System

Abstract

To provide a system.SOLUTION: A system includes: first means for acquiring answer data indicating an answer to a learning subject of a user; second means for generating a first prompt sentence for outputting a next learning subject of the user on the basis of the answer data and past learning data of the user stored in a database; third means for inputting the generated first prompt sentence into a generative AI model to generate the next learning subject; and fourth means for transmitting the generated next learning subject to a terminal of the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Currently, in order for parents to support their children's studies, they have to create their children's assignments and mark the results after finishing their own work or housework. In addition, children have to wait for their parents to be free, which creates problems for parents and children, as well as waiting time. [Means for solving the problem]

[0005] To solve this problem, we provide a system that uses generative AI to generate learning tasks based on user input and presents them to the user. Furthermore, the system adjusts the difficulty of the learning tasks based on the user's learning progress and the questions they got wrong, marks the user's learning results, and presents the results to the user. This reduces the burden on parents and allows children to actively engage in learning. [Brief explanation of the drawings]

[0006] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 2 is a sequence diagram showing a flow of processing in the data processing system according to the first embodiment of the first form example. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Embodiment 1. [Figure 13] FIG. 10 is a sequence diagram showing a processing flow of a data processing system in a second embodiment of the second form example. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Embodiment Example 2. [Figure 15]FIG. 10 is a sequence diagram showing the flow of processing in a data processing system according to a third embodiment of the third embodiment. [Figure 16] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Embodiment 3. [Figure 17] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the first embodiment of the first form example when an emotion engine is combined. [Figure 18] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Form Example 1 when an emotion engine is combined. [Figure 19] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the second embodiment of the second form example when an emotion engine is combined. [Figure 20] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Form Example 2 when an emotion engine is combined. [Figure 21] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the third embodiment of the third form example when an emotion engine is combined. [Figure 22] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Form Example 3 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0007] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0008] First, the terms used in the following description will be explained.

[0009] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, the processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), or a TPU (TENSOR PROCESSING UNIT (registered trademark)).

[0010] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0011] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0012] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0013] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0014] [First embodiment]

[0015] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0016] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0017] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0018] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0019] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0020] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0021] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0022] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0023] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0024] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0025] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0026] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0027] "Example 1"

[0028] One embodiment of the present invention is a learning support system that uses a generative AI. This system generates learning tasks based on user input and presents the task to the user. Specifically, when the user asks, "What is today's learning task?", the generative AI generates a specific learning task such as, "Today, let's do 10 math problems on adding and subtracting fractions," and presents it to the user.

[0029] "Example 2"

[0030] Furthermore, this system adjusts the difficulty of the learning tasks based on the user's learning progress and the problems they got wrong. For example, if the user gets many fraction addition problems wrong, the generation AI will optimize the learning content by focusing on fraction addition as the next learning task. "Example 3"

[0031] The system also marks the user's learning results and presents the results to the user. Specifically, the answers to the problems the user solves are input into the generation AI, which then marks the answers based on those answers. The results are then presented to the user, and for any problems the user got wrong, an explanation is provided along with the correct answer. This allows the user to understand their own learning progress and deepen their understanding.

[0032] The processing flow of each embodiment will be described below.

[0033] "Example 1"

[0034] Step 1: The user asks the system, "What is today's study topic?"

[0035] Step 2: The generative AI receives input from the user and generates learning tasks, such as "Today, let's do 10 addition and subtraction problems with fractions in arithmetic."

[0036] Step 3: The learning task generated by the generative AI is presented to the user.

[0037] "Example 2"

[0038] Step 1: After the user completes the learning task, they enter the results into the system.

[0039] Step 2: The generative AI receives input from the user and analyzes the questions they got wrong.

[0040] Step 3: The AI ​​adjusts the difficulty of the next lesson. For example, if the user makes many mistakes on fraction addition, the AI ​​will optimize the learning content by focusing on fraction addition in the next lesson.

[0041] "Example 3"

[0042] Step 1: The user enters the answer to the problem into the system.

[0043] Step 2: The generative AI receives input from the user and marks the answers.

[0044] Step 3: The AI ​​generator presents the results of the marking to the user. For incorrect questions, it provides an explanation along with the correct answer. This allows the user to understand their own learning progress and deepen their understanding.

[0045] Example 1

[0046] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0047] Conventional learning support systems have difficulty providing appropriate learning tasks that correspond to the user's learning progress and individual needs. Furthermore, they lack effective means of providing feedback on the user's learning results or motivating them to learn. This makes it difficult for users to progress efficiently in their studies.

[0048] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0049] In this invention, the server includes means for receiving input from a user, means for analyzing the received input data and sending it as a prompt to a generative AI model, means for generating study tasks based on the user's input using the generative AI model and presenting the study tasks to the user, means for adjusting the difficulty of the study tasks based on the user's learning progress and questions that the user got wrong, means for grading the user's study results and presenting the results to the user, means for sending the generated study tasks to the user, and means for displaying the study tasks to the user. This makes it possible to provide appropriate study tasks according to the individual needs of the user and to increase motivation to learn through feedback on the study results.

[0050] "Means for receiving input from the user" refers to the function of providing an interface for the user to input information to the system in the form of text, voice, etc., and receiving that input.

[0051] "Means for analyzing received input data and sending it to the generative AI model as a prompt" refers to the function of analyzing input data received from a user and sending the analysis results to the generative AI model as a prompt in a format that the generative AI model can understand.

[0052] "Means for generating learning tasks based on user input using a generative AI model and presenting the learning tasks to the user" refers to the function of a generative AI model generating appropriate learning tasks based on user input and presenting the generated learning tasks to the user.

[0053] "A means to adjust the difficulty of the learning task based on the user's learning progress and the questions they got wrong" refers to a function that takes into account the user's past learning history and the questions they got wrong and appropriately adjusts the difficulty of the next learning task presented.

[0054] "Means for marking the user's learning results and presenting the results to the user" refers to a function that judges whether the learning tasks answered by the user are correct or incorrect and presents the results to the user as feedback.

[0055] "Means for sending generated learning tasks to users" refers to the function of sending learning tasks generated by the generative AI model to the user's device.

[0056] "Means for displaying study assignments to the user" refers to a function that visually displays received study assignments on the user's device.

[0057] This invention relates to a learning support system using a generative AI model. This system has the function of generating learning tasks based on user input and presenting the learning tasks to the user. Specific embodiments of this system are described below.

[0058] Hardware and software used

[0059] The system uses the following hardware and software:

[0060] Server: Receives input from the user, sends prompts to the generative AI model, and sends the generated learning tasks to the user.

[0061] Terminal: Provides an interface for users to input information and displays the learning tasks received from the server. Terminals include smartphones, tablets, and PCs.

[0062] Generative AI model: Generates learning tasks based on user input. Specifically, we use a generative AI model that uses natural language processing technology (e.g., GPT-4 (registered trademark) from OpenAI (registered trademark)).

[0063] System Operation

[0064] Receiving user input

[0065] A user uses a terminal to provide text input to the system, for example, a user might type, "What's my lesson today?" This input is sent to the server through the terminal interface.

[0066] Parsing input data and sending prompts

[0067] The server analyzes the user's input data received from the device. The analyzed data is sent to the generative AI model as a prompt. For example, the input "What is today's study topic?" is sent to the generative AI model as a prompt.

[0068] Generating learning tasks

[0069] The generative AI model generates learning tasks based on the received prompts. The generated learning tasks are specific and actionable for the user. For example, a learning task might be generated such as, "Today, let's do 10 addition and subtraction problems of fractions in arithmetic."

[0070] Submit and view coursework

[0071] The server sends the learning tasks received from the generative AI model to the device, which then displays the received learning tasks to the user. For example, the smartphone screen might display the following text: "Today, let's do 10 math problems on adding and subtracting fractions."

[0072] Specific examples

[0073] Example 1

[0074] The user opens the app on their smartphone, enters "What is today's lesson topic?" and presses the send button. The device sends this input to the server. The server receives the request and sends a prompt to the generative AI model saying, "What is today's lesson topic?" The generative AI model generates a lesson topic, such as "Today, let's do 10 math problems on adding and subtracting fractions," and returns it to the server. The server sends this lesson topic to the device, which displays it on the app screen.

[0075] Example 2

[0076] The user opens a web browser on their computer, enters "What is today's study topic?" and presses the send button. The device sends this input to the server. The server receives the request and sends a prompt to the generative AI model saying "What is today's study topic?" The generative AI model generates a study topic such as "Today, let's research Japan's Warring States period in history," and returns it to the server. The server sends this study topic to the device, which displays it on the web browser screen.

[0077] Prompt Sentence Examples

[0078] Below are some example prompts to input to a generative AI model:

[0079] "What's today's lesson?"

[0080] "Please tell me what we'll be studying tomorrow."

[0081] "What should I study for my next test?"

[0082] Using these prompts, the generative AI model can provide the user with appropriate learning tasks.

[0083] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0084] Step 1:

[0085] Receiving user input

[0086] A user uses a terminal to input text into the system. For example, the user types, "What is today's study topic?" This input is sent to the server through the terminal interface. The input data is in text format, and the terminal sends this data to the server as an HTTP POST request.

[0087] Step 2:

[0088] Receiving and parsing input data

[0089] The server receives the user's input data sent from the device. The received data is analyzed by the text analysis module. As a result of the analysis, the user's intent is understood and a prompt is generated to be sent to the generative AI model. For example, the input "What are your study goals for today?" is converted into the prompt "Please tell me your study goals for today."

[0090] Step 3:

[0091] Sending prompts to generative AI models

[0092] The server sends the parsed prompt to the generative AI model. This prompt must be in a format that the generative AI model can understand. The generative AI model receives the prompt and performs data calculations to generate an appropriate learning task. For example, the prompt sent might be, "Please tell me today's learning task."

[0093] Step 4:

[0094] Generation of learning tasks

[0095] The generative AI model generates learning tasks based on the received prompt. The generated learning tasks are specific and actionable for the user. For example, a learning task might be generated such as, "Today, let's do 10 addition and subtraction problems of fractions in arithmetic." The generative AI model returns this learning task to the server.

[0096] Step 5:

[0097] Receiving and sending assignments

[0098] The server receives the learning task from the generative AI model. It then prepares the received learning task to send to the user. Specifically, it formats the learning task as an HTTP response and prepares to send it to the device. For example, it prepares the text "Today, let's do 10 math problems on adding and subtracting fractions" as an HTTP response.

[0099] Step 6:

[0100] Submit your coursework

[0101] The server sends the prepared learning tasks to the device. This is done using a communication protocol such as an HTTP response. For example, the server sends the following text to the device as an HTTP response: "Today, let's do 10 math problems on adding and subtracting fractions."

[0102] Step 7:

[0103] Viewing learning assignments

[0104] The device receives the learning tasks from the server and displays them to the user. The way they are displayed depends on the interface the user is accessing. For example, a smartphone screen might display the following text: "Today, let's do 10 math problems on adding and subtracting fractions."

[0105] (Application example 1)

[0106] Next, a description will be given of Application Example 1 of Embodiment Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0107] Conventional learning support systems can only be used by users at home or in specific locations, and do not provide sufficient learning support in physical stores. Furthermore, they lacked the functionality to allow users to generate and display learning tasks in physical stores, resulting in low learning efficiency and convenience. Furthermore, they lacked the functionality to adjust the difficulty of learning tasks based on the user's learning progress or incorrect questions, making it difficult to meet individual learning needs. To address these issues, there is a need for a system that strengthens learning support in physical stores and allows users to generate and display learning tasks using their mobile information devices.

[0108] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0109] In this invention, the server includes means for generating study tasks based on input from the user using a generation AI and presenting the study tasks to the user, means for adjusting the difficulty of the study tasks based on the user's learning progress and questions that they have gotten wrong, means for grading the user's learning results and presenting the results to the user, means for the user to generate and present study tasks using a mobile information terminal in a physical store, and means for displaying the generated study tasks to the user in the physical store. This enhances learning support in physical stores and enables users to efficiently generate and present study tasks using their mobile information terminals.

[0110] "Generative AI" is an artificial intelligence system that generates learning tasks based on user input and presents them to the user.

[0111] "User" means an individual or organization that receives learning assignments using the learning support system.

[0112] A "learning assignment" is a collection of specific learning content or problems that a user must tackle.

[0113] A "mobile information terminal" is a portable information processing device such as a smartphone or tablet.

[0114] A "physical store" is a commercial establishment or service provider that exists in a physical location and can be visited by users.

[0115] "Study progress" is an indicator that shows how far the user has progressed in the learning tasks.

[0116] "Difficulty adjustment" is the process of changing the difficulty of a learning task based on the user's learning progress and the questions they get wrong.

[0117] "Marking" is the process of determining whether a user's learning results are correct or incorrect and evaluating them.

[0118] "Feedback" refers to evaluations and advice provided to users regarding their learning results.

[0119] "Learning motivation" refers to the degree of interest and motivation a user has in learning.

[0120] A system for implementing this invention uses a generative AI model to generate study tasks based on user input and present them to the user. The system adjusts the difficulty of the study tasks based on the user's learning progress and incorrect questions, marks the user's learning results, and presents the results to the user. The system also has a function that allows users to generate and present study tasks using their mobile information terminals in physical stores.

[0121] The server receives input from the user using a generative AI model and generates a prompt sentence. For example, if the user inputs "What is today's study topic?", the server generates the prompt sentence "The user input 'What is today's study topic?' Please generate today's study topic." The generated prompt sentence is sent to the generative AI model, which generates a specific study topic.

[0122] The generated learning tasks are sent from the server to the user's mobile information device and displayed on the device. If the user is in a physical store, they can check and work on the learning tasks through their device. The progress of the learning tasks and any questions that they got wrong are sent to the server, which then adjusts the difficulty of the learning tasks based on this information.

[0123] The server automatically marks the user's learning results and presents the results to the user, allowing the user to check their learning progress in real time.The server also provides feedback based on the user's learning results, increasing the user's motivation to learn.

[0124] As a concrete example, if a user visits a brick-and-mortar bookstore and uses a mobile information terminal to type, "What is today's study assignment?", the server will use the generative AI model to generate a study assignment such as, "Today, let's do 10 math problems on adding and subtracting fractions," and display it on the terminal. The user can then work on this assignment and send the results back to the server, which will adjust the difficulty of the next study assignment.

[0125] Example prompt sentence:

[0126] The user types, "What's today's lesson plan?" Generate today's lesson plan.

[0127] In this way, learning support in brick-and-mortar stores is enhanced, and users can efficiently generate and present learning tasks using mobile information terminals.

[0128] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0129] Step 1:

[0130] A user uses a mobile information terminal to input a request for a learning assignment.

[0131] Input: User types "What should I study today?"

[0132] Output: The user's input data.

[0133] Specific operation: The user launches the application on the device, enters "What is today's study topic?" in the text box, and presses the submit button.

[0134] Step 2:

[0135] The device sends the user's input to the server.

[0136] Input: User input data.

[0137] Output: The request data sent to the server.

[0138] Specific operation: The terminal sends the user's input data to the server as an HTTP request.

[0139] Step 3:

[0140] The server receives the user's input and generates a prompt.

[0141] Input: User input data.

[0142] Output: The generated prompt statement.

[0143] Specific operation: The server analyzes the user's input data and generates a prompt sentence such as, "The user entered 'What are today's study topics?' Please generate today's study topics."

[0144] Step 4:

[0145] The server sends prompt sentences to the generative AI model to generate learning tasks.

[0146] Input: The generated prompt statement.

[0147] Output: The generated learning task.

[0148] Specific operation: The server sends a prompt to a generative AI model (e.g., OpenAI's GPT-3 (registered trademark)) to generate a learning task. Based on the prompt, the generative AI model generates a learning task such as "Today, let's do 10 addition and subtraction problems of fractions in arithmetic."

[0149] Step 5:

[0150] The server transmits the generated learning assignments to the terminal.

[0151] Input: The generated learning task.

[0152] Output: The learning task data sent to the device.

[0153] Specific operation: The server sends the generated learning task to the terminal as an HTTP response.

[0154] Step 6:

[0155] The device displays the received learning assignments to the user.

[0156] Input: Learning task data received from the server.

[0157] Output: The learning task that is displayed to the user.

[0158] Specific operation: The device analyzes the received learning task data and displays it on the screen. The user can check the learning task on the device screen.

[0159] Step 7:

[0160] The user works on the learning tasks and enters the results into the device.

[0161] Input: User's learning result data.

[0162] Output: Learning result data input to the terminal.

[0163] Specific operation: The user uses the input form on the device to enter the answer to the learning task and presses the submit button.

[0164] Step 8:

[0165] The device sends the user's learning results to the server.

[0166] Input: User's learning result data.

[0167] Output: Training result data sent to the server.

[0168] Specific operation: The device sends the user's learning result data to the server as an HTTP request.

[0169] Step 9:

[0170] The server marks the user's learning results and generates the results.

[0171] Input: User's learning result data.

[0172] Output: Rounding result data.

[0173] Specific operation: The server analyzes the user's learning result data, determines whether the answer is correct or incorrect, and generates a marking result.

[0174] Step 10:

[0175] The server sends the rounding results to the terminal.

[0176] Input: Rounding result data.

[0177] Output: Rounding result data sent to the terminal.

[0178] Specific operation: The server sends the rounding result data to the terminal as an HTTP response.

[0179] Step 11:

[0180] The terminal displays the received marking results to the user.

[0181] Input: Rounding result data received from the server.

[0182] Output: Rounding results displayed to the user.

[0183] Specific operation: The terminal analyzes the received marking result data and displays it on the screen. The user can check the marking result on the terminal screen.

[0184] Example 2

[0185] Next, a description will be given of Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0186] Conventional learning systems lacked the ability to adjust the difficulty of learning tasks based on the user's learning progress and the questions they got wrong, making it difficult to provide an optimized learning experience for each individual user. Furthermore, they lacked a mechanism for efficiently managing users' answer data and reflecting it in the next learning task, preventing improvements in learning effectiveness. Furthermore, there was a need for a system that could consistently generate and present learning tasks using a generative AI model.

[0187] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0188] In this invention, the server includes a means for storing the user's answer data in a database, a means for sending a prompt to the generative AI model to generate the next learning task, and a means for presenting the generated learning task to the user. This makes it possible to adjust the difficulty of the learning task based on the user's learning progress and questions that they got wrong, providing an optimized learning experience for each individual user.

[0189] "Generative AI" refers to artificial intelligence technology that generates learning tasks based on user input and presents them to the user.

[0190] "User" refers to an individual or group working on a learning task.

[0191] "Study assignments" refer to problems or exercises that users must complete.

[0192] "Study progress" refers to an indicator that shows how far a user has progressed in their studies.

[0193] "Incorrect questions" refer to questions that the user was unable to answer correctly.

[0194] "Difficulty adjustment" refers to changing the difficulty of learning tasks based on the user's learning progress and the questions they got wrong.

[0195] "Marking" refers to determining whether a user's learning results are correct or incorrect.

[0196] "Answer data" refers to information about the answers given by a user to a learning task.

[0197] "Database" refers to an information management system for storing user answer data.

[0198] "Generative AI model" refers to artificial intelligence technology that generates the next learning task based on a prompt.

[0199] A "prompt" refers to an instruction given to a generative AI model to generate the next learning task.

[0200] "Presenting" refers to displaying the generated learning task to the user.

[0201] This invention is a system that adjusts the difficulty of learning tasks based on the user's learning progress and the questions they have answered incorrectly, providing an optimized learning experience for each individual user. Specific embodiments of this system are described below.

[0202] Hardware and software used

[0203] server

[0204] The server is responsible for collecting user answer data and storing it in a database. The server uses a database management system such as MySQL (registered trademark). It also sends prompts to the generative AI model and performs processing to generate the next learning task. For example, OpenAI's GPT-4 is used as the generative AI model.

[0205] Terminal

[0206] The terminal provides an interface for users to work on learning tasks. The terminal is built using web technologies such as HTML, CSS, and JavaScript (registered trademark). It has the function of sending the user's answers to the server in real time.

[0207] user

[0208] Users work on learning tasks provided through their devices. When they answer the questions, the results are sent to the server and used to adjust the difficulty of the next learning task.

[0209] Explanation of program processing

[0210] Server Processing

[0211] The server receives the user's answer data and stores it in a database such as MySQL. For example, it stores information such as the user ID, question ID, answer result, and answer time. The server then sends a prompt to the generative AI model. The prompt describes what questions the user made the most mistakes on and what questions should be presented as the next learning assignment.

[0212] Processing generative AI models

[0213] The generative AI model generates the next learning task based on the prompt received from the server. For example, if the user makes many mistakes in fraction addition, the next learning task will focus on fraction addition.

[0214] Terminal handling

[0215] The device displays the new learning tasks received from the server to the user, who can then work on the learning tasks again.

[0216] Specific examples

[0217] For example, if a user answers "1 / 2 + 1 / 3 = 5 / 6", the answer is sent to the server and saved in the database as "User ID: 12345, Question ID: 67890, Answer: 5 / 6, Correct / Incorrect: Incorrect, Answer Time: 30 seconds". The server then sends the following prompt to the generative AI model:

[0218] Prompt Sentence Examples

[0219] Users made a lot of mistakes on fraction addition problems, so please focus on fraction addition in the next lesson. Specifically, please include problems like:

[0220] 1. 1 / 2 + 1 / 3 = ?

[0221] 2. 3 / 4 + 2 / 5 = ?

[0222] 3. 5 / 6 + 1 / 2 = ?

[0223] The generative AI model generates the next learning task based on this prompt and returns it to the server. The server then sends the problem set to the device, which then displays it as a new learning task for the user. This makes it possible to adjust the difficulty of the learning task based on the user's learning progress and the questions they got wrong, providing an optimized learning experience for each individual user.

[0224] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0225] Step 1:

[0226] Users work on learning tasks

[0227] The user works on a learning task provided through the device. For example, the user answers a fraction addition problem. The user's answer (e.g., "1 / 2 + 1 / 3 = 5 / 6") is obtained as input. The user's answer data is generated as output.

[0228] Step 2:

[0229] The device sends the user's answer to the server

[0230] The terminal sends the user's answers to the server in real time. Specifically, it sends the user's answer data (e.g., "User ID: 12345, Question ID: 67890, Answer: 5 / 6") to the server. The user's answer data is obtained as input. The answer data sent to the server is generated as output.

[0231] Step 3:

[0232] The server saves the answer data in a database

[0233] The server stores the received answer data in a database. Specifically, it uses a database management system such as MySQL to store information such as user ID, question ID, answer result, and answer time. The answer data sent to the server is obtained as input. The answer data stored in the database is generated as output.

[0234] Step 4:

[0235] The server sends a prompt to the generative AI model

[0236] The server sends a prompt to the generative AI model based on the user's past learning data. Specifically, it generates a prompt such as, "The user made many mistakes on fraction addition problems, so please focus on fraction addition in the next learning assignment," and sends it to the generative AI model. The user's past learning data is obtained as input. The prompt sent to the generative AI model is generated as output.

[0237] Step 5:

[0238] The generative AI model generates the next learning task

[0239] A generative AI model (for example, OpenAI's GPT-4) generates the next training problem based on the received prompt. Specifically, it generates problems such as "1 / 2 + 1 / 3 = ?", "3 / 4 + 2 / 5 = ?", and "5 / 6 + 1 / 2 = ?". The input is the prompt. The output is a generated training problem.

[0240] Step 6:

[0241] The server sends the generated learning assignments to the device.

[0242] The server sends the learning tasks received from the generative AI model to the device. Specifically, it sends the generated problem set to the device in JSON format. The generated learning tasks are obtained as input. The learning tasks sent to the device are generated as output.

[0243] Step 7:

[0244] The device displays new learning tasks to the user.

[0245] The terminal displays the new learning task received from the server to the user. The user can then try the learning task again. As input, the learning task sent from the server is obtained. As output, the learning task displayed to the user is generated.

