system

Abstract

We provide the system. [Solution] A means for generating a response to user input using a generative model, A means of generating custom learning content based on students' learning progress and interests, A means of planning and managing the progress of a user's learning plan, A learning support system that includes this.

Description

Technical Field

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

Background Art

[0002] [[ID=eleven]] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, the method including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In modern educational settings, there is a demand for individualized learning support and efficient guidance. However, teachers are burdened, and it is difficult to provide sufficient individual guidance to all students. In addition, with the spread of online learning, there is a need for an environment where students can effectively learn at home, but existing tools cannot fully meet individual needs. Furthermore, it is difficult to provide custom learning content according to different learning progress and interests of each student, and there is a lack of means to efficiently plan and manage the learning progress.

Means for Solving the Problems

[0005] The present invention solves the above problems by providing a learning support system that includes means for generating responses to user input using a generative model, means for generating custom learning content based on students' learning progress and interests, and means for formulating and managing users' learning plans. With this system, users can receive individualized instruction and effectively progress in their learning even in online learning. Furthermore, by providing customized learning content for each student, it becomes easier to maintain motivation for learning and to formulate concrete and realistic learning plans.

[0006] A "generative model" is an artificial intelligence model that generates responses in natural language based on user input.

[0007] "User" refers to a person who uses this learning support system to engage in learning activities.

[0008] "Response" refers to the explanations or answers that a generative model generates in response to user input.

[0009] "Learning progress" is an indicator that shows how far a student is progressing towards a specific learning objective.

[0010] "Interest" refers to the interest or curiosity that users or students have towards specific learning content or fields.

[0011] "Custom learning content" refers to learning materials and exercises that are individually generated based on the user's learning progress and interests.

[0012] A "learning plan" refers to a plan that includes a specific learning schedule and methodology for achieving the user's learning goals.

[0013] "Progress management" refers to the process of checking and managing how far a user is progressing in their learning according to their learning plan.

[0014] A "learning support system" refers to a system that includes generative models, functions for generating custom learning content based on students' learning progress and interests, and functions for planning and managing learning progress.

[0015] "Individualized instruction support" refers to personalized educational support provided by the system to individual users.

[0016] An "online learning platform" refers to an environment or system that provides learning content and support via the internet. [Brief explanation of the drawing]

[0017] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11]It is a sequence diagram showing the processing flow of the data processing system in Embodiment 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Embodiment 2 when the emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when the emotion engine is combined.

Mode for Carrying Out the Invention

[0018] Hereinafter, an example of an embodiment of the system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0019] First, the language used in the following description will be explained.

[0020] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0021] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0022] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0023] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0024] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0025] [First Embodiment]

[0026] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0027] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0028] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0029] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0030] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0031] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0032] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0033] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0035] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0036] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0037] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0038] The specific embodiments of the present invention are described below. This system is a learning support system that includes a generative model, a function for generating custom learning content based on students' learning progress and interests, and a function for planning and managing the progress of learning plans. This system operates with three main parties: the server, the terminal, and the user.

[0039] Overall system flow

[0040] server

[0041] The server implements the generative model, which is the core of this system. The server first loads the generative model, configures various settings, and prepares to receive requests. Based on requests sent from users and terminals, the server uses the generative model to generate responses and customized learning materials, and sends the results to the terminals. The server also maintains information on each student's learning progress and handles the planning and management of learning plans.

[0042] terminal

[0043] The terminal provides the user interface and is responsible for sending user input to the server. When a user logs in and submits questions or requests, the terminal displays responses and generated content from the server that received them. The terminal provides an appropriate interface to make it easier for the user to use.

[0044] User

[0045] Users are primarily students and teachers who input questions and requests into the system. They also request the creation of learning plans and monitor their progress as they learn.

[0046] Program processing

[0047] Loading and initializing the generative model

[0048] The server loads and initializes the generated model upon startup. It reads the model's parameters and configuration files, preparing it to function correctly. It starts a listener to await requests and waits for connections from clients.

[0049] User authentication and login

[0050] The device displays a login screen, and the user enters their authentication information (username and password). The device sends this information to the server, which then authenticates the user by comparing it against the authentication database. If authentication is successful, a token is generated and sent to the device. If authentication fails, an error message is returned.

[0051] Use as a supplementary tool for individualized instruction

[0052] The user inputs a question about a mathematical problem, and the terminal sends it to the server. The server uses a generative model to generate a response to the question and sends it back to the terminal. The terminal then displays the received response to the user.

[0053] Generating custom learning content

[0054] The user requests the generation of learning content based on a specific theme or difficulty level. The device sends this request to the server. The server uses a generative model to generate custom learning content and sends it to the device. This content is displayed on the device, and the user uses it for their learning.

[0055] Planning and managing study progress

[0056] The user submits a request for a learning plan and sends it to the server via their device. The server generates a learning plan using generative models and other algorithms based on the user's learning history and progress information, and sends it to the device. The user proceeds with their learning based on this plan and reports their progress to the server via their device. The server manages the progress and adjusts or improves the learning plan as needed.

[0057] Specific example

[0058] 1. When a user asks a question about a mathematical formula

[0059] The user enters "Please tell me the quadratic formula" into the question input screen on their device and submits it.

[0060] The device sends this question to the server.

[0061] The server uses a generative model to generate an explanation of the quadratic formula and produces a response such as, "The quadratic formula for solving the quadratic equation ax^2 + bx + c = 0 is..."

[0062] The device displays this response to the user, who then uses the explanation for learning.

[0063] 2. When a user requests custom learning content

[0064] The user enters a request saying, "Please create a workbook of problems on the basics of vectors."

[0065] The device sends this request to the server.

[0066] The server uses a generative model to generate a set of problems on the fundamentals of vectors and sends them to the terminal.

[0067] The user uses the problem set to progress with their studies.

[0068] 3. When the user creates a learning plan

[0069] The user requests, "Please create a study plan for the next math exam," in order to develop a study plan for the next exam.

[0070] The device sends this request to the server.

[0071] The server uses a generative model to generate an appropriate learning plan based on the user's learning history and progress data, and sends it to the terminal.

[0072] The device displays a learning plan to the user, and the user proceeds with their studies according to that plan.

[0073] As described above, the present invention provides a system that provides efficient and personalized learning support by using a generative model. This system allows users to receive high-quality learning support and improve their learning outcomes.

[0074] The following describes the processing flow.

[0075] Step 1: Load and initialize the generative model

[0076] The server loads the generated model upon startup, reads the necessary parameters and configuration files, and initializes it. It allocates resources for the model to function correctly and starts a listener to receive requests.

[0077] Step 2: User Authentication and Login

[0078] The terminal displays a login screen to the user. The user enters their authentication information (username, password) and presses the submit button. The terminal sends this authentication information to the server. The server compares the received authentication information with its database, and if authentication is successful, generates a JWT token and sends it back to the terminal. If authentication fails, it sends an appropriate error message back to the terminal.

[0079] Step 3: Enter Questions

[0080] The user enters text about their question or problem into the question input screen on the device and presses the submit button. The device then sends the entered question to the server.

[0081] Step 4: Question analysis and response generation

[0082] The server analyzes the received question and provides input to the generative model. The generative model generates a response to the question and returns the result to the server.

[0083] Step 5: Send the generated response

[0084] The server sends the response returned from the generative model to the terminal. The terminal displays the received response to the user. The user reviews the displayed response and enters additional questions as needed.

[0085] Step 6: Request custom learning content

[0086] The user wishes to generate custom learning content based on a specific learning theme or difficulty level, and enters a request into their device. The device then sends this request to the server.

[0087] Step 7: Generating Custom Content

[0088] The server receives a custom learning content generation request and uses a generative model to generate learning content based on the request. It then constructs a response containing the generated content and sends it to the terminal.

[0089] Step 8: Displaying the generated content

[0090] The device receives the generated learning content and displays it to the user. The user then uses the displayed content to continue learning.

[0091] Step 9: Develop a study plan

[0092] The user requests assistance with exam preparation or the creation of a study plan for specific learning objectives. The device sends this request to the server.

[0093] Step 10: Generate and submit your study plan.

[0094] The server generates an appropriate learning plan using generative models and other algorithms based on the user's learning history and progress information. It then sends the generated learning plan to the user's device.

[0095] Step 11: Display and implement your study plan

[0096] The device receives the generated learning plan and displays it to the user. The user proceeds with their studies based on the displayed learning plan and inputs their progress into the device as needed.

[0097] Step 12: Progress Management and Feedback

[0098] The server receives progress information sent by the user and manages the progress of the learning plan. If necessary, it generates feedback using a generative model and sends it to the terminal. The terminal displays the feedback to the user and helps them with their next learning session.

[0099] (Example 1)

[0100] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0101] Conventional learning support systems struggle to individually address the diverse learning needs of users. Furthermore, they fail to effectively manage learning progress, creating a need for efficient means to support planned learning. Additionally, the security of user authentication and the automation of on-demand learning content generation using generative models are insufficient, posing challenges to improving the quality and efficiency of learning.

[0102] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0103] In this invention, the server includes means for generating a response to user input using a generative model, means for the learning support server to read the generative model, perform various settings, and await requests, means for receiving user authentication information and issuing authentication tokens, means for generating custom learning content using the generative model in response to a learning content generation request from the terminal and sending it to the terminal, and means for managing student learning progress information, formulating a learning plan based on that information, and managing progress. This enables responses to individual learning needs, resulting in planned and effective learning support, secure user authentication, and on-demand generation and delivery of learning content.

[0104] A "generative model" is a system that generates text or responses using neural networks or machine learning algorithms.

[0105] "Responses to user input" refer to text-based answers and explanations generated based on questions and requests received from the user.

[0106] A "learning support server" is the core of a learning support system, handling tasks such as loading generative models, configuring various settings, processing requests, generating and sending learning content, and managing learning progress.

[0107] "Means for waiting for requests" refers to the function that allows a server to accept connections and data transmissions from users and terminals.

[0108] "User authentication information" refers to data such as usernames and passwords that are used to identify a user and grant them permissions.

[0109] An "authentication token" is a unique identifier issued to a user who has successfully authenticated, and is used to authenticate the user in subsequent communications.

[0110] A "learning content generation request" refers to a user requesting the creation of new learning materials or problem sets based on a specific theme or difficulty level.

[0111] "Custom learning content" refers to original learning materials and workbooks that are generated based on the user's requests and learning progress.

[0112] "Learning progress information" refers to data that records how far students have progressed in their studies.

[0113] A "study plan" is a document that outlines the learning content and progress schedule that students aim to achieve.

[0114] "Means of managing progress" refers to a function that tracks students' learning progress based on their learning plan and makes adjustments or improvements as needed.

[0115] "On-demand" refers to a service delivery method that provides services immediately in response to user requests.

[0116] This invention relates to a learning support system that includes a generative model, a learning support server, a function for generating custom learning content based on students' learning progress and interests, and a function for planning and managing learning plans. This invention is a system that operates primarily with three parties: the server, the terminal, and the user.

[0117] Server configuration and operation

[0118] The server implements the generative model (e.g., GPT-3® or BERT) that forms the core of this system. Upon startup, the server loads and initializes the generative model. During this process, it reads the model's parameters and configuration files (e.g., model_config.json and model_weights.h5) to prepare for correct operation. It also starts a listener to await requests and waits for connections from clients (terminals).

[0119] User Authentication

[0120] The server receives user authentication information (username, password) sent from the terminal and compares it with the authentication database. If authentication is successful, an authentication token is generated and sent to the terminal. If authentication fails, an error message is returned.

[0121] Learning content generation

[0122] The server responds to learning content generation requests from the terminal and generates custom learning content using a generative model. This content generation is based on user requests and individual learning progress information. The generated content is sent to the terminal and made available to the user.

[0123] Learning progress management and planning

[0124] The server manages each student's learning progress information and creates a learning plan based on it. In this process, generative models and other algorithms are used to analyze the user's learning history and progress data to generate an appropriate learning plan. The learning plan is sent to the user's device, where they can review and execute it.

[0125] Terminal configuration and operation

[0126] The terminal provides the user interface and is responsible for sending user input to the server. When a user logs in and enters questions or requests, the terminal displays the responses and generated content from the server that received them. The terminal provides a user-friendly interface, supporting smooth learning.

[0127] User actions

[0128] Users are primarily students and teachers, and they can input questions and requests into the system. For example, consider the following specific examples:

[0129] Questions about mathematical formulas

[0130] The user types "Please tell me the quadratic formula" into the terminal and sends it. The terminal sends this question to the server, which uses a generative model to generate a response and sends it back to the terminal. The terminal displays the generated response to the user.

[0131] Request for custom learning content

[0132] The user enters a request: "Please create a set of problems on the basics of vectors." The device sends this request to the server, which uses a generative model to generate custom learning content and sends it to the device. The user then uses the generated set of problems for learning.

[0133] Planning a study schedule

[0134] The user requests, "Please create a study plan for the next math exam." The device sends this request to the server, which uses a generative model to generate a study plan based on the user's learning history and progress data. The device displays this plan to the user, who then proceeds with their studies based on the plan.

[0135] Thus, the present invention uses a generative model to provide high-quality learning support and realize efficient and personalized learning.

[0136] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0137] Step 1:

[0138] The server starts up and loads the generative model. During this process, it reads the configuration file (e.g., model_config.json) and the model parameter file (e.g., model_weights.h5) and initializes the generative model. The server starts a listener to await requests and waits for connections and requests from clients. This process takes the server startup command and configuration files as input and outputs the state where the generative model initialization is complete.

[0139] Step 2:

[0140] The user accesses the login screen on their device and enters their username and password. The device sends this information to the server. The server compares it with the authentication database, and if authentication is successful, generates a token and sends it to the device. If authentication fails, it returns an error message. The input to this process is the user's authentication information, and the output is either an authentication token or an error message.

[0141] In terms of specific operations, the terminal receives user input and sends it to the server. On the server side, for example, an SQL query is executed to verify the user information (e.g., SELECT FROM users WHERE username = ? AND password_hash = ?).

[0142] Step 3:

[0143] The user enters "Please tell me the quadratic formula" on the terminal's question input screen and submits it. The terminal sends this question to the server. The server inputs the question as a prompt to the generative model, and the generative model generates a response. This generated response is sent to the terminal and displayed to the user. The input to this process is the user's question, and the output is the response generated by the generative model.

[0144] In terms of specific operation, the server inputs the response to the prompt "Please tell me the quadratic formula" into the generative model, and then sends the resulting response to the terminal.

[0145] Step 4:

[0146] A user requests and sends a request from their device saying, "Please create a set of problems on the basics of vectors." The device sends this request to the server. The server uses a generative model to generate custom learning content based on the theme. This generated content is sent to the device and made available to the user. The input to this process is the user's request to generate custom learning content, and the output is the generated custom learning content.

[0147] In terms of specific operation, the server inputs a request based on the prompt "Create a problem set on the basics of vectors" into the generative model, and then sends the resulting problem set to the terminal.

[0148] Step 5:

[0149] A user requests from their device, "Please create a study plan for the next math exam." The device sends this request to the server. The server retrieves the user's learning history and progress data from a database and generates a study plan using generative models and other algorithms. This study plan is sent to the device and displayed to the user. The input to this process is the user's study plan creation request and learning history data, and the output is the generated study plan.

[0150] In terms of specific operations, the server retrieves user progress data from the database, inputs it into a generative model to generate an appropriate learning plan, and then sends it to the terminal.

[0151] Step 6:

[0152] The user progresses through their learning based on a learning plan and reports their progress to the server from their terminal. The server stores this progress information in a database and adjusts or improves the learning plan as needed. The input to this process is the user's learning progress information, and the output is an updated learning plan and progress data.

[0153] In terms of specific operations, the user progresses through the learning plan, reports their progress via their device, the server records this in a database, and the learning plan is reviewed.

[0154] (Application Example 1)

[0155] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0156] Online education platforms require personalized learning support and an environment that enables users to learn efficiently. Furthermore, generating custom learning content in real time based on user progress and developing and managing learning plans based on individual needs are also crucial challenges. Current systems struggle to comprehensively provide these elements, resulting in insufficient support for maximizing learning effectiveness.

[0157] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0158] In this invention, the server includes means for generating responses to user input using a generative model, means for generating custom learning content based on the user's learning progress and interests, means for formulating and managing the user's learning plan, means for providing an individualized learning plan, and means for delivering the generated learning content to the user's terminal. As a result, the user can receive individualized learning support in real time and learn efficiently.

[0159] A "generative model" is a machine learning algorithm that automatically generates appropriate responses or content based on user input.

[0160] "User learning progress" refers to information that indicates the progress and level of proficiency achieved by learners in accordance with their learning plan.

[0161] "Custom learning content" refers to learning materials that are individually generated based on the needs, interests, and progress of a specific user.

[0162] A "user learning plan" refers to the specific steps and schedule that a user will use to achieve their learning goals.

[0163] A "personalized learning plan" refers to a learning schedule optimized for each user's individual learning history, progress, and goals.

[0164] A "user terminal" is an electronic device used by a user to communicate with a server and receive learning content.

[0165] "Real-time" means responding to or processing user requests immediately.

[0166] An "online education platform" is a system that provides learning content and educational services via the internet.

[0167] A "learning support system" refers to a set of computer programs and hardware that provide functions to effectively support a user's learning.

[0168] The specific embodiments of this invention are described below. This invention is a learning support system that utilizes a generative model and operates with three main parties: a server, a terminal, and a user. Various hardware and software and their processing are described below.

[0169] server

[0170] The server implements the generative model, which forms the core of this system. The server loads and initializes the generative model upon startup. The server is implemented using a server-side framework such as Flask and starts a listener to respond to user requests. The server also manages user authentication information using JWT (JSON Web Token). Specifically, a server machine with a high-performance processor and memory is used.

[0171] Generative model

[0172] A generative model is a machine learning algorithm that generates responses or custom learning content based on user requests. For example, it uses generative AI models such as OpenAI® GPT-4®. The model is loaded when the server starts up, and when it receives a request, it generates text based on its content.

[0173] terminal

[0174] The terminal provides the user interface and transmits user input to the server. Users log in using electronic devices such as smartphones, tablets, or PCs and enter questions or requests. The entered information is sent to the server, and the server's response and generated learning content are displayed on the terminal. The terminal provides a user-friendly interface to ensure easy operation for the user.

[0175] User

[0176] Users are primarily students and educators who input questions and requests into the system. Users can request the creation of learning plans and progress through their studies while checking progress information.

[0177] Data processing

[0178] Authentication: When a user logs in, the device sends authentication information to the server. The server compares this information with the authentication database, and if authentication is successful, generates a JWT token and sends it to the device.

[0179] Response generation: When a user enters a question about a mathematical problem into the terminal, the server uses a generative model to generate a response to the question and sends it to the terminal.

[0180] Generating custom learning content: When a user requests learning content based on a specific theme or difficulty level, the server uses a generative model to generate custom learning content and sends it to the device.

[0181] Learning plan creation: When a user requests the generation of a learning plan, the server generates a learning plan using generative models and other algorithms based on the user's learning history and progress information, and sends it to the terminal.

[0182] Specific example

[0183] 1. If a user wants to learn the basics of Python programming

[0184] The user types "I want to learn the basics of Python programming" into their terminal and sends a request.

[0185] The server uses a generative model to generate custom learning content on the fundamentals of Python programming and sends it to the device.

[0186] The device displays the generated learning content to the user, and the user begins learning.

[0187] 2. When a user plans their studies for the next math exam.

[0188] The user enters the request, "Please create a study plan for my next math exam."

[0189] The server generates an appropriate learning plan using a generative model based on the user's learning history and progress data, and sends it to the terminal.

[0190] The device displays the generated learning plan to the user, and the user proceeds with learning according to that plan.

[0191] Example of a prompt

[0192] Custom learning content generation:

[0193] "Generate learning content on the fundamentals of Python programming. Includes examples of videos, quizzes, and explanatory articles."

[0194] Learning plan generation:

[0195] "Create a study plan for the next math exam. Include daily tasks and key points based on the user's current progress and past learning history."

[0196] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0197] Step 1:

[0198] The server loads and initializes the generative model upon startup. The input is the model's configuration file and parameters, and the output is the initialized generative model. Specifically, the server reads the model file from the specified directory, sets the necessary parameters, and prepares the generative model.

[0199] Step 2:

[0200] The terminal displays a login screen, and the user enters their authentication information (username, password). The input is the username and password, and the output is a request containing the authentication information. The terminal sends this authentication information to the server.

[0201] Step 3:

[0202] The server compares the received authentication information with the authentication database, and if authentication is successful, generates a JWT token and sends it to the terminal. The input is the authentication information, and the output is either a JWT token or an error message. Specifically, the server compares it with a hashed password, and if authentication is successful, generates a token.

[0203] Step 4:

[0204] The user enters and submits questions and requests for learning content generation on their device. The input is the user's request, and the output is the request data. Specifically, the device sends data to the server using the submit button in the user's input field.

[0205] Step 5:

[0206] The server uses a generative model to generate responses or custom training content based on the received request. The input is the request data, and the output is the generated response or content. Specifically, the server inputs a prompt into the generative model and retrieves the generated result from the model.

[0207] Step 6:

[0208] The server sends the generated response or content to the terminal. The input is the generated response or content, and the output is the data sent to the terminal. Specifically, the server sends data to the terminal as an HTTP response.

[0209] Step 7:

[0210] The terminal displays received responses and content to the user. Input is data sent from the server, and output is information displayed on the user interface. Specifically, the terminal uses HTML and GUI components to display the data appropriately.

[0211] Step 8:

[0212] The user progresses through the learning process using the displayed responses and content, and reports progress information from the device to the server. The input is the user's learning progress information, and the output is the progress data sent to the server. Specifically, the device provides a field for entering progress information, and the user sends the data to the server using a submit button.

