system

Abstract

To provide a system.SOLUTION: A system includes: first means for transmitting initial setting information based on learning contents selected by a user; second means for generating a first prompt sentence for outputting a question corresponding to understanding of the user on the basis of the received initial setting information; third means for inputting the generated first prompt sentence into a generative AI model to generate a question corresponding to the understanding of the user; fourth means for displaying the generated question corresponding to the understanding of the user on 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] While there is an increasing demand for creative practical lessons in today's educational settings, a decline in the quality of lessons due to a lack of resources is an issue. [Means for solving the problem]

[0005] By using generative AI to provide feedback and supplementary information based on learning progress and level of understanding, and by automatically generating practice problems suited to each individual and providing assistance with answering them, we can improve students' learning effectiveness while supporting their individual needs, thereby contributing to improving the quality of practical classes. [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] The present invention is an educational platform that uses generative AI. Specifically, it grasps the learner's learning progress and level of understanding and provides feedback and supplementary information accordingly. This feedback and supplementary information is provided in the form of text, audio, video, etc., and helps deepen the learner's understanding. It also automatically generates exercises based on the learner's level of understanding, interests, and learning style. This provides questions that are tailored to each learner, maximizing learning effectiveness. It also provides an answer support function to support the learner as they work through the questions.

[0029] "Example 2"

[0030] As a concrete example, let's imagine a mathematics seminar class. Suppose a student is studying calculus. The generative AI grasps the student's level of understanding and automatically generates problems that correspond to the student's level of understanding, from basic concepts to applications of calculus. It also provides hints and explanations for the answers as the student works on the problems, deepening the student's understanding.

[0031] "Example 3"

[0032] Furthermore, generative AI accumulates learners' learning history and progress, and uses this information to analyze their level of understanding and learning style, thereby providing optimal learning support for each learner and maximizing learning effectiveness.

[0033] The processing flow of each embodiment will be described below.

[0034] "Example 1"

[0035] Step 1: The generative AI understands the learner's learning progress and level of understanding based on the learner's activities on the platform and test results.

[0036] Step 2: Generate and provide feedback and supplementary information based on the learner's level of understanding. This can be in the form of text, audio, or video to deepen the learner's understanding.

[0037] Step 3: Automatically generate exercises tailored to the learner's level of understanding, interests, and learning style. This provides questions that are tailored to each individual learner, maximizing learning effectiveness.

[0038] Step 4: As students work through the problems, provide them with hints and explanations to deepen their understanding.

[0039] "Example 2"

[0040] Step 1: The student begins learning calculus.

[0041] Step 2: The generative AI grasps the learner's level of understanding and automatically generates questions that correspond to the learner's level of understanding, from basic concepts of calculus to applications.

[0042] Step 3: As students work through the problems, provide them with hints and explanations to deepen their understanding.

[0043] "Example 3"

[0044] Step 1: The generative AI accumulates the learner's learning history and progress.

[0045] Step 2: The generative AI analyzes the learner's level of understanding and learning style based on the accumulated data.

[0046] Step 3: Based on the analysis results, provide learning support optimized for each learner to maximize learning effectiveness.

[0047] Example 1

[0048] 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."

[0049] With conventional educational platforms, it is difficult to provide appropriate feedback and supplementary information according to each learner's progress and level of understanding, and there is a lack of automatic generation of exercises and answer support that meet individual needs. This makes it difficult to maximize learning effectiveness, and the challenge is to provide an optimal learning experience that matches the learner's level of understanding and interests.

[0050] 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.

[0051] In this invention, the server includes means for collecting learner progress data, means for analyzing the collected data and evaluating the learner's level of understanding, means for generating feedback and supplementary information based on the evaluation results, means for automatically generating exercises according to the learner's level of understanding and interests, and means for providing support for answering the exercises. This makes it possible to provide optimal feedback and supplementary information to each learner, automatically generate exercises that meet the learner's individual needs, and provide support for answering the exercises.

[0052] "Student progress data" refers to a record of the learning activities that a learner has performed on the educational platform, and includes information such as study time, percentage of correct answers, and number of problems attempted.

[0053] A "means for collecting" is a method or device for obtaining learner progress data from a learning management system or other educational platform.

[0054] "Means for analyzing and assessing comprehension" refers to methods and devices for assessing learners' comprehension based on collected data, and involves analyzing the data using a generative AI model.

[0055] The "means for generating feedback and supplementary information" refers to a method or device for generating appropriate feedback and supplementary information based on the results of the learner's comprehension assessment.

[0056] "Means for automatically generating exercises" refers to a method or device for automatically creating exercises that correspond to the learner's level of understanding and interests.

[0057] A "means for providing solution support" is a method or device for providing hints and step-by-step explanations of solutions to learners as they work through practice problems.

[0058] This invention relates to an educational platform that uses a generative AI model. Specifically, it is a system that grasps a learner's learning progress and level of understanding and provides feedback and supplementary information accordingly. The system aims to maximize learning effectiveness by automatically generating exercises tailored to each learner and providing answer support functions.

[0059] The server collects learner progress data from learning management systems (LMS) and other educational platforms. Collected data includes study time, correct answer rate, and number of problems attempted. The server inputs this data into a generative AI model (e.g., OpenAI's GPT-4®) to assess the learner's level of understanding. The generative AI model analyzes the learner's response patterns and progress to identify which concepts and skills the learner is lacking.

[0060] Based on the analysis results, the server generates appropriate feedback and supplementary information for the learner. For example, if the learner's understanding of a particular concept is lacking, the server generates detailed explanations and examples in the form of text or video. This allows the learner to supplement their missing knowledge and deepen their understanding.

[0061] Furthermore, the server automatically generates exercises tailored to the learner's level of understanding, interests, and learning style. Using a generative AI model, it creates exercises with the optimal difficulty and content for each learner. For example, it starts with basic integral problems and gradually increases the difficulty level. This allows learners to progress at their own pace.

[0062] The server provides solution support functions for students as they work on practice problems. Specifically, it generates hints and step-by-step explanations of solutions to support the problem-solving process. This helps students understand the problem-solving process more easily, improving their learning effectiveness.

[0063] As a concrete example, consider a user studying a mathematics unit on calculus. The server collects the user's progress data and inputs it into a generative AI model. The generative AI model detects that the user has insufficient understanding of a particular concept (e.g., the Fundamental Theorem of Integration). Based on this information, the server generates supplementary information about the Fundamental Theorem of Integration in the form of text and video and provides it to the user.

[0064] Furthermore, the server automatically generates exercises tailored to the user's level of understanding. For example, it starts with basic integral problems and provides problems of gradually increasing difficulty. As the user works on the problems, a solution assistance function is applied, providing hints and step-by-step explanations of the solutions.

[0065] Example prompt sentence:

[0066] Generate supplementary information about the Fundamental Theorem of Integration based on the user's learning progress. Create text and video content to explain the theory in a way that is easy for users to understand.

[0067] In this way, the server leverages generative AI models to provide each learner with an optimal learning experience.

[0068] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0069] Step 1:

[0070] The server collects learner progress data from learning management systems (LMS) and other educational platforms. Specifically, it obtains data such as study time, correct answer rate, and number of questions attempted through API. The input is the learner progress data, and the output is the collected progress data.

[0071] Step 2:

[0072] The server inputs the collected progress data into a generative AI model to evaluate the learner's level of understanding. The generative AI model analyzes the learner's response patterns and progress status to identify which concepts and skills are lacking. The input is the collected progress data, and the output is the comprehension assessment results.

[0073] Step 3:

[0074] The server generates feedback and supplementary information based on the results of comprehension assessment. For example, if a user's understanding of a particular concept is insufficient, it generates detailed explanations and examples of that concept in text or video format. The input is the comprehension assessment result, and the output is the generated feedback and supplementary information.

[0075] Step 4:

[0076] The server automatically generates exercises based on the learner's level of comprehension and interests. Using a generative AI model, it creates problems with the optimal difficulty and content for each learner. For example, it starts with basic integral problems and provides problems of gradually increasing difficulty. The input is the comprehension assessment results, and the output is the automatically generated exercises.

[0077] Step 5:

[0078] The server provides answer support functions when students work on exercises. Specifically, it generates hints and step-by-step explanations of solutions to support the process of solving the problems. The input is the automatically generated exercises, and the output is the provided answer support information.

[0079] In this way, the server leverages generative AI models to provide each learner with an optimal learning experience.

[0080] (Application example 1)

[0081] 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."

[0082] Conventional educational platforms have difficulty providing individual feedback and supplementary information based on learners' progress and level of understanding, which prevents them from fully improving learning outcomes. Furthermore, in the education and training of factory workers, there is a lack of individual progress management and feedback based on their level of understanding, making it difficult to improve work efficiency and quality. Furthermore, the automatic generation of exercises and answer support functions are inadequate, making it difficult to provide education that meets the individual needs of learners and workers.

[0083] 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.

[0084] In this invention, the server includes means for using generative AI to provide feedback and supplementary information according to the progress and level of understanding of learning, means for automatically generating practice problems suited to each worker, means for providing answer support, means for monitoring the worker's progress in real time, means for providing feedback according to the worker's level of understanding in the form of text, audio, and video, means for automatically generating practice problems based on the worker's level of understanding and interests, and means for supporting the worker when working on the problems. This enables education and training that meets the individual needs of learners and workers, improving learning effectiveness, work efficiency, and quality.

[0085] "Generative AI" is an artificial intelligence technology that generates new information and content based on data.

[0086] "Learning progress" is an indicator of how far a learner has progressed in the learning process.

[0087] "Level of understanding" is an indicator that shows how well a learner understands a particular content.

[0088] "Feedback" is evaluation or advice given to a learner regarding their behavior or performance.

[0089] "Supplementary information" is additional information provided to enhance the learner's understanding.

[0090] "Practice problems" are problems that learners solve to confirm what they have learned.

[0091] "Auto-generation" means that a system automatically creates content or information without human intervention.

[0092] "Solution support" refers to advice and hints provided to learners when solving problems.

[0093] "Workers" refers to people who perform work in factories or work sites.

[0094] "Real-time monitoring" means monitoring the ongoing situation immediately.

[0095] "Text" is information written in text.

[0096] "Sound" is audible information.

[0097] "Video" is a moving image that is visually displayed.

[0098] "Interests" are areas or content in which a learner or worker is particularly interested.

[0099] This invention is a system that applies an educational platform using generative AI to the education and training of factory workers. A specific embodiment of this system is shown below.

[0100] System configuration

[0101] The system consists of the following major components:

[0102] 1. Server: Runs the generative AI and provides feedback and supplementary information based on learning progress and level of understanding.

[0103] 2. Devices: Tablets or smartphones used by workers, allowing them to receive real-time feedback and practice questions.