[0246] (Application example 2)

[0247] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0248] Conventional learning support systems lacked the ability to adjust the difficulty of learning tasks based on the user's learning progress and the questions they got wrong, making it difficult for users to focus on areas in which they were weak. Furthermore, they lacked the ability to record the user's learning progress and optimize the next learning task, which led to problems with reduced learning efficiency. Furthermore, the lack of a function to focus questions on the category in which the user was weakest limited the effectiveness of learning.

[0249] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0250] In this invention, the server includes means for generating study tasks based on user input using a generation AI and presenting the study tasks to the user, means for adjusting the difficulty of the study tasks based on the user's learning progress and questions that they got wrong, means for grading the user's learning results and presenting the results to the user, means for recording the user's learning progress and optimizing the next study task, and means for focusing questions on the category in which the user is weakest. This makes it possible to provide optimal study tasks according to the user's learning progress, improving the efficiency and effectiveness of learning.

[0251] "Generative AI" is an artificial intelligence system that generates learning tasks based on user input and presents them to the user.

[0252] "Learning assignments" are problems or exercises that users must work on, and are generated by generative AI.

[0253] "User's learning progress" is an index that indicates how far the user has progressed in their learning, and represents the progress of their learning.

[0254] "Wrong questions" are questions that the user answered incorrectly, and are data used to evaluate the user's level of understanding.

[0255] "Difficulty of learning task" is an indicator of how difficult the learning task is, and is adjusted based on the user's learning progress and the questions they got wrong.

[0256] "Marking" is an evaluation method that judges whether a user's learning results are correct or incorrect and presents the results to the user.

[0257] "User's learning results" are data that indicate how the user answered the learning tasks, and are the subject of marking.

[0258] "Recording the user's learning progress" means saving data on how far the user has progressed in their studies, which serves as the basis for optimizing the next learning task.

[0259] "Optimizing the next learning task" means providing the next learning task in the most optimal form based on the user's learning progress and the questions they got wrong.

[0260] A "category" is a specific field or theme to which a learning task belongs, and indicates the classification of the content that the user is learning.

[0261] "Focusing on questions" means providing many questions on categories that the user is particularly weak at, and is a means of improving the user's level of understanding.

[0262] As an embodiment of the present invention, the following system is constructed.

[0263] System Overview

[0264] The system adjusts the difficulty of the learning tasks based on the user's learning progress and the questions they got wrong, and has the ability to optimize the next learning task. Specifically, it uses a generative AI model to generate learning tasks based on user input and presents them to the user. It also marks the user's learning results and presents the results to the user.

[0265] Hardware and software used

[0266] Hardware: Smartphones, servers

[0267] Software: Python (registered trademark), database (e.g., SQLite (registered trademark)), generative AI model

[0268] Data processing and calculation

[0269] The server receives input data from the user and generates learning tasks using a generative AI model. The generated learning tasks are presented to the user's smartphone. The user works on the learning tasks and sends the results to the server. The server records the user's learning results and processes the data to optimize the next learning task.

[0270] Specific examples

[0271] For example, consider a situation where a user answers a quiz about robot operating procedures. If the user makes a mistake in a particular operating procedure, the next learning task will focus on that operating procedure.

[0272] Example prompt sentence:

[0273] User ID: user1

[0274] Category: robot_operation

[0275] Correct Answer: False

[0276] When this prompt is entered, the server records the user's learning progress and focuses questions related to "robot_operation" on the next learning assignment.

[0277] Processing flow

[0278] 1. The user works on a learning task using a smartphone.

[0279] 2. The user's answer is sent to the server.

[0280] 3. The server records the user's learning progress and the questions they got wrong.

[0281] 4. The generative AI model generates the next learning task and presents it to the user.

[0282] In this way, it is possible to provide optimal learning tasks according to the user's learning progress, thereby improving the efficiency and effectiveness of learning.

[0283] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0284] Step 1:

[0285] Users use their smartphones to complete learning tasks.

[0286] Input: User's answer data (e.g., user ID, category, answer result)

[0287] Output: The user's answer results are sent to the server.

[0288] Specific operation: The user opens the application on their smartphone and enters answers to the presented learning tasks. Once the answers are completed, the answer data is sent to the server.

[0289] Step 2:

[0290] The server receives the user's answer results and records them in a database.

[0291] Input: User's answer data (e.g., user ID, category, answer result)

[0292] Output: User learning progress data recorded in a database

[0293] Specific operation: The server analyzes the received answer data and records the user's learning progress and incorrect questions in a database, thereby accumulating the user's learning history.

[0294] Step 3:

[0295] The server generates the next study task based on the user's learning progress and the questions they got wrong.

[0296] Input: User learning progress data, data on questions answered incorrectly

[0297] Output: Generated next learning task data

[0298] How it works: The server analyzes the user's learning progress and incorrect questions recorded in the database, and uses the generative AI model to generate the next learning task. The generated learning task is adjusted to include a focus on the categories in which the user is weak.

[0299] Step 4:

[0300] The server sends the generated next learning assignment to the user's smartphone.

[0301] Input: Generated next learning task data

[0302] Output: Learning tasks displayed on the user's smartphone

[0303] Specific operation: The server sends the generated learning task data to the user's smartphone, where the user can then work on the task in their next learning session.

[0304] Step 5:

[0305] The user then works on the next learning task and sends the answer to the server again.

[0306] Input: User's answer data (e.g., user ID, category, answer result)

[0307] Output: The user's answer results are sent to the server.

[0308] Specific operation: The user opens the smartphone application and enters the answer to the next learning task that is presented. Once the answer is completed, the answer data is sent back to the server.

[0309] In this way, optimal learning tasks are continuously provided according to the user's learning progress, improving the efficiency and effectiveness of learning.

[0310] Example 3

[0311] Next, a third embodiment of the third embodiment will be described. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0312] In conventional learning systems, evaluation and feedback on user answers is often done manually, which is problematic in terms of efficiency. It is also difficult to provide appropriate feedback based on the user's learning progress and level of understanding, which can lead to issues such as insufficient motivation to learn or deepening of understanding. Furthermore, there was a need for a system that could properly process user answer data and quickly present evaluation results.

[0313] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0314] In this invention, the server includes means for generating learning tasks based on user input using a generative AI model and presenting the learning tasks to the user, means for adjusting the difficulty of the learning tasks based on the user's learning progress and questions that the user got wrong, means for marking the user's learning results and presenting the results to the user, means for receiving the user's answer data, generating prompt sentences and sending them to the generative AI model, and means for receiving evaluation results and explanations from the generative AI model, storing them in a database, and presenting them to the user. This enables quick and accurate evaluation of the user's answers, and by providing appropriate feedback, it is possible to improve motivation to learn and deepen understanding.

[0315] A "generative AI model" is a type of artificial intelligence that generates learning tasks and evaluates answers based on input data from users.

[0316] A "user" is an individual or group that uses the learning system to solve learning tasks and input answers.

[0317] "Learning tasks" are problems or exercises that users must solve, and are generated by the generative AI model.

[0318] "Answer data" is information about the answer that the user inputs to the study task.

[0319] A "prompt sentence" is an instruction sentence generated by the server to request the generative AI model to evaluate the answer.

[0320] The "evaluation result" is the result of the generative AI model evaluating the user's answer, including a judgment of correctness or incorrectness and an explanation.

[0321] A "database" is an information management system for storing information such as evaluation results and explanations.

[0322] A "server" is a computer system that receives user answer data and performs processes such as sending it to the generative AI model.

[0323] "Feedback" refers to evaluations and explanations provided to users regarding their learning results, and is intended to encourage the user's learning progress and increase their motivation to learn.

[0324] The present invention relates to a system that allows a user to solve a learning task, evaluates the answer, and provides feedback. A specific embodiment of this system will be described below.

[0325] Hardware and software used

[0326] Hardware: Servers, devices (PCs, smartphones, tablets, etc.)

[0327] Software: Generative AI models (e.g., large-scale language models), web browsers, database management systems (e.g., MySQL)

[0328] System configuration

[0329] 1. The user solves a problem

[0330] The user uses a terminal to access the learning platform and inputs answers to the questions displayed.

[0331] For example, a user answers "4" to "Question 1: 2 + 2 = ?"

[0332] 2. The device sends the answer data to the server

[0333] The terminal transmits the answer data entered by the user to the server.

[0334] Specifically, the terminal uses an HTTP request to send the answer data to the server.

[0335] For example, the answer data "Question 1: 4" is sent to the server.

[0336] 3. The server generates a prompt and sends it to the generative AI model.

[0337] The server generates a prompt sentence based on the received answer data.

[0338] The generated prompt sentence is sent to the generative AI model.

[0339] For example, the prompt sentence "The user's answer is '4'. Please determine whether this answer is correct, and if it is correct, mark it as 'Correct'. If it is incorrect, mark it as 'Incorrect'. Please provide the correct answer and an explanation." is sent to the generative AI model.

[0340] 4. The generative AI model evaluates the answer and returns the result to the server

[0341] The generative AI model evaluates the user's answer based on the prompt sent from the server.

[0342] The generative AI model generates evaluation results and explanations and returns them to the server.

[0343] For example, the generative AI model returns the result "Correct. Explanation: 2 + 2 is 4" to the server.

[0344] 5. The server stores the results in a database and presents them to the user.

[0345] The server stores the evaluation results and explanations received from the generative AI model in a database.

[0346] The server sends the saved results to the user's terminal and presents them to the user.

[0347] For example, the server sends the result "Question 1: Correct. Explanation: 2 + 2 is 4" to the user's terminal.

[0348] 6. The user checks the results and continues learning

[0349] The user checks the results on the device.

[0350] Users will learn the correct answers and explanations for questions they get wrong.

[0351] For example, the user confirms the result "Question 1: Correct. Explanation: 2 + 2 is 4." and moves on to the next question.

[0352] In this way, the user can check whether their answer is correct, and if they are wrong, they can learn the correct answer and explanation. This system makes it possible to efficiently grasp the user's learning progress and deepen their understanding. The flow of the identification process in Example 3 will be described with reference to FIG. 15.

[0353] Step 1:

[0354] The user solves the problem

[0355] Input: The user accesses the learning platform using a device and enters answers to the questions displayed.

[0356] Specific action: For example, the user answers "4" to "Question 1: 2 + 2 = ?"

[0357] Output: The user's answer data is saved on the device.

[0358] Step 2:

[0359] The device sends the answer data to the server

[0360] Input: User's answer data

[0361] Specific operation: The device sends the answer data to the server using an HTTP request.

[0362] Output: For example, the answer data "Question 1: 4" is sent to the server.

[0363] Step 3:

[0364] The server generates a prompt and sends it to the generative AI model.

[0365] Input: Answer data received by the server

[0366] Specific operation: The server generates a prompt based on the received answer data. For example, it generates a prompt like this: "The user's answer is '4'. Please determine whether this answer is correct. If it is correct, mark it as 'Correct'. If it is incorrect, mark it as 'Incorrect'. Please provide the correct answer and an explanation."

[0367] Output: The generated prompt sentence is sent to the generative AI model.

[0368] Step 4:

[0369] The generative AI model evaluates the answer and returns the result to the server

[0370] Input: The prompt received by the generative AI model

[0371] Specific operation: The generative AI model evaluates the user's answer based on the prompt sentence and generates an evaluation result and explanation. For example, it generates a result such as "Correct. Explanation: 2 + 2 is 4."

[0372] Output: The evaluation result and explanation are returned to the server.

[0373] Step 5:

[0374] The server stores the results in a database and presents them to the user.

[0375] Input: Evaluation results and explanations received by the server

[0376] Specific operation: The server saves the evaluation results and explanations in a database and sends the saved results to the user's device. For example, it sends the result "Question 1: Correct. Explanation: 2 + 2 is 4" to the user's device.

[0377] Output: The evaluation results and explanations are displayed on the user's terminal.

[0378] Step 6:

[0379] Users can check the results and continue learning.

[0380] Input: Evaluation results and explanations displayed on the user's device

[0381] Specific operation: The user checks the results on the device and learns the correct answer and explanation for the questions they got wrong. For example, the user sees the result "Question 1: Correct. Explanation: 2 + 2 is 4" and moves on to the next question.

[0382] Output: User progress and understanding improves.

[0383] (Application example 3)

[0384] Next, a description will be given of Application Example 3 of Form Example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0385] Conventional learning systems lacked the ability to automatically mark the answers to questions solved by users and provide feedback. They also lacked the ability to provide the correct answer and explanation if the user's answer was incorrect. This made it difficult for users to accurately grasp their own learning progress, which in turn hindered their motivation to learn.

[0386] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes means for generating study tasks based on user input using a generation AI and presenting the study tasks to the user, means for adjusting the difficulty of the study tasks based on the user's learning progress and questions that were answered incorrectly, means for marking the user's learning results and presenting the results to the user, means for receiving the user's answers in list form, comparing them with correct answers, and generating results, and means for providing the correct answers and explanations in the case of incorrect answers. This allows the user to accurately grasp their own learning progress and deepen their understanding.

[0387] "Generative AI" is an artificial intelligence system that generates learning tasks based on user input and presents them to the user.

[0388] "Learning assignments" are questions or exercises for users to answer, which are generated by generative AI.

[0389] "User's learning progress" refers to the progress a user makes in answering learning tasks.

[0390] "Marking" is the act of determining whether the user's answer is correct and showing the result.

[0391] "List format" is a format in which the user's answers are organized into a series of items and displayed in an easy-to-compare format.

[0392] A "correct answer" is an accurate response to a learning task.

[0393] An "explanation" provides the correct answer for a question that the user got wrong, and explains the reason and background behind the answer.

[0394] "Feedback" refers to evaluations and advice provided based on the user's learning results.

[0395] The system for implementing this invention mainly uses a server, a user terminal, and a generative AI model. Specific embodiments will be described below.

[0396] System Configuration

[0397] 1. Server:

[0398] The server hosts the generative AI model and receives input data from the user.

[0399] The server receives the user's answers in the form of a list, compares them with the correct answers, and generates a result.

[0400] The server provides the correct answer and explanation in case of incorrect answer.

[0401] 2. User Device:

[0402] A user terminal is a device such as a smartphone or tablet that a user uses to answer study questions.

[0403] The user terminal communicates with the server, displays the generated learning tasks, and transmits the user's answers to the server.

[0404] 3. Generative AI Model:

[0405] The generative AI model generates learning tasks based on user input and presents them to the user.

[0406] The generative AI model adjusts the difficulty of the learning tasks based on the user's learning progress and the questions they got wrong.

[0407] The generative AI model marks the user's learning results and presents the results to the user.

[0408] Program processing explanation

[0409] The server receives the answer data sent from the user device and performs the following processing using the generative AI model.

[0410] 1. Data reception and analysis:

[0411] The server receives the answer data sent from the user terminal.

[0412] Analyze the received data and convert it into a list format.

[0413] 2. Compare answers and generate results:

[0414] The server uses a generative AI model to compare the user's answer with the correct answer.

[0415] Based on the comparison result, a determination is made as to whether the answer is correct or incorrect, and a result is generated.

[0416] 3. Providing Feedback:

[0417] The server generates the correct answer and explanation for incorrect answers.

[0418] The generated feedback is sent to the user terminal and presented to the user.

[0419] Hardware and software used

[0420] Hardware: Servers, smartphones, tablets

[0421] Software: Python, OpenAI API

[0422] Specific examples

[0423] When a user answers a learning task using a smartphone, they input prompt sentences like the following into the generative AI model:

[0424] Example prompt sentence:

[0425] User Answer: Answer 2

[0426] Correct Answer: Correct Answer 2

[0427] Explanation: Explanation 2

[0428] This answer is incorrect. The correct answer is "Correct Answer 2". Explanation: Explanation 2

[0429] By inputting this prompt into the generative AI model, appropriate feedback can be provided to the user, allowing them to accurately grasp their own learning progress and deepen their understanding.

[0430] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0431] Step 1:

[0432] The user answers the learning questions using the device. The user inputs the answer through the application on the device and presses the submit button. The input data is the user's answer, which is then sent to the server.

[0433] Step 2:

[0434] The server receives the answer data sent from the user's device. The received data is stored in the server as a list of the user's answers. The input data is the user's answer list, and the output data is a list of answers ready for analysis.

[0435] Step 3:

[0436] The server uses a generative AI model to compare the user's answers with the correct answers. Specifically, it matches the user's answer list with the correct answer list and determines whether each answer is correct or incorrect. The input data is the user's answer list and the correct answer list, and the output data is the determination result.

[0437] Step 4:

[0438] The server generates feedback based on the judgment result. If the answer is correct, it displays "Correct." If the answer is incorrect, it generates "Incorrect," along with the correct answer and explanation. The input data is the judgment result, and the output data is the feedback message.

[0439] Step 5:

[0440] The server sends the generated feedback message to the user terminal, and the user terminal displays the received feedback message to the user. The input data is the feedback message, and the output data is the feedback presented to the user.

[0441] Step 6:

[0442] Users can check the feedback on their devices and learn the correct answers and explanations for questions they got wrong. This allows users to understand their own learning progress and deepen their understanding. The input data is the feedback message, and the output data is the user's improvement in understanding.

[0443] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0444] "Example 1"

[0445] In one embodiment of the present invention, the generation AI uses an emotion engine that recognizes the user's emotions. Specifically, the emotion engine infers the user's emotional state from the user's reactions, facial expressions, tone of voice, and other factors while solving learning tasks. This inference is provided to the generation AI, which then adjusts the generation and presentation of learning tasks and the method of presenting grading results based on this information. For example, when the user is facing a difficult problem, the generation AI can send encouraging messages to maintain the user's motivation to study.

[0446] "Example 2"

[0447] In another form, the emotion engine infers the user's emotional state from their learning progress and reactions to questions they got wrong. Specifically, the emotion engine analyzes how the user reacted to questions they got wrong and how that reaction changed. The results of this analysis are provided to the generation AI, which then adjusts the difficulty of the next learning task based on this. For example, if the user is disappointed about a question they got wrong, the generation AI may lower the difficulty of the next learning task to maintain the user's motivation to learn.

[0448] "Example 3"

[0449] Furthermore, there is also a form in which the generative AI adjusts the feedback and encouragement it provides to the user based on the user's emotional state provided by the emotion engine. Specifically, after the user completes a learning task, the generative AI provides feedback taking into account the user's emotional state provided by the emotion engine. For example, if the user feels joy after completing a learning task, the generative AI provides feedback that reinforces that emotion. This further increases the user's motivation to learn.

[0450] The processing flow of each embodiment will be described below.

[0451] "Example 1"

[0452] Step 1: The user solves the learning task.

[0453] Step 2: The emotion engine estimates the user's emotional state based on their reactions, facial expressions, tone of voice, etc.

[0454] Step 3: Provide the estimation results to the generation AI.

[0455] Step 4: The generative AI adjusts how it generates and presents learning tasks and marks the results based on the provided emotional state.

[0456] Step 5: The generative AI sends encouraging messages to keep the user motivated to learn.

[0457] "Example 2"

[0458] Step 1: The user solves the learning task.

[0459] Step 2: The emotion engine infers the user's emotional state from their reaction to the incorrect question.

[0460] Step 3: Provide the estimation results to the generation AI.

[0461] Step 4: The AI ​​adjusts the difficulty of the next learning task based on the provided emotional state. Step 5: The AI ​​sends encouraging messages to keep the user motivated to learn.

[0462] "Example 3"

[0463] Step 1: The user solves the learning task.

[0464] Step 2: The emotion engine estimates the user's emotional state.

[0465] Step 3: Provide the estimation results to the generation AI.

[0466] Step 4: The generative AI provides feedback taking into account the provided emotional state.

[0467] Step 5: The generative AI provides feedback to improve the user's motivation to learn.

[0468] Example 1

[0469] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0470] Conventional learning support systems have the ability to adjust the difficulty of learning tasks based on the user's learning progress and the problems they have gotten wrong, but they lack learning support that takes into account the user's emotional state. This can lead to users losing motivation when they encounter difficult problems. In addition, they are unable to provide feedback or encouraging messages that reflect the user's emotions, which can lead to a decrease in learning continuity.

[0471] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0472] In this invention, the server includes means for generating study tasks based on user input using a generation AI and presenting the study tasks to the user, means for adjusting the difficulty of the study tasks based on the user's learning progress and questions that they got wrong, means for grading the user's learning results and presenting the results to the user, means for collecting user emotion data and analyzing it with an emotion engine, and means for adjusting the presentation method of the study tasks based on the analysis results of the emotion engine. This enables learning support that takes the user's emotional state into consideration, maintaining the user's motivation to learn and improving continuity in learning.

[0473] "Generative AI" refers to artificial intelligence that generates learning tasks based on user input and presents them to the user.

[0474] "Study assignments" are specific problems or exercises that users must complete.

[0475] "User's learning progress" is an indicator that shows how far the user has progressed in their learning.

[0476] "Difficulty of a learning task" is an index that indicates how difficult a learning task is.

[0477] "Marking" refers to the act of determining whether a learning task completed by a user is correct or incorrect and showing the result.

[0478] "Emotional data" refers to data that indicates the user's emotional state, such as the user's facial expression or tone of voice.

[0479] An "emotion engine" is a system that analyzes collected emotional data and estimates the user's emotional state.

[0480] "Feedback" refers to evaluations and comments provided on a user's learning results.

[0481] "Encouraging messages" are words of encouragement or encouragement sent to users to motivate them to learn.

[0482] The present invention relates to a learning support system using a generative AI model. Specific embodiments of this system will be described below.

[0483] First, the server generates a program for the learning support system using the generative AI model. This program has the function of generating learning tasks based on user input and presenting them to the user.

[0484] The user inputs "What is today's study topic?" through the device interface, which can be a web browser or a mobile application. The server receives the request from the user and analyzes the request.

[0485] Next, the server sends the user's request as a prompt to the generative AI model. The prompt is in the form of "Please tell me today's study task." The generative AI model analyzes the prompt and generates specific study tasks taking into account the user's learning history and progress. For example, it generates a task such as "Today, let's do 10 addition and subtraction problems of fractions in arithmetic."

[0486] The server presents the learning tasks received from the generative AI model to the user through the device's interface. As the user solves the learning tasks, the device uses a camera and microphone to collect emotional data such as the user's facial expressions and tone of voice.

[0487] The emotion engine analyzes the collected emotion data and estimates the user's emotional state. For example, if the user is facing a difficult problem, the emotion engine estimates that the user is confused. The server sends this estimation result to the generative AI model.

[0488] The generative AI model adjusts the presentation of learning tasks and additional messages based on the estimated emotional state. For example, if the user is confused, it generates an encouraging message. The server then displays this message to the user to keep them motivated to learn.

[0489] For example, if a user types "What is today's study topic?" into their device, the server receives this request and sends a prompt to the generative AI model saying, "Please tell me today's study topic." The generative AI model generates a study topic such as "Today, let's do 10 math problems on adding and subtracting fractions," and returns it to the server. The server then presents this study topic to the user.

[0490] While the user is solving a learning task, the device uses a camera and microphone to collect the user's facial expressions and tone of voice. The emotion engine analyzes this data and infers that the user is confused. The server sends this inference to the generative AI model, which then generates an encouraging message such as "Good luck! You're almost there!" The server displays this message to the user to maintain their motivation to learn.

[0491] Examples of prompts include:

[0492] "What's today's lesson?"

[0493] "Tell me the next question."

[0494] Send a message of encouragement

[0495] In this way, a learning support system that combines a generative AI model and an emotion engine supports users' learning.

[0496] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0497] Step 1:

[0498] A user requests a learning assignment

[0499] The user inputs "What is today's study topic?" through the terminal interface. This input is sent to the server. The input data is the user's request.

[0500] Step 2:

[0501] The server receives the user request

[0502] The server receives a request from a user and analyzes the request content. The input data is the user's request content, and the output data is the analyzed request content.