[0213] Step 9:

[0214] The server adjusts the learning plan based on the received progress information and generates a new learning plan as needed. The input is the user's progress data, and the output is a new or adjusted learning plan. Specifically, the server updates the database and generates or adjusts the plan using a generative model.

[0215] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0216] The specific embodiments of the present invention are described below. This system is a learning support system that includes a generative model, a function for generating custom learning content based on students' learning progress and interests, a function for planning and managing learning plans, and an emotion engine that recognizes user emotions and adjusts responses. This system operates with three main parties: the server, the terminal, and the user.

[0217] Overall system flow

[0218] server

[0219] The server implements the generative model and sentiment engine, which are the core of this system. The server first loads the generative model, configures various settings, and prepares to accept requests. Based on the user input and sentiment data included in the request, it uses the generative model to generate responses and customized learning materials, and sends the results to the terminal. The server also maintains each student's learning progress information and handles the planning and management of learning plans.

[0220] terminal

[0221] The terminal provides the user interface and is responsible for sending user input (text and sentiment data) to the server. When a user logs in and submits a question or request, the terminal displays the response from the server, generated content, and sentiment-based adjustments. The terminal provides an appropriate interface to make it easier for the user to use.

[0222] User

[0223] Users are primarily students and teachers who input questions and requests into the system. They also request the creation of learning plans and monitor their progress as they learn. Furthermore, the user's emotions are recognized through the terminal's input interface and transmitted to the server.

[0224] Program processing

[0225] Loading and initializing the generative model

[0226] Upon startup, the server loads the generative model and sentiment engine, and initializes by reading the necessary parameters and configuration files. It allocates resources for the model to function correctly and starts a listener to receive requests.

[0227] User authentication and login

[0228] The terminal displays a login screen to the user. The user enters their authentication information (username, password) and presses the submit button. The terminal sends this authentication information to the server. The server compares the received authentication information with its database, and if authentication is successful, generates a JWT token and sends it back to the terminal. If authentication fails, it sends an appropriate error message back to the terminal.

[0229] Question entry and submission of sentiment data

[0230] The user enters text about their question or problem into the terminal's question input screen and presses the submit button. The terminal sends the entered question, along with emotional data obtained from the user's facial recognition and voice tone analysis, to the server.

[0231] Question analysis and response generation

[0232] The server analyzes the received question and provides input to the generative model, taking into account the sentiment data recognized by the sentiment engine. The generative model generates a response based on the question and sentiment data and returns the result to the server.

[0233] Sending the generated response

[0234] The server sends the response returned from the generative model to the terminal after making appropriate adjustments based on the user's sentiment. The terminal displays the received response to the user. The user reviews the displayed response and enters additional questions if necessary.

[0235] Request for custom learning content

[0236] The user requests the generation of custom learning content based on a specific learning theme or difficulty level, and enters this request into their device. Along with this request, the device sends the user's sentiment data to the server.

[0237] Generating custom content

[0238] The server receives a custom learning content generation request and uses a generative model to generate learning content based on the request. It constructs a response containing the generated content, adjusts the difficulty and tone based on the user's sentiment data, and sends it to the device.

[0239] Displaying generated content

[0240] The device receives the generated learning content and displays it to the user. The user then uses the displayed content to continue learning.

[0241] Planning a study schedule

[0242] The user requests assistance with exam preparation or the creation of a study plan for specific learning objectives. The device sends this request to the server.

[0243] Generating and submitting a study plan

[0244] The server generates an appropriate learning plan using a generative model and sentiment engine based on the user's learning history and progress information. The generated learning plan is then sent to the device.

[0245] Displaying and implementing the study plan

[0246] The device receives the generated learning plan and displays it to the user. The user proceeds with their studies based on the displayed learning plan and inputs their progress into the device as needed.

[0247] Progress management and feedback

[0248] The server receives progress information and sentiment data sent by the user and manages the progress of the learning plan. If necessary, it generates feedback using a generative model and sends it to the device. The device displays the feedback to the user, who can then use it to improve their next learning session.

[0249] Specific example

[0250] 1. When using sentiment data when users ask questions about mathematical formulas.

[0251] The user enters "Please tell me the quadratic formula" into the question input screen on their device and submits it.

[0252] Along with this question, the device sends the user's facial expression recognition results (e.g., confused expression) and voice tone analysis results to the server.

[0253] The server uses a generative model to generate responses to questions, taking into account feelings of confusion and producing responses such as "You should explain with more specific examples."

[0254] The device displays this response to the user, who then uses the explanation for learning.

[0255] 2. When sentiment data is used when users request custom learning content.

[0256] The user requests, "Please create a workbook of problems on the basics of vectors."

[0257] Along with this request, the device sends the user's sentiment data to the server.

[0258] The server uses a generative model to generate a set of problems on the fundamentals of vectors, and adjusts the difficulty level if the user's motivation to learn is high.

[0259] The device displays the generated problem sets to the user, who then uses them to progress with their studies.

[0260] 3. When users utilize sentiment data when developing learning plans.

[0261] The user requests, "Please create a study plan for the next math exam," in order to develop a study plan for the next exam.

[0262] Along with this request, the device sends the user's sentiment data to the server.

[0263] The server uses generative models and a sentiment engine to generate an appropriate learning plan based on the user's learning history, progress data, and sentiment data.

[0264] For example, if a user is experiencing stress, the server will adjust its schedule accordingly.

[0265] The device displays the generated learning plan to the user, and the user proceeds with learning according to that plan.

[0266] This invention provides a system that uses a generative model and an emotion engine to provide personalized learning support that takes into account the user's emotions. This system allows users to learn more effectively and with less stress.

[0267] The following describes the processing flow.

[0268] Step 1: Load and initialize the generative model

[0269] Upon startup, the server loads the generative model and sentiment engine, and initializes by reading the necessary parameters and configuration files. It allocates resources for the model to function correctly and starts a listener to receive requests.

[0270] Step 2: User Authentication and Login

[0271] The terminal displays a login screen to the user. The user enters their authentication information (username, password) and presses the submit button. The terminal sends this authentication information to the server. The server compares the received authentication information with its database, and if authentication is successful, generates a JWT token and sends it back to the terminal. If authentication fails, it sends an appropriate error message back to the terminal.

[0272] Step 3: Enter the questions and submit sentiment data.

[0273] The user enters text about their question or problem into the terminal's question input screen and presses the submit button. The terminal sends the entered question, along with emotional data obtained from the user's facial recognition and voice tone analysis, to the server.

[0274] Step 4: Question analysis and response generation

[0275] The server analyzes the received question and provides an input to the generation model considering the sentiment data recognized by the sentiment engine. The generation model generates a response based on the question and the sentiment data and returns the result to the server.

[0276] Step 5: Sending the Generated Response

[0277] The server sends the response returned from the generation model to the terminal after making appropriate adjustments based on the user's sentiment. The terminal displays the received response to the user. The user checks the displayed response and enters additional questions if necessary.

[0278] Step 6: Request for Custom Learning Content

[0279] The user wishes to generate custom learning content based on a specific learning theme or difficulty level and enters a request into the terminal. The terminal sends this request along with the user's sentiment data to the server.

[0280] Step 7: Generation of Custom Content

[0281] The server receives the custom learning content generation request, uses the generation model to generate learning content based on the request, constructs a response containing the generated content, adjusts the difficulty level and tone by referring to the user's sentiment data, and sends it to the terminal.

[0282] Step 8: Display of Generated Content

[0283] The terminal receives the generated learning content and displays it to the user. The user proceeds with learning using the displayed content.

[0284] Step 9: Formulation of a Learning Plan

[0285] The user requests the formulation of a learning plan for exam preparation or specific learning goals. The terminal sends this request to the server.

[0286] Step 10: Generation and Transmission of Learning Plan

[0287] Based on the user's learning history and progress information, the server uses the generation model and emotion engine to generate an appropriate learning plan. The generated learning plan is transmitted to the terminal.

[0288] Step 11: Display and Practice of Learning Plan

[0289] The terminal receives the generated learning plan and displays it to the user. The user proceeds with learning based on the displayed learning plan and inputs the progress status into the terminal as appropriate.

[0290] Step 12: Progress Management and Feedback

[0291] The server receives the progress information and emotion data transmitted from the user, and manages the progress of the learning plan. If necessary, feedback is generated using the generation model and transmitted to the terminal. The terminal displays the feedback to the user and uses it for the next learning.

[0292] Specific Example

[0293] 1. When the user asks a question about a math formula

[0294] The user inputs "Please teach me the quadratic formula for the solutions of a quadratic equation" on the question input screen of the terminal and sends it.

[0295] The terminal sends this question along with the user's facial expression recognition result (e.g., a confused expression) and voice tone analysis result to the server.

[0296] The server uses the generation model to generate a response to the question, and generates a response such as "You should give more specific examples to explain" considering the confused emotion.

[0297] The terminal displays this response to the user, and the user uses the explanation for learning.

[0298] 2. When the user requests custom learning content

[0299] The user requests, "Please create a problem set on the basics of vectors."

[0300] The terminal sends the user's sentiment data to the server along with this request.

[0301] The server uses the generation model to generate a problem set on the basics of vectors and makes adjustments such as increasing the difficulty level if the user's learning motivation is high.

[0302] The terminal displays the generated problem set to the user, and the user proceeds with learning using it.

[0303] 3. When the user formulates a learning plan

[0304] The user requests, "Please create a learning plan required until the next math exam." in order to formulate a learning plan for the next exam.

[0305] The terminal sends the user's sentiment data to the server along with this request.

[0306] The server uses the generation model and sentiment engine to generate an appropriate learning plan based on the user's learning history, progress data, and sentiment data.

[0307] For example, if the user is feeling stressed, the server adjusts the schedule gently.

[0308] The terminal displays the generated learning plan to the user, and the user proceeds with learning according to the plan.

[0309] (Example 2)

[0310] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0311] Traditional learning support systems provided features such as generating custom learning content based on individual users' learning progress and interests, developing learning plans, and managing progress. However, they lacked the ability to adjust responses based on user emotions. As a result, it was difficult to provide optimal learning support tailored to the user's emotional state, leading to challenges in maintaining learning effectiveness and motivation.

[0312] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0313] In this invention, the server includes means for generating responses to user input using a generative model, means for generating custom learning content based on the student's learning progress and interests, means for formulating a user's learning plan and managing its progress, means for recognizing the user's emotional data and adjusting the response, means for performing user authentication and matching authentication information with a database, means for adjusting the generated response based on the user's emotions and transmitting it to the terminal, and means for generating a learning plan and managing its progress based on the user's progress information and emotional data. This enables personalized learning support that takes the user's emotions into consideration.

[0314] A "generative model" is a pre-trained algorithm that generates natural language responses or content based on input data.

[0315] A "user" is a person, such as a student or teacher, who uses the system and is the entity that makes questions or learning requests through the learning support system.

[0316] "Custom learning content" refers to educational content such as learning materials and workbooks that are created individually based on a student's learning progress and interests.

[0317] A "learning plan" is a schedule or plan built based on the learning goals set by the user, and it also includes the management of progress.

[0318] "Emotional data" refers to data that indicates the user's current emotional state, obtained from user facial recognition and voice tone analysis.

[0319] A "response" refers to the natural language response or explanation that a generative model outputs in response to a user's question or request.

[0320] "User authentication" is the process of verifying the identity of users accessing a system, and it uses authentication information such as usernames and passwords.

[0321] A "terminal" is a device used by a user to interact with a system, and includes personal computers, tablets, smartphones, and other similar devices.

[0322] "Progress information" refers to data that shows the user's progress in their learning, and is recorded and managed through an online system.

[0323] A "database" is a digital repository for storing user authentication information, learning progress information, generated learning content, and other similar data.

[0324] A "listener" refers to the state in which a server is waiting for requests from external sources and is the process of accepting user input.

[0325] Specific embodiments of the present invention are described below. This learning support system is broadly composed of three components: a server, a terminal, and a user. The role of each component will be described in detail below.

[0326] Server configuration and operation

[0327] The server plays a central role in this system. It implements the following main functions:

[0328] 1. Loading and initializing the generative model:

[0329] When the server starts up, it loads pre-trained generative models and sentiment engines. This uses libraries such as Python and Tensorflow®.

[0330] It reads the necessary parameters and configuration files, allocates resources, and starts a listener to wait for requests.

[0331] 2. User Authentication:

[0332] The server receives user authentication information (username, password) sent from the terminal and compares it with the database.

[0333] If authentication is successful, a JWT token is generated and sent back to the device; if authentication fails, an error message is sent back.

[0334] 3. Question analysis and response generation:

[0335] The server receives the question text and sentiment data sent by the user.

[0336] The question text and sentiment data are given to a generative model as input, and an appropriate response is generated. This response is adjusted considering the question content and sentiment data.

[0337] 4. Generating custom learning content:

[0338] The server generates custom learning content based on user requests and sentiment data.

[0339] The generated content is adjusted in difficulty and tone based on the user's learning motivation and emotional state.

[0340] 5. Planning and managing study progress:

[0341] The server generates an appropriate learning plan based on the user's learning history, progress data, and sentiment data.

[0342] The generated learning plan is sent to the device, and feedback and adjustments are made based on the user's progress.

[0343] Terminal configuration and operation

[0344] A terminal is a device that provides a user interface and performs the following functions:

[0345] 1. Displaying the login screen and sending user authentication information:

[0346] The device displays the login screen and sends the authentication information entered by the user to the server.

[0347] 2. Question entry and submission of sentiment data:

[0348] The user enters text into the question input screen on the device and presses the submit button. At the same time, emotional data is sent to the server based on facial recognition and voice tone analysis.

[0349] 3. Display of response from the server:

[0350] The terminal receives a response from the server and displays it to the user. The user reviews the displayed response and enters additional questions as needed.

[0351] 4. Displaying custom learning content:

[0352] The device displays custom learning content received from the server, enabling users to efficiently progress in their learning.

[0353] 5. Displaying and implementing the study plan:

[0354] The device receives the learning plan sent from the server and displays it to the user. The user then proceeds with their learning based on this learning plan.

[0355] User roles and actions

[0356] The user is the subject who uses the system to progress through the learning process, and takes the following actions:

[0357] 1. Enter your question or request:

[0358] Users input learning-related questions, requests for custom learning content, and learning plans into the system.

[0359] 2. Implementing the study plan and inputting progress:

[0360] The user progresses through the learning process based on the provided learning plan and enters progress information into their device.

[0361] 3. Use of learning content:

[0362] Students use the generated custom learning content to progress through their studies.

[0363] Specific example

[0364] If a user asks a question about a mathematical formula:

[0365] The user types the question "Please tell me the quadratic formula" into their device and sends it.

[0366] The facial recognition results (confused expression) and voice tone analysis results are also sent to the server at the same time.

[0367] The server uses a generative model to generate responses to questions, taking into account feelings of confusion and producing responses such as "You should explain with specific examples."

[0368] The terminal displays this response to the user.

[0369] When a user requests custom learning content:

[0370] I requested that you create a workbook of problems on the fundamentals of vectors.

[0371] Along with this request, the device sends the user's sentiment data to the server.

[0372] The server uses a generative model to generate a set of problems based on the request, and increases the difficulty level if the user's motivation is high.

[0373] The terminal displays the generated problem set to the user.

[0374] When a user creates a learning plan:

[0375] "Please create a study plan that will be necessary before the next math exam," I requested.

[0376] Along with this request, the device sends the user's sentiment data to the server.

[0377] The server uses generative models and an emotion engine to generate an appropriate learning plan based on the user's learning history, progress data, and sentiment data.

[0378] For example, if a user is experiencing stress, the server will adjust its schedule accordingly.

[0379] The device displays the generated learning plan to the user, and the user proceeds with learning according to that plan.

[0380] Example of a prompt:

[0381] "Please create a study plan for math and English for the next semester."

[0382] "If the user is confused by the tone of voice, please generate an easy-to-understand explanation."

[0383] "Please generate a set of problems on the fundamentals of vectors and set the difficulty level to intermediate."

[0384] In this way, a personalized learning support system using generative models and emotion engines can be realized.

[0385] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0386] Step 1: Load and initialize the generative model

[0387] When the server starts up, it loads a pre-trained generative model and sentiment engine. Specifically, it creates instances of the generative model using libraries such as Python and TensorFlow, and then loads the pre-trained model.

[0388] Input: Generative models and emotion engines, and their configuration files.

[0389] Data processing and calculations: Model loading and resource allocation.

[0390] Output: Start of listener to accept requests.

[0391] Step 2: User Authentication

[0392] The device displays a login screen, and the user enters their username and password and presses the submit button.

[0393] Enter: Username and password.

[0394] Data processing and calculations: Send authentication information to the server.

[0395] The server then compares this authentication information with the database.

[0396] Input: Authentication information.

[0397] Data processing and calculations: Database matching.

[0398] Output: If authentication is successful, a JWT token is generated and sent back to the device. If authentication fails, an error message is returned.

[0399] Step 3: Enter the questions and submit sentiment data.

[0400] The user enters text about their question or problem into the question input screen on their device and presses the submit button (e.g., "Please tell me the quadratic formula").

[0401] Input: Question text.

[0402] Data processing and computation: facial expression recognition and voice tone analysis.

[0403] The terminal sends the entered question along with emotional data obtained from facial recognition results and voice tone analysis to the server.

[0404] Input: Question text and sentiment data.

[0405] Data processing and calculations: Data transmission.

[0406] Output: Sending data to the server.

[0407] Step 4: Question analysis and response generation

[0408] The server analyzes the received question text and sentiment data.

[0409] Input: Question text and sentiment data.

[0410] Data processing and computation: Input to generative models.

[0411] Generating response text that takes emotional data into account (e.g., recognizing the emotion of confusion generates a response that includes a detailed explanation).

[0412] Output: The generated response text.

[0413] Step 5: Send the generated response

[0414] The server sends the generated response to the terminal after appropriately adjusting its emotional tone.

[0415] Input: The generated response text.

[0416] Data processing and calculation: Adjustments based on emotional data.

[0417] Output: Sends the adjusted response text to the terminal.

[0418] Step 6: Display and confirm the response

[0419] The terminal displays the received response to the user.

[0420] Input: Adjusted response text.

[0421] Data processing and calculations: Display processing.

[0422] Output: Display the response to the user.

[0423] Step 7: Request custom learning content

[0424] Users can request the creation of custom learning content based on specific learning themes and difficulty levels (e.g., "Please create a set of problems on the basics of vectors").

[0425] Input: Request for custom learning content.

[0426] Data processing and computation: facial expression recognition and voice tone analysis.

[0427] The device sends emotion data to the server along with the request.

[0428] Input: Request and sentiment data.

[0429] Data processing and calculations: Data transmission.

[0430] Output: Sending requests and sentiment data to the server.

[0431] Step 8: Generating Custom Content

[0432] The server receives a custom training content generation request and uses a generative model to generate the training content.

[0433] Input: Request and sentiment data.

[0434] Data processing and computation: Generating custom content using generative models.

[0435] The difficulty level and tone are adjusted based on emotional data.

[0436] Output: Generated training content.

[0437] Step 9: Displaying the generated content

[0438] The device receives the generated learning content and displays it to the user.

[0439] Input: Generated learning content.

[0440] Data processing and calculations: Display processing.

[0441] Output: Display learning content to the user.

[0442] Step 10: Develop a study plan

[0443] The user requests the creation of a study plan (e.g., "Please create a study plan for the next math exam").

[0444] Input: Request for a study plan.

[0445] Data processing and computation: facial expression recognition and voice tone analysis.

[0446] The device sends emotion data to the server along with the request.

[0447] Input: Request and sentiment data.

[0448] Data processing and calculations: Data transmission.

[0449] Output: Sending a training plan request and sentiment data to the server.

[0450] Step 11: Generate and submit your study plan.

[0451] The server uses generative models and a sentiment engine to generate an appropriate learning plan based on the user's learning history, progress data, and sentiment data.

[0452] Inputs: Learning history, progress data, sentiment data.

[0453] Data processing and computation: Generating learning plans using generative models.

[0454] Adjusting schedules based on emotional data.

[0455] Output: The generated training plan.

[0456] Step 12: Display and implement your study plan

[0457] The device receives the generated learning plan and displays it to the user.

[0458] Input: The generated training plan.

[0459] Data processing and calculations: Display processing.

[0460] Output: Displays the learning plan to the user.

[0461] Step 13: Progress Management and Feedback

[0462] The user progresses through the learning process according to the learning plan and enters progress information into the device.

[0463] Input: Progress information.

[0464] Data processing and calculations: Data transmission.

[0465] The server receives progress information and sentiment data sent by the user and manages the progress of the learning plan.

[0466] Input: Progress information and sentiment data.

[0467] Data processing and computation: progress management and feedback generation.

[0468] The device displays feedback to the user, which can be used to improve future learning.

[0469] Output: Feedback to the user.

[0470] (Application Example 2)

[0471] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0472] Conventional learning support systems have difficulty generating customized learning content tailored to each student's individual learning progress and interests, and they also lack the ability to provide responses that take into account students' emotional states. As a result, it has been difficult for students to properly plan their studies, manage their progress, and learn effectively, leading to a decline in the quality of learning.

[0473] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0474] In this invention, the server includes means for generating responses to user input using a generative model, means for generating custom learning content based on the learner's learning progress and interests, means for formulating a user's learning plan and managing its progress, and means for recognizing the user's emotions and adjusting responses and learning content based on that data. This enables personalized learning support for each individual learner, taking into account their learning progress and emotions.

[0475] A "generative model" is an algorithm that generates appropriate responses or content based on user input.

[0476] "Customized learning content" refers to learning materials and exercises tailored to each learner's individual learning progress and interests.

[0477] A "learning plan" is a schedule and set of action guidelines created to help users effectively progress towards their learning goals.

[0478] "Progress management" is a method for checking the results and progress of learners' learning activities and making adjustments as needed.

