System

Abstract

To provide a system capable of promoting the independent learning of a learner and reducing the job burden of a teacher.SOLUTION: The system includes a means for allowing a user to input lesson contents, a means for analyzing the input lesson contents and retrieving related resources, a means for calling a generation AI model for generating erroneous information on the basis of an analysis result, a means for adjusting the generated erroneous information, a means for generating a lesson script and teaching materials on the basis of the generated information, and a means for providing the generated lesson script and teaching materials to the user. A system comprising: means for prompting a learner to think; means for recording reactions of the learner during a lesson and transmitting the reactions to a server; means for analyzing the recorded reactions and providing feedback in real time; and means for generating a summary of lesson content after the lesson ends.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] In today's educational environment, passive education, in which teachers unilaterally impart knowledge, is the norm, and students' independent thinking and problem-solving abilities are not sufficiently developed. Furthermore, the serious teacher shortage makes it difficult to provide high-quality educational services. Given this background, new educational methods are needed that promote learners' independent learning and reduce the burden on teachers. [Means for solving the problem]

[0005] The present invention solves these problems by providing a system that includes a means for a user to input lesson content, a means for analyzing the input lesson content and searching for related resources, a means for calling a generative AI model that generates erroneous information based on the analysis results, a means for adjusting the generated erroneous information, a means for generating lesson scripts and teaching materials based on the generated information, a means for a user to present erroneous information during class to encourage learners to think, a means for recording learners' responses during class and transmitting them to a server, a means for analyzing the recorded responses and providing feedback in real time, and a means for generating a summary of the lesson content after the class has ended.

[0006] Furthermore, the means for analyzing learner responses includes a means for recording learner comments using voice recognition technology, and the generative AI model that generates erroneous information is characterized by being able to adjust the difficulty of the information according to the learner's level of understanding. This makes it possible to promote learners' independent learning and reduce the workload of teachers.

[0007] "User" refers to a teacher or educator who uses the system to plan and deliver lessons.

[0008] "Class content" refers to information such as subjects, target grades, and topics that users enter into the system.

[0009] "Related resources" refers to teaching materials, past data, reference materials, etc. related to the course content.

[0010] "Analysis" refers to the process of identifying and organizing relevant resources based on the input lesson content.

[0011] A "generative AI model" refers to an artificial intelligence algorithm that generates false information based on input information.

[0012] "Misinformation" refers to intentionally incorrect answers or information, intended to provide learners with an opportunity to think independently.

[0013] A "lesson script" refers to instructions or a plan that a user uses to conduct a lesson.

[0014] "Teaching materials" refers to the materials and tools used during lessons.

[0015] "Learner" refers to pupils who participate in classes and study.

[0016] "Reactions" refer to behaviors such as opinions, questions, and answers shown by learners during class.

[0017] A "server" refers to a computer that processes and stores data at the center of a system.

[0018] "Terminal" refers to a device (e.g., PC, tablet, smartphone, etc.) used by a user or learner to access the system.

[0019] "Recording" refers to the process of saving learners' reactions and behavior during class as data.

[0020] A "summary" refers to a report that briefly summarizes the content of the lesson and the learners' responses.

[0021] "Speech recognition technology" refers to technology that converts speech into text and is used to record what learners say.

[0022] "Feedback" refers to advice or information provided to learners in response to their reactions or behavior during a lesson. [Brief explanation of the drawings]

[0023] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3]FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0024] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0025] First, the terms used in the following description will be explained.

[0026] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0027] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0028] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0029] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0030] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0031] [First embodiment]

[0032] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0033] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0034] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0035] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0036] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0037] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0038] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0039] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0040] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0041] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0042] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0043] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0044] This invention relates to "Empower GPT," a system for supporting teachers in the educational field. This system uses generative AI to provide false information, aiming to develop learners' ability to solve problems independently.

[0045] System Overview

[0046] The system includes the following main components:

[0047] 1. User interface (terminal): A device that allows teachers to input lesson content and check and modify the generated scripts and teaching materials.

[0048] 2. Server: A central computer that processes and stores data, calls generative AI models, and analyzes learner responses.

[0049] 3. Generative AI model: An artificial intelligence algorithm that generates misinformation based on lesson content, providing learners with an opportunity to think for themselves.

[0050] Program processing procedure

[0051] 1. User login and authentication:

[0052] The user logs into the system from a terminal and enters the necessary authentication information.

[0053] The server verifies the authentication information and sets up the appropriate environment for each user.

[0054] 2. Input lesson content:

[0055] The user inputs information such as the class subject, target grade, topic, etc. from the terminal.

[0056] The terminal transmits this information to the server.

[0057] 3. Analyzing information and finding resources:

[0058] The server analyzes the input lesson content and searches for related resources (teaching materials, past lesson content, etc.).

[0059] 4. Creation of misinformation:

[0060] The server calls the generative AI model and generates false information based on the analysis results.

[0061] The generated misinformation is tailored to the learner's level of understanding and grade level.

[0062] 5. Generating lesson scripts and teaching materials:

[0063] The server generates lesson scripts and teaching materials based on the generated incorrect information and correct information.

[0064] The generated scripts and teaching materials are sent to the terminal and checked by the user.

[0065] 6. Class Conduct:

[0066] A user presents false information during a lesson to initiate a process of thinking in the learners.

[0067] Generative AI models respond to users' questions and concerns in real time.

[0068] 7. Recording and analyzing learner responses:

[0069] The device uses voice recognition technology to record the learner's responses.

[0070] The recorded data is sent to a server, which analyzes it.

[0071] 8. Providing Feedback:

[0072] The server provides real-time feedback based on the analysis results and suggests next steps to the learner.

[0073] 9. Post-class summary generation:

[0074] After the lesson ends, the server generates a summary based on the lesson content and the learners' responses.

[0075] The summary is sent to the terminal, where the user can review it and reflect it in the next lesson plan.

[0076] Specific examples

[0077] For example, consider a case where the topic of a second-year junior high school math class is "quadratic equations." When a user selects "quadratic equations" as the topic, the server searches for related resources, and the generative AI model generates an incorrect answer (e.g., $x^2 + 4x + 4 = (x+2)^2 - 4$, an incorrect expansion). The user presents this incorrect answer during class, encouraging students to derive the correct answer on their own. After the class, the server generates a summary based on the content of the class and the students' responses, which can be used for the next class.

[0078] In this way, the present invention provides a system that promotes independent learning by learners in educational settings and reduces the burden on teachers.

[0079] The processing flow will be explained below.

[0080] Step 1: User Login and Authentication

[0081] The user enters the information (user name, password) required to log in to the system from the terminal.

[0082] The terminal sends the entered login information to the server.

[0083] The server checks the received login information against a database and performs authentication.

[0084] If authentication is successful, the server obtains the user's personal settings and history and sends them to the terminal.

[0085] The user confirms successful authentication through the terminal and is then able to access the system.

[0086] Step 2: Enter lesson content

[0087] The user enters information such as the class subject, target grade, and topic from the terminal.

[0088] The terminal sends the input lesson content to the server.

[0089] Step 3: Analyze the information and find resources

[0090] The server analyzes the input lesson content.

[0091] Based on the analysis results, the server searches for relevant resources (teaching materials, past lesson contents, reference materials, etc.) and retrieves them from the database.

[0092] Step 4: Creating misinformation

[0093] The server calls the generative AI model and generates incorrect information based on the lesson content and analysis results.

[0094] The generated erroneous information is appropriately adjusted according to the learner's level of understanding and the target grade level.

[0095] Step 5: Generate lesson scripts and teaching materials

[0096] The server combines the generated incorrect information with the correct information to generate lesson scripts and teaching materials.

[0097] The generated lesson scripts and teaching materials are sent to the terminal, where the user can check and modify them.

[0098] Step 6: Conducting lessons

[0099] The user starts the lesson through the terminal.

[0100] The user follows a generated lesson script to intentionally present incorrect information and encourage learners to think.

[0101] During lessons, the generative AI model responds to and supports users' questions and concerns in real time.

[0102] Step 7: Record and analyze learner responses

[0103] The device uses voice recognition technology to record students' comments and reactions during class.

[0104] The recorded data is sent to a server, which analyzes it.

[0105] Based on the analysis results, the server provides real-time feedback and suggests next learning steps to the user.

[0106] Step 8: Generate a summary after the lesson

[0107] After the lesson ends, the server generates a summary based on the data from the lesson and the learners' responses.

[0108] The generated summary is sent to the terminal, where the user can check the lesson content and the learners' level of understanding and reflect it in future lesson plans.

[0109] Example 1

[0110] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0111] In today's educational settings, there are no fully established methods for teachers to provide appropriate teaching materials and misinformation according to each student's level of understanding, and for students to develop the ability to solve problems independently based on that information. Furthermore, it is difficult to analyze students' responses in real time and provide appropriate feedback. This increases the burden on teachers and hinders the promotion of independent learning among students, creating the problem of a lack of understanding.

[0112] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0113] In this invention, the server includes: means for a user to input lesson content; means for analyzing the input lesson content and searching for related resources; means for calling a generative AI model that generates erroneous information based on the analysis results; means for adjusting the generated erroneous information; means for generating lesson scripts and teaching materials based on the generated information; means for providing the generated lesson scripts and teaching materials to the user; means for a user to present erroneous information during class to encourage learners to think; means for recording learners' responses during class and transmitting them to the server; means for analyzing the recorded responses and providing feedback in real time; means for generating a summary of the lesson content after the class ends; means for adjusting the difficulty level of the generated erroneous information according to the learners' level of understanding; and means including a generative AI model that responds to user questions and doubts in real time during class. This reduces the burden on teachers and promotes learners' independent learning.

[0114] "Users" refers to teachers and educators who use this system in educational settings.

[0115] "Lesson content" refers to information such as subject, target grade, and topic that teachers enter into the system.

[0116] "Analysis" refers to the process of breaking down input lesson content using natural language processing techniques or other methods to identify relevant resources.

[0117] "Resources" refers to teaching materials, reference materials, past lesson data, etc. related to the content of the lesson.

[0118] A "generative AI model" refers to the part of a system that uses artificial intelligence algorithms to generate false information based on specific inputs.

[0119] A "prompt" refers to input text used to instruct a generative AI model on what information to generate.

[0120] "Misinformation" refers to incorrect data or answers that are intentionally generated as a starting point for learners to use their own understanding to derive accurate information.

[0121] A "script" refers to a script or instruction manual that helps guide a lesson.

[0122] "Instructional materials" refers to visual, textual, or hands-on learning materials used to supplement course content.

[0123] "Terminal" refers to a device such as a computer or tablet that teachers use to access and operate the system.

[0124] "Server" refers to the central computing system that processes and stores data, invokes generative AI models, analyzes learner responses, etc.

[0125] "Speech recognition technology" refers to technology that converts learners' speech into text.

[0126] "Feedback" refers to instructions, hints, advice, etc. provided based on the learner's responses.

[0127] A "summary" refers to a short report generated based on the lesson content and learner responses.

[0128] "Responses" refer to the level of understanding, questions, comments, etc. that learners show during class.

[0129] This invention relates to a system called "Empower GPT" that supports teachers in the educational field. This system uses generative AI to provide false information and aims to cultivate learners' ability to solve problems independently. The following describes in detail the mode for carrying out the invention.

[0130] System configuration

[0131] The system includes the following main components:

[0132] 1. User interface (terminal): A device on which teachers input lesson content and check and edit generated scripts and teaching materials. Examples include PCs, tablets, and smartphones.

[0133] 2. Server: This is the central computer that processes and stores data, calls the generative AI model, and analyzes learner responses. In this invention, the server is assumed to be a high-performance computer with a Linux (registered trademark)-based OS. MySQL (registered trademark) or PostgreSQL is used as the database.

[0134] 3. Generative AI model: An artificial intelligence algorithm that generates false information based on the lesson content input by the teacher, providing learners with an opportunity to think for themselves. Specific examples include OpenAI's (registered trademark) GPT-3 (registered trademark) and GPT-4 (registered trademark).

[0135] Program processing procedure

[0136] The system begins when the user (teacher) logs in and inputs the lesson content. The specific processing steps are explained below.

[0137] 1. User login and authentication:

[0138] A user logs in to the system from a terminal and enters the necessary authentication information, such as a user ID and password.

[0139] The server verifies the authentication information and sets up the appropriate environment for each individual user, a process that involves the server verifying user information against a database.

[0140] 2. Input lesson content:

[0141] The user inputs information such as the class subject, target grade, topic, etc., into the terminal. For example, they input "mathematics," "second-year junior high school student," and "quadratic equations."

[0142] The device sends this information to the server, and the data is sent in JSON format.

[0143] 3. Analyzing information and finding resources:

[0144] The server analyzes the input lesson content and searches the database for related resources (teaching materials, past lesson content, etc.) using natural language processing libraries (e.g., NLTK and spaCy).

[0145] 4. Creation of misinformation:

[0146] The server calls the generative AI model and generates false information based on the analysis results. The generative AI model is instructed using prompts.

[0147] The generated misinformation is tailored to the learner's level of understanding and grade level.

[0148] 5. Generating lesson scripts and teaching materials:

[0149] The server generates lesson scripts and teaching materials based on the generated incorrect information and correct information. For example, the teaching materials are constructed using Python or JavaScript (registered trademark) scripts.

[0150] The generated scripts and teaching materials are sent to the terminal and checked by the user.

[0151] 6. Class Conduct:

[0152] The user presents incorrect information during a lesson to initiate the thinking process in the learner, for example, by using an incorrect formula such as "x^2 + 4x + 4 = (x+2)^2 - 4".

[0153] The generative AI model responds to user questions and concerns in real time, including when users type questions into the generative AI, which then responds instantly.

[0154] 7. Recording and analyzing learner responses:

[0155] The device records the learner's responses using voice recognition technology, such as Google® Speech-to-Text or Amazon Transcribe.

[0156] The recorded data is sent to a server, which analyzes it using machine learning algorithms (e.g., scikit-learn or TENSORFLOW®).

[0157] 8. Providing Feedback:

[0158] The server provides real-time feedback based on the analysis results and suggests next steps to the learner, such as "Many students got this question wrong, so let's explain it again."

[0159] 9. Post-class summary generation:

[0160] After the lesson, the server generates a summary based on the lesson content and the learners' responses. The summary is created using natural language generation technology.

[0161] The summary is sent to the terminal, where the user can review it and reflect it in the next lesson plan.

[0162] Specific examples

[0163] For example, if the topic of a second-year junior high school math class is "quadratic equations," when the user selects "quadratic equations" as the topic, the server searches for related resources and the generative AI model generates an incorrect answer (e.g., $x^2 + 4x + 4 = (x+2)^2 - 4$, an incorrect expansion). The user presents this incorrect answer during class, encouraging the learners to derive the correct answer on their own. After the class, the server generates a summary based on the content of the class and the learners' responses, which can be used for the next class.

[0164] Examples of prompt statements

[0165] Topic: 8th Grade Math - Quadratic Equations

[0166] Objectives: Students will understand how to solve quadratic equations and be able to correct incorrect information themselves.

[0167] Examples of misinformation it generates:

[0168] $x^2 ​​+ 4x + 4 = (x+2)^2 - 4$

[0169] "The only solution to a quadratic equation is x = 2"

[0170] In this way, the present invention provides a system that promotes independent learning by learners in educational settings and reduces the burden on teachers.

[0171] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0172] Step 1:

[0173] A user accesses the system from a terminal and enters login information (user ID and password). The input data is encrypted on the terminal as authentication information and sent to the server. The server receives the authentication information and checks it against a database. Based on the check result, if authentication is successful, the environment settings corresponding to the user are loaded and a dashboard screen is displayed on the terminal. The output is a dashboard screen that the user can access.

[0174] Step 2:

[0175] The user inputs the class subject, target grade, and specific topic into the device. The input data includes information such as "Mathematics," "Second Year of Junior High School," and "Quadratic Equations." The device sends the input data in JSON format to the server. The server temporarily stores the received data and prepares for the analysis step. The output is a notification that the data has been saved on the server.

[0176] Step 3:

[0177] The server analyzes the received lesson content data. A natural language processing library (e.g., NLTK or spaCy) is used for the analysis. The input for the analysis is JSON data about the lesson content. As a result of the analysis, search keywords and tags for related resources (teaching materials, past lesson content, etc.) are generated. The output is the search keywords and tags.

[0178] Step 4:

[0179] The server searches for related resources from databases and cloud storage based on the analysis results. The input in this step is search keywords and tags. Related resources (PDF teaching materials, video links, past lesson scripts, etc.) are obtained as search results. The output is a list of the obtained resources.

[0180] Step 5:

[0181] The server invokes a generative AI model to generate false information using a prompt. The input in this step is the analysis result of the lesson content and the prompt. The generative AI model (e.g., GPT-3 or GPT-4) generates false information based on the input prompt. The output is the generated false information.

[0182] Step 6:

[0183] The server adjusts the generated misinformation according to the learner's level of understanding and grade level. The input is the generated misinformation and the learner's profile data. The adjusted information is generated and output. The adjustment may use machine learning algorithms.

[0184] Step 7:

[0185] The server generates lesson scripts and teaching materials based on the adjusted misinformation and correct information. The input data is the adjusted misinformation and related correct information. The lesson scripts and teaching materials are generated using programming languages ​​such as Python and JavaScript. The output is the completed lesson scripts and teaching materials.

[0186] Step 8:

[0187] The server sends the generated lesson script and teaching materials to the terminal, where the user can review them. The output is a script and teaching materials that the user can review on the terminal. The user can then modify the content as needed.

[0188] Step 9:

[0189] The user presents false information generated during class to prompt students to think. The input is the adjusted false information. As a concrete example, a teacher writes "$x^2 + 4x + 4 = (x+2)^2 - 4$" on the blackboard and asks students to think about what is wrong. The output is the student's response.

[0190] Step 10:

[0191] The device uses voice recognition technology (e.g., Google Speech-to-Text or Amazon Transcribe) to record the learner's responses and send them to the server. The input is the learner's speech. The voice data is converted into text format, and the output is the text data and its analysis results.

[0192] Step 11:

[0193] The server analyzes the recorded responses of the learner and provides feedback in real time. The input is the learner's response data in text format. Based on the analysis results, appropriate feedback is generated, and the output is the content of the feedback.

[0194] Step 12:

[0195] After the lesson ends, the server generates a summary based on the lesson content and learner responses. The input is the lesson script, teaching materials, and learner response data. The summary is created using natural language generation technology. The output is a summary that is sent to the terminal and reviewed by the user.

[0196] (Application example 1)

[0197] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0198] In conventional educational systems, when a single teacher instructs a large number of students, it is difficult to grasp each student's level of understanding and reactions in real time and provide appropriate feedback. Furthermore, in order for students to develop their independent problem-solving skills, they need teaching materials that contain not only accurate information but also incorrect information, but generating and adjusting such materials is time-consuming. This increases the burden on teachers and makes it difficult to improve students' independent learning abilities.

[0199] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0200] In this invention, the server includes a means for users to input lesson content, a means for analyzing the input lesson content and searching for related resources, and a means for calling a generative AI model that generates erroneous information based on the analysis results. This makes it possible to generate and adjust erroneous information and answers and provide appropriate feedback in real time. Furthermore, the server includes a means for users to select the content they want to learn and provide teaching materials containing erroneous information, a means for providing real-time feedback to students based on the generated erroneous information and answers, and a means for recording and saving learned content and answers so that they can be reviewed later. This facilitates the improvement of learners' independent learning abilities while reducing the burden on teachers.

[0201] "Means for users to input lesson content" refers to interfaces and applications that allow teachers and students to input the subjects or topics they want to learn into digital devices.

[0202] "Means for analyzing input lesson content and searching for related resources" refers to an algorithm that searches and retrieves related information and materials from databases or networks based on the input teaching materials or topics.

[0203] "Means for invoking a generative AI model that generates erroneous information based on the analysis results" refers to a mechanism that uses the analysis results of the input content to execute a process that intentionally creates erroneous information using a generative AI model.

[0204] The "means for adjusting generated erroneous information" is a function for appropriately adjusting the difficulty and content of the generated erroneous information according to the learner's level of understanding and comprehension.

[0205] The "means for generating lesson scripts and teaching materials based on the generated information" is a system that combines incorrect information with correct information to create and provide scripts and educational materials to be used in actual lessons.

[0206] The "means for providing the generated lesson scripts and teaching materials to users" refers to a mechanism for distributing the generated lesson scripts and teaching materials to teachers and learners via digital devices.

[0207] "Means for users to present incorrect information during class and encourage students to think" is a process in which a teacher intentionally presents incorrect information during class and has students point out the errors.

[0208] "Means for recording learners' responses during class and sending them to a server" refers to a system that has the function of recording how learners respond during class and sending and saving them to a server.

[0209] The "means for analyzing recorded responses and providing real-time feedback" refers to an algorithm that analyzes recorded responses of learners and provides appropriate feedback immediately.

[0210] The "means for generating a summary of the content of a lesson after the lesson has ended" is a system that summarizes the content of the lesson after the lesson has ended, generates a summary, and provides it.

[0211] "Means for providing students with real-time feedback based on incorrect information or answers" refers to a system that provides supplementary information or advice in real time to learners as they solve problems, based on the incorrect information or answers they generate.

[0212] "A means to record and save what has been learned and answers, and to review them later" is a system that has the function of saving what learners have learned and answers in a database, allowing them to check and review them later.

[0213] This invention relates to an educational support system that utilizes a generative AI model. The purpose of this invention is to promote learners' independent learning and reduce the burden on teachers by cooperating with three main components: a server, a terminal, and a user.

[0214] System configuration

[0215] 1. Server:

[0216] The server has the following functions:

[0217] Provides a means for users to input lesson content.

[0218] It has the function of analyzing the input lesson content and searching for related resources.

[0219] Based on the analysis results, a generative AI model is invoked to generate false information.

[0220] The generated incorrect information is adjusted and lesson scripts and teaching materials are generated.

[0221] Analyze recorded learner responses and provide feedback in real time.

[0222] 2. Terminal:

[0223] Terminals are devices used by teachers and students and have the following functions:

[0224] It provides a user interface for inputting, checking, and modifying lesson content.

[0225] Display the generated lesson scripts and teaching materials.

[0226] Learners' responses are recorded during the lesson and sent to the server.

[0227] 3. User:

[0228] Users are mainly teachers and students, and use the system in the following steps:

[0229] Enter the lesson content from the terminal.

[0230] Check and modify the generated scripts and teaching materials.

[0231] Present misinformation during class and encourage students to problem-solve.

[0232] Record learners' responses and review their feedback after the lesson.

[0233] Software and Hardware Configuration

[0234] On the server side, it uses high-performance cloud servers such as Amazon Web Services and Google Cloud Platform, and the backend is built using Node.js and Express, with MongoDB as the database.

[0235] The front-end part will be developed using React Native, taking into consideration use on smartphones.

[0236] The generative AI model uses OpenAI GPT-4, which generates prompts based on the input lesson content and then creates false information as a response to those prompts.

[0237] As a use case, consider a 8th grade math lesson on "quadratic equations," where the following prompt is sent to the generative AI model:

[0238] "Generate false information and answers on the following topic: Quadratic Equations"

[0239] The generative AI model generates incorrect information for this prompt (e.g., an incorrect expansion such as $x^2 + 4x + 4 = (x+2)^2 - 4$), which is then provided as part of the learning material.