[0503] Step 3:

[0504] The server sends a prompt to the generative AI model

[0505] The server generates a prompt based on the analyzed request content and sends it to the generative AI model. The input data is the analyzed request content, and the output data is the generated prompt. The specific prompt is "Please tell me what today's learning topic is."

[0506] Step 4:

[0507] Generative AI model generates learning tasks

[0508] The generative AI model analyzes the prompt and generates specific learning tasks by taking into account the user's learning history and progress. The input data is the prompt and the user's learning history and progress data, and the output data is the generated learning task. For example, it generates a task such as, "Today, let's do 10 addition and subtraction problems of fractions in arithmetic."

[0509] Step 5:

[0510] The server presents the generated learning tasks to the user.

[0511] The server presents the learning tasks received from the generative AI model to the user. The input data is the generated learning tasks, and the output data is the learning tasks presented to the user. This presentation is done through the terminal interface.

[0512] Step 6:

[0513] The user solves the learning task

[0514] The user begins to solve the presented learning problem. The answer is entered into the terminal. The input data is the user's answer, and the output data is the answer entered into the terminal.

[0515] Step 7:

[0516] The device collects user emotional data

[0517] The device uses a camera and a microphone to collect emotional data such as the user's facial expression and tone of voice. The input data is the user's facial expression and tone of voice, and the output data is the collected emotional data.

[0518] Step 8:

[0519] The emotion engine analyzes the emotion data

[0520] The emotion engine analyzes the collected emotion data and estimates the user's emotional state. The input data is the collected emotion data, and the output data is the estimated emotional state. For example, if the user is facing a difficult problem, the emotion engine estimates that the user is confused.

[0521] Step 9:

[0522] The server sends the results of the emotion engine to the generative AI model.

[0523] The server sends the estimated emotional state received from the emotion engine to the generative AI model. The input data is the estimated emotional state, and the output data is the emotional state data sent to the generative AI model.

[0524] Step 10:

[0525] Generative AI models adjust how learning tasks are presented

[0526] The generative AI model adjusts the presentation of learning tasks and additional messages based on the estimated emotional state. The input data is the estimated emotional state, and the output data is the adjusted learning tasks and messages. For example, if the user is confused, it generates an encouraging message.

[0527] Step 11:

[0528] The server presents tailored learning tasks and messages to the user.

[0529] The server presents the adjusted learning tasks and messages received from the generative AI model to the user. The input data are the adjusted learning tasks and messages, and the output data are the learning tasks and messages presented to the user. For example, it displays a message such as "Good luck! You're almost there!"

[0530] (Application example 1)

[0531] Next, a description will be given of Application Example 1 of Embodiment Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0532] Conventional learning support systems can generate and present learning tasks based on the user's learning progress and incorrect questions, but they have the problem of being unable to provide feedback or adjust learning tasks that take into account the user's emotional state. Furthermore, when supporting factory workers, there are concerns that a decline in work efficiency and motivation can occur because appropriate feedback and support that takes into account the worker's emotional state is not provided. To solve these problems, a system that can recognize the user's emotional state and adjust learning tasks and work content based on that is needed.

[0533] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0534] In this invention, the server includes means for generating study tasks based on user input using a generation AI and presenting the study tasks to the user, means for adjusting the difficulty of the study tasks based on the user's learning progress and questions that they got wrong, means for grading the user's study results and presenting the results to the user, means for estimating the user's emotional state using an emotion engine and adjusting the method for generating and presenting study tasks based on the emotional state, and means for providing appropriate feedback according to the user's emotional state. This enables learning support and task support that takes the user's emotional state into consideration.

[0535] "Generative AI" is artificial intelligence that generates learning tasks and work content based on input from the user and presents them to the user.

[0536] "Study assignments" are specific learning content or problems that users must address.

[0537] The "emotion engine" is a technology for estimating a user's emotional state from facial expressions, tone of voice, etc.

[0538] "Feedback" refers to evaluations and advice provided to users regarding their learning and work results.

[0539] "User emotional state" refers to the emotional state exhibited by the user, including, for example, tiredness, confusion, joy, etc.

[0540] "Study progress" is an indicator that shows how far the user has progressed in the learning tasks.

[0541] "Marking" is the act of determining whether a user's learning results are correct or incorrect and showing the result.

[0542] A "means" is a method or device for achieving a specific purpose.

[0543] "Presenting" is the act of displaying or informing a user of information.

[0544] "Adjusting difficulty" means changing the difficulty of the learning tasks based on the user's learning progress and the questions they got wrong.

[0545] As an embodiment of the present invention, we will explain an example of a "work support system" installed in a factory robot. This system is designed to improve the work efficiency of factory workers by using generative AI and an emotion engine.

[0546] System configuration

[0547] 1. Hardware:

[0548] Factory robots

[0549] camera

[0550] microphone

[0551] 2. Software:

[0552] Python

[0553] EmotionEngine Library

[0554] TaskGenerator Library

[0555] Program processing explanation

[0556] Generate work

[0557] The server uses the TaskGenerator library to generate work content based on input from the user (factory worker). For example, when a worker asks, "What will you be doing today?", the generation AI generates and presents specific work content such as, "Today, assemble 50 units of part A."

[0558] Emotion Recognition and Feedback

[0559] The server uses the EmotionEngine library to analyze data from the camera and microphone and estimate the worker's emotional state. For example, if the server estimates that the worker is tired, the generative AI will send a message such as "Take a short break."

[0560] Specific examples

[0561] 1. Work generation:

[0562] When a worker asks, "What will you be doing today?", the generative AI generates and presents specific work content, such as "Today, you will assemble 50 units of part A."

[0563] 2. Emotion Recognition and Feedback:

[0564] If the generator estimates that the worker is tired, it will send a message such as "Take a short break."

[0565] Prompt Sentence Examples

[0566] "What are you working on today?"

[0567] User input data

[0568] In this way, it becomes possible to provide learning and work support that takes into account the user's emotional state.

[0569] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0570] Step 1:

[0571] The user types into the terminal, "What will I be working on today?"

[0572] Input: User asks "What are you working on today?"

[0573] Output: User question data

[0574] Step 2:

[0575] The server receives the user's question data and generates task content using the TaskGenerator library.

[0576] Input: User question data

[0577] Data processing: Generate work content using the TaskGenerator library

[0578] Output: Generated work content (e.g., "Today, let's assemble 50 units of part A")

[0579] Step 3:

[0580] The server transmits the generated work content to the terminal and presents it to the user.

[0581] Input: Generated work

[0582] Output: The work presented to the user

[0583] Step 4:

[0584] The user begins working, and the camera and microphone capture the user's facial expressions and tone of voice in real time.

[0585] Input: User's facial expression data, voice tone data

[0586] Output: Captured emotion data

[0587] Step 5:

[0588] The server uses the EmotionEngine library to analyze the captured emotion data and estimate the user's emotional state.

[0589] Input: Captured emotion data

[0590] Data Computation: Estimating Emotional States Using the EmotionEngine Library

[0591] Output: Estimated emotional state (e.g., "Tired")

[0592] Step 6:

[0593] The server uses a generation AI to generate an appropriate feedback message based on the estimated emotional state.

[0594] Input: Estimated emotional state

[0595] Data processing: Generative AI is used to generate feedback messages

[0596] Output: The generated feedback message (e.g., "Let's take a short break")

[0597] Step 7:

[0598] The server transmits the generated feedback message to the terminal and presents it to the user.

[0599] Input: The generated feedback message

[0600] Output: The feedback message that is presented to the user.

[0601] Example 2

[0602] Next, a description will be given of Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0603] Conventional learning systems lack the ability to adjust the difficulty of learning tasks based on the user's learning progress and incorrect questions, making it difficult to maintain the user's motivation to learn. Furthermore, they do not provide learning tasks that take the user's emotional state into consideration, which results in the user not being able to maximize their learning effectiveness. Furthermore, they lack a mechanism for effectively collecting and analyzing the user's learning data and reflecting it in the next learning task.

[0604] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0605] In this invention, the server includes means for generating study tasks based on user input using a generative AI and presenting the study tasks to the user, means for adjusting the difficulty of the study tasks based on the user's learning progress and the questions answered incorrectly, means for grading the user's study results and presenting the results to the user, means for collecting the user's study data and saving it in a database, means for analyzing the collected study data and extracting the user's correct answer rate and trends in the questions answered incorrectly, means for analyzing the user's reactions during study and estimating their emotional state, means for generating prompt sentences to be input into the generative AI model based on the analysis results of the study data and the emotional state, and means for providing the generated study tasks to the user and collecting the user's answers. This makes it possible to provide optimal study tasks that take into account the user's learning progress and emotional state.

[0606] "Generative AI" refers to artificial intelligence that generates learning tasks based on user input and presents them to the user.

[0607] A "user" is an individual who uses the learning system to complete learning tasks.

[0608] "Learning assignments" are questions and exercises created by the generative AI for users to answer.

[0609] "Study progress" is a metric that shows how much progress a user has made through a learning task.

[0610] A "mistake" is a question that the user did not answer correctly in the learning task.

[0611] "Difficulty adjustment" refers to changing the difficulty of the next learning task based on the user's learning progress and the questions they got wrong.

[0612] "Marking" refers to determining whether a user's learning results are correct or incorrect and presenting the results to the user.

[0613] "Learning data" refers to data that includes information such as the results and reactions of users when they answer learning tasks.

[0614] A "database" is a system for storing collected learning data.

[0615] "Analysis" refers to analyzing the collected learning data and extracting the user's correct answer rate and trends in questions they got wrong.

[0616] "Emotional state" refers to the emotions a user exhibits while learning, including, for example, happiness, disappointment, excitement, etc.

[0617] A "prompt sentence" is an instruction sentence to be input into a generative AI model, and is generated based on the results of an analysis of the learning data and emotional state.

[0618] "Answer result" refers to the result of the answer given by the user to the learning task.

[0619] The present invention is a system that provides optimal study tasks in consideration of a user's learning progress and emotional state. Specific embodiments of this system will be described below.

[0620] The server collects the results of the questions that users answer on the learning platform. Specifically, it records data such as the questions that users answered correctly, the questions that they answered incorrectly, and the time it took to answer, and stores this data in a database. This data collection is done using programming languages ​​such as Python.

[0621] Next, the server analyzes the collected training data. Using the Python pandas library, it analyzes the user's accuracy rate and trends in questions they got wrong. For example, if a user makes a lot of mistakes when adding fractions, it can extract that information.

[0622] Furthermore, the server uses an emotion engine to analyze the user's reactions while studying. A general emotion recognition service is used as the emotion recognition software. While the user is answering the study questions, the device's camera and microphone record the user's facial expressions and voice, and this data is sent to the server in real time. The emotion engine estimates the user's emotional state, such as whether they are disappointed about a question they got wrong.

[0623] The server generates prompt sentences to be input to the generative AI model based on the learning data and the analysis results of the emotion engine. For example, it generates prompt sentences like the following:

[0624] According to your learning data, you're making a lot of mistakes with fraction addition. Please focus on fraction addition in your next learning exercise. Also, please lower the difficulty a bit, as users are frustrated with the problems they get wrong.

[0625] The server inputs the generated prompt sentence into a generative AI model to generate the next learning task. The generative AI model uses common generative AI technology. The generative AI model creates learning tasks appropriate for the user based on the prompt sentence. For example, it generates a learning task that includes many problems related to adding fractions.

[0626] Finally, the device provides the generated learning tasks to the user. The user answers the learning tasks, and the device sends the results back to the server. The server uses the collected feedback to help generate the next learning task. For example, it analyzes the trends in questions the user recently got wrong and reflects this in the next learning task.

[0627] In this way, it is possible to provide optimal learning tasks that take into account the user's learning progress and emotional state.

[0628] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0629] Step 1:

[0630] The server collects the results of questions answered by users on the learning platform. Specifically, it records data such as the questions answered correctly and incorrectly by the user, and the time it took to answer, and stores this data in a database. The input is the user's answer, and the output is the learning data stored in the database.

[0631] Step 2:

[0632] The server analyzes the collected training data. It uses the Python pandas library to analyze the user's correct answer rate and the tendency of questions that were answered incorrectly. The input is the training data stored in the database, and the output is the analysis results of the user's correct answer rate and the tendency of questions that were answered incorrectly.

[0633] Step 3:

[0634] The server analyzes the user's reactions during learning using an emotion engine. Using emotion recognition software, it infers the user's emotional state from their facial expressions and voice. The input is the user's facial expressions and voice data sent from the device, and the output is the inferred emotional state.

[0635] Step 4:

[0636] The server generates prompt sentences to be input to the generative AI model based on the learning data and the analysis results of the emotion engine. Specifically, it generates prompt sentences like the following:

[0637] Based on your learning data, you've made a lot of mistakes with fraction addition. Please focus on fraction addition in your next learning exercise. Also, please lower the difficulty a bit, as your users are disappointed with the problems they get wrong.

[0638] The input is the training data and the analysis result of the emotional state, and the output is the generated prompt sentence.

[0639] Step 5:

[0640] The server inputs the generated prompt sentence into the generative AI model to generate the next learning task. The generative AI model creates a learning task appropriate for the user based on the prompt sentence. The input is the generated prompt sentence, and the output is the generated learning task.

[0641] Step 6:

[0642] The terminal provides the generated learning tasks to the user. The user answers the learning tasks, and the terminal sends the results back to the server. The input is the generated learning tasks, and the output is the user's answer.

[0643] Step 7:

[0644] The server uses the collected feedback to help generate the next study assignment. For example, it analyzes the trends of questions that the user has recently gotten wrong and reflects this in the next study assignment. The input is the user's answer, and the output is the analysis results that will help generate the next study assignment.

[0645] (Application example 2)

[0646] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0647] Conventional learning support systems have a function to adjust the difficulty of learning tasks based on the user's learning progress and the questions they got wrong, but they do not take into account the user's emotional state, which can lead to users feeling stressed or losing motivation to learn. Furthermore, in work environments such as factories, there is a lack of functionality to monitor the progress of workers and the frequency of mistakes in real time and adjust the work content and difficulty, which can lead to problems with reduced work efficiency.

[0648] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0649] In this invention, the server includes means for generating learning tasks based on user input using a generation AI and presenting the learning tasks to the user, means for adjusting the difficulty of the learning tasks based on the user's learning progress and questions answered incorrectly, means for marking the user's learning results and presenting the results to the user, means for estimating the user's emotional state using an emotion engine and adjusting the difficulty of the learning tasks based on the results, and means for monitoring the worker's progress and frequency of mistakes in real time and adjusting the task content and difficulty level. This makes it possible to provide learning tasks that take the user's emotional state into consideration and to optimize the task content in work environments such as factories.

[0650] "Generative AI" is artificial intelligence that generates learning tasks based on user input and presents them to the user.

[0651] "Study assignments" are specific problems or exercises that users must complete.

[0652] "User's learning progress" is an indicator that shows how far the user has progressed in the learning tasks.

[0653] "Wrong questions" are learning tasks that the user was unable to answer correctly.

[0654] "Difficulty of learning task" is an index that indicates how difficult the learning task is.

[0655] "Marking" is the act of determining whether a user's learning results are correct or incorrect and providing feedback.

[0656] An "emotion engine" is a technology for estimating a user's emotional state.

[0657] "Emotional state" refers to the emotions a user experiences while learning or working.

[0658] "Worker progress" is an index that indicates how far a worker has progressed in the work.

[0659] "Frequency of mistakes" refers to the number and frequency of mistakes that workers make while working.

[0660] "Real-time monitoring" means monitoring the ongoing situation immediately.

[0661] "Work content" refers to the specific work that a worker will perform.

[0662] "Work environment" refers to the location and conditions under which work is performed.

[0663] The system for implementing this invention includes a generative AI, an emotion engine, and a real-time monitoring function. Specific embodiments of the system are described below.

[0664] System configuration

[0665] 1. Generation AI:

[0666] The server generates learning tasks based on user input and presents them to the user. The AI ​​analyzes the user's learning progress and the questions they got wrong, and adjusts the difficulty of the next learning task.

[0667] 2. Emotion Engine:

[0668] The server uses an emotion engine to estimate the user's emotional state. The emotion engine receives the user's reaction data (e.g., sigh, frown, etc.) as input and outputs the user's emotional state.

[0669] 3. Real-time monitoring:

[0670] The server monitors the worker's progress and frequency of mistakes in real time and adjusts the task content and difficulty accordingly, providing optimal tasks according to the worker's skill level and emotional state.

[0671] Hardware and software used

[0672] Hardware:

[0673] Server: Processes data and runs generative AI.

[0674] User device: Presents learning tasks and collects user response data.

[0675] Sensors: Used to monitor workers' progress and frequency of mistakes in real time.

[0676] software:

[0677] Generative AI model: Runs algorithms to generate learning tasks and adjust their difficulty.

[0678] Emotion engine: Runs algorithms to estimate the user's emotional state.

[0679] Real-time monitoring system: Monitors workers' progress and frequency of mistakes in real time and sends the data to a server.

[0680] Specific examples

[0681] For example, if a user is studying mathematics, the generative AI will analyze the questions the user got wrong and focus on questions from the same subject for the next study assignment. The emotion engine will also analyze the user's reaction data and, if the user is feeling stressed, will lower the difficulty of the next study assignment.

[0682] In the factory work environment, a real-time monitoring system monitors the progress of workers and the frequency of mistakes, and an emotion engine estimates the worker's emotional state. If a worker is feeling stressed, the system adjusts the task, such as by lowering the difficulty of the work.

[0683] Prompt Sentence Examples

[0684] Below are some example prompts to input to the generative AI model:

[0685] Create a program that adjusts the difficulty of the next task based on the worker's progress and emotional state. This system takes the worker's progress (progress rate and frequency of mistakes) and emotional state (reactions such as sighing or frowning) as input and adjusts the difficulty of the next task. Using an emotion engine and an AI task generator, it estimates the worker's emotional state and generates the next task.

[0686] In this way, it is possible to provide learning tasks that take into account the user's learning progress and emotional state, and to optimize work content in work environments such as factories.

[0687] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0688] Step 1:

[0689] The server receives input from the user and sends it to the generative AI model. The input data includes the results of the learning tasks the user answered and reaction data. The generative AI model generates the next learning task based on this input data. Specifically, it analyzes the questions the user got wrong and their learning progress, and selects learning tasks of an appropriate level of difficulty.

[0690] Step 2:

[0691] The server sends the generated learning tasks to the user's device. The user's device presents the received learning tasks to the user. The user then works on the presented learning tasks. In this step, how the user responds to the learning tasks is recorded in real time.

[0692] Step 3:

[0693] The user device sends the user's learning results and reaction data to the server. The server uses an emotion engine to estimate the user's emotional state. The emotion engine receives the user's reaction data (e.g., sighing, frowning, etc.) as input and outputs the user's emotional state. Emotional states include stress, disappointment, and satisfaction.

[0694] Step 4:

[0695] The server adjusts the difficulty of the next learning task based on the output of the emotion engine. Specifically, it lowers the difficulty if the user is feeling stressed and raises it if the user is satisfied. This adjustment result is fed back to the generative AI model and reflected in the generation of the next learning task.

[0696] Step 5:

[0697] The server uses a real-time monitoring system in the factory's work environment to monitor the progress of workers and the frequency of mistakes. It receives data from sensors and analyzes the progress rate and frequency of mistakes in real time, providing optimal work content according to the worker's skill level and emotional state.

[0698] Step 6:

[0699] The server adjusts the task content and difficulty level based on data obtained from the real-time monitoring system. It uses an emotion engine to estimate the worker's emotional state and lowers the difficulty of the task if the worker is feeling stressed. The adjusted task content is sent to the worker's device and presented to the worker.

[0700] Step 7:

[0701] The server integrates all data and comprehensively evaluates the learning progress and work efficiency of users and workers. This provides data for optimizing long-term learning and work plans. The evaluation results are also used to improve the generative AI model and emotion engine algorithms.

[0702] In this way, it is possible to provide learning tasks that take into account the user's learning progress and emotional state, and to optimize work content in work environments such as factories.

[0703] Example 3

[0704] Next, a third embodiment of the third embodiment will be described. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0705] Conventional learning support systems provide uniform evaluations and feedback on users' answers, making it difficult to respond to individual users' emotional states. It was also difficult to accurately grasp a user's learning progress and level of understanding and provide appropriate feedback.

[0706] The identification process by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means. In this invention, the server includes means for generating study tasks based on user input using a generation AI and presenting the study tasks to the user, means for adjusting the difficulty of the study tasks based on the user's learning progress and questions that were answered incorrectly, means for marking the user's learning results and presenting the results to the user, means for adjusting feedback based on the user's emotional state provided by the emotion engine, means for evaluating the user's answers using a generation AI model and determining whether they are correct, and means for generating feedback using the generation AI model that takes the user's emotional state into consideration. This makes it possible to accurately grasp the user's learning progress and level of understanding and provide appropriate feedback according to the individual emotional state.

[0707] "Generative AI" is artificial intelligence that generates learning tasks based on user input and evaluates the user's answers.

[0708] "Study assignments" are questions or exercises for users to answer.

[0709] "User learning progress" is a metric that indicates how much progress a user has made through a learning task.

[0710] "Difficulty of learning task" is a level that indicates how difficult the learning task is for the user.

[0711] "Marking" is an evaluation action that determines whether the user's answer is correct or incorrect.

[0712] An "emotion engine" is a system that detects a user's emotional state and provides that information.

[0713] "Feedback" is the evaluation and advice provided to a user on their answers and learning progress.

[0714] A "generative AI model" is an artificial intelligence model that generates evaluations and feedback based on the user's answers and emotional state.

[0715] "Determining correctness" means determining whether the user's answer is correct or incorrect.

[0716] "Emotional state" refers to the psychological state, such as joy or depression, that a user feels while learning.

[0717] This invention is a learning support system that tracks a user's learning progress and deepens their understanding. The system uses a generative AI model and an emotion engine to evaluate the user's answers and provide feedback.

[0718] The server receives the answers to the problems solved by the user and inputs them into the generative AI model. For example, a general artificial intelligence model is used as the generative AI model. The generative AI model evaluates the user's answer and determines whether it is correct or incorrect. The server then presents the evaluation results from the generative AI model to the user. For problems that were answered incorrectly, an explanation is provided along with the correct answer.

[0719] Furthermore, the server uses an emotion engine to obtain the user's current emotional state. The emotion engine can use, for example, a general emotion recognition API. The generative AI model takes this emotional state into account and adjusts the feedback and encouragement provided to the user. For example, if the user feels joy after completing a learning task, the generative AI model provides feedback that reinforces that emotion.

[0720] As a concrete example, let's say a user solves a math problem. If the user answers "2 + 2 = 5," the server inputs this answer into the generative AI model. The generative AI model determines that "2 + 2 = 4" is the correct answer and tells the user, "Your answer is wrong. The correct answer is '2 + 2 = 4'." If the emotion engine detects that the user is feeling down, the generative AI model provides encouraging feedback such as, "It's okay to make a mistake. You'll definitely get it right next time!"

[0721] Example prompt sentence:

[0722] User Answer: 2 + 2 = 5

[0723] Correct answer: 2 + 2 = 4

[0724] User's emotional state: Depressed

[0725] Prompt to generative AI model: The user's answer is "2 + 2 = 5". The correct answer is "2 + 2 = 4". The user is currently feeling depressed. Please provide encouraging feedback to the user.

[0726] In this way, the server, the terminal, and the user work together to grasp the user's learning progress and provide support for deepening the user's understanding. The flow of the identification process in the third embodiment will be described with reference to FIG.

[0727] Step 1:

[0728] The user solves the problem

[0729] The user solves a learning task (e.g., a math problem) presented on the device. For example, the answer to the question "2 + 2 = ?" is "5." The input is the user's answer "5," and the output is the answer data entered into the device.

[0730] Step 2:

[0731] The device sends the user's answer to the server

[0732] The terminal sends the answers entered by the user to the server, where the answer data is properly formatted and sent in a form that the server can process. The input is the user's answer data, and the output is the answer data sent to the server.