[0479] An "emotion engine" is an algorithm that recognizes a user's emotions and adjusts responses and content based on that data.

[0480] "User" refers to individuals who use this system, such as learners and teachers.

[0481] A "server" is a computer system that handles the core functions of a system and implements generative models, emotion engines, and other related components.

[0482] A "terminal" is a device that provides a user interface and transmits user input to a server.

[0483] "Interface" refers to the user interface or input method that allows a user to interact with a system.

[0484] The learning support system of the present invention consists of a server, a terminal, and a user. This system includes a generative model, the generation of custom learning content based on learning progress and interests, the planning and progress management of learning plans, and an emotion engine.

[0485] The server is the core of the entire system, implementing the generative model and sentiment engine. The server first loads the generative model, configures various settings, and prepares to receive requests. Requests include user input and sentiment data, and based on this, the server uses the generative model to generate responses and customized learning materials. The generated results are sent to the terminal. The server also maintains each student's learning progress information and handles the planning and management of learning progress.

[0486] The terminal provides the user interface and transmits user input (text and sentiment data) to the server. Users log in to the terminal and enter questions or requests. Responses from the server, generated content, and sentiment-based adjustments are displayed on the terminal. The terminal provides an appropriate interface to make it easier for users to use.

[0487] Users are primarily learners and instructors who input questions and requests into the system. They also request the creation of learning plans and monitor their progress as they learn. Furthermore, the user's emotions are recognized through the terminal's input interface and transmitted to the server.

[0488] Program Processing Description

[0489] The server loads and initializes the generative model. When the server starts, it loads the generative model and sentiment engine, and initializes the necessary parameters and configuration files. This allocates resources for the model to function correctly and starts a listener to accept requests.

[0490] The terminal supports user authentication and login. The terminal displays a login screen to the user, who enters their authentication information (username, password). The authentication information is sent from the terminal to the server, which compares the received information with its database. If authentication is successful, a JWT token is generated and sent back to the terminal.

[0491] Furthermore, when a user enters a question or requests custom learning content, the device sends sentiment data along with the content to the server. The server analyzes the received questions and requests and provides input to the generative model based on the sentiment data recognized by the sentiment engine. The generative model generates responses and custom learning content based on the questions and sentiment data, and returns the results to the server. The server then sends this to the device for display to the user.

[0492] Hardware and software used

[0493] Server: A computer system that performs the core functions of a system (e.g., "Nginx", "Docker").

[0494] Generative models and sentiment engines: Algorithms and libraries (e.g., TensorFlow, PyTorch)

[0495] Device: A device that provides a user interface (e.g., "iOS device", "ANDROID® device")

[0496] Database: Stores user information and learning progress (e.g., "PostgreSQL").

[0497] Examples of specific cases and prompt statements

[0498] As a concrete example, when a user asks a question about the equations of motion in physics, a fearful expression is detected through facial recognition. This information is sent from the terminal to the server, which uses a generative model to provide a detailed and easy-to-understand answer.

[0499] Example prompt: "Please answer the following question considering the sentiment data: The user has requested an explanation of the equations of motion in physics. The sentiment data indicates 'fear'."

[0500] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0501] Step 1:

[0502] The server loads and initializes the generative model and sentiment engine. Specifically, at startup, the server loads the generative model and sentiment engine into memory and reads the necessary parameters and configuration files. This ensures that the generative model and sentiment engine have the resources to function correctly. It also starts a listener to receive requests. The input is the configuration files and parameters of the generative model and sentiment engine, and the output is the initialized generative model and sentiment engine.

[0503] Step 2:

[0504] The terminal receives the user's login information and sends it to the server. Specifically, the user enters their username and password on the terminal's login screen and presses the submit button. The terminal sends this authentication information to the server. The input is the user's login information (username, password), and the output is the authentication request sent to the server.

[0505] Step 3:

[0506] The server authenticates the login information. Specifically, the server compares the received authentication information with the database, and if successful, generates a JWT token and sends it back to the terminal. If it fails, it sends an appropriate error message back to the terminal. The input is the authentication information received from the terminal, and the output is a JWT token if authentication is successful, and an error message if it fails.

[0507] Step 4:

[0508] The device collects user questions and sentiment data and sends them to the server. Specifically, the user enters text about their questions or problems into the question input screen and presses the submit button. The device simultaneously sends sentiment data acquired from the camera and voice input to the server. The input consists of the user's question text and sentiment data, and the output is the request sent to the server.

[0509] Step 5:

[0510] The server receives questions and sentiment data, analyzes them, and generates responses. Specifically, the server analyzes the received questions and inputs the recognized sentiment data into a generative model. The generative model generates appropriate responses based on the questions and sentiment data. The input is the questions and sentiment data received from the terminal, and the output is the generated response data.

[0511] Step 6:

[0512] The server adjusts the generated response based on sentiment data and sends it to the terminal. Specifically, the server adjusts the tone and level of detail of the response returned from the generative model according to the user's emotions, and then sends it to the terminal. The input is the generated response data and sentiment data, and the output is the adjusted response.

[0513] Step 7:

[0514] The terminal displays the received response to the user. Specifically, the terminal displays the response sent from the server on the screen, allowing the user to confirm the response. The input is the adjusted response received from the server, and the output is the response displayed on the screen.

[0515] Step 8:

[0516] Users request custom learning content. Specifically, users input a request on their device screen to generate custom learning content based on a specific learning theme and difficulty level. The input is the user's request, and the output is the request sent from the device to the server.

[0517] Step 9:

[0518] The server generates custom learning content and adjusts it based on sentiment data. Specifically, the server uses a generative model to generate learning content based on requests and adjusts the difficulty and tone by referring to the user's sentiment data. The input is the user's request and sentiment data, and the output is the generated custom learning content.

[0519] Step 10:

[0520] The device displays the generated learning content to the user. Specifically, it receives the generated content from the server and displays it on the screen. The input is the generated learning content, and the output is the learning content displayed to the user.

[0521] Example of a prompt

[0522] "Please answer the following question considering the emotion data: The user has asked for an explanation of the equations of motion in physics. The emotion data indicates 'fear'."

[0523] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0524] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0525] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0526] [Second Embodiment]

[0527] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0528] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0529] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0530] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0531] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0532] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0533] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0534] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0535] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0536] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0537] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0538] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0539] The specific embodiments of the present invention are described below. This system is a learning support system that includes a generative model, a function for generating custom learning content based on students' learning progress and interests, and a function for planning and managing the progress of learning plans. This system operates with three main parties: the server, the terminal, and the user.

[0540] Overall system flow

[0541] server

[0542] The server implements the generative model, which is the core of this system. The server first loads the generative model, configures various settings, and prepares to receive requests. Based on requests sent from users and terminals, the server uses the generative model to generate responses and customized learning materials, and sends the results to the terminals. The server also maintains information on each student's learning progress and handles the planning and management of learning plans.

[0543] terminal

[0544] The terminal provides the user interface and is responsible for sending user input to the server. When a user logs in and submits questions or requests, the terminal displays responses and generated content from the server that received them. The terminal provides an appropriate interface to make it easier for the user to use.

[0545] User

[0546] Users are primarily students and teachers who input questions and requests into the system. They also request the creation of learning plans and monitor their progress as they learn.

[0547] Program processing

[0548] Loading and initializing the generative model

[0549] The server loads and initializes the generated model upon startup. It reads the model's parameters and configuration files, preparing it to function correctly. It starts a listener to await requests and waits for connections from clients.

[0550] User authentication and login

[0551] The device displays a login screen, and the user enters their authentication information (username and password). The device sends this information to the server, which then authenticates the user by comparing it against the authentication database. If authentication is successful, a token is generated and sent to the device. If authentication fails, an error message is returned.

[0552] Use as a supplementary tool for individualized instruction

[0553] The user inputs a question about a mathematical problem, and the terminal sends it to the server. The server uses a generative model to generate a response to the question and sends it back to the terminal. The terminal then displays the received response to the user.

[0554] Generating custom learning content

[0555] The user requests the generation of learning content based on a specific theme or difficulty level. The device sends this request to the server. The server uses a generative model to generate custom learning content and sends it to the device. This content is displayed on the device, and the user uses it for their learning.

[0556] Planning and managing study progress

[0557] The user submits a request for a learning plan and sends it to the server via their device. The server generates a learning plan using generative models and other algorithms based on the user's learning history and progress information, and sends it to the device. The user proceeds with their learning based on this plan and reports their progress to the server via their device. The server manages the progress and adjusts or improves the learning plan as needed.

[0558] Specific example

[0559] 1. When a user asks a question about a mathematical formula

[0560] The user enters "Please tell me the quadratic formula" into the question input screen on their device and submits it.

[0561] The device sends this question to the server.

[0562] The server uses a generative model to generate an explanation of the quadratic formula and produces a response such as, "The quadratic formula for solving the quadratic equation ax^2 + bx + c = 0 is..."

[0563] The device displays this response to the user, who then uses the explanation for learning.

[0564] 2. When a user requests custom learning content

[0565] The user enters a request saying, "Please create a workbook of problems on the basics of vectors."

[0566] The device sends this request to the server.

[0567] The server uses a generative model to generate a set of problems on the fundamentals of vectors and sends them to the terminal.

[0568] The user uses the problem set to progress with their studies.

[0569] 3. When the user creates a learning plan

[0570] The user requests, "Please create a study plan for the next math exam," in order to develop a study plan for the next exam.

[0571] The device sends this request to the server.

[0572] The server uses a generative model to generate an appropriate learning plan based on the user's learning history and progress data, and sends it to the terminal.

[0573] The device displays a learning plan to the user, and the user proceeds with their studies according to that plan.

[0574] As described above, the present invention provides a system that provides efficient and personalized learning support by using a generative model. This system allows users to receive high-quality learning support and improve their learning outcomes.

[0575] The following describes the processing flow.

[0576] Step 1: Load and initialize the generative model

[0577] The server loads the generated model upon startup, reads the necessary parameters and configuration files, and initializes it. It allocates resources for the model to function correctly and starts a listener to receive requests.

[0578] Step 2: User Authentication and Login

[0579] The terminal displays a login screen to the user. The user enters their authentication information (username, password) and presses the submit button. The terminal sends this authentication information to the server. The server compares the received authentication information with its database, and if authentication is successful, generates a JWT token and sends it back to the terminal. If authentication fails, it sends an appropriate error message back to the terminal.

[0580] Step 3: Enter Questions

[0581] The user enters text about their question or problem into the question input screen on the device and presses the submit button. The device then sends the entered question to the server.

[0582] Step 4: Question analysis and response generation

[0583] The server analyzes the received question and provides input to the generative model. The generative model generates a response to the question and returns the result to the server.

[0584] Step 5: Send the generated response

[0585] The server sends the response returned from the generative model to the terminal. The terminal displays the received response to the user. The user reviews the displayed response and enters additional questions as needed.

[0586] Step 6: Request custom learning content

[0587] The user wishes to generate custom learning content based on a specific learning theme or difficulty level, and enters a request into their device. The device then sends this request to the server.

[0588] Step 7: Generating Custom Content

[0589] The server receives a custom learning content generation request and uses a generative model to generate learning content based on the request. It then constructs a response containing the generated content and sends it to the terminal.

[0590] Step 8: Displaying the generated content

[0591] The device receives the generated learning content and displays it to the user. The user then uses the displayed content to continue learning.

[0592] Step 9: Develop a study plan

[0593] The user requests assistance with exam preparation or the creation of a study plan for specific learning objectives. The device sends this request to the server.

[0594] Step 10: Generate and submit your study plan.

[0595] The server generates an appropriate learning plan using generative models and other algorithms based on the user's learning history and progress information. It then sends the generated learning plan to the user's device.

[0596] Step 11: Display and implement your study plan

[0597] The device receives the generated learning plan and displays it to the user. The user proceeds with their studies based on the displayed learning plan and inputs their progress into the device as needed.

[0598] Step 12: Progress Management and Feedback

[0599] The server receives progress information sent by the user and manages the progress of the learning plan. If necessary, it generates feedback using a generative model and sends it to the terminal. The terminal displays the feedback to the user and helps them with their next learning session.

[0600] (Example 1)

[0601] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0602] Conventional learning support systems struggle to individually address the diverse learning needs of users. Furthermore, they fail to effectively manage learning progress, creating a need for efficient means to support planned learning. Additionally, the security of user authentication and the automation of on-demand learning content generation using generative models are insufficient, posing challenges to improving the quality and efficiency of learning.

[0603] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0604] In this invention, the server includes means for generating a response to user input using a generative model, means for the learning support server to read the generative model, perform various settings, and await requests, means for receiving user authentication information and issuing authentication tokens, means for generating custom learning content using the generative model in response to a learning content generation request from the terminal and sending it to the terminal, and means for managing student learning progress information, formulating a learning plan based on that information, and managing progress. This enables responses to individual learning needs, resulting in planned and effective learning support, secure user authentication, and on-demand generation and delivery of learning content.

[0605] A "generative model" is a system that generates text or responses using neural networks or machine learning algorithms.

[0606] "Responses to user input" refer to text-based answers and explanations generated based on questions and requests received from the user.

[0607] A "learning support server" is the core of a learning support system, handling tasks such as loading generative models, configuring various settings, processing requests, generating and sending learning content, and managing learning progress.

[0608] "Means for waiting for requests" refers to the function that allows a server to accept connections and data transmissions from users and terminals.

[0609] "User authentication information" refers to data such as usernames and passwords that are used to identify a user and grant them permissions.

[0610] An "authentication token" is a unique identifier issued to a user who has successfully authenticated, and is used to authenticate the user in subsequent communications.

[0611] A "learning content generation request" refers to a user requesting the creation of new learning materials or problem sets based on a specific theme or difficulty level.

[0612] "Custom learning content" refers to original learning materials and workbooks that are generated based on the user's requests and learning progress.

[0613] "Learning progress information" refers to data that records how far students have progressed in their studies.

[0614] A "study plan" is a document that outlines the learning content and progress schedule that students aim to achieve.

[0615] "Means of managing progress" refers to a function that tracks students' learning progress based on their learning plan and makes adjustments or improvements as needed.

[0616] "On-demand" refers to a service delivery method that provides services immediately in response to user requests.

[0617] This invention relates to a learning support system that includes a generative model, a learning support server, a function for generating custom learning content based on students' learning progress and interests, and a function for planning and managing learning plans. This invention is a system that operates primarily with three parties: the server, the terminal, and the user.

[0618] Server configuration and operation

[0619] The server implements the generative model (e.g., GPT-3 or BERT) that forms the core of this system. Upon startup, the server loads and initializes the generative model. During this process, it reads the model's parameters and configuration files (e.g., model_config.json and model_weights.h5) to prepare for correct operation. It also starts a listener to await requests and waits for connections from clients (terminals).

[0620] User Authentication

[0621] The server receives user authentication information (username, password) sent from the terminal and compares it with the authentication database. If authentication is successful, an authentication token is generated and sent to the terminal. If authentication fails, an error message is returned.

[0622] Learning content generation

[0623] The server responds to learning content generation requests from the terminal and generates custom learning content using a generative model. This content generation is based on user requests and individual learning progress information. The generated content is sent to the terminal and made available to the user.

[0624] Learning progress management and planning

[0625] The server manages each student's learning progress information and creates a learning plan based on it. In this process, generative models and other algorithms are used to analyze the user's learning history and progress data to generate an appropriate learning plan. The learning plan is sent to the user's device, where they can review and execute it.

[0626] Terminal configuration and operation

[0627] The terminal provides the user interface and is responsible for sending user input to the server. When a user logs in and enters questions or requests, the terminal displays the responses and generated content from the server that received them. The terminal provides a user-friendly interface, supporting smooth learning.

[0628] User actions

[0629] Users are primarily students and teachers, and they can input questions and requests into the system. For example, consider the following specific examples:

[0630] Questions about mathematical formulas

[0631] The user types "Please tell me the quadratic formula" into the terminal and sends it. The terminal sends this question to the server, which uses a generative model to generate a response and sends it back to the terminal. The terminal displays the generated response to the user.

[0632] Request for custom learning content

[0633] The user enters a request: "Please create a set of problems on the basics of vectors." The device sends this request to the server, which uses a generative model to generate custom learning content and sends it to the device. The user then uses the generated set of problems for learning.

[0634] Planning a study schedule

[0635] The user requests, "Please create a study plan for the next math exam." The device sends this request to the server, which uses a generative model to generate a study plan based on the user's learning history and progress data. The device displays this plan to the user, who then proceeds with their studies based on the plan.

[0636] Thus, the present invention uses a generative model to provide high-quality learning support and realize efficient and personalized learning.

[0637] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0638] Step 1:

[0639] The server starts up and loads the generative model. During this process, it reads the configuration file (e.g., model_config.json) and the model parameter file (e.g., model_weights.h5) and initializes the generative model. The server starts a listener to await requests and waits for connections and requests from clients. This process takes the server startup command and configuration files as input and outputs the state where the generative model initialization is complete.

[0640] Step 2:

[0641] The user accesses the login screen on their device and enters their username and password. The device sends this information to the server. The server compares it with the authentication database, and if authentication is successful, generates a token and sends it to the device. If authentication fails, it returns an error message. The input to this process is the user's authentication information, and the output is either an authentication token or an error message.

[0642] In terms of specific operations, the terminal receives user input and sends it to the server. On the server side, for example, an SQL query is executed to verify the user information (e.g., SELECT FROM users WHERE username = ? AND password_hash = ?).

[0643] Step 3:

[0644] The user enters "Please tell me the quadratic formula" on the terminal's question input screen and submits it. The terminal sends this question to the server. The server inputs the question as a prompt to the generative model, and the generative model generates a response. This generated response is sent to the terminal and displayed to the user. The input to this process is the user's question, and the output is the response generated by the generative model.

[0645] In terms of specific operation, the server inputs the response to the prompt "Please tell me the quadratic formula" into the generative model, and then sends the resulting response to the terminal.

[0646] Step 4:

[0647] A user requests and sends a request from their device saying, "Please create a set of problems on the basics of vectors." The device sends this request to the server. The server uses a generative model to generate custom learning content based on the theme. This generated content is sent to the device and made available to the user. The input to this process is the user's request to generate custom learning content, and the output is the generated custom learning content.

[0648] In terms of specific operation, the server inputs a request based on the prompt "Create a problem set on the basics of vectors" into the generative model, and then sends the resulting problem set to the terminal.

[0649] Step 5:

[0650] A user requests from their device, "Please create a study plan for the next math exam." The device sends this request to the server. The server retrieves the user's learning history and progress data from a database and generates a study plan using generative models and other algorithms. This study plan is sent to the device and displayed to the user. The input to this process is the user's study plan creation request and learning history data, and the output is the generated study plan.

[0651] In terms of specific operations, the server retrieves user progress data from the database, inputs it into a generative model to generate an appropriate learning plan, and then sends it to the terminal.

[0652] Step 6:

[0653] The user progresses through their learning based on a learning plan and reports their progress to the server from their terminal. The server stores this progress information in a database and adjusts or improves the learning plan as needed. The input to this process is the user's learning progress information, and the output is an updated learning plan and progress data.

[0654] In terms of specific operations, the user progresses through the learning plan, reports their progress via their device, the server records this in a database, and the learning plan is reviewed.

[0655] (Application Example 1)

[0656] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0657] Online education platforms require personalized learning support and an environment that enables users to learn efficiently. Furthermore, generating custom learning content in real time based on user progress and developing and managing learning plans based on individual needs are also crucial challenges. Current systems struggle to comprehensively provide these elements, resulting in insufficient support for maximizing learning effectiveness.

[0658] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0659] In this invention, the server includes means for generating responses to user input using a generative model, means for generating custom learning content based on the user's learning progress and interests, means for formulating and managing the user's learning plan, means for providing an individualized learning plan, and means for delivering the generated learning content to the user's terminal. As a result, the user can receive individualized learning support in real time and learn efficiently.

[0660] A "generative model" is a machine learning algorithm that automatically generates appropriate responses or content based on user input.

[0661] "User learning progress" refers to information that indicates the progress and level of proficiency achieved by learners in accordance with their learning plan.

[0662] "Custom learning content" refers to learning materials that are individually generated based on the needs, interests, and progress of a specific user.

[0663] A "user learning plan" refers to the specific steps and schedule that a user will use to achieve their learning goals.

[0664] A "personalized learning plan" refers to a learning schedule optimized for each user's individual learning history, progress, and goals.

[0665] A "user terminal" is an electronic device used by a user to communicate with a server and receive learning content.

[0666] "Real-time" means responding to or processing user requests immediately.

[0667] An "online education platform" is a system that provides learning content and educational services via the internet.

[0668] A "learning support system" refers to a set of computer programs and hardware that provide functions to effectively support a user's learning.

[0669] The specific embodiments of this invention are described below. This invention is a learning support system that utilizes a generative model and operates with three main parties: a server, a terminal, and a user. Various hardware and software and their processing are described below.

[0670] server

[0671] The server implements the generative model, which forms the core of this system. The server loads and initializes the generative model upon startup. The server is implemented using a server-side framework such as Flask and starts a listener to respond to user requests. The server also manages user authentication information using JWT (JSON Web Token). Specifically, a server machine with a high-performance processor and memory is used.

[0672] Generative model

[0673] A generative model is a machine learning algorithm that generates responses or custom learning content based on user requests. For example, it uses generative AI models such as OpenAI GPT-4. The model is loaded when the server starts up, and when it receives a request, it generates text based on its content.

[0674] terminal

[0675] The terminal provides the user interface and transmits user input to the server. Users log in using electronic devices such as smartphones, tablets, or PCs and enter questions or requests. The entered information is sent to the server, and the server's response and generated learning content are displayed on the terminal. The terminal provides a user-friendly interface to ensure easy operation for the user.