[0240] The server analyzes the learner's responses and provides real-time feedback, for example, helping the learner recognize incorrect information and guide them in the process of arriving at the correct answer.

[0241] This system will enable students to develop the ability to correct their own mistakes and reduce the burden on teachers. It is also expected that the inclusion of incorrect information will dramatically improve students' thinking and problem-solving abilities.

[0242] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0243] Step 1:

[0244] The user inputs the lesson content.

[0245] In this step, the user enters information such as the class subject, target grade, topic, etc. into the application via their device. The entered information is sent from the front end (React Native) to the server (Node.js).

[0246] Input: Course subject, target grade, topic

[0247] Output: The input data is sent to the server

[0248] Step 2:

[0249] The server analyzes the content of the lesson and searches for related resources.

[0250] The server analyzes the received lesson content and searches a database (MongoDB) for teaching materials and past resources related to that topic.

[0251] Input: Lesson content sent in Step 1

[0252] Output: A list of resources related to the analysis results

[0253] Step 3:

[0254] The server calls a generative AI model based on the analysis results and generates incorrect information.

[0255] Based on the analysis results, the server sends a prompt to the generative AI model (OpenAI GPT-4) to generate incorrect information.

[0256] Input: Analysis results, prompt (e.g. "Generate incorrect information about quadratic equations")

[0257] Output: The incorrect information generated

[0258] Step 4:

[0259] The server will adjust the generated incorrect information.

[0260] The server runs an algorithm that adjusts the generated misinformation according to the learner's level of comprehension and grade level, optimizing the difficulty of the misinformation.

[0261] Input: Generated misinformation, learner comprehension information

[0262] Output: Adjusted misinformation

[0263] Step 5:

[0264] The server generates lesson scripts and teaching materials based on the generated information.

[0265] The server generates lesson scripts and teaching materials based on the adjusted incorrect information and correct information and provides them to users.

[0266] Input: Adjusted incorrect information, correct information

[0267] Output: Lesson scripts, teaching materials

[0268] Step 6:

[0269] The user presents incorrect information during class to prompt students to think.

[0270] The user uses the generated teaching materials in class, presents incorrect information to students, and has them point out the errors, thereby encouraging the students to think independently.

[0271] Input: lesson scripts, teaching materials

[0272] Output: Learner response

[0273] Step 7:

[0274] The device records the learners' responses during class and sends them to the server.

[0275] Learners' responses and reactions are recorded by the device and sent to the server in real time. Speech recognition technology is also used to record utterances.

[0276] Input: Learner response data

[0277] Output: Response data sent to the server

[0278] Step 8:

[0279] The server analyzes the recorded responses and provides real-time feedback.

[0280] The server analyzes the received learner response data and uses an AI model to generate and provide instant feedback.

[0281] Input: Reaction data

[0282] Output: Real-time feedback

[0283] Step 9:

[0284] After the lesson ends, the server generates a summary of the lesson content.

[0285] After the lesson is over, the server creates a summary based on the lesson content and the learners' reactions and provides it to the user to help plan the next lesson.

[0286] Input: Lesson content, learner response data

[0287] Output: Lesson summary

[0288] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0289] This invention relates to "Empower GPT," an educational support system with the ability to recognize users' emotions and adapt lesson scripts and teaching materials. The system aims to optimize lessons based on the user's emotional state while providing false information using generative AI to foster learners' ability to solve problems independently.

[0290] System Overview

[0291] The system includes the following main components:

[0292] 1. User interface (terminal): A device that allows teachers to input lesson content and check and modify the generated scripts and teaching materials.

[0293] 2. Server: The central computer that processes and stores data, invokes generative AI models, and analyzes learner responses and user sentiment.

[0294] 3. Generative AI model: An artificial intelligence algorithm that generates misinformation based on lesson content, providing learners with an opportunity to think independently.

[0295] 4. Emotion Engine: An algorithm to recognize the user's emotional state and adjust the lesson script and teaching materials.

[0296] Program processing procedure

[0297] 1. User login and authentication:

[0298] The user logs into the system from a terminal and enters their authentication information.

[0299] The terminal sends the entered login information to the server, and the server performs authentication.

[0300] If authentication is successful, the server obtains the user's personal settings and history and sends them to the device.

[0301] 2. Input lesson content:

[0302] The user enters information such as the class subject, target grade, and topic into the terminal.

[0303] The terminal transmits this information to the server.

[0304] 3. Analyzing information and finding resources:

[0305] The server analyzes the input lesson content and searches for and organizes related resources.

[0306] 4. Creation of misinformation:

[0307] The server calls the generative AI model and generates incorrect information based on the lesson content and analysis results.

[0308] The generative AI model adjusts the generated incorrect answers based on the user's emotional data.

[0309] 5. Generating lesson scripts and teaching materials:

[0310] The server combines the generated incorrect information with the correct information to generate lesson scripts and teaching materials.

[0311] The generated lesson scripts and teaching materials are sent to the terminal, where the user can check and modify them.

[0312] 6. Class Conduct:

[0313] The user starts the lesson through the terminal.

[0314] The user presents incorrect information according to the generated lesson script, prompting learners to think.

[0315] Generative AI models respond to users' questions and concerns in real time.

[0316] 7. Recording and analyzing learner responses and user emotions:

[0317] The device uses voice recognition technology to record students' comments and reactions during class.

[0318] The terminal recognizes the user's emotional state and transmits the emotional data to the server.

[0319] The recorded data and emotional data are sent to a server for analysis.

[0320] Based on the analysis results, the server provides real-time feedback and suggests next learning steps to the user.

[0321] 8. Post-class summary generation:

[0322] After the lesson ends, the server generates a summary based on the data from the lesson, the learners' reactions, and the user's emotional data.

[0323] A summary is sent to the terminal, and the user checks the lesson content and the learners' level of understanding, and reflects this in the next lesson plan.

[0324] Specific examples

[0325] For example, consider a case where the topic of a math class for second-year junior high school students is "quadratic equations." The user selects "quadratic equations" as the topic and enters the necessary information into the system. The server searches for related resources, and the generative AI model generates an incorrect answer (e.g., $x^2 + 4x + 4 = (x+2)^2 - 4$, an incorrect expansion). The emotion engine adjusts the generated incorrect answers and lesson scripts based on the user's emotional data.

[0326] The user presents the incorrect answer during the lesson, encouraging the learner to find the correct answer on their own. The device uses voice recognition technology and an emotion recognition engine to record the learner's responses and the user's emotions during the lesson. After the lesson ends, the server analyzes this data, generates a summary, and provides it to the user. In this way, it is possible to promote learners' independent learning and conduct lessons that are tailored to the user's emotional state.

[0327] The processing flow will be explained below.

[0328] Step 1: User Login and Authentication

[0329] The user enters the information (user name, password) required to log in to the system from the terminal.

[0330] The terminal sends the entered login information to the server.

[0331] The server checks the received login information against a database and performs authentication.

[0332] If authentication is successful, the server obtains the user's personal settings and history and sends them to the terminal.

[0333] The user can then verify their authentication through the terminal and gain access to the system.

[0334] Step 2: Enter lesson content

[0335] The user enters information such as the class subject, target grade, and topic from the terminal.

[0336] The terminal sends the input lesson content to the server.

[0337] Step 3: Analyze the information and find resources

[0338] The server analyzes the input lesson content.

[0339] The server searches for and retrieves relevant resources (teaching materials, past lesson contents, reference materials, etc.) based on the analysis results.

[0340] Step 4: Creating misinformation

[0341] The server calls the generative AI model and generates incorrect information based on the lesson content and analysis results.

[0342] The server adjusts the difficulty and content of the generated erroneous information according to the learner's level of understanding.

[0343] Step 5: Emotional Engine Alignment

[0344] The device monitors the user's emotional state in real time and transmits the emotional data to a server.

[0345] The server uses an emotion engine to analyze the emotional data and adjust the generated incorrect answers and lesson scripts.

[0346] Step 6: Generate lesson scripts and teaching materials

[0347] The server combines the generated incorrect information with the correct information to generate lesson scripts and teaching materials.

[0348] The generated lesson scripts and teaching materials are sent to the terminal, where the user can check and modify them.

[0349] Step 7: Conducting lessons

[0350] The user starts the lesson through the terminal.

[0351] The user follows a generated lesson script to intentionally present incorrect information and encourage learners to think.

[0352] During lessons, the generative AI model responds to users' questions and concerns in real time.

[0353] Step 8: Recording and analyzing learner responses and user emotions

[0354] The device uses voice recognition technology to record students' comments and reactions during class.

[0355] The terminal recognizes the user's emotional state and transmits the emotional data to the server.

[0356] The server analyzes the recorded learner responses and user emotion data.

[0357] Based on the analysis results, the server provides real-time feedback and suggests next learning steps to the user.

[0358] Step 9: Generate a summary after the lesson

[0359] After the lesson ends, the server generates a summary based on the data from the lesson, the learners' reactions, and the user's emotional data.

[0360] The generated summary is sent to the terminal, where the user can check the lesson content and the learners' level of understanding and reflect it in the next lesson plan.

[0361] Example 2

[0362] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0363] Traditional educational systems lacked effective methods for cultivating students' independent problem-solving skills. They also lacked mechanisms for adapting lesson content based on the user's emotional state, making it difficult to maximize educational effectiveness. As a result, they were unable to respond to fluctuations in students' interests and comprehension levels, limiting their ability to provide consistent educational content.

[0364] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0365] In this invention, the server includes means for a user to input lesson content, means for analyzing the input lesson content and searching for related resources, means for calling an AI algorithm that generates erroneous information based on the analysis results, means for adjusting the generated erroneous information, means for generating lesson scripts and teaching materials based on the generated information, means for providing the generated lesson scripts and teaching materials to the user, means for the user to present erroneous information during the lesson to encourage students to think, means for recording students' responses during the lesson and transmitting them to a central computer, means for analyzing the recorded responses and providing feedback in real time, means for generating a summary of the lesson content after the lesson ends, and means for recognizing the user's emotional state and adjusting the lesson scripts and teaching materials, thereby cultivating students' independent problem-solving skills and enabling the lesson content to be adapted based on the user's emotional state.

[0366] "Users" are educators and administrators who operate the system and input lesson content.

[0367] "Class content" refers to educational information such as the subjects, topics, and target grades covered in the class.

[0368] "Input means" refers to an interface or device that allows users to input lesson content into the system.

[0369] "Analysis means" refers to algorithms or devices that analyze the input lesson content and search for related resources.

[0370] "Misinformation" refers to intentionally inaccurate educational information provided to learners in order to solve problems independently.

[0371] "Artificial intelligence algorithms" refer to machine learning and natural language processing technologies that generate false information based on input information.

[0372] "Adjustment means" refers to processes or functions for adjusting erroneous information generated by artificial intelligence algorithms.

[0373] A "lesson script" is a script or plan used in a lesson, and is a document or digital data that contains the content as the lesson progresses.

[0374] "Teaching materials" refers to the teaching materials and learning resources used in classes (e.g., textbooks, workbooks, video materials, etc.).

[0375] "Recording means" refers to voice recognition technology or other digital recording devices used to record learner responses during lessons.

[0376] The "central computer" is the main device in the system that processes and stores data, calls generative AI models, and analyzes learner responses and user emotions.

[0377] "Analysis tools" are algorithms or devices that analyze recorded learner responses and provide real-time feedback.

[0378] An "emotional state" is a state that recognizes and classifies a user's emotions (e.g., joy, sadness, anger, surprise, etc.).

[0379] A "summary generation means" refers to an algorithm or device that creates a summary after a lesson based on the lesson content and learners' responses.

[0380] "Real-time feedback" refers to guidance and advice provided immediately or briefly during or after a lesson.

[0381] MODE FOR CARRYING OUT THE INVENTION

[0382] In the present invention, the following hardware and software are used to effectively implement the education support system.

[0383] (System Configuration)

[0384] 1. User Interface (Terminal)

[0385] 2. Server

[0386] 3. Generative AI Models

[0387] 4. Emotion Engine

[0388] (User Interface (Terminal))

[0389] A user interface is a device that allows users to input lesson content and review and edit generated scripts and teaching materials. This can be a computer, tablet, smartphone, or other digital device. It has a solid graphical user interface (GUI) and is designed to allow users to operate it intuitively.

[0390] (server)

[0391] The server is the central computer that processes and stores data, calls the generative AI model, and analyzes learner responses and user emotions. It uses a cloud server with powerful computing capabilities, and is able to exchange data with devices in real time via the communications infrastructure. The server integrates a database, analysis engine, and various algorithms.

[0392] (Generative AI model)

[0393] The generative AI model is an artificial intelligence algorithm that generates false information based on input lesson content, providing learners with an opportunity to think independently. It incorporates natural language processing (NLP) and machine learning (ML) technologies and has the ability to adaptively adjust the information it generates based on the user's emotional data.

[0394] (Emotion Engine)

[0395] The emotion engine is an algorithm that recognizes the user's emotional state and adjusts lesson scripts and teaching materials accordingly. It constantly monitors the user's emotional state in real time using voice recognition and facial expression analysis technologies.

[0396] Specific examples

[0397] For example, consider a second-year junior high school math class where the topic is "quadratic equations."

[0398] 1. The user enters the necessary information about the "quadratic equation" into the terminal.

[0399] 2. The terminal sends the entered information to the server.

[0400] 3. The server searches the database for relevant resources and organizes them, including external resources.

[0401] 4. The generative AI model generates an incorrect answer (e.g., $x^2 + 4x + 4 = (x+2)^2 - 4$, an incorrect expansion).

[0402] 5. The emotion engine adjusts generated incorrect answers and lesson scripts based on the user's emotional data.

[0403] 6. The user checks the lesson script and teaching materials and makes any necessary corrections.

[0404] 7. The user conducts the lesson according to the generated lesson script, presenting incorrect information and encouraging students to solve the problem themselves.

[0405] 8. The device uses voice recognition technology and an emotion engine to record learners' responses and the user's emotions during class.

[0406] 9. The server analyzes this data and provides feedback during and after the lesson.

[0407] Prompt Sentence Examples

[0408] "In an eighth-grade math class, explain what a quadratic equation is and generate an incorrect answer. For example, include the expansion of $(x + 3)^2 = x^2 + 6x + 9$."

[0409] "If the user's emotions are anxiety, explain how to adjust the incorrect answer to the quadratic equation."

[0410] In this way, the present invention aims to adapt lesson content based on the user's emotional state and foster learners' independent problem-solving abilities.

[0411] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0412] Specific processing steps of the system program

[0413] Step 1:

[0414] The user logs into the system from a terminal and enters their authentication information.

[0415] Input: User login information (ID and password)

[0416] Operation:

[0417] The device sends the login information to the server.

[0418] The server receives the login information and authenticates it against a database.

[0419] Output: Authentication result (success or failure), and if successful, the user's personal settings data

[0420] If the authentication is successful, the server sends the user's personal setting information to the terminal, and the terminal displays the user's dashboard.

[0421] Step 2:

[0422] The user enters information such as the class subject, target grade, and topic into the terminal.

[0423] Input: Course content information (subject, target grade, topic, etc.)

[0424] Operation:

[0425] The terminal sends lesson content information to the server.

[0426] The server prepares the received data for analysis.

[0427] Output: Analysis results of lesson content data

[0428] The server checks your input to make sure you have all the necessary information.

[0429] Step 3:

[0430] The server analyzes the input lesson content and searches for related resources.

[0431] Input: Lesson content data

[0432] Operation:

[0433] The server analyzes the lesson content using natural language processing (NLP).

[0434] The server searches the database and the internet for relevant educational resources.

[0435] Output: List of related resources

[0436] The list of found resources is sent to the terminal and made available for the user to review.

[0437] Step 4:

[0438] The server calls the generative AI model and generates incorrect information based on the lesson content and analysis results.

[0439] Input: lesson content data, analysis results

[0440] Operation:

[0441] The generative AI model is called and begins the process of generating incorrect information (wrong answers) related to the lesson content.

[0442] Output: Generated incorrect answer data

[0443] As an example of an incorrect answer, "incorrect expansion of a quadratic equation" is generated.

[0444] Step 5:

[0445] An emotion engine adjusts generated incorrect answers based on the user's emotion data.

[0446] Input: wrong answer data, user emotion data

[0447] Operation:

[0448] The emotional engine analyzes the user's emotional state and adjusts the difficulty and type of incorrect answers.

[0449] Output: Adjusted error data

[0450] For example, if the user shows signs of anxiety, the difficulty level for incorrect answers is set low.

[0451] Step 6:

[0452] The server combines the generated incorrect information with the correct information to generate lesson scripts and teaching materials.

[0453] Input: Adjusted incorrect answer data, correct information data

[0454] Operation:

[0455] The server combines incorrect and correct answers to generate lesson scripts and teaching materials.

[0456] Output: Generated lesson scripts and teaching materials

[0457] The generated teaching materials and scripts are sent to the terminal, and the user confirms them on a confirmation screen.

[0458] Step 7:

[0459] The user starts the lesson through the terminal.

[0460] Input: Verified lesson scripts and materials

[0461] Operation:

[0462] Lessons are conducted based on lesson scripts created by the user, and incorrect information is presented to encourage learners to solve the problems themselves.

[0463] Output: Lesson data, student response data

[0464] The device supports the progress of the class in real time.

[0465] Step 8:

[0466] The device uses voice recognition technology to record students' comments and reactions during class and recognize the user's emotional state.

[0467] Input: Learner's speech, user's facial expression and tone of voice

[0468] Operation:

[0469] The device collects data in real time using voice and emotion recognition technology.

[0470] Output: Learner's speech data, user's emotion data

[0471] The collected data is sent to a server.

[0472] Step 9:

[0473] The server analyzes the recorded data and provides real-time feedback and suggestions for next learning steps to the user.

[0474] Input: Learner's speech data, user's emotion data

[0475] Operation:

[0476] The server analyzes the data and generates next steps and feedback.

[0477] Output: Real-time feedback, suggestions for next learning steps

[0478] Feedback is sent to the terminal and the user is notified.

[0479] Step 10:

[0480] After the lesson ends, the server generates a summary based on the data from the lesson, the learners' reactions, and the user's emotional data.

[0481] Input: lesson data, learner response data, user emotion data

[0482] Operation:

[0483] The server generates a summary of the lesson content to help plan the next lesson.

[0484] Output: Summary report

[0485] A summary is sent to the terminal for the user to review.

[0486] (Application example 2)

[0487] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0488] Conventional educational support systems have difficulty optimizing lesson content according to the user's emotional state, and providing real-time feedback based on the learner's reactions is insufficient. Furthermore, because they do not recommend individually optimized content based on the learner's emotional state, there are issues such as a decrease in learning efficiency and difficulty in maintaining motivation.

[0489] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for a user to input lesson content; means for analyzing the input lesson content and searching for related resources; means for calling a generative AI model that generates erroneous information based on the analysis results; means for adjusting the generated erroneous information; means for generating lesson scripts and teaching materials based on the generated information; means for providing the generated lesson scripts and teaching materials to the user; means for a user to present erroneous information during a lesson to encourage the learner to think; means for recording the learner's responses during the lesson and transmitting them to the server; means for analyzing the recorded responses and providing feedback in real time; means for generating a summary of the lesson content after the lesson ends; emotion recognition means for recognizing the user's emotional state and adapting the lesson content based on the state; and means including a generative AI model that recommends content based on the learner's emotional state in real time. This enables real-time feedback and individually optimized content recommendations based on the user's emotional state and the learner's responses.

[0490] The "means for users to input lesson content" refers to an interface that allows educators to input information such as lesson content, subjects, target grades, and topics.

[0491] "Means for analyzing the input lesson content and searching for related resources" refers to the function of analyzing the input lesson content and automatically searching for and collecting resources such as reference materials, literature, and problem sets related to that content.

[0492] "Means for invoking a generative AI model that generates erroneous information based on the analysis results" refers to the process of using the analysis results of the lesson content to launch an AI model that intentionally generates erroneous information and applying it to that model.

[0493] The "means for adjusting the generated erroneous information" is a function that appropriately adjusts the erroneous information generated by the generative AI model according to the user's emotional state and the learner's level of understanding.

[0494] The "means for generating lesson scripts and teaching materials based on the generated information" is a function that automatically generates scripts and teaching materials to be used in lessons based on the generated erroneous information and accurate information.

[0495] "Means for providing the generated lesson scripts and teaching materials to the user" is a function that provides the generated lesson scripts and teaching materials through an interface so that the educator can check and modify them.

[0496] "Means for users to present incorrect information during class to encourage students to think" refers to a process in which an educator intentionally presents incorrect information as part of a lesson and encourages students to think about and correct the error themselves.

[0497] "Means for recording learners' responses during lessons and sending them to a server" refers to a function that records learners' reactions, comments, facial expressions, and other responses, and sends them to a server for storage.

[0498] "Means for analyzing recorded responses and providing feedback in real time" refers to a function that analyzes recorded responses of learners and provides feedback to educators in real time based on the results.

[0499] "Means for generating a summary of the lesson content after the lesson has ended" is a function that generates a report summarizing the lesson content after the lesson has ended, based on data from the lesson and the students' responses.

[0500] "Emotion recognition means that recognizes the user's emotional state and adapts the lesson content based on that state" is a function that analyzes the educator's emotional state and optimizes the lesson content and progress based on the results.

[0501] "Means including a generative AI model that recommends content based on the learner's emotional state in real time" refers to a process that uses a generative AI model to grasp the learner's emotional state in real time and recommend content appropriate to that state.

[0502] This invention is an educational support system that recognizes the user's emotional state and optimizes lesson content accordingly. The system uses a generative AI model and emotion recognition means to provide real-time feedback and content recommendations based on the learner's reactions and emotions.

[0503] System configuration

[0504] The system includes the following major hardware and software components:

[0505] 1. User Interface (Terminal):

[0506] This is a device that allows educators to input lesson content and check and correct the generated scripts and teaching materials. This interface can be configured as a smartphone, tablet, or PC.

[0507] 2. Server:

[0508] This is the central computer that processes and stores data, calls generative AI models, and analyzes learner responses and user emotions. It runs on a cloud server.

[0509] 3. Generative AI Model:

[0510] This algorithm generates false information based on the content of lessons, providing learners with an opportunity to think independently. This model is implemented using AI frameworks such as Python and TensorFlow.

[0511] 4. Emotion Recognition Engine:

[0512] This algorithm recognizes the emotional state of users and learners from data acquired through cameras and microphones, and adjusts lesson content accordingly. It uses libraries such as OpenCV and dlib.

[0513] System Operation

[0514] User Authentication

[0515] The educator logs in to the system through the terminal and enters authentication information. The terminal sends the entered information to the server, which then authenticates the user. If authentication is successful, the server retrieves the user's personal settings and history and sends them to the terminal.

[0516] Input and generation of lesson content

[0517] The educator inputs the lesson content into the device. The input content is sent to the server, which analyzes the information and calls a generative AI model to generate incorrect answers. The generated incorrect answers are adjusted based on the user's emotional state using an emotion recognition engine. Lesson scripts and teaching materials are then automatically generated and provided to the educator.

[0518] Class implementation and response analysis

[0519] During the lesson, the educator follows the generated script and presents incorrect answers to encourage students to think. The device uses voice recognition technology to record what the student says and simultaneously recognizes the user's emotional state. This data is sent to the server and analyzed in real time. The server then provides real-time feedback to the educator based on this data.