[0104] 3. Network: Connects the server and the terminal to send and receive data.

[0105] Program processing

[0106] The server processes the data in the following steps:

[0107] 1. Collecting progress data: Collect progress data from the workers' devices, including the progress and understanding of the work.

[0108] 2. Feedback generation: Based on the progress data collected, generative AI is used to generate appropriate feedback, which can be in the form of text, audio, or video.

[0109] 3. Automatic generation of exercises: Generative AI is used to automatically generate exercises based on the worker's level of understanding and interests.

[0110] 4. Solution support: When workers tackle practice problems, generative AI is used to provide solution support.

[0111] Hardware and software used

[0112] Hardware: Factory robots, tablets and smartphones used by workers

[0113] Software: Python (registered trademark), OpenAI API

[0114] Specific examples

[0115] Progress data collection

[0116] The server collects the following progress data from the worker's terminal:

[0117] Worker ID: 12345

[0118] Learning progress: 50%

[0119] Comprehension: Medium

[0120] Interests: Machine operation

[0121] Generate feedback

[0122] The server generates the following feedback based on the progress data collected:

[0123] "You're making good progress. Now let's review the basics of machine operation."

[0124] Automatic generation of exercises

[0125] The server generates the following exercises based on the worker's level of understanding:

[0126] "Please answer the following questions regarding the basics of machine operation: 1. Explain the procedure for starting a machine."

[0127] Prompt Sentence Examples

[0128] Below are some examples of prompts used with generative AI:

[0129] Feedback generation prompt: "Generate appropriate feedback based on the following worker progress data: Worker ID: 12345, Learning Progress: 50%, Understanding: Medium, Interest: Machine Operation."

[0130] Exercise generation prompt: "Generate exercises suitable for workers with a medium level of understanding."

[0131] In this way, the server can use generative AI to support worker education and training, improving learning effectiveness, work efficiency, and quality.

[0132] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0133] Step 1:

[0134] The server collects progress data from the worker's device. Specifically, it acquires data such as the worker's ID, learning progress, level of understanding, and interest. The input is the progress data sent from the worker's device, and the output is the progress data saved on the server.

[0135] Step 2:

[0136] The server sends prompts to the generative AI based on the collected progress data and generates feedback. Specifically, it converts the progress data into prompts in text format and sends them to the generative AI. The inputs are the progress data and prompts, and the output is the generated feedback.

[0137] Step 3:

[0138] The server sends the generated feedback to the worker's terminal. Specifically, the feedback is delivered to the worker's terminal in the form of text, audio, or video. The input is the generated feedback, and the output is the feedback displayed on the worker's terminal.

[0139] Step 4:

[0140] The server automatically generates exercises based on the worker's level of understanding and interest. Specifically, it generates prompts based on the data on level of understanding and interest and sends them to the generative AI. The input is the data on level of understanding and interest and the prompts, and the output is the generated exercises.

[0141] Step 5:

[0142] The server sends the generated exercises to the worker's terminal. Specifically, the exercises are delivered to the worker's terminal in text format. The input is the generated exercise, and the output is the exercise displayed on the worker's terminal.

[0143] Step 6:

[0144] The server provides answer support when workers tackle practice problems. Specifically, it collects the workers' answer data and sends prompts to the generative AI to generate answer support. The input is the worker's answer data and prompts, and the output is the generated answer support.

[0145] Step 7:

[0146] The server sends the generated answer support to the worker's terminal. Specifically, the answer support is delivered to the worker's terminal in the form of text, audio, and video. The input is the generated answer support, and the output is the answer support displayed on the worker's terminal.

[0147] Example 2

[0148] 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."

[0149] Conventional learning support systems have had issues with the difficulty of providing appropriate questions according to the learner's level of understanding, and the lack of feedback on answers and supplementary information limits the learning effect. In particular, there is a need to automatically generate questions that meet individual learning needs and improve the quality of answer support.

[0150] 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.

[0151] In this invention, the server includes means for collecting data such as the learner's past answer history and answer time, and evaluating the learner's level of understanding using a generative AI model, means for inputting prompts into the generative AI model to automatically generate questions according to the learner's level of understanding, means for displaying questions to the learner and providing an interface for inputting answers, and means for evaluating the learner's answers and generating and providing answer hints and explanations as necessary. This makes it possible to provide appropriate questions according to the learner's level of understanding and provide high-quality answer support.

[0152] "Learner" refers to an individual receiving education or training.

[0153] "Answer history" refers to a record of questions that a learner has answered in the past and the results of those answers.

[0154] "Response time" refers to the time it takes a learner to complete a response to a particular question.

[0155] A "generative AI model" refers to an algorithm or system that uses artificial intelligence techniques to analyze data and perform a specific task (e.g., automatically generating questions or evaluating answers).

[0156] "Understanding" refers to an indicator of how well a learner understands specific knowledge or skills.

[0157] A "prompt" is an instruction given to a generative AI model to perform a specific task.

[0158] "Interface" refers to the means or screen through which a learner interacts with a system.

[0159] "Hints" refer to additional information or advice that can be helpful to learners when solving a problem.

[0160] "Explanation" refers to information that provides detailed explanations of how to answer a question and related concepts.

[0161] The present invention is a system for automatically generating questions according to a learner's level of understanding and providing support for answering the questions. A specific embodiment of this system will be described below.

[0162] First, the server collects data such as the learner's past answer history and answer time. This data is stored in a database and managed for each learner. The server then inputs the collected data into a generative AI model (e.g., a general natural language processing model) to evaluate the learner's level of understanding.

[0163] Next, the server inputs prompt sentences into the generative AI model to automatically generate questions according to the learner's level of understanding. For example, the following prompt sentences can be used:

[0164] To assess students' understanding of the basic concept of definite integrals, generate a problem like this: 'Find the definite integral of the following function: ∫(2x + 3)dx, from x = 0 to x = 2.'

[0165] The generated questions are sent from the server to the terminal. The terminal displays the questions to the learner and provides an interface for inputting answers. The learner inputs their answer to the displayed question. For example, if the learner inputs the answer "7", the terminal sends the answer to the server.

[0166] The server receives the learner's answer and uses the generative AI model to determine whether the answer is correct. For example, the generative AI model receives a prompt such as, "Please evaluate whether the learner's answer '7' is correct." The generative AI model determines that the answer is correct. The server then sends this evaluation result to the device, which then displays "That's correct!" to the learner.

[0167] Furthermore, if the learner enters an incorrect answer, the server inputs a prompt to the generative AI model saying, "Please generate a hint or explanation for the solution to this problem." The generative AI model generates a hint saying, "As a hint for the solution to this problem, first integrate the function, and then explain how to apply the range of the definite integral." The server sends this hint to the device, which then displays the hint to the learner.

[0168] In this way, a system is realized in which the server, terminal, and user work together to provide questions that correspond to the learner's level of understanding and support their learning. This system allows learners to tackle questions that are appropriate for their level of understanding and receive high-quality answer support.

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

[0170] Step 1:

[0171] The server collects data such as the learner's past answer history and answer time.

[0172] Input: Learner's answer history data, answer time data

[0173] Data processing: Obtain each learner's answer history and answer time from the database and format them into an analyzable format.

[0174] Output: Formatted answer history data and answer time data of the learner

[0175] Step 2:

[0176] The server inputs the collected data into a generative AI model to evaluate the learner's level of understanding.

[0177] Input: Formatted answer history data of learners, answer time data

[0178] Data computation: Input data into the generative AI model and evaluate comprehension using the prompt, "Please rate this learner's comprehension."

[0179] Output: Learner comprehension assessment results

[0180] Step 3:

[0181] The server inputs prompt sentences into the generative AI model and automatically generates questions based on the learner's level of understanding.

[0182] Input: Learner comprehension assessment results, prompt text

[0183] Data calculation: The generative AI model is given a prompt, "Please generate questions that correspond to the learner's level of understanding," and an appropriate question is generated.

[0184] Output: Generated problem

[0185] Step 4:

[0186] The server sends the generated questions to the terminal.

[0187] Input: Generated question

[0188] Data processing: Convert the generated questions into a format that can be sent to the terminal.

[0189] Output: Problems sent to terminal

[0190] Step 5:

[0191] The terminal displays questions to the learner and provides an interface for entering answers.

[0192] Input: Question sent to terminal

[0193] Specific operation: The device displays the questions on the screen and provides a form where the learner can enter their answers.

[0194] Output: The answer entered by the learner

[0195] Step 6:

[0196] The terminal transmits the learner's answers to the server.

[0197] Input: The answer entered by the learner

[0198] Data processing: Convert the learner's answers into a format that can be sent to the server.

[0199] Output: The learner's answer sent to the server

[0200] Step 7:

[0201] The server receives the learner's answer and uses the generative AI model to determine whether the answer is correct.

[0202] Input: Learner's answer

[0203] Data calculation: The generative AI model is given a prompt, "Please evaluate whether the learner's answer is correct," and the model determines whether the answer is correct or not.

[0204] Output: Evaluation result of the answer

[0205] Step 8:

[0206] The server transmits the evaluation result of the answer to the terminal.

[0207] Input: Answer evaluation result

[0208] Data processing: The evaluation results of the answers are converted into a format that can be sent to the terminal.

[0209] Output: Evaluation result of the answer sent to the device

[0210] Step 9:

[0211] The terminal displays the evaluation results of the answers to the learner.

[0212] Input: Evaluation result of the answer sent to the terminal

[0213] Specific operation: The device displays the evaluation results of the answers on the screen and provides feedback to the learner.

[0214] Output: The evaluation result of the answer displayed to the learner

[0215] Step 10:

[0216] If a learner enters an incorrect answer, the server inputs a prompt to the generative AI model saying, "Please generate hints and explanations for the answer to this question."

[0217] Input: Learner's incorrect answer, prompt

[0218] Data calculation: A prompt sentence is input into the generative AI model to generate hints and explanations for the answer.

[0219] Output: Generated hints and explanations

[0220] Step 11:

[0221] The server sends the generated hints and explanations to the terminal.

[0222] Input: Generated hints and explanations

[0223] Data processing: Convert the generated hints and explanations into a format that can be sent to the device.

[0224] Output: Hints and explanations sent to the terminal

[0225] Step 12:

[0226] The device displays hints and explanations to the learner.

[0227] Input: Hints and explanations sent to your device

[0228] Specific operation: The device displays hints and explanations on the screen to deepen the learner's understanding.

[0229] Output: Hints and explanations displayed to the learner

[0230] (Application example 2)

[0231] 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."