[0676] User

[0677] Users are primarily students and educators who input questions and requests into the system. Users can request the creation of learning plans and progress through their studies while checking progress information.

[0678] Data processing

[0679] Authentication: When a user logs in, the device sends authentication information to the server. The server compares this information with the authentication database, and if authentication is successful, generates a JWT token and sends it to the device.

[0680] Response generation: When a user enters a question about a mathematical problem into the terminal, the server uses a generative model to generate a response to the question and sends it to the terminal.

[0681] Generating custom learning content: When a user requests learning content based on a specific theme or difficulty level, the server uses a generative model to generate custom learning content and sends it to the device.

[0682] Learning plan creation: When a user requests the generation of a learning plan, the server generates a learning plan using generative models and other algorithms based on the user's learning history and progress information, and sends it to the terminal.

[0683] Specific example

[0684] 1. If a user wants to learn the basics of Python programming

[0685] The user types "I want to learn the basics of Python programming" into their terminal and sends a request.

[0686] The server uses a generative model to generate custom learning content on the fundamentals of Python programming and sends it to the device.

[0687] The device displays the generated learning content to the user, and the user begins learning.

[0688] 2. When a user plans their studies for the next math exam.

[0689] The user enters the request, "Please create a study plan for my next math exam."

[0690] The server generates an appropriate learning plan using a generative model based on the user's learning history and progress data, and sends it to the terminal.

[0691] The device displays the generated learning plan to the user, and the user proceeds with learning according to that plan.

[0692] Example of a prompt

[0693] Custom learning content generation:

[0694] "Generate learning content on the fundamentals of Python programming. Includes examples of videos, quizzes, and explanatory articles."

[0695] Learning plan generation:

[0696] "Create a study plan for the next math exam. Include daily tasks and key points based on the user's current progress and past learning history."

[0697] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0698] Step 1:

[0699] The server loads and initializes the generative model upon startup. The input is the model's configuration file and parameters, and the output is the initialized generative model. Specifically, the server reads the model file from the specified directory, sets the necessary parameters, and prepares the generative model.

[0700] Step 2:

[0701] The terminal displays a login screen, and the user enters their authentication information (username, password). The input is the username and password, and the output is a request containing the authentication information. The terminal sends this authentication information to the server.

[0702] Step 3:

[0703] The server compares the received authentication information with the authentication database, and if authentication is successful, generates a JWT token and sends it to the terminal. The input is the authentication information, and the output is either a JWT token or an error message. Specifically, the server compares it with a hashed password, and if authentication is successful, generates a token.

[0704] Step 4:

[0705] The user enters and submits questions and requests for learning content generation on their device. The input is the user's request, and the output is the request data. Specifically, the device sends data to the server using the submit button in the user's input field.

[0706] Step 5:

[0707] The server uses a generative model to generate responses or custom training content based on the received request. The input is the request data, and the output is the generated response or content. Specifically, the server inputs a prompt into the generative model and retrieves the generated result from the model.

[0708] Step 6:

[0709] The server sends the generated response or content to the terminal. The input is the generated response or content, and the output is the data sent to the terminal. Specifically, the server sends data to the terminal as an HTTP response.

[0710] Step 7:

[0711] The terminal displays received responses and content to the user. Input is data sent from the server, and output is information displayed on the user interface. Specifically, the terminal uses HTML and GUI components to display the data appropriately.

[0712] Step 8:

[0713] The user progresses through the learning process using the displayed responses and content, and reports progress information from the device to the server. The input is the user's learning progress information, and the output is the progress data sent to the server. Specifically, the device provides a field for entering progress information, and the user sends the data to the server using a submit button.

[0714] Step 9:

[0715] The server adjusts the learning plan based on the received progress information and generates a new learning plan as needed. The input is the user's progress data, and the output is a new or adjusted learning plan. Specifically, the server updates the database and generates or adjusts the plan using a generative model.

[0716] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0717] The specific embodiments of the present invention are described below. This system is a learning support system that includes a generative model, a function for generating custom learning content based on students' learning progress and interests, a function for planning and managing learning plans, and an emotion engine that recognizes user emotions and adjusts responses. This system operates with three main parties: the server, the terminal, and the user.

[0718] Overall system flow

[0719] server

[0720] The server implements the generative model and sentiment engine, which are the core of this system. The server first loads the generative model, configures various settings, and prepares to accept requests. Based on the user input and sentiment data included in the request, it uses the generative model to generate responses and customized learning materials, and sends the results to the terminal. The server also maintains each student's learning progress information and handles the planning and management of learning plans.

[0721] terminal

[0722] The terminal provides the user interface and is responsible for sending user input (text and sentiment data) to the server. When a user logs in and submits a question or request, the terminal displays the response from the server, generated content, and sentiment-based adjustments. The terminal provides an appropriate interface to make it easier for the user to use.

[0723] User

[0724] Users are primarily students and teachers who input questions and requests into the system. They also request the creation of learning plans and monitor their progress as they learn. Furthermore, the user's emotions are recognized through the terminal's input interface and transmitted to the server.

[0725] Program processing

[0726] Loading and initializing the generative model

[0727] Upon startup, the server loads the generative model and sentiment engine, and initializes by reading the necessary parameters and configuration files. It allocates resources for the model to function correctly and starts a listener to receive requests.

[0728] User authentication and login

[0729] The terminal displays a login screen to the user. The user enters their authentication information (username, password) and presses the submit button. The terminal sends this authentication information to the server. The server compares the received authentication information with its database, and if authentication is successful, generates a JWT token and sends it back to the terminal. If authentication fails, it sends an appropriate error message back to the terminal.

[0730] Question entry and submission of sentiment data

[0731] The user enters text about their question or problem into the terminal's question input screen and presses the submit button. The terminal sends the entered question, along with emotional data obtained from the user's facial recognition and voice tone analysis, to the server.

[0732] Question analysis and response generation

[0733] The server analyzes the received question and provides input to the generative model, taking into account the sentiment data recognized by the sentiment engine. The generative model generates a response based on the question and sentiment data and returns the result to the server.

[0734] Sending the generated response

[0735] The server sends the response returned from the generative model to the terminal after making appropriate adjustments based on the user's sentiment. The terminal displays the received response to the user. The user reviews the displayed response and enters additional questions if necessary.

[0736] Request for custom learning content

[0737] The user requests the generation of custom learning content based on a specific learning theme or difficulty level, and enters this request into their device. Along with this request, the device sends the user's sentiment data to the server.

[0738] Generating custom content

[0739] The server receives a custom learning content generation request and uses a generative model to generate learning content based on the request. It constructs a response containing the generated content, adjusts the difficulty and tone based on the user's sentiment data, and sends it to the device.

[0740] Displaying generated content

[0741] The device receives the generated learning content and displays it to the user. The user then uses the displayed content to continue learning.

[0742] Planning a study schedule

[0743] The user requests assistance with exam preparation or the creation of a study plan for specific learning objectives. The device sends this request to the server.

[0744] Generating and submitting a study plan

[0745] The server generates an appropriate learning plan using a generative model and sentiment engine based on the user's learning history and progress information. The generated learning plan is then sent to the device.

[0746] Displaying and implementing the study plan

[0747] The device receives the generated learning plan and displays it to the user. The user proceeds with their studies based on the displayed learning plan and inputs their progress into the device as needed.

[0748] Progress management and feedback

[0749] The server receives progress information and sentiment data sent by the user and manages the progress of the learning plan. If necessary, it generates feedback using a generative model and sends it to the device. The device displays the feedback to the user, who can then use it to improve their next learning session.

[0750] Specific example

[0751] 1. When using sentiment data when users ask questions about mathematical formulas.

[0752] The user enters "Please tell me the quadratic formula" into the question input screen on their device and submits it.

[0753] Along with this question, the device sends the user's facial expression recognition results (e.g., confused expression) and voice tone analysis results to the server.

[0754] The server uses a generative model to generate responses to questions, taking into account feelings of confusion and producing responses such as "You should explain with more specific examples."

[0755] The device displays this response to the user, who then uses the explanation for learning.

[0756] 2. When sentiment data is used when users request custom learning content.

[0757] The user requests, "Please create a workbook of problems on the basics of vectors."

[0758] Along with this request, the device sends the user's sentiment data to the server.

[0759] The server uses a generative model to generate a set of problems on the fundamentals of vectors, and adjusts the difficulty level if the user's motivation to learn is high.

[0760] The device displays the generated problem sets to the user, who then uses them to progress with their studies.

[0761] 3. When users utilize sentiment data when developing learning plans.

[0762] The user requests, "Please create a study plan for the next math exam," in order to develop a study plan for the next exam.

[0763] Along with this request, the device sends the user's sentiment data to the server.

[0764] The server uses generative models and a sentiment engine to generate an appropriate learning plan based on the user's learning history, progress data, and sentiment data.

[0765] For example, if a user is experiencing stress, the server will adjust its schedule accordingly.

[0766] The device displays the generated learning plan to the user, and the user proceeds with learning according to that plan.

[0767] This invention provides a system that uses a generative model and an emotion engine to provide personalized learning support that takes into account the user's emotions. This system allows users to learn more effectively and with less stress.

[0768] The following describes the processing flow.

[0769] Step 1: Load and initialize the generative model

[0770] Upon startup, the server loads the generative model and sentiment engine, and initializes by reading the necessary parameters and configuration files. It allocates resources for the model to function correctly and starts a listener to receive requests.

[0771] Step 2: User Authentication and Login

[0772] The terminal displays a login screen to the user. The user enters their authentication information (username, password) and presses the submit button. The terminal sends this authentication information to the server. The server compares the received authentication information with its database, and if authentication is successful, generates a JWT token and sends it back to the terminal. If authentication fails, it sends an appropriate error message back to the terminal.

[0773] Step 3: Enter the questions and submit sentiment data.

[0774] The user enters text about their question or problem into the terminal's question input screen and presses the submit button. The terminal sends the entered question, along with emotional data obtained from the user's facial recognition and voice tone analysis, to the server.

[0775] Step 4: Question analysis and response generation

[0776] The server analyzes the received question and provides input to the generative model, taking into account the sentiment data recognized by the sentiment engine. The generative model generates a response based on the question and sentiment data and returns the result to the server.

[0777] Step 5: Send the generated response

[0778] The server sends the response returned from the generative model to the terminal after making appropriate adjustments based on the user's sentiment. The terminal displays the received response to the user. The user reviews the displayed response and enters additional questions if necessary.

[0779] Step 6: Request custom learning content

[0780] The user requests the generation of custom learning content based on a specific learning theme or difficulty level, and enters this request into their device. Along with this request, the device sends the user's sentiment data to the server.

[0781] Step 7: Generating Custom Content

[0782] The server receives a custom learning content generation request and uses a generative model to generate learning content based on the request. It constructs a response containing the generated content, adjusts the difficulty and tone based on the user's sentiment data, and sends it to the device.

[0783] Step 8: Displaying the generated content

[0784] The device receives the generated learning content and displays it to the user. The user then uses the displayed content to continue learning.

[0785] Step 9: Develop a study plan

[0786] The user requests assistance with exam preparation or the creation of a study plan for specific learning objectives. The device sends this request to the server.

[0787] Step 10: Generate and submit your study plan.

[0788] The server generates an appropriate learning plan using a generative model and sentiment engine based on the user's learning history and progress information. The generated learning plan is then sent to the device.

[0789] Step 11: Display and implement your study plan

[0790] The device receives the generated learning plan and displays it to the user. The user proceeds with their studies based on the displayed learning plan and inputs their progress into the device as needed.

[0791] Step 12: Progress Management and Feedback

[0792] The server receives progress information and sentiment data sent by the user and manages the progress of the learning plan. If necessary, it generates feedback using a generative model and sends it to the device. The device displays the feedback to the user, who can then use it to improve their next learning session.

[0793] Specific example

[0794] 1. When a user asks a question about a mathematical formula

[0795] The user enters "Please tell me the quadratic formula" into the question input screen on their device and submits it.

[0796] Along with this question, the device sends the user's facial expression recognition results (e.g., confused expression) and voice tone analysis results to the server.

[0797] The server uses a generative model to generate responses to questions, taking into account feelings of confusion and producing responses such as "You should explain with more specific examples."

[0798] The device displays this response to the user, who then uses the explanation for learning.

[0799] 2. When a user requests custom learning content

[0800] The user requests, "Please create a workbook of problems on the basics of vectors."

[0801] Along with this request, the device sends the user's sentiment data to the server.

[0802] The server uses a generative model to generate a set of problems on the fundamentals of vectors, and adjusts the difficulty level if the user's motivation to learn is high.

[0803] The device displays the generated problem sets to the user, who then uses them to progress with their studies.

[0804] 3. When the user creates a learning plan

[0805] The user requests, "Please create a study plan for the next math exam," in order to develop a study plan for the next exam.

[0806] Along with this request, the device sends the user's sentiment data to the server.

[0807] The server uses generative models and a sentiment engine to generate an appropriate learning plan based on the user's learning history, progress data, and sentiment data.

[0808] For example, if a user is experiencing stress, the server will adjust its schedule accordingly.

[0809] The device displays the generated learning plan to the user, and the user proceeds with learning according to that plan.

[0810] (Example 2)

[0811] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0812] Traditional learning support systems provided features such as generating custom learning content based on individual users' learning progress and interests, developing learning plans, and managing progress. However, they lacked the ability to adjust responses based on user emotions. As a result, it was difficult to provide optimal learning support tailored to the user's emotional state, leading to challenges in maintaining learning effectiveness and motivation.

[0813] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0814] In this invention, the server includes means for generating responses to user input using a generative model, means for generating custom learning content based on the student's learning progress and interests, means for formulating a user's learning plan and managing its progress, means for recognizing the user's emotional data and adjusting the response, means for performing user authentication and matching authentication information with a database, means for adjusting the generated response based on the user's emotions and transmitting it to the terminal, and means for generating a learning plan and managing its progress based on the user's progress information and emotional data. This enables personalized learning support that takes the user's emotions into consideration.

[0815] A "generative model" is a pre-trained algorithm that generates natural language responses or content based on input data.

[0816] A "user" is a person, such as a student or teacher, who uses the system and is the entity that makes questions or learning requests through the learning support system.

[0817] "Custom learning content" refers to educational content such as learning materials and workbooks that are created individually based on a student's learning progress and interests.

[0818] A "learning plan" is a schedule or plan built based on the learning goals set by the user, and it also includes the management of progress.

[0819] "Emotional data" refers to data that indicates the user's current emotional state, obtained from user facial recognition and voice tone analysis.

[0820] A "response" refers to the natural language response or explanation that a generative model outputs in response to a user's question or request.

[0821] "User authentication" is the process of verifying the identity of users accessing a system, and it uses authentication information such as usernames and passwords.

[0822] A "terminal" is a device used by a user to interact with a system, and includes personal computers, tablets, smartphones, and other similar devices.

[0823] "Progress information" refers to data that shows the user's progress in their learning, and is recorded and managed through an online system.

[0824] A "database" is a digital repository for storing user authentication information, learning progress information, generated learning content, and other similar data.

[0825] A "listener" refers to the state in which a server is waiting for requests from external sources and is the process of accepting user input.

[0826] Specific embodiments of the present invention are described below. This learning support system is broadly composed of three components: a server, a terminal, and a user. The role of each component will be described in detail below.

[0827] Server configuration and operation

[0828] The server plays a central role in this system. It implements the following main functions:

[0829] 1. Loading and initializing the generative model:

[0830] When the server starts up, it loads pre-trained generative models and sentiment engines. This uses libraries such as Python and TensorFlow.

[0831] It reads the necessary parameters and configuration files, allocates resources, and starts a listener to wait for requests.

[0832] 2. User Authentication:

[0833] The server receives user authentication information (username, password) sent from the terminal and compares it with the database.

[0834] If authentication is successful, a JWT token is generated and sent back to the device; if authentication fails, an error message is sent back.

[0835] 3. Question analysis and response generation:

[0836] The server receives the question text and sentiment data sent by the user.

[0837] The question text and sentiment data are given to a generative model as input, and an appropriate response is generated. This response is adjusted considering the question content and sentiment data.

[0838] 4. Generating custom learning content:

[0839] The server generates custom learning content based on user requests and sentiment data.

[0840] The generated content is adjusted in difficulty and tone based on the user's learning motivation and emotional state.

[0841] 5. Planning and managing study progress:

[0842] The server generates an appropriate learning plan based on the user's learning history, progress data, and sentiment data.

[0843] The generated learning plan is sent to the device, and feedback and adjustments are made based on the user's progress.

[0844] Terminal configuration and operation

[0845] A terminal is a device that provides a user interface and performs the following functions:

[0846] 1. Displaying the login screen and sending user authentication information:

[0847] The device displays the login screen and sends the authentication information entered by the user to the server.

[0848] 2. Question entry and submission of sentiment data:

[0849] The user enters text into the question input screen on the device and presses the submit button. At the same time, emotional data is sent to the server based on facial recognition and voice tone analysis.

[0850] 3. Display of response from the server:

[0851] The terminal receives a response from the server and displays it to the user. The user reviews the displayed response and enters additional questions as needed.

[0852] 4. Displaying custom learning content:

[0853] The device displays custom learning content received from the server, enabling users to efficiently progress in their learning.

[0854] 5. Displaying and implementing the study plan:

[0855] The device receives the learning plan sent from the server and displays it to the user. The user then proceeds with their learning based on this learning plan.

[0856] User roles and actions

[0857] The user is the subject who uses the system to progress through the learning process, and takes the following actions:

[0858] 1. Enter your question or request:

[0859] Users input learning-related questions, requests for custom learning content, and learning plans into the system.

[0860] 2. Implementing the study plan and inputting progress:

[0861] The user progresses through the learning process based on the provided learning plan and enters progress information into their device.

[0862] 3. Use of learning content:

[0863] Students use the generated custom learning content to progress through their studies.

[0864] Specific example

[0865] If a user asks a question about a mathematical formula:

[0866] The user types the question "Please tell me the quadratic formula" into their device and sends it.

[0867] The facial recognition results (confused expression) and voice tone analysis results are also sent to the server at the same time.

[0868] The server uses a generative model to generate responses to questions, taking into account feelings of confusion and producing responses such as "You should explain with specific examples."

[0869] The terminal displays this response to the user.

[0870] When a user requests custom learning content:

[0871] I requested that you create a workbook of problems on the fundamentals of vectors.

[0872] Along with this request, the device sends the user's sentiment data to the server.

[0873] The server uses a generative model to generate a set of problems based on the request, and increases the difficulty level if the user's motivation is high.

[0874] The terminal displays the generated problem set to the user.

[0875] When a user creates a learning plan:

[0876] "Please create a study plan that will be necessary before the next math exam," I requested.

[0877] Along with this request, the device sends the user's sentiment data to the server.

[0878] The server uses generative models and an emotion engine to generate an appropriate learning plan based on the user's learning history, progress data, and sentiment data.

[0879] For example, if a user is experiencing stress, the server will adjust its schedule accordingly.

[0880] The device displays the generated learning plan to the user, and the user proceeds with learning according to that plan.

[0881] Example of a prompt:

[0882] "Please create a study plan for math and English for the next semester."

[0883] "If the user is confused by the tone of voice, please generate an easy-to-understand explanation."

[0884] "Please generate a set of problems on the fundamentals of vectors and set the difficulty level to intermediate."

[0885] In this way, a personalized learning support system using generative models and emotion engines can be realized.

[0886] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0887] Step 1: Load and initialize the generative model

[0888] When the server starts up, it loads a pre-trained generative model and sentiment engine. Specifically, it creates instances of the generative model using libraries such as Python and TensorFlow, and then loads the pre-trained model.

[0889] Input: Generative models and emotion engines, and their configuration files.

[0890] Data processing and calculations: Model loading and resource allocation.

[0891] Output: Start of listener to accept requests.

[0892] Step 2: User Authentication

[0893] The device displays a login screen, and the user enters their username and password and presses the submit button.

[0894] Enter: Username and password.

[0895] Data processing and calculations: Send authentication information to the server.

[0896] The server then compares this authentication information with the database.

[0897] Input: Authentication information.

[0898] Data processing and calculations: Database matching.

[0899] Output: If authentication is successful, a JWT token is generated and sent back to the device. If authentication fails, an error message is returned.

[0900] Step 3: Enter the questions and submit sentiment data.

[0901] The user enters text about their question or problem into the question input screen on their device and presses the submit button (e.g., "Please tell me the quadratic formula").

[0902] Input: Question text.

[0903] Data processing and computation: facial expression recognition and voice tone analysis.

[0904] The terminal sends the entered question along with emotional data obtained from facial recognition results and voice tone analysis to the server.

[0905] Input: Question text and sentiment data.

[0906] Data processing and calculations: Data transmission.

[0907] Output: Sending data to the server.

[0908] Step 4: Question analysis and response generation

[0909] The server analyzes the received question text and sentiment data.

[0910] Input: Question text and sentiment data.

[0911] Data processing and computation: Input to generative models.

[0912] Generating response text that takes emotional data into account (e.g., recognizing the emotion of confusion generates a response that includes a detailed explanation).

[0913] Output: The generated response text.

[0914] Step 5: Send the generated response

[0915] The server sends the generated response to the terminal after appropriately adjusting its emotional tone.

[0916] Input: The generated response text.

[0917] Data processing and calculation: Adjustments based on emotional data.

[0918] Output: Sends the adjusted response text to the terminal.

[0919] Step 6: Display and confirm the response

[0920] The terminal displays the received response to the user.

[0921] Input: Adjusted response text.

[0922] Data processing and calculations: Display processing.

[0923] Output: Display the response to the user.

[0924] Step 7: Request custom learning content

[0925] Users can request the creation of custom learning content based on specific learning themes and difficulty levels (e.g., "Please create a set of problems on the basics of vectors").

[0926] Input: Request for custom learning content.