[0733] Step 3:

[0734] The server inputs the answer into the generative AI model

[0735] The server inputs the received user answer into the generative AI model. For example, it generates and sends a prompt to the generative AI model saying, "The user's answer is '2 + 2 = 5'. Please tell me the correct answer." The input is the user's answer data, and the output is the prompt sent to the generative AI model.

[0736] Step 4:

[0737] The generative AI model evaluates the answer and returns the result to the server

[0738] The generative AI model evaluates the user's answer and returns the correct answer and the evaluation result to the server. For example, it may return a result such as "The correct answer is '2 + 2 = 4'. The user's answer is incorrect." The input is the prompt sent to the generative AI model, and the output is the evaluation result returned to the server.

[0739] Step 5:

[0740] The server presents the evaluation results to the user.

[0741] The server presents the evaluation results received from the generative AI model to the user. For example, it sends a message to the terminal saying, "Your answer is incorrect. The correct answer is '2 + 2 = 4'" and displays it to the user. The input is the evaluation result from the generative AI model, and the output is the evaluation result presented to the user.

[0742] Step 6:

[0743] The server obtains the user's emotional state from the emotion engine.

[0744] The server uses the emotion engine to obtain the user's current emotional state. The emotion engine detects that the user is depressed and returns the result "The user is currently depressed" to the server. The input is the user's current state, and the output is the emotional state data from the emotion engine.

[0745] Step 7:

[0746] The server allows the generative AI model to generate feedback that takes into account the emotional state.

[0747] The server inputs the user's emotional state obtained from the emotion engine into the generative AI model, which then generates feedback that takes the emotional state into account. For example, a prompt such as "The user is currently feeling depressed. Please provide encouraging feedback" is sent to the generative AI model. The input is the emotional state data from the emotion engine, and the output is the prompt sent to the generative AI model.

[0748] Step 8:

[0749] The server provides feedback to the user

[0750] The generative AI model generates feedback such as "It's okay if you make a mistake. You'll definitely get it right next time!" and returns it to the server. The server sends this feedback to the device, which then displays it to the user. The input is the feedback from the generative AI model, and the output is the feedback presented to the user.

[0751] (Application example 3)

[0752] Next, a description will be given of Application Example 3 of Form Example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0753] Conventional learning systems have difficulty accurately grasping a user's learning progress and level of understanding and providing appropriate feedback. They also lack the ability to provide feedback and encouragement that takes into account the user's emotional state, which can lead to insufficient motivation to learn. Furthermore, learning management within the virtual store is inadequate, leaving a lack of means to improve the user's learning experience.

[0754] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes: means for generating study tasks based on user input using a generation AI and presenting the study tasks to the user; means for adjusting the difficulty of the study tasks based on the user's learning progress and questions that the user got wrong; means for grading the user's learning results and presenting the results to the user; means for analyzing the user's emotional state using an emotion engine and providing feedback and encouragement based on the emotional state; and means for managing the user's learning progress using a smart device when studying in the virtual store. This makes it possible to accurately grasp the user's learning progress and level of understanding and provide appropriate feedback and encouragement, thereby increasing motivation to study and improving the learning experience in the virtual store.

[0755] "Generative AI" is an artificial intelligence system that generates learning tasks based on user input and presents them to the user.

[0756] "Learning tasks" are problems or tasks that users must solve, and are generated by generative AI.

[0757] "User's learning progress" refers to the progress a user makes in the course of learning.

[0758] A "wrong question" is a question that the user answered incorrectly.

[0759] "Difficulty of learning task" indicates the degree of difficulty of the learning task, and is adjusted based on the user's learning progress and the questions they got wrong.

[0760] "Marking" refers to determining whether a user's learning results are correct or incorrect and evaluating them.

[0761] An "emotion engine" is a system for analyzing a user's emotional state.

[0762] "Emotional state" refers to the emotions such as joy, sadness, excitement, etc. that a user feels while learning.

[0763] "Feedback" refers to evaluations and comments provided on a user's learning results.

[0764] "Encouragement" refers to words of encouragement or messages provided to motivate users to learn more.

[0765] A "virtual store" is a virtual store that exists on the Internet and provides an environment for users to study.

[0766] "Smart devices" refer to electronic devices with advanced functions such as smartphones and head-mounted displays.

[0767] "Managing learning progress" means tracking a user's learning progress and making appropriate adjustments.

[0768] The system for implementing this invention includes a generative AI, an emotion engine, a smart device, and a virtual store. The specific operation of each component and their interaction will be described below.

[0769] 1. Generating learning tasks using generative AI

[0770] The server uses a generation AI to generate learning tasks based on user input. When the user enters an answer, the generation AI analyzes the answer and generates the next learning task to be presented. The generation AI adjusts the difficulty of the learning task, taking into account the user's learning progress and the questions they got wrong.

[0771] 2. Marking and presenting learning results

[0772] When the user inputs the answer to the problem they solved into the AI ​​generator, the AI ​​marks the answers based on the answers. The marks are then presented to the user, allowing them to understand their own learning progress.

[0773] 3. Emotion analysis and feedback using an emotion engine

[0774] The emotion engine analyzes the user's emotional state. After the user completes a learning task, the emotion engine detects the user's emotional state from their facial expressions and voice. Based on the detected emotional state, the generative AI provides appropriate feedback and encouragement. For example, if the user is feeling happy, the generative AI will provide feedback such as, "Great! Keep it up!"

[0775] 4. Learning management within the virtual store

[0776] Users study in a virtual store. Smart devices such as smartphones and head-mounted displays are used to manage their learning progress. The virtual store provides an environment for users to solve learning tasks, improving their learning experience.

[0777] Specific examples

[0778] The user inputs answers such as "1. 2+2=4," "2. 3+5=8," and "3. 7-2=5." If the emotion engine interprets the user's facial expression as "happy," the generative AI provides feedback such as "Great! Keep it up!"

[0779] Prompt Sentence Examples

[0780] Please circle your answer below:

[0781] 1. 2+2=4

[0782] 2. 3+5=8

[0783] 3. 7-2=5

[0784] In this way, users can track their learning progress and deepen their understanding, and the feedback provided by the emotion engine enhances their motivation to learn and improves the learning experience within the virtual store.

[0785] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0786] Step 1:

[0787] The user solves the learning tasks. The user uses a smart device to input answers to the learning tasks presented in the virtual store. The input answers are sent to the server.

[0788] Step 2:

[0789] The server uses a generation AI to analyze the user's answers. Based on the input answers, the generation AI marks them. Specifically, the generation AI generates a prompt and evaluates the user's answers. For example, it generates a prompt such as "Please mark the following answers: 1. 2+2=4 2. 3+5=8 3. 7-2=5." The analysis results are judged as correct or incorrect, and the result is saved on the server.

[0790] Step 3:

[0791] The server uses an emotion engine to analyze the user's emotional state. The user's facial expressions and voice data are sent from the smart device to the server, and the emotion engine analyzes them. The analysis result is the user's emotional state (e.g., "happy" or "sad").

[0792] Step 4:

[0793] The server uses a generation AI to provide feedback to the user. Based on the emotional state obtained from the emotion engine, the generation AI generates appropriate feedback and encouraging messages. For example, if the user is analyzed as "happy," feedback such as "Great! Keep it up!" is generated. The generated feedback is stored on the server.

[0794] Step 5:

[0795] The server provides feedback to the user. The generated feedback and marking results are sent to the smart device and displayed to the user. The user can grasp their own learning progress and level of understanding and receive appropriate feedback.

[0796] Step 6:

[0797] The server adjusts the difficulty of the next learning task based on the user's learning progress and the questions they got wrong. The generation AI generates the next learning task based on the user's learning history and adjusts its difficulty. The generated learning task is saved on the server and presented to the user the next time they study.

[0798] In this way, users can effectively learn within the virtual store, and feedback from the emotion engine increases their motivation to learn.

[0799] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0800] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0801] Another example of generative AI is Gemini (registered trademark) (Internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[0802] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0803] [Second embodiment]

[0804] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0805] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0806] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0807] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0808] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0809] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0810] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0811] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0812] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0813] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0814] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0815] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0816] "Example 1"

[0817] One embodiment of the present invention is a learning support system that uses a generative AI. This system generates learning tasks based on user input and presents the task to the user. Specifically, when the user asks, "What is today's learning task?", the generative AI generates a specific learning task such as, "Today, let's do 10 math problems on adding and subtracting fractions," and presents it to the user.

[0818] "Example 2"

[0819] Furthermore, this system adjusts the difficulty of the learning tasks based on the user's learning progress and the problems they got wrong. For example, if the user gets many fraction addition problems wrong, the generation AI will optimize the learning content by focusing on fraction addition as the next learning task. "Example 3"

[0820] The system also marks the user's learning results and presents the results to the user. Specifically, the answers to the problems the user solves are input into the generation AI, which then marks the answers based on those answers. The results are then presented to the user, and for any problems the user got wrong, an explanation is provided along with the correct answer. This allows the user to understand their own learning progress and deepen their understanding.

[0821] The processing flow of each embodiment will be described below.

[0822] "Example 1"

[0823] Step 1: The user asks the system, "What is today's study topic?"

[0824] Step 2: The generative AI receives input from the user and generates learning tasks, such as "Today, let's do 10 addition and subtraction problems with fractions in arithmetic."

[0825] Step 3: The learning task generated by the generative AI is presented to the user.

[0826] "Example 2"

[0827] Step 1: After the user completes the learning task, they enter the results into the system.

[0828] Step 2: The generative AI receives input from the user and analyzes the questions they got wrong.

[0829] Step 3: The AI ​​adjusts the difficulty of the next lesson. For example, if the user makes many mistakes on fraction addition, the AI ​​will optimize the learning content by focusing on fraction addition in the next lesson.

[0830] "Example 3"

[0831] Step 1: The user enters the answer to the problem into the system.

[0832] Step 2: The generative AI receives input from the user and marks the answers.

[0833] Step 3: The AI ​​generator presents the results of the marking to the user. For incorrect questions, it provides an explanation along with the correct answer. This allows the user to understand their own learning progress and deepen their understanding.

[0834] Example 1

[0835] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0836] Conventional learning support systems have difficulty providing appropriate learning tasks that correspond to the user's learning progress and individual needs. Furthermore, they lack effective means of providing feedback on the user's learning results or motivating them to learn. This makes it difficult for users to progress efficiently in their studies.

[0837] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0838] In this invention, the server includes means for receiving input from a user, means for analyzing the received input data and sending it as a prompt to a generative AI model, means for generating study tasks based on the user's input using the generative AI model and presenting the study tasks to the user, means for adjusting the difficulty of the study tasks based on the user's learning progress and questions that the user got wrong, means for grading the user's study results and presenting the results to the user, means for sending the generated study tasks to the user, and means for displaying the study tasks to the user. This makes it possible to provide appropriate study tasks according to the individual needs of the user and to increase motivation to learn through feedback on the study results.

[0839] "Means for receiving input from the user" refers to the function of providing an interface for the user to input information to the system in the form of text, voice, etc., and receiving that input.

[0840] "Means for analyzing received input data and sending it to the generative AI model as a prompt" refers to the function of analyzing input data received from a user and sending the analysis results to the generative AI model as a prompt in a format that the generative AI model can understand.

[0841] "Means for generating learning tasks based on user input using a generative AI model and presenting the learning tasks to the user" refers to the function of a generative AI model generating appropriate learning tasks based on user input and presenting the generated learning tasks to the user.

[0842] "A means to adjust the difficulty of the learning task based on the user's learning progress and the questions they got wrong" refers to a function that takes into account the user's past learning history and the questions they got wrong and appropriately adjusts the difficulty of the next learning task presented.

[0843] "Means for marking the user's learning results and presenting the results to the user" refers to a function that judges whether the learning tasks answered by the user are correct or incorrect and presents the results to the user as feedback.

[0844] "Means for sending generated learning tasks to users" refers to the function of sending learning tasks generated by the generative AI model to the user's device.

[0845] "Means for displaying study assignments to the user" refers to a function that visually displays received study assignments on the user's device.

[0846] This invention relates to a learning support system using a generative AI model. This system has the function of generating learning tasks based on user input and presenting the learning tasks to the user. Specific embodiments of this system are described below.

[0847] Hardware and software used

[0848] The system uses the following hardware and software:

[0849] Server: Receives input from the user, sends prompts to the generative AI model, and sends the generated learning tasks to the user.

[0850] Terminal: Provides an interface for users to input information and displays the learning tasks received from the server. Terminals include smartphones, tablets, and PCs.

[0851] Generative AI model: Generates learning tasks based on user input. Specifically, we use a generative AI model that uses natural language processing techniques (e.g., OpenAI's GPT-4).

[0852] System Operation

[0853] Receiving user input

[0854] A user uses a terminal to provide text input to the system, for example, a user might type, "What's my lesson today?" This input is sent to the server through the terminal interface.

[0855] Parsing input data and sending prompts

[0856] The server analyzes the user's input data received from the device. The analyzed data is sent to the generative AI model as a prompt. For example, the input "What is today's study topic?" is sent to the generative AI model as a prompt.

[0857] Generating learning tasks

[0858] The generative AI model generates learning tasks based on the received prompts. The generated learning tasks are specific and actionable for the user. For example, a learning task might be generated such as, "Today, let's do 10 addition and subtraction problems of fractions in arithmetic."

[0859] Submit and view coursework

[0860] The server sends the learning tasks received from the generative AI model to the device, which then displays the received learning tasks to the user. For example, the smartphone screen might display the following text: "Today, let's do 10 math problems on adding and subtracting fractions."

[0861] Specific examples

[0862] Example 1

[0863] The user opens the app on their smartphone, enters "What is today's lesson topic?" and presses the send button. The device sends this input to the server. The server receives the request and sends a prompt to the generative AI model saying, "What is today's lesson topic?" The generative AI model generates a lesson topic, such as "Today, let's do 10 math problems on adding and subtracting fractions," and returns it to the server. The server sends this lesson topic to the device, which displays it on the app screen.

[0864] Example 2

[0865] The user opens a web browser on their computer, enters "What is today's study topic?" and presses the send button. The device sends this input to the server. The server receives the request and sends a prompt to the generative AI model saying "What is today's study topic?" The generative AI model generates a study topic such as "Today, let's research Japan's Warring States period in history," and returns it to the server. The server sends this study topic to the device, which displays it on the web browser screen.

[0866] Prompt Sentence Examples

[0867] Below are some example prompts to input to a generative AI model:

[0868] "What's today's lesson?"

[0869] "Please tell me what we'll be studying tomorrow."

[0870] "What should I study for my next test?"

[0871] Using these prompts, the generative AI model can provide the user with appropriate learning tasks.

[0872] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0873] Step 1:

[0874] Receiving user input

[0875] A user uses a terminal to input text into the system. For example, the user types, "What is today's study topic?" This input is sent to the server through the terminal interface. The input data is in text format, and the terminal sends this data to the server as an HTTP POST request.

[0876] Step 2:

[0877] Receiving and parsing input data

[0878] The server receives the user's input data sent from the device. The received data is analyzed by the text analysis module. As a result of the analysis, the user's intent is understood and a prompt is generated to be sent to the generative AI model. For example, the input "What are your study goals for today?" is converted into the prompt "Please tell me your study goals for today."

[0879] Step 3:

[0880] Sending prompts to generative AI models

[0881] The server sends the parsed prompt to the generative AI model. This prompt must be in a format that the generative AI model can understand. The generative AI model receives the prompt and performs data calculations to generate an appropriate learning task. For example, the prompt sent might be, "Please tell me today's learning task."

[0882] Step 4:

[0883] Generating learning tasks

[0884] The generative AI model generates learning tasks based on the received prompt. The generated learning tasks are specific and actionable for the user. For example, a learning task might be generated such as, "Today, let's do 10 addition and subtraction problems of fractions in arithmetic." The generative AI model returns this learning task to the server.

[0885] Step 5:

[0886] Receiving and sending assignments

[0887] The server receives the learning task from the generative AI model. It then prepares the received learning task to send to the user. Specifically, it formats the learning task as an HTTP response and prepares to send it to the device. For example, it prepares the text "Today, let's do 10 math problems on adding and subtracting fractions" as an HTTP response.

[0888] Step 6:

[0889] Submit your coursework

[0890] The server sends the prepared learning tasks to the device. This is done using a communication protocol such as an HTTP response. For example, the server sends the following text to the device as an HTTP response: "Today, let's do 10 math problems on adding and subtracting fractions."

[0891] Step 7:

[0892] Viewing learning assignments

[0893] The device receives the learning tasks from the server and displays them to the user. The way they are displayed depends on the interface the user is accessing. For example, a smartphone screen might display the following text: "Today, let's do 10 math problems on adding and subtracting fractions."

[0894] (Application example 1)

[0895] Next, a description will be given of Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0896] Conventional learning support systems can only be used by users at home or in specific locations, and do not provide sufficient learning support in physical stores. Furthermore, they lacked the functionality to allow users to generate and display learning tasks in physical stores, resulting in low learning efficiency and convenience. Furthermore, they lacked the functionality to adjust the difficulty of learning tasks based on the user's learning progress or incorrect questions, making it difficult to meet individual learning needs. To address these issues, there is a need for a system that strengthens learning support in physical stores and allows users to generate and display learning tasks using their mobile information devices.

[0897] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0898] In this invention, the server includes means for generating study tasks based on input from the user using a generation AI and presenting the study tasks to the user, means for adjusting the difficulty of the study tasks based on the user's learning progress and questions that they have gotten wrong, means for grading the user's learning results and presenting the results to the user, means for the user to generate and present study tasks using a mobile information terminal in a physical store, and means for displaying the generated study tasks to the user in the physical store. This enhances learning support in physical stores and enables users to efficiently generate and present study tasks using their mobile information terminals.

[0899] "Generative AI" is an artificial intelligence system that generates learning tasks based on user input and presents them to the user.

[0900] "User" means an individual or organization that receives learning assignments using the learning support system.

[0901] A "learning assignment" is a collection of specific learning content or problems that a user must tackle.

[0902] A "mobile information terminal" is a portable information processing device such as a smartphone or tablet.

[0903] A "physical store" is a commercial establishment or service provider that exists in a physical location and can be visited by users.

[0904] "Study progress" is an indicator that shows how far the user has progressed in the learning tasks.

[0905] "Difficulty adjustment" is the process of changing the difficulty of a learning task based on the user's learning progress and the questions they get wrong.

[0906] "Marking" is the process of determining whether a user's learning results are correct or incorrect and evaluating them.

[0907] "Feedback" refers to evaluations and advice provided to users regarding their learning results.

[0908] "Learning motivation" refers to the degree of interest and motivation a user has in learning.

[0909] A system for implementing this invention uses a generative AI model to generate study tasks based on user input and present them to the user. The system adjusts the difficulty of the study tasks based on the user's learning progress and incorrect questions, marks the user's learning results, and presents the results to the user. The system also has a function that allows users to generate and present study tasks using their mobile information terminals in physical stores.

[0910] The server receives input from the user using a generative AI model and generates a prompt sentence. For example, if the user inputs "What is today's study topic?", the server generates the prompt sentence "The user input 'What is today's study topic?' Please generate today's study topic." The generated prompt sentence is sent to the generative AI model, which generates a specific study topic.

[0911] The generated learning tasks are sent from the server to the user's mobile information device and displayed on the device. If the user is in a physical store, they can check and work on the learning tasks through their device. The progress of the learning tasks and any questions that they got wrong are sent to the server, which then adjusts the difficulty of the learning tasks based on this information.

[0912] The server automatically marks the user's learning results and presents the results to the user, allowing the user to check their learning progress in real time.The server also provides feedback based on the user's learning results, increasing the user's motivation to learn.

[0913] As a concrete example, if a user visits a brick-and-mortar bookstore and uses a mobile information terminal to type, "What is today's study assignment?", the server will use the generative AI model to generate a study assignment such as, "Today, let's do 10 math problems on adding and subtracting fractions," and display it on the terminal. The user can then work on this assignment and send the results back to the server, which will adjust the difficulty of the next study assignment.

[0914] Example prompt sentence:

[0915] The user types, "What's today's lesson plan?" Generate today's lesson plan.

[0916] In this way, learning support in brick-and-mortar stores is enhanced, and users can efficiently generate and present learning tasks using mobile information terminals.

[0917] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0918] Step 1:

[0919] A user uses a mobile information terminal to input a request for a learning assignment.

[0920] Input: User types "What should I study today?"

[0921] Output: The user's input data.

[0922] Specific operation: The user launches the application on the device, enters "What is today's study topic?" in the text box, and presses the submit button.

[0923] Step 2:

[0924] The device sends the user's input to the server.

[0925] Input: User input data.

[0926] Output: The request data sent to the server.

[0927] Specific operation: The terminal sends the user's input data to the server as an HTTP request.

[0928] Step 3:

[0929] The server receives the user's input and generates a prompt.

[0930] Input: User input data.

[0931] Output: The generated prompt statement.

[0932] Specific operation: The server analyzes the user's input data and generates a prompt sentence such as, "The user entered 'What are today's study topics?' Please generate today's study topics."

[0933] Step 4:

[0934] The server sends prompt sentences to the generative AI model to generate learning tasks.

[0935] Input: The generated prompt statement.

[0936] Output: The generated learning task.

[0937] How it works: The server sends a prompt to a generative AI model (such as OpenAI's GPT-3) to generate a learning task. Based on the prompt, the generative AI model generates a learning task such as "Today, let's do 10 addition and subtraction problems with fractions in arithmetic."

[0938] Step 5:

[0939] The server transmits the generated learning assignments to the terminal.

[0940] Input: The generated learning task.

[0941] Output: The learning task data sent to the device.

[0942] Specific operation: The server sends the generated learning task to the terminal as an HTTP response.

[0943] Step 6:

[0944] The device displays the received learning assignments to the user.

[0945] Input: Learning task data received from the server.

[0946] Output: The learning task that is displayed to the user.

[0947] Specific operation: The device analyzes the received learning task data and displays it on the screen. The user can check the learning task on the device screen.

[0948] Step 7:

[0949] The user works on the learning tasks and enters the results into the device.

[0950] Input: User's learning result data.

[0951] Output: Learning result data input to the terminal.

[0952] Specific operation: The user uses the input form on the device to enter the answer to the learning task and presses the submit button.

[0953] Step 8:

[0954] The device sends the user's learning results to the server.

[0955] Input: User's learning result data.

[0956] Output: Training result data sent to the server.

[0957] Specific operation: The device sends the user's learning result data to the server as an HTTP request.

[0958] Step 9:

[0959] The server marks the user's learning results and generates the results.

[0960] Input: User's learning result data.

[0961] Output: Rounding result data.

[0962] Specific operation: The server analyzes the user's learning result data, determines whether the answer is correct or incorrect, and generates a marking result.

[0963] Step 10:

[0964] The server sends the rounding results to the terminal.

[0965] Input: Rounding result data.

[0966] Output: Rounding result data sent to the terminal.

[0967] Specific operation: The server sends the rounding result data to the terminal as an HTTP response.

[0968] Step 11:

[0969] The terminal displays the received marking results to the user.

[0970] Input: Rounding result data received from the server.

[0971] Output: Rounding results displayed to the user.

[0972] Specific operation: The terminal analyzes the received marking result data and displays it on the screen. The user can check the marking result on the terminal screen.

[0973] Example 2

[0974] Next, a description will be given of Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0975] Conventional learning systems lacked the ability to adjust the difficulty of learning tasks based on the user's learning progress and the questions they got wrong, making it difficult to provide an optimized learning experience for each individual user. Furthermore, they lacked a mechanism for efficiently managing users' answer data and reflecting it in the next learning task, preventing improvements in learning effectiveness. Furthermore, there was a need for a system that could consistently generate and present learning tasks using a generative AI model.

[0976] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0977] In this invention, the server includes a means for storing the user's answer data in a database, a means for sending a prompt to the generative AI model to generate the next learning task, and a means for presenting the generated learning task to the user. This makes it possible to adjust the difficulty of the learning task based on the user's learning progress and questions that they got wrong, providing an optimized learning experience for each individual user.