[0520] Generate a summary after the lesson

[0521] After the lesson, the server generates a summary based on the data from the lesson, the learners' reactions, and the user's emotional data, and sends it to the device. The educator can then review the summary and incorporate it into their next lesson plan.

[0522] Specific examples

[0523] For example, consider a second-year junior high school math class with the topic of "quadratic equations." The educator selects "quadratic equations" as the topic and inputs the necessary information into the system. The server searches for related resources, and the generative AI model generates an incorrect answer (e.g., $x^2 + 4x + 4 = (x+2)^2 - 4$, an incorrect expansion). The emotion recognition engine adjusts the generated incorrect answer and lesson script based on the user's emotional data. The educator presents this incorrect answer during the lesson, encouraging the learner to derive the correct answer on their own. The device uses voice recognition technology and an emotion recognition engine to record the learner's reactions and the user's emotions during the lesson. After the lesson ends, the server analyzes this data and generates a summary to provide to the educator.

[0524] Prompt Sentence Examples

[0525] "If an emotion recognition engine detects that a user is feeling stressed, create a program that recommends relaxing music or movies to watch in that state."

[0526] As described above, the present invention is a system that can improve learning effectiveness and motivation by optimizing the method of providing lessons and content based on the emotional states of both the user and the learner.

[0527] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0528] Step 1:

[0529] The user inputs the lesson content (input content: lesson subject, target grade, topic, etc.)

[0530] The user inputs the lesson contents into the terminal interface and sends the input data to the server, which receives the data and prepares it for the next analysis step.

[0531] Step 2:

[0532] The server analyzes the input lesson content and searches for related resources (input content: lesson content data, output content: related resources)

[0533] The server analyzes the input lesson content using a text analysis algorithm, searches for related educational resources (reference materials, workbooks, etc.), and creates a list.

[0534] Step 3:

[0535] The server calls a generative AI model that generates incorrect information based on the analysis results (input: analysis results, output: incorrect information).

[0536] The server inputs the analysis results into a generative AI model, which is built using AI frameworks such as Python and TensorFlow, to generate false information related to the lesson content.

[0537] Step 4:

[0538] Adjusting the generated misinformation (input: misinformation, user's emotional state, output: adjusted misinformation)

[0539] The server uses an emotion recognition engine to analyze the user's emotional state and adjusts the generated erroneous information based on that state, adjusting the level of erroneous answers to an appropriate level according to the user's stress or excitement level.

[0540] Step 5:

[0541] Generate lesson scripts and teaching materials based on the generated information (input: adjusted incorrect information, correct information; output: lesson scripts, teaching materials)

[0542] The server automatically generates lesson scripts and teaching materials by combining the correct information with the adjusted incorrect information. These scripts and teaching materials are output in a format that users can review and edit.

[0543] Step 6:

[0544] The generated lesson scripts and teaching materials are provided to the user (input content: lesson script, teaching materials, output content: displayed on the user's device)

[0545] The server sends the generated lesson scripts and teaching materials to the user's terminal, where the user can check them and make corrections if necessary.

[0546] Step 7:

[0547] The user presents incorrect information during class to encourage students to think about it (input: class script, teaching materials, output: encourage students to think)

[0548] During class, the user presents incorrect information according to the generated script and prompts the learner to solve it, thereby developing the learner's independent problem-solving ability.

[0549] Step 8:

[0550] During the lesson, learners' responses are recorded and sent to the server (input: learners' responses, output: recorded data)

[0551] The device uses a microphone and camera to record the learner's responses and transmits the data to a server, using voice recognition and video analysis technology.

[0552] Step 9:

[0553] Analyze recorded responses and provide real-time feedback (input: recorded data, output: feedback)

[0554] The server analyzes the recorded student response data, generates real-time feedback, and sends it to the device, helping the user to progress through the lesson.

[0555] Step 10:

[0556] Generate a summary of the lesson content after the lesson ends (input content: data from the lesson, output content: summary)

[0557] The server generates a summary of the lesson content based on data from the lesson, learners' reactions, and user emotional data, and sends it to the user's device. The user can review this summary and reflect it in their next lesson plan.

[0558] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0559] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0560] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0561] [Second embodiment]

[0562] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0563] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0564] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0565] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0566] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0567] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0568] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0569] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0570] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0571] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0572] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0573] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0574] This invention relates to "Empower GPT," a system for supporting teachers in the educational field. This system uses generative AI to provide false information, aiming to develop learners' ability to solve problems independently.

[0575] System Overview

[0576] The system includes the following main components:

[0577] 1. User interface (terminal): A device that allows teachers to input lesson content and check and modify the generated scripts and teaching materials.

[0578] 2. Server: A central computer that processes and stores data, calls generative AI models, and analyzes learner responses.

[0579] 3. Generative AI model: An artificial intelligence algorithm that generates misinformation based on lesson content, providing learners with an opportunity to think for themselves.

[0580] Program processing procedure

[0581] 1. User login and authentication:

[0582] The user logs into the system from a terminal and enters the necessary authentication information.

[0583] The server verifies the authentication information and sets up the appropriate environment for each user.

[0584] 2. Input lesson content:

[0585] The user inputs information such as the class subject, target grade, topic, etc. from the terminal.

[0586] The terminal transmits this information to the server.

[0587] 3. Analyzing information and finding resources:

[0588] The server analyzes the input lesson content and searches for related resources (teaching materials, past lesson content, etc.).

[0589] 4. Creation of misinformation:

[0590] The server calls the generative AI model and generates false information based on the analysis results.

[0591] The generated misinformation is tailored to the learner's level of understanding and grade level.

[0592] 5. Generating lesson scripts and teaching materials:

[0593] The server generates lesson scripts and teaching materials based on the generated incorrect information and correct information.

[0594] The generated scripts and teaching materials are sent to the terminal and checked by the user.

[0595] 6. Class Conduct:

[0596] A user presents false information during a lesson to initiate a process of thinking in the learners.

[0597] Generative AI models respond to users' questions and concerns in real time.

[0598] 7. Recording and analyzing learner responses:

[0599] The device uses voice recognition technology to record the learner's responses.

[0600] The recorded data is sent to a server, which analyzes it.

[0601] 8. Providing Feedback:

[0602] The server provides real-time feedback based on the analysis results and suggests next steps to the learner.

[0603] 9. Post-class summary generation:

[0604] After the lesson ends, the server generates a summary based on the lesson content and the learners' responses.

[0605] The summary is sent to the terminal, where the user can review it and reflect it in the next lesson plan.

[0606] Specific examples

[0607] For example, consider a case where the topic of a second-year junior high school math class is "quadratic equations." When a user selects "quadratic equations" as the topic, the server searches for related resources, and the generative AI model generates an incorrect answer (e.g., $x^2 + 4x + 4 = (x+2)^2 - 4$, an incorrect expansion). The user presents this incorrect answer during class, encouraging students to derive the correct answer on their own. After the class, the server generates a summary based on the content of the class and the students' responses, which can be used for the next class.

[0608] In this way, the present invention provides a system that promotes independent learning by learners in educational settings and reduces the burden on teachers.

[0609] The processing flow will be explained below.

[0610] Step 1: User Login and Authentication

[0611] The user enters the information (user name, password) required to log in to the system from the terminal.

[0612] The terminal sends the entered login information to the server.

[0613] The server checks the received login information against a database and performs authentication.

[0614] If authentication is successful, the server obtains the user's personal settings and history and sends them to the terminal.

[0615] The user confirms successful authentication through the terminal and is then able to access the system.

[0616] Step 2: Enter lesson content

[0617] The user enters information such as the class subject, target grade, and topic from the terminal.

[0618] The terminal sends the input lesson content to the server.

[0619] Step 3: Analyze the information and find resources

[0620] The server analyzes the input lesson content.

[0621] Based on the analysis results, the server searches for relevant resources (teaching materials, past lesson contents, reference materials, etc.) and retrieves them from the database.

[0622] Step 4: Creating misinformation

[0623] The server calls the generative AI model and generates incorrect information based on the lesson content and analysis results.

[0624] The generated erroneous information is appropriately adjusted according to the learner's level of understanding and the target grade level.

[0625] Step 5: Generate lesson scripts and teaching materials

[0626] The server combines the generated incorrect information with the correct information to generate lesson scripts and teaching materials.

[0627] The generated lesson scripts and teaching materials are sent to the terminal, where the user can check and modify them.

[0628] Step 6: Conducting lessons

[0629] The user starts the lesson through the terminal.

[0630] The user follows a generated lesson script to intentionally present incorrect information and encourage learners to think.

[0631] During lessons, the generative AI model responds to and supports users' questions and concerns in real time.

[0632] Step 7: Record and analyze learner responses

[0633] The device uses voice recognition technology to record students' comments and reactions during class.

[0634] The recorded data is sent to a server, which analyzes it.

[0635] Based on the analysis results, the server provides real-time feedback and suggests next learning steps to the user.

[0636] Step 8: Generate a summary after the lesson

[0637] After the lesson ends, the server generates a summary based on the data from the lesson and the learners' responses.

[0638] The generated summary is sent to the terminal, where the user can check the lesson content and the learners' level of understanding and reflect it in future lesson plans.

[0639] Example 1

[0640] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0641] In today's educational settings, there are no fully established methods for teachers to provide appropriate teaching materials and misinformation according to each student's level of understanding, and for students to develop the ability to solve problems independently based on that information. Furthermore, it is difficult to analyze students' responses in real time and provide appropriate feedback. This increases the burden on teachers and hinders the promotion of independent learning among students, creating the problem of a lack of understanding.

[0642] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0643] In this invention, the server includes: means for a user to input lesson content; means for analyzing the input lesson content and searching for related resources; means for calling a generative AI model that generates erroneous information based on the analysis results; means for adjusting the generated erroneous information; means for generating lesson scripts and teaching materials based on the generated information; means for providing the generated lesson scripts and teaching materials to the user; means for a user to present erroneous information during class to encourage learners to think; means for recording learners' responses during class and transmitting them to the server; means for analyzing the recorded responses and providing feedback in real time; means for generating a summary of the lesson content after the class ends; means for adjusting the difficulty level of the generated erroneous information according to the learners' level of understanding; and means including a generative AI model that responds to user questions and doubts in real time during class. This reduces the burden on teachers and promotes learners' independent learning.

[0644] "Users" refers to teachers and educators who use this system in educational settings.

[0645] "Lesson content" refers to information such as subject, target grade, and topic that teachers enter into the system.

[0646] "Analysis" refers to the process of breaking down input lesson content using natural language processing techniques or other methods to identify relevant resources.

[0647] "Resources" refers to teaching materials, reference materials, past lesson data, etc. related to the content of the lesson.

[0648] A "generative AI model" refers to the part of a system that uses artificial intelligence algorithms to generate false information based on specific inputs.

[0649] A "prompt" refers to input text used to instruct a generative AI model on what information to generate.

[0650] "Misinformation" refers to incorrect data or answers that are intentionally generated as a starting point for learners to use their own understanding to derive accurate information.

[0651] A "script" refers to a script or instruction manual that helps guide a lesson.

[0652] "Instructional materials" refers to visual, textual, or hands-on learning materials used to supplement course content.

[0653] "Terminal" refers to a device such as a computer or tablet that teachers use to access and operate the system.

[0654] "Server" refers to the central computing system that processes and stores data, invokes generative AI models, analyzes learner responses, etc.

[0655] "Speech recognition technology" refers to technology that converts learners' speech into text.

[0656] "Feedback" refers to instructions, hints, advice, etc. provided based on the learner's responses.

[0657] A "summary" refers to a short report generated based on the lesson content and learner responses.

[0658] "Responses" refer to the level of understanding, questions, comments, etc. that learners show during class.

[0659] This invention relates to a system called "Empower GPT" that supports teachers in the educational field. This system uses generative AI to provide false information and aims to cultivate learners' ability to solve problems independently. The following describes in detail the mode for carrying out the invention.

[0660] System configuration

[0661] The system includes the following main components:

[0662] 1. User interface (terminal): A device on which teachers input lesson content and check and edit generated scripts and teaching materials. Examples include PCs, tablets, and smartphones.

[0663] 2. Server: This is the central computer that processes and stores data, calls the generative AI model, and analyzes learner responses. In this invention, the server is assumed to be a high-performance computer with a Linux-based OS. MySQL, PostgreSQL, or other databases are used.

[0664] 3. Generative AI model: An artificial intelligence algorithm that generates false information based on the lesson content input by the teacher, providing learners with an opportunity to think for themselves. Specific examples include OpenAI's GPT-3 and GPT-4.

[0665] Program processing procedure

[0666] The system begins when the user (teacher) logs in and inputs the lesson content. The specific processing steps are explained below.

[0667] 1. User login and authentication:

[0668] A user logs in to the system from a terminal and enters the necessary authentication information, such as a user ID and password.

[0669] The server verifies the authentication information and sets up the appropriate environment for each individual user, a process that involves the server verifying user information against a database.

[0670] 2. Input lesson content:

[0671] The user inputs information such as the class subject, target grade, topic, etc., into the terminal. For example, they input "mathematics," "second-year junior high school student," and "quadratic equations."

[0672] The device sends this information to the server, and the data is sent in JSON format.

[0673] 3. Analyzing information and finding resources:

[0674] The server analyzes the input lesson content and searches the database for related resources (teaching materials, past lesson content, etc.) using natural language processing libraries (e.g., NLTK and spaCy).

[0675] 4. Creation of misinformation:

[0676] The server calls the generative AI model and generates false information based on the analysis results. The generative AI model is instructed using prompts.

[0677] The generated misinformation is tailored to the learner's level of understanding and grade level.

[0678] 5. Generating lesson scripts and teaching materials:

[0679] The server generates lesson scripts and teaching materials based on the generated incorrect information and correct information. For example, it uses Python or JavaScript scripts to build teaching materials.

[0680] The generated scripts and teaching materials are sent to the terminal and checked by the user.

[0681] 6. Class Conduct:

[0682] The user presents incorrect information during a lesson to initiate the thinking process in the learner, for example, by using an incorrect formula such as "x^2 + 4x + 4 = (x+2)^2 - 4".

[0683] The generative AI model responds to user questions and concerns in real time, including when users type questions into the generative AI, which then responds instantly.

[0684] 7. Recording and analyzing learner responses:

[0685] The device records the learner's responses using voice recognition technology, such as Google Speech-to-Text or Amazon Transcribe.

[0686] The recorded data is sent to a server, which analyzes it using machine learning algorithms (e.g., scikit-learn or TensorFlow).

[0687] 8. Providing Feedback:

[0688] The server provides real-time feedback based on the analysis results and suggests next steps to the learner, such as "Many students got this question wrong, so let's explain it again."

[0689] 9. Post-class summary generation:

[0690] After the lesson, the server generates a summary based on the lesson content and the learners' responses. The summary is created using natural language generation technology.

[0691] The summary is sent to the terminal, where the user can review it and reflect it in the next lesson plan.

[0692] Specific examples

[0693] For example, if the topic of a second-year junior high school math class is "quadratic equations," when the user selects "quadratic equations" as the topic, the server searches for related resources and the generative AI model generates an incorrect answer (e.g., $x^2 + 4x + 4 = (x+2)^2 - 4$, an incorrect expansion). The user presents this incorrect answer during class, encouraging the learners to derive the correct answer on their own. After the class, the server generates a summary based on the content of the class and the learners' responses, which can be used for the next class.

[0694] Examples of prompt statements

[0695] Topic: 8th Grade Math - Quadratic Equations

[0696] Objectives: Students will understand how to solve quadratic equations and be able to correct incorrect information themselves.

[0697] Examples of misinformation it generates:

[0698] $x^2 ​​+ 4x + 4 = (x+2)^2 - 4$

[0699] "The only solution to a quadratic equation is x = 2"

[0700] In this way, the present invention provides a system that promotes independent learning by learners in educational settings and reduces the burden on teachers.

[0701] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0702] Step 1:

[0703] A user accesses the system from a terminal and enters login information (user ID and password). The input data is encrypted on the terminal as authentication information and sent to the server. The server receives the authentication information and checks it against a database. Based on the check result, if authentication is successful, the environment settings corresponding to the user are loaded and a dashboard screen is displayed on the terminal. The output is a dashboard screen that the user can access.

[0704] Step 2:

[0705] The user inputs the class subject, target grade, and specific topic into the device. The input data includes information such as "Mathematics," "Second Year of Junior High School," and "Quadratic Equations." The device sends the input data in JSON format to the server. The server temporarily stores the received data and prepares for the analysis step. The output is a notification that the data has been saved on the server.

[0706] Step 3:

[0707] The server analyzes the received lesson content data. A natural language processing library (e.g., NLTK or spaCy) is used for the analysis. The input for the analysis is JSON data about the lesson content. As a result of the analysis, search keywords and tags for related resources (teaching materials, past lesson content, etc.) are generated. The output is the search keywords and tags.

[0708] Step 4:

[0709] The server searches for related resources from databases and cloud storage based on the analysis results. The input in this step is search keywords and tags. Related resources (PDF teaching materials, video links, past lesson scripts, etc.) are obtained as search results. The output is a list of the obtained resources.

[0710] Step 5:

[0711] The server invokes a generative AI model to generate false information using a prompt. The input in this step is the analysis result of the lesson content and the prompt. The generative AI model (e.g., GPT-3 or GPT-4) generates false information based on the input prompt. The output is the generated false information.

[0712] Step 6:

[0713] The server adjusts the generated misinformation according to the learner's level of understanding and grade level. The input is the generated misinformation and the learner's profile data. The adjusted information is generated and output. The adjustment may use machine learning algorithms.

[0714] Step 7:

[0715] The server generates lesson scripts and teaching materials based on the adjusted misinformation and correct information. The input data is the adjusted misinformation and related correct information. The lesson scripts and teaching materials are generated using programming languages ​​such as Python and JavaScript. The output is the completed lesson scripts and teaching materials.

[0716] Step 8:

[0717] The server sends the generated lesson script and teaching materials to the terminal, where the user can review them. The output is a script and teaching materials that the user can review on the terminal. The user can then modify the content as needed.

[0718] Step 9:

[0719] The user presents false information generated during class to prompt students to think. The input is the adjusted false information. As a concrete example, a teacher writes "$x^2 + 4x + 4 = (x+2)^2 - 4$" on the blackboard and asks students to think about what is wrong. The output is the student's response.

[0720] Step 10:

[0721] The device uses voice recognition technology (e.g., Google Speech-to-Text or Amazon Transcribe) to record the learner's responses and send them to the server. The input is the learner's speech. The voice data is converted into text format, and the output is the text data and its analysis results.

[0722] Step 11:

[0723] The server analyzes the recorded responses of the learner and provides feedback in real time. The input is the learner's response data in text format. Based on the analysis results, appropriate feedback is generated, and the output is the content of the feedback.

[0724] Step 12:

[0725] After the lesson ends, the server generates a summary based on the lesson content and learner responses. The input is the lesson script, teaching materials, and learner response data. The summary is created using natural language generation technology. The output is a summary that is sent to the terminal and reviewed by the user.

[0726] (Application example 1)

[0727] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0728] In conventional educational systems, when a single teacher instructs a large number of students, it is difficult to grasp each student's level of understanding and reactions in real time and provide appropriate feedback. Furthermore, in order for students to develop their independent problem-solving skills, they need teaching materials that contain not only accurate information but also incorrect information, but generating and adjusting such materials is time-consuming. This increases the burden on teachers and makes it difficult to improve students' independent learning abilities.

[0729] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0730] In this invention, the server includes a means for users to input lesson content, a means for analyzing the input lesson content and searching for related resources, and a means for calling a generative AI model that generates erroneous information based on the analysis results. This makes it possible to generate and adjust erroneous information and answers and provide appropriate feedback in real time. Furthermore, the server includes a means for users to select the content they want to learn and provide teaching materials containing erroneous information, a means for providing real-time feedback to students based on the generated erroneous information and answers, and a means for recording and saving learned content and answers so that they can be reviewed later. This facilitates the improvement of learners' independent learning abilities while reducing the burden on teachers.

[0731] "Means for users to input lesson content" refers to interfaces and applications that allow teachers and students to input the subjects or topics they want to learn into digital devices.

[0732] "Means for analyzing input lesson content and searching for related resources" refers to an algorithm that searches and retrieves related information and materials from databases or networks based on the input teaching materials or topics.

[0733] "Means for invoking a generative AI model that generates erroneous information based on the analysis results" refers to a mechanism that uses the analysis results of the input content to execute a process that intentionally creates erroneous information using a generative AI model.

[0734] The "means for adjusting generated erroneous information" is a function for appropriately adjusting the difficulty and content of the generated erroneous information according to the learner's level of understanding and comprehension.

[0735] The "means for generating lesson scripts and teaching materials based on the generated information" is a system that combines incorrect information with correct information to create and provide scripts and educational materials to be used in actual lessons.

[0736] The "means for providing the generated lesson scripts and teaching materials to users" refers to a mechanism for distributing the generated lesson scripts and teaching materials to teachers and learners via digital devices.

[0737] "Means for users to present incorrect information during class and encourage students to think" is a process in which a teacher intentionally presents incorrect information during class and has students point out the errors.

[0738] "Means for recording learners' responses during class and sending them to a server" refers to a system that has the function of recording how learners respond during class and sending and saving them to a server.

[0739] The "means for analyzing recorded responses and providing real-time feedback" refers to an algorithm that analyzes recorded responses of learners and provides appropriate feedback immediately.

[0740] The "means for generating a summary of the content of a lesson after the lesson has ended" is a system that summarizes the content of the lesson after the lesson has ended, generates a summary, and provides it.

[0741] "Means for providing students with real-time feedback based on incorrect information or answers" refers to a system that provides supplementary information or advice in real time to learners as they solve problems, based on the incorrect information or answers they generate.

[0742] "A means to record and save what has been learned and answers, and to review them later" is a system that has the function of saving what learners have learned and answers in a database, allowing them to check and review them later.

[0743] This invention relates to an educational support system that utilizes a generative AI model. The purpose of this invention is to promote learners' independent learning and reduce the burden on teachers by cooperating with three main components: a server, a terminal, and a user.

[0744] System configuration

[0745] 1. Server:

[0746] The server has the following functions:

[0747] Provides a means for users to input lesson content.

[0748] It has the function of analyzing the input lesson content and searching for related resources.

[0749] Based on the analysis results, a generative AI model is invoked to generate false information.

[0750] The generated incorrect information is adjusted and lesson scripts and teaching materials are generated.

[0751] Analyze recorded learner responses and provide feedback in real time.

[0752] 2. Terminal:

[0753] Terminals are devices used by teachers and students and have the following functions:

[0754] It provides a user interface for inputting, checking, and modifying lesson content.

[0755] Display the generated lesson scripts and teaching materials.

[0756] Learners' responses are recorded during the lesson and sent to the server.

[0757] 3. User:

[0758] Users are mainly teachers and students, and use the system in the following steps:

[0759] Enter the lesson content from the terminal.

[0760] Check and modify the generated scripts and teaching materials.

[0761] Present misinformation during class and encourage students to problem-solve.

[0762] Record learners' responses and review their feedback after the lesson.

[0763] Software and Hardware Configuration

[0764] On the server side, it uses high-performance cloud servers such as Amazon Web Services and Google Cloud Platform, and the backend is built using Node.js and Express, with MongoDB as the database.