[0927] Data processing and computation: facial expression recognition and voice tone analysis.

[0928] The device sends emotion data to the server along with the request.

[0929] Input: Request and sentiment data.

[0930] Data processing and calculations: Data transmission.

[0931] Output: Sending requests and sentiment data to the server.

[0932] Step 8: Generating Custom Content

[0933] The server receives a custom training content generation request and uses a generative model to generate the training content.

[0934] Input: Request and sentiment data.

[0935] Data processing and computation: Generating custom content using generative models.

[0936] The difficulty level and tone are adjusted based on emotional data.

[0937] Output: Generated training content.

[0938] Step 9: Displaying the generated content

[0939] The device receives the generated learning content and displays it to the user.

[0940] Input: Generated learning content.

[0941] Data processing and calculations: Display processing.

[0942] Output: Display learning content to the user.

[0943] Step 10: Develop a study plan

[0944] The user requests the creation of a study plan (e.g., "Please create a study plan for the next math exam").

[0945] Input: Request for a study plan.

[0946] Data processing and computation: facial expression recognition and voice tone analysis.

[0947] The device sends emotion data to the server along with the request.

[0948] Input: Request and sentiment data.

[0949] Data processing and calculations: Data transmission.

[0950] Output: Sending a training plan request and sentiment data to the server.

[0951] Step 11: Generate and submit your study plan.

[0952] The server uses generative models and a sentiment engine to generate an appropriate learning plan based on the user's learning history, progress data, and sentiment data.

[0953] Inputs: Learning history, progress data, sentiment data.

[0954] Data processing and computation: Generating learning plans using generative models.

[0955] Adjusting schedules based on emotional data.

[0956] Output: The generated training plan.

[0957] Step 12: Display and implement your study plan

[0958] The device receives the generated learning plan and displays it to the user.

[0959] Input: The generated training plan.

[0960] Data processing and calculations: Display processing.

[0961] Output: Displays the learning plan to the user.

[0962] Step 13: Progress Management and Feedback

[0963] The user progresses through the learning process according to the learning plan and enters progress information into the device.

[0964] Input: Progress information.

[0965] Data processing and calculations: Data transmission.

[0966] The server receives progress information and sentiment data sent by the user and manages the progress of the learning plan.

[0967] Input: Progress information and sentiment data.

[0968] Data processing and computation: progress management and feedback generation.

[0969] The device displays feedback to the user, which can be used to improve future learning.

[0970] Output: Feedback to the user.

[0971] (Application Example 2)

[0972] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0973] Conventional learning support systems have difficulty generating customized learning content tailored to each student's individual learning progress and interests, and they also lack the ability to provide responses that take into account students' emotional states. As a result, it has been difficult for students to properly plan their studies, manage their progress, and learn effectively, leading to a decline in the quality of learning.

[0974] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0975] In this invention, the server includes means for generating responses to user input using a generative model, means for generating custom learning content based on the learner's learning progress and interests, means for formulating a user's learning plan and managing its progress, and means for recognizing the user's emotions and adjusting responses and learning content based on that data. This enables personalized learning support for each individual learner, taking into account their learning progress and emotions.

[0976] A "generative model" is an algorithm that generates appropriate responses or content based on user input.

[0977] "Customized learning content" refers to learning materials and exercises tailored to each learner's individual learning progress and interests.

[0978] A "learning plan" is a schedule and set of action guidelines created to help users effectively progress towards their learning goals.

[0979] "Progress management" is a method for checking the results and progress of learners' learning activities and making adjustments as needed.

[0980] An "emotion engine" is an algorithm that recognizes a user's emotions and adjusts responses and content based on that data.

[0981] "User" refers to individuals who use this system, such as learners and teachers.

[0982] A "server" is a computer system that handles the core functions of a system and implements generative models, emotion engines, and other related components.

[0983] A "terminal" is a device that provides a user interface and transmits user input to a server.

[0984] "Interface" refers to the user interface or input method that allows a user to interact with a system.

[0985] The learning support system of the present invention consists of a server, a terminal, and a user. This system includes a generative model, the generation of custom learning content based on learning progress and interests, the planning and progress management of learning plans, and an emotion engine.

[0986] The server is the core of the entire system, implementing the generative model and sentiment engine. The server first loads the generative model, configures various settings, and prepares to receive requests. Requests include user input and sentiment data, and based on this, the server uses the generative model to generate responses and customized learning materials. The generated results are sent to the terminal. The server also maintains each student's learning progress information and handles the planning and management of learning progress.

[0987] The terminal provides the user interface and transmits user input (text and sentiment data) to the server. Users log in to the terminal and enter questions or requests. Responses from the server, generated content, and sentiment-based adjustments are displayed on the terminal. The terminal provides an appropriate interface to make it easier for users to use.

[0988] Users are primarily learners and instructors who input questions and requests into the system. They also request the creation of learning plans and monitor their progress as they learn. Furthermore, the user's emotions are recognized through the terminal's input interface and transmitted to the server.

[0989] Program Processing Description

[0990] The server loads and initializes the generative model. When the server starts, it loads the generative model and sentiment engine, and initializes the necessary parameters and configuration files. This allocates resources for the model to function correctly and starts a listener to accept requests.

[0991] The terminal supports user authentication and login. The terminal displays a login screen to the user, who enters their authentication information (username, password). The authentication information is sent from the terminal to the server, which compares the received information with its database. If authentication is successful, a JWT token is generated and sent back to the terminal.

[0992] Furthermore, when a user enters a question or requests custom learning content, the device sends sentiment data along with the content to the server. The server analyzes the received questions and requests and provides input to the generative model based on the sentiment data recognized by the sentiment engine. The generative model generates responses and custom learning content based on the questions and sentiment data, and returns the results to the server. The server then sends this to the device for display to the user.

[0993] Hardware and software used

[0994] Server: A computer system that performs the core functions of a system (e.g., "Nginx", "Docker").

[0995] Generative models and sentiment engines: Algorithms and libraries (e.g., TensorFlow, PyTorch)

[0996] Device: A device that provides a user interface (e.g., "iOS device", "Android device")

[0997] Database: Stores user information and learning progress (e.g., "PostgreSQL").

[0998] Examples of specific cases and prompt statements

[0999] As a concrete example, when a user asks a question about the equations of motion in physics, a fearful expression is detected through facial recognition. This information is sent from the terminal to the server, which uses a generative model to provide a detailed and easy-to-understand answer.

[1000] Example prompt: "Please answer the following question considering the sentiment data: The user has requested an explanation of the equations of motion in physics. The sentiment data indicates 'fear'."

[1001] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1002] Step 1:

[1003] The server loads and initializes the generative model and sentiment engine. Specifically, at startup, the server loads the generative model and sentiment engine into memory and reads the necessary parameters and configuration files. This ensures that the generative model and sentiment engine have the resources to function correctly. It also starts a listener to receive requests. The input is the configuration files and parameters of the generative model and sentiment engine, and the output is the initialized generative model and sentiment engine.

[1004] Step 2:

[1005] The terminal receives the user's login information and sends it to the server. Specifically, the user enters their username and password on the terminal's login screen and presses the submit button. The terminal sends this authentication information to the server. The input is the user's login information (username, password), and the output is the authentication request sent to the server.

[1006] Step 3:

[1007] The server authenticates the login information. Specifically, the server compares the received authentication information with the database, and if successful, generates a JWT token and sends it back to the terminal. If it fails, it sends an appropriate error message back to the terminal. The input is the authentication information received from the terminal, and the output is a JWT token if authentication is successful, and an error message if it fails.

[1008] Step 4:

[1009] The device collects user questions and sentiment data and sends them to the server. Specifically, the user enters text about their questions or problems into the question input screen and presses the submit button. The device simultaneously sends sentiment data acquired from the camera and voice input to the server. The input consists of the user's question text and sentiment data, and the output is the request sent to the server.

[1010] Step 5:

[1011] The server receives questions and sentiment data, analyzes them, and generates responses. Specifically, the server analyzes the received questions and inputs the recognized sentiment data into a generative model. The generative model generates appropriate responses based on the questions and sentiment data. The input is the questions and sentiment data received from the terminal, and the output is the generated response data.

[1012] Step 6:

[1013] The server adjusts the generated response based on sentiment data and sends it to the terminal. Specifically, the server adjusts the tone and level of detail of the response returned from the generative model according to the user's emotions, and then sends it to the terminal. The input is the generated response data and sentiment data, and the output is the adjusted response.

[1014] Step 7:

[1015] The terminal displays the received response to the user. Specifically, the terminal displays the response sent from the server on the screen, allowing the user to confirm the response. The input is the adjusted response received from the server, and the output is the response displayed on the screen.

[1016] Step 8:

[1017] Users request custom learning content. Specifically, users input a request on their device screen to generate custom learning content based on a specific learning theme and difficulty level. The input is the user's request, and the output is the request sent from the device to the server.

[1018] Step 9:

[1019] The server generates custom learning content and adjusts it based on sentiment data. Specifically, the server uses a generative model to generate learning content based on requests and adjusts the difficulty and tone by referring to the user's sentiment data. The input is the user's request and sentiment data, and the output is the generated custom learning content.

[1020] Step 10:

[1021] The device displays the generated learning content to the user. Specifically, it receives the generated content from the server and displays it on the screen. The input is the generated learning content, and the output is the learning content displayed to the user.

[1022] Example of a prompt

[1023] "Please answer the following question considering the emotion data: The user has asked for an explanation of the equations of motion in physics. The emotion data indicates 'fear'."

[1024] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1025] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1026] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[1027] [Third Embodiment]

[1028] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[1029] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1030] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1031] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[1032] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1033] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1034] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1035] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1036] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1037] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1038] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1039] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[1040] The specific embodiments of the present invention are described below. This system is a learning support system that includes a generative model, a function for generating custom learning content based on students' learning progress and interests, and a function for planning and managing the progress of learning plans. This system operates with three main parties: the server, the terminal, and the user.

[1041] Overall system flow

[1042] server

[1043] The server implements the generative model, which is the core of this system. The server first loads the generative model, configures various settings, and prepares to receive requests. Based on requests sent from users and terminals, the server uses the generative model to generate responses and customized learning materials, and sends the results to the terminals. The server also maintains information on each student's learning progress and handles the planning and management of learning plans.

[1044] terminal

[1045] The terminal provides the user interface and is responsible for sending user input to the server. When a user logs in and submits questions or requests, the terminal displays responses and generated content from the server that received them. The terminal provides an appropriate interface to make it easier for the user to use.

[1046] User

[1047] Users are primarily students and teachers who input questions and requests into the system. They also request the creation of learning plans and monitor their progress as they learn.

[1048] Program processing

[1049] Loading and initializing the generative model

[1050] The server loads and initializes the generated model upon startup. It reads the model's parameters and configuration files, preparing it to function correctly. It starts a listener to await requests and waits for connections from clients.

[1051] User authentication and login

[1052] The device displays a login screen, and the user enters their authentication information (username and password). The device sends this information to the server, which then authenticates the user by comparing it against the authentication database. If authentication is successful, a token is generated and sent to the device. If authentication fails, an error message is returned.

[1053] Use as a supplementary tool for individualized instruction

[1054] The user inputs a question about a mathematical problem, and the terminal sends it to the server. The server uses a generative model to generate a response to the question and sends it back to the terminal. The terminal then displays the received response to the user.

[1055] Generating custom learning content

[1056] The user requests the generation of learning content based on a specific theme or difficulty level. The device sends this request to the server. The server uses a generative model to generate custom learning content and sends it to the device. This content is displayed on the device, and the user uses it for their learning.

[1057] Planning and managing study progress

[1058] The user submits a request for a learning plan and sends it to the server via their device. The server generates a learning plan using generative models and other algorithms based on the user's learning history and progress information, and sends it to the device. The user proceeds with their learning based on this plan and reports their progress to the server via their device. The server manages the progress and adjusts or improves the learning plan as needed.

[1059] Specific example

[1060] 1. When a user asks a question about a mathematical formula

[1061] The user enters "Please tell me the quadratic formula" into the question input screen on their device and submits it.

[1062] The device sends this question to the server.

[1063] The server uses a generative model to generate an explanation of the quadratic formula and produces a response such as, "The quadratic formula for solving the quadratic equation ax^2 + bx + c = 0 is..."

[1064] The device displays this response to the user, who then uses the explanation for learning.

[1065] 2. When a user requests custom learning content

[1066] The user enters a request saying, "Please create a workbook of problems on the basics of vectors."

[1067] The device sends this request to the server.

[1068] The server uses a generative model to generate a set of problems on the fundamentals of vectors and sends them to the terminal.

[1069] The user uses the problem set to progress with their studies.

[1070] 3. When the user creates a learning plan

[1071] The user requests, "Please create a study plan for the next math exam," in order to develop a study plan for the next exam.

[1072] The device sends this request to the server.

[1073] The server uses a generative model to generate an appropriate learning plan based on the user's learning history and progress data, and sends it to the terminal.

[1074] The device displays a learning plan to the user, and the user proceeds with their studies according to that plan.

[1075] As described above, the present invention provides a system that provides efficient and personalized learning support by using a generative model. This system allows users to receive high-quality learning support and improve their learning outcomes.

[1076] The following describes the processing flow.

[1077] Step 1: Load and initialize the generative model

[1078] The server loads the generated model upon startup, reads the necessary parameters and configuration files, and initializes it. It allocates resources for the model to function correctly and starts a listener to receive requests.

[1079] Step 2: User Authentication and Login

[1080] The terminal displays a login screen to the user. The user enters their authentication information (username, password) and presses the submit button. The terminal sends this authentication information to the server. The server compares the received authentication information with its database, and if authentication is successful, generates a JWT token and sends it back to the terminal. If authentication fails, it sends an appropriate error message back to the terminal.

[1081] Step 3: Enter Questions

[1082] The user enters text about their question or problem into the question input screen on the device and presses the submit button. The device then sends the entered question to the server.

[1083] Step 4: Question analysis and response generation

[1084] The server analyzes the received question and provides input to the generative model. The generative model generates a response to the question and returns the result to the server.

[1085] Step 5: Send the generated response

[1086] The server sends the response returned from the generative model to the terminal. The terminal displays the received response to the user. The user reviews the displayed response and enters additional questions as needed.

[1087] Step 6: Request custom learning content

[1088] The user wishes to generate custom learning content based on a specific learning theme or difficulty level, and enters a request into their device. The device then sends this request to the server.

[1089] Step 7: Generating Custom Content

[1090] The server receives a custom learning content generation request and uses a generative model to generate learning content based on the request. It then constructs a response containing the generated content and sends it to the terminal.

[1091] Step 8: Displaying the generated content

[1092] The device receives the generated learning content and displays it to the user. The user then uses the displayed content to continue learning.

[1093] Step 9: Develop a study plan

[1094] The user requests assistance with exam preparation or the creation of a study plan for specific learning objectives. The device sends this request to the server.

[1095] Step 10: Generate and submit your study plan.

[1096] The server generates an appropriate learning plan using generative models and other algorithms based on the user's learning history and progress information. It then sends the generated learning plan to the user's device.

[1097] Step 11: Display and implement your study plan

[1098] The device receives the generated learning plan and displays it to the user. The user proceeds with their studies based on the displayed learning plan and inputs their progress into the device as needed.

[1099] Step 12: Progress Management and Feedback

[1100] The server receives progress information sent by the user and manages the progress of the learning plan. If necessary, it generates feedback using a generative model and sends it to the terminal. The terminal displays the feedback to the user and helps them with their next learning session.

[1101] (Example 1)

[1102] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1103] Conventional learning support systems struggle to individually address the diverse learning needs of users. Furthermore, they fail to effectively manage learning progress, creating a need for efficient means to support planned learning. Additionally, the security of user authentication and the automation of on-demand learning content generation using generative models are insufficient, posing challenges to improving the quality and efficiency of learning.

[1104] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1105] In this invention, the server includes means for generating a response to user input using a generative model, means for the learning support server to read the generative model, perform various settings, and await requests, means for receiving user authentication information and issuing authentication tokens, means for generating custom learning content using the generative model in response to a learning content generation request from the terminal and sending it to the terminal, and means for managing student learning progress information, formulating a learning plan based on that information, and managing progress. This enables responses to individual learning needs, resulting in planned and effective learning support, secure user authentication, and on-demand generation and delivery of learning content.

[1106] A "generative model" is a system that generates text or responses using neural networks or machine learning algorithms.

[1107] "Responses to user input" refer to text-based answers and explanations generated based on questions and requests received from the user.

[1108] A "learning support server" is the core of a learning support system, handling tasks such as loading generative models, configuring various settings, processing requests, generating and sending learning content, and managing learning progress.

[1109] "Means for waiting for requests" refers to the function that allows a server to accept connections and data transmissions from users and terminals.

[1110] "User authentication information" refers to data such as usernames and passwords that are used to identify a user and grant them permissions.

[1111] An "authentication token" is a unique identifier issued to a user who has successfully authenticated, and is used to authenticate the user in subsequent communications.

[1112] A "learning content generation request" refers to a user requesting the creation of new learning materials or problem sets based on a specific theme or difficulty level.

[1113] "Custom learning content" refers to original learning materials and workbooks that are generated based on the user's requests and learning progress.

[1114] "Learning progress information" refers to data that records how far students have progressed in their studies.

[1115] A "study plan" is a document that outlines the learning content and progress schedule that students aim to achieve.

[1116] "Means of managing progress" refers to a function that tracks students' learning progress based on their learning plan and makes adjustments or improvements as needed.

[1117] "On-demand" refers to a service delivery method that provides services immediately in response to user requests.

[1118] This invention relates to a learning support system that includes a generative model, a learning support server, a function for generating custom learning content based on students' learning progress and interests, and a function for planning and managing learning plans. This invention is a system that operates primarily with three parties: the server, the terminal, and the user.

[1119] Server configuration and operation

[1120] The server implements the generative model (e.g., GPT-3 or BERT) that forms the core of this system. Upon startup, the server loads and initializes the generative model. During this process, it reads the model's parameters and configuration files (e.g., model_config.json and model_weights.h5) to prepare for correct operation. It also starts a listener to await requests and waits for connections from clients (terminals).

[1121] User Authentication

[1122] The server receives user authentication information (username, password) sent from the terminal and compares it with the authentication database. If authentication is successful, an authentication token is generated and sent to the terminal. If authentication fails, an error message is returned.

[1123] Learning content generation

[1124] The server responds to learning content generation requests from the terminal and generates custom learning content using a generative model. This content generation is based on user requests and individual learning progress information. The generated content is sent to the terminal and made available to the user.

[1125] Learning progress management and planning

[1126] The server manages each student's learning progress information and creates a learning plan based on it. In this process, generative models and other algorithms are used to analyze the user's learning history and progress data to generate an appropriate learning plan. The learning plan is sent to the user's device, where they can review and execute it.

[1127] Terminal configuration and operation

[1128] The terminal provides the user interface and is responsible for sending user input to the server. When a user logs in and enters questions or requests, the terminal displays the responses and generated content from the server that received them. The terminal provides a user-friendly interface, supporting smooth learning.

[1129] User actions

[1130] Users are primarily students and teachers, and they can input questions and requests into the system. For example, consider the following specific examples:

[1131] Questions about mathematical formulas

[1132] The user types "Please tell me the quadratic formula" into the terminal and sends it. The terminal sends this question to the server, which uses a generative model to generate a response and sends it back to the terminal. The terminal displays the generated response to the user.

[1133] Request for custom learning content

[1134] The user enters a request: "Please create a set of problems on the basics of vectors." The device sends this request to the server, which uses a generative model to generate custom learning content and sends it to the device. The user then uses the generated set of problems for learning.

[1135] Planning a study schedule

[1136] The user requests, "Please create a study plan for the next math exam." The device sends this request to the server, which uses a generative model to generate a study plan based on the user's learning history and progress data. The device displays this plan to the user, who then proceeds with their studies based on the plan.

[1137] Thus, the present invention uses a generative model to provide high-quality learning support and realize efficient and personalized learning.

[1138] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1139] Step 1:

[1140] The server starts up and loads the generative model. During this process, it reads the configuration file (e.g., model_config.json) and the model parameter file (e.g., model_weights.h5) and initializes the generative model. The server starts a listener to await requests and waits for connections and requests from clients. This process takes the server startup command and configuration files as input and outputs the state where the generative model initialization is complete.

[1141] Step 2:

[1142] The user accesses the login screen on their device and enters their username and password. The device sends this information to the server. The server compares it with the authentication database, and if authentication is successful, generates a token and sends it to the device. If authentication fails, it returns an error message. The input to this process is the user's authentication information, and the output is either an authentication token or an error message.

[1143] In terms of specific operations, the terminal receives user input and sends it to the server. On the server side, for example, an SQL query is executed to verify the user information (e.g., SELECT FROM users WHERE username = ? AND password_hash = ?).

[1144] Step 3:

[1145] The user enters "Please tell me the quadratic formula" on the terminal's question input screen and submits it. The terminal sends this question to the server. The server inputs the question as a prompt to the generative model, and the generative model generates a response. This generated response is sent to the terminal and displayed to the user. The input to this process is the user's question, and the output is the response generated by the generative model.

[1146] In terms of specific operation, the server inputs the response to the prompt "Please tell me the quadratic formula" into the generative model, and then sends the resulting response to the terminal.

[1147] Step 4:

[1148] A user requests and sends a request from their device saying, "Please create a set of problems on the basics of vectors." The device sends this request to the server. The server uses a generative model to generate custom learning content based on the theme. This generated content is sent to the device and made available to the user. The input to this process is the user's request to generate custom learning content, and the output is the generated custom learning content.

[1149] In terms of specific operation, the server inputs a request based on the prompt "Create a problem set on the basics of vectors" into the generative model, and then sends the resulting problem set to the terminal.