[0978] "Generative AI" refers to artificial intelligence technology that generates learning tasks based on user input and presents them to the user.

[0979] "User" refers to an individual or group working on a learning task.

[0980] "Study assignments" refer to problems or exercises that users must complete.

[0981] "Study progress" refers to an indicator that shows how far a user has progressed in their studies.

[0982] "Incorrect questions" refer to questions that the user was unable to answer correctly.

[0983] "Difficulty adjustment" refers to changing the difficulty of learning tasks based on the user's learning progress and the questions they got wrong.

[0984] "Marking" refers to determining whether a user's learning results are correct or incorrect.

[0985] "Answer data" refers to information about the answers given by a user to a learning task.

[0986] "Database" refers to an information management system for storing user answer data.

[0987] "Generative AI model" refers to artificial intelligence technology that generates the next learning task based on a prompt.

[0988] A "prompt" refers to an instruction given to a generative AI model to generate the next learning task.

[0989] "Presenting" refers to displaying the generated learning task to the user.

[0990] This invention is a system that adjusts the difficulty of learning tasks based on the user's learning progress and the questions they have answered incorrectly, providing an optimized learning experience for each individual user. Specific embodiments of this system are described below.

[0991] Hardware and software used

[0992] server

[0993] The server is responsible for collecting user answer data and storing it in a database. A database management system such as MySQL is used for the server. It also sends prompts to the generative AI model and processes the data to generate the next learning task. For example, OpenAI's GPT-4 is used as the generative AI model.

[0994] Terminal

[0995] The device provides an interface for users to work on learning tasks. The device is built using web technologies such as HTML, CSS, and JavaScript. The device has the function of sending the user's answers to the server in real time.

[0996] user

[0997] Users work on learning tasks provided through their devices. When they answer the questions, the results are sent to the server and used to adjust the difficulty of the next learning task.

[0998] Explanation of program processing

[0999] Server Processing

[1000] The server receives the user's answer data and stores it in a database such as MySQL. For example, it stores information such as the user ID, question ID, answer result, and answer time. The server then sends a prompt to the generative AI model. The prompt describes what questions the user made the most mistakes on and what questions should be presented as the next learning assignment.

[1001] Processing generative AI models

[1002] The generative AI model generates the next learning task based on the prompt received from the server. For example, if the user makes many mistakes in fraction addition, the next learning task will focus on fraction addition.

[1003] Terminal handling

[1004] The device displays the new learning tasks received from the server to the user, who can then work on the learning tasks again.

[1005] Specific examples

[1006] For example, if a user answers "1 / 2 + 1 / 3 = 5 / 6", the answer is sent to the server and saved in the database as "User ID: 12345, Question ID: 67890, Answer: 5 / 6, Correct / Incorrect: Incorrect, Answer Time: 30 seconds". The server then sends the following prompt to the generative AI model:

[1007] Prompt Sentence Examples

[1008] Users made a lot of mistakes on fraction addition problems, so please focus on fraction addition in the next lesson. Specifically, please include problems like:

[1009] 1. 1 / 2 + 1 / 3 = ?

[1010] 2. 3 / 4 + 2 / 5 = ?

[1011] 3. 5 / 6 + 1 / 2 = ?

[1012] The generative AI model generates the next learning task based on this prompt and returns it to the server. The server then sends the problem set to the device, which then displays it as a new learning task for the user. This makes it possible to adjust the difficulty of the learning task based on the user's learning progress and the questions they got wrong, providing an optimized learning experience for each individual user.

[1013] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1014] Step 1:

[1015] Users work on learning tasks

[1016] The user works on a learning task provided through the device. For example, the user answers a fraction addition problem. The user's answer (e.g., "1 / 2 + 1 / 3 = 5 / 6") is obtained as input. The user's answer data is generated as output.

[1017] Step 2:

[1018] The device sends the user's answer to the server

[1019] The terminal sends the user's answers to the server in real time. Specifically, it sends the user's answer data (e.g., "User ID: 12345, Question ID: 67890, Answer: 5 / 6") to the server. The user's answer data is obtained as input. The answer data sent to the server is generated as output.

[1020] Step 3:

[1021] The server saves the answer data in a database

[1022] The server stores the received answer data in a database. Specifically, it uses a database management system such as MySQL to store information such as user ID, question ID, answer result, and answer time. The answer data sent to the server is obtained as input. The answer data stored in the database is generated as output.

[1023] Step 4:

[1024] The server sends a prompt to the generative AI model

[1025] The server sends a prompt to the generative AI model based on the user's past learning data. Specifically, it generates a prompt such as, "The user made many mistakes on fraction addition problems, so please focus on fraction addition in the next learning assignment," and sends it to the generative AI model. The user's past learning data is obtained as input. The prompt sent to the generative AI model is generated as output.

[1026] Step 5:

[1027] The generative AI model generates the next learning task

[1028] A generative AI model (for example, OpenAI's GPT-4) generates the next training problem based on the received prompt. Specifically, it generates problems such as "1 / 2 + 1 / 3 = ?", "3 / 4 + 2 / 5 = ?", and "5 / 6 + 1 / 2 = ?". The input is the prompt. The output is a generated training problem.

[1029] Step 6:

[1030] The server sends the generated learning assignments to the device.

[1031] The server sends the learning tasks received from the generative AI model to the device. Specifically, it sends the generated problem set to the device in JSON format. The generated learning tasks are obtained as input. The learning tasks sent to the device are generated as output.

[1032] Step 7:

[1033] The device displays new learning tasks to the user.

[1034] The terminal displays the new learning task received from the server to the user. The user can then try the learning task again. As input, the learning task sent from the server is obtained. As output, the learning task displayed to the user is generated.

[1035] (Application example 2)

[1036] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1037] Conventional learning support systems lacked the ability to adjust the difficulty of learning tasks based on the user's learning progress and the questions they got wrong, making it difficult for users to focus on areas in which they were weak. Furthermore, they lacked the ability to record the user's learning progress and optimize the next learning task, which led to problems with reduced learning efficiency. Furthermore, the lack of a function to focus questions on the category in which the user was weakest limited the effectiveness of learning.

[1038] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1039] In this invention, the server includes means for generating study tasks based on user input using a generation AI and presenting the study tasks to the user, means for adjusting the difficulty of the study tasks based on the user's learning progress and questions that they got wrong, means for grading the user's learning results and presenting the results to the user, means for recording the user's learning progress and optimizing the next study task, and means for focusing questions on the category in which the user is weakest. This makes it possible to provide optimal study tasks according to the user's learning progress, improving the efficiency and effectiveness of learning.

[1040] "Generative AI" is an artificial intelligence system that generates learning tasks based on user input and presents them to the user.

[1041] "Learning assignments" are problems or exercises that users must work on, and are generated by generative AI.

[1042] "User's learning progress" is an index that indicates how far the user has progressed in their learning, and represents the progress of their learning.

[1043] "Wrong questions" are questions that the user answered incorrectly, and are data used to evaluate the user's level of understanding.

[1044] "Difficulty of learning task" is an indicator of how difficult the learning task is, and is adjusted based on the user's learning progress and the questions they got wrong.

[1045] "Marking" is an evaluation method that judges whether a user's learning results are correct or incorrect and presents the results to the user.

[1046] "User's learning results" are data that indicate how the user answered the learning tasks, and are the subject of marking.

[1047] "Recording the user's learning progress" means saving data on how far the user has progressed in their studies, which serves as the basis for optimizing the next learning task.

[1048] "Optimizing the next learning task" means providing the next learning task in the most optimal form based on the user's learning progress and the questions they got wrong.

[1049] A "category" is a specific field or theme to which a learning task belongs, and indicates the classification of the content that the user is learning.

[1050] "Focusing on questions" means providing many questions on categories that the user is particularly weak at, and is a means of improving the user's level of understanding.

[1051] As an embodiment of the present invention, the following system is constructed.

[1052] System Overview

[1053] The system adjusts the difficulty of the learning tasks based on the user's learning progress and the questions they got wrong, and has the ability to optimize the next learning task. Specifically, it uses a generative AI model to generate learning tasks based on user input and presents them to the user. It also marks the user's learning results and presents the results to the user.

[1054] Hardware and software used

[1055] Hardware: Smartphones, servers

[1056] Software: Python, database (e.g., SQLite), generative AI model

[1057] Data processing and calculation

[1058] The server receives input data from the user and generates learning tasks using a generative AI model. The generated learning tasks are presented to the user's smartphone. The user works on the learning tasks and sends the results to the server. The server records the user's learning results and processes the data to optimize the next learning task.

[1059] Specific examples

[1060] For example, consider a situation where a user answers a quiz about robot operating procedures. If the user makes a mistake in a particular operating procedure, the next learning task will focus on that operating procedure.

[1061] Example prompt sentence:

[1062] User ID: user1

[1063] Category: robot_operation

[1064] Correct Answer: False

[1065] When this prompt is entered, the server records the user's learning progress and focuses questions related to "robot_operation" on the next learning assignment.

[1066] Processing flow

[1067] 1. The user works on a learning task using a smartphone.

[1068] 2. The user's answer is sent to the server.

[1069] 3. The server records the user's learning progress and the questions they got wrong.

[1070] 4. The generative AI model generates the next learning task and presents it to the user.

[1071] In this way, it is possible to provide optimal learning tasks according to the user's learning progress, thereby improving the efficiency and effectiveness of learning.

[1072] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1073] Step 1:

[1074] Users use their smartphones to complete learning tasks.

[1075] Input: User's answer data (e.g., user ID, category, answer result)

[1076] Output: The user's answer results are sent to the server.

[1077] Specific operation: The user opens the application on their smartphone and enters answers to the presented learning tasks. Once the answers are completed, the answer data is sent to the server.

[1078] Step 2:

[1079] The server receives the user's answer results and records them in a database.

[1080] Input: User's answer data (e.g., user ID, category, answer result)

[1081] Output: User learning progress data recorded in a database

[1082] Specific operation: The server analyzes the received answer data and records the user's learning progress and incorrect questions in a database, thereby accumulating the user's learning history.

[1083] Step 3:

[1084] The server generates the next study task based on the user's learning progress and the questions they got wrong.

[1085] Input: User learning progress data, data on questions answered incorrectly

[1086] Output: Generated next learning task data

[1087] How it works: The server analyzes the user's learning progress and incorrect questions recorded in the database, and uses the generative AI model to generate the next learning task. The generated learning task is adjusted to include a focus on the categories in which the user is weak.

[1088] Step 4:

[1089] The server sends the generated next learning assignment to the user's smartphone.

[1090] Input: Generated next learning task data

[1091] Output: Learning tasks displayed on the user's smartphone

[1092] Specific operation: The server sends the generated learning task data to the user's smartphone, where the user can then work on the task in their next learning session.

[1093] Step 5:

[1094] The user then works on the next learning task and sends the answer to the server again.

[1095] Input: User's answer data (e.g., user ID, category, answer result)

[1096] Output: The user's answer results are sent to the server.

[1097] Specific operation: The user opens the smartphone application and enters the answer to the next learning task that is presented. Once the answer is completed, the answer data is sent back to the server.

[1098] In this way, optimal learning tasks are continuously provided according to the user's learning progress, improving the efficiency and effectiveness of learning.

[1099] Example 3

[1100] Next, a description will be given of Example 3 of Form Example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1101] In conventional learning systems, evaluation and feedback on user answers is often done manually, which is problematic in terms of efficiency. It is also difficult to provide appropriate feedback based on the user's learning progress and level of understanding, which can lead to issues such as insufficient motivation to learn or deepening of understanding. Furthermore, there was a need for a system that could properly process user answer data and quickly present evaluation results.

[1102] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1103] In this invention, the server includes means for generating learning tasks based on user input using a generative AI model and presenting the learning tasks to the user, means for adjusting the difficulty of the learning tasks based on the user's learning progress and questions that the user got wrong, means for marking the user's learning results and presenting the results to the user, means for receiving the user's answer data, generating prompt sentences and sending them to the generative AI model, and means for receiving evaluation results and explanations from the generative AI model, storing them in a database, and presenting them to the user. This enables quick and accurate evaluation of the user's answers, and by providing appropriate feedback, it is possible to improve motivation to learn and deepen understanding.

[1104] A "generative AI model" is a type of artificial intelligence that generates learning tasks and evaluates answers based on input data from users.

[1105] A "user" is an individual or group that uses the learning system to solve learning tasks and input answers.

[1106] "Learning tasks" are problems or exercises that users must solve, and are generated by the generative AI model.

[1107] "Answer data" is information about the answer that the user inputs to the study task.

[1108] A "prompt sentence" is an instruction sentence generated by the server to request the generative AI model to evaluate the answer.

[1109] The "evaluation result" is the result of the generative AI model evaluating the user's answer, including a judgment of correctness or incorrectness and an explanation.

[1110] A "database" is an information management system for storing information such as evaluation results and explanations.

[1111] A "server" is a computer system that receives user answer data and performs processes such as sending it to the generative AI model.

[1112] "Feedback" refers to evaluations and explanations provided to users regarding their learning results, and is intended to encourage the user's learning progress and increase their motivation to learn.

[1113] The present invention relates to a system that allows a user to solve a learning task, evaluates the answer, and provides feedback. A specific embodiment of this system will be described below.

[1114] Hardware and software used

[1115] Hardware: Servers, devices (PCs, smartphones, tablets, etc.)

[1116] Software: Generative AI models (e.g., large-scale language models), web browsers, database management systems (e.g., MySQL)

[1117] System configuration

[1118] 1. The user solves a problem

[1119] The user uses a terminal to access the learning platform and inputs answers to the questions displayed.

[1120] For example, a user answers "4" to "Question 1: 2 + 2 = ?"

[1121] 2. The device sends the answer data to the server

[1122] The terminal transmits the answer data entered by the user to the server.

[1123] Specifically, the terminal uses an HTTP request to send the answer data to the server.

[1124] For example, the answer data "Question 1: 4" is sent to the server.

[1125] 3. The server generates a prompt and sends it to the generative AI model.

[1126] The server generates a prompt sentence based on the received answer data.

[1127] The generated prompt sentence is sent to the generative AI model.

[1128] For example, the prompt sentence "The user's answer is '4'. Please determine whether this answer is correct, and if it is correct, mark it as 'Correct'. If it is incorrect, mark it as 'Incorrect'. Please provide the correct answer and an explanation." is sent to the generative AI model.

[1129] 4. The generative AI model evaluates the answer and returns the result to the server

[1130] The generative AI model evaluates the user's answer based on the prompt sent from the server.

[1131] The generative AI model generates evaluation results and explanations and returns them to the server.

[1132] For example, the generative AI model returns the result "Correct. Explanation: 2 + 2 is 4" to the server.

[1133] 5. The server stores the results in a database and presents them to the user.

[1134] The server stores the evaluation results and explanations received from the generative AI model in a database.

[1135] The server sends the saved results to the user's terminal and presents them to the user.

[1136] For example, the server sends the result "Question 1: Correct. Explanation: 2 + 2 is 4" to the user's terminal.

[1137] 6. The user checks the results and continues learning

[1138] The user checks the results on the device.

[1139] Users will learn the correct answers and explanations for questions they get wrong.

[1140] For example, the user confirms the result "Question 1: Correct. Explanation: 2 + 2 is 4." and moves on to the next question.

[1141] In this way, the user can check whether their answer is correct, and if they are wrong, they can learn the correct answer and explanation. This system makes it possible to efficiently grasp the user's learning progress and deepen their understanding. The flow of the identification process in Example 3 will be described with reference to FIG. 15.

[1142] Step 1:

[1143] The user solves the problem

[1144] Input: The user accesses the learning platform using a device and enters answers to the questions displayed.

[1145] Specific action: For example, the user answers "4" to "Question 1: 2 + 2 = ?"

[1146] Output: The user's answer data is saved on the device.

[1147] Step 2:

[1148] The device sends the answer data to the server

[1149] Input: User's answer data

[1150] Specific operation: The device sends the answer data to the server using an HTTP request.

[1151] Output: For example, the answer data "Question 1: 4" is sent to the server.

[1152] Step 3:

[1153] The server generates a prompt and sends it to the generative AI model.

[1154] Input: Answer data received by the server

[1155] Specific operation: The server generates a prompt based on the received answer data. For example, it generates a prompt like this: "The user's answer is '4'. Please determine whether this answer is correct. If it is correct, mark it as 'Correct'. If it is incorrect, mark it as 'Incorrect'. Please provide the correct answer and an explanation."

[1156] Output: The generated prompt sentence is sent to the generative AI model.

[1157] Step 4:

[1158] The generative AI model evaluates the answer and returns the result to the server

[1159] Input: The prompt received by the generative AI model

[1160] Specific operation: The generative AI model evaluates the user's answer based on the prompt sentence and generates an evaluation result and explanation. For example, it generates a result such as "Correct. Explanation: 2 + 2 is 4."

[1161] Output: The evaluation result and explanation are returned to the server.

[1162] Step 5:

[1163] The server stores the results in a database and presents them to the user.

[1164] Input: Evaluation results and explanations received by the server

[1165] Specific operation: The server saves the evaluation results and explanations in a database and sends the saved results to the user's device. For example, it sends the result "Question 1: Correct. Explanation: 2 + 2 is 4" to the user's device.

[1166] Output: The evaluation results and explanations are displayed on the user's terminal.

[1167] Step 6:

[1168] Users can check the results and continue learning.

[1169] Input: Evaluation results and explanations displayed on the user's device

[1170] Specific operation: The user checks the results on the device and learns the correct answer and explanation for the questions they got wrong. For example, the user sees the result "Question 1: Correct. Explanation: 2 + 2 is 4" and moves on to the next question.

[1171] Output: User progress and understanding improves.

[1172] (Application example 3)

[1173] Next, a description will be given of Application Example 3 of Form Example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1174] Conventional learning systems lacked the ability to automatically mark the answers to questions solved by users and provide feedback. They also lacked the ability to provide the correct answer and explanation if the user's answer was incorrect. This made it difficult for users to accurately grasp their own learning progress, which in turn hindered their motivation to learn.

[1175] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes means for generating study tasks based on user input using a generation AI and presenting the study tasks to the user, means for adjusting the difficulty of the study tasks based on the user's learning progress and questions that were answered incorrectly, means for marking the user's learning results and presenting the results to the user, means for receiving the user's answers in list form, comparing them with correct answers, and generating results, and means for providing the correct answers and explanations in the case of incorrect answers. This allows the user to accurately grasp their own learning progress and deepen their understanding.

[1176] "Generative AI" is an artificial intelligence system that generates learning tasks based on user input and presents them to the user.

[1177] "Learning assignments" are questions or exercises for users to answer, which are generated by generative AI.

[1178] "User's learning progress" refers to the progress a user makes in answering learning tasks.

[1179] "Marking" is the act of determining whether the user's answer is correct and showing the result.

[1180] "List format" is a format in which the user's answers are organized into a series of items and displayed in an easy-to-compare format.

[1181] A "correct answer" is an accurate response to a learning task.

[1182] An "explanation" provides the correct answer for a question that the user got wrong, and explains the reason and background behind the answer.

[1183] "Feedback" refers to evaluations and advice provided based on the user's learning results.

[1184] The system for implementing this invention mainly uses a server, a user terminal, and a generative AI model. Specific embodiments will be described below.

[1185] System Configuration

[1186] 1. Server:

[1187] The server hosts the generative AI model and receives input data from the user.

[1188] The server receives the user's answers in the form of a list, compares them with the correct answers, and generates a result.

[1189] The server provides the correct answer and explanation in case of incorrect answer.

[1190] 2. User Device:

[1191] A user terminal is a device such as a smartphone or tablet that a user uses to answer study questions.

[1192] The user terminal communicates with the server, displays the generated learning tasks, and transmits the user's answers to the server.

[1193] 3. Generative AI Model:

[1194] The generative AI model generates learning tasks based on user input and presents them to the user.

[1195] The generative AI model adjusts the difficulty of the learning tasks based on the user's learning progress and the questions they got wrong.

[1196] The generative AI model marks the user's learning results and presents the results to the user.

[1197] Program processing explanation

[1198] The server receives the answer data sent from the user device and performs the following processing using the generative AI model.

[1199] 1. Data reception and analysis:

[1200] The server receives the answer data sent from the user terminal.

[1201] Analyze the received data and convert it into a list format.

[1202] 2. Compare answers and generate results:

[1203] The server uses a generative AI model to compare the user's answer with the correct answer.

[1204] Based on the comparison result, a determination is made as to whether the answer is correct or incorrect, and a result is generated.

[1205] 3. Providing Feedback:

[1206] The server generates the correct answer and explanation for incorrect answers.

[1207] The generated feedback is sent to the user terminal and presented to the user.

[1208] Hardware and software used

[1209] Hardware: Servers, smartphones, tablets

[1210] Software: Python, OpenAI API

[1211] Specific examples

[1212] When a user answers a learning task using a smartphone, they input prompt sentences like the following into the generative AI model:

[1213] Example prompt sentence:

[1214] User Answer: Answer 2

[1215] Correct Answer: Correct Answer 2

[1216] Explanation: Explanation 2

[1217] This answer is incorrect. The correct answer is "Correct Answer 2". Explanation: Explanation 2

[1218] By inputting this prompt into the generative AI model, appropriate feedback can be provided to the user, allowing them to accurately grasp their own learning progress and deepen their understanding.

[1219] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1220] Step 1:

[1221] The user answers the learning questions using the device. The user inputs the answer through the application on the device and presses the submit button. The input data is the user's answer, which is then sent to the server.

[1222] Step 2:

[1223] The server receives the answer data sent from the user's device. The received data is stored in the server as a list of the user's answers. The input data is the user's answer list, and the output data is a list of answers ready for analysis.

[1224] Step 3:

[1225] The server uses a generative AI model to compare the user's answers with the correct answers. Specifically, it matches the user's answer list with the correct answer list and determines whether each answer is correct or incorrect. The input data is the user's answer list and the correct answer list, and the output data is the determination result.

[1226] Step 4:

[1227] The server generates feedback based on the judgment result. If the answer is correct, it displays "Correct." If the answer is incorrect, it generates "Incorrect," along with the correct answer and explanation. The input data is the judgment result, and the output data is the feedback message.

[1228] Step 5:

[1229] The server sends the generated feedback message to the user terminal, and the user terminal displays the received feedback message to the user. The input data is the feedback message, and the output data is the feedback presented to the user.

[1230] Step 6:

[1231] Users can check the feedback on their devices and learn the correct answers and explanations for questions they got wrong. This allows users to understand their own learning progress and deepen their understanding. The input data is the feedback message, and the output data is the user's improvement in understanding.

[1232] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1233] "Example 1"

[1234] In one embodiment of the present invention, the generation AI uses an emotion engine that recognizes the user's emotions. Specifically, the emotion engine infers the user's emotional state from the user's reactions, facial expressions, tone of voice, and other factors while solving learning tasks. This inference is provided to the generation AI, which then adjusts the generation and presentation of learning tasks and the method of presenting grading results based on this information. For example, when the user is facing a difficult problem, the generation AI can send encouraging messages to maintain the user's motivation to study.

[1235] "Example 2"

[1236] In another form, the emotion engine infers the user's emotional state from their learning progress and reactions to questions they got wrong. Specifically, the emotion engine analyzes how the user reacted to questions they got wrong and how that reaction changed. The results of this analysis are provided to the generation AI, which then adjusts the difficulty of the next learning task based on this. For example, if the user is disappointed about a question they got wrong, the generation AI may lower the difficulty of the next learning task to maintain the user's motivation to learn.

[1237] "Example 3"

[1238] Furthermore, there is also a form in which the generative AI adjusts the feedback and encouragement it provides to the user based on the user's emotional state provided by the emotion engine. Specifically, after the user completes a learning task, the generative AI provides feedback taking into account the user's emotional state provided by the emotion engine. For example, if the user feels joy after completing a learning task, the generative AI provides feedback that reinforces that emotion. This further increases the user's motivation to learn.