[0765] The front-end part will be developed using React Native, taking into consideration use on smartphones.

[0766] The generative AI model uses OpenAI GPT-4, which generates prompts based on the input lesson content and then creates false information as a response to those prompts.

[0767] As a use case, consider a 8th grade math lesson on "quadratic equations," where the following prompt is sent to the generative AI model:

[0768] "Generate false information and answers on the following topic: Quadratic Equations"

[0769] The generative AI model generates incorrect information for this prompt (e.g., an incorrect expansion such as $x^2 + 4x + 4 = (x+2)^2 - 4$), which is then provided as part of the learning material.

[0770] The server analyzes the learner's responses and provides real-time feedback, for example, helping the learner recognize incorrect information and guide them in the process of arriving at the correct answer.

[0771] This system will enable students to develop the ability to correct their own mistakes and reduce the burden on teachers. It is also expected that the inclusion of incorrect information will dramatically improve students' thinking and problem-solving abilities.

[0772] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0773] Step 1:

[0774] The user inputs the lesson content.

[0775] In this step, the user enters information such as the class subject, target grade, topic, etc. into the application via their device. The entered information is sent from the front end (React Native) to the server (Node.js).

[0776] Input: Course subject, target grade, topic

[0777] Output: The input data is sent to the server

[0778] Step 2:

[0779] The server analyzes the content of the lesson and searches for related resources.

[0780] The server analyzes the received lesson content and searches a database (MongoDB) for teaching materials and past resources related to that topic.

[0781] Input: Lesson content sent in Step 1

[0782] Output: A list of resources related to the analysis results

[0783] Step 3:

[0784] The server calls a generative AI model based on the analysis results and generates incorrect information.

[0785] Based on the analysis results, the server sends a prompt to the generative AI model (OpenAI GPT-4) to generate incorrect information.

[0786] Input: Analysis results, prompt (e.g. "Generate incorrect information about quadratic equations")

[0787] Output: The incorrect information generated

[0788] Step 4:

[0789] The server will adjust the generated incorrect information.

[0790] The server runs an algorithm that adjusts the generated misinformation according to the learner's level of comprehension and grade level, optimizing the difficulty of the misinformation.

[0791] Input: Generated misinformation, learner comprehension information

[0792] Output: Adjusted misinformation

[0793] Step 5:

[0794] The server generates lesson scripts and teaching materials based on the generated information.

[0795] The server generates lesson scripts and teaching materials based on the adjusted incorrect information and correct information and provides them to users.

[0796] Input: Adjusted incorrect information, correct information

[0797] Output: Lesson scripts, teaching materials

[0798] Step 6:

[0799] The user presents incorrect information during class to prompt students to think.

[0800] The user uses the generated teaching materials in class, presents incorrect information to students, and has them point out the errors, thereby encouraging the students to think independently.

[0801] Input: lesson scripts, teaching materials

[0802] Output: Learner response

[0803] Step 7:

[0804] The device records the learners' responses during class and sends them to the server.

[0805] Learners' responses and reactions are recorded by the device and sent to the server in real time. Speech recognition technology is also used to record utterances.

[0806] Input: Learner response data

[0807] Output: Response data sent to the server

[0808] Step 8:

[0809] The server analyzes the recorded responses and provides real-time feedback.

[0810] The server analyzes the received learner response data and uses an AI model to generate and provide instant feedback.

[0811] Input: Reaction data

[0812] Output: Real-time feedback

[0813] Step 9:

[0814] After the lesson ends, the server generates a summary of the lesson content.

[0815] After the lesson is over, the server creates a summary based on the lesson content and the learners' reactions and provides it to the user to help plan the next lesson.

[0816] Input: Lesson content, learner response data

[0817] Output: Lesson summary

[0818] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0819] This invention relates to "Empower GPT," an educational support system with the ability to recognize users' emotions and adapt lesson scripts and teaching materials. The system aims to optimize lessons based on the user's emotional state while providing false information using generative AI to foster learners' ability to solve problems independently.

[0820] System Overview

[0821] The system includes the following main components:

[0822] 1. User interface (terminal): A device that allows teachers to input lesson content and check and modify the generated scripts and teaching materials.

[0823] 2. Server: The central computer that processes and stores data, invokes generative AI models, and analyzes learner responses and user sentiment.

[0824] 3. Generative AI model: An artificial intelligence algorithm that generates misinformation based on lesson content, providing learners with an opportunity to think independently.

[0825] 4. Emotion Engine: An algorithm to recognize the user's emotional state and adjust the lesson script and teaching materials.

[0826] Program processing procedure

[0827] 1. User login and authentication:

[0828] The user logs into the system from a terminal and enters their authentication information.

[0829] The terminal sends the entered login information to the server, and the server performs authentication.

[0830] If authentication is successful, the server obtains the user's personal settings and history and sends them to the device.

[0831] 2. Input lesson content:

[0832] The user enters information such as the class subject, target grade, and topic into the terminal.

[0833] The terminal transmits this information to the server.

[0834] 3. Analyzing information and finding resources:

[0835] The server analyzes the input lesson content and searches for and organizes related resources.

[0836] 4. Creation of misinformation:

[0837] The server calls the generative AI model and generates incorrect information based on the lesson content and analysis results.

[0838] The generative AI model adjusts the generated incorrect answers based on the user's emotional data.

[0839] 5. Generating lesson scripts and teaching materials:

[0840] The server combines the generated incorrect information with the correct information to generate lesson scripts and teaching materials.

[0841] The generated lesson scripts and teaching materials are sent to the terminal, where the user can check and modify them.

[0842] 6. Class Conduct:

[0843] The user starts the lesson through the terminal.

[0844] The user presents incorrect information according to the generated lesson script, prompting learners to think.

[0845] Generative AI models respond to users' questions and concerns in real time.

[0846] 7. Recording and analyzing learner responses and user emotions:

[0847] The device uses voice recognition technology to record students' comments and reactions during class.

[0848] The terminal recognizes the user's emotional state and transmits the emotional data to the server.

[0849] The recorded data and emotional data are sent to a server for analysis.

[0850] Based on the analysis results, the server provides real-time feedback and suggests next learning steps to the user.

[0851] 8. Post-class summary generation:

[0852] After the lesson ends, the server generates a summary based on the data from the lesson, the learners' reactions, and the user's emotional data.

[0853] A summary is sent to the terminal, and the user checks the lesson content and the learners' level of understanding, and reflects this in the next lesson plan.

[0854] Specific examples

[0855] For example, consider a case where the topic of a math class for second-year junior high school students is "quadratic equations." The user selects "quadratic equations" as the topic and enters the necessary information into the system. The server searches for related resources, and the generative AI model generates an incorrect answer (e.g., $x^2 + 4x + 4 = (x+2)^2 - 4$, an incorrect expansion). The emotion engine adjusts the generated incorrect answers and lesson scripts based on the user's emotional data.

[0856] The user presents the incorrect answer during the lesson, encouraging the learner to find the correct answer on their own. The device uses voice recognition technology and an emotion recognition engine to record the learner's responses and the user's emotions during the lesson. After the lesson ends, the server analyzes this data, generates a summary, and provides it to the user. In this way, it is possible to promote learners' independent learning and conduct lessons that are tailored to the user's emotional state.

[0857] The processing flow will be explained below.

[0858] Step 1: User Login and Authentication

[0859] The user enters the information (user name, password) required to log in to the system from the terminal.

[0860] The terminal sends the entered login information to the server.

[0861] The server checks the received login information against a database and performs authentication.

[0862] If authentication is successful, the server obtains the user's personal settings and history and sends them to the terminal.

[0863] The user can then verify their authentication through the terminal and gain access to the system.

[0864] Step 2: Enter lesson content

[0865] The user enters information such as the class subject, target grade, and topic from the terminal.

[0866] The terminal sends the input lesson content to the server.

[0867] Step 3: Analyze the information and find resources

[0868] The server analyzes the input lesson content.

[0869] The server searches for and retrieves relevant resources (teaching materials, past lesson contents, reference materials, etc.) based on the analysis results.

[0870] Step 4: Creating misinformation

[0871] The server calls the generative AI model and generates incorrect information based on the lesson content and analysis results.

[0872] The server adjusts the difficulty and content of the generated erroneous information according to the learner's level of understanding.

[0873] Step 5: Emotional Engine Alignment

[0874] The device monitors the user's emotional state in real time and transmits the emotional data to a server.

[0875] The server uses an emotion engine to analyze the emotional data and adjust the generated incorrect answers and lesson scripts.

[0876] Step 6: Generate lesson scripts and teaching materials

[0877] The server combines the generated incorrect information with the correct information to generate lesson scripts and teaching materials.

[0878] The generated lesson scripts and teaching materials are sent to the terminal, where the user can check and modify them.

[0879] Step 7: Conducting lessons

[0880] The user starts the lesson through the terminal.

[0881] The user follows a generated lesson script to intentionally present incorrect information and encourage learners to think.

[0882] During lessons, the generative AI model responds to users' questions and concerns in real time.

[0883] Step 8: Recording and analyzing learner responses and user emotions

[0884] The device uses voice recognition technology to record students' comments and reactions during class.

[0885] The terminal recognizes the user's emotional state and transmits the emotional data to the server.

[0886] The server analyzes the recorded learner responses and user emotion data.

[0887] Based on the analysis results, the server provides real-time feedback and suggests next learning steps to the user.

[0888] Step 9: Generate a summary after the lesson

[0889] After the lesson ends, the server generates a summary based on the data from the lesson, the learners' reactions, and the user's emotional data.

[0890] The generated summary is sent to the terminal, where the user can check the lesson content and the learners' level of understanding and reflect it in the next lesson plan.

[0891] Example 2

[0892] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0893] Traditional educational systems lacked effective methods for cultivating students' independent problem-solving skills. They also lacked mechanisms for adapting lesson content based on the user's emotional state, making it difficult to maximize educational effectiveness. As a result, they were unable to respond to fluctuations in students' interests and comprehension levels, limiting their ability to provide consistent educational content.

[0894] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0895] In this invention, the server includes means for a user to input lesson content, means for analyzing the input lesson content and searching for related resources, means for calling an AI algorithm that generates erroneous information based on the analysis results, means for adjusting the generated erroneous information, means for generating lesson scripts and teaching materials based on the generated information, means for providing the generated lesson scripts and teaching materials to the user, means for the user to present erroneous information during the lesson to encourage students to think, means for recording students' responses during the lesson and transmitting them to a central computer, means for analyzing the recorded responses and providing feedback in real time, means for generating a summary of the lesson content after the lesson ends, and means for recognizing the user's emotional state and adjusting the lesson scripts and teaching materials, thereby cultivating students' independent problem-solving skills and enabling the lesson content to be adapted based on the user's emotional state.

[0896] "Users" are educators and administrators who operate the system and input lesson content.

[0897] "Class content" refers to educational information such as the subjects, topics, and target grades covered in the class.

[0898] "Input means" refers to an interface or device that allows users to input lesson content into the system.

[0899] "Analysis means" refers to algorithms or devices that analyze the input lesson content and search for related resources.

[0900] "Misinformation" refers to intentionally inaccurate educational information provided to learners in order to solve problems independently.

[0901] "Artificial intelligence algorithms" refer to machine learning and natural language processing technologies that generate false information based on input information.

[0902] "Adjustment means" refers to processes or functions for adjusting erroneous information generated by artificial intelligence algorithms.

[0903] A "lesson script" is a script or plan used in a lesson, and is a document or digital data that contains the content as the lesson progresses.

[0904] "Teaching materials" refers to the teaching materials and learning resources used in classes (e.g., textbooks, workbooks, video materials, etc.).

[0905] "Recording means" refers to voice recognition technology or other digital recording devices used to record learner responses during lessons.

[0906] The "central computer" is the main device in the system that processes and stores data, calls generative AI models, and analyzes learner responses and user emotions.

[0907] "Analysis tools" are algorithms or devices that analyze recorded learner responses and provide real-time feedback.

[0908] An "emotional state" is a state that recognizes and classifies a user's emotions (e.g., joy, sadness, anger, surprise, etc.).

[0909] A "summary generation means" refers to an algorithm or device that creates a summary after a lesson based on the lesson content and learners' responses.

[0910] "Real-time feedback" refers to guidance and advice provided immediately or briefly during or after a lesson.

[0911] MODE FOR CARRYING OUT THE INVENTION

[0912] In the present invention, the following hardware and software are used to effectively implement the education support system.

[0913] (System Configuration)

[0914] 1. User Interface (Terminal)

[0915] 2. Server

[0916] 3. Generative AI Models

[0917] 4. Emotion Engine

[0918] (User Interface (Terminal))

[0919] A user interface is a device that allows users to input lesson content and review and edit generated scripts and teaching materials. This can be a computer, tablet, smartphone, or other digital device. It has a solid graphical user interface (GUI) and is designed to allow users to operate it intuitively.

[0920] (server)

[0921] The server is the central computer that processes and stores data, calls the generative AI model, and analyzes learner responses and user emotions. It uses a cloud server with powerful computing capabilities, and is able to exchange data with devices in real time via the communications infrastructure. The server integrates a database, analysis engine, and various algorithms.

[0922] (Generative AI model)

[0923] The generative AI model is an artificial intelligence algorithm that generates false information based on input lesson content, providing learners with an opportunity to think independently. It incorporates natural language processing (NLP) and machine learning (ML) technologies and has the ability to adaptively adjust the information it generates based on the user's emotional data.

[0924] (Emotion Engine)

[0925] The emotion engine is an algorithm that recognizes the user's emotional state and adjusts lesson scripts and teaching materials accordingly. It constantly monitors the user's emotional state in real time using voice recognition and facial expression analysis technologies.

[0926] Specific examples

[0927] For example, consider a second-year junior high school math class where the topic is "quadratic equations."

[0928] 1. The user enters the necessary information about the "quadratic equation" into the terminal.

[0929] 2. The terminal sends the entered information to the server.

[0930] 3. The server searches the database for relevant resources and organizes them, including external resources.

[0931] 4. The generative AI model generates an incorrect answer (e.g., $x^2 + 4x + 4 = (x+2)^2 - 4$, an incorrect expansion).

[0932] 5. The emotion engine adjusts generated incorrect answers and lesson scripts based on the user's emotional data.

[0933] 6. The user checks the lesson script and teaching materials and makes any necessary corrections.

[0934] 7. The user conducts the lesson according to the generated lesson script, presenting incorrect information and encouraging students to solve the problem themselves.

[0935] 8. The device uses voice recognition technology and an emotion engine to record learners' responses and the user's emotions during class.

[0936] 9. The server analyzes this data and provides feedback during and after the lesson.

[0937] Prompt Sentence Examples

[0938] "In an eighth-grade math class, explain what a quadratic equation is and generate an incorrect answer. For example, include the expansion of $(x + 3)^2 = x^2 + 6x + 9$."

[0939] "If the user's emotions are anxiety, explain how to adjust the incorrect answer to the quadratic equation."

[0940] In this way, the present invention aims to adapt lesson content based on the user's emotional state and foster learners' independent problem-solving abilities.

[0941] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0942] Specific processing steps of the system program

[0943] Step 1:

[0944] The user logs into the system from a terminal and enters their authentication information.

[0945] Input: User login information (ID and password)

[0946] Operation:

[0947] The device sends the login information to the server.

[0948] The server receives the login information and authenticates it against a database.

[0949] Output: Authentication result (success or failure), and if successful, the user's personal settings data

[0950] If the authentication is successful, the server sends the user's personal setting information to the terminal, and the terminal displays the user's dashboard.

[0951] Step 2:

[0952] The user enters information such as the class subject, target grade, and topic into the terminal.

[0953] Input: Course content information (subject, target grade, topic, etc.)

[0954] Operation:

[0955] The terminal sends lesson content information to the server.

[0956] The server prepares the received data for analysis.

[0957] Output: Analysis results of lesson content data

[0958] The server checks your input to make sure you have all the necessary information.

[0959] Step 3:

[0960] The server analyzes the input lesson content and searches for related resources.

[0961] Input: Lesson content data

[0962] Operation:

[0963] The server analyzes the lesson content using natural language processing (NLP).

[0964] The server searches the database and the internet for relevant educational resources.

[0965] Output: List of related resources

[0966] The list of found resources is sent to the terminal and made available for the user to review.

[0967] Step 4:

[0968] The server calls the generative AI model and generates incorrect information based on the lesson content and analysis results.

[0969] Input: lesson content data, analysis results

[0970] Operation:

[0971] The generative AI model is called and begins the process of generating incorrect information (wrong answers) related to the lesson content.

[0972] Output: Generated incorrect answer data

[0973] As an example of an incorrect answer, "incorrect expansion of a quadratic equation" is generated.

[0974] Step 5:

[0975] An emotion engine adjusts generated incorrect answers based on the user's emotion data.

[0976] Input: wrong answer data, user emotion data

[0977] Operation:

[0978] The emotional engine analyzes the user's emotional state and adjusts the difficulty and type of incorrect answers.

[0979] Output: Adjusted error data

[0980] For example, if the user shows signs of anxiety, the difficulty level for incorrect answers is set low.

[0981] Step 6:

[0982] The server combines the generated incorrect information with the correct information to generate lesson scripts and teaching materials.

[0983] Input: Adjusted incorrect answer data, correct information data

[0984] Operation:

[0985] The server combines incorrect and correct answers to generate lesson scripts and teaching materials.

[0986] Output: Generated lesson scripts and teaching materials

[0987] The generated teaching materials and scripts are sent to the terminal, and the user confirms them on a confirmation screen.

[0988] Step 7:

[0989] The user starts the lesson through the terminal.

[0990] Input: Verified lesson scripts and materials

[0991] Operation:

[0992] Lessons are conducted based on lesson scripts created by the user, and incorrect information is presented to encourage learners to solve the problems themselves.

[0993] Output: Lesson data, student response data

[0994] The device supports the progress of the class in real time.

[0995] Step 8:

[0996] The device uses voice recognition technology to record students' comments and reactions during class and recognize the user's emotional state.

[0997] Input: Learner's speech, user's facial expression and tone of voice

[0998] Operation:

[0999] The device collects data in real time using voice and emotion recognition technology.

[1000] Output: Learner's speech data, user's emotion data

[1001] The collected data is sent to a server.

[1002] Step 9:

[1003] The server analyzes the recorded data and provides real-time feedback and suggestions for next learning steps to the user.

[1004] Input: Learner's speech data, user's emotion data

[1005] Operation:

[1006] The server analyzes the data and generates next steps and feedback.

[1007] Output: Real-time feedback, suggestions for next learning steps

[1008] Feedback is sent to the terminal and the user is notified.

[1009] Step 10:

[1010] After the lesson ends, the server generates a summary based on the data from the lesson, the learners' reactions, and the user's emotional data.

[1011] Input: lesson data, learner response data, user emotion data

[1012] Operation:

[1013] The server generates a summary of the lesson content to help plan the next lesson.

[1014] Output: Summary report

[1015] A summary is sent to the terminal for the user to review.

[1016] (Application example 2)

[1017] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1018] Conventional educational support systems have difficulty optimizing lesson content according to the user's emotional state, and providing real-time feedback based on the learner's reactions is insufficient. Furthermore, because they do not recommend individually optimized content based on the learner's emotional state, there are issues such as a decrease in learning efficiency and difficulty in maintaining motivation.

[1019] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for a user to input lesson content; means for analyzing the input lesson content and searching for related resources; means for calling a generative AI model that generates erroneous information based on the analysis results; means for adjusting the generated erroneous information; means for generating lesson scripts and teaching materials based on the generated information; means for providing the generated lesson scripts and teaching materials to the user; means for a user to present erroneous information during a lesson to encourage the learner to think; means for recording the learner's responses during the lesson and transmitting them to the server; means for analyzing the recorded responses and providing feedback in real time; means for generating a summary of the lesson content after the lesson ends; emotion recognition means for recognizing the user's emotional state and adapting the lesson content based on the state; and means including a generative AI model that recommends content based on the learner's emotional state in real time. This enables real-time feedback and individually optimized content recommendations based on the user's emotional state and the learner's responses.

[1020] The "means for users to input lesson content" refers to an interface that allows educators to input information such as lesson content, subjects, target grades, and topics.

[1021] "Means for analyzing the input lesson content and searching for related resources" refers to the function of analyzing the input lesson content and automatically searching for and collecting resources such as reference materials, literature, and problem sets related to that content.

[1022] "Means for invoking a generative AI model that generates erroneous information based on the analysis results" refers to the process of using the analysis results of the lesson content to launch an AI model that intentionally generates erroneous information and applying it to that model.

[1023] The "means for adjusting the generated erroneous information" is a function that appropriately adjusts the erroneous information generated by the generative AI model according to the user's emotional state and the learner's level of understanding.

[1024] The "means for generating lesson scripts and teaching materials based on the generated information" is a function that automatically generates scripts and teaching materials to be used in lessons based on the generated erroneous information and accurate information.

[1025] "Means for providing the generated lesson scripts and teaching materials to the user" is a function that provides the generated lesson scripts and teaching materials through an interface so that the educator can check and modify them.

[1026] "Means for users to present incorrect information during class to encourage students to think" refers to a process in which an educator intentionally presents incorrect information as part of a lesson and encourages students to think about and correct the error themselves.

[1027] "Means for recording learners' responses during lessons and sending them to a server" refers to a function that records learners' reactions, comments, facial expressions, and other responses, and sends them to a server for storage.

[1028] "Means for analyzing recorded responses and providing feedback in real time" refers to a function that analyzes recorded responses of learners and provides feedback to educators in real time based on the results.

[1029] "Means for generating a summary of the lesson content after the lesson has ended" is a function that generates a report summarizing the lesson content after the lesson has ended, based on data from the lesson and the students' responses.

[1030] "Emotion recognition means that recognizes the user's emotional state and adapts the lesson content based on that state" is a function that analyzes the educator's emotional state and optimizes the lesson content and progress based on the results.

[1031] "Means including a generative AI model that recommends content based on the learner's emotional state in real time" refers to a process that uses a generative AI model to grasp the learner's emotional state in real time and recommend content appropriate to that state.

[1032] This invention is an educational support system that recognizes the user's emotional state and optimizes lesson content accordingly. The system uses a generative AI model and emotion recognition means to provide real-time feedback and content recommendations based on the learner's reactions and emotions.

[1033] System configuration

[1034] The system includes the following major hardware and software components:

[1035] 1. User Interface (Terminal):

[1036] This is a device that allows educators to input lesson content and check and correct the generated scripts and teaching materials. This interface can be configured as a smartphone, tablet, or PC.

[1037] 2. Server:

[1038] This is the central computer that processes and stores data, calls generative AI models, and analyzes learner responses and user emotions. It runs on a cloud server.

[1039] 3. Generative AI Model:

[1040] This algorithm generates false information based on the content of lessons, providing learners with an opportunity to think independently. This model is implemented using AI frameworks such as Python and TensorFlow.

[1041] 4. Emotion Recognition Engine:

[1042] This algorithm recognizes the emotional state of users and learners from data acquired through cameras and microphones, and adjusts lesson content accordingly. It uses libraries such as OpenCV and dlib.

[1043] System Operation

[1044] User Authentication

[1045] The educator logs in to the system through the terminal and enters authentication information. The terminal sends the entered information to the server, which then authenticates the user. If authentication is successful, the server retrieves the user's personal settings and history and sends them to the terminal.