[1150] Step 5:

[1151] A user requests from their device, "Please create a study plan for the next math exam." The device sends this request to the server. The server retrieves the user's learning history and progress data from a database and generates a study plan using generative models and other algorithms. This study plan is sent to the device and displayed to the user. The input to this process is the user's study plan creation request and learning history data, and the output is the generated study plan.

[1152] In terms of specific operations, the server retrieves user progress data from the database, inputs it into a generative model to generate an appropriate learning plan, and then sends it to the terminal.

[1153] Step 6:

[1154] The user progresses through their learning based on a learning plan and reports their progress to the server from their terminal. The server stores this progress information in a database and adjusts or improves the learning plan as needed. The input to this process is the user's learning progress information, and the output is an updated learning plan and progress data.

[1155] In terms of specific operations, the user progresses through the learning plan, reports their progress via their device, the server records this in a database, and the learning plan is reviewed.

[1156] (Application Example 1)

[1157] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1158] Online education platforms require personalized learning support and an environment that enables users to learn efficiently. Furthermore, generating custom learning content in real time based on user progress and developing and managing learning plans based on individual needs are also crucial challenges. Current systems struggle to comprehensively provide these elements, resulting in insufficient support for maximizing learning effectiveness.

[1159] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1160] In this invention, the server includes means for generating responses to user input using a generative model, means for generating custom learning content based on the user's learning progress and interests, means for formulating and managing the user's learning plan, means for providing an individualized learning plan, and means for delivering the generated learning content to the user's terminal. As a result, the user can receive individualized learning support in real time and learn efficiently.

[1161] A "generative model" is a machine learning algorithm that automatically generates appropriate responses or content based on user input.

[1162] "User learning progress" refers to information that indicates the progress and level of proficiency achieved by learners in accordance with their learning plan.

[1163] "Custom learning content" refers to learning materials that are individually generated based on the needs, interests, and progress of a specific user.

[1164] A "user learning plan" refers to the specific steps and schedule that a user will use to achieve their learning goals.

[1165] A "personalized learning plan" refers to a learning schedule optimized for each user's individual learning history, progress, and goals.

[1166] A "user terminal" is an electronic device used by a user to communicate with a server and receive learning content.

[1167] "Real-time" means responding to or processing user requests immediately.

[1168] An "online education platform" is a system that provides learning content and educational services via the internet.

[1169] A "learning support system" refers to a set of computer programs and hardware that provide functions to effectively support a user's learning.

[1170] The specific embodiments of this invention are described below. This invention is a learning support system that utilizes a generative model and operates with three main parties: a server, a terminal, and a user. Various hardware and software and their processing are described below.

[1171] server

[1172] The server implements the generative model, which forms the core of this system. The server loads and initializes the generative model upon startup. The server is implemented using a server-side framework such as Flask and starts a listener to respond to user requests. The server also manages user authentication information using JWT (JSON Web Token). Specifically, a server machine with a high-performance processor and memory is used.

[1173] Generative model

[1174] A generative model is a machine learning algorithm that generates responses or custom learning content based on user requests. For example, it uses generative AI models such as OpenAI GPT-4. The model is loaded when the server starts up, and when it receives a request, it generates text based on its content.

[1175] terminal

[1176] The terminal provides the user interface and transmits user input to the server. Users log in using electronic devices such as smartphones, tablets, or PCs and enter questions or requests. The entered information is sent to the server, and the server's response and generated learning content are displayed on the terminal. The terminal provides a user-friendly interface to ensure easy operation for the user.

[1177] User

[1178] Users are primarily students and educators who input questions and requests into the system. Users can request the creation of learning plans and progress through their studies while checking progress information.

[1179] Data processing

[1180] Authentication: When a user logs in, the device sends authentication information to the server. The server compares this information with the authentication database, and if authentication is successful, generates a JWT token and sends it to the device.

[1181] Response generation: When a user enters a question about a mathematical problem into the terminal, the server uses a generative model to generate a response to the question and sends it to the terminal.

[1182] Generating custom learning content: When a user requests learning content based on a specific theme or difficulty level, the server uses a generative model to generate custom learning content and sends it to the device.

[1183] Learning plan creation: When a user requests the generation of a learning plan, the server generates a learning plan using generative models and other algorithms based on the user's learning history and progress information, and sends it to the terminal.

[1184] Specific example

[1185] 1. If a user wants to learn the basics of Python programming

[1186] The user types "I want to learn the basics of Python programming" into their terminal and sends a request.

[1187] The server uses a generative model to generate custom learning content on the fundamentals of Python programming and sends it to the device.

[1188] The device displays the generated learning content to the user, and the user begins learning.

[1189] 2. When a user plans their studies for the next math exam.

[1190] The user enters the request, "Please create a study plan for my next math exam."

[1191] The server generates an appropriate learning plan using a generative model based on the user's learning history and progress data, and sends it to the terminal.

[1192] The device displays the generated learning plan to the user, and the user proceeds with learning according to that plan.

[1193] Example of a prompt

[1194] Custom learning content generation:

[1195] "Generate learning content on the fundamentals of Python programming. Includes examples of videos, quizzes, and explanatory articles."

[1196] Learning plan generation:

[1197] "Create a study plan for the next math exam. Include daily tasks and key points based on the user's current progress and past learning history."

[1198] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1199] Step 1:

[1200] The server loads and initializes the generative model upon startup. The input is the model's configuration file and parameters, and the output is the initialized generative model. Specifically, the server reads the model file from the specified directory, sets the necessary parameters, and prepares the generative model.

[1201] Step 2:

[1202] The terminal displays a login screen, and the user enters their authentication information (username, password). The input is the username and password, and the output is a request containing the authentication information. The terminal sends this authentication information to the server.

[1203] Step 3:

[1204] The server compares the received authentication information with the authentication database, and if authentication is successful, generates a JWT token and sends it to the terminal. The input is the authentication information, and the output is either a JWT token or an error message. Specifically, the server compares it with a hashed password, and if authentication is successful, generates a token.

[1205] Step 4:

[1206] The user enters and submits questions and requests for learning content generation on their device. The input is the user's request, and the output is the request data. Specifically, the device sends data to the server using the submit button in the user's input field.

[1207] Step 5:

[1208] The server uses a generative model to generate responses or custom training content based on the received request. The input is the request data, and the output is the generated response or content. Specifically, the server inputs a prompt into the generative model and retrieves the generated result from the model.

[1209] Step 6:

[1210] The server sends the generated response or content to the terminal. The input is the generated response or content, and the output is the data sent to the terminal. Specifically, the server sends data to the terminal as an HTTP response.

[1211] Step 7:

[1212] The terminal displays received responses and content to the user. Input is data sent from the server, and output is information displayed on the user interface. Specifically, the terminal uses HTML and GUI components to display the data appropriately.

[1213] Step 8:

[1214] The user progresses through the learning process using the displayed responses and content, and reports progress information from the device to the server. The input is the user's learning progress information, and the output is the progress data sent to the server. Specifically, the device provides a field for entering progress information, and the user sends the data to the server using a submit button.

[1215] Step 9:

[1216] The server adjusts the learning plan based on the received progress information and generates a new learning plan as needed. The input is the user's progress data, and the output is a new or adjusted learning plan. Specifically, the server updates the database and generates or adjusts the plan using a generative model.

[1217] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1218] The specific embodiments of the present invention are described below. This system is a learning support system that includes a generative model, a function for generating custom learning content based on students' learning progress and interests, a function for planning and managing learning plans, and an emotion engine that recognizes user emotions and adjusts responses. This system operates with three main parties: the server, the terminal, and the user.

[1219] Overall system flow

[1220] server

[1221] The server implements the generative model and sentiment engine, which are the core of this system. The server first loads the generative model, configures various settings, and prepares to accept requests. Based on the user input and sentiment data included in the request, it uses the generative model to generate responses and customized learning materials, and sends the results to the terminal. The server also maintains each student's learning progress information and handles the planning and management of learning plans.

[1222] terminal

[1223] The terminal provides the user interface and is responsible for sending user input (text and sentiment data) to the server. When a user logs in and submits a question or request, the terminal displays the response from the server, generated content, and sentiment-based adjustments. The terminal provides an appropriate interface to make it easier for the user to use.

[1224] User

[1225] Users are primarily students and teachers who input questions and requests into the system. They also request the creation of learning plans and monitor their progress as they learn. Furthermore, the user's emotions are recognized through the terminal's input interface and transmitted to the server.

[1226] Program processing

[1227] Loading and initializing the generative model

[1228] Upon startup, the server loads the generative model and sentiment engine, and initializes by reading the necessary parameters and configuration files. It allocates resources for the model to function correctly and starts a listener to receive requests.

[1229] User authentication and login

[1230] The terminal displays a login screen to the user. The user enters their authentication information (username, password) and presses the submit button. The terminal sends this authentication information to the server. The server compares the received authentication information with its database, and if authentication is successful, generates a JWT token and sends it back to the terminal. If authentication fails, it sends an appropriate error message back to the terminal.

[1231] Question entry and submission of sentiment data

[1232] The user enters text about their question or problem into the terminal's question input screen and presses the submit button. The terminal sends the entered question, along with emotional data obtained from the user's facial recognition and voice tone analysis, to the server.

[1233] Question analysis and response generation

[1234] The server analyzes the received question and provides input to the generative model, taking into account the sentiment data recognized by the sentiment engine. The generative model generates a response based on the question and sentiment data and returns the result to the server.

[1235] Sending the generated response

[1236] The server sends the response returned from the generative model to the terminal after making appropriate adjustments based on the user's sentiment. The terminal displays the received response to the user. The user reviews the displayed response and enters additional questions if necessary.

[1237] Request for custom learning content

[1238] The user requests the generation of custom learning content based on a specific learning theme or difficulty level, and enters this request into their device. Along with this request, the device sends the user's sentiment data to the server.

[1239] Generating custom content

[1240] The server receives a custom learning content generation request and uses a generative model to generate learning content based on the request. It constructs a response containing the generated content, adjusts the difficulty and tone based on the user's sentiment data, and sends it to the device.

[1241] Displaying generated content

[1242] The device receives the generated learning content and displays it to the user. The user then uses the displayed content to continue learning.

[1243] Planning a study schedule

[1244] The user requests assistance with exam preparation or the creation of a study plan for specific learning objectives. The device sends this request to the server.

[1245] Generating and submitting a study plan

[1246] The server generates an appropriate learning plan using a generative model and sentiment engine based on the user's learning history and progress information. The generated learning plan is then sent to the device.

[1247] Displaying and implementing the study plan

[1248] The device receives the generated learning plan and displays it to the user. The user proceeds with their studies based on the displayed learning plan and inputs their progress into the device as needed.

[1249] Progress management and feedback

[1250] The server receives progress information and sentiment data sent by the user and manages the progress of the learning plan. If necessary, it generates feedback using a generative model and sends it to the device. The device displays the feedback to the user, who can then use it to improve their next learning session.

[1251] Specific example

[1252] 1. When using sentiment data when users ask questions about mathematical formulas.

[1253] The user enters "Please tell me the quadratic formula" into the question input screen on their device and submits it.

[1254] Along with this question, the device sends the user's facial expression recognition results (e.g., confused expression) and voice tone analysis results to the server.

[1255] The server uses a generative model to generate responses to questions, taking into account feelings of confusion and producing responses such as "You should explain with more specific examples."

[1256] The device displays this response to the user, who then uses the explanation for learning.

[1257] 2. When sentiment data is used when users request custom learning content.

[1258] The user requests, "Please create a workbook of problems on the basics of vectors."

[1259] Along with this request, the device sends the user's sentiment data to the server.

[1260] The server uses a generative model to generate a set of problems on the fundamentals of vectors, and adjusts the difficulty level if the user's motivation to learn is high.

[1261] The device displays the generated problem sets to the user, who then uses them to progress with their studies.

[1262] 3. When users utilize sentiment data when developing learning plans.

[1263] The user requests, "Please create a study plan for the next math exam," in order to develop a study plan for the next exam.

[1264] Along with this request, the device sends the user's sentiment data to the server.

[1265] The server uses generative models and a sentiment engine to generate an appropriate learning plan based on the user's learning history, progress data, and sentiment data.

[1266] For example, if a user is experiencing stress, the server will adjust its schedule accordingly.

[1267] The device displays the generated learning plan to the user, and the user proceeds with learning according to that plan.

[1268] This invention provides a system that uses a generative model and an emotion engine to provide personalized learning support that takes into account the user's emotions. This system allows users to learn more effectively and with less stress.

[1269] The following describes the processing flow.

[1270] Step 1: Load and initialize the generative model

[1271] Upon startup, the server loads the generative model and sentiment engine, and initializes by reading the necessary parameters and configuration files. It allocates resources for the model to function correctly and starts a listener to receive requests.

[1272] Step 2: User Authentication and Login

[1273] The terminal displays a login screen to the user. The user enters their authentication information (username, password) and presses the submit button. The terminal sends this authentication information to the server. The server compares the received authentication information with its database, and if authentication is successful, generates a JWT token and sends it back to the terminal. If authentication fails, it sends an appropriate error message back to the terminal.

[1274] Step 3: Enter the questions and submit sentiment data.

[1275] The user enters text about their question or problem into the terminal's question input screen and presses the submit button. The terminal sends the entered question, along with emotional data obtained from the user's facial recognition and voice tone analysis, to the server.

[1276] Step 4: Question analysis and response generation

[1277] The server analyzes the received question and provides input to the generative model, taking into account the sentiment data recognized by the sentiment engine. The generative model generates a response based on the question and sentiment data and returns the result to the server.

[1278] Step 5: Send the generated response

[1279] The server sends the response returned from the generative model to the terminal after making appropriate adjustments based on the user's sentiment. The terminal displays the received response to the user. The user reviews the displayed response and enters additional questions if necessary.

[1280] Step 6: Request custom learning content

[1281] The user requests the generation of custom learning content based on a specific learning theme or difficulty level, and enters this request into their device. Along with this request, the device sends the user's sentiment data to the server.

[1282] Step 7: Generating Custom Content

[1283] The server receives a custom learning content generation request and uses a generative model to generate learning content based on the request. It constructs a response containing the generated content, adjusts the difficulty and tone based on the user's sentiment data, and sends it to the device.

[1284] Step 8: Displaying the generated content

[1285] The device receives the generated learning content and displays it to the user. The user then uses the displayed content to continue learning.

[1286] Step 9: Develop a study plan

[1287] The user requests assistance with exam preparation or the creation of a study plan for specific learning objectives. The device sends this request to the server.

[1288] Step 10: Generate and submit your study plan.

[1289] The server generates an appropriate learning plan using a generative model and sentiment engine based on the user's learning history and progress information. The generated learning plan is then sent to the device.

[1290] Step 11: Display and implement your study plan

[1291] The device receives the generated learning plan and displays it to the user. The user proceeds with their studies based on the displayed learning plan and inputs their progress into the device as needed.

[1292] Step 12: Progress Management and Feedback

[1293] The server receives progress information and sentiment data sent by the user and manages the progress of the learning plan. If necessary, it generates feedback using a generative model and sends it to the device. The device displays the feedback to the user, who can then use it to improve their next learning session.

[1294] Specific example

[1295] 1. When a user asks a question about a mathematical formula

[1296] The user enters "Please tell me the quadratic formula" into the question input screen on their device and submits it.

[1297] Along with this question, the device sends the user's facial expression recognition results (e.g., confused expression) and voice tone analysis results to the server.

[1298] The server uses a generative model to generate responses to questions, taking into account feelings of confusion and producing responses such as "You should explain with more specific examples."

[1299] The device displays this response to the user, who then uses the explanation for learning.

[1300] 2. When a user requests custom learning content

[1301] The user requests, "Please create a workbook of problems on the basics of vectors."

[1302] Along with this request, the device sends the user's sentiment data to the server.

[1303] The server uses a generative model to generate a set of problems on the fundamentals of vectors, and adjusts the difficulty level if the user's motivation to learn is high.

[1304] The device displays the generated problem sets to the user, who then uses them to progress with their studies.

[1305] 3. When the user creates a learning plan

[1306] The user requests, "Please create a study plan for the next math exam," in order to develop a study plan for the next exam.

[1307] Along with this request, the device sends the user's sentiment data to the server.

[1308] The server uses generative models and a sentiment engine to generate an appropriate learning plan based on the user's learning history, progress data, and sentiment data.

[1309] For example, if a user is experiencing stress, the server will adjust its schedule accordingly.

[1310] The device displays the generated learning plan to the user, and the user proceeds with learning according to that plan.

[1311] (Example 2)

[1312] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1313] Traditional learning support systems provided features such as generating custom learning content based on individual users' learning progress and interests, developing learning plans, and managing progress. However, they lacked the ability to adjust responses based on user emotions. As a result, it was difficult to provide optimal learning support tailored to the user's emotional state, leading to challenges in maintaining learning effectiveness and motivation.

[1314] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1315] In this invention, the server includes means for generating responses to user input using a generative model, means for generating custom learning content based on the student's learning progress and interests, means for formulating a user's learning plan and managing its progress, means for recognizing the user's emotional data and adjusting the response, means for performing user authentication and matching authentication information with a database, means for adjusting the generated response based on the user's emotions and transmitting it to the terminal, and means for generating a learning plan and managing its progress based on the user's progress information and emotional data. This enables personalized learning support that takes the user's emotions into consideration.

[1316] A "generative model" is a pre-trained algorithm that generates natural language responses or content based on input data.

[1317] A "user" is a person, such as a student or teacher, who uses the system and is the entity that makes questions or learning requests through the learning support system.

[1318] "Custom learning content" refers to educational content such as learning materials and workbooks that are created individually based on a student's learning progress and interests.

[1319] A "learning plan" is a schedule or plan built based on the learning goals set by the user, and it also includes the management of progress.

[1320] "Emotional data" refers to data that indicates the user's current emotional state, obtained from user facial recognition and voice tone analysis.

[1321] A "response" refers to the natural language response or explanation that a generative model outputs in response to a user's question or request.

[1322] "User authentication" is the process of verifying the identity of users accessing a system, and it uses authentication information such as usernames and passwords.

[1323] A "terminal" is a device used by a user to interact with a system, and includes personal computers, tablets, smartphones, and other similar devices.

[1324] "Progress information" refers to data that shows the user's progress in their learning, and is recorded and managed through an online system.

[1325] A "database" is a digital repository for storing user authentication information, learning progress information, generated learning content, and other similar data.

[1326] A "listener" refers to the state in which a server is waiting for requests from external sources and is the process of accepting user input.

[1327] Specific embodiments of the present invention are described below. This learning support system is broadly composed of three components: a server, a terminal, and a user. The role of each component will be described in detail below.

[1328] Server configuration and operation

[1329] The server plays a central role in this system. It implements the following main functions:

[1330] 1. Loading and initializing the generative model:

[1331] When the server starts up, it loads pre-trained generative models and sentiment engines. This uses libraries such as Python and TensorFlow.

[1332] It reads the necessary parameters and configuration files, allocates resources, and starts a listener to wait for requests.

[1333] 2. User Authentication:

[1334] The server receives user authentication information (username, password) sent from the terminal and compares it with the database.

[1335] If authentication is successful, a JWT token is generated and sent back to the device; if authentication fails, an error message is sent back.

[1336] 3. Question analysis and response generation:

[1337] The server receives the question text and sentiment data sent by the user.

[1338] The question text and sentiment data are given to a generative model as input, and an appropriate response is generated. This response is adjusted considering the question content and sentiment data.

[1339] 4. Generating custom learning content:

[1340] The server generates custom learning content based on user requests and sentiment data.

[1341] The generated content is adjusted in difficulty and tone based on the user's learning motivation and emotional state.

[1342] 5. Planning and managing study progress:

[1343] The server generates an appropriate learning plan based on the user's learning history, progress data, and sentiment data.

[1344] The generated learning plan is sent to the device, and feedback and adjustments are made based on the user's progress.

[1345] Terminal configuration and operation

[1346] A terminal is a device that provides a user interface and performs the following functions:

[1347] 1. Displaying the login screen and sending user authentication information:

[1348] The device displays the login screen and sends the authentication information entered by the user to the server.

[1349] 2. Question entry and submission of sentiment data:

[1350] The user enters text into the question input screen on the device and presses the submit button. At the same time, emotional data is sent to the server based on facial recognition and voice tone analysis.

[1351] 3. Display of response from the server:

[1352] The terminal receives a response from the server and displays it to the user. The user reviews the displayed response and enters additional questions as needed.

[1353] 4. Displaying custom learning content:

[1354] The device displays custom learning content received from the server, enabling users to efficiently progress in their learning.

[1355] 5. Displaying and implementing the study plan:

[1356] The device receives the learning plan sent from the server and displays it to the user. The user then proceeds with their learning based on this learning plan.

[1357] User roles and actions

[1358] The user is the subject who uses the system to progress through the learning process, and takes the following actions:

[1359] 1. Enter your question or request:

[1360] Users input learning-related questions, requests for custom learning content, and learning plans into the system.

[1361] 2. Implementing the study plan and inputting progress:

[1362] The user progresses through the learning process based on the provided learning plan and enters progress information into their device.

[1363] 3. Use of learning content:

[1364] Students use the generated custom learning content to progress through their studies.

[1365] Specific example

[1366] If a user asks a question about a mathematical formula:

[1367] The user types the question "Please tell me the quadratic formula" into their device and sends it.

[1368] The facial recognition results (confused expression) and voice tone analysis results are also sent to the server at the same time.

[1369] The server uses a generative model to generate responses to questions, taking into account feelings of confusion and producing responses such as "You should explain with specific examples."

[1370] The terminal displays this response to the user.

[1371] When a user requests custom learning content:

[1372] I requested that you create a workbook of problems on the fundamentals of vectors.

[1373] Along with this request, the device sends the user's sentiment data to the server.

[1374] The server uses a generative model to generate a set of problems based on the request, and increases the difficulty level if the user's motivation is high.