[1239] The processing flow of each embodiment will be described below.

[1240] "Example 1"

[1241] Step 1: The user solves the learning task.

[1242] Step 2: The emotion engine estimates the user's emotional state based on their reactions, facial expressions, tone of voice, etc.

[1243] Step 3: Provide the estimation results to the generation AI.

[1244] Step 4: The generative AI adjusts how it generates and presents learning tasks and marks the results based on the provided emotional state.

[1245] Step 5: The generative AI sends encouraging messages to keep the user motivated to learn.

[1246] "Example 2"

[1247] Step 1: The user solves the learning task.

[1248] Step 2: The emotion engine infers the user's emotional state from their reaction to the incorrect question.

[1249] Step 3: Provide the estimation results to the generation AI.

[1250] Step 4: The AI ​​adjusts the difficulty of the next learning task based on the provided emotional state. Step 5: The AI ​​sends encouraging messages to keep the user motivated to learn.

[1251] "Example 3"

[1252] Step 1: The user solves the learning task.

[1253] Step 2: The emotion engine estimates the user's emotional state.

[1254] Step 3: Provide the estimation results to the generation AI.

[1255] Step 4: The generative AI provides feedback taking into account the provided emotional state.

[1256] Step 5: The generative AI provides feedback to improve the user's motivation to learn.

[1257] Example 1

[1258] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1259] Conventional learning support systems have the ability to adjust the difficulty of learning tasks based on the user's learning progress and the problems they have gotten wrong, but they lack learning support that takes into account the user's emotional state. This can lead to users losing motivation when they encounter difficult problems. In addition, they are unable to provide feedback or encouraging messages that reflect the user's emotions, which can lead to a decrease in learning continuity.

[1260] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1261] In this invention, the server includes means for generating study tasks based on user input using a generation AI and presenting the study tasks to the user, means for adjusting the difficulty of the study tasks based on the user's learning progress and questions that they got wrong, means for grading the user's learning results and presenting the results to the user, means for collecting user emotion data and analyzing it with an emotion engine, and means for adjusting the presentation method of the study tasks based on the analysis results of the emotion engine. This enables learning support that takes the user's emotional state into consideration, maintaining the user's motivation to learn and improving continuity in learning.

[1262] "Generative AI" refers to artificial intelligence that generates learning tasks based on user input and presents them to the user.

[1263] "Study assignments" are specific problems or exercises that users must complete.

[1264] "User's learning progress" is an indicator that shows how far the user has progressed in their learning.

[1265] "Difficulty of a learning task" is an index that indicates how difficult a learning task is.

[1266] "Marking" refers to the act of determining whether a learning task completed by a user is correct or incorrect and showing the result.

[1267] "Emotional data" refers to data that indicates the user's emotional state, such as the user's facial expression or tone of voice.

[1268] An "emotion engine" is a system that analyzes collected emotional data and estimates the user's emotional state.

[1269] "Feedback" refers to evaluations and comments provided on a user's learning results.

[1270] "Encouraging messages" are words of encouragement or encouragement sent to users to motivate them to learn.

[1271] The present invention relates to a learning support system using a generative AI model. Specific embodiments of this system will be described below.

[1272] First, the server generates a program for the learning support system using the generative AI model. This program has the function of generating learning tasks based on user input and presenting them to the user.

[1273] The user inputs "What is today's study topic?" through the device interface, which can be a web browser or a mobile application. The server receives the request from the user and analyzes the request.

[1274] Next, the server sends the user's request as a prompt to the generative AI model. The prompt is in the form of "Please tell me today's study task." The generative AI model analyzes the prompt and generates specific study tasks taking into account the user's learning history and progress. For example, it generates a task such as "Today, let's do 10 addition and subtraction problems of fractions in arithmetic."

[1275] The server presents the learning tasks received from the generative AI model to the user through the device's interface. As the user solves the learning tasks, the device uses a camera and microphone to collect emotional data such as the user's facial expressions and tone of voice.

[1276] The emotion engine analyzes the collected emotion data and estimates the user's emotional state. For example, if the user is facing a difficult problem, the emotion engine estimates that the user is confused. The server sends this estimation result to the generative AI model.

[1277] The generative AI model adjusts the presentation of learning tasks and additional messages based on the estimated emotional state. For example, if the user is confused, it generates an encouraging message. The server then displays this message to the user to keep them motivated to learn.

[1278] For example, if a user types "What is today's study topic?" into their device, the server receives this request and sends a prompt to the generative AI model saying, "Please tell me today's study topic." The generative AI model generates a study topic such as "Today, let's do 10 math problems on adding and subtracting fractions," and returns it to the server. The server then presents this study topic to the user.

[1279] While the user is solving a learning task, the device uses a camera and microphone to collect the user's facial expressions and tone of voice. The emotion engine analyzes this data and infers that the user is confused. The server sends this inference to the generative AI model, which then generates an encouraging message such as "Good luck! You're almost there!" The server displays this message to the user to maintain their motivation to learn.

[1280] Examples of prompts include:

[1281] "What's today's lesson?"

[1282] "Tell me the next question."

[1283] Send a message of encouragement

[1284] In this way, a learning support system that combines a generative AI model and an emotion engine supports users' learning.

[1285] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1286] Step 1:

[1287] A user requests a learning assignment

[1288] The user inputs "What is today's study topic?" through the terminal interface. This input is sent to the server. The input data is the user's request.

[1289] Step 2:

[1290] The server receives the user request

[1291] The server receives a request from a user and analyzes the request content. The input data is the user's request content, and the output data is the analyzed request content.

[1292] Step 3:

[1293] The server sends a prompt to the generative AI model

[1294] The server generates a prompt based on the analyzed request content and sends it to the generative AI model. The input data is the analyzed request content, and the output data is the generated prompt. The specific prompt is "Please tell me what today's learning topic is."

[1295] Step 4:

[1296] Generative AI model generates learning tasks

[1297] The generative AI model analyzes the prompt and generates specific learning tasks by taking into account the user's learning history and progress. The input data is the prompt and the user's learning history and progress data, and the output data is the generated learning task. For example, it generates a task such as, "Today, let's do 10 addition and subtraction problems of fractions in arithmetic."

[1298] Step 5:

[1299] The server presents the generated learning tasks to the user.

[1300] The server presents the learning tasks received from the generative AI model to the user. The input data is the generated learning tasks, and the output data is the learning tasks presented to the user. This presentation is done through the terminal interface.

[1301] Step 6:

[1302] The user solves the learning task

[1303] The user begins to solve the presented learning problem. The answer is entered into the terminal. The input data is the user's answer, and the output data is the answer entered into the terminal.

[1304] Step 7:

[1305] The device collects user emotional data

[1306] The device uses a camera and a microphone to collect emotional data such as the user's facial expression and tone of voice. The input data is the user's facial expression and tone of voice, and the output data is the collected emotional data.

[1307] Step 8:

[1308] The emotion engine analyzes the emotion data

[1309] The emotion engine analyzes the collected emotion data and estimates the user's emotional state. The input data is the collected emotion data, and the output data is the estimated emotional state. For example, if the user is facing a difficult problem, the emotion engine estimates that the user is confused.

[1310] Step 9:

[1311] The server sends the results of the emotion engine to the generative AI model.

[1312] The server sends the estimated emotional state received from the emotion engine to the generative AI model. The input data is the estimated emotional state, and the output data is the emotional state data sent to the generative AI model.

[1313] Step 10:

[1314] Generative AI models adjust how learning tasks are presented

[1315] The generative AI model adjusts the presentation of learning tasks and additional messages based on the estimated emotional state. The input data is the estimated emotional state, and the output data is the adjusted learning tasks and messages. For example, if the user is confused, it generates an encouraging message.

[1316] Step 11:

[1317] The server presents tailored learning tasks and messages to the user.

[1318] The server presents the adjusted learning tasks and messages received from the generative AI model to the user. The input data are the adjusted learning tasks and messages, and the output data are the learning tasks and messages presented to the user. For example, it displays a message such as "Good luck! You're almost there!"

[1319] (Application example 1)

[1320] Next, a description will be given of Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1321] Conventional learning support systems can generate and present learning tasks based on the user's learning progress and incorrect questions, but they have the problem of being unable to provide feedback or adjust learning tasks that take into account the user's emotional state. Furthermore, when supporting factory workers, there are concerns that a decline in work efficiency and motivation can occur because appropriate feedback and support that takes into account the worker's emotional state is not provided. To solve these problems, a system that can recognize the user's emotional state and adjust learning tasks and work content based on that is needed.

[1322] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1323] In this invention, the server includes means for generating study tasks based on user input using a generation AI and presenting the study tasks to the user, means for adjusting the difficulty of the study tasks based on the user's learning progress and questions that they got wrong, means for grading the user's study results and presenting the results to the user, means for estimating the user's emotional state using an emotion engine and adjusting the method for generating and presenting study tasks based on the emotional state, and means for providing appropriate feedback according to the user's emotional state. This enables learning support and task support that takes the user's emotional state into consideration.

[1324] "Generative AI" is artificial intelligence that generates learning tasks and work content based on input from the user and presents them to the user.

[1325] "Study assignments" are specific learning content or problems that users must address.

[1326] The "emotion engine" is a technology for estimating a user's emotional state from facial expressions, tone of voice, etc.

[1327] "Feedback" refers to evaluations and advice provided to users regarding their learning and work results.

[1328] "User emotional state" refers to the emotional state exhibited by the user, including, for example, tiredness, confusion, joy, etc.

[1329] "Study progress" is an indicator that shows how far the user has progressed in the learning tasks.

[1330] "Marking" is the act of determining whether a user's learning results are correct or incorrect and showing the result.

[1331] A "means" is a method or device for achieving a specific purpose.

[1332] "Presenting" is the act of displaying or informing a user of information.

[1333] "Adjusting difficulty" means changing the difficulty of the learning tasks based on the user's learning progress and the questions they got wrong.

[1334] As an embodiment of the present invention, we will explain an example of a "work support system" installed in a factory robot. This system is designed to improve the work efficiency of factory workers by using generative AI and an emotion engine.

[1335] System configuration

[1336] 1. Hardware:

[1337] Factory robots

[1338] camera

[1339] microphone

[1340] 2. Software:

[1341] Python

[1342] EmotionEngine Library

[1343] TaskGenerator Library

[1344] Program processing explanation

[1345] Generate work

[1346] The server uses the TaskGenerator library to generate work content based on input from the user (factory worker). For example, when a worker asks, "What will you be doing today?", the generation AI generates and presents specific work content such as, "Today, assemble 50 units of part A."

[1347] Emotion Recognition and Feedback

[1348] The server uses the EmotionEngine library to analyze data from the camera and microphone and estimate the worker's emotional state. For example, if the server estimates that the worker is tired, the generative AI will send a message such as "Take a short break."

[1349] Specific examples

[1350] 1. Work generation:

[1351] When a worker asks, "What will you be doing today?", the generative AI generates and presents specific work content, such as "Today, you will assemble 50 units of part A."

[1352] 2. Emotion Recognition and Feedback:

[1353] If the generator estimates that the worker is tired, it will send a message such as "Take a short break."

[1354] Prompt Sentence Examples

[1355] "What are you working on today?"

[1356] User input data

[1357] In this way, it becomes possible to provide learning and work support that takes into account the user's emotional state.

[1358] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1359] Step 1:

[1360] The user types into the terminal, "What will I be working on today?"

[1361] Input: User asks "What are you working on today?"

[1362] Output: User question data

[1363] Step 2:

[1364] The server receives the user's question data and generates task content using the TaskGenerator library.

[1365] Input: User question data

[1366] Data processing: Generate work content using the TaskGenerator library

[1367] Output: Generated work content (e.g., "Today, let's assemble 50 units of part A")

[1368] Step 3:

[1369] The server transmits the generated work content to the terminal and presents it to the user.

[1370] Input: Generated work

[1371] Output: The work presented to the user

[1372] Step 4:

[1373] The user begins working, and the camera and microphone capture the user's facial expressions and tone of voice in real time.

[1374] Input: User's facial expression data, voice tone data

[1375] Output: Captured emotion data

[1376] Step 5:

[1377] The server uses the EmotionEngine library to analyze the captured emotion data and estimate the user's emotional state.

[1378] Input: Captured emotion data

[1379] Data Computation: Estimating Emotional States Using the EmotionEngine Library

[1380] Output: Estimated emotional state (e.g., "Tired")

[1381] Step 6:

[1382] The server uses a generation AI to generate an appropriate feedback message based on the estimated emotional state.

[1383] Input: Estimated emotional state

[1384] Data processing: Generative AI is used to generate feedback messages

[1385] Output: The generated feedback message (e.g., "Let's take a short break")

[1386] Step 7:

[1387] The server transmits the generated feedback message to the terminal and presents it to the user.

[1388] Input: The generated feedback message

[1389] Output: The feedback message that is presented to the user.

[1390] Example 2

[1391] Next, a description will be given of Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1392] Conventional learning systems lack the ability to adjust the difficulty of learning tasks based on the user's learning progress and incorrect questions, making it difficult to maintain the user's motivation to learn. Furthermore, they do not provide learning tasks that take the user's emotional state into consideration, which results in the user not being able to maximize their learning effectiveness. Furthermore, they lack a mechanism for effectively collecting and analyzing the user's learning data and reflecting it in the next learning task.

[1393] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1394] In this invention, the server includes means for generating study tasks based on user input using a generative AI and presenting the study tasks to the user, means for adjusting the difficulty of the study tasks based on the user's learning progress and the questions answered incorrectly, means for grading the user's study results and presenting the results to the user, means for collecting the user's study data and saving it in a database, means for analyzing the collected study data and extracting the user's correct answer rate and trends in the questions answered incorrectly, means for analyzing the user's reactions during study and estimating their emotional state, means for generating prompt sentences to be input into the generative AI model based on the analysis results of the study data and the emotional state, and means for providing the generated study tasks to the user and collecting the user's answers. This makes it possible to provide optimal study tasks that take into account the user's learning progress and emotional state.

[1395] "Generative AI" refers to artificial intelligence that generates learning tasks based on user input and presents them to the user.

[1396] A "user" is an individual who uses the learning system to complete learning tasks.

[1397] "Learning assignments" are questions and exercises created by the generative AI for users to answer.

[1398] "Study progress" is a metric that shows how much progress a user has made through a learning task.

[1399] A "mistake" is a question that the user did not answer correctly in the learning task.

[1400] "Difficulty adjustment" refers to changing the difficulty of the next learning task based on the user's learning progress and the questions they got wrong.

[1401] "Marking" refers to determining whether a user's learning results are correct or incorrect and presenting the results to the user.

[1402] "Learning data" refers to data that includes information such as the results and reactions of users when they answer learning tasks.

[1403] A "database" is a system for storing collected learning data.

[1404] "Analysis" refers to analyzing the collected learning data and extracting the user's correct answer rate and trends in questions they got wrong.

[1405] "Emotional state" refers to the emotions a user exhibits while learning, including, for example, happiness, disappointment, excitement, etc.

[1406] A "prompt sentence" is an instruction sentence to be input into a generative AI model, and is generated based on the results of an analysis of the learning data and emotional state.

[1407] "Answer result" refers to the result of the answer given by the user to the learning task.

[1408] The present invention is a system that provides optimal study tasks in consideration of a user's learning progress and emotional state. Specific embodiments of this system will be described below.

[1409] The server collects the results of the questions that users answer on the learning platform. Specifically, it records data such as the questions that users answered correctly, the questions that they answered incorrectly, and the time it took to answer, and stores this data in a database. This data collection is done using programming languages ​​such as Python.

[1410] Next, the server analyzes the collected training data. Using the Python pandas library, it analyzes the user's accuracy rate and trends in questions they get wrong. For example, if a user makes many mistakes when adding fractions, it can extract that information.

[1411] Furthermore, the server uses an emotion engine to analyze the user's reactions while studying. A general emotion recognition service is used as the emotion recognition software. While the user is answering the study questions, the device's camera and microphone record the user's facial expressions and voice, and this data is sent to the server in real time. The emotion engine estimates the user's emotional state, such as whether they are disappointed about a question they got wrong.

[1412] The server generates prompt sentences to be input to the generative AI model based on the learning data and the analysis results of the emotion engine. For example, it generates prompt sentences like the following:

[1413] According to your learning data, you're making a lot of mistakes with fraction addition. Please focus on fraction addition in your next learning exercise. Also, please lower the difficulty a bit, as users are frustrated with the problems they get wrong.

[1414] The server inputs the generated prompt sentence into a generative AI model to generate the next learning task. The generative AI model uses common generative AI technology. The generative AI model creates learning tasks appropriate for the user based on the prompt sentence. For example, it generates a learning task that includes many problems related to adding fractions.

[1415] Finally, the device provides the generated learning tasks to the user. The user answers the learning tasks, and the device sends the results back to the server. The server uses the collected feedback to help generate the next learning task. For example, it analyzes the trends in questions the user recently got wrong and reflects this in the next learning task.

[1416] In this way, it is possible to provide optimal learning tasks that take into account the user's learning progress and emotional state.

[1417] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1418] Step 1:

[1419] The server collects the results of questions answered by users on the learning platform. Specifically, it records data such as the questions answered correctly and incorrectly by the user, and the time it took to answer, and stores this data in a database. The input is the user's answer, and the output is the learning data stored in the database.

[1420] Step 2:

[1421] The server analyzes the collected training data. It uses the Python pandas library to analyze the user's correct answer rate and the tendency of questions that were answered incorrectly. The input is the training data stored in the database, and the output is the analysis results of the user's correct answer rate and the tendency of questions that were answered incorrectly.

[1422] Step 3:

[1423] The server analyzes the user's reactions during learning using an emotion engine. Using emotion recognition software, it infers the user's emotional state from their facial expressions and voice. The input is the user's facial expressions and voice data sent from the device, and the output is the inferred emotional state.

[1424] Step 4:

[1425] The server generates prompt sentences to be input to the generative AI model based on the learning data and the analysis results of the emotion engine. Specifically, it generates prompt sentences like the following:

[1426] Based on your learning data, you've made a lot of mistakes with fraction addition. Please focus on fraction addition in your next learning exercise. Also, please lower the difficulty a bit, as your users are disappointed with the problems they get wrong.

[1427] The input is the training data and the analysis result of the emotional state, and the output is the generated prompt sentence.

[1428] Step 5:

[1429] The server inputs the generated prompt sentence into the generative AI model to generate the next learning task. The generative AI model creates a learning task appropriate for the user based on the prompt sentence. The input is the generated prompt sentence, and the output is the generated learning task.

[1430] Step 6:

[1431] The terminal provides the generated learning tasks to the user. The user answers the learning tasks, and the terminal sends the results back to the server. The input is the generated learning tasks, and the output is the user's answer.

[1432] Step 7:

[1433] The server uses the collected feedback to help generate the next study assignment. For example, it analyzes the trends of questions that the user has recently gotten wrong and reflects this in the next study assignment. The input is the user's answer, and the output is the analysis results that will help generate the next study assignment.

[1434] (Application example 2)

[1435] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1436] Conventional learning support systems have a function to adjust the difficulty of learning tasks based on the user's learning progress and the questions they got wrong, but they do not take into account the user's emotional state, which can lead to users feeling stressed or losing motivation to learn. Furthermore, in work environments such as factories, there is a lack of functionality to monitor the progress of workers and the frequency of mistakes in real time and adjust the work content and difficulty, which can lead to problems with reduced work efficiency.

[1437] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1438] In this invention, the server includes means for generating learning tasks based on user input using a generation AI and presenting the learning tasks to the user, means for adjusting the difficulty of the learning tasks based on the user's learning progress and questions answered incorrectly, means for marking the user's learning results and presenting the results to the user, means for estimating the user's emotional state using an emotion engine and adjusting the difficulty of the learning tasks based on the results, and means for monitoring the worker's progress and frequency of mistakes in real time and adjusting the task content and difficulty level. This makes it possible to provide learning tasks that take the user's emotional state into consideration and to optimize the task content in work environments such as factories.

[1439] "Generative AI" is artificial intelligence that generates learning tasks based on user input and presents them to the user.

[1440] "Study assignments" are specific problems or exercises that users must complete.

[1441] "User's learning progress" is an indicator that shows how far the user has progressed in the learning tasks.

[1442] "Wrong questions" are learning tasks that the user was unable to answer correctly.

[1443] "Difficulty of learning task" is an index that indicates how difficult the learning task is.

[1444] "Marking" is the act of determining whether a user's learning results are correct or incorrect and providing feedback.

[1445] An "emotion engine" is a technology for estimating a user's emotional state.

[1446] "Emotional state" refers to the emotions a user experiences while learning or working.

[1447] "Worker progress" is an index that indicates how far a worker has progressed in the work.

[1448] "Frequency of mistakes" refers to the number and frequency of mistakes that workers make while working.

[1449] "Real-time monitoring" means monitoring the ongoing situation immediately.

[1450] "Work content" refers to the specific work that a worker will perform.

[1451] "Work environment" refers to the location and conditions under which work is performed.

[1452] The system for implementing this invention includes a generative AI, an emotion engine, and a real-time monitoring function. Specific embodiments of the system are described below.

[1453] System configuration

[1454] 1. Generation AI:

[1455] The server generates learning tasks based on user input and presents them to the user. The AI ​​analyzes the user's learning progress and the questions they got wrong, and adjusts the difficulty of the next learning task.

[1456] 2. Emotion Engine:

[1457] The server uses an emotion engine to estimate the user's emotional state. The emotion engine receives the user's reaction data (e.g., sigh, frown, etc.) as input and outputs the user's emotional state.

[1458] 3. Real-time monitoring:

[1459] The server monitors the worker's progress and frequency of mistakes in real time and adjusts the task content and difficulty accordingly, providing optimal tasks according to the worker's skill level and emotional state.

[1460] Hardware and software used

[1461] Hardware:

[1462] Server: Processes data and runs generative AI.

[1463] User device: Presents learning tasks and collects user response data.

[1464] Sensors: Used to monitor workers' progress and frequency of mistakes in real time.

[1465] software:

[1466] Generative AI model: Runs algorithms to generate learning tasks and adjust their difficulty.

[1467] Emotion engine: Runs algorithms to estimate the user's emotional state.

[1468] Real-time monitoring system: Monitors workers' progress and frequency of mistakes in real time and sends the data to a server.

[1469] Specific examples

[1470] For example, if a user is studying mathematics, the generative AI will analyze the questions the user got wrong and focus on questions from the same subject for the next study assignment. The emotion engine will also analyze the user's reaction data and, if the user is feeling stressed, will lower the difficulty of the next study assignment.

[1471] In the factory work environment, a real-time monitoring system monitors the progress of workers and the frequency of mistakes, and an emotion engine estimates the worker's emotional state. If a worker is feeling stressed, the system adjusts the task, such as by lowering the difficulty of the work.

[1472] Prompt Sentence Examples

[1473] Below are some example prompts to input to the generative AI model:

[1474] Create a program that adjusts the difficulty of the next task based on the worker's progress and emotional state. This system takes the worker's progress (progress rate and frequency of mistakes) and emotional state (reactions such as sighing or frowning) as input and adjusts the difficulty of the next task. Using an emotion engine and an AI task generator, it estimates the worker's emotional state and generates the next task.

[1475] In this way, it is possible to provide learning tasks that take into account the user's learning progress and emotional state, and to optimize work content in work environments such as factories.

[1476] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1477] Step 1:

[1478] The server receives input from the user and sends it to the generative AI model. The input data includes the results of the learning tasks the user answered and reaction data. The generative AI model generates the next learning task based on this input data. Specifically, it analyzes the questions the user got wrong and their learning progress, and selects learning tasks of an appropriate level of difficulty.

[1479] Step 2:

[1480] The server sends the generated learning tasks to the user's device. The user's device presents the received learning tasks to the user. The user then works on the presented learning tasks. In this step, how the user responds to the learning tasks is recorded in real time.