[0232] Conventional learning systems have the problem of not providing enough questions or feedback that correspond to the learner's level of understanding, limiting the learning effect. Furthermore, there is a lack of appropriate support for learners to progress through their studies at their own pace, making it difficult to improve the quality of their learning. Furthermore, in a learning environment using mobile devices such as smartphones, there is a need for a system that can generate questions and provide answer support that correspond to individual needs.

[0233] 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.

[0234] In this invention, the server includes means for providing feedback and supplementary information according to the learner's learning progress and level of understanding using generative AI, means for automatically generating practice problems suited to each individual, means for providing answer support, means for generating questions according to the learner's level of understanding and providing hints and explanations through an application installed on a smartphone, and means for inputting prompt sentences using a generative AI model and generating questions, hints, and explanations. This enables individualized support according to the learner's level of understanding, improving both the effectiveness and quality of learning.

[0235] "Generative AI" is an artificial intelligence technology that analyzes a learner's level of understanding and progress, and automatically generates appropriate feedback, supplementary information, questions, hints, and explanations based on that analysis.

[0236] "Feedback" means providing learners with evaluations and advice based on their level of understanding and progress in response to the problems and tasks they have tackled.

[0237] "Supplementary information" is information that provides additional knowledge or explanation that learners need to deepen their understanding.

[0238] "Exercises" are problems that learners work on to understand and master specific learning content.

[0239] "Automatic generation" refers to the use of artificial intelligence or algorithms to generate questions or information without manual intervention.

[0240] "Solution support" means providing hints and explanations that learners need to solve problems, helping them arrive at the answer.

[0241] A "smartphone" is a type of mobile phone, a multi-function device that can connect to the Internet and use applications.

[0242] An "application" is a software program designed to provide a particular function or service.

[0243] A "prompt" is an instruction given to a generative AI to generate a specific output.

[0244] "Problems, hints, and explanations" are practice problems that students need to solve in order to advance their studies, clues to help them find the answers, and detailed explanations of the answers to the problems.

[0245] The system for implementing this invention uses generative AI to provide feedback and supplemental information according to the progress and level of understanding of the learning, automatically generate practice problems suited to each individual, and provide answer support. A specific embodiment of this system is shown below.

[0246] The server uses a generative AI model to analyze the learner's level of understanding and provides appropriate feedback and supplementary information based on that analysis. Specifically, as the learner works on problems through an application installed on their smartphone, the server evaluates their answers in real time to determine their level of understanding. The server then automatically generates the next problem based on the learner's level of understanding and sends it to the application.

[0247] This system uses "text-davinci-003 (registered trademark)" as the generative AI model. The server receives prompts as input and generates questions, hints, and explanations. For example, if a learner's level of comprehension is "intermediate," the following prompts are used:

[0248] Problem generation prompt: "Generate calculus problems appropriate for learners with an intermediate level of comprehension."

[0249] Hint prompt: "Please provide a hint for the following problem: {generated problem}"

[0250] Explanation prompt: "Please provide an explanation for the following problem: {generated problem}"

[0251] The server inputs these prompts into a generative AI model and sends the resulting output to a smartphone application, which displays the generated questions, hints, and explanations to the learner to support their learning.

[0252] As a concrete example, consider the case where a learner is working on a calculus problem. When the learner answers the problem, the answer is sent to the server, where a generative AI model evaluates the answer. Based on the evaluation results, the next problem is automatically generated and provided to the learner. Hints and detailed explanations for the answer are also provided at the same time, allowing the learner to deepen their understanding at their own pace.

[0253] In this way, systems using generative AI can respond individually to learners' levels of understanding, improving both the effectiveness and quality of learning.

[0254] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0255] Step 1:

[0256] The user launches the smartphone application and begins learning. The user selects the content to study (e.g., calculus) on the application. The input is the user's selected learning content, and the output is initial setting information based on that content. The application sends the initial setting information to the server based on the user's selection.

[0257] Step 2:

[0258] The server receives the user's initial setting information and inputs a prompt statement to the generative AI model. The prompt statement is an instruction statement for generating a question according to the user's level of understanding. The input is the initial setting information and the prompt statement, and the output is the generated question. The server generates a question using the generative AI model and sends the question to the application.

[0259] Step 3:

[0260] The application displays the questions received from the server to the user. The user inputs the answers to the displayed questions. The input is the user's answer, and the output is the answer data. The application sends the user's answer data to the server.

[0261] Step 4:

[0262] The server receives the user's answer data and inputs it into the generative AI model. The generative AI model evaluates the answer data and analyzes the user's level of understanding. The input is the user's answer data, and the output is the evaluation results and comprehension information. The server creates a prompt sentence to generate the next question based on the evaluation results and comprehension information.

[0263] Step 5:

[0264] The server inputs a new prompt into the generative AI model to generate the next problem. The input is the new prompt, and the output is the next problem. The server then sends the generated next problem to the application.

[0265] Step 6:

[0266] The application displays the next question received from the server to the user. The user again inputs the answer to the displayed question. The input is the user's answer, and the output is the answer data. The application sends the user's answer data to the server.

[0267] Step 7:

[0268] The server receives the user's answer data again and inputs it into the generative AI model. The generative AI model evaluates the answer data and further analyzes the user's level of understanding. The input is the user's answer data, and the output is the evaluation results and comprehension information. The server creates prompt sentences to generate hints and explanations based on the evaluation results and comprehension information.

[0269] Step 8:

[0270] The server inputs prompt sentences to generate hints and explanations into the generative AI model, and generates the hints and explanations. The input is the prompt sentence for the hint or explanation, and the output is the generated hint or explanation. The server sends the generated hint or explanation to the application.

[0271] Step 9:

[0272] The application displays hints and explanations received from the server to the user. The user then uses the displayed hints and explanations to tackle the next problem. The input is the hints and explanations, and the output is an improvement in the user's understanding. The application reports the user's learning progress to the server.

[0273] In this way, systems using generative AI can respond individually to the user's level of understanding, improving both the effectiveness and quality of learning.

[0274] Example 3

[0275] Next, a description will be given of a third embodiment of the third embodiment. 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."

[0276] Conventional learning support systems have had difficulty providing optimal feedback and supplementary information tailored to each learner's level of understanding and learning style. Furthermore, they lacked sufficient support for automatically generating practice questions and answering them to meet individual needs, limiting the improvement of learning effectiveness. This led to problems such as a decline in learner motivation and a decline in learning efficiency.

[0277] 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.

[0278] In this invention, the server includes means for collecting learners' learning history and progress data from a learning management system or online learning platform, means for saving the collected data in a database, means for preprocessing the saved data, means for analyzing the learners' level of understanding and learning style based on the preprocessed data, means for generating prompt sentences using a generative AI model to provide learning support optimized for each learner, and means for providing the generated prompt sentences to the learner. This makes it possible to provide optimal feedback and supplemental information according to the individual needs of the learner, automatically generate practice questions, and provide answer support.

[0279] A "learning management system" is a software platform that manages learners' learning history and progress data and supports educational activities.

[0280] An "online learning platform" is a system that provides learning content via the Internet and allows learners to progress with their studies online.

[0281] "Learning history" is a record of a learner's past learning activities, and is data including test scores, study time, and percentage of correct answers to questions.

[0282] "Progress data" is data that indicates the status of a learner's current learning activities, and is information that includes the learner's learning progress and achievement level.

[0283] A "database" is a system for efficiently storing, managing, and searching collected data.

[0284] "Preprocessing" refers to operations such as cleaning and shaping data, filling in missing values, and handling outliers that are performed before data analysis or machine learning.

[0285] "Level of understanding" is an indicator that shows how well a learner understands a particular learning content.

[0286] "Learning style" is a characteristic that indicates the method or approach a learner takes to learn.

[0287] A "generative AI model" is a model that uses artificial intelligence technology to generate text and data, and includes, for example, GPT-4 and BERT (registered trademark).

[0288] A "prompt" is a piece of text generated by a generative AI model that includes feedback to the learner, supplementary information, suggested practice questions, etc.

[0289] This invention is a system that collects learners' learning history and progress data from learning management systems and online learning platforms, and provides optimized learning support for each learner based on that data. Specific embodiments of this system are described below.

[0290] The server collects learners' learning history and progress data from learning management systems and online learning platforms (e.g., Moodle (registered trademark), Canvas (registered trademark)). This includes obtaining data such as test scores, study time, and percentage of questions answered correctly through APIs. For example, the server uses the Moodle API to obtain Learner A's latest test results.

[0291] Next, the server stores the collected data in a database (e.g., MySQL (registered trademark), PostgreSQL (registered trademark)). The database has a table for each learner, and stores each learner's progress data. For example, it executes an SQL query to insert the test results of learner A into the "test_results" table.

[0292] The server then uses Python to preprocess the data. Specifically, it uses the Pandas library (registered trademark) to clean and format the data, and to fill in missing values ​​and process outliers. For example, if the study time for learner A is missing, it is filled in with the average study time.

[0293] Based on the preprocessed data, the server uses machine learning libraries such as Scikit-learn (registered trademark) and TENSORFLOW (registered trademark) to analyze the learner's level of understanding and learning style. Specifically, it performs clustering and regression analysis to evaluate the learner's performance. For example, it uses K-means clustering to determine which cluster Learner A belongs to.

[0294] Next, the server uses a generative AI model (e.g., GPT-4, BERT) to generate prompts to provide learning support optimized for each learner. The generated prompts include questions and explanations based on the learner's level of understanding, as well as suggested learning methods tailored to the learner's learning style. For example, the server might generate a prompt that reads, "Student A lacks understanding of differentiation, so we recommend that he or she relearn the basic concepts of differentiation. Also, visual explanations are effective, so please provide explanations using graphs."

[0295] Finally, the terminal displays the prompt received from the server to the learner. The user (learner) proceeds with their learning based on the presented learning support. For example, learner A re-learns the basic concept of differentiation according to the presented prompt and checks the explanation using graphs.

[0296] In this way, the present invention can provide learning support that is optimized for each learner, thereby maximizing the learning effect.The flow of the identification process in the third embodiment will be described with reference to FIG.

[0297] Step 1:

[0298] The server collects learners' learning history and progress data from learning management systems and online learning platforms. Specifically, it obtains data such as test scores, study time, and percentage of correct answers via API. For example, it uses the Moodle API to obtain Learner A's latest test results.

[0299] Input: Learning history and progress data from your learning management system or online learning platform

[0300] Output: Collected learning history and progress data

[0301] Step 2:

[0302] The server saves the collected data in a database, such as MySQL or PostgreSQL. For example, it executes an SQL query to insert the test results of learner A into the "test_results" table.

[0303] Input: Collected learning history and progress data

[0304] Output: Data stored in the database

[0305] Step 3:

[0306] The server uses Python to preprocess the data. Specifically, it uses the Pandas library to clean and format the data, and to fill in missing values ​​and handle outliers. For example, if the study time for learner A is missing, it is filled in with the average study time.