[1375] The terminal displays the generated problem set to the user.

[1376] When a user creates a learning plan:

[1377] "Please create a study plan that will be necessary before the next math exam," I requested.

[1378] Along with this request, the device sends the user's sentiment data to the server.

[1379] The server uses generative models and an emotion engine to generate an appropriate learning plan based on the user's learning history, progress data, and sentiment data.

[1380] For example, if a user is experiencing stress, the server will adjust its schedule accordingly.

[1381] The device displays the generated learning plan to the user, and the user proceeds with learning according to that plan.

[1382] Example of a prompt:

[1383] "Please create a study plan for math and English for the next semester."

[1384] "If the user is confused by the tone of voice, please generate an easy-to-understand explanation."

[1385] "Please generate a set of problems on the fundamentals of vectors and set the difficulty level to intermediate."

[1386] In this way, a personalized learning support system using generative models and emotion engines can be realized.

[1387] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1388] Step 1: Load and initialize the generative model

[1389] When the server starts up, it loads a pre-trained generative model and sentiment engine. Specifically, it creates instances of the generative model using libraries such as Python and TensorFlow, and then loads the pre-trained model.

[1390] Input: Generative models and emotion engines, and their configuration files.

[1391] Data processing and calculations: Model loading and resource allocation.

[1392] Output: Start of listener to accept requests.

[1393] Step 2: User Authentication

[1394] The device displays a login screen, and the user enters their username and password and presses the submit button.

[1395] Enter: Username and password.

[1396] Data processing and calculations: Send authentication information to the server.

[1397] The server then compares this authentication information with the database.

[1398] Input: Authentication information.

[1399] Data processing and calculations: Database matching.

[1400] Output: If authentication is successful, a JWT token is generated and sent back to the device. If authentication fails, an error message is returned.

[1401] Step 3: Enter the questions and submit sentiment data.

[1402] The user enters text about their question or problem into the question input screen on their device and presses the submit button (e.g., "Please tell me the quadratic formula").

[1403] Input: Question text.

[1404] Data processing and computation: facial expression recognition and voice tone analysis.

[1405] The terminal sends the entered question along with emotional data obtained from facial recognition results and voice tone analysis to the server.

[1406] Input: Question text and sentiment data.

[1407] Data processing and calculations: Data transmission.

[1408] Output: Sending data to the server.

[1409] Step 4: Question analysis and response generation

[1410] The server analyzes the received question text and sentiment data.

[1411] Input: Question text and sentiment data.

[1412] Data processing and computation: Input to generative models.

[1413] Generating response text that takes emotional data into account (e.g., recognizing the emotion of confusion generates a response that includes a detailed explanation).

[1414] Output: The generated response text.

[1415] Step 5: Send the generated response

[1416] The server sends the generated response to the terminal after appropriately adjusting its emotional tone.

[1417] Input: The generated response text.

[1418] Data processing and calculation: Adjustments based on emotional data.

[1419] Output: Sends the adjusted response text to the terminal.

[1420] Step 6: Display and confirm the response

[1421] The terminal displays the received response to the user.

[1422] Input: Adjusted response text.

[1423] Data processing and calculations: Display processing.

[1424] Output: Display the response to the user.

[1425] Step 7: Request custom learning content

[1426] Users can request the creation of custom learning content based on specific learning themes and difficulty levels (e.g., "Please create a set of problems on the basics of vectors").

[1427] Input: Request for custom learning content.

[1428] Data processing and computation: facial expression recognition and voice tone analysis.

[1429] The device sends emotion data to the server along with the request.

[1430] Input: Request and sentiment data.

[1431] Data processing and calculations: Data transmission.

[1432] Output: Sending requests and sentiment data to the server.

[1433] Step 8: Generating Custom Content

[1434] The server receives a custom training content generation request and uses a generative model to generate the training content.

[1435] Input: Request and sentiment data.

[1436] Data processing and computation: Generating custom content using generative models.

[1437] The difficulty level and tone are adjusted based on emotional data.

[1438] Output: Generated training content.

[1439] Step 9: Displaying the generated content

[1440] The device receives the generated learning content and displays it to the user.

[1441] Input: Generated learning content.

[1442] Data processing and calculations: Display processing.

[1443] Output: Display learning content to the user.

[1444] Step 10: Develop a study plan

[1445] The user requests the creation of a study plan (e.g., "Please create a study plan for the next math exam").

[1446] Input: Request for a study plan.

[1447] Data processing and computation: facial expression recognition and voice tone analysis.

[1448] The device sends emotion data to the server along with the request.

[1449] Input: Request and sentiment data.

[1450] Data processing and calculations: Data transmission.

[1451] Output: Sending a training plan request and sentiment data to the server.

[1452] Step 11: Generate and submit your study plan.

[1453] The server uses generative models and a sentiment engine to generate an appropriate learning plan based on the user's learning history, progress data, and sentiment data.

[1454] Inputs: Learning history, progress data, sentiment data.

[1455] Data processing and computation: Generating learning plans using generative models.

[1456] Adjusting schedules based on emotional data.

[1457] Output: The generated training plan.

[1458] Step 12: Display and implement your study plan

[1459] The device receives the generated learning plan and displays it to the user.

[1460] Input: The generated training plan.

[1461] Data processing and calculations: Display processing.

[1462] Output: Displays the learning plan to the user.

[1463] Step 13: Progress Management and Feedback

[1464] The user progresses through the learning process according to the learning plan and enters progress information into the device.

[1465] Input: Progress information.

[1466] Data processing and calculations: Data transmission.

[1467] The server receives progress information and sentiment data sent by the user and manages the progress of the learning plan.

[1468] Input: Progress information and sentiment data.

[1469] Data processing and computation: progress management and feedback generation.

[1470] The device displays feedback to the user, which can be used to improve future learning.

[1471] Output: Feedback to the user.

[1472] (Application Example 2)

[1473] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1474] Conventional learning support systems have difficulty generating customized learning content tailored to each student's individual learning progress and interests, and they also lack the ability to provide responses that take into account students' emotional states. As a result, it has been difficult for students to properly plan their studies, manage their progress, and learn effectively, leading to a decline in the quality of learning.

[1475] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1476] In this invention, the server includes means for generating responses to user input using a generative model, means for generating custom learning content based on the learner's learning progress and interests, means for formulating a user's learning plan and managing its progress, and means for recognizing the user's emotions and adjusting responses and learning content based on that data. This enables personalized learning support for each individual learner, taking into account their learning progress and emotions.

[1477] A "generative model" is an algorithm that generates appropriate responses or content based on user input.

[1478] "Customized learning content" refers to learning materials and exercises tailored to each learner's individual learning progress and interests.

[1479] A "learning plan" is a schedule and set of action guidelines created to help users effectively progress towards their learning goals.

[1480] "Progress management" is a method for checking the results and progress of learners' learning activities and making adjustments as needed.

[1481] An "emotion engine" is an algorithm that recognizes a user's emotions and adjusts responses and content based on that data.

[1482] "User" refers to individuals who use this system, such as learners and teachers.

[1483] A "server" is a computer system that handles the core functions of a system and implements generative models, emotion engines, and other related components.

[1484] A "terminal" is a device that provides a user interface and transmits user input to a server.

[1485] "Interface" refers to the user interface or input method that allows a user to interact with a system.

[1486] The learning support system of the present invention consists of a server, a terminal, and a user. This system includes a generative model, the generation of custom learning content based on learning progress and interests, the planning and progress management of learning plans, and an emotion engine.

[1487] The server is the core of the entire system, implementing the generative model and sentiment engine. The server first loads the generative model, configures various settings, and prepares to receive requests. Requests include user input and sentiment data, and based on this, the server uses the generative model to generate responses and customized learning materials. The generated results are sent to the terminal. The server also maintains each student's learning progress information and handles the planning and management of learning progress.

[1488] The terminal provides the user interface and transmits user input (text and sentiment data) to the server. Users log in to the terminal and enter questions or requests. Responses from the server, generated content, and sentiment-based adjustments are displayed on the terminal. The terminal provides an appropriate interface to make it easier for users to use.

[1489] Users are primarily learners and instructors who input questions and requests into the system. They also request the creation of learning plans and monitor their progress as they learn. Furthermore, the user's emotions are recognized through the terminal's input interface and transmitted to the server.

[1490] Program Processing Description

[1491] The server loads and initializes the generative model. When the server starts, it loads the generative model and sentiment engine, and initializes the necessary parameters and configuration files. This allocates resources for the model to function correctly and starts a listener to accept requests.

[1492] The terminal supports user authentication and login. The terminal displays a login screen to the user, who enters their authentication information (username, password). The authentication information is sent from the terminal to the server, which compares the received information with its database. If authentication is successful, a JWT token is generated and sent back to the terminal.

[1493] Furthermore, when a user enters a question or requests custom learning content, the device sends sentiment data along with the content to the server. The server analyzes the received questions and requests and provides input to the generative model based on the sentiment data recognized by the sentiment engine. The generative model generates responses and custom learning content based on the questions and sentiment data, and returns the results to the server. The server then sends this to the device for display to the user.

[1494] Hardware and software used

[1495] Server: A computer system that performs the core functions of a system (e.g., "Nginx", "Docker").

[1496] Generative models and sentiment engines: Algorithms and libraries (e.g., TensorFlow, PyTorch)

[1497] Device: A device that provides a user interface (e.g., "iOS device", "Android device")

[1498] Database: Stores user information and learning progress (e.g., "PostgreSQL").

[1499] Examples of specific cases and prompt statements

[1500] As a concrete example, when a user asks a question about the equations of motion in physics, a fearful expression is detected through facial recognition. This information is sent from the terminal to the server, which uses a generative model to provide a detailed and easy-to-understand answer.

[1501] Example prompt: "Please answer the following question considering the sentiment data: The user has requested an explanation of the equations of motion in physics. The sentiment data indicates 'fear'."

[1502] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1503] Step 1:

[1504] The server loads and initializes the generative model and sentiment engine. Specifically, at startup, the server loads the generative model and sentiment engine into memory and reads the necessary parameters and configuration files. This ensures that the generative model and sentiment engine have the resources to function correctly. It also starts a listener to receive requests. The input is the configuration files and parameters of the generative model and sentiment engine, and the output is the initialized generative model and sentiment engine.

[1505] Step 2:

[1506] The terminal receives the user's login information and sends it to the server. Specifically, the user enters their username and password on the terminal's login screen and presses the submit button. The terminal sends this authentication information to the server. The input is the user's login information (username, password), and the output is the authentication request sent to the server.

[1507] Step 3:

[1508] The server authenticates the login information. Specifically, the server compares the received authentication information with the database, and if successful, generates a JWT token and sends it back to the terminal. If it fails, it sends an appropriate error message back to the terminal. The input is the authentication information received from the terminal, and the output is a JWT token if authentication is successful, and an error message if it fails.

[1509] Step 4:

[1510] The device collects user questions and sentiment data and sends them to the server. Specifically, the user enters text about their questions or problems into the question input screen and presses the submit button. The device simultaneously sends sentiment data acquired from the camera and voice input to the server. The input consists of the user's question text and sentiment data, and the output is the request sent to the server.

[1511] Step 5:

[1512] The server receives questions and sentiment data, analyzes them, and generates responses. Specifically, the server analyzes the received questions and inputs the recognized sentiment data into a generative model. The generative model generates appropriate responses based on the questions and sentiment data. The input is the questions and sentiment data received from the terminal, and the output is the generated response data.

[1513] Step 6:

[1514] The server adjusts the generated response based on sentiment data and sends it to the terminal. Specifically, the server adjusts the tone and level of detail of the response returned from the generative model according to the user's emotions, and then sends it to the terminal. The input is the generated response data and sentiment data, and the output is the adjusted response.

[1515] Step 7:

[1516] The terminal displays the received response to the user. Specifically, the terminal displays the response sent from the server on the screen, allowing the user to confirm the response. The input is the adjusted response received from the server, and the output is the response displayed on the screen.

[1517] Step 8:

[1518] Users request custom learning content. Specifically, users input a request on their device screen to generate custom learning content based on a specific learning theme and difficulty level. The input is the user's request, and the output is the request sent from the device to the server.

[1519] Step 9:

[1520] The server generates custom learning content and adjusts it based on sentiment data. Specifically, the server uses a generative model to generate learning content based on requests and adjusts the difficulty and tone by referring to the user's sentiment data. The input is the user's request and sentiment data, and the output is the generated custom learning content.

[1521] Step 10:

[1522] The device displays the generated learning content to the user. Specifically, it receives the generated content from the server and displays it on the screen. The input is the generated learning content, and the output is the learning content displayed to the user.

[1523] Example of a prompt

[1524] "Please answer the following question considering the emotion data: The user has asked for an explanation of the equations of motion in physics. The emotion data indicates 'fear'."

[1525] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1526] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1527] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1528] [Fourth Embodiment]

[1529] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1530] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1531] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1532] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1533] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1534] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1535] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1536] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1537] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1538] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1539] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1540] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1541] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1542] The specific embodiments of the present invention are described below. This system is a learning support system that includes a generative model, a function for generating custom learning content based on students' learning progress and interests, and a function for planning and managing the progress of learning plans. This system operates with three main parties: the server, the terminal, and the user.

[1543] Overall system flow

[1544] server

[1545] The server implements the generative model, which is the core of this system. The server first loads the generative model, configures various settings, and prepares to receive requests. Based on requests sent from users and terminals, the server uses the generative model to generate responses and customized learning materials, and sends the results to the terminals. The server also maintains information on each student's learning progress and handles the planning and management of learning plans.

[1546] terminal

[1547] The terminal provides the user interface and is responsible for sending user input to the server. When a user logs in and submits questions or requests, the terminal displays responses and generated content from the server that received them. The terminal provides an appropriate interface to make it easier for the user to use.

[1548] User

[1549] Users are primarily students and teachers who input questions and requests into the system. They also request the creation of learning plans and monitor their progress as they learn.

[1550] Program processing

[1551] Loading and initializing the generative model

[1552] The server loads and initializes the generated model upon startup. It reads the model's parameters and configuration files, preparing it to function correctly. It starts a listener to await requests and waits for connections from clients.

[1553] User authentication and login

[1554] The device displays a login screen, and the user enters their authentication information (username and password). The device sends this information to the server, which then authenticates the user by comparing it against the authentication database. If authentication is successful, a token is generated and sent to the device. If authentication fails, an error message is returned.

[1555] Use as a supplementary tool for individualized instruction

[1556] The user inputs a question about a mathematical problem, and the terminal sends it to the server. The server uses a generative model to generate a response to the question and sends it back to the terminal. The terminal then displays the received response to the user.

[1557] Generating custom learning content

[1558] The user requests the generation of learning content based on a specific theme or difficulty level. The device sends this request to the server. The server uses a generative model to generate custom learning content and sends it to the device. This content is displayed on the device, and the user uses it for their learning.

[1559] Planning and managing study progress

[1560] The user submits a request for a learning plan and sends it to the server via their device. The server generates a learning plan using generative models and other algorithms based on the user's learning history and progress information, and sends it to the device. The user proceeds with their learning based on this plan and reports their progress to the server via their device. The server manages the progress and adjusts or improves the learning plan as needed.

[1561] Specific example

[1562] 1. When a user asks a question about a mathematical formula

[1563] The user enters "Please tell me the quadratic formula" into the question input screen on their device and submits it.

[1564] The device sends this question to the server.

[1565] The server uses a generative model to generate an explanation of the quadratic formula and produces a response such as, "The quadratic formula for solving the quadratic equation ax^2 + bx + c = 0 is..."

[1566] The device displays this response to the user, who then uses the explanation for learning.

[1567] 2. When a user requests custom learning content

[1568] The user enters a request saying, "Please create a workbook of problems on the basics of vectors."

[1569] The device sends this request to the server.

[1570] The server uses a generative model to generate a set of problems on the fundamentals of vectors and sends them to the terminal.

[1571] The user uses the problem set to progress with their studies.

[1572] 3. When the user creates a learning plan

[1573] The user requests, "Please create a study plan for the next math exam," in order to develop a study plan for the next exam.

[1574] The device sends this request to the server.

[1575] The server uses a generative model to generate an appropriate learning plan based on the user's learning history and progress data, and sends it to the terminal.

[1576] The device displays a learning plan to the user, and the user proceeds with their studies according to that plan.

[1577] As described above, the present invention provides a system that provides efficient and personalized learning support by using a generative model. This system allows users to receive high-quality learning support and improve their learning outcomes.

[1578] The following describes the processing flow.

[1579] Step 1: Load and initialize the generative model

[1580] The server loads the generated model upon startup, reads the necessary parameters and configuration files, and initializes it. It allocates resources for the model to function correctly and starts a listener to receive requests.

[1581] Step 2: User Authentication and Login

[1582] The terminal displays a login screen to the user. The user enters their authentication information (username, password) and presses the submit button. The terminal sends this authentication information to the server. The server compares the received authentication information with its database, and if authentication is successful, generates a JWT token and sends it back to the terminal. If authentication fails, it sends an appropriate error message back to the terminal.

[1583] Step 3: Enter Questions

[1584] The user enters text about their question or problem into the question input screen on the device and presses the submit button. The device then sends the entered question to the server.

[1585] Step 4: Question analysis and response generation

[1586] The server analyzes the received question and provides input to the generative model. The generative model generates a response to the question and returns the result to the server.

[1587] Step 5: Send the generated response

[1588] The server sends the response returned from the generative model to the terminal. The terminal displays the received response to the user. The user reviews the displayed response and enters additional questions as needed.

[1589] Step 6: Request custom learning content

[1590] The user wishes to generate custom learning content based on a specific learning theme or difficulty level, and enters a request into their device. The device then sends this request to the server.

[1591] Step 7: Generating Custom Content

[1592] The server receives a custom learning content generation request and uses a generative model to generate learning content based on the request. It then constructs a response containing the generated content and sends it to the terminal.

[1593] Step 8: Displaying the generated content

[1594] The device receives the generated learning content and displays it to the user. The user then uses the displayed content to continue learning.

[1595] Step 9: Develop a study plan

[1596] The user requests assistance with exam preparation or the creation of a study plan for specific learning objectives. The device sends this request to the server.

[1597] Step 10: Generate and submit your study plan.

[1598] The server generates an appropriate learning plan using generative models and other algorithms based on the user's learning history and progress information. It then sends the generated learning plan to the user's device.

[1599] Step 11: Display and implement your study plan

[1600] The device receives the generated learning plan and displays it to the user. The user proceeds with their studies based on the displayed learning plan and inputs their progress into the device as needed.

[1601] Step 12: Progress Management and Feedback

[1602] The server receives progress information sent by the user and manages the progress of the learning plan. If necessary, it generates feedback using a generative model and sends it to the terminal. The terminal displays the feedback to the user and helps them with their next learning session.

[1603] (Example 1)

[1604] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1605] Conventional learning support systems struggle to individually address the diverse learning needs of users. Furthermore, they fail to effectively manage learning progress, creating a need for efficient means to support planned learning. Additionally, the security of user authentication and the automation of on-demand learning content generation using generative models are insufficient, posing challenges to improving the quality and efficiency of learning.

[1606] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1607] In this invention, the server includes means for generating a response to user input using a generative model, means for the learning support server to read the generative model, perform various settings, and await requests, means for receiving user authentication information and issuing authentication tokens, means for generating custom learning content using the generative model in response to a learning content generation request from the terminal and sending it to the terminal, and means for managing student learning progress information, formulating a learning plan based on that information, and managing progress. This enables responses to individual learning needs, resulting in planned and effective learning support, secure user authentication, and on-demand generation and delivery of learning content.

[1608] A "generative model" is a system that generates text or responses using neural networks or machine learning algorithms.

[1609] "Responses to user input" refer to text-based answers and explanations generated based on questions and requests received from the user.

[1610] A "learning support server" is the core of a learning support system, handling tasks such as loading generative models, configuring various settings, processing requests, generating and sending learning content, and managing learning progress.

[1611] "Means for waiting for requests" refers to the function that allows a server to accept connections and data transmissions from users and terminals.

[1612] "User authentication information" refers to data such as usernames and passwords that are used to identify a user and grant them permissions.

[1613] An "authentication token" is a unique identifier issued to a user who has successfully authenticated, and is used to authenticate the user in subsequent communications.

[1614] A "learning content generation request" refers to a user requesting the creation of new learning materials or problem sets based on a specific theme or difficulty level.

[1615] "Custom learning content" refers to original learning materials and workbooks that are generated based on the user's requests and learning progress.

[1616] "Learning progress information" refers to data that records how far students have progressed in their studies.

[1617] A "study plan" is a document that outlines the learning content and progress schedule that students aim to achieve.

[1618] "Means of managing progress" refers to a function that tracks students' learning progress based on their learning plan and makes adjustments or improvements as needed.

[1619] "On-demand" refers to a service delivery method that provides services immediately in response to user requests.

[1620] This invention relates to a learning support system that includes a generative model, a learning support server, a function for generating custom learning content based on students' learning progress and interests, and a function for planning and managing learning plans. This invention is a system that operates primarily with three parties: the server, the terminal, and the user.

[1621] Server configuration and operation

[1622] The server implements the generative model (e.g., GPT-3 or BERT) that forms the core of this system. Upon startup, the server loads and initializes the generative model. During this process, it reads the model's parameters and configuration files (e.g., model_config.json and model_weights.h5) to prepare for correct operation. It also starts a listener to await requests and waits for connections from clients (terminals).

[1623] User Authentication

[1624] The server receives user authentication information (username, password) sent from the terminal and compares it with the authentication database. If authentication is successful, an authentication token is generated and sent to the terminal. If authentication fails, an error message is returned.

[1625] Learning content generation

[1626] The server responds to learning content generation requests from the terminal and generates custom learning content using a generative model. This content generation is based on user requests and individual learning progress information. The generated content is sent to the terminal and made available to the user.