[1481] Step 3:

[1482] The user device sends the user's learning results and reaction data to the server. The server uses an emotion engine to estimate the user's emotional state. The emotion engine receives the user's reaction data (e.g., sighing, frowning, etc.) as input and outputs the user's emotional state. Emotional states include stress, disappointment, and satisfaction.

[1483] Step 4:

[1484] The server adjusts the difficulty of the next learning task based on the output of the emotion engine. Specifically, it lowers the difficulty if the user is feeling stressed and raises it if the user is satisfied. This adjustment result is fed back to the generative AI model and reflected in the generation of the next learning task.

[1485] Step 5:

[1486] The server uses a real-time monitoring system to monitor the progress of workers and the frequency of mistakes in the factory work environment. It receives data from sensors and analyzes the progress rate and frequency of mistakes in real time, providing optimal work content according to the worker's skill level and emotional state.

[1487] Step 6:

[1488] The server adjusts the task content and difficulty level based on data obtained from the real-time monitoring system. It uses an emotion engine to estimate the worker's emotional state and lowers the difficulty of the task if the worker is feeling stressed. The adjusted task content is sent to the worker's device and presented to the worker.

[1489] Step 7:

[1490] The server integrates all data and comprehensively evaluates the learning progress and work efficiency of users and workers. This provides data for optimizing long-term learning and work plans. The evaluation results are also used to improve the generative AI model and emotion engine algorithms.

[1491] In this way, it is possible to provide learning tasks that take into account the user's learning progress and emotional state, and to optimize work content in work environments such as factories.

[1492] Example 3

[1493] Next, a description will be given of Example 3 of Form Example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1494] Conventional learning support systems provide uniform evaluations and feedback on users' answers, making it difficult to respond to individual users' emotional states. It was also difficult to accurately grasp a user's learning progress and level of understanding and provide appropriate feedback.

[1495] The identification process by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means. In this invention, the server includes: means for generating study tasks based on user input using a generation AI and presenting the study tasks to the user; means for adjusting the difficulty of the study tasks based on the user's learning progress and questions that were answered incorrectly; means for grading the user's learning results and presenting the results to the user; means for adjusting feedback based on the user's emotional state provided by the emotion engine; means for evaluating the user's answers using a generation AI model and determining whether they are correct; and means for generating feedback that takes the user's emotional state into consideration using the generation AI model. This makes it possible to accurately grasp the user's learning progress and level of understanding and provide appropriate feedback according to the individual emotional state.

[1496] "Generative AI" is artificial intelligence that generates learning tasks based on user input and evaluates the user's answers.

[1497] "Study assignments" are questions or exercises for users to answer.

[1498] "User learning progress" is a metric that indicates how much progress a user has made through a learning task.

[1499] "Difficulty of learning task" is a level that indicates how difficult the learning task is for the user.

[1500] "Marking" is an evaluation action that determines whether the user's answer is correct or incorrect.

[1501] An "emotion engine" is a system that detects a user's emotional state and provides that information.

[1502] "Feedback" is the evaluation and advice provided to a user on their answers and learning progress.

[1503] A "generative AI model" is an artificial intelligence model that generates evaluations and feedback based on the user's answers and emotional state.

[1504] "Determining correctness" means determining whether the user's answer is correct or incorrect.

[1505] "Emotional state" refers to the psychological state, such as joy or depression, that a user feels while learning.

[1506] This invention is a learning support system that tracks a user's learning progress and deepens their understanding. The system uses a generative AI model and an emotion engine to evaluate the user's answers and provide feedback.

[1507] The server receives the answers to the problems solved by the user and inputs them into the generative AI model. For example, a general artificial intelligence model is used as the generative AI model. The generative AI model evaluates the user's answer and determines whether it is correct or incorrect. The server then presents the evaluation results from the generative AI model to the user. For problems that were answered incorrectly, an explanation is provided along with the correct answer.

[1508] Furthermore, the server uses an emotion engine to obtain the user's current emotional state. The emotion engine can use, for example, a general emotion recognition API. The generative AI model takes this emotional state into account and adjusts the feedback and encouragement provided to the user. For example, if the user feels joy after completing a learning task, the generative AI model provides feedback that reinforces that emotion.

[1509] As a concrete example, let's say a user solves a math problem. If the user answers "2 + 2 = 5," the server inputs this answer into the generative AI model. The generative AI model determines that "2 + 2 = 4" is the correct answer and tells the user, "Your answer is wrong. The correct answer is '2 + 2 = 4'." If the emotion engine detects that the user is feeling down, the generative AI model provides encouraging feedback such as, "It's okay to make a mistake. You'll definitely get it right next time!"

[1510] Example prompt sentence:

[1511] User Answer: 2 + 2 = 5

[1512] Correct answer: 2 + 2 = 4

[1513] User's emotional state: Depressed

[1514] Prompt to generative AI model: The user's answer is "2 + 2 = 5". The correct answer is "2 + 2 = 4". The user is currently feeling depressed. Please provide encouraging feedback to the user.

[1515] In this way, the server, the terminal, and the user work together to grasp the user's learning progress and provide support for deepening the user's understanding. The flow of the identification process in the third embodiment will be described with reference to FIG.

[1516] Step 1:

[1517] The user solves the problem

[1518] The user solves a learning task (e.g., a math problem) presented on the device. For example, the answer to the question "2 + 2 = ?" is "5." The input is the user's answer "5," and the output is the answer data entered into the device.

[1519] Step 2:

[1520] The device sends the user's answer to the server

[1521] The terminal sends the answers entered by the user to the server, where the answer data is properly formatted and sent in a form that the server can process. The input is the user's answer data, and the output is the answer data sent to the server.

[1522] Step 3:

[1523] The server inputs the answer into the generative AI model

[1524] The server inputs the received user answer into the generative AI model. For example, it generates and sends a prompt to the generative AI model saying, "The user's answer is '2 + 2 = 5'. Please tell me the correct answer." The input is the user's answer data, and the output is the prompt sent to the generative AI model.

[1525] Step 4:

[1526] The generative AI model evaluates the answer and returns the result to the server

[1527] The generative AI model evaluates the user's answer and returns the correct answer and the evaluation result to the server. For example, it may return a result such as "The correct answer is '2 + 2 = 4'. The user's answer is incorrect." The input is the prompt sent to the generative AI model, and the output is the evaluation result returned to the server.

[1528] Step 5:

[1529] The server presents the evaluation results to the user.

[1530] The server presents the evaluation results received from the generative AI model to the user. For example, it sends a message to the terminal saying, "Your answer is incorrect. The correct answer is '2 + 2 = 4'" and displays it to the user. The input is the evaluation result from the generative AI model, and the output is the evaluation result presented to the user.

[1531] Step 6:

[1532] The server obtains the user's emotional state from the emotion engine.

[1533] The server uses the emotion engine to obtain the user's current emotional state. The emotion engine detects that the user is depressed and returns the result "The user is currently depressed" to the server. The input is the user's current state, and the output is the emotional state data from the emotion engine.

[1534] Step 7:

[1535] The server allows the generative AI model to generate feedback that takes into account the emotional state.

[1536] The server inputs the user's emotional state obtained from the emotion engine into the generative AI model, which then generates feedback that takes the emotional state into account. For example, a prompt such as "The user is currently feeling depressed. Please provide encouraging feedback" is sent to the generative AI model. The input is the emotional state data from the emotion engine, and the output is the prompt sent to the generative AI model.

[1537] Step 8:

[1538] The server provides feedback to the user

[1539] The generative AI model generates feedback such as "It's okay if you make a mistake. You'll definitely get it right next time!" and returns it to the server. The server sends this feedback to the device, which then displays it to the user. The input is the feedback from the generative AI model, and the output is the feedback presented to the user.

[1540] (Application example 3)

[1541] Next, a description will be given of Application Example 3 of Form Example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1542] Conventional learning systems have difficulty accurately grasping a user's learning progress and level of understanding and providing appropriate feedback. They also lack the ability to provide feedback and encouragement that takes into account the user's emotional state, which can lead to insufficient motivation to learn. Furthermore, learning management within the virtual store is inadequate, leaving a lack of means to improve the user's learning experience.

[1543] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes: means for generating study tasks based on user input using a generation AI and presenting the study tasks to the user; means for adjusting the difficulty of the study tasks based on the user's learning progress and questions that the user got wrong; means for grading the user's learning results and presenting the results to the user; means for analyzing the user's emotional state using an emotion engine and providing feedback and encouragement based on the emotional state; and means for managing the user's learning progress using a smart device when studying in the virtual store. This makes it possible to accurately grasp the user's learning progress and level of understanding and provide appropriate feedback and encouragement, thereby increasing motivation to study and improving the learning experience in the virtual store.

[1544] "Generative AI" is an artificial intelligence system that generates learning tasks based on user input and presents them to the user.

[1545] "Learning tasks" are problems or tasks that users must solve, and are generated by generative AI.

[1546] "User's learning progress" refers to the progress a user makes in the course of learning.

[1547] A "wrong question" is a question that the user answered incorrectly.

[1548] "Difficulty of learning task" indicates the degree of difficulty of the learning task, and is adjusted based on the user's learning progress and the questions they got wrong.

[1549] "Marking" refers to determining whether a user's learning results are correct or incorrect and evaluating them.

[1550] An "emotion engine" is a system for analyzing a user's emotional state.

[1551] "Emotional state" refers to the emotions such as joy, sadness, excitement, etc. that a user feels while learning.

[1552] "Feedback" refers to evaluations and comments provided on a user's learning results.

[1553] "Encouragement" refers to words of encouragement or messages provided to motivate users to learn more.

[1554] A "virtual store" is a virtual store that exists on the Internet and provides an environment for users to study.

[1555] "Smart devices" refer to electronic devices with advanced functions such as smartphones and head-mounted displays.

[1556] "Managing learning progress" means tracking a user's learning progress and making appropriate adjustments.

[1557] The system for implementing this invention includes a generative AI, an emotion engine, a smart device, and a virtual store. The specific operation of each component and their interaction will be described below.

[1558] 1. Generating learning tasks using generative AI

[1559] The server uses a generation AI to generate learning tasks based on user input. When the user enters an answer, the generation AI analyzes the answer and generates the next learning task to be presented. The generation AI adjusts the difficulty of the learning task, taking into account the user's learning progress and the questions they got wrong.

[1560] 2. Marking and presenting learning results

[1561] When the user inputs the answer to the problem they solved into the AI ​​generator, the AI ​​marks the answers based on the answers. The marks are then presented to the user, allowing them to understand their own learning progress.

[1562] 3. Emotion analysis and feedback using an emotion engine

[1563] The emotion engine analyzes the user's emotional state. After the user completes a learning task, the emotion engine detects the user's emotional state from their facial expressions and voice. Based on the detected emotional state, the generative AI provides appropriate feedback and encouragement. For example, if the user is feeling happy, the generative AI will provide feedback such as, "Great! Keep it up!"

[1564] 4. Learning management within the virtual store

[1565] Users study in a virtual store. Smart devices such as smartphones and head-mounted displays are used to manage their learning progress. The virtual store provides an environment for users to solve learning tasks, improving their learning experience.

[1566] Specific examples

[1567] The user inputs answers such as "1. 2+2=4," "2. 3+5=8," and "3. 7-2=5." If the emotion engine interprets the user's facial expression as "happy," the generative AI provides feedback such as "Great! Keep it up!"

[1568] Prompt Sentence Examples

[1569] Please circle your answer below:

[1570] 1. 2+2=4

[1571] 2. 3+5=8

[1572] 3. 7-2=5

[1573] In this way, users can track their learning progress and deepen their understanding, and the feedback provided by the emotion engine enhances their motivation to learn and improves the learning experience within the virtual store.

[1574] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1575] Step 1:

[1576] The user solves the learning tasks. The user uses a smart device to input answers to the learning tasks presented in the virtual store. The input answers are sent to the server.

[1577] Step 2:

[1578] The server uses a generation AI to analyze the user's answers. Based on the input answers, the generation AI marks them. Specifically, the generation AI generates a prompt and evaluates the user's answers. For example, it generates a prompt such as "Please mark the following answers: 1. 2+2=4 2. 3+5=8 3. 7-2=5." The analysis results are judged as correct or incorrect, and the result is saved on the server.

[1579] Step 3:

[1580] The server uses an emotion engine to analyze the user's emotional state. The user's facial expressions and voice data are sent from the smart device to the server, and the emotion engine analyzes them. The analysis result is the user's emotional state (e.g., "happy" or "sad").

[1581] Step 4:

[1582] The server uses a generation AI to provide feedback to the user. Based on the emotional state obtained from the emotion engine, the generation AI generates appropriate feedback and encouraging messages. For example, if the user is analyzed as "happy," feedback such as "Great! Keep it up!" is generated. The generated feedback is stored on the server.

[1583] Step 5:

[1584] The server provides feedback to the user. The generated feedback and marking results are sent to the smart device and displayed to the user. The user can grasp their own learning progress and level of understanding and receive appropriate feedback.

[1585] Step 6:

[1586] The server adjusts the difficulty of the next learning task based on the user's learning progress and the questions they got wrong. The generation AI generates the next learning task based on the user's learning history and adjusts its difficulty. The generated learning task is saved on the server and presented to the user the next time they study.

[1587] In this way, users can effectively learn within the virtual store, and feedback from the emotion engine increases their motivation to learn.

[1588] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1589] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1590] Another example of generative AI is Gemini (internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[1591] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1592] [Third embodiment]

[1593] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1594] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[1595] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1596] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[1597] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1598] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1599] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1600] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1601] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1602] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1603] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1604] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1605] "Example 1"

[1606] One embodiment of the present invention is a learning support system that uses a generative AI. This system generates learning tasks based on user input and presents the task to the user. Specifically, when the user asks, "What is today's learning task?", the generative AI generates a specific learning task such as, "Today, let's do 10 math problems on adding and subtracting fractions," and presents it to the user.

[1607] "Example 2"

[1608] Furthermore, this system adjusts the difficulty of the learning tasks based on the user's learning progress and the problems they got wrong. For example, if the user gets many fraction addition problems wrong, the generation AI will optimize the learning content by focusing on fraction addition as the next learning task. "Example 3"

[1609] The system also marks the user's learning results and presents the results to the user. Specifically, the answers to the problems the user solves are input into the generation AI, which then marks the answers based on those answers. The results are then presented to the user, and for any problems the user got wrong, an explanation is provided along with the correct answer. This allows the user to understand their own learning progress and deepen their understanding.

[1610] The processing flow of each embodiment will be described below.

[1611] "Example 1"

[1612] Step 1: The user asks the system, "What is today's study topic?"

[1613] Step 2: The generative AI receives input from the user and generates learning tasks, such as "Today, let's do 10 addition and subtraction problems with fractions in arithmetic."

[1614] Step 3: The learning task generated by the generative AI is presented to the user.

[1615] "Example 2"

[1616] Step 1: After the user completes the learning task, they enter the results into the system.

[1617] Step 2: The generative AI receives input from the user and analyzes the questions they got wrong.

[1618] Step 3: The AI ​​adjusts the difficulty of the next lesson. For example, if the user makes many mistakes on fraction addition, the AI ​​will optimize the learning content by focusing on fraction addition in the next lesson.

[1619] "Example 3"

[1620] Step 1: The user enters the answer to the problem into the system.

[1621] Step 2: The generative AI receives input from the user and marks the answers.

[1622] Step 3: The AI ​​generator presents the results of the marking to the user. For incorrect questions, it provides an explanation along with the correct answer. This allows the user to understand their own learning progress and deepen their understanding.

[1623] Example 1

[1624] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1625] Conventional learning support systems have difficulty providing appropriate learning tasks that correspond to the user's learning progress and individual needs. Furthermore, they lack effective means of providing feedback on the user's learning results or motivating them to learn. This makes it difficult for users to progress efficiently in their studies.

[1626] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1627] In this invention, the server includes means for receiving input from a user, means for analyzing the received input data and sending it as a prompt to a generative AI model, means for generating study tasks based on the user's input using the generative AI model and presenting the study tasks to the user, means for adjusting the difficulty of the study tasks based on the user's learning progress and questions that the user got wrong, means for grading the user's study results and presenting the results to the user, means for sending the generated study tasks to the user, and means for displaying the study tasks to the user. This makes it possible to provide appropriate study tasks according to the individual needs of the user and to increase motivation to learn through feedback on the study results.

[1628] "Means for receiving input from the user" refers to the function of providing an interface for the user to input information to the system in the form of text, voice, etc., and receiving that input.

[1629] "Means for analyzing received input data and sending it to the generative AI model as a prompt" refers to the function of analyzing input data received from a user and sending the analysis results to the generative AI model as a prompt in a format that the generative AI model can understand.

[1630] "Means for generating learning tasks based on user input using a generative AI model and presenting the learning tasks to the user" refers to the function of a generative AI model generating appropriate learning tasks based on user input and presenting the generated learning tasks to the user.

[1631] "A means to adjust the difficulty of the learning task based on the user's learning progress and the questions they got wrong" refers to a function that takes into account the user's past learning history and the questions they got wrong and appropriately adjusts the difficulty of the next learning task presented.

[1632] "Means for marking the user's learning results and presenting the results to the user" refers to a function that judges whether the learning tasks answered by the user are correct or incorrect and presents the results to the user as feedback.

[1633] "Means for sending generated learning tasks to users" refers to the function of sending learning tasks generated by the generative AI model to the user's device.

[1634] "Means for displaying study assignments to the user" refers to a function that visually displays received study assignments on the user's device.

[1635] This invention relates to a learning support system using a generative AI model. This system has the function of generating learning tasks based on user input and presenting the learning tasks to the user. Specific embodiments of this system are described below.

[1636] Hardware and software used

[1637] The system uses the following hardware and software:

[1638] Server: Receives input from the user, sends prompts to the generative AI model, and sends the generated learning tasks to the user.

[1639] Terminal: Provides an interface for users to input information and displays the learning tasks received from the server. Terminals include smartphones, tablets, and PCs.

[1640] Generative AI model: Generates learning tasks based on user input. Specifically, we use a generative AI model that uses natural language processing techniques (e.g., OpenAI's GPT-4).

[1641] System Operation

[1642] Receiving user input

[1643] A user uses a terminal to provide text input to the system, for example, a user might type, "What's my lesson today?" This input is sent to the server through the terminal interface.

[1644] Parsing input data and sending prompts

[1645] The server analyzes the user's input data received from the device. The analyzed data is sent to the generative AI model as a prompt. For example, the input "What is today's study topic?" is sent to the generative AI model as a prompt.

[1646] Generating learning tasks

[1647] The generative AI model generates learning tasks based on the received prompts. The generated learning tasks are specific and actionable for the user. For example, a learning task might be generated such as, "Today, let's do 10 addition and subtraction problems of fractions in arithmetic."

[1648] Submit and view coursework

[1649] The server sends the learning tasks received from the generative AI model to the device, which then displays the received learning tasks to the user. For example, the smartphone screen might display the following text: "Today, let's do 10 math problems on adding and subtracting fractions."

[1650] Specific examples

[1651] Example 1

[1652] The user opens the app on their smartphone, enters "What is today's lesson topic?" and presses the send button. The device sends this input to the server. The server receives the request and sends a prompt to the generative AI model saying, "What is today's lesson topic?" The generative AI model generates a lesson topic, such as "Today, let's do 10 math problems on adding and subtracting fractions," and returns it to the server. The server sends this lesson topic to the device, which displays it on the app screen.

[1653] Example 2

[1654] The user opens a web browser on their computer, enters "What is today's study topic?" and presses the send button. The device sends this input to the server. The server receives the request and sends a prompt to the generative AI model saying "What is today's study topic?" The generative AI model generates a study topic such as "Today, let's research Japan's Warring States period in history," and returns it to the server. The server sends this study topic to the device, which displays it on the web browser screen.

[1655] Prompt Sentence Examples

[1656] Below are some example prompts to input to a generative AI model:

[1657] "What's today's lesson?"

[1658] "Please tell me what we'll be studying tomorrow."

[1659] "What should I study for my next test?"

[1660] Using these prompts, the generative AI model can provide the user with appropriate learning tasks.

[1661] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1662] Step 1:

[1663] Receiving user input

[1664] A user uses a terminal to input text into the system. For example, the user types, "What is today's study topic?" This input is sent to the server through the terminal interface. The input data is in text format, and the terminal sends this data to the server as an HTTP POST request.

[1665] Step 2:

[1666] Receiving and parsing input data

[1667] The server receives the user's input data sent from the device. The received data is analyzed by the text analysis module. As a result of the analysis, the user's intent is understood and a prompt is generated to be sent to the generative AI model. For example, the input "What are your study goals for today?" is converted into the prompt "Please tell me your study goals for today."

[1668] Step 3:

[1669] Sending prompts to generative AI models

[1670] The server sends the parsed prompt to the generative AI model. This prompt must be in a format that the generative AI model can understand. The generative AI model receives the prompt and performs data calculations to generate an appropriate learning task. For example, the prompt sent might be, "Please tell me today's learning task."

[1671] Step 4:

[1672] Generating learning tasks

[1673] The generative AI model generates learning tasks based on the received prompt. The generated learning tasks are specific and actionable for the user. For example, a learning task might be generated such as, "Today, let's do 10 addition and subtraction problems of fractions in arithmetic." The generative AI model returns this learning task to the server.

[1674] Step 5:

[1675] Receiving and sending assignments

[1676] The server receives the learning task from the generative AI model. It then prepares the received learning task to send to the user. Specifically, it formats the learning task as an HTTP response and prepares to send it to the device. For example, it prepares the text "Today, let's do 10 math problems on adding and subtracting fractions" as an HTTP response.

[1677] Step 6:

[1678] Submit your coursework

[1679] The server sends the prepared learning tasks to the device. This is done using a communication protocol such as an HTTP response. For example, the server sends the following text to the device as an HTTP response: "Today, let's do 10 math problems on adding and subtracting fractions."

[1680] Step 7:

[1681] Viewing learning assignments

[1682] The device receives the learning tasks from the server and displays them to the user. The way they are displayed depends on the interface the user is accessing. For example, a smartphone screen might display the following text: "Today, let's do 10 math problems on adding and subtracting fractions."

[1683] (Application example 1)

[1684] Next, a description will be given of Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1685] Conventional learning support systems can only be used by users at home or in specific locations, and do not provide sufficient learning support in physical stores. Furthermore, they lacked the functionality to allow users to generate and display learning tasks in physical stores, resulting in low learning efficiency and convenience. Furthermore, they lacked the functionality to adjust the difficulty of learning tasks based on the user's learning progress or incorrect questions, making it difficult to meet individual learning needs. To address these issues, there is a need for a system that strengthens learning support in physical stores and allows users to generate and display learning tasks using their mobile information devices.

[1686] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1687] In this invention, the server includes means for generating study tasks based on input from the user using a generation AI and presenting the study tasks to the user, means for adjusting the difficulty of the study tasks based on the user's learning progress and questions that they have gotten wrong, means for grading the user's learning results and presenting the results to the user, means for the user to generate and present study tasks using a mobile information terminal in a physical store, and means for displaying the generated study tasks to the user in the physical store. This enhances learning support in physical stores and enables users to efficiently generate and present study tasks using their mobile information terminals.

[1688] "Generative AI" is an artificial intelligence system that generates learning tasks based on user input and presents them to the user.

[1689] "User" means an individual or organization that receives learning assignments using the learning support system.

[1690] A "learning assignment" is a collection of specific learning content or problems that a user must tackle.

[1691] A "mobile information terminal" is a portable information processing device such as a smartphone or tablet.

[1692] A "physical store" is a commercial establishment or service provider that exists in a physical location and can be visited by users.

[1693] "Study progress" is an indicator that shows how far the user has progressed in the learning tasks.

[1694] "Difficulty adjustment" is the process of changing the difficulty of a learning task based on the user's learning progress and the questions they get wrong.