[0307] Input: Data stored in a database

[0308] Output: Preprocessed data

[0309] Step 4:

[0310] The server analyzes the learner's level of understanding and learning style based on the preprocessed data. Specifically, it uses machine learning libraries such as Scikit-learn and TensorFlow to perform clustering and regression analysis to evaluate the learner's performance. For example, it uses K-means clustering to determine which cluster Learner A belongs to.

[0311] Input: Preprocessed data

[0312] Output: Analysis of learner's comprehension and learning style

[0313] Step 5:

[0314] The server uses a generative AI model to generate prompts to provide learning support optimized for each learner. Specifically, it uses generative AI models such as GPT-4 and BERT to generate prompts that include questions and explanations based on the learner's level of understanding and suggestions for learning methods tailored to the learner's learning style. For example, it generates a prompt that reads, "Student A lacks understanding of differentiation, so we recommend that he / she relearn the basic concepts of differentiation. Also, visual explanations are effective, so please provide explanations using graphs."

[0315] Input: Analysis results of learner comprehension and learning style

[0316] Output: Generated prompt statement

[0317] Step 6:

[0318] The terminal displays the prompt received from the server to the learner. The user (learner) proceeds with their learning based on the presented learning support. For example, learner A re-learns the basic concept of differentiation according to the presented prompt and checks the explanation using graphs.

[0319] Input: Generated prompt text

[0320] Output: The prompt displayed to the learner

[0321] (Application example 3)

[0322] 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."

[0323] Conventional learning support systems are limited to providing feedback and supplementary information according to the learner's progress and level of understanding, and also have limitations in automatically generating exercises and providing answer support that meet the individual needs of the learner. Furthermore, there is a lack of means to optimize the work efficiency and work style of robots working in factories, making it difficult to improve productivity throughout the factory.

[0324] 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 using generative AI to provide feedback and supplemental information according to the learning progress and level of understanding, means for automatically generating practice problems suited to each learner, means for providing answer support, means for accumulating work history data and analyzing work efficiency and work style, and means for generating optimized work instructions. This makes it possible to provide optimized learning support to each learner and maximize learning effectiveness, as well as improve the work efficiency of robots in factories and overall productivity.

[0325] "Generative AI" refers to artificial intelligence that generates new information and content based on data.

[0326] "Learning progress" is an indicator of how far a learner has progressed in the learning process.

[0327] "Level of understanding" is an indicator of how well a learner understands the learning content.

[0328] "Feedback" refers to evaluation and advice provided to learners regarding their behavior and performance.

[0329] "Supplemental information" is additional information provided to complement the learning content.

[0330] "Practice problems" are problems that learners solve to gain a practical understanding of the learning content.

[0331] "Solution support" refers to advice and hints provided to learners when solving practice problems.

[0332] "Work history data" is a record of work that robots and machines have done in the past.

[0333] "Work efficiency" is an index that indicates how much work can be completed within a certain time.

[0334] A "work style" is a specific method or pattern used by a robot or machine to perform a task.

[0335] "Optimized work instructions" are instructions that take into account work efficiency and work style to perform work most effectively.

[0336] A system for implementing this invention includes means for using generative AI to provide feedback and supplementary information according to learning progress and level of understanding, means for automatically generating practice problems suited to each individual, means for providing answer support, means for accumulating work history data and analyzing work efficiency and work style, and means for generating optimized work instructions.

[0337] The server collects learners' learning history and progress data and analyzes this data using generative AI. Specifically, it uses software such as Python, Pandas, and Scikit-learn to preprocess the data and train machine learning models. This allows it to understand the learner's level of understanding and learning style and provide optimal feedback and supplementary information.

[0338] The server also automatically generates exercises tailored to the individual needs of each learner and provides support in answering them, allowing learners to tackle exercises that are optimized for them and maximizing their learning effectiveness.

[0339] Furthermore, the server accumulates work history data of the robots working in the factory and analyzes their work efficiency and work style. Specifically, it uses machine learning algorithms such as KMeans clustering to classify the robots' work styles and generate optimized work instructions. This improves the work efficiency of the robots in the factory and increases overall productivity.

[0340] For example, based on the work history data of a factory robot over the past month, the robot's working style can be analyzed, and based on the results, optimized work instructions such as a "high-speed work mode" or a "precision work mode" can be provided.

[0341] Examples of prompts to input to a generative AI model include:

[0342] "Based on the work history data of factory robots over the past month, analyze the robot's work style and generate optimized work instructions."

[0343] In this way, the server can provide optimized learning support to each learner, maximizing the learning effect, while also improving the work efficiency of robots in the factory and increasing overall productivity.

[0344] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0345] Step 1:

[0346] The server collects learners' learning history and progress data. Specifically, it acquires log data of the learners' learning activities on the online platform and stores it in a database. The input is the learner's log data, and the output is the saved learning history data.

[0347] Step 2:

[0348] The server preprocesses the collected learning history data. Specifically, it cleans and formats the data using the Python Pandas library. The input is the saved learning history data, and the output is the preprocessed data.

[0349] Step 3:

[0350] The server trains a generative AI model based on the preprocessed data. Specifically, it uses the Scikit-learn library to build a machine learning model and analyzes the learner's level of understanding and learning style. The input is the preprocessed data, and the output is the trained generative AI model.

[0351] Step 4:

[0352] The server uses a trained generative AI model to generate optimal feedback and supplemental information for the learner. Specifically, it inputs a prompt sentence into the generative AI model and generates appropriate feedback and supplemental information. The input is the prompt sentence and the trained generative AI model, and the output is the generated feedback and supplemental information.

[0353] Step 5:

[0354] The server automatically generates exercises that meet the individual needs of each learner. Specifically, it uses a generative AI model to generate exercises that correspond to the learner's level of understanding. The input is the learner's comprehension data and the generative AI model, and the output is the generated exercises.

[0355] Step 6:

[0356] The server provides answer support for the generated exercises. Specifically, it uses a generative AI model to generate hints and advice for learners when solving the exercises. The inputs are the generated exercises and the generative AI model, and the output is answer support information.

[0357] Step 7:

[0358] The server collects work history data from the robots working in the factory. Specifically, it stores data acquired from the robots' sensors in a database. The input is the robot's sensor data, and the output is the stored work history data.

[0359] Step 8:

[0360] The server preprocesses the collected work history data. Specifically, it cleans and formats the data using the Python Pandas library. The input is the saved work history data, and the output is the preprocessed data.

[0361] Step 9:

[0362] The server trains a generative AI model based on the preprocessed data. Specifically, it uses the Scikit-learn library to perform KMeans clustering to analyze the robot's working style. The input is the preprocessed data, and the output is the trained generative AI model.

[0363] Step 10:

[0364] The server uses a trained generative AI model to generate optimized work instructions. Specifically, a prompt sentence is input into the generative AI model to generate optimal work instructions for the robot. The input is the prompt sentence and the trained generative AI model, and the output is the generated work instructions.

[0365] Step 11:

[0366] The server sends the generated work instructions to the robot. Specifically, the work instructions are sent to the robot via the network, and the robot performs the work according to the instructions. The input is the generated work instructions, and the output is the robot's work actions.

[0367] 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.

[0368] "Example 1"

[0369] One embodiment of the present invention is an educational support system that combines generative AI and an emotion engine. This system uses the emotion engine to grasp the learner's emotional state in addition to their learning progress and level of understanding. Specifically, the system estimates the learner's emotional state from their facial expressions, tone of voice, and behavioral patterns during learning. The system then provides feedback and supplemental information according to the learner's emotional state, improving the learner's learning effectiveness.

[0370] "Example 2"

[0371] As a concrete example, imagine a learner is working on a math problem and the emotion engine detects frustration. In this case, the generative AI can use that information to provide the learner with easier problems or hints on how to solve them, thereby reducing frustration and improving learning outcomes.

[0372] "Example 3"

[0373] The emotion engine can also detect a learner's excitement. For example, if it detects a learner's joy after solving a problem, the generative AI can use that information to provide problems of similar difficulty, thereby maintaining the learner's excitement and increasing their motivation to learn.

[0374] The processing flow of each embodiment will be described below.

[0375] "Example 1"

[0376] Step 1: The learner activates the learning system and engages with the learning materials.

[0377] Step 2: The emotion engine estimates the learner's emotional state from their facial expressions, tone of voice, and behavioral patterns during learning.

[0378] Step 3: The generative AI receives information from the emotion engine and provides feedback and supplemental information based on the learner's emotional state.

[0379] "Example 2"

[0380] Step 1: Learners tackle math problems.

[0381] Step 2: The emotion engine detects learner frustration.

[0382] Step 3: The generative AI uses that information to provide the learner with easier questions or hints on how to solve them.

[0383] "Example 3"

[0384] Step 1: Learners solve the problem.

[0385] Step 2: The emotion engine detects the learner's joy.

[0386] Step 3: The generative AI uses that information to provide problems of similar difficulty.

[0387] Example 1

[0388] 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."

[0389] Conventional educational support systems provide insufficient feedback and supplementary information based on learners' learning progress and level of understanding, and the automatic generation of practice problems and answer support tailored to individual needs are also limited. Furthermore, feedback does not take into account the learner's emotional state, making it difficult to maximize learning effectiveness. This makes it difficult to provide optimal learning support for each learner.

[0390] 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.

[0391] In this invention, the server includes means for grasping a learner's learning progress and level of understanding, means for analyzing the learner's emotional state, means for providing feedback and supplemental information based on the learner's learning progress, level of understanding, and emotional state using generative AI, means for automatically generating practice problems tailored to each individual, and means for providing answer support. This makes it possible to provide optimal feedback and supplemental information based on the learner's learning progress, level of understanding, and emotional state, and also realizes the automatic generation of practice problems and answer support tailored to individual needs. This makes it possible to provide optimal learning support to each learner and maximize learning effectiveness.

[0392] "Learning progress" is a measure of how far a learner has progressed through a learning activity.

[0393] "Level of understanding" is an indicator that shows how well a learner understands a particular learning content.

[0394] "Emotional state" refers to the learner's emotional and psychological state during learning, and can be estimated from facial expressions, tone of voice, behavioral patterns, etc.

[0395] "Generative AI" is a system that uses artificial intelligence technology to analyze data and automatically generate feedback, supplementary information, practice questions, and more.

[0396] "Feedback" refers to evaluations, advice, supplementary information, etc. provided to learners regarding their learning activities.

[0397] "Supplementary information" refers to additional information or materials provided to help learners deepen their understanding.

[0398] "Exercises" refer to problems and tasks that learners tackle to gain a practical understanding of the learning content.