[1046] Input and generation of lesson content

[1047] The educator inputs the lesson content into the device. The input content is sent to the server, which analyzes the information and calls a generative AI model to generate incorrect answers. The generated incorrect answers are adjusted based on the user's emotional state using an emotion recognition engine. Lesson scripts and teaching materials are then automatically generated and provided to the educator.

[1048] Class implementation and response analysis

[1049] During the lesson, the educator follows the generated script and presents incorrect answers to encourage students to think. The device uses voice recognition technology to record what the student says and simultaneously recognizes the user's emotional state. This data is sent to the server and analyzed in real time. The server then provides real-time feedback to the educator based on this data.

[1050] Generate a summary after the lesson

[1051] After the lesson, the server generates a summary based on the data from the lesson, the learners' reactions, and the user's emotional data, and sends it to the device. The educator can then review the summary and incorporate it into their next lesson plan.

[1052] Specific examples

[1053] For example, consider a second-year junior high school math class with the topic of "quadratic equations." The educator selects "quadratic equations" as the topic and inputs the necessary information into the system. The server searches for related resources, and the generative AI model generates an incorrect answer (e.g., $x^2 + 4x + 4 = (x+2)^2 - 4$, an incorrect expansion). The emotion recognition engine adjusts the generated incorrect answer and lesson script based on the user's emotional data. The educator presents this incorrect answer during the lesson, encouraging the learner to derive the correct answer on their own. The device uses voice recognition technology and an emotion recognition engine to record the learner's reactions and the user's emotions during the lesson. After the lesson ends, the server analyzes this data and generates a summary to provide to the educator.

[1054] Prompt Sentence Examples

[1055] "If an emotion recognition engine detects that a user is feeling stressed, create a program that recommends relaxing music or movies to watch in that state."

[1056] As described above, the present invention is a system that can improve learning effectiveness and motivation by optimizing the method of providing lessons and content based on the emotional states of both the user and the learner.

[1057] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1058] Step 1:

[1059] The user inputs the lesson content (input content: lesson subject, target grade, topic, etc.)

[1060] The user inputs the lesson contents into the terminal interface and sends the input data to the server, which receives the data and prepares it for the next analysis step.

[1061] Step 2:

[1062] The server analyzes the input lesson content and searches for related resources (input content: lesson content data, output content: related resources)

[1063] The server analyzes the input lesson content using a text analysis algorithm, searches for related educational resources (reference materials, workbooks, etc.), and creates a list.

[1064] Step 3:

[1065] The server calls a generative AI model that generates incorrect information based on the analysis results (input: analysis results, output: incorrect information).

[1066] The server inputs the analysis results into a generative AI model, which is built using AI frameworks such as Python and TensorFlow, to generate false information related to the lesson content.

[1067] Step 4:

[1068] Adjusting the generated misinformation (input: misinformation, user's emotional state, output: adjusted misinformation)

[1069] The server uses an emotion recognition engine to analyze the user's emotional state and adjusts the generated erroneous information based on that state, adjusting the level of erroneous answers to an appropriate level according to the user's stress or excitement level.

[1070] Step 5:

[1071] Generate lesson scripts and teaching materials based on the generated information (input: adjusted incorrect information, correct information; output: lesson scripts, teaching materials)

[1072] The server automatically generates lesson scripts and teaching materials by combining the correct information with the adjusted incorrect information. These scripts and teaching materials are output in a format that users can review and edit.

[1073] Step 6:

[1074] The generated lesson scripts and teaching materials are provided to the user (input content: lesson script, teaching materials, output content: displayed on the user's device)

[1075] The server sends the generated lesson scripts and teaching materials to the user's terminal, where the user can check them and make corrections if necessary.

[1076] Step 7:

[1077] The user presents incorrect information during class to encourage students to think about it (input: class script, teaching materials, output: encourage students to think)

[1078] During class, the user presents incorrect information according to the generated script and prompts the learner to solve it, thereby developing the learner's independent problem-solving ability.

[1079] Step 8:

[1080] During the lesson, learners' responses are recorded and sent to the server (input: learners' responses, output: recorded data)

[1081] The device uses a microphone and camera to record the learner's responses and transmits the data to a server, using voice recognition and video analysis technology.

[1082] Step 9:

[1083] Analyze recorded responses and provide real-time feedback (input: recorded data, output: feedback)

[1084] The server analyzes the recorded student response data, generates real-time feedback, and sends it to the device, helping the user to progress through the lesson.

[1085] Step 10:

[1086] Generate a summary of the lesson content after the lesson ends (input content: data from the lesson, output content: summary)

[1087] The server generates a summary of the lesson content based on data from the lesson, learners' reactions, and user emotional data, and sends it to the user's device. The user can review this summary and reflect it in their next lesson plan.

[1088] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1089] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1090] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1091] [Third embodiment]

[1092] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1093] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[1094] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1095] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[1096] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1097] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1098] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1099] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1100] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1101] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1102] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1103] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1104] This invention relates to "Empower GPT," a system for supporting teachers in the educational field. This system uses generative AI to provide false information, aiming to develop learners' ability to solve problems independently.

[1105] System Overview

[1106] The system includes the following main components:

[1107] 1. User interface (terminal): A device that allows teachers to input lesson content and check and modify the generated scripts and teaching materials.

[1108] 2. Server: A central computer that processes and stores data, calls generative AI models, and analyzes learner responses.

[1109] 3. Generative AI model: An artificial intelligence algorithm that generates misinformation based on lesson content, providing learners with an opportunity to think for themselves.

[1110] Program processing procedure

[1111] 1. User login and authentication:

[1112] The user logs into the system from a terminal and enters the necessary authentication information.

[1113] The server verifies the authentication information and sets up the appropriate environment for each user.

[1114] 2. Input lesson content:

[1115] The user inputs information such as the class subject, target grade, topic, etc. from the terminal.

[1116] The terminal transmits this information to the server.

[1117] 3. Analyzing information and finding resources:

[1118] The server analyzes the input lesson content and searches for related resources (teaching materials, past lesson content, etc.).

[1119] 4. Creation of misinformation:

[1120] The server calls the generative AI model and generates false information based on the analysis results.

[1121] The generated misinformation is tailored to the learner's level of understanding and grade level.

[1122] 5. Generating lesson scripts and teaching materials:

[1123] The server generates lesson scripts and teaching materials based on the generated incorrect information and correct information.

[1124] The generated scripts and teaching materials are sent to the terminal and checked by the user.

[1125] 6. Class Conduct:

[1126] A user presents false information during a lesson to initiate a process of thinking in the learners.

[1127] Generative AI models respond to users' questions and concerns in real time.

[1128] 7. Recording and analyzing learner responses:

[1129] The device uses voice recognition technology to record the learner's responses.

[1130] The recorded data is sent to a server, which analyzes it.

[1131] 8. Providing Feedback:

[1132] The server provides real-time feedback based on the analysis results and suggests next steps to the learner.

[1133] 9. Post-class summary generation:

[1134] After the lesson ends, the server generates a summary based on the lesson content and the learners' responses.

[1135] The summary is sent to the terminal, where the user can review it and reflect it in the next lesson plan.

[1136] Specific examples

[1137] For example, consider a case where the topic of a second-year junior high school math class is "quadratic equations." When a user selects "quadratic equations" as the topic, the server searches for related resources, and the generative AI model generates an incorrect answer (e.g., $x^2 + 4x + 4 = (x+2)^2 - 4$, an incorrect expansion). The user presents this incorrect answer during class, encouraging students to derive the correct answer on their own. After the class, the server generates a summary based on the content of the class and the students' responses, which can be used for the next class.

[1138] In this way, the present invention provides a system that promotes independent learning by learners in educational settings and reduces the burden on teachers.

[1139] The processing flow will be explained below.

[1140] Step 1: User Login and Authentication

[1141] The user enters the information (user name, password) required to log in to the system from the terminal.

[1142] The terminal sends the entered login information to the server.

[1143] The server checks the received login information against a database and performs authentication.

[1144] If authentication is successful, the server obtains the user's personal settings and history and sends them to the terminal.

[1145] The user confirms successful authentication through the terminal and is then able to access the system.

[1146] Step 2: Enter lesson content

[1147] The user enters information such as the class subject, target grade, and topic from the terminal.

[1148] The terminal sends the input lesson content to the server.

[1149] Step 3: Analyze the information and find resources

[1150] The server analyzes the input lesson content.

[1151] Based on the analysis results, the server searches for relevant resources (teaching materials, past lesson contents, reference materials, etc.) and retrieves them from the database.

[1152] Step 4: Creating misinformation

[1153] The server calls the generative AI model and generates incorrect information based on the lesson content and analysis results.

[1154] The generated erroneous information is appropriately adjusted according to the learner's level of understanding and the target grade level.

[1155] Step 5: Generate lesson scripts and teaching materials

[1156] The server combines the generated incorrect information with the correct information to generate lesson scripts and teaching materials.

[1157] The generated lesson scripts and teaching materials are sent to the terminal, where the user can check and modify them.

[1158] Step 6: Conducting lessons

[1159] The user starts the lesson through the terminal.

[1160] The user follows a generated lesson script to intentionally present incorrect information and encourage learners to think.

[1161] During lessons, the generative AI model responds to and supports users' questions and concerns in real time.

[1162] Step 7: Record and analyze learner responses

[1163] The device uses voice recognition technology to record students' comments and reactions during class.

[1164] The recorded data is sent to a server, which analyzes it.

[1165] Based on the analysis results, the server provides real-time feedback and suggests next learning steps to the user.

[1166] Step 8: Generate a summary after the lesson

[1167] After the lesson ends, the server generates a summary based on the data from the lesson and the learners' responses.

[1168] The generated summary is sent to the terminal, where the user can check the lesson content and the learners' level of understanding and reflect it in future lesson plans.

[1169] Example 1

[1170] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1171] In today's educational settings, there are no fully established methods for teachers to provide appropriate teaching materials and misinformation according to each student's level of understanding, and for students to develop the ability to solve problems independently based on that information. Furthermore, it is difficult to analyze students' responses in real time and provide appropriate feedback. This increases the burden on teachers and hinders the promotion of independent learning among students, creating the problem of a lack of understanding.

[1172] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1173] In this invention, the server includes: means for a user to input lesson content; means for analyzing the input lesson content and searching for related resources; means for calling a generative AI model that generates erroneous information based on the analysis results; means for adjusting the generated erroneous information; means for generating lesson scripts and teaching materials based on the generated information; means for providing the generated lesson scripts and teaching materials to the user; means for a user to present erroneous information during class to encourage learners to think; means for recording learners' responses during class and transmitting them to the server; means for analyzing the recorded responses and providing feedback in real time; means for generating a summary of the lesson content after the class ends; means for adjusting the difficulty level of the generated erroneous information according to the learners' level of understanding; and means including a generative AI model that responds to user questions and doubts in real time during class. This reduces the burden on teachers and promotes learners' independent learning.

[1174] "Users" refers to teachers and educators who use this system in educational settings.

[1175] "Lesson content" refers to information such as subject, target grade, and topic that teachers enter into the system.

[1176] "Analysis" refers to the process of breaking down input lesson content using natural language processing techniques or other methods to identify relevant resources.

[1177] "Resources" refers to teaching materials, reference materials, past lesson data, etc. related to the content of the lesson.

[1178] A "generative AI model" refers to the part of a system that uses artificial intelligence algorithms to generate false information based on specific inputs.

[1179] A "prompt" refers to input text used to instruct a generative AI model on what information to generate.

[1180] "Misinformation" refers to incorrect data or answers that are intentionally generated as a starting point for learners to use their own understanding to derive accurate information.

[1181] A "script" refers to a script or instruction manual that helps guide a lesson.

[1182] "Instructional materials" refers to visual, textual, or hands-on learning materials used to supplement course content.

[1183] "Terminal" refers to a device such as a computer or tablet that teachers use to access and operate the system.

[1184] "Server" refers to the central computing system that processes and stores data, invokes generative AI models, analyzes learner responses, etc.

[1185] "Speech recognition technology" refers to technology that converts learners' speech into text.

[1186] "Feedback" refers to instructions, hints, advice, etc. provided based on the learner's responses.

[1187] A "summary" refers to a short report generated based on the lesson content and learner responses.

[1188] "Responses" refer to the level of understanding, questions, comments, etc. that learners show during class.

[1189] This invention relates to a system called "Empower GPT" that supports teachers in the educational field. This system uses generative AI to provide false information and aims to cultivate learners' ability to solve problems independently. The following describes in detail the mode for carrying out the invention.

[1190] System configuration

[1191] The system includes the following main components:

[1192] 1. User interface (terminal): A device on which teachers input lesson content and check and edit generated scripts and teaching materials. Examples include PCs, tablets, and smartphones.

[1193] 2. Server: This is the central computer that processes and stores data, calls the generative AI model, and analyzes learner responses. In this invention, the server is assumed to be a high-performance computer with a Linux-based OS. MySQL, PostgreSQL, or other databases are used.

[1194] 3. Generative AI model: An artificial intelligence algorithm that generates false information based on the lesson content input by the teacher, providing learners with an opportunity to think for themselves. Specific examples include OpenAI's GPT-3 and GPT-4.

[1195] Program processing procedure

[1196] The system begins when the user (teacher) logs in and inputs the lesson content. The specific processing steps are explained below.

[1197] 1. User login and authentication:

[1198] A user logs in to the system from a terminal and enters the necessary authentication information, such as a user ID and password.

[1199] The server verifies the authentication information and sets up the appropriate environment for each individual user, a process that involves the server verifying user information against a database.

[1200] 2. Input lesson content:

[1201] The user inputs information such as the class subject, target grade, topic, etc., into the terminal. For example, they input "mathematics," "second-year junior high school student," and "quadratic equations."

[1202] The device sends this information to the server, and the data is sent in JSON format.

[1203] 3. Analyzing information and finding resources:

[1204] The server analyzes the input lesson content and searches the database for related resources (teaching materials, past lesson content, etc.) using natural language processing libraries (e.g., NLTK and spaCy).

[1205] 4. Creation of misinformation:

[1206] The server calls the generative AI model and generates false information based on the analysis results. The generative AI model is instructed using prompts.

[1207] The generated misinformation is tailored to the learner's level of understanding and grade level.

[1208] 5. Generating lesson scripts and teaching materials:

[1209] The server generates lesson scripts and teaching materials based on the generated incorrect information and correct information. For example, it uses Python or JavaScript scripts to build teaching materials.

[1210] The generated scripts and teaching materials are sent to the terminal and checked by the user.

[1211] 6. Class Conduct:

[1212] The user presents incorrect information during a lesson to initiate the thinking process in the learner, for example, by using an incorrect formula such as "x^2 + 4x + 4 = (x+2)^2 - 4".

[1213] The generative AI model responds to user questions and concerns in real time, including when users type questions into the generative AI, which then responds instantly.

[1214] 7. Recording and analyzing learner responses:

[1215] The device records the learner's responses using voice recognition technology, such as Google Speech-to-Text or Amazon Transcribe.

[1216] The recorded data is sent to a server, which analyzes it using machine learning algorithms (e.g., scikit-learn or TensorFlow).

[1217] 8. Providing Feedback:

[1218] The server provides real-time feedback based on the analysis results and suggests next steps to the learner, such as "Many students got this question wrong, so let's explain it again."

[1219] 9. Post-class summary generation:

[1220] After the lesson, the server generates a summary based on the lesson content and the learners' responses. The summary is created using natural language generation technology.

[1221] The summary is sent to the terminal, where the user can review it and reflect it in the next lesson plan.

[1222] Specific examples

[1223] For example, if the topic of a second-year junior high school math class is "quadratic equations," when the user selects "quadratic equations" as the topic, the server searches for related resources and the generative AI model generates an incorrect answer (e.g., $x^2 + 4x + 4 = (x+2)^2 - 4$, an incorrect expansion). The user presents this incorrect answer during class, encouraging the learners to derive the correct answer on their own. After the class, the server generates a summary based on the content of the class and the learners' responses, which can be used for the next class.

[1224] Examples of prompt statements

[1225] Topic: 8th Grade Math - Quadratic Equations

[1226] Objectives: Students will understand how to solve quadratic equations and be able to correct incorrect information themselves.

[1227] Examples of misinformation it generates:

[1228] $x^2 ​​+ 4x + 4 = (x+2)^2 - 4$

[1229] "The only solution to a quadratic equation is x = 2"

[1230] In this way, the present invention provides a system that promotes independent learning by learners in educational settings and reduces the burden on teachers.

[1231] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1232] Step 1:

[1233] A user accesses the system from a terminal and enters login information (user ID and password). The input data is encrypted on the terminal as authentication information and sent to the server. The server receives the authentication information and checks it against a database. Based on the check result, if authentication is successful, the environment settings corresponding to the user are loaded and a dashboard screen is displayed on the terminal. The output is a dashboard screen that the user can access.

[1234] Step 2:

[1235] The user inputs the class subject, target grade, and specific topic into the device. The input data includes information such as "Mathematics," "Second Year of Junior High School," and "Quadratic Equations." The device sends the input data in JSON format to the server. The server temporarily stores the received data and prepares for the analysis step. The output is a notification that the data has been saved on the server.

[1236] Step 3:

[1237] The server analyzes the received lesson content data. A natural language processing library (e.g., NLTK or spaCy) is used for the analysis. The input for the analysis is JSON data about the lesson content. As a result of the analysis, search keywords and tags for related resources (teaching materials, past lesson content, etc.) are generated. The output is the search keywords and tags.

[1238] Step 4:

[1239] The server searches for related resources from databases and cloud storage based on the analysis results. The input in this step is search keywords and tags. Related resources (PDF teaching materials, video links, past lesson scripts, etc.) are obtained as search results. The output is a list of the obtained resources.

[1240] Step 5:

[1241] The server invokes a generative AI model to generate false information using a prompt. The input in this step is the analysis result of the lesson content and the prompt. The generative AI model (e.g., GPT-3 or GPT-4) generates false information based on the input prompt. The output is the generated false information.

[1242] Step 6:

[1243] The server adjusts the generated misinformation according to the learner's level of understanding and grade level. The input is the generated misinformation and the learner's profile data. The adjusted information is generated and output. The adjustment may use machine learning algorithms.

[1244] Step 7:

[1245] The server generates lesson scripts and teaching materials based on the adjusted misinformation and correct information. The input data is the adjusted misinformation and related correct information. The lesson scripts and teaching materials are generated using programming languages ​​such as Python and JavaScript. The output is the completed lesson scripts and teaching materials.

[1246] Step 8:

[1247] The server sends the generated lesson script and teaching materials to the terminal, where the user can review them. The output is a script and teaching materials that the user can review on the terminal. The user can then modify the content as needed.

[1248] Step 9:

[1249] The user presents false information generated during class to prompt students to think. The input is the adjusted false information. As a concrete example, a teacher writes "$x^2 + 4x + 4 = (x+2)^2 - 4$" on the blackboard and asks students to think about what is wrong. The output is the student's response.

[1250] Step 10:

[1251] The device uses voice recognition technology (e.g., Google Speech-to-Text or Amazon Transcribe) to record the learner's responses and send them to the server. The input is the learner's speech. The voice data is converted into text format, and the output is the text data and its analysis results.

[1252] Step 11:

[1253] The server analyzes the recorded responses of the learner and provides feedback in real time. The input is the learner's response data in text format. Based on the analysis results, appropriate feedback is generated, and the output is the content of the feedback.

[1254] Step 12:

[1255] After the lesson ends, the server generates a summary based on the lesson content and learner responses. The input is the lesson script, teaching materials, and learner response data. The summary is created using natural language generation technology. The output is a summary that is sent to the terminal and reviewed by the user.

[1256] (Application example 1)

[1257] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1258] In conventional educational systems, when a single teacher instructs a large number of students, it is difficult to grasp each student's level of understanding and reactions in real time and provide appropriate feedback. Furthermore, in order for students to develop their independent problem-solving skills, they need teaching materials that contain not only accurate information but also incorrect information, but generating and adjusting such materials is time-consuming. This increases the burden on teachers and makes it difficult to improve students' independent learning abilities.

[1259] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1260] In this invention, the server includes a means for users to input lesson content, a means for analyzing the input lesson content and searching for related resources, and a means for calling a generative AI model that generates erroneous information based on the analysis results. This makes it possible to generate and adjust erroneous information and answers and provide appropriate feedback in real time. Furthermore, the server includes a means for users to select the content they want to learn and provide teaching materials containing erroneous information, a means for providing real-time feedback to students based on the generated erroneous information and answers, and a means for recording and saving learned content and answers so that they can be reviewed later. This facilitates the improvement of learners' independent learning abilities while reducing the burden on teachers.

[1261] "Means for users to input lesson content" refers to interfaces and applications that allow teachers and students to input the subjects or topics they want to learn into digital devices.

[1262] "Means for analyzing input lesson content and searching for related resources" refers to an algorithm that searches and retrieves related information and materials from databases or networks based on the input teaching materials or topics.

[1263] "Means for invoking a generative AI model that generates erroneous information based on the analysis results" refers to a mechanism that uses the analysis results of the input content to execute a process that intentionally creates erroneous information using a generative AI model.

[1264] The "means for adjusting generated erroneous information" is a function for appropriately adjusting the difficulty and content of the generated erroneous information according to the learner's level of understanding and comprehension.

[1265] The "means for generating lesson scripts and teaching materials based on the generated information" is a system that combines incorrect information with correct information to create and provide scripts and educational materials to be used in actual lessons.

[1266] The "means for providing the generated lesson scripts and teaching materials to users" refers to a mechanism for distributing the generated lesson scripts and teaching materials to teachers and learners via digital devices.

[1267] "Means for users to present incorrect information during class and encourage students to think" is a process in which a teacher intentionally presents incorrect information during class and has students point out the errors.

[1268] "Means for recording learners' responses during class and sending them to a server" refers to a system that has the function of recording how learners respond during class and sending and saving them to a server.

[1269] The "means for analyzing recorded responses and providing real-time feedback" refers to an algorithm that analyzes recorded responses of learners and provides appropriate feedback immediately.

[1270] The "means for generating a summary of the content of a lesson after the lesson has ended" is a system that summarizes the content of the lesson after the lesson has ended, generates a summary, and provides it.

[1271] "Means for providing students with real-time feedback based on incorrect information or answers" refers to a system that provides supplementary information or advice in real time to learners as they solve problems, based on the incorrect information or answers they generate.

[1272] "A means to record and save what has been learned and answers, and to review them later" is a system that has the function of saving what learners have learned and answers in a database, allowing them to check and review them later.

[1273] This invention relates to an educational support system that utilizes a generative AI model. The purpose of this invention is to promote learners' independent learning and reduce the burden on teachers by cooperating with three main components: a server, a terminal, and a user.

[1274] System configuration

[1275] 1. Server:

[1276] The server has the following functions:

[1277] Provides a means for users to input lesson content.

[1278] It has the function of analyzing the input lesson content and searching for related resources.

[1279] Based on the analysis results, a generative AI model is invoked to generate false information.

[1280] The generated incorrect information is adjusted and lesson scripts and teaching materials are generated.

[1281] Analyze recorded learner responses and provide feedback in real time.

[1282] 2. Terminal:

[1283] Terminals are devices used by teachers and students and have the following functions:

[1284] It provides a user interface for inputting, checking, and modifying lesson content.

[1285] Display the generated lesson scripts and teaching materials.

[1286] Learners' responses are recorded during the lesson and sent to the server.