[1627] Learning progress management and planning

[1628] The server manages each student's learning progress information and creates a learning plan based on it. In this process, generative models and other algorithms are used to analyze the user's learning history and progress data to generate an appropriate learning plan. The learning plan is sent to the user's device, where they can review and execute it.

[1629] Terminal configuration and operation

[1630] The terminal provides the user interface and is responsible for sending user input to the server. When a user logs in and enters questions or requests, the terminal displays the responses and generated content from the server that received them. The terminal provides a user-friendly interface, supporting smooth learning.

[1631] User actions

[1632] Users are primarily students and teachers, and they can input questions and requests into the system. For example, consider the following specific examples:

[1633] Questions about mathematical formulas

[1634] The user types "Please tell me the quadratic formula" into the terminal and sends it. The terminal sends this question to the server, which uses a generative model to generate a response and sends it back to the terminal. The terminal displays the generated response to the user.

[1635] Request for custom learning content

[1636] The user enters a request: "Please create a set of problems on the basics of vectors." The device sends this request to the server, which uses a generative model to generate custom learning content and sends it to the device. The user then uses the generated set of problems for learning.

[1637] Planning a study schedule

[1638] The user requests, "Please create a study plan for the next math exam." The device sends this request to the server, which uses a generative model to generate a study plan based on the user's learning history and progress data. The device displays this plan to the user, who then proceeds with their studies based on the plan.

[1639] Thus, the present invention uses a generative model to provide high-quality learning support and realize efficient and personalized learning.

[1640] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1641] Step 1:

[1642] The server starts up and loads the generative model. During this process, it reads the configuration file (e.g., model_config.json) and the model parameter file (e.g., model_weights.h5) and initializes the generative model. The server starts a listener to await requests and waits for connections and requests from clients. This process takes the server startup command and configuration files as input and outputs the state where the generative model initialization is complete.

[1643] Step 2:

[1644] The user accesses the login screen on their device and enters their username and password. The device sends this information to the server. The server compares it with the authentication database, and if authentication is successful, generates a token and sends it to the device. If authentication fails, it returns an error message. The input to this process is the user's authentication information, and the output is either an authentication token or an error message.

[1645] In terms of specific operations, the terminal receives user input and sends it to the server. On the server side, for example, an SQL query is executed to verify the user information (e.g., SELECT FROM users WHERE username = ? AND password_hash = ?).

[1646] Step 3:

[1647] The user enters "Please tell me the quadratic formula" on the terminal's question input screen and submits it. The terminal sends this question to the server. The server inputs the question as a prompt to the generative model, and the generative model generates a response. This generated response is sent to the terminal and displayed to the user. The input to this process is the user's question, and the output is the response generated by the generative model.

[1648] In terms of specific operation, the server inputs the response to the prompt "Please tell me the quadratic formula" into the generative model, and then sends the resulting response to the terminal.

[1649] Step 4:

[1650] A user requests and sends a request from their device saying, "Please create a set of problems on the basics of vectors." The device sends this request to the server. The server uses a generative model to generate custom learning content based on the theme. This generated content is sent to the device and made available to the user. The input to this process is the user's request to generate custom learning content, and the output is the generated custom learning content.

[1651] In terms of specific operation, the server inputs a request based on the prompt "Create a problem set on the basics of vectors" into the generative model, and then sends the resulting problem set to the terminal.

[1652] Step 5:

[1653] A user requests from their device, "Please create a study plan for the next math exam." The device sends this request to the server. The server retrieves the user's learning history and progress data from a database and generates a study plan using generative models and other algorithms. This study plan is sent to the device and displayed to the user. The input to this process is the user's study plan creation request and learning history data, and the output is the generated study plan.

[1654] In terms of specific operations, the server retrieves user progress data from the database, inputs it into a generative model to generate an appropriate learning plan, and then sends it to the terminal.

[1655] Step 6:

[1656] The user progresses through their learning based on a learning plan and reports their progress to the server from their terminal. The server stores this progress information in a database and adjusts or improves the learning plan as needed. The input to this process is the user's learning progress information, and the output is an updated learning plan and progress data.

[1657] In terms of specific operations, the user progresses through the learning plan, reports their progress via their device, the server records this in a database, and the learning plan is reviewed.

[1658] (Application Example 1)

[1659] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1660] Online education platforms require personalized learning support and an environment that enables users to learn efficiently. Furthermore, generating custom learning content in real time based on user progress and developing and managing learning plans based on individual needs are also crucial challenges. Current systems struggle to comprehensively provide these elements, resulting in insufficient support for maximizing learning effectiveness.

[1661] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1662] In this invention, the server includes means for generating responses to user input using a generative model, means for generating custom learning content based on the user's learning progress and interests, means for formulating and managing the user's learning plan, means for providing an individualized learning plan, and means for delivering the generated learning content to the user's terminal. As a result, the user can receive individualized learning support in real time and learn efficiently.

[1663] A "generative model" is a machine learning algorithm that automatically generates appropriate responses or content based on user input.

[1664] "User learning progress" refers to information that indicates the progress and level of proficiency achieved by learners in accordance with their learning plan.

[1665] "Custom learning content" refers to learning materials that are individually generated based on the needs, interests, and progress of a specific user.

[1666] A "user learning plan" refers to the specific steps and schedule that a user will use to achieve their learning goals.

[1667] A "personalized learning plan" refers to a learning schedule optimized for each user's individual learning history, progress, and goals.

[1668] A "user terminal" is an electronic device used by a user to communicate with a server and receive learning content.

[1669] "Real-time" means responding to or processing user requests immediately.

[1670] An "online education platform" is a system that provides learning content and educational services via the internet.

[1671] A "learning support system" refers to a set of computer programs and hardware that provide functions to effectively support a user's learning.

[1672] The specific embodiments of this invention are described below. This invention is a learning support system that utilizes a generative model and operates with three main parties: a server, a terminal, and a user. Various hardware and software and their processing are described below.

[1673] server

[1674] The server implements the generative model, which forms the core of this system. The server loads and initializes the generative model upon startup. The server is implemented using a server-side framework such as Flask and starts a listener to respond to user requests. The server also manages user authentication information using JWT (JSON Web Token). Specifically, a server machine with a high-performance processor and memory is used.

[1675] Generative model

[1676] A generative model is a machine learning algorithm that generates responses or custom learning content based on user requests. For example, it uses generative AI models such as OpenAI GPT-4. The model is loaded when the server starts up, and when it receives a request, it generates text based on its content.

[1677] terminal

[1678] The terminal provides the user interface and transmits user input to the server. Users log in using electronic devices such as smartphones, tablets, or PCs and enter questions or requests. The entered information is sent to the server, and the server's response and generated learning content are displayed on the terminal. The terminal provides a user-friendly interface to ensure easy operation for the user.

[1679] User

[1680] Users are primarily students and educators who input questions and requests into the system. Users can request the creation of learning plans and progress through their studies while checking progress information.

[1681] Data processing

[1682] Authentication: When a user logs in, the device sends authentication information to the server. The server compares this information with the authentication database, and if authentication is successful, generates a JWT token and sends it to the device.

[1683] Response generation: When a user enters a question about a mathematical problem into the terminal, the server uses a generative model to generate a response to the question and sends it to the terminal.

[1684] Generating custom learning content: When a user requests learning content based on a specific theme or difficulty level, the server uses a generative model to generate custom learning content and sends it to the device.

[1685] Learning plan creation: When a user requests the generation of a learning plan, the server generates a learning plan using generative models and other algorithms based on the user's learning history and progress information, and sends it to the terminal.

[1686] Specific example

[1687] 1. If a user wants to learn the basics of Python programming

[1688] The user types "I want to learn the basics of Python programming" into their terminal and sends a request.

[1689] The server uses a generative model to generate custom learning content on the fundamentals of Python programming and sends it to the device.

[1690] The device displays the generated learning content to the user, and the user begins learning.

[1691] 2. When a user plans their studies for the next math exam.

[1692] The user enters the request, "Please create a study plan for my next math exam."

[1693] The server generates an appropriate learning plan using a generative model based on the user's learning history and progress data, and sends it to the terminal.

[1694] The device displays the generated learning plan to the user, and the user proceeds with learning according to that plan.

[1695] Example of a prompt

[1696] Custom learning content generation:

[1697] "Generate learning content on the fundamentals of Python programming. Includes examples of videos, quizzes, and explanatory articles."

[1698] Learning plan generation:

[1699] "Create a study plan for the next math exam. Include daily tasks and key points based on the user's current progress and past learning history."

[1700] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1701] Step 1:

[1702] The server loads and initializes the generative model upon startup. The input is the model's configuration file and parameters, and the output is the initialized generative model. Specifically, the server reads the model file from the specified directory, sets the necessary parameters, and prepares the generative model.

[1703] Step 2:

[1704] The terminal displays a login screen, and the user enters their authentication information (username, password). The input is the username and password, and the output is a request containing the authentication information. The terminal sends this authentication information to the server.

[1705] Step 3:

[1706] The server compares the received authentication information with the authentication database, and if authentication is successful, generates a JWT token and sends it to the terminal. The input is the authentication information, and the output is either a JWT token or an error message. Specifically, the server compares it with a hashed password, and if authentication is successful, generates a token.

[1707] Step 4:

[1708] The user enters and submits questions and requests for learning content generation on their device. The input is the user's request, and the output is the request data. Specifically, the device sends data to the server using the submit button in the user's input field.

[1709] Step 5:

[1710] The server uses a generative model to generate responses or custom training content based on the received request. The input is the request data, and the output is the generated response or content. Specifically, the server inputs a prompt into the generative model and retrieves the generated result from the model.

[1711] Step 6:

[1712] The server sends the generated response or content to the terminal. The input is the generated response or content, and the output is the data sent to the terminal. Specifically, the server sends data to the terminal as an HTTP response.

[1713] Step 7:

[1714] The terminal displays received responses and content to the user. Input is data sent from the server, and output is information displayed on the user interface. Specifically, the terminal uses HTML and GUI components to display the data appropriately.

[1715] Step 8:

[1716] The user progresses through the learning process using the displayed responses and content, and reports progress information from the device to the server. The input is the user's learning progress information, and the output is the progress data sent to the server. Specifically, the device provides a field for entering progress information, and the user sends the data to the server using a submit button.

[1717] Step 9:

[1718] The server adjusts the learning plan based on the received progress information and generates a new learning plan as needed. The input is the user's progress data, and the output is a new or adjusted learning plan. Specifically, the server updates the database and generates or adjusts the plan using a generative model.

[1719] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1720] The specific embodiments of the present invention are described below. This system is a learning support system that includes a generative model, a function for generating custom learning content based on students' learning progress and interests, a function for planning and managing learning plans, and an emotion engine that recognizes user emotions and adjusts responses. This system operates with three main parties: the server, the terminal, and the user.

[1721] Overall system flow

[1722] server

[1723] The server implements the generative model and sentiment engine, which are the core of this system. The server first loads the generative model, configures various settings, and prepares to accept requests. Based on the user input and sentiment data included in the request, it uses the generative model to generate responses and customized learning materials, and sends the results to the terminal. The server also maintains each student's learning progress information and handles the planning and management of learning plans.

[1724] terminal

[1725] The terminal provides the user interface and is responsible for sending user input (text and sentiment data) to the server. When a user logs in and submits a question or request, the terminal displays the response from the server, generated content, and sentiment-based adjustments. The terminal provides an appropriate interface to make it easier for the user to use.

[1726] User

[1727] Users are primarily students and teachers who input questions and requests into the system. They also request the creation of learning plans and monitor their progress as they learn. Furthermore, the user's emotions are recognized through the terminal's input interface and transmitted to the server.

[1728] Program processing

[1729] Loading and initializing the generative model

[1730] Upon startup, the server loads the generative model and sentiment engine, and initializes by reading the necessary parameters and configuration files. It allocates resources for the model to function correctly and starts a listener to receive requests.

[1731] User authentication and login

[1732] The terminal displays a login screen to the user. The user enters their authentication information (username, password) and presses the submit button. The terminal sends this authentication information to the server. The server compares the received authentication information with its database, and if authentication is successful, generates a JWT token and sends it back to the terminal. If authentication fails, it sends an appropriate error message back to the terminal.

[1733] Question entry and submission of sentiment data

[1734] The user enters text about their question or problem into the terminal's question input screen and presses the submit button. The terminal sends the entered question, along with emotional data obtained from the user's facial recognition and voice tone analysis, to the server.

[1735] Question analysis and response generation

[1736] The server analyzes the received question and provides input to the generative model, taking into account the sentiment data recognized by the sentiment engine. The generative model generates a response based on the question and sentiment data and returns the result to the server.

[1737] Sending the generated response

[1738] The server sends the response returned from the generative model to the terminal after making appropriate adjustments based on the user's sentiment. The terminal displays the received response to the user. The user reviews the displayed response and enters additional questions if necessary.

[1739] Request for custom learning content

[1740] The user requests the generation of custom learning content based on a specific learning theme or difficulty level, and enters this request into their device. Along with this request, the device sends the user's sentiment data to the server.

[1741] Generating custom content

[1742] The server receives a custom learning content generation request and uses a generative model to generate learning content based on the request. It constructs a response containing the generated content, adjusts the difficulty and tone based on the user's sentiment data, and sends it to the device.

[1743] Displaying generated content

[1744] The device receives the generated learning content and displays it to the user. The user then uses the displayed content to continue learning.

[1745] Planning a study schedule

[1746] The user requests assistance with exam preparation or the creation of a study plan for specific learning objectives. The device sends this request to the server.

[1747] Generating and submitting a study plan

[1748] The server generates an appropriate learning plan using a generative model and sentiment engine based on the user's learning history and progress information. The generated learning plan is then sent to the device.

[1749] Displaying and implementing the study plan

[1750] The device receives the generated learning plan and displays it to the user. The user proceeds with their studies based on the displayed learning plan and inputs their progress into the device as needed.

[1751] Progress management and feedback

[1752] The server receives progress information and sentiment data sent by the user and manages the progress of the learning plan. If necessary, it generates feedback using a generative model and sends it to the device. The device displays the feedback to the user, who can then use it to improve their next learning session.

[1753] Specific example

[1754] 1. When using sentiment data when users ask questions about mathematical formulas.

[1755] The user enters "Please tell me the quadratic formula" into the question input screen on their device and submits it.

[1756] Along with this question, the device sends the user's facial expression recognition results (e.g., confused expression) and voice tone analysis results to the server.

[1757] The server uses a generative model to generate responses to questions, taking into account feelings of confusion and producing responses such as "You should explain with more specific examples."

[1758] The device displays this response to the user, who then uses the explanation for learning.

[1759] 2. When sentiment data is used when users request custom learning content.

[1760] The user requests, "Please create a workbook of problems on the basics of vectors."

[1761] Along with this request, the device sends the user's sentiment data to the server.

[1762] The server uses a generative model to generate a set of problems on the fundamentals of vectors, and adjusts the difficulty level if the user's motivation to learn is high.

[1763] The device displays the generated problem sets to the user, who then uses them to progress with their studies.

[1764] 3. When users utilize sentiment data when developing learning plans.

[1765] The user requests, "Please create a study plan for the next math exam," in order to develop a study plan for the next exam.

[1766] Along with this request, the device sends the user's sentiment data to the server.

[1767] The server uses generative models and a sentiment engine to generate an appropriate learning plan based on the user's learning history, progress data, and sentiment data.

[1768] For example, if a user is experiencing stress, the server will adjust its schedule accordingly.

[1769] The device displays the generated learning plan to the user, and the user proceeds with learning according to that plan.

[1770] This invention provides a system that uses a generative model and an emotion engine to provide personalized learning support that takes into account the user's emotions. This system allows users to learn more effectively and with less stress.

[1771] The following describes the processing flow.

[1772] Step 1: Load and initialize the generative model

[1773] Upon startup, the server loads the generative model and sentiment engine, and initializes by reading the necessary parameters and configuration files. It allocates resources for the model to function correctly and starts a listener to receive requests.

[1774] Step 2: User Authentication and Login

[1775] The terminal displays a login screen to the user. The user enters their authentication information (username, password) and presses the submit button. The terminal sends this authentication information to the server. The server compares the received authentication information with its database, and if authentication is successful, generates a JWT token and sends it back to the terminal. If authentication fails, it sends an appropriate error message back to the terminal.

[1776] Step 3: Enter the questions and submit sentiment data.

[1777] The user enters text about their question or problem into the terminal's question input screen and presses the submit button. The terminal sends the entered question, along with emotional data obtained from the user's facial recognition and voice tone analysis, to the server.

[1778] Step 4: Question analysis and response generation

[1779] The server analyzes the received question and provides input to the generative model, taking into account the sentiment data recognized by the sentiment engine. The generative model generates a response based on the question and sentiment data and returns the result to the server.

[1780] Step 5: Send the generated response

[1781] The server sends the response returned from the generative model to the terminal after making appropriate adjustments based on the user's sentiment. The terminal displays the received response to the user. The user reviews the displayed response and enters additional questions if necessary.

[1782] Step 6: Request custom learning content

[1783] The user requests the generation of custom learning content based on a specific learning theme or difficulty level, and enters this request into their device. Along with this request, the device sends the user's sentiment data to the server.

[1784] Step 7: Generating Custom Content

[1785] The server receives a custom learning content generation request and uses a generative model to generate learning content based on the request. It constructs a response containing the generated content, adjusts the difficulty and tone based on the user's sentiment data, and sends it to the device.

[1786] Step 8: Displaying the generated content

[1787] The device receives the generated learning content and displays it to the user. The user then uses the displayed content to continue learning.

[1788] Step 9: Develop a study plan

[1789] The user requests assistance with exam preparation or the creation of a study plan for specific learning objectives. The device sends this request to the server.

[1790] Step 10: Generate and submit your study plan.

[1791] The server generates an appropriate learning plan using a generative model and sentiment engine based on the user's learning history and progress information. The generated learning plan is then sent to the device.

[1792] Step 11: Display and implement your study plan

[1793] The device receives the generated learning plan and displays it to the user. The user proceeds with their studies based on the displayed learning plan and inputs their progress into the device as needed.

[1794] Step 12: Progress Management and Feedback

[1795] The server receives progress information and sentiment data sent by the user and manages the progress of the learning plan. If necessary, it generates feedback using a generative model and sends it to the device. The device displays the feedback to the user, who can then use it to improve their next learning session.

[1796] Specific example

[1797] 1. When a user asks a question about a mathematical formula

[1798] The user enters "Please tell me the quadratic formula" into the question input screen on their device and submits it.

[1799] Along with this question, the device sends the user's facial expression recognition results (e.g., confused expression) and voice tone analysis results to the server.

[1800] The server uses a generative model to generate responses to questions, taking into account feelings of confusion and producing responses such as "You should explain with more specific examples."

[1801] The device displays this response to the user, who then uses the explanation for learning.

[1802] 2. When a user requests custom learning content

[1803] The user requests, "Please create a workbook of problems on the basics of vectors."

[1804] Along with this request, the device sends the user's sentiment data to the server.

[1805] The server uses a generative model to generate a set of problems on the fundamentals of vectors, and adjusts the difficulty level if the user's motivation to learn is high.

[1806] The device displays the generated problem sets to the user, who then uses them to progress with their studies.

[1807] 3. When the user creates a learning plan

[1808] The user requests, "Please create a study plan for the next math exam," in order to develop a study plan for the next exam.

[1809] Along with this request, the device sends the user's sentiment data to the server.

[1810] The server uses generative models and a sentiment engine to generate an appropriate learning plan based on the user's learning history, progress data, and sentiment data.

[1811] For example, if a user is experiencing stress, the server will adjust its schedule accordingly.

[1812] The device displays the generated learning plan to the user, and the user proceeds with learning according to that plan.

[1813] (Example 2)

[1814] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1815] Traditional learning support systems provided features such as generating custom learning content based on individual users' learning progress and interests, developing learning plans, and managing progress. However, they lacked the ability to adjust responses based on user emotions. As a result, it was difficult to provide optimal learning support tailored to the user's emotional state, leading to challenges in maintaining learning effectiveness and motivation.

[1816] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1817] In this invention, the server includes means for generating responses to user input using a generative model, means for generating custom learning content based on the student's learning progress and interests, means for formulating a user's learning plan and managing its progress, means for recognizing the user's emotional data and adjusting the response, means for performing user authentication and matching authentication information with a database, means for adjusting the generated response based on the user's emotions and transmitting it to the terminal, and means for generating a learning plan and managing its progress based on the user's progress information and emotional data. This enables personalized learning support that takes the user's emotions into consideration.

[1818] A "generative model" is a pre-trained algorithm that generates natural language responses or content based on input data.

[1819] A "user" is a person, such as a student or teacher, who uses the system and is the entity that makes questions or learning requests through the learning support system.

[1820] "Custom learning content" refers to educational content such as learning materials and workbooks that are created individually based on a student's learning progress and interests.

[1821] A "learning plan" is a schedule or plan built based on the learning goals set by the user, and it also includes the management of progress.

[1822] "Emotional data" refers to data that indicates the user's current emotional state, obtained from user facial recognition and voice tone analysis.

[1823] A "response" refers to the natural language response or explanation that a generative model outputs in response to a user's question or request.

[1824] "User authentication" is the process of verifying the identity of users accessing a system, and it uses authentication information such as usernames and passwords.

[1825] A "terminal" is a device used by a user to interact with a syst...

Claims

1. A means for generating a response to user input using a generative model, A means of generating custom learning content based on students' learning progress and interests, A means of planning and managing the progress of a user's learning plan, A learning support system that includes this.

2. A learning support system according to claim 1, which generates responses to students' questions and provides individualized instruction support.

3. A learning support system according to claim 1, which enables users to access generative models on demand as part of an online learning platform.

Images