[0399] "Solution support" refers to support such as hints and solution procedures provided to learners when they are working on practice problems.

[0400] "Learning effectiveness" is an indicator that shows the degree to which learners improve their knowledge and skills through learning activities.

[0401] "Individual needs" refers to the specific demands and requirements of each learner based on their learning style, level of understanding, interests, etc.

[0402] This invention relates to an educational support system that combines generative AI and an emotion engine. Specifically, it is a system that grasps a learner's learning progress, level of understanding, and emotional state, and provides feedback and supplementary information accordingly. The system aims to maximize learning effectiveness by automatically generating exercises that are optimal for each learner and providing answer support.

[0403] Hardware and software used

[0404] The server monitors learners' activities on the learning platform and collects data, such as the questions they answer, the learning materials they view, and the amount of time they spend studying. This allows the server to understand the learners' learning progress and level of understanding.

[0405] The device uses a camera and microphone to collect the learner's facial expressions and tone of voice, and the server sends this data to an emotion engine to analyze the learner's emotional state.

[0406] The server uses a generative AI model to generate feedback and supplemental information based on the collected learning data and the analysis of the learner's emotional state, and provides the learner with the feedback and supplemental information in the form of text, audio, video, etc.

[0407] Furthermore, the server automatically generates exercises tailored to the learner's level of understanding, interests, and learning style, providing each learner with the most appropriate questions.An answer support function is also provided to assist learners as they work through the questions.

[0408] Specific examples

[0409] For example, suppose a student is solving a math problem. The server collects the student's correct answer rate and answer time to grasp the student's learning progress. At the same time, the device uses a camera and microphone to analyze the student's facial expressions and tone of voice to estimate their emotional state.

[0410] If a learner is struggling with a problem, the server uses a generative AI model to generate a text message asking, "Can you explain how to solve this problem again?" and provides it to the learner. It also automatically generates related problems that the learner might be interested in and presents them as the next practice problem.

[0411] Prompt Sentence Examples

[0412] "When a learner is solving a math problem, what kind of feedback should be provided if their answer time is increasing and their facial expression shows confusion?"

[0413] In this way, the server helps the learner to maximize the learning effect.

[0414] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0415] Step 1:

[0416] The server monitors learners' activities on the learning platform and collects data, such as the questions they answer, the learning materials they view, and the amount of time they spend studying. This allows the server to understand the learners' learning progress and level of understanding.

[0417] Input: Learner's answer history, viewed materials, study time

[0418] Data processing: Analysis of log data

[0419] Output: Learning progress data, comprehension data

[0420] Specific operation: The server analyzes the log data of the learning platform to obtain the learner's answer history, and also records the type and time of the learning materials viewed by the learner.

[0421] Step 2:

[0422] The device uses a camera and microphone to collect the learner's facial expressions and tone of voice, and the server sends this data to an emotion engine to analyze the learner's emotional state.

[0423] Input: Learner's facial expression data, voice tone data

[0424] Data processing: Analysis using emotion engine

[0425] Output: Emotional state data

[0426] Specific operation: While the learner is solving the problem, the device captures facial expressions with a camera and records the tone of voice with a microphone. The server sends the collected data to an emotion engine to estimate the learner's emotional state (e.g., confusion, excitement, fatigue).

[0427] Step 3:

[0428] The server uses a generative AI model to generate feedback and supplemental information based on the collected learning data and the analysis of the learner's emotional state, and provides the learner with the feedback and supplemental information in the form of text, audio, video, etc.

[0429] Input: learning progress data, comprehension data, emotional state data

[0430] Data processing: Feedback generation using generative AI models

[0431] Output: Feedback, supplementary information

[0432] Specific behavior: If the learner is confused, the server generates a text message saying, "Can you explain how to solve this problem again?" It also recommends related videos that the learner may be interested in.

[0433] Step 4:

[0434] The server automatically generates exercises tailored to the learner's level of understanding, interests, and learning style, providing each learner with the most appropriate exercises.

[0435] Input: learning progress data, comprehension data, interest data

[0436] Data processing: Generative AI model for generating exercises

[0437] Output: Auto-generated exercises

[0438] Specific operation: The server analyzes the learner's past answer history and generates new questions based on their level of understanding. Using a generative AI model, it creates questions based on the learner's interests.

[0439] Step 5:

[0440] The server assists the learner in working through the problems, providing hints and step-by-step guides to the solutions.

[0441] Input: automatically generated exercises, learner's answer status

[0442] Data processing: Answer support generation using generative AI models

[0443] Output: Hints, solution guide

[0444] Specific operation: When a learner gets stuck on a problem, the server displays a hint such as "Here's the next step." It also provides a video that explains in detail the steps to solve the problem.

[0445] (Application example 1)

[0446] 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."

[0447] Conventional educational platforms provide insufficient feedback and supplementary information according to the learner's learning progress and level of understanding, and the automatic generation of practice problems and answer support tailored to individual needs are also limited. Furthermore, there is a problem in that the feedback provided does not take into account the learner's emotional state, which means that the learning effect is not maximized. Furthermore, when it comes to educational support for factory workers, there is a lack of support when learning how to operate and maintain robots.

[0448] 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.

[0449] In this invention, the server includes means for using generative AI to provide feedback and supplemental information according to the progress and level of understanding of the learner, means for automatically generating practice problems suited to each individual, means for providing answer support, means for grasping the learner's emotional state using an emotion engine and providing feedback and supplemental information according to that, and means for providing educational support for employees learning how to operate and maintain robots in factories. This not only maximizes the learner's learning effect and enables the provision of practice problems and answer support tailored to individual needs, but also makes it possible to maintain the learner's motivation by providing appropriate feedback according to the learner's emotional state, thereby improving the quality of employee training in factories.

[0450] "Generative AI" is an artificial intelligence technology that generates new information and content based on data.

[0451] "Learning progress" is an indicator of how far a learner has progressed in the learning process.

[0452] "Level of understanding" is an index that indicates how well a learner understands the learning content.

[0453] "Feedback" refers to evaluations and advice provided to learners based on their learning progress and level of understanding.

[0454] "Supplementary information" is additional information provided to help learners gain a deeper understanding of the content they are learning.

[0455] "Practice problems" are problems that learners solve to gain a practical understanding of the learning content.

[0456] "Auto-generation" refers to the use of artificial intelligence and algorithms to create new information and content without human intervention.

[0457] "Solution support" refers to advice and hints provided to learners when solving practice problems.

[0458] The "emotion engine" is a technology that analyzes the learner's emotional state and provides appropriate feedback and supplementary information based on that.

[0459] "Emotional state" refers to a state that indicates a learner's current feelings or mood.

[0460] "Robot operation" refers to the control and operation of robots used in factories.

[0461] "Maintenance" refers to the maintenance work carried out to keep machines and equipment operating normally.

[0462] "Employee education support" refers to education and training provided to employees working in factories to help them acquire the necessary knowledge and skills.

[0463] As an embodiment of the present invention, a smart factory training support system will be described as an example. This system combines generative AI and an emotion engine to provide training support for employees working in a factory.

[0464] The server includes a means for providing feedback and supplemental information based on learning progress and understanding using generative AI. Specifically, the server collects employee learning data and evaluates progress using a generative AI model. Based on the evaluation results, the server provides appropriate feedback and supplemental information.

[0465] The server also includes a means for automatically generating exercises tailored to each individual. Using a generative AI model, the server automatically generates and provides exercises tailored to each employee's learning content and progress. This allows employees to tackle problems tailored to their individual needs.

[0466] Furthermore, the server includes a means for providing solution support. When employees work on practice problems, the server uses a generative AI model to provide solution support. Specifically, the server provides hints and directions for solving problems.

[0467] The system also includes a means for grasping the learner's emotional state using an emotion engine and providing appropriate feedback and supplementary information. The server uses a camera to capture the employee's facial expressions and analyzes their emotional state using EmotionRecognizer. Based on the analysis results, the system provides appropriate feedback and supplementary information.

[0468] This also includes a means of providing educational support to employees learning how to operate and maintain the robots that work in factories. The server provides the necessary information and practice questions when employees learn how to operate and maintain the robots, maximizing the effectiveness of their learning.

[0469] For example, if an employee is learning how to operate a robot and their progress is slow or their emotional state is estimated to be "frustrated," the server will provide feedback such as, "Your progress is slow. Please check the additional information below," or "Take a short break to refresh yourself."

[0470] An example of a prompt is as follows:

[0471] "Enter employee learning progress data and generate feedback if progress is lagging."

[0472] In this way, the smart factory training support system can maximize employee learning effectiveness and improve efficiency within the factory.

[0473] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0474] Step 1:

[0475] The server collects employee learning data. Specifically, it obtains data on progress and understanding that employees enter into the learning platform. The input data includes study time, the percentage of correct answers to questions, and self-evaluation of the learning content. Based on this data, the server prepares basic data for evaluating learning progress.

[0476] Step 2:

[0477] The server uses a generative AI model to evaluate learning progress. Using collected learning data as input, the generative AI model analyzes progress and outputs evaluation results. Specifically, it quantifies the degree of learning progress and level of understanding and saves them as evaluation results. These evaluation results are used to provide feedback in the next step.

[0478] Step 3:

[0479] The server generates feedback and supplemental information based on the evaluation results. Using a generative AI model, it creates feedback based on the employee's progress and level of understanding. For example, if progress is behind, it generates feedback such as "Your progress is behind. Please check the supplemental information below." The generated feedback is sent to the employee's device.

[0480] Step 4:

[0481] The server automatically generates exercises suited to each individual. Using a generative AI model, exercises are generated based on the employee's learning content and progress. The employee's learning history and evaluation results are used as input data. The generated exercises are sent to the employee's device, where they can then work on them.

[0482] Step 5:

[0483] The server provides answer support. When employees work on practice problems, it uses a generative AI model to provide answer support. Specifically, it generates hints for the problems and directions for the solutions, and sends them to the employee's device. This allows the employee to receive help in solving the problems.

[0484] Step 6:

[0485] The server uses an emotion engine to understand the employee's emotional state. It uses a camera to capture the employee's facial expression and analyzes the emotional state using EmotionRecognizer. Image data of the facial expression is used as input data. The analysis results are output as the employee's emotional state.

[0486] Step 7:

[0487] The server provides feedback and supplementary information according to the employee's emotional state. Based on the analysis results of the emotion engine, the server generates appropriate feedback and supplementary information. For example, if the employee is estimated to be "frustrated," the server generates feedback such as "Take a short break to refresh yourself." The generated feedback is sent to the employee's device.

[0488] Step 8:

[0489] The server provides educational support to employees who are learning how to operate and maintain the robots that work in the factory. It provides the information and practice questions that employees need when learning how to operate and maintain the robots. Employee learning history and evaluation results are used as input data. The provided information and practice questions are sent to the employee's device, allowing the employee to continue their learning.

