562] The server analyzes the user's learning tendencies based on the evaluation results and identifies the user's weaknesses using data mining techniques and machine learning models. It generates new problems related to the identified weaknesses and sends them to the terminal as an HTTP response. The terminal displays the new problems on its screen, allowing the user to continue learning.

[2563] Chat function

[2564] When a user types a question into the chat box and clicks the send button, the device sends the entered question to the server in real time. The server inputs the received question into an AI chatbot model and generates a response in real time. If emotion recognition determines that the user's emotional state is "frustrated," the response is adjusted to something like, "This may be difficult to understand, but I will explain it carefully." The server then sends the generated response to the device as an HTTP response in real time, and the device displays it in the chat box.

[2565] As described above, the present invention provides a more effective learning experience by providing real-time responses to user questions and answ...

Claims

1. A means of entering questions, A means of receiving the entered questions, A means for generating answers and explanations to received questions, A means of providing the generated answers and explanations to the user, A means for evaluating the generated answer, A means of analyzing user trends based on evaluation results, A means of generating the next problem based on user trends, A means of providing the generated problems to the user, A means of interacting with users in a chat format, A system that includes this.

2. The system according to claim 1, further comprising means for identifying user weaknesses based on evaluation results and generating problems that focus on the identified weaknesses.

3. The system according to claim 1, further comprising means for providing the generated answers and explanations to the user in real time.

Images