[1287] 3. User:

[1288] Users are mainly teachers and students, and use the system in the following steps:

[1289] Enter the lesson content from the terminal.

[1290] Check and modify the generated scripts and teaching materials.

[1291] Present misinformation during class and encourage students to problem-solve.

[1292] Record learners' responses and review their feedback after the lesson.

[1293] Software and Hardware Configuration

[1294] On the server side, it uses high-performance cloud servers such as Amazon Web Services and Google Cloud Platform, and the backend is built using Node.js and Express, with MongoDB as the database.

[1295] The front-end part will be developed using React Native, taking into consideration use on smartphones.

[1296] The generative AI model uses OpenAI GPT-4, which generates prompts based on the input lesson content and then creates false information as a response to those prompts.

[1297] As a use case, consider a 8th grade math lesson on "quadratic equations," where the following prompt is sent to the generative AI model:

[1298] "Generate false information and answers on the following topic: Quadratic Equations"

[1299] The generative AI model generates incorrect information for this prompt (e.g., an incorrect expansion such as $x^2 + 4x + 4 = (x+2)^2 - 4$), which is then provided as part of the learning material.

[1300] The server analyzes the learner's responses and provides real-time feedback, for example, helping the learner recognize incorrect information and guide them in the process of arriving at the correct answer.

[1301] This system will enable students to develop the ability to correct their own mistakes and reduce the burden on teachers. It is also expected that the inclusion of incorrect information will dramatically improve students' thinking and problem-solving abilities.

[1302] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1303] Step 1:

[1304] The user inputs the lesson content.

[1305] In this step, the user enters information such as the class subject, target grade, topic, etc. into the application via their device. The entered information is sent from the front end (React Native) to the server (Node.js).

[1306] Input: Course subject, target grade, topic

[1307] Output: The input data is sent to the server

[1308] Step 2:

[1309] The server analyzes the content of the lesson and searches for related resources.

[1310] The server analyzes the received lesson content and searches a database (MongoDB) for teaching materials and past resources related to that topic.

[1311] Input: Lesson content sent in Step 1

[1312] Output: A list of resources related to the analysis results

[1313] Step 3:

[1314] The server calls a generative AI model based on the analysis results and generates incorrect information.

[1315] Based on the analysis results, the server sends a prompt to the generative AI model (OpenAI GPT-4) to generate incorrect information.

[1316] Input: Analysis results, prompt (e.g. "Generate incorrect information about quadratic equations")

[1317] Output: The incorrect information generated

[1318] Step 4:

[1319] The server will adjust the generated incorrect information.

[1320] The server runs an algorithm that adjusts the generated misinformation according to the learner's level of comprehension and grade level, optimizing the difficulty of the misinformation.

[1321] Input: Generated misinformation, learner comprehension information

[1322] Output: Adjusted misinformation

[1323] Step 5:

[1324] The server generates lesson scripts and teaching materials based on the generated information.

[1325] The server generates lesson scripts and teaching materials based on the adjusted incorrect information and correct information and provides them to users.

[1326] Input: Adjusted incorrect information, correct information

[1327] Output: Lesson scripts, teaching materials

[1328] Step 6:

[1329] The user presents incorrect information during class to prompt students to think.

[1330] The user uses the generated teaching materials in class, presents incorrect information to students, and has them point out the errors, thereby encouraging the students to think independently.

[1331] Input: lesson scripts, teaching materials

[1332] Output: Learner response

[1333] Step 7:

[1334] The device records the learners' responses during class and sends them to the server.

[1335] Learners' responses and reactions are recorded by the device and sent to the server in real time. Speech recognition technology is also used to record utterances.

[1336] Input: Learner response data

[1337] Output: Response data sent to the server

[1338] Step 8:

[1339] The server analyzes the recorded responses and provides real-time feedback.

[1340] The server analyzes the received learner response data and uses an AI model to generate and provide instant feedback.

[1341] Input: Reaction data

[1342] Output: Real-time feedback

[1343] Step 9:

[1344] After the lesson ends, the server generates a summary of the lesson content.

[1345] After the lesson is over, the server creates a summary based on the lesson content and the learners' reactions and provides it to the user to help plan the next lesson.

[1346] Input: Lesson content, learner response data

[1347] Output: Lesson summary

[1348] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1349] This invention relates to "Empower GPT," an educational support system with the ability to recognize users' emotions and adapt lesson scripts and teaching materials. The system aims to optimize lessons based on the user's emotional state while providing false information using generative AI to foster learners' ability to solve problems independently.

[1350] System Overview

[1351] The system includes the following main components:

[1352] 1. User interface (terminal): A device that allows teachers to input lesson content and check and modify the generated scripts and teaching materials.

[1353] 2. Server: The central computer that processes and stores data, invokes generative AI models, and analyzes learner responses and user sentiment.

[1354] 3. Generative AI model: An artificial intelligence algorithm that generates misinformation based on lesson content, providing learners with an opportunity to think independently.

[1355] 4. Emotion Engine: An algorithm to recognize the user's emotional state and adjust the lesson script and teaching materials.

[1356] Program processing procedure

[1357] 1. User login and authentication:

[1358] The user logs into the system from a terminal and enters their authentication information.

[1359] The terminal sends the entered login information to the server, and the server performs authentication.

[1360] If authentication is successful, the server obtains the user's personal settings and history and sends them to the device.

[1361] 2. Input lesson content:

[1362] The user enters information such as the class subject, target grade, and topic into the terminal.

[1363] The terminal transmits this information to the server.

[1364] 3. Analyzing information and finding resources:

[1365] The server analyzes the input lesson content and searches for and organizes related resources.

[1366] 4. Creation of misinformation:

[1367] The server calls the generative AI model and generates incorrect information based on the lesson content and analysis results.

[1368] The generative AI model adjusts the generated incorrect answers based on the user's emotional data.

[1369] 5. Generating lesson scripts and teaching materials:

[1370] The server combines the generated incorrect information with the correct information to generate lesson scripts and teaching materials.

[1371] The generated lesson scripts and teaching materials are sent to the terminal, where the user can check and modify them.

[1372] 6. Class Conduct:

[1373] The user starts the lesson through the terminal.

[1374] The user presents incorrect information according to the generated lesson script, prompting learners to think.

[1375] Generative AI models respond to users' questions and concerns in real time.

[1376] 7. Recording and analyzing learner responses and user emotions:

[1377] The device uses voice recognition technology to record students' comments and reactions during class.

[1378] The terminal recognizes the user's emotional state and transmits the emotional data to the server.

[1379] The recorded data and emotional data are sent to a server for analysis.

[1380] Based on the analysis results, the server provides real-time feedback and suggests next learning steps to the user.

[1381] 8. Post-class summary generation:

[1382] After the lesson ends, the server generates a summary based on the data from the lesson, the learners' reactions, and the user's emotional data.

[1383] A summary is sent to the terminal, and the user checks the lesson content and the learners' level of understanding, and reflects this in the next lesson plan.

[1384] Specific examples

[1385] For example, consider a case where the topic of a math class for second-year junior high school students is "quadratic equations." The user selects "quadratic equations" as the topic and enters the necessary information into the system. The server searches for related resources, and the generative AI model generates an incorrect answer (e.g., $x^2 + 4x + 4 = (x+2)^2 - 4$, an incorrect expansion). The emotion engine adjusts the generated incorrect answers and lesson scripts based on the user's emotional data.

[1386] The user presents the incorrect answer during the lesson, encouraging the learner to find the correct answer on their own. The device uses voice recognition technology and an emotion recognition engine to record the learner's responses and the user's emotions during the lesson. After the lesson ends, the server analyzes this data, generates a summary, and provides it to the user. In this way, it is possible to promote learners' independent learning and conduct lessons that are tailored to the user's emotional state.

[1387] The processing flow will be explained below.

[1388] Step 1: User Login and Authentication

[1389] The user enters the information (user name, password) required to log in to the system from the terminal.

[1390] The terminal sends the entered login information to the server.

[1391] The server checks the received login information against a database and performs authentication.

[1392] If authentication is successful, the server obtains the user's personal settings and history and sends them to the terminal.

[1393] The user can then verify their authentication through the terminal and gain access to the system.

[1394] Step 2: Enter lesson content

[1395] The user enters information such as the class subject, target grade, and topic from the terminal.

[1396] The terminal sends the input lesson content to the server.

[1397] Step 3: Analyze the information and find resources

[1398] The server analyzes the input lesson content.

[1399] The server searches for and retrieves relevant resources (teaching materials, past lesson contents, reference materials, etc.) based on the analysis results.

[1400] Step 4: Creating misinformation

[1401] The server calls the generative AI model and generates incorrect information based on the lesson content and analysis results.

[1402] The server adjusts the difficulty and content of the generated erroneous information according to the learner's level of understanding.

[1403] Step 5: Emotional Engine Alignment

[1404] The device monitors the user's emotional state in real time and transmits the emotional data to a server.

[1405] The server uses an emotion engine to analyze the emotional data and adjust the generated incorrect answers and lesson scripts.

[1406] Step 6: Generate lesson scripts and teaching materials

[1407] The server combines the generated incorrect information with the correct information to generate lesson scripts and teaching materials.

[1408] The generated lesson scripts and teaching materials are sent to the terminal, where the user can check and modify them.

[1409] Step 7: Conducting lessons

[1410] The user starts the lesson through the terminal.

[1411] The user follows a generated lesson script to intentionally present incorrect information and encourage learners to think.

[1412] During lessons, the generative AI model responds to users' questions and concerns in real time.

[1413] Step 8: Recording and analyzing learner responses and user emotions

[1414] The device uses voice recognition technology to record students' comments and reactions during class.

[1415] The terminal recognizes the user's emotional state and transmits the emotional data to the server.

[1416] The server analyzes the recorded learner responses and user emotion data.

[1417] Based on the analysis results, the server provides real-time feedback and suggests next learning steps to the user.

[1418] Step 9: Generate a summary after the lesson

[1419] After the lesson ends, the server generates a summary based on the data from the lesson, the learners' reactions, and the user's emotional data.

[1420] The generated summary is sent to the terminal, where the user can check the lesson content and the learners' level of understanding and reflect it in the next lesson plan.

[1421] Example 2

[1422] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1423] Traditional educational systems lacked effective methods for cultivating students' independent problem-solving skills. They also lacked mechanisms for adapting lesson content based on the user's emotional state, making it difficult to maximize educational effectiveness. As a result, they were unable to respond to fluctuations in students' interests and comprehension levels, limiting their ability to provide consistent educational content.

[1424] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1425] In this invention, the server includes means for a user to input lesson content, means for analyzing the input lesson content and searching for related resources, means for calling an AI algorithm that generates erroneous information based on the analysis results, means for adjusting the generated erroneous information, means for generating lesson scripts and teaching materials based on the generated information, means for providing the generated lesson scripts and teaching materials to the user, means for the user to present erroneous information during the lesson to encourage students to think, means for recording students' responses during the lesson and transmitting them to a central computer, means for analyzing the recorded responses and providing feedback in real time, means for generating a summary of the lesson content after the lesson ends, and means for recognizing the user's emotional state and adjusting the lesson scripts and teaching materials, thereby cultivating students' independent problem-solving skills and enabling the lesson content to be adapted based on the user's emotional state.

[1426] "Users" are educators and administrators who operate the system and input lesson content.

[1427] "Class content" refers to educational information such as the subjects, topics, and target grades covered in the class.

[1428] "Input means" refers to an interface or device that allows users to input lesson content into the system.

[1429] "Analysis means" refers to algorithms or devices that analyze the input lesson content and search for related resources.

[1430] "Misinformation" refers to intentionally inaccurate educational information provided to learners in order to solve problems independently.

[1431] "Artificial intelligence algorithms" refer to machine learning and natural language processing technologies that generate false information based on input information.

[1432] "Adjustment means" refers to processes or functions for adjusting erroneous information generated by artificial intelligence algorithms.

[1433] A "lesson script" is a script or plan used in a lesson, and is a document or digital data that contains the content as the lesson progresses.

[1434] "Teaching materials" refers to the teaching materials and learning resources used in classes (e.g., textbooks, workbooks, video materials, etc.).

[1435] "Recording means" refers to voice recognition technology or other digital recording devices used to record learner responses during lessons.

[1436] The "central computer" is the main device in the system that processes and stores data, calls generative AI models, and analyzes learner responses and user emotions.

[1437] "Analysis tools" are algorithms or devices that analyze recorded learner responses and provide real-time feedback.

[1438] An "emotional state" is a state that recognizes and classifies a user's emotions (e.g., joy, sadness, anger, surprise, etc.).

[1439] A "summary generation means" refers to an algorithm or device that creates a summary after a lesson based on the lesson content and learners' responses.

[1440] "Real-time feedback" refers to guidance and advice provided immediately or briefly during or after a lesson.

[1441] MODE FOR CARRYING OUT THE INVENTION

[1442] In the present invention, the following hardware and software are used to effectively implement the education support system.

[1443] (System Configuration)

[1444] 1. User Interface (Terminal)

[1445] 2. Server

[1446] 3. Generative AI Models

[1447] 4. Emotion Engine

[1448] (User Interface (Terminal))

[1449] A user interface is a device that allows users to input lesson content and review and edit generated scripts and teaching materials. This can be a computer, tablet, smartphone, or other digital device. It has a solid graphical user interface (GUI) and is designed to allow users to operate it intuitively.

[1450] (server)

[1451] The server is the central computer that processes and stores data, calls the generative AI model, and analyzes learner responses and user emotions. It uses a cloud server with powerful computing capabilities, and is able to exchange data with devices in real time via the communications infrastructure. The server integrates a database, analysis engine, and various algorithms.

[1452] (Generative AI model)

[1453] The generative AI model is an artificial intelligence algorithm that generates false information based on input lesson content, providing learners with an opportunity to think independently. It incorporates natural language processing (NLP) and machine learning (ML) technologies and has the ability to adaptively adjust the information it generates based on the user's emotional data.

[1454] (Emotion Engine)

[1455] The emotion engine is an algorithm that recognizes the user's emotional state and adjusts lesson scripts and teaching materials accordingly. It constantly monitors the user's emotional state in real time using voice recognition and facial expression analysis technologies.

[1456] Specific examples

[1457] For example, consider a second-year junior high school math class where the topic is "quadratic equations."

[1458] 1. The user enters the necessary information about the "quadratic equation" into the terminal.

[1459] 2. The terminal sends the entered information to the server.

[1460] 3. The server searches the database for relevant resources and organizes them, including external resources.

[1461] 4. The generative AI model generates an incorrect answer (e.g., $x^2 + 4x + 4 = (x+2)^2 - 4$, an incorrect expansion).

[1462] 5. The emotion engine adjusts generated incorrect answers and lesson scripts based on the user's emotional data.

[1463] 6. The user checks the lesson script and teaching materials and makes any necessary corrections.

[1464] 7. The user conducts the lesson according to the generated lesson script, presenting incorrect information and encouraging students to solve the problem themselves.

[1465] 8. The device uses voice recognition technology and an emotion engine to record learners' responses and the user's emotions during class.

[1466] 9. The server analyzes this data and provides feedback during and after the lesson.

[1467] Prompt Sentence Examples

[1468] "In an eighth-grade math class, explain what a quadratic equation is and generate an incorrect answer. For example, include the expansion of $(x + 3)^2 = x^2 + 6x + 9$."

[1469] "If the user's emotions are anxiety, explain how to adjust the incorrect answer to the quadratic equation."

[1470] In this way, the present invention aims to adapt lesson content based on the user's emotional state and foster learners' independent problem-solving abilities.

[1471] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1472] Specific processing steps of the system program

[1473] Step 1:

[1474] The user logs into the system from a terminal and enters their authentication information.

[1475] Input: User login information (ID and password)

[1476] Operation:

[1477] The device sends the login information to the server.

[1478] The server receives the login information and authenticates it against a database.

[1479] Output: Authentication result (success or failure), and if successful, the user's personal settings data

[1480] If the authentication is successful, the server sends the user's personal setting information to the terminal, and the terminal displays the user's dashboard.

[1481] Step 2:

[1482] The user enters information such as the class subject, target grade, and topic into the terminal.

[1483] Input: Course content information (subject, target grade, topic, etc.)

[1484] Operation:

[1485] The terminal sends lesson content information to the server.

[1486] The server prepares the received data for analysis.

[1487] Output: Analysis results of lesson content data

[1488] The server checks your input to make sure you have all the necessary information.

[1489] Step 3:

[1490] The server analyzes the input lesson content and searches for related resources.

[1491] Input: Lesson content data

[1492] Operation:

[1493] The server analyzes the lesson content using natural language processing (NLP).

[1494] The server searches the database and the internet for relevant educational resources.

[1495] Output: List of related resources

[1496] The list of found resources is sent to the terminal and made available for the user to review.

[1497] Step 4:

[1498] The server calls the generative AI model and generates incorrect information based on the lesson content and analysis results.

[1499] Input: lesson content data, analysis results

[1500] Operation:

[1501] The generative AI model is called and begins the process of generating incorrect information (wrong answers) related to the lesson content.

[1502] Output: Generated incorrect answer data

[1503] As an example of an incorrect answer, "incorrect expansion of a quadratic equation" is generated.

[1504] Step 5:

[1505] An emotion engine adjusts generated incorrect answers based on the user's emotion data.

[1506] Input: wrong answer data, user emotion data

[1507] Operation:

[1508] The emotional engine analyzes the user's emotional state and adjusts the difficulty and type of incorrect answers.

[1509] Output: Adjusted error data

[1510] For example, if the user shows signs of anxiety, the difficulty level for incorrect answers is set low.

[1511] Step 6:

[1512] The server combines the generated incorrect information with the correct information to generate lesson scripts and teaching materials.

[1513] Input: Adjusted incorrect answer data, correct information data

[1514] Operation:

[1515] The server combines incorrect and correct answers to generate lesson scripts and teaching materials.

[1516] Output: Generated lesson scripts and teaching materials

[1517] The generated teaching materials and scripts are sent to the terminal, and the user confirms them on a confirmation screen.

[1518] Step 7:

[1519] The user starts the lesson through the terminal.

[1520] Input: Verified lesson scripts and materials

[1521] Operation:

[1522] Lessons are conducted based on lesson scripts created by the user, and incorrect information is presented to encourage learners to solve the problems themselves.

[1523] Output: Lesson data, student response data

[1524] The device supports the progress of the class in real time.

[1525] Step 8:

[1526] The device uses voice recognition technology to record students' comments and reactions during class and recognize the user's emotional state.

[1527] Input: Learner's speech, user's facial expression and tone of voice

[1528] Operation:

[1529] The device collects data in real time using voice and emotion recognition technology.

[1530] Output: Learner's speech data, user's emotion data

[1531] The collected data is sent to a server.

[1532] Step 9:

[1533] The server analyzes the recorded data and provides real-time feedback and suggestions for next learning steps to the user.

[1534] Input: Learner's speech data, user's emotion data

[1535] Operation:

[1536] The server analyzes the data and generates next steps and feedback.

[1537] Output: Real-time feedback, suggestions for next learning steps

[1538] Feedback is sent to the terminal and the user is notified.

[1539] Step 10:

[1540] After the lesson ends, the server generates a summary based on the data from the lesson, the learners' reactions, and the user's emotional data.

[1541] Input: lesson data, learner response data, user emotion data

[1542] Operation:

[1543] The server generates a summary of the lesson content to help plan the next lesson.

[1544] Output: Summary report

[1545] A summary is sent to the terminal for the user to review.

[1546] (Application example 2)

[1547] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1548] Conventional educational support systems have difficulty optimizing lesson content according to the user's emotional state, and providing real-time feedback based on the learner's reactions is insufficient. Furthermore, because they do not recommend individually optimized content based on the learner's emotional state, there are issues such as a decrease in learning efficiency and difficulty in maintaining motivation.

[1549] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for a user to input lesson content; means for analyzing the input lesson content and searching for related resources; means for calling a generative AI model that generates erroneous information based on the analysis results; means for adjusting the generated erroneous information; means for generating lesson scripts and teaching materials based on the generated information; means for providing the generated lesson scripts and teaching materials to the user; means for a user to present erroneous information during a lesson to encourage the learner to think; means for recording the learner's responses during the lesson and transmitting them to the server; means for analyzing the recorded responses and providing feedback in real time; means for generating a summary of the lesson content after the lesson ends; emotion recognition means for recognizing the user's emotional state and adapting the lesson content based on the state; and means including a generative AI model that recommends content based on the learner's emotional state in real time. This enables real-time feedback and individually optimized content recommendations based on the user's emotional state and the learner's responses.

[1550] The "means for users to input lesson content" refers to an interface that allows educators to input information such as lesson content, subjects, target grades, and topics.

[1551] "Means for analyzing the input lesson content and searching for related resources" refers to the function of analyzing the input lesson content and automatically searching for and collecting resources such as reference materials, literature, and problem sets related to that content.

[1552] "Means for invoking a generative AI model that generates erroneous information based on the analysis results" refers to the process of using the analysis results of the lesson content to launch an AI model that intentionally generates erroneous information and applying it to that model.

[1553] The "means for adjusting the generated erroneous information" is a function that appropriately adjusts the erroneous information generated by the generative AI model according to the user's emotional state and the learner's level of understanding.

[1554] The "means for generating lesson scripts and teaching materials based on the generated information" is a function that automatically generates scripts and teaching materials to be used in lessons based on the generated erroneous information and accurate information.

[1555] "Means for providing the generated lesson scripts and teaching materials to the user" is a function that provides the generated lesson scripts and teaching materials through an interface so that the educator can check and modify them.

[1556] "Means for users to present incorrect information during class to encourage students to think" refers to a process in which an educator intentionally presents incorrect information as part of a lesson and encourages students to think about and correct the error themselves.

[1557] "Means for recording learners' responses during lessons and sending them to a server" refers to a function that records learners' reactions, comments, facial expressions, and other responses, and sends them to a server for storage.

[1558] "Means for analyzing recorded responses and providing feedback in real time" refers to a function that analyzes recorded responses of learners and provides feedback to educators in real time based on the results.

[1559] "Means for generating a summary of the lesson content after the lesson has ended" is a function that generates a report summarizing the lesson content after the lesson has ended, based on data from the lesson and the students' responses.

[1560] "Emotion recognition means that recognizes the user's emotional state and adapts the lesson content based on that state" is a function that analyzes the educator's emotional state and optimizes the lesson content and progress based on the results.

[1561] "Means including a generative AI model that recommends content based on the learner's emotional state in real time" refers to a process that uses a generative AI model to grasp the learner's emotional state in real time and recommend content appropriate to that state.

[1562] This invention is an educational support system that recognizes the user's emotional state and optimizes lesson content accordingly. The system uses a generative AI model and emotion recognition means to provide real-time feedback and content recommendations based on the learner's reactions and emotions.

[1563] System configuration

[1564] The system includes the following major hardware and software components:

[1565] 1. User Interface (Terminal):

[1566] This is a device that allows educators to input lesson content and check and correct the generated scripts and teaching materials. This interface can be configured as a smartphone, tablet, or PC.

[1567] 2. Server:

[1568] This is the central computer that processes and stores data, calls generative AI models, and analyzes learner responses and user emotions. It runs on a cloud server.

[1569] 3. Generative AI Model:

[1570] This algorithm generates false information based on the content of lessons, providing learners with an opportunity to think independently. This model is implemented using AI frameworks such as Python and TensorFlow.