[0490] Example 2

[0491] 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."

[0492] Conventional learning support systems have had the problem of being unable to sufficiently improve learning effectiveness because it is difficult to provide appropriate feedback and questions based on the learner's level of understanding and emotional state. In addition, they do not adequately respond when the learner becomes frustrated, which can lead to a decrease in motivation to learn.

[0493] 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.

[0494] In this invention, the server includes means for collecting learner data, means for evaluating the learner's level of understanding based on the collected data, means for automatically generating exercises according to the learner's level of understanding using a generative AI model, means for presenting the generated exercises to the learner, means for recording and evaluating the learner's answers, means for providing hints and explanations based on the results of the answer evaluation, and means for detecting the learner's emotions and taking measures to reduce frustration. This makes it possible to provide appropriate feedback and questions according to the learner's level of understanding and emotional state, thereby improving learning effectiveness and maintaining motivation to learn.

[0495] "Means for collecting learner data" refers to a device or program for collecting questions that learners have answered in the past, their answer data, answer time, answer process, etc.

[0496] A "means for assessing a learner's level of understanding" is a device or program that analyzes and scores a learner's level of understanding based on collected data.

[0497] "Means for automatically generating exercises appropriate to a learner's level of understanding using a generative AI model" refers to a device or program that uses a generative AI model to automatically create exercises appropriate to a learner's level of understanding.

[0498] The "means for presenting the generated exercises to the learner" is a device or program for displaying the generated exercises to the learner.

[0499] "Means for recording and evaluating learner's answers" refers to a device or program for recording the answers given by learners and determining whether they are correct or incorrect.

[0500] The "means for providing hints and explanations based on the evaluation results of the answers" refers to a device or program for generating and providing appropriate hints and explanations to the learner based on the evaluation results of the answers.

[0501] "Means for detecting learners' emotions and taking measures to reduce frustration" refers to a device or program that detects emotions from a learner's facial expressions and voice, and takes appropriate measures if the learner is feeling frustrated.

[0502] The present invention is a system for automatically generating exercises according to a learner's level of understanding and providing support for solving the exercises. A specific embodiment of this system will be described below.

[0503] The server uses a database to collect learner data. Specifically, it uses a database management system such as MySQL or PostgreSQL to store questions that learners have previously answered, their answer data, answer times, answering processes, and so on.

[0504] The server then evaluates the learner's level of understanding based on the collected data. It preprocesses the data using Python's pandas library and models the learner's level of understanding using a machine learning library such as scikit-learn. For example, it can score the learner's level of understanding based on their past accuracy rate and answer time.

[0505] The server uses a generative AI model (e.g., GPT-4) to automatically generate exercises appropriate for the learner's level of comprehension. Specifically, the server incorporates the comprehension score into a prompt, and inputs a prompt such as "Please generate calculus problems suitable for a learner with a comprehension score of 70" into the generative AI model.

[0506] The device presents the questions received from the server to the learner. Specifically, the questions are displayed on the screen through a web application (for example, a front end using React (registered trademark) or Vue.js (registered trademark)).

[0507] The user (learner) answers the questions presented to them. The device records the learner's answers in real time and sends them to the server. The server evaluates the received answers and determines whether they are correct. Specifically, the server uses a Python mathematical processing library (e.g., SymPy (registered trademark)) to determine whether the answers are correct.

[0508] The server generates hints and explanations as needed based on the evaluation results of the answers. Using the generative AI model, it generates prompts such as "Please give me a hint for the answer to this problem" and inputs them into the AI ​​model. The generated hints and explanations are sent to the device and presented to the learner.

[0509] Furthermore, the device monitors the learner's facial expressions and voice in real time and detects frustration using an emotion engine (e.g., Emotion API). The server receives data from the emotion engine and provides easier questions or additional hints if the learner is frustrated.

[0510] As a concrete example, suppose a learner is working on the problem "Find the derivative of f(x) = x^2." If the device detects that the learner is struggling to find the answer and the emotion engine detects frustration, the server inputs a prompt message to the generative AI model saying, "The learner is feeling frustrated. Please provide an easier problem." The generative AI model then generates a hint such as, "Please tell me the basic steps to find the derivative of f(x) = x^2," and presents it to the learner via the device.

[0511] In this way, the server, terminal, and user work together to provide questions and support according to the learner's level of understanding, thereby improving learning effectiveness.

[0512] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0513] Step 1: Collect learner data

[0514] The server stores in a database the questions that learners have answered in the past, their answer data, the time it took to answer, the process of answering, etc. Specifically, every time a learner submits an answer, the data is recorded in a database management system such as MySQL or PostgreSQL. The input is the learner's answer data, and the output is the learning history data stored in the database.

[0515] Step 2: Assess learner comprehension

[0516] The server evaluates the learner's level of understanding based on the collected data. It preprocesses the data using Python's pandas library and models the level of understanding using machine learning libraries such as scikit-learn. The input is learning history data obtained from the database, and the output is the learner's comprehension score. Specifically, it scores the learner's level of understanding based on past correct answer rates and answer times.

[0517] Step 3: Automatic question generation

[0518] The server uses a generative AI model to automatically generate exercises that correspond to the learner's level of comprehension. Specifically, the server incorporates the comprehension score into a prompt, and inputs a prompt such as "Please generate calculus problems suitable for a learner with a comprehension score of 70" into the generative AI model. The input is the learner's comprehension score and the prompt, and the output is the generated exercises.

[0519] Step 4: State the problem

[0520] The terminal presents the questions received from the server to the learner. Specifically, the questions are displayed on the screen via a web application. The input is the exercise questions sent from the server, and the output is the questions presented to the learner.

[0521] Step 5: Record and evaluate your answers

[0522] The user (learner) answers the questions presented to them. The device records the learner's answers in real time and sends them to the server. The server evaluates the received answers and determines whether they are correct. Specifically, it uses a Python mathematical processing library (e.g., SymPy) to determine whether the answers are correct. The input is the learner's answer data, and the output is the evaluation result of the answer.

[0523] Step 6: Providing hints and explanations

[0524] The server generates hints and explanations as needed based on the evaluation results of the answers. Using the generative AI model, it generates prompts such as "Please give me a hint for the answer to this problem" and inputs them into the AI ​​model. The generated hints and explanations are sent to the device and presented to the learner. The inputs are the evaluation results of the answers and the prompt, and the output is the generated hints and explanations.

[0525] Step 7: Detect and respond to emotions

[0526] The device monitors the learner's facial expressions and voice in real time and detects frustration using an emotion engine. The server receives data from the emotion engine and, if the learner is feeling frustrated, provides easier questions or additional hints. The input is the learner's facial and voice data, and the output is the emotion detection results and countermeasures. Specifically, when the emotion engine detects frustration, the server inputs a prompt statement into the generative AI model saying, "The learner is feeling frustrated. Please provide easier questions," and presents the generated countermeasures to the learner via the device.

[0527] (Application example 2)

[0528] 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."

[0529] Conventional learning support systems and customer service support systems have had difficulty understanding users' levels of understanding and emotional states in real time and providing appropriate feedback and suggestions accordingly. Furthermore, they lacked the means to detect users' frustrations and respond appropriately, which meant that learning effects and customer satisfaction could not be fully improved.

[0530] 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.

[0531] In this invention, the server includes means for providing feedback and supplementary information according to the learning progress and level of understanding using generative AI, means for automatically generating practice problems suited to each user, means for providing answer support, means for detecting the user's emotional state using an emotion engine and taking appropriate action, and means for automatically generating and presenting suggestions according to the user's needs. This enables appropriate feedback and suggestions according to the user's level of understanding and emotional state, thereby improving learning effectiveness and customer satisfaction.

[0532] "Generative AI" is artificial intelligence that automatically generates appropriate feedback and suggestions based on user input and the situation.

[0533] "Feedback" refers to evaluations and advice provided based on a user's behavior and understanding.

[0534] "Supplementary information" is additional information or explanation provided to enhance the user's understanding.

[0535] "Exercises" are problems that learners work on to acquire specific knowledge or skills.

[0536] "Solution support" refers to hints and explanations provided to users when solving problems.

[0537] The "emotion engine" is a technology that analyzes a user's facial expressions and behavior to detect their emotional state.

[0538] "Suggestions" are recommendations of products or services provided based on the user's needs and circumstances.

[0539] "User" means an individual or customer who uses the system.

[0540] "Understanding" refers to the degree to which a user has understood a particular piece of knowledge or skill.

[0541] "Frustration" refers to the dissatisfaction and stress felt by users.

[0542] "Needs" refers to the demands and requirements of users.

[0543] A system for implementing the present invention includes a generative AI, an emotion engine, smart glasses, a camera, and a server. A specific embodiment of this system will be described below.

[0544] The server includes means for using generative AI to provide feedback and supplementary information according to the progress and level of understanding of the learning, means for automatically generating practice problems suited to each individual, means for providing answer support, means for using an emotion engine to detect the user's emotional state and take appropriate action, and means for automatically generating and presenting suggestions according to the user's needs.

[0545] The smart glasses have a built-in camera that captures the user's facial expressions and behavior in real time. The video data acquired from the camera is sent to a server and analyzed by an emotion engine. The emotion engine analyzes the user's facial expressions and behavior to detect their emotional state. For example, if the user is feeling frustrated, the emotion engine sends that information to the generative AI.

[0546] The generative AI automatically generates suggestions based on the user's needs based on the information received from the emotion engine. For example, if the user is feeling frustrated, the generative AI inputs a prompt such as, "The customer is feeling frustrated. Please make an appropriate product suggestion." This suggestion is then displayed on the smart glasses' display and presented to the user.

[0547] As a concrete example, the following prompt sentence can be input to a generative AI model:

[0548] "The customer is frustrated. Please provide appropriate product suggestions."

[0549] This system enables appropriate feedback and suggestions based on the user's level of understanding and emotional state, improving learning effectiveness and customer satisfaction.

[0550] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0551] Step 1:

[0552] The user wears the smart glasses, and the camera captures the user's facial expressions and actions in real time.

[0553] Input: User's facial expressions and actions

[0554] Output: Captured video data

[0555] Specific operation: The camera in the smart glasses detects the user's face and captures video data.

[0556] Step 2:

[0557] The terminal transmits the captured video data to the server.

[0558] Input: Captured video data

[0559] Output: Video data sent to the server

[0560] Specific operation: The smart glasses transmit video data to the server via Wi-Fi (registered trademark) or Bluetooth (registered trademark).

[0561] Step 3:

[0562] The video data received by the server is analyzed using an emotion engine to detect the user's emotional state.