[1695] "Marking" is the process of determining whether a user's learning results are correct or incorrect and evaluating them.

[1696] "Feedback" refers to evaluations and advice provided to users regarding their learning results.

[1697] "Learning motivation" refers to the degree of interest and motivation a user has in learning.

[1698] A system for implementing this invention uses a generative AI model to generate study tasks based on user input and present them to the user. The system adjusts the difficulty of the study tasks based on the user's learning progress and incorrect questions, marks the user's learning results, and presents the results to the user. The system also has a function that allows users to generate and present study tasks using their mobile information terminals in physical stores.

[1699] The server receives input from the user using a generative AI model and generates a prompt sentence. For example, if the user inputs "What is today's study topic?", the server generates the prompt sentence "The user input 'What is today's study topic?' Please generate today's study topic." The generated prompt sentence is sent to the generative AI model, which generates a specific study topic.

[1700] The generated learning tasks are sent from the server to the user's mobile information device and displayed on the device. If the user is in a physical store, they can check and work on the learning tasks through their device. The progress of the learning tasks and any questions that they got wrong are sent to the server, which then adjusts the difficulty of the learning tasks based on this information.

[1701] The server automatically marks the user's learning results and presents the results to the user, allowing the user to check their learning progress in real time.The server also provides feedback based on the user's learning results, increasing the user's motivation to learn.

[1702] As a concrete example, if a user visits a brick-and-mortar bookstore and uses a mobile information terminal to type, "What is today's study assignment?", the server will use the generative AI model to generate a study assignment such as, "Today, let's do 10 math problems on adding and subtracting fractions," and display it on the terminal. The user can then work on this assignment and send the results back to the server, which will adjust the difficulty of the next study assignment.

[1703] Example prompt sentence:

[1704] The user types, "What's today's lesson plan?" Generate today's lesson plan.

[1705] In this way, learning support in brick-and-mortar stores is enhanced, and users can efficiently generate and present learning tasks using mobile information terminals.

[1706] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1707] Step 1:

[1708] A user uses a mobile information terminal to input a request for a learning assignment.

[1709] Input: User types "What should I study today?"

[1710] Output: The user's input data.

[1711] Specific operation: The user launches the application on the device, enters "What is today's study topic?" in the text box, and presses the submit button.

[1712] Step 2:

[1713] The device sends the user's input to the server.

[1714] Input: User input data.

[1715] Output: The request data sent to the server.

[1716] Specific operation: The terminal sends the user's input data to the server as an HTTP request.

[1717] Step 3:

[1718] The server receives the user's input and generates a prompt.

[1719] Input: User input data.

[1720] Output: The generated prompt statement.

[1721] Specific operation: The server analyzes the user's input data and generates a prompt sentence such as, "The user entered 'What are today's study topics?' Please generate today's study topics."

[1722] Step 4:

[1723] The server sends prompt sentences to the generative AI model to generate learning tasks.

[1724] Input: The generated prompt statement.

[1725] Output: The generated learning task.

[1726] How it works: The server sends a prompt to a generative AI model (such as OpenAI's GPT-3) to generate a learning task. Based on the prompt, the generative AI model generates a learning task such as "Today, let's do 10 addition and subtraction problems with fractions in arithmetic."

[1727] Step 5:

[1728] The server transmits the generated learning assignments to the terminal.

[1729] Input: The generated learning task.

[1730] Output: The learning task data sent to the device.

[1731] Specific operation: The server sends the generated learning task to the terminal as an HTTP response.

[1732] Step 6:

[1733] The device displays the received learning assignments to the user.

[1734] Input: Learning task data received from the server.

[1735] Output: The learning task that is displayed to the user.

[1736] Specific operation: The device analyzes the received learning task data and displays it on the screen. The user can check the learning task on the device screen.

[1737] Step 7:

[1738] The user works on the learning tasks and enters the results into the device.

[1739] Input: User's learning result data.

[1740] Output: Learning result data input to the terminal.

[1741] Specific operation: The user uses the input form on the device to enter the answer to the learning task and presses the submit button.

[1742] Step 8:

[1743] The device sends the user's learning results to the server.

[1744] Input: User's learning result data.

[1745] Output: Training result data sent to the server.

[1746] Specific operation: The device sends the user's learning result data to the server as an HTTP request.

[1747] Step 9:

[1748] The server marks the user's learning results and generates the results.

[1749] Input: User's learning result data.

[1750] Output: Rounding result data.

[1751] Specific operation: The server analyzes the user's learning result data, determines whether the answer is correct or incorrect, and generates a marking result.

[1752] Step 10:

[1753] The server sends the rounding results to the terminal.

[1754] Input: Rounding result data.

[1755] Output: Rounding result data sent to the terminal.

[1756] Specific operation: The server sends the rounding result data to the terminal as an HTTP response.

[1757] Step 11:

[1758] The terminal displays the received marking results to the user.

[1759] Input: Marking result data received from the server.

[1760] Output: Rounding results displayed to the user.

[1761] Specific operation: The terminal analyzes the received marking result data and displays it on the screen. The user can check the marking result on the terminal screen.

[1762] Example 2

[1763] Next, a description will be given of Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1764] Conventional learning systems lacked the ability to adjust the difficulty of learning tasks based on the user's learning progress and the questions they got wrong, making it difficult to provide an optimized learning experience for each individual user. Furthermore, they lacked a mechanism for efficiently managing users' answer data and reflecting it in the next learning task, preventing improvements in learning effectiveness. Furthermore, there was a need for a system that could consistently generate and present learning tasks using a generative AI model.

[1765] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1766] In this invention, the server includes a means for storing the user's answer data in a database, a means for sending a prompt to the generative AI model to generate the next learning task, and a means for presenting the generated learning task to the user. This makes it possible to adjust the difficulty of the learning task based on the user's learning progress and questions that they got wrong, providing an optimized learning experience for each individual user.

[1767] "Generative AI" refers to artificial intelligence technology that generates learning tasks based on user input and presents them to the user.

[1768] "User" refers to an individual or group working on a learning task.

[1769] "Study assignments" refer to problems or exercises that users must complete.

[1770] "Study progress" refers to an indicator that shows how far a user has progressed in their studies.

[1771] "Incorrect questions" refer to questions that the user was unable to answer correctly.

[1772] "Difficulty adjustment" refers to changing the difficulty of learning tasks based on the user's learning progress and the questions they got wrong.

[1773] "Marking" refers to determining whether a user's learning results are correct or incorrect.

[1774] "Answer data" refers to information about the answers given by a user to a learning task.

[1775] "Database" refers to an information management system for storing user answer data.

[1776] "Generative AI model" refers to artificial intelligence technology that generates the next learning task based on a prompt.

[1777] A "prompt sentence" refers to an instruction sentence that tells the generative AI model to generate the next learning task.

[1778] "Presenting" refers to displaying the generated learning task to the user.

[1779] This invention is a system that adjusts the difficulty of learning tasks based on the user's learning progress and the questions they have answered incorrectly, providing an optimized learning experience for each individual user. Specific embodiments of this system are described below.

[1780] Hardware and software used

[1781] server

[1782] The server is responsible for collecting user answer data and storing it in a database. A database management system such as MySQL is used for the server. It also sends prompts to the generative AI model and processes the data to generate the next learning task. For example, OpenAI's GPT-4 is used as the generative AI model.

[1783] Terminal

[1784] The device provides an interface for users to work on learning tasks. The device is built using web technologies such as HTML, CSS, and JavaScript. The device has the function of sending the user's answers to the server in real time.

[1785] user

[1786] Users work on learning tasks provided through their devices. When they answer the questions, the results are sent to the server and used to adjust the difficulty of the next learning task.

[1787] Explanation of program processing

[1788] Server Processing

[1789] The server receives the user's answer data and stores it in a database such as MySQL. For example, it stores information such as the user ID, question ID, answer result, and answer time. The server then sends a prompt to the generative AI model. The prompt describes what questions the user made the most mistakes on and what questions should be presented as the next learning assignment.

[1790] Processing generative AI models

[1791] The generative AI model generates the next learning task based on the prompt received from the server. For example, if the user makes many mistakes in fraction addition, the next learning task will focus on fraction addition.

[1792] Terminal handling

[1793] The device displays the new learning tasks received from the server to the user, who can then work on the learning tasks again.

[1794] Specific examples

[1795] For example, if a user answers "1 / 2 + 1 / 3 = 5 / 6", the answer is sent to the server and saved in the database as "User ID: 12345, Question ID: 67890, Answer: 5 / 6, Correct / Incorrect: Incorrect, Answer Time: 30 seconds". The server then sends the following prompt to the generative AI model:

[1796] Prompt Sentence Examples

[1797] Users made a lot of mistakes on fraction addition problems, so please focus on fraction addition in the next lesson. Specifically, please include problems like:

[1798] 1. 1 / 2 + 1 / 3 = ?

[1799] 2. 3 / 4 + 2 / 5 = ?

[1800] 3. 5 / 6 + 1 / 2 = ?

[1801] The generative AI model generates the next learning task based on this prompt and returns it to the server. The server then sends the problem set to the device, which then displays it as a new learning task for the user. This makes it possible to adjust the difficulty of the learning task based on the user's learning progress and the questions they got wrong, providing an optimized learning experience for each individual user.

[1802] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1803] Step 1:

[1804] Users work on learning tasks

[1805] The user works on a learning task provided through the device. For example, the user answers a fraction addition problem. The user's answer (e.g., "1 / 2 + 1 / 3 = 5 / 6") is obtained as input. The user's answer data is generated as output.

[1806] Step 2:

[1807] The device sends the user's answer to the server

[1808] The terminal sends the user's answers to the server in real time. Specifically, it sends the user's answer data (e.g., "User ID: 12345, Question ID: 67890, Answer: 5 / 6") to the server. The user's answer data is obtained as input. The answer data sent to the server is generated as output.

[1809] Step 3:

[1810] The server saves the answer data in a database

[1811] The server stores the received answer data in a database. Specifically, it uses a database management system such as MySQL to store information such as user ID, question ID, answer result, and answer time. The answer data sent to the server is obtained as input. The answer data stored in the database is generated as output.

[1812] Step 4:

[1813] The server sends a prompt to the generative AI model

[1814] The server sends a prompt to the generative AI model based on the user's past learning data. Specifically, it generates a prompt such as, "The user made many mistakes on fraction addition problems, so please focus on fraction addition in the next learning assignment," and sends it to the generative AI model. The user's past learning data is obtained as input. The prompt sent to the generative AI model is generated as output.

[1815] Step 5:

[1816] The generative AI model generates the next learning task

[1817] A generative AI model (for example, OpenAI's GPT-4) generates the next training problem based on the received prompt. Specifically, it generates problems such as "1 / 2 + 1 / 3 = ?", "3 / 4 + 2 / 5 = ?", and "5 / 6 + 1 / 2 = ?". The input is the prompt. The output is a generated training problem.

[1818] Step 6:

[1819] The server sends the generated learning assignments to the device.

[1820] The server sends the learning tasks received from the generative AI model to the terminal. Specifically, it sends the generated problem set to the terminal in JSON format. The generated learning tasks are obtained as input. The learning tasks sent to the terminal are generated as output.

[1821] Step 7:

[1822] The device displays new learning tasks to the user.

[1823] The terminal displays the new learning task received from the server to the user. The user can then try the learning task again. As input, the learning task sent from the server is obtained. As output, the learning task displayed to the user is generated.

[1824] (Application example 2)

[1825] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1826] Conventional learning support systems lacked the ability to adjust the difficulty of learning tasks based on the user's learning progress and the questions they got wrong, making it difficult for users to focus on areas in which they were weak. Furthermore, they lacked the ability to record the user's learning progress and optimize the next learning task, which led to problems with reduced learning efficiency. Furthermore, the lack of a function to focus questions on the category in which the user was weakest limited the effectiveness of learning.

[1827] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1828] In this invention, the server includes means for generating study tasks based on user input using a generation AI and presenting the study tasks to the user, means for adjusting the difficulty of the study tasks based on the user's learning progress and questions that they got wrong, means for grading the user's learning results and presenting the results to the user, means for recording the user's learning progress and optimizing the next study task, and means for focusing questions on the category in which the user is weakest. This makes it possible to provide optimal study tasks according to the user's learning progress, improving the efficiency and effectiveness of learning.

[1829] "Generative AI" is an artificial intelligence system that generates learning tasks based on user input and presents them to the user.

[1830] "Learning assignments" are problems or exercises that users must work on, and are generated by generative AI.

[1831] "User's learning progress" is an index that indicates how far the user has progressed in their learning, and represents the progress of their learning.

[1832] "Wrong questions" are questions that the user answered incorrectly, and are data used to evaluate the user's level of understanding.

[1833] "Difficulty of learning task" is an indicator of how difficult the learning task is, and is adjusted based on the user's learning progress and the questions they got wrong.

[1834] "Marking" is an evaluation method that judges whether a user's learning results are correct or incorrect and presents the results to the user.

[1835] "User's learning results" are data that indicate how the user answered the learning tasks, and are the subject of marking.

[1836] "Recording the user's learning progress" means saving data on how far the user has progressed in their studies, which serves as the basis for optimizing the next learning task.

[1837] "Optimizing the next learning task" means providing the next learning task in the most optimal form based on the user's learning progress and the questions they got wrong.

[1838] A "category" is a specific field or theme to which a learning task belongs, and indicates the classification of the content that the user is learning.

[1839] "Focusing on questions" means providing many questions on categories that the user is particularly weak at, and is a means of improving the user's level of understanding.

[1840] As an embodiment of the present invention, the following system is constructed.

[1841] System Overview

[1842] The system adjusts the difficulty of the learning tasks based on the user's learning progress and the questions they got wrong, and has the ability to optimize the next learning task. Specifically, it uses a generative AI model to generate learning tasks based on user input and presents them to the user. It also marks the user's learning results and presents the results to the user.

[1843] Hardware and software used

[1844] Hardware: Smartphones, servers

[1845] Software: Python, database (e.g., SQLite), generative AI model

[1846] Data processing and calculation

[1847] The server receives input data from the user and generates learning tasks using a generative AI model. The generated learning tasks are presented to the user's smartphone. The user works on the learning tasks and sends the results to the server. The server records the user's learning results and processes the data to optimize the next learning task.

[1848] Specific examples

[1849] For example, consider a situation where a user answers a quiz about robot operating procedures. If the user makes a mistake in a particular operating procedure, the next learning task will focus on that operating procedure.

[1850] Example prompt sentence:

[1851] User ID: user1

[1852] Category: robot_operation

[1853] Correct Answer: False

[1854] When this prompt is entered, the server records the user's learning progress and focuses questions related to "robot_operation" on the next learning assignment.

[1855] Processing flow

[1856] 1. The user works on a learning task using a smartphone.

[1857] 2. The user's answer is sent to the server.

[1858] 3. The server records the user's learning progress and the questions they got wrong.

[1859] 4. The generative AI model generates the next learning task and presents it to the user.

[1860] In this way, it is possible to provide optimal learning tasks according to the user's learning progress, thereby improving the efficiency and effectiveness of learning.

[1861] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1862] Step 1:

[1863] Users use their smartphones to complete learning tasks.

[1864] Input: User's answer data (e.g., user ID, category, answer result)

[1865] Output: The user's answer results are sent to the server.

[1866] Specific operation: The user opens the application on their smartphone and enters answers to the presented learning tasks. Once the answers are completed, the answer data is sent to the server.

[1867] Step 2:

[1868] The server receives the user's answer results and records them in a database.

[1869] Input: User's answer data (e.g., user ID, category, answer result)

[1870] Output: User learning progress data recorded in a database

[1871] Specific operation: The server analyzes the received answer data and records the user's learning progress and incorrect questions in a database, thereby accumulating the user's learning history.

[1872] Step 3:

[1873] The server generates the next study task based on the user's learning progress and the questions they got wrong.

[1874] Input: User learning progress data, data on questions answered incorrectly

[1875] Output: Generated next learning task data

[1876] How it works: The server analyzes the user's learning progress and incorrect questions recorded in the database, and uses the generative AI model to generate the next learning task. The generated learning task is adjusted to include a focus on the categories in which the user is weak.

[1877] Step 4:

[1878] The server sends the generated next learning assignment to the user's smartphone.

[1879] Input: Generated next learning task data

[1880] Output: Learning tasks displayed on the user's smartphone

[1881] Specific operation: The server sends the generated learning task data to the user's smartphone, where the user can then work on the task in their next learning session.

[1882] Step 5:

[1883] The user then works on the next learning task and sends the answer to the server again.

[1884] Input: User's answer data (e.g., user ID, category, answer result)

[1885] Output: The user's answer results are sent to the server.

[1886] Specific operation: The user opens the smartphone application and enters the answer to the next learning task that is presented. Once the answer is completed, the answer data is sent back to the server.

[1887] In this way, optimal learning tasks are continuously provided according to the user's learning progress, improving the efficiency and effectiveness of learning.

[1888] Example 3

[1889] Next, a third embodiment of the third embodiment will be described. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1890] In conventional learning systems, evaluation and feedback on user answers is often done manually, which is problematic in terms of efficiency. It is also difficult to provide appropriate feedback based on the user's learning progress and level of understanding, which can lead to issues such as insufficient motivation to learn or deepening of understanding. Furthermore, there was a need for a system that could properly process user answer data and quickly present evaluation results.

[1891] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1892] In this invention, the server includes means for generating learning tasks based on user input using a generative AI model and presenting the learning tasks to the user, means for adjusting the difficulty of the learning tasks based on the user's learning progress and questions that the user got wrong, means for marking the user's learning results and presenting the results to the user, means for receiving the user's answer data, generating prompt sentences and sending them to the generative AI model, and means for receiving evaluation results and explanations from the generative AI model, storing them in a database, and presenting them to the user. This enables quick and accurate evaluation of the user's answers, and by providing appropriate feedback, it is possible to improve motivation to learn and deepen understanding.

[1893] A "generative AI model" is a type of artificial intelligence that generates learning tasks and evaluates answers based on input data from users.

[1894] A "user" is an individual or group that uses the learning system to solve learning tasks and input answers.

[1895] "Learning tasks" are problems or exercises that users must solve, and are generated by the generative AI model.

[1896] "Answer data" is information about the answer that the user inputs to the study task.

[1897] A "prompt sentence" is an instruction sentence generated by the server to request the generative AI model to evaluate the answer.

[1898] The "evaluation result" is the result of the generative AI model evaluating the user's answer, including a judgment of correctness or incorrectness and an explanation.

[1899] A "database" is an information management system for storing information such as evaluation results and explanations.

[1900] A "server" is a computer system that receives user answer data and performs processes such as sending it to the generative AI model.

[1901] "Feedback" refers to evaluations and explanations provided to users regarding their learning results, and is intended to encourage the user's learning progress and increase their motivation to learn.

[1902] The present invention relates to a system that allows a user to solve a learning task, evaluates the answer, and provides feedback. A specific embodiment of this system will be described below.

[1903] Hardware and software used

[1904] Hardware: Servers, devices (PCs, smartphones, tablets, etc.)

[1905] Software: Generative AI models (e.g., large-scale language models), web browsers, database management systems (e.g., MySQL)

[1906] System configuration

[1907] 1. The user solves a problem

[1908] The user uses a terminal to access the learning platform and inputs answers to the questions displayed.

[1909] For example, a user answers "4" to "Question 1: 2 + 2 = ?"

[1910] 2. The device sends the answer data to the server

[1911] The terminal transmits the answer data entered by the user to the server.

[1912] Specifically, the terminal uses an HTTP request to send the answer data to the server.

[1913] For example, the answer data "Question 1: 4" is sent to the server.

[1914] 3. The server generates a prompt and sends it to the generative AI model.

[1915] The server generates a prompt sentence based on the received answer data.

[1916] The generated prompt sentence is sent to the generative AI model.

[1917] For example, the prompt sentence "The user's answer is '4'. Please determine whether this answer is correct, and if it is correct, mark it as 'Correct'. If it is incorrect, mark it as 'Incorrect'. Please provide the correct answer and an explanation." is sent to the generative AI model.

[1918] 4. The generative AI model evaluates the answer and returns the result to the server

[1919] The generative AI model evaluates the user's answer based on the prompt sent from the server.

[1920] The generative AI model generates evaluation results and explanations and returns them to the server.

[1921] For example, the generative AI model returns the result "Correct. Explanation: 2 + 2 is 4" to the server.

[1922] 5. The server stores the results in a database and presents them to the user.

[1923] The server stores the evaluation results and explanations received from the generative AI model in a database.

[1924] The server sends the saved results to the user's terminal and presents them to the user.

[1925] For example, the server sends the result "Question 1: Correct. Explanation: 2 + 2 is 4" to the user's terminal.

[1926] 6. The user checks the results and continues learning

[1927] The user checks the results on the device.

[1928] Users will learn the correct answers and explanations for questions they get wrong.

[1929] For example, the user confirms the result "Question 1: Correct. Explanation: 2 + 2 is 4." and moves on to the next question.

[1930] In this way, the user can check whether their answer is correct, and if they are wrong, they can learn the correct answer and explanation. This system makes it possible to efficiently grasp the user's learning progress and deepen their understanding. The flow of the identification process in Example 3 will be described with reference to FIG. 15.

[1931] Step 1:

[1932] The user solves the problem

[1933] Input: The user accesses the learning platform using a device and enters answers to the questions displayed.

[1934] Specific action: For example, the user answers "4" to "Question 1: 2 + 2 = ?"

[1935] Output: The user's answer data is saved on the device.

[1936] Step 2:

[1937] The device sends the answer data to the server

[1938] Input: User's answer data

[1939] Specific operation: The device sends the answer data to the server using an HTTP request.

[1940] Output: For example, the answer data "Question 1: 4" is sent to the server.

[1941] Step 3:

[1942] The server generates a prompt and sends it to the generative AI model.

[1943] Input: Answer data received by the server

[1944] Specific operation: The server generates a prompt based on the received answer data. For example, it generates a prompt like this: "The user's answer is '4'. Please determine whether this answer is correct. If it is correct, mark it as 'Correct'. If it is incorrect, mark it as 'Incorrect'. Please provide the correct answer and an explanation."

[1945] Output: The generated prompt sentence is sent to the generative AI model.

[1946] Step 4:

[1947] The generative AI model evaluates the answer and returns the result to the server

[1948] Input: The prompt received by the generative AI model

[1949] Specific operation: The generative AI model evaluates the user's answer based on the prompt sentence and generates an evaluation result and explanation. For example, it generates a result such as "Correct. Explanation: 2 + 2 is 4."

[1950] Output: The evaluation result and explanation are returned to the server.

[1951] Step 5:

[1952] The server stores the results in a database and presents them to the user.

[1953] Input: Evaluation results and explanations received by the server

[1954] Specific operation: The server saves the evaluation results and explanations in a database and sends the saved results to the user's device. For example, it sends the result "Question 1: Correct. Explanation: 2 + 2 is 4" to the user's device.

[1955] Output: The evaluation results and explanations are displayed on the user's terminal.

[1956] Step 6:

[1957] Users can check the results and continue learning.

[1958] Input: Evaluation results and explanations displayed on the user's device

[1959] Specific operation: The user checks the results on the dev...

Claims

1. A first means for acquiring answer data indicating an answer to a learning task by a user; a second means for generating a first prompt sentence for outputting a next learning task for the user based on the answer data and past learning data of the user stored in a database; a third means for inputting the generated first prompt sentence into a generative AI model to generate the next learning task; A fourth means for transmitting the generated next learning assignment to the user's terminal; A system including:

2. The third means adjusts the difficulty level of the next learning task to be generated based on past learning data of the user indicated in the first prompt sentence. The system of claim 1 .

3. a fifth means for estimating an emotional state of the user from the user's facial expression and tone of voice using an emotion engine; a sixth means for adjusting how the next learning task is presented to the user based on the emotional state of the user; 3. The system according to claim 1 or 2, comprising:

Images