[1571] 4. Emotion Recognition Engine:

[1572] This algorithm recognizes the emotional state of users and learners from data acquired through cameras and microphones, and adjusts lesson content accordingly. It uses libraries such as OpenCV and dlib.

[1573] System Operation

[1574] User Authentication

[1575] The educator logs in to the system through the terminal and enters authentication information. The terminal sends the entered information to the server, which then authenticates the user. If authentication is successful, the server retrieves the user's personal settings and history and sends them to the terminal.

[1576] Input and generation of lesson content

[1577] The educator inputs the lesson content into the device. The input content is sent to the server, which analyzes the information and calls a generative AI model to generate incorrect answers. The generated incorrect answers are adjusted based on the user's emotional state using an emotion recognition engine. Lesson scripts and teaching materials are then automatically generated and provided to the educator.

[1578] Class implementation and response analysis

[1579] During the lesson, the educator follows the generated script and presents incorrect answers to encourage students to think. The device uses voice recognition technology to record what the student says and simultaneously recognizes the user's emotional state. This data is sent to the server and analyzed in real time. The server then provides real-time feedback to the educator based on this data.

[1580] Generate a summary after the lesson

[1581] After the lesson, the server generates a summary based on the data from the lesson, the learners' reactions, and the user's emotional data, and sends it to the device. The educator can then review the summary and incorporate it into their next lesson plan.

[1582] Specific examples

[1583] For example, consider a second-year junior high school math class with the topic of "quadratic equations." The educator selects "quadratic equations" as the topic and inputs the necessary information into the system. The server searches for related resources, and the generative AI model generates an incorrect answer (e.g., $x^2 + 4x + 4 = (x+2)^2 - 4$, an incorrect expansion). The emotion recognition engine adjusts the generated incorrect answer and lesson script based on the user's emotional data. The educator presents this incorrect answer during the lesson, encouraging the learner to derive the correct answer on their own. The device uses voice recognition technology and an emotion recognition engine to record the learner's reactions and the user's emotions during the lesson. After the lesson ends, the server analyzes this data and generates a summary to provide to the educator.

[1584] Prompt Sentence Examples

[1585] "If an emotion recognition engine detects that a user is feeling stressed, create a program that recommends relaxing music or movies to watch in that state."

[1586] As described above, the present invention is a system that can improve learning effectiveness and motivation by optimizing the method of providing lessons and content based on the emotional states of both the user and the learner.

[1587] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1588] Step 1:

[1589] The user inputs the lesson content (input content: lesson subject, target grade, topic, etc.)

[1590] The user inputs the lesson contents into the terminal interface and sends the input data to the server, which receives the data and prepares it for the next analysis step.

[1591] Step 2:

[1592] The server analyzes the input lesson content and searches for related resources (input content: lesson content data, output content: related resources)

[1593] The server analyzes the input lesson content using a text analysis algorithm, searches for related educational resources (reference materials, workbooks, etc.), and creates a list.

[1594] Step 3:

[1595] The server calls a generative AI model that generates incorrect information based on the analysis results (input: analysis results, output: incorrect information).

[1596] The server inputs the analysis results into a generative AI model, which is built using AI frameworks such as Python and TensorFlow, to generate false information related to the lesson content.

[1597] Step 4:

[1598] Adjusting the generated misinformation (input: misinformation, user's emotional state, output: adjusted misinformation)

[1599] The server uses an emotion recognition engine to analyze the user's emotional state and adjusts the generated erroneous information based on that state, adjusting the level of erroneous answers to an appropriate level according to the user's stress or excitement level.

[1600] Step 5:

[1601] Generate lesson scripts and teaching materials based on the generated information (input: adjusted incorrect information, correct information; output: lesson scripts, teaching materials)

[1602] The server automatically generates lesson scripts and teaching materials by combining the correct information with the adjusted incorrect information. These scripts and teaching materials are output in a format that users can review and edit.

[1603] Step 6:

[1604] The generated lesson scripts and teaching materials are provided to the user (input content: lesson script, teaching materials, output content: displayed on the user's device)

[1605] The server sends the generated lesson scripts and teaching materials to the user's terminal, where the user can check them and make corrections if necessary.

[1606] Step 7:

[1607] The user presents incorrect information during class to encourage students to think about it (input: class script, teaching materials, output: encourage students to think)

[1608] During class, the user presents incorrect information according to the generated script and prompts the learner to solve it, thereby developing the learner's independent problem-solving ability.

[1609] Step 8:

[1610] During the lesson, learners' responses are recorded and sent to the server (input: learners' responses, output: recorded data)

[1611] The device uses a microphone and camera to record the learner's responses and transmits the data to a server, using voice recognition and video analysis technology.

[1612] Step 9:

[1613] Analyze recorded responses and provide real-time feedback (input: recorded data, output: feedback)

[1614] The server analyzes the recorded student response data, generates real-time feedback, and sends it to the device, helping the user to progress through the lesson.

[1615] Step 10:

[1616] Generate a summary of the lesson content after the lesson ends (input content: data from the lesson, output content: summary)

[1617] The server generates a summary of the lesson content based on data from the lesson, learners' reactions, and user emotional data, and sends it to the user's device. The user can review this summary and reflect it in their next lesson plan.

[1618] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1619] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1620] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1621] [Fourth embodiment]

[1622] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1623] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1624] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1625] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1626] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1627] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1628] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1629] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1630] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1631] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1632] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1633] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1634] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1635] This invention relates to "Empower GPT," a system for supporting teachers in the educational field. This system uses generative AI to provide false information, aiming to develop learners' ability to solve problems independently.

[1636] System Overview

[1637] The system includes the following main components:

[1638] 1. User interface (terminal): A device that allows teachers to input lesson content and check and modify the generated scripts and teaching materials.

[1639] 2. Server: A central computer that processes and stores data, calls generative AI models, and analyzes learner responses.

[1640] 3. Generative AI model: An artificial intelligence algorithm that generates misinformation based on lesson content, providing learners with an opportunity to think for themselves.

[1641] Program processing procedure

[1642] 1. User login and authentication:

[1643] The user logs into the system from a terminal and enters the necessary authentication information.

[1644] The server verifies the authentication information and sets up the appropriate environment for each user.

[1645] 2. Input lesson content:

[1646] The user inputs information such as the class subject, target grade, topic, etc. from the terminal.

[1647] The terminal transmits this information to the server.

[1648] 3. Analyzing information and finding resources:

[1649] The server analyzes the input lesson content and searches for related resources (teaching materials, past lesson content, etc.).

[1650] 4. Creation of misinformation:

[1651] The server calls the generative AI model and generates false information based on the analysis results.

[1652] The generated misinformation is tailored to the learner's level of understanding and grade level.

[1653] 5. Generating lesson scripts and teaching materials:

[1654] The server generates lesson scripts and teaching materials based on the generated incorrect information and correct information.

[1655] The generated scripts and teaching materials are sent to the terminal and checked by the user.

[1656] 6. Class Conduct:

[1657] A user presents false information during a lesson to initiate a process of thinking in the learners.

[1658] Generative AI models respond to users' questions and concerns in real time.

[1659] 7. Recording and analyzing learner responses:

[1660] The device uses voice recognition technology to record the learner's responses.

[1661] The recorded data is sent to a server, which analyzes it.

[1662] 8. Providing Feedback:

[1663] The server provides real-time feedback based on the analysis results and suggests next steps to the learner.

[1664] 9. Post-class summary generation:

[1665] After the lesson ends, the server generates a summary based on the lesson content and the learners' responses.

[1666] The summary is sent to the terminal, where the user can review it and reflect it in the next lesson plan.

[1667] Specific examples

[1668] For example, consider a case where the topic of a second-year junior high school math class is "quadratic equations." When a user selects "quadratic equations" as the topic, the server searches for related resources, and the generative AI model generates an incorrect answer (e.g., $x^2 + 4x + 4 = (x+2)^2 - 4$, an incorrect expansion). The user presents this incorrect answer during class, encouraging students to derive the correct answer on their own. After the class, the server generates a summary based on the content of the class and the students' responses, which can be used for the next class.

[1669] In this way, the present invention provides a system that promotes independent learning by learners in educational settings and reduces the burden on teachers.

[1670] The processing flow will be explained below.

[1671] Step 1: User Login and Authentication

[1672] The user enters the information (user name, password) required to log in to the system from the terminal.

[1673] The terminal sends the entered login information to the server.

[1674] The server checks the received login information against a database and performs authentication.

[1675] If authentication is successful, the server obtains the user's personal settings and history and sends them to the terminal.

[1676] The user confirms successful authentication through the terminal and is then able to access the system.

[1677] Step 2: Enter lesson content

[1678] The user enters information such as the class subject, target grade, and topic from the terminal.

[1679] The terminal sends the input lesson content to the server.

[1680] Step 3: Analyze the information and find resources

[1681] The server analyzes the input lesson content.

[1682] Based on the analysis results, the server searches for relevant resources (teaching materials, past lesson contents, reference materials, etc.) and retrieves them from the database.

[1683] Step 4: Creating misinformation

[1684] The server calls the generative AI model and generates incorrect information based on the lesson content and analysis results.

[1685] The generated erroneous information is appropriately adjusted according to the learner's level of understanding and the target grade level.

[1686] Step 5: Generate lesson scripts and teaching materials

[1687] The server combines the generated incorrect information with the correct information to generate lesson scripts and teaching materials.

[1688] The generated lesson scripts and teaching materials are sent to the terminal, where the user can check and modify them.

[1689] Step 6: Conducting lessons

[1690] The user starts the lesson through the terminal.

[1691] The user follows a generated lesson script to intentionally present incorrect information and encourage learners to think.

[1692] During lessons, the generative AI model responds to and supports users' questions and concerns in real time.

[1693] Step 7: Record and analyze learner responses

[1694] The device uses voice recognition technology to record students' comments and reactions during class.

[1695] The recorded data is sent to a server, which analyzes it.

[1696] Based on the analysis results, the server provides real-time feedback and suggests next learning steps to the user.

[1697] Step 8: Generate a summary after the lesson

[1698] After the lesson ends, the server generates a summary based on the data from the lesson and the learners' responses.

[1699] The generated summary is sent to the terminal, where the user can check the lesson content and the learners' level of understanding and reflect it in future lesson plans.

[1700] Example 1

[1701] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1702] In today's educational settings, there are no fully established methods for teachers to provide appropriate teaching materials and misinformation according to each student's level of understanding, and for students to develop the ability to solve problems independently based on that information. Furthermore, it is difficult to analyze students' responses in real time and provide appropriate feedback. This increases the burden on teachers and hinders the promotion of independent learning among students, creating the problem of a lack of understanding.

[1703] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1704] In this invention, the server includes: means for a user to input lesson content; means for analyzing the input lesson content and searching for related resources; means for calling a generative AI model that generates erroneous information based on the analysis results; means for adjusting the generated erroneous information; means for generating lesson scripts and teaching materials based on the generated information; means for providing the generated lesson scripts and teaching materials to the user; means for a user to present erroneous information during class to encourage learners to think; means for recording learners' responses during class and transmitting them to the server; means for analyzing the recorded responses and providing feedback in real time; means for generating a summary of the lesson content after the class ends; means for adjusting the difficulty level of the generated erroneous information according to the learners' level of understanding; and means including a generative AI model that responds to user questions and doubts in real time during class. This reduces the burden on teachers and promotes learners' independent learning.

[1705] "Users" refers to teachers and educators who use this system in educational settings.

[1706] "Lesson content" refers to information such as subject, target grade, and topic that teachers enter into the system.

[1707] "Analysis" refers to the process of breaking down input lesson content using natural language processing techniques or other methods to identify relevant resources.

[1708] "Resources" refers to teaching materials, reference materials, past lesson data, etc. related to the content of the lesson.

[1709] A "generative AI model" refers to the part of a system that uses artificial intelligence algorithms to generate false information based on specific inputs.

[1710] A "prompt" refers to input text used to instruct a generative AI model on what information to generate.

[1711] "Misinformation" refers to incorrect data or answers that are intentionally generated as a starting point for learners to use their own understanding to derive accurate information.

[1712] A "script" refers to a script or instruction manual that helps guide a lesson.

[1713] "Instructional materials" refers to visual, textual, or hands-on learning materials used to supplement course content.

[1714] "Terminal" refers to a device such as a computer or tablet that teachers use to access and operate the system.

[1715] "Server" refers to the central computing system that processes and stores data, invokes generative AI models, analyzes learner responses, etc.

[1716] "Speech recognition technology" refers to technology that converts learners' speech into text.

[1717] "Feedback" refers to instructions, hints, advice, etc. provided based on the learner's responses.

[1718] A "summary" refers to a short report generated based on the lesson content and learner responses.

[1719] "Responses" refer to the level of understanding, questions, comments, etc. that learners show during class.

[1720] This invention relates to a system called "Empower GPT" that supports teachers in the educational field. This system uses generative AI to provide false information and aims to cultivate learners' ability to solve problems independently. The following describes in detail the mode for carrying out the invention.

[1721] System configuration

[1722] The system includes the following main components:

[1723] 1. User interface (terminal): A device on which teachers input lesson content and check and edit generated scripts and teaching materials. Examples include PCs, tablets, and smartphones.

[1724] 2. Server: This is the central computer that processes and stores data, calls the generative AI model, and analyzes learner responses. In this invention, the server is assumed to be a high-performance computer with a Linux-based OS. MySQL, PostgreSQL, or other databases are used.

[1725] 3. Generative AI model: An artificial intelligence algorithm that generates false information based on the lesson content input by the teacher, providing learners with an opportunity to think for themselves. Specific examples include OpenAI's GPT-3 and GPT-4.

[1726] Program processing procedure

[1727] The system begins when the user (teacher) logs in and inputs the lesson content. The specific processing steps are explained below.

[1728] 1. User login and authentication:

[1729] A user logs in to the system from a terminal and enters the necessary authentication information, such as a user ID and password.

[1730] The server verifies the authentication information and sets up the appropriate environment for each individual user, a process that involves the server verifying user information against a database.

[1731] 2. Input lesson content:

[1732] The user inputs information such as the class subject, target grade, topic, etc., into the terminal. For example, they input "mathematics," "second-year junior high school student," and "quadratic equations."

[1733] The device sends this information to the server, and the data is sent in JSON format.

[1734] 3. Analyzing information and finding resources:

[1735] The server analyzes the input lesson content and searches the database for related resources (teaching materials, past lesson content, etc.) using natural language processing libraries (e.g., NLTK and spaCy).

[1736] 4. Creation of misinformation:

[1737] The server calls the generative AI model and generates false information based on the analysis results. The generative AI model is instructed using prompts.

[1738] The generated misinformation is tailored to the learner's level of understanding and grade level.

[1739] 5. Generating lesson scripts and teaching materials:

[1740] The server generates lesson scripts and teaching materials based on the generated incorrect information and correct information. For example, it uses Python or JavaScript scripts to build teaching materials.

[1741] The generated scripts and teaching materials are sent to the terminal and checked by the user.

[1742] 6. Class Conduct:

[1743] The user presents incorrect information during a lesson to initiate the thinking process in the learner, for example, by using an incorrect formula such as "x^2 + 4x + 4 = (x+2)^2 - 4".

[1744] The generative AI model responds to user questions and concerns in real time, including when users type questions into the generative AI, which then responds instantly.

[1745] 7. Recording and analyzing learner responses:

[1746] The device records the learner's responses using voice recognition technology, such as Google Speech-to-Text or Amazon Transcribe.

[1747] The recorded data is sent to a server, which analyzes it using machine learning algorithms (e.g., scikit-learn or TensorFlow).

[1748] 8. Providing Feedback:

[1749] The server provides real-time feedback based on the analysis results and suggests next steps to the learner, such as "Many students got this question wrong, so let's explain it again."

[1750] 9. Post-class summary generation:

[1751] After the lesson, the server generates a summary based on the lesson content and the learners' responses. The summary is created using natural language generation technology.

[1752] The summary is sent to the terminal, where the user can review it and reflect it in the next lesson plan.

[1753] Specific examples

[1754] For example, if the topic of a second-year junior high school math class is "quadratic equations," when the user selects "quadratic equations" as the topic, the server searches for related resources and the generative AI model generates an incorrect answer (e.g., $x^2 + 4x + 4 = (x+2)^2 - 4$, an incorrect expansion). The user presents this incorrect answer during class, encouraging the learners to derive the correct answer on their own. After the class, the server generates a summary based on the content of the class and the learners' responses, which can be used for the next class.

[1755] Examples of prompt statements

[1756] Topic: 8th Grade Math - Quadratic Equations

[1757] Objectives: Students will understand how to solve quadratic equations and be able to correct incorrect information themselves.

[1758] Examples of misinformation it generates:

[1759] $x^2 ​​+ 4x + 4 = (x+2)^2 - 4$

[1760] "The only solution to a quadratic equation is x = 2"

[1761] In this way, the present invention provides a system that promotes independent learning by learners in educational settings and reduces the burden on teachers.

[1762] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1763] Step 1:

[1764] A user accesses the system from a terminal and enters login information (user ID and password). The input data is encrypted on the terminal as authentication information and sent to the server. The server receives the authentication information and checks it against a database. Based on the check result, if authentication is successful, the environment settings corresponding to the user are loaded and a dashboard screen is displayed on the terminal. The output is a dashboard screen that the user can access.

[1765] Step 2:

[1766] The user inputs the class subject, target grade, and specific topic into the device. The input data includes information such as "Mathematics," "Second Year of Junior High School," and "Quadratic Equations." The device sends the input data in JSON format to the server. The server temporarily stores the received data and prepares for the analysis step. The output is a notification that the data has been saved on the server.

[1767] Step 3:

[1768] The server analyzes the received lesson content data. A natural language processing library (e.g., NLTK or spaCy) is used for the analysis. The input for the analysis is JSON data about the lesson content. As a result of the analysis, search keywords and tags for related resources (teaching materials, past lesson content, etc.) are generated. The output is the search keywords and tags.

[1769] Step 4:

[1770] The server searches for related resources from databases and cloud storage based on the analysis results. The input in this step is search keywords and tags. Related resources (PDF teaching materials, video links, past lesson scripts, etc.) are obtained as search results. The output is a list of the obtained resources.

[1771] Step 5:

[1772] The server invokes a generative AI model to generate false information using a prompt. The input in this step is the analysis result of the lesson content and the prompt. The generative AI model (e.g., GPT-3 or GPT-4) generates false information based on the input prompt. The output is the generated false information.

[1773] Step 6:

[1774] The server adjusts the generated misinformation according to the learner's level of understanding and grade level. The input is the generated misinformation and the learner's profile data. The adjusted information is generated and output. The adjustment may use machine learning algorithms.

[1775] Step 7:

[1776] The server generates lesson scripts and teaching materials based on the adjusted misinformation and correct information. The input data is the adjusted misinformation and related correct information. The lesson scripts and teaching materials are generated using programming languages ​​such as Python and JavaScript. The output is the completed lesson scripts and teaching materials.

[1777] Step 8:

[1778] The server sends the generated lesson script and teaching materials to the terminal, where the user can review them. The output is a script and teaching materials that the user can review on the terminal. The user can then modify the content as needed.

[1779] Step 9:

[1780] The user presents false information generated during class to prompt students to think. The input is the adjusted false information. As a concrete example, a teacher writes "$x^2 + 4x + 4 = (x+2)^2 - 4$" on the blackboard and asks students to think about what is wrong. The output is the student's response.

[1781] Step 10:

[1782] The device uses voice recognition technology (e.g., Google Speech-to-Text or Amazon Transcribe) to record the learner's responses and send them to the server. The input is the learner's speech. The voice data is converted into text format, and the output is the text data and its analysis results.

[1783] Step 11:

[1784] The server analyzes the recorded responses of the learner and provides feedback in real time. The input is the learner's response data in text format. Based on the analysis results, appropriate feedback is generated, and the output is the content of the feedback.

[1785] Step 12:

[1786] After the lesson ends, the server generates a summary based on the lesson content and learner responses. The input is the lesson script, teaching materials, and learner response data. The summary is created using natural language generation technology. The output is a summary that is sent to the terminal and reviewed by the user.

[1787] (Application example 1)

[1788] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1789] In conventional educational systems, when a single teacher instructs a large number of students, it is difficult to grasp each student's level of understanding and reactions in real time and provide appropriate feedback. Furthermore, in order for students to develop their independent problem-solving skills, they need teaching materials that contain not only accurate information but also incorrect information, but generating and adjusting such materials is time-consuming. This increases the burden on teachers and makes it difficult to improve students' independent learning abilities.

[1790] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1791] In this invention, the server includes a means for users to input lesson content, a means for analyzing the input lesson content and searching for related resources, and a means for calling a generative AI model that generates erroneous information based on the analysis results. This makes it possible to generate and adjust erroneous information and answers and provide appropriate feedback in real time. Furthermore, the server includes a means for users to select the content they want to learn and provide teaching materials containing erroneous information, a means for providing real-time feedback to students based on the generated erroneous information and answers, and a means for recording and saving learned content and answers so that they can be reviewed later. This facilitates the improvement of learners' independent learning abilities while reducing the burden on teachers.

[1792] "Means for users to input lesson content" refers to interfaces and applications that allow teachers and students to input the subjects or topics they want to learn into digital devices.

[1793] "Means for analyzing input lesson content and searching for related resources" refers to an algorithm that searches and retrieves related information and materials from databases or networks based on the input teaching materials or topics.

[1794] "Means for invoking a generative AI model that generates erroneous information based on the analysis results" refers to a mechanism that uses the analysis results of the input content to execute a process that intentionally creates erroneous information using a generative AI model.

[1795] The "means for adjusting generated erroneous information" is a function for appropriately adjusting the difficulty and content of the generated erroneous information according to the learner's level of understanding and comprehension.

[1796] The "means for generating lesson scripts and teaching materials based on the generated information" is a system that combines incorrect information with correct information to create and provide scripts and educational materials to be used in actual lessons.

[1797] The "means for providing the generated lesson scripts and teaching materials to users" refers to a mechanism for distributing the generated lesson scripts and teaching materials to teachers and learners via digital devices.

[1798] "Means for users to present incorrect information during class and encourage students to think" is a process in which a teacher intentionally presents incorrect information during class and has students point out the errors.

[1799] "Means for recording learners' responses during class and sending them to a server" refers to a system that has the function of recording how learners respond during class and sending and saving them to a server.

[1800] The "means for analyzing recorded responses and providing real-time feedback" refers to an algorithm that analyzes recorded responses of learners and provides appropriate feedback immediately.

[1801] The "means for generating a summary of the content of a lesson after the lesson has ended" is a system that summarizes the content of the lesson after the lesson has ended, generates a summary, and provides it.

[1802] "Means for providing students with real-time feedback based on incorrect information or answers" refers to a system that provides supplementary information or advice in real time to learners as they solve problems, based on the incorrect information or answers they generate.

[1803] "A means to record and save what has been learned and answers, and to review them later" is a system that has the function of saving what learners have learned and answers in a database, allowing them to check and review them later.

[1804] This invention relates to an educational support system that utilizes a generative AI model. The purpose of this invention is to promote learners' independent learning and reduce the burden on teachers by cooperating with three main components: a server, a terminal, and a user.

[1805] System configuration

[1806] 1. Server:

[1807] The server has the following functions:

[1808] Provides a means for users to input lesson content.