[0563] Input: Video data sent to the server

[0564] Output: User's emotional state (e.g., frustration)

[0565] Specific operation: The server inputs video data into the emotion engine and detects the emotional state using a facial expression analysis algorithm.

[0566] Step 4:

[0567] The server sends the emotional state obtained from the emotion engine to the generative AI.

[0568] Input: User's emotional state

[0569] Output: Emotional state sent to the generative AI

[0570] Specific operation: The server passes the output of the emotion engine to the generative AI.

[0571] Step 5:

[0572] Generative AI automatically generates appropriate suggestions based on your emotional state.

[0573] Input: User's emotional state

[0574] Output: Generated suggestions (e.g. product suggestions)

[0575] Specific operation: The generative AI inputs the prompt sentence, "The customer is frustrated. Please provide an appropriate product suggestion." and generates an appropriate suggestion.

[0576] Step 6:

[0577] The server sends the generated proposal to the smart glasses.

[0578] Input: Generated proposals

[0579] Output: Suggestions sent to the smart glasses

[0580] Specific operation: The server sends the output of the generative AI to the smart glasses.

[0581] Step 7:

[0582] The smart glasses display the suggestions to the user.

[0583] Input: Suggestion sent to smart glasses

[0584] Output: The proposal displayed to the user

[0585] Specific operation: The smart glasses display shows the suggestion and presents it to the user.

[0586] Example 3

[0587] Next, a description will be given of a third embodiment of the third embodiment. 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."

[0588] Conventional learning support systems have difficulty providing optimal feedback and supplementary information according to each learner's individual level of understanding and learning style, making it impossible to maximize learning effectiveness. They also do not adequately provide appropriate questions to maintain learners' motivation. Furthermore, learning support does not take into account the learner's emotional state, which can lead to a decline in the quality of learning.

[0589] 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.

[0590] In this invention, the server includes means for providing feedback and supplementary information according to the progress and level of understanding of the learner using generative AI, means for automatically generating practice problems suited to each learner, means for providing answer support, means for acquiring the learner's learning history and progress data, means for analyzing the learner's level of understanding and learning style, means for monitoring the learner's facial expressions and behavior and detecting their excitement level, and means for generating new questions based on the detected excitement level. This allows the provision of learning support optimized for each learner, maximizing learning effectiveness and maintaining motivation to learn.

[0591] "Generative AI" is an artificial intelligence technology that provides optimal feedback and supplementary information based on the learner's progress and level of understanding, and automatically generates practice problems that meet individual needs.

[0592] "Feedback" refers to evaluations and advice provided to learners regarding their learning activities, and is information used to improve learning outcomes.

[0593] "Supplementary information" refers to additional knowledge or explanation provided to deepen a learner's understanding.

[0594] "Practice problems" are problems that students should solve in order to gain a practical understanding of the learning content.

[0595] "Solution support" refers to hints and explanations provided to learners when solving practice problems.

[0596] A "learning history" is a record of the learning activities that a learner has undertaken up to now.

[0597] "Progress data" refers to data that indicates how far a learner has progressed in their studies.

[0598] "Level of understanding" is an indicator that shows how well a learner understands a particular learning content.

[0599] "Learning style" refers to the characteristics of a learner that indicate the method and environment in which they learn most effectively.

[0600] "Facial expressions" refer to the learner's facial expressions, which provide clues to their emotional state.

[0601] "Behavior" refers to the actions and reactions that learners make while learning.

[0602] "Excitement" refers to the joy and sense of accomplishment that learners feel when they solve a problem.

[0603] This invention is a system that uses generative AI to provide feedback and supplementary information according to learning progress and level of understanding, automatically generates practice problems suited to each individual, and provides answer support. Furthermore, by acquiring the learner's learning history and progress data and analyzing their level of understanding and learning style, it provides learning support optimized for each learner. It also generates new problems to maintain motivation by monitoring the learner's facial expressions and behavior and detecting their excitement.

[0604] Hardware and software used

[0605] Hardware: Servers, devices (PCs, tablets, smartphones)

[0606] Software: Generative AI models (e.g., GPT-4), emotion engines, database management systems (e.g., MySQL)

[0607] Program processing

[0608] 1. The user logs in to the learning platform. The user accesses the learning platform using a device (PC, tablet, smartphone) and enters their login information.

[0609] 2. The server retrieves the user's learning history and progress data from the database. The server connects to a database management system (e.g., MySQL) and queries the user's learning history and progress data.

[0610] 3. The server sends learning history and progress data to the generative AI model. The server then calls an API to send the acquired data to the generative AI model.

[0611] 4. The generative AI model analyzes the user's level of understanding and learning style. The generative AI model (e.g., GPT-4) analyzes the received data and evaluates the user's level of understanding and learning style.

[0612] 5. The generative AI model generates optimized learning content based on the analysis results. The generative AI model generates learning content (e.g., new questions and explanations) that is optimal for the user.

[0613] 6. The server sends the generated learning content to the device. The server sends the learning content received from the generative AI model to the user's device.

[0614] 7. The device displays the learning content to the user. The device displays the received learning content to the user.

[0615] 8. The user uses the learning content to solve the problem. The user uses the displayed learning content to solve the problem.

[0616] 9. The emotion engine monitors the user's facial expressions and behavior to detect excitement. The emotion engine analyzes the user's facial expressions and tone of voice via the user's camera and microphone to detect excitement.

[0617] 10. The emotion engine sends the excited state information detected to the generative AI model. The emotion engine sends the excited state data detected to the generative AI model.

[0618] 11. The generative AI model generates new questions based on the excitement state information. The generative AI model generates new questions based on the excitement state information to maintain the user's motivation.

[0619] 12. The server sends the new problem to the device. The server sends the new problem received from the generative AI model to the user's device.

[0620] 13. The device presents a new problem to the user. The device displays a new problem to the user and prompts them to take the next learning step.

[0621] Specific examples

[0622] For example, suppose a user is solving a math problem. When the user solves the problem, the emotion engine detects the user's joy. Based on this information, the generative AI model generates the next problem to be solved and displays it on the device.

[0623] Prompt Sentence Examples

[0624] "If you detect the joy the user feels after solving a math problem, generate the next problem for them to solve."

[0625] This system provides learning support that is optimized for each learner, maximizing the learning effect. The flow of the specific processing in the third embodiment will be described with reference to FIG.

[0626] Step 1:

[0627] A user logs into the learning platform.

[0628] Input: User login information (user ID, password)

[0629] Specific operation: The user accesses the learning platform using a device (PC, tablet, smartphone) and enters login information.

[0630] Output: Login success message, user session information

[0631] Step 2:

[0632] The server retrieves the user's learning history and progress data from the database.

[0633] Input: User session information

[0634] Specific operation: The server connects to a database management system (e.g., MySQL) and queries the user's learning history and progress data.

[0635] Output: User learning history data, progress data

[0636] Step 3:

[0637] The server sends learning history and progress data to the generated AI model.

[0638] Input: User learning history data, progress data

[0639] Specific operation: The server calls an API to send the acquired data to the generative AI model.

[0640] Output: Message that data has been sent to the generative AI model

[0641] Step 4:

[0642] The generative AI model analyzes the user's level of understanding and learning style.

[0643] Input: User learning history data, progress data

[0644] What it does: A generative AI model (e.g., GPT-4) analyzes the data it receives and evaluates the user's level of understanding and learning style.

[0645] Output: User comprehension assessment results, learning style analysis results

[0646] Step 5:

[0647] The generative AI model generates optimized learning content based on the analysis results.

[0648] Input: User comprehension assessment results, learning style analysis results

[0649] Specific operation: The generative AI model generates optimal learning content (e.g., new questions and explanations) for the user.

[0650] Output: Optimized learning content

[0651] Step 6:

[0652] The server transmits the generated learning content to the terminal.

[0653] Input: Optimized learning content

[0654] Specific operation: The server sends the learning content received from the generative AI model to the user's device.

[0655] Output: Learning content submission completion message

[0656] Step 7:

[0657] The terminal displays the learning content to the user.

[0658] Input: Optimized learning content

[0659] Specific operation: The device displays the received learning content to the user.

[0660] Output: User views learning content

[0661] Step 8:

[0662] The user uses the learning content to solve the problem.

[0663] Input: Optimized learning content

[0664] Specific Action: The user solves a problem using the displayed learning content.

[0665] Output: Problem-solving results

[0666] Step 9:

[0667] The emotion engine monitors the user's facial expressions and behavior to detect excitement.

[0668] Input: User facial expression data, behavior data

[0669] Specific operation: The emotion engine analyzes the user's facial expressions and tone of voice via the camera and microphone to detect excitement.

[0670] Output: Excitation state detection result

[0671] Step 10:

[0672] The emotion engine sends the detected excitement information to the generative AI model.

[0673] Input: Excitation state detection result

[0674] Specific operation: The emotion engine sends the excited state data detected to the generative AI model.

[0675] Output: Message that data has been sent to the generative AI model

[0676] Step 11:

[0677] The generative AI model generates new problems based on the excited state information.

[0678] Input: Excitation state detection result

[0679] Specific operation: The generative AI model generates new problems based on the excitement state information to maintain the user's motivation.

[0680] Output: New problem

[0681] Step 12:

[0682] The server sends a new question to the device.

[0683] Input: New problem

[0684] Specific operation: The server sends the new problem received from the generative AI model to the user's device.

[0685] Output: New issue submission successful message

[0686] Step 13:

[0687] The terminal presents the user with a new problem.

[0688] Input: New problem

[0689] Specific operation: The device displays a new problem to the user and prompts them for the next learning step.

[0690] Output: User views new question

[0691] (Application example 3)

[0692] 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."

[0693] Conventional learning support systems are insufficient in providing feedback and supplementary information according to the learner's progress and level of understanding, and the automatic generation of practice problems and answer support tailored to individual needs are also limited. Furthermore, when it comes to learning support for workers in factories, there is a lack of optimized training programs based on work history and progress, and a lack of means to maintain motivation that takes emotional state into account. This makes it difficult to maximize learning effectiveness and work efficiency.

[0694] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[0695] In this invention, the server includes means for using generative AI to provide feedback and supplementary information according to the learning progress and level of understanding, means for automatically generating practice problems suited to each individual, means for providing answer support, means for recording the worker's work history and progress and storing it in a database, means for providing the worker with an optimized training program based on the stored data, and means for detecting the worker's emotional state using an emotion engine and adjusting the next task content based on that information. This makes it possible to provide learning support and training programs optimized for each learner and worker, maximizing learning effectiveness and work efficiency.

[0696] "Generative AI" is an artificial intelligence technology that provides optimal feedback and supplementary information based on the progress and level of understanding of learners and workers, and automatically generates exercises and training programs that meet individual needs.