[1809] It has the function of analyzing the input lesson content and searching for related resources.

[1810] Based on the analysis results, a generative AI model is invoked to generate false information.

[1811] The generated incorrect information is adjusted and lesson scripts and teaching materials are generated.

[1812] Analyze recorded learner responses and provide feedback in real time.

[1813] 2. Terminal:

[1814] Terminals are devices used by teachers and students and have the following functions:

[1815] It provides a user interface for inputting, checking, and modifying lesson content.

[1816] Display the generated lesson scripts and teaching materials.

[1817] Learners' responses are recorded during the lesson and sent to the server.

[1818] 3. User:

[1819] Users are mainly teachers and students, and use the system in the following steps:

[1820] Enter the lesson content from the terminal.

[1821] Check and modify the generated scripts and teaching materials.

[1822] Present misinformation during class and encourage students to problem-solve.

[1823] Record learners' responses and review their feedback after the lesson.

[1824] Software and Hardware Configuration

[1825] On the server side, it uses high-performance cloud servers such as Amazon Web Services and Google Cloud Platform, and the backend is built using Node.js and Express, with MongoDB as the database.

[1826] The front-end part will be developed using React Native, taking into consideration use on smartphones.

[1827] The generative AI model uses OpenAI GPT-4, which generates prompts based on the input lesson content and then creates false information as a response to those prompts.

[1828] As a use case, consider a 8th grade math lesson on "quadratic equations," where the following prompt is sent to the generative AI model:

[1829] "Generate false information and answers on the following topic: Quadratic Equations"

[1830] The generative AI model generates incorrect information for this prompt (e.g., an incorrect expansion such as $x^2 + 4x + 4 = (x+2)^2 - 4$), which is then provided as part of the learning material.

[1831] The server analyzes the learner's responses and provides real-time feedback, for example, helping the learner recognize incorrect information and guide them in the process of arriving at the correct answer.

[1832] This system will enable students to develop the ability to correct their own mistakes and reduce the burden on teachers. It is also expected that the inclusion of incorrect information will dramatically improve students' thinking and problem-solving abilities.

[1833] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1834] Step 1:

[1835] The user inputs the lesson content.

[1836] In this step, the user enters information such as the class subject, target grade, topic, etc. into the application via their device. The entered information is sent from the front end (React Native) to the server (Node.js).

[1837] Input: Course subject, target grade, topic

[1838] Output: The input data is sent to the server

[1839] Step 2:

[1840] The server analyzes the content of the lesson and searches for related resources.

[1841] The server analyzes the received lesson content and searches a database (MongoDB) for teaching materials and past resources related to that topic.

[1842] Input: Lesson content sent in Step 1

[1843] Output: A list of resources related to the analysis results

[1844] Step 3:

[1845] The server calls a generative AI model based on the analysis results and generates incorrect information.

[1846] Based on the analysis results, the server sends a prompt to the generative AI model (OpenAI GPT-4) to generate incorrect information.

[1847] Input: Analysis results, prompt (e.g. "Generate incorrect information about quadratic equations")

[1848] Output: The incorrect information generated

[1849] Step 4:

[1850] The server will adjust the generated incorrect information.

[1851] The server runs an algorithm that adjusts the generated misinformation according to the learner's level of comprehension and grade level, optimizing the difficulty of the misinformation.

[1852] Input: Generated misinformation, learner comprehension information

[1853] Output: Adjusted misinformation

[1854] Step 5:

[1855] The server generates lesson scripts and teaching materials based on the generated information.

[1856] The server generates lesson scripts and teaching materials based on the adjusted incorrect information and correct information and provides them to users.

[1857] Input: Adjusted incorrect information, correct information

[1858] Output: Lesson scripts, teaching materials

[1859] Step 6:

[1860] The user presents incorrect information during class to prompt students to think.

[1861] The user uses the generated teaching materials in class, presents incorrect information to students, and has them point out the errors, thereby encouraging the students to think independently.

[1862] Input: lesson scripts, teaching materials

[1863] Output: Learner response

[1864] Step 7:

[1865] The device records the learners' responses during class and sends them to the server.

[1866] Learners' responses and reactions are recorded by the device and sent to the server in real time. Speech recognition technology is also used to record utterances.

[1867] Input: Learner response data

[1868] Output: Response data sent to the server

[1869] Step 8:

[1870] The server analyzes the recorded responses and provides real-time feedback.

[1871] The server analyzes the received learner response data and uses an AI model to generate and provide instant feedback.

[1872] Input: Reaction data

[1873] Output: Real-time feedback

[1874] Step 9:

[1875] After the lesson ends, the server generates a summary of the lesson content.

[1876] After the lesson is over, the server creates a summary based on the lesson content and the learners' reactions and provides it to the user to help plan the next lesson.

[1877] Input: Lesson content, learner response data

[1878] Output: Lesson summary

[1879] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1880] This invention relates to "Empower GPT," an educational support system with the ability to recognize users' emotions and adapt lesson scripts and teaching materials. The system aims to optimize lessons based on the user's emotional state while providing false information using generative AI to foster learners' ability to solve problems independently.

[1881] System Overview

[1882] The system includes the following main components:

[1883] 1. User interface (terminal): A device that allows teachers to input lesson content and check and modify the generated scripts and teaching materials.

[1884] 2. Server: The central computer that processes and stores data, invokes generative AI models, and analyzes learner responses and user sentiment.

[1885] 3. Generative AI model: An artificial intelligence algorithm that generates misinformation based on lesson content, providing learners with an opportunity to think independently.

[1886] 4. Emotion Engine: An algorithm to recognize the user's emotional state and adjust the lesson script and teaching materials.

[1887] Program processing procedure

[1888] 1. User login and authentication:

[1889] The user logs into the system from a terminal and enters their authentication information.

[1890] The terminal sends the entered login information to the server, and the server performs authentication.

[1891] If authentication is successful, the server obtains the user's personal settings and history and sends them to the device.

[1892] 2. Input lesson content:

[1893] The user enters information such as the class subject, target grade, and topic into the terminal.

[1894] The terminal transmits this information to the server.

[1895] 3. Analyzing information and finding resources:

[1896] The server analyzes the input lesson content and searches for and organizes related resources.

[1897] 4. Creation of misinformation:

[1898] The server calls the generative AI model and generates incorrect information based on the lesson content and analysis results.

[1899] The generative AI model adjusts the generated incorrect answers based on the user's emotional data.

[1900] 5. Generating lesson scripts and teaching materials:

[1901] The server combines the generated incorrect information with the correct information to generate lesson scripts and teaching materials.

[1902] The generated lesson scripts and teaching materials are sent to the terminal, where the user can check and modify them.

[1903] 6. Class Conduct:

[1904] The user starts the lesson through the terminal.

[1905] The user presents incorrect information according to the generated lesson script, prompting learners to think.

[1906] Generative AI models respond to users' questions and concerns in real time.

[1907] 7. Recording and analyzing learner responses and user emotions:

[1908] The device uses voice recognition technology to record students' comments and reactions during class.

[1909] The terminal recognizes the user's emotional state and transmits the emotional data to the server.

[1910] The recorded data and emotional data are sent to a server for analysis.

[1911] Based on the analysis results, the server provides real-time feedback and suggests next learning steps to the user.

[1912] 8. Post-class summary generation:

[1913] After the lesson ends, the server generates a summary based on the data from the lesson, the learners' reactions, and the user's emotional data.

[1914] A summary is sent to the terminal, and the user checks the lesson content and the learners' level of understanding, and reflects this in the next lesson plan.

[1915] Specific examples

[1916] For example, consider a case where the topic of a math class for second-year junior high school students is "quadratic equations." The user selects "quadratic equations" as the topic and enters the necessary information into the system. The server searches for related resources, and the generative AI model generates an incorrect answer (e.g., $x^2 + 4x + 4 = (x+2)^2 - 4$, an incorrect expansion). The emotion engine adjusts the generated incorrect answers and lesson scripts based on the user's emotional data.

[1917] The user presents the incorrect answer during the lesson, encouraging the learner to find the correct answer on their own. The device uses voice recognition technology and an emotion recognition engine to record the learner's responses and the user's emotions during the lesson. After the lesson ends, the server analyzes this data, generates a summary, and provides it to the user. In this way, it is possible to promote learners' independent learning and conduct lessons that are tailored to the user's emotional state.

[1918] The processing flow will be explained below.

[1919] Step 1: User Login and Authentication

[1920] The user enters the information (user name, password) required to log in to the system from the terminal.

[1921] The terminal sends the entered login information to the server.

[1922] The server checks the received login information against a database and performs authentication.

[1923] If authentication is successful, the server obtains the user's personal settings and history and sends them to the terminal.

[1924] The user can then verify their authentication through the terminal and gain access to the system.

[1925] Step 2: Enter lesson content

[1926] The user enters information such as the class subject, target grade, and topic from the terminal.

[1927] The terminal sends the input lesson content to the server.

[1928] Step 3: Analyze the information and find resources

[1929] The server analyzes the input lesson content.

[1930] The server searches for and retrieves relevant resources (teaching materials, past lesson contents, reference materials, etc.) based on the analysis results.

[1931] Step 4: Creating misinformation

[1932] The server calls the generative AI model and generates incorrect information based on the lesson content and analysis results.

[1933] The server adjusts the difficulty and content of the generated erroneous information according to the learner's level of understanding.

[1934] Step 5: Emotional Engine Alignment

[1935] The device monitors the user's emotional state in real time and transmits the emotional data to a server.

[1936] The server uses an emotion engine to analyze the emotional data and adjust the generated incorrect answers and lesson scripts.

[1937] Step 6: Generate lesson scripts and teaching materials

[1938] The server combines the generated incorrect information with the correct information to generate lesson scripts and teaching materials.

[1939] The generated lesson scripts and teaching materials are sent to the terminal, where the user can check and modify them.

[1940] Step 7: Conducting lessons

[1941] The user starts the lesson through the terminal.

[1942] The user follows a generated lesson script to intentionally present incorrect information and encourage learners to think.

[1943] During lessons, the generative AI model responds to users' questions and concerns in real time.

[1944] Step 8: Recording and analyzing learner responses and user emotions

[1945] The device uses voice recognition technology to record students' comments and reactions during class.

[1946] The terminal recognizes the user's emotional state and transmits the emotional data to the server.

[1947] The server analyzes the recorded learner responses and user emotion data.

[1948] Based on the analysis results, the server provides real-time feedback and suggests next learning steps to the user.

[1949] Step 9: Generate a summary after the lesson

[1950] After the lesson ends, the server generates a summary based on the data from the lesson, the learners' reactions, and the user's emotional data.

[1951] The generated summary is sent to the terminal, where the user can check the lesson content and the learners' level of understanding and reflect it in the next lesson plan.

[1952] Example 2

[1953] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1954] Traditional educational systems lacked effective methods for cultivating students' independent problem-solving skills. They also lacked mechanisms for adapting lesson content based on the user's emotional state, making it difficult to maximize educational effectiveness. As a result, they were unable to respond to fluctuations in students' interests and comprehension levels, limiting their ability to provide consistent educational content.

[1955] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1956] In this invention, the server includes means for a user to input lesson content, means for analyzing the input lesson content and searching for related resources, means for calling an AI algorithm that generates erroneous information based on the analysis results, means for adjusting the generated erroneous information, means for generating lesson scripts and teaching materials based on the generated information, means for providing the generated lesson scripts and teaching materials to the user, means for the user to present erroneous information during the lesson to encourage students to think, means for recording students' responses during the lesson and transmitting them to a central computer, means for analyzing the recorded responses and providing feedback in real time, means for generating a summary of the lesson content after the lesson ends, and means for recognizing the user's emotional state and adjusting the lesson scripts and teaching materials, thereby cultivating students' independent problem-solving skills and enabling the lesson content to be adapted based on the user's emotional state.

[1957] "Users" are educators and administrators who operate the system and input lesson content.

[1958] "Class content" refers to educational information such as the subjects, topics, and target grades covered in the class.

[1959] "Input means" refers to an interface or device that allows users to input lesson content into the system.

[1960] "Analysis means" refers to algorithms or devices that analyze the input lesson content and search for related resources.

[1961] "Misinformation" refers to intentionally inaccurate educational information provided to learners in order to solve problems independently.

[1962] "Artificial intelligence algorithms" refer to machine learning and natural language processing technologies that generate false information based on input information.

[1963] "Adjustment means" refers to processes or functions for adjusting erroneous information generated by artificial intelligence algorithms.

[1964] A "lesson script" is a script or plan used in a lesson, and is a document or digital data that contains the content as the lesson progresses.

[1965] "Teaching materials" refers to the teaching materials and learning resources used in classes (e.g., textbooks, workbooks, video materials, etc.).

[1966] "Recording means" refers to voice recognition technology or other digital recording devices used to record learner responses during lessons.

[1967] The "central computer" is the main device in the system that processes and stores data, calls generative AI models, and analyzes learner responses and user emotions.

[1968] "Analysis tools" are algorithms or devices that analyze recorded learner responses and provide real-time feedback.

[1969] An "emotional state" is a state that recognizes and classifies a user's emotions (e.g., joy, sadness, anger, surprise, etc.).

[1970] A "summary generation means" refers to an algorithm or device that creates a summary after a lesson based on the lesson content and learners' responses.

[1971] "Real-time feedback" refers to guidance and advice provided immediately or briefly during or after a lesson.

[1972] MODE FOR CARRYING OUT THE INVENTION

[1973] In the present invention, the following hardware and software are used to effectively implement the education support system.

[1974] (System Configuration)

[1975] 1. User Interface (Terminal)

[1976] 2. Server

[1977] 3. Generative AI Models

[1978] 4. Emotion Engine

[1979] (User Interface (Terminal))

[1980] A user interface is a device that allows users to input lesson content and review and edit generated scripts and teaching materials. This can be a computer, tablet, smartphone, or other digital device. It has a solid graphical user interface (GUI) and is designed to allow users to operate it intuitively.

[1981] (server)

[1982] The server is the central computer that processes and stores data, calls the generative AI model, and analyzes learner responses and user emotions. It uses a cloud server with powerful computing capabilities, and is able to exchange data with devices in real time via the communications infrastructure. The server integrates a database, analysis engine, and various algorithms.

[1983] (Generative AI model)

[1984] The generative AI model is an artificial intelligence algorithm that generates false information based on input lesson content, providing learners with an opportunity to think independently. It incorporates natural language processing (NLP) and machine learning (ML) technologies and has the ability to adaptively adjust the information it generates based on the user's emotional data.

[1985] (Emotion Engine)

[1986] The emotion engine is an algorithm that recognizes the user's emotional state and adjusts lesson scripts and teaching materials accordingly. It constantly monitors the user's emotional state in real time using voice recognition and facial expression analysis technologies.

[1987] Specific examples

[1988] For example, consider a second-year junior high school math class where the topic is "quadratic equations."

[1989] 1. The user enters the necessary information about the "quadratic equation" into the terminal.

[1990] 2. The terminal sends the entered information to the server.

[1991] 3. The server searches the database for relevant resources and organizes them, including external resources.

[1992] 4. The generative AI model generates an incorrect answer (e.g., $x^2 + 4x + 4 = (x+2)^2 - 4$, an incorrect expansion).

[1993] 5. The emotion engine adjusts generated incorrect answers and lesson scripts based on the user's emotional data.

[1994] 6. The user checks the lesson script and teaching materials and makes any necessary corrections.

[1995] 7. The user conducts the lesson according to the generated lesson script, presenting incorrect information and encouraging students to solve the problem themselves.

[1996] 8. The device uses voice recognition technology and an emotion engine to record learners' responses and the user's emotions during class.

[1997] 9. The server analyzes this data and provides feedback during and after the lesson.

[1998] Prompt Sentence Examples

[1999] "In an eighth-grade math class, explain what a quadratic equation is and generate an incorrect answer. For example, include the expansion of $(x + 3)^2 = x^2 + 6x + 9$."

[2000] "If the user's emotions are anxiety, explain how to adjust the incorrect answer to the quadratic equation."

[2001] In this way, the present invention aims to adapt lesson content based on the user's emotional state and foster learners' independent problem-solving abilities.

[2002] The flow of the identification process in the second embodiment will be described with reference to FIG.

[2003] Specific processing steps of the system program

[2004] Step 1:

[2005] The user logs into the system from a terminal and enters their authentication information.

[2006] Input: User login information (ID and password)

[2007] Operation:

[2008] The device sends the login information to the server.

[2009] The server receives the login information and authenticates it against a database.

[2010] Output: Authentication result (success or failure), and if successful, the user's personal settings data

[2011] If the authentication is successful, the server sends the user's personal setting information to the terminal, and the terminal displays the user's dashboard.

[2012] Step 2:

[2013] The user enters information such as the class subject, target grade, and topic into the terminal.

[2014] Input: Course content information (subject, target grade, topic, etc.)

[2015] Operation:

[2016] The terminal sends lesson content information to the server.

[2017] The server prepares the received data for analysis.

[2018] Output: Analysis results of lesson content data

[2019] The server checks your input to make sure you have all the necessary information.

[2020] Step 3:

[2021] The server analyzes the input lesson content and searches for related resources.

[2022] Input: Lesson content data

[2023] Operation:

[2024] The server analyzes the lesson content using natural language processing (NLP).

[2025] The server searches the database and the internet for relevant educational resources.

[2026] Output: List of related resources

[2027] The list of found resources is sent to the terminal and made available for the user to review.

[2028] Step 4:

[2029] The server calls the generative AI model and generates incorrect information based on the lesson content and analysis results.

[2030] Input: lesson content data, analysis results

[2031] Operation:

[2032] The generative AI model is called and begins the process of generating incorrect information (wrong answers) related to the lesson content.

[2033] Output: Generated incorrect answer data

[2034] As an example of an incorrect answer, "incorrect expansion of a quadratic equation" is generated.

[2035] Step 5:

[2036] An emotion engine adjusts generated incorrect answers based on the user's emotion data.

[2037] Input: wrong answer data, user emotion data

[2038] Operation:

[2039] The emotional engine analyzes the user's emotional state and adjusts the difficulty and type of incorrect answers.

[2040] Output: Adjusted error data

[2041] For example, if the user shows signs of anxiety, the difficulty level for incorrect answers is set low.

[2042] Step 6:

[2043] The server combines the generated incorrect information with the correct information to generate lesson scripts and teaching materials.

[2044] Input: Adjusted incorrect answer data, correct information data

[2045] Operation:

[2046] The server combines incorrect and correct answers to generate lesson scripts and teaching materials.

[2047] Output: Generated lesson scripts and teaching materials

[2048] The generated teaching materials and scripts are sent to the terminal, and the user confirms them on a confirmation screen.

[2049] Step 7:

[2050] The user starts the lesson through the terminal.

[2051] Input: Verified lesson scripts and materials

[2052] Operation:

[2053] Lessons are conducted based on lesson scripts created by the user, and incorrect information is presented to encourage learners to solve the problems themselves.

[2054] Output: Lesson data, student response data

[2055] The device supports the progress of the class in real time.

[2056] Step 8:

[2057] The device uses voice recognition technology to record students' comments and reactions during class and recognize the user's emotional state.

[2058] Input: Learner's speech, user's facial expression and tone of voice

[2059] Operation:

[2060] The device collects data in real time using voice and emotion recognition technology.

[2061] Output: Learner's speech data, user's emotion data

[2062] The collected data is sent to a server.

[2063] Step 9:

[2064] The server analyzes the recorded data and provides real-time feedback and suggestions for next learning steps to the user.

[2065] Input: Learner's speech data, user's emotion data

[2066] Operation:

[2067] The server analyzes the data and generates next steps and feedback.

[2068] Output: Real-time feedback, suggestions for next learning steps

[2069] Feedback is sent to the terminal and the user is notified.

[2070] Step 10:

[2071] After the lesson ends, the server generates a summary based on the data from the lesson, the learners' reactions, and the user's emotional data.

[2072] Input: lesson data, learner response data, user emotion data

[2073] Operation:

[2074] The server generates a summary of the lesson content to help plan the next lesson.

[2075] Output: Summary report

[2076] A summary is sent to the terminal for the user to review.

[2077] (Application example 2)

[2078] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[2079] Conventional educational support systems have difficulty optimizing lesson content according to the user's emotional state, and providing real-time feedback based on the learner's reactions is insufficient. Furthermore, because they do not recommend individually optimized content based on the learner's emotional state, there are issues such as a decrease in learning efficiency and difficulty in maintaining motivation.

[2080] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for a user to input lesson content; means for analyzing the input lesson content and searching for related resources; means for calling a generative AI model that generates erroneous information based on the analysis results; means for adjusting the generated erroneous information; means for generating lesson scripts and teaching materials based on the generated information; means for providing the generated lesson scripts and teaching materials to the user; means for a user to present erroneous information during a lesson to encourage the learner to think; means for recording the learner's responses during the lesson and transmitting them to the server; means for analyzing the recorded responses and providing feedback in real time; means for generating a summary of the lesson content after the lesson ends; emotion recognition means for recognizing the user's emotional state and adapting the lesson content based on the state; and means including a generative AI model that recommends content based on the learner's emotional state in real time. This enables real-time feedback and individually optimized content recommendations based on the user's emotional state and the learner's responses.

[2081] The "means for users to input lesson content" refers to an interface that allows educators to input information such as lesson content, subjects, target grades, and topics.

[2082] "Means for analyzing the input lesson content and searching for related resources" refers to the function of analyzing the input lesson content and automatically searching for and collecting resources such as reference materials, literature, and problem sets related to that content.

[2083] "Means for invoking a generative AI model that generates erroneous information based on the analysis results" refers to the process of using the analysis results of the lesson content to launch an AI model that intentionally generates erroneous information and applying it to that model.

[2084] The "means for adjusting the generated erroneous information" is a function that appropriately adjusts the erroneous information generated by the generative AI model according to the user's emotional state and the learner's level of understanding.

[2085] The "means for generating lesson scripts and teaching materials based on the generated information" is a function that automatically generates scripts and teaching materials to be used in lessons based on the generated erroneous information and accurate information.

[2086] "Means for providing the generated lesson scripts and teaching materials to the user" is a function that provides the generated lesson scripts and teaching materials through an interface so that the educator can check and modify them.

[2087] "Means for users to present incorrect information during class to encourage students to think" refers to a process in which an educator intentionally presents incorrect information as part of a lesson and encourages students to think about and correct the error themselves.

[2088] "Means for recording learners' responses during lessons and sending them to a server" refers to a function that records learners' reactions, comments, facial expressions, and other responses, and sends them to a server for ...

Claims

1. A means for a user to input lesson content; A means of analyzing the input lesson content and searching for related resources; A means for invoking a generative AI model that generates false information based on the analysis results; a means of reconciling any misinformation generated; A means for generating lesson scripts and teaching materials based on the generated information; A means for providing the generated lesson scripts and teaching materials to users; A means for users to present incorrect information during class and encourage students to think about it. a means for recording learner responses during the lesson and transmitting the recorded responses to a server; a means of analyzing the recorded responses and providing real-time feedback; A means for generating a summary of the content of a lesson after the lesson has ended; A system including:

2. 2. The system of claim 1, wherein the means for analyzing the learner's response includes means for recording the learner's utterances using voice recognition technology.

3. The system according to claim 1, characterized in that the generative AI model that generates erroneous information is capable of adjusting the difficulty of the information according to the learner's level of understanding.

Images