[0697] "Feedback" refers to information and advice provided to learners and workers based on their progress and level of understanding, with the aim of improving learning effectiveness and work efficiency.

[0698] "Supplementary information" refers to additional information or materials provided to enhance a learner's or worker's understanding.

[0699] "Practice problems" are problems that learners must solve to acquire specific knowledge or skills, and are automatically generated by generative AI.

[0700] "Solution support" refers to hints and explanations provided to learners when they are solving practice problems, and is intended to enhance learning effectiveness.

[0701] "Work history" refers to a record of work that a worker has done in the past, and is stored in a database.

[0702] "Progress" is a measure of how far a learner or worker has progressed toward a particular task or goal.

[0703] A "database" is a system for efficiently storing, managing, and searching collected data.

[0704] "Training Program" means a series of learning activities or exercises designed to enable a worker to acquire specific skills or knowledge.

[0705] An "emotion engine" is a technology that detects the emotional state of learners and workers and uses that information to provide appropriate feedback and adjust the next task.

[0706] An "emotional state" is a psychological state such as joy, excitement, or stress that a learner or worker feels in a particular situation.

[0707] A system for implementing this invention has the following configuration. First, the server uses generative AI to provide feedback and supplementary information according to the learning progress and level of understanding. Specifically, the server collects progress data of learners and workers and stores it in a database. Next, the generative AI generates a training program optimized for the learner or worker based on the stored data.

[0708] The server uses the following hardware and software:

[0709] Hardware:

[0710] Factory robots (e.g. general-purpose robotic arms)

[0711] Emotion detection sensors (e.g. emotion recognition sensors)

[0712] Database server (e.g. MySQL server)

[0713] software:

[0714] Generative AI models (e.g., GPT-4)

[0715] Emotion engine (e.g. emotion recognition SDK)

[0716] Data analysis tools (e.g., Python's Pandas library)

[0717] The server first uses the factory robot to collect the worker's work history and progress in real time and stores it in a MySQL database. It then uses Python's Pandas library to analyze the collected data and evaluate the worker's level of understanding and work style. It then uses a generative AI model (GPT-4) to generate a training program optimized for the worker and provides it to the worker through the factory robot's interface.

[0718] Furthermore, an emotion engine (emotion recognition SDK) is used to detect the emotional state of the worker in real time. It detects the joy of the worker when they have a successful experience and adjusts the next task based on that information. This helps maintain the worker's motivation and maximizes learning effects and work efficiency.

[0719] As a concrete example, if Worker A is learning a new work procedure and has difficulty with a particular step, the server analyzes Worker A's progress data and uses a generative AI model to generate training content specific to the step he or she is having difficulty with. The emotion engine detects the joy Worker A feels when he or she successfully completes that step and provides the next step of the same difficulty level.

[0720] An example of a prompt sentence is as follows:

[0721] Generate training content tailored to specific procedures based on the progress data of Worker A. Also consider the emotional data of Worker A when he had a successful experience, and adjust the difficulty of the next procedure to be provided.

[0722] In this way, the server provides learning support and training programs optimized for each learner and worker, maximizing learning effectiveness and work efficiency.

[0723] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0724] Step 1:

[0725] The server uses factory robots to collect workers' work history and progress data in real time.

[0726] Input: Work history and progress data of workers

[0727] Output: Collected work history and progress data

[0728] Specific operation: Using sensors and cameras installed on factory robots, workers' movements and work content are recorded and stored in a database.

[0729] Step 2:

[0730] The server stores the collected work history and progress data in a database.

[0731] Input: Collected work history and progress data

[0732] Output: Data stored in the database

[0733] What it does: Connects to a MySQL database and inserts collected data into the appropriate tables.

[0734] Step 3:

[0735] The server uses Python's Pandas library to analyze the accumulated data and evaluate the worker's level of understanding and working style.

[0736] Input: Work history and progress data stored in the database

[0737] Output: Evaluation results of worker's understanding and work style

[0738] Specific operations: Use the Pandas library to read data and apply statistical methods and machine learning algorithms to evaluate worker performance.

[0739] Step 4:

[0740] The server uses a generative AI model (GPT-4) to generate a training program optimized for the worker.

[0741] Input: Evaluation results of worker's understanding and work style

[0742] Output: Optimized training program

[0743] Specific operation: The evaluation results are input as prompts into GPT-4 to obtain the generated training program.

[0744] Step 5:

[0745] The server provides the generated training program to the worker through the interface of the factory robot.

[0746] Input: Optimized training program

[0747] Output: Training program provided to the worker

[0748] Specific operation: The contents of the training program are presented to workers using the factory robot's display and audio output.

[0749] Step 6:

[0750] The server uses an emotion engine (emotion recognition SDK) to detect the worker's emotional state in real time.

[0751] Input: Real-time video and audio data of the worker

[0752] Output: Emotional state data of the worker

[0753] Specific operation: Using the emotion recognition SDK, the worker's facial expressions and tone of voice are analyzed to detect their emotional state.

[0754] Step 7:

[0755] The server adjusts the next task based on the emotional state data.

[0756] Input: Worker emotional state data

[0757] Output: Adjusted next steps

[0758] Specific operation: Analyzes emotional state data, detects the joy felt when a worker has a successful experience, and adjusts the difficulty of the next task provided.

[0759] In this way, the server provides learning support and training programs optimized for each learner and worker, maximizing learning effectiveness and work efficiency.

[0760] 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.

[0761] 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.

[0762] Another example of generative AI is Gemini (registered trademark) (Internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[0763] 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.

[0764] [Second embodiment]

[0765] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0766] 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.

[0767] 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).

[0768] 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.

[0769] 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.

[0770] 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).

[0771] 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.

[0772] 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.

[0773] 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.

[0774] 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.

[0775] 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.

[0776] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0777] "Example 1"

[0778] The present invention is an educational platform that uses generative AI. Specifically, it grasps the learner's learning progress and level of understanding and provides feedback and supplementary information accordingly. This feedback and supplementary information is provided in the form of text, audio, video, etc., and helps deepen the learner's understanding. It also automatically generates exercises based on the learner's level of understanding, interests, and learning style. This provides questions that are tailored to each learner, maximizing learning effectiveness. It also provides an answer support function to support the learner as they work through the questions.

[0779] "Example 2"

[0780] As a concrete example, let's imagine a mathematics seminar class. Suppose a student is studying calculus. The generative AI grasps the student's level of understanding and automatically generates problems that correspond to the student's level of understanding, from basic concepts to applications of calculus. It also provides hints and explanations for the answers as the student works on the problems, deepening the student's understanding.

[0781] "Example 3"

[0782] Furthermore, generative AI accumulates learners' learning history and progress, and uses this information to analyze their level of understanding and learning style, thereby providing optimal learning support for each learner and maximizing learning effectiveness.

[0783] The processing flow of each embodiment will be described below.

[0784] "Example 1"

[0785] Step 1: The generative AI understands the learner's learning progress and level of understanding based on the learner's activities on the platform and test results.

[0786] Step 2: Generate and provide feedback and supplementary information based on the learner's level of understanding. This can be in the form of text, audio, or video to deepen the learner's understanding.

[0787] Step 3: Automatically generate exercises tailored to the learner's level of understanding, interests, and learning style. This provides questions that are tailored to each individual learner, maximizing learning effectiveness.

[0788] Step 4: As students work through the problems, provide them with hints and explanations to deepen their understanding.

[0789] "Example 2"

[0790] Step 1: The student begins learning calculus.

[0791] Step 2: The generative AI grasps the learner's level of understanding and automatically generates questions that correspond to the learner's level of understanding, from basic concepts of calculus to applications.

[0792] Step 3: As students work through the problems, provide them with hints and explanations to deepen their understanding.

[0793] "Example 3"

[0794] Step 1: The generative AI accumulates the learner's learning history and progress.

[0795] Step 2: The generative AI analyzes the learner's level of understanding and learning style based on the accumulated data.

[0796] Step 3: Based on the analysis results, provide learning support optimized for each learner to maximize learning effectiveness.

[0797] Example 1

[0798] 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."

[0799] With conventional educational platforms, it is difficult to provide appropriate feedback and supplementary information according to each learner's progress and level of understanding, and there is a lack of automatic generation of exercises and answer support that meet individual needs. This makes it difficult to maximize learning effectiveness, and the challenge is to provide an optimal learning experience that matches the learner's level of understanding and interests.

[0800] 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.

[0801] In this invention, the server includes means for collecting learner progress data, means for analyzing the collected data and evaluating the learner's level of understanding, means for generating feedback and supplementary information based on the evaluation results, means for automatically generating exercises according to the learner's level of understanding and interests, and means for providing support for answering the exercises. This makes it possible to provide optimal feedback and supplementary information to each learner, automatically generate exercises that meet the learner's individual needs, and provide support for answering the exercises.

[0802] "Student progress data" refers to a record of the learning activities that a learner has performed on the educational platform, and includes information such as study time, percentage of correct answers, and number of problems attempted.

[0803] A "means for collecting" is a method or device for obtaining learner progress data from a learning management system or other educational platform.

[0804] "Means for analyzing and assessing comprehension" refers to methods and devices for assessing learners' comprehension based on collected data, and involves analyzing the data using a generative AI model.

[0805] The "means for generating feedback and supplementary information" refers to a method or device for generating appropriate feedback and supplementary information based on the results of the learner's comprehension assessment.

[0806] "Means for automatically generating exercises" refers to a method or device for automatically creating exercises that correspond to the learner's level of understanding and interests.

[0807] A "means for providing solution support" is a method or device for providing hints and step-by-step explanations of solutions to learners as they work through practice problems.

[0808] This invention relates to an educational platform that uses a generative AI model. Specifically, it is a system that grasps a learner's learning progress and level of understanding and provides feedback and supplementary information accordingly. The system aims to maximize learning effectiveness by automatically generating exercises tailored to each learner and providing answer support functions.

[0809] The server collects learner progress data from learning management systems (LMS) and other educational platforms. Collected data includes study time, correct answer rate, and number of problems attempted. The server inputs this data into a generative AI model (e.g., OpenAI's GPT-4) to assess the learner's level of understanding. The generative AI model analyzes the learner's response patterns and progress to identify which concepts and skills the learner is lacking.

[0810] Based on the analysis results, the server generates appropriate feedback and supplementary information for the learner. For example, if the learner's understanding of a particular concept is lacking, the server generates detailed explanations and examples in the form of text or video. This allows the learner to supplement their missing knowledge and deepen their understanding.

[0811] Furthermore, the server automatically generates exercises tailored to the learner's level of understanding, interests, and learning style. Using a generative AI model, it creates exercises with the optimal difficulty and content for each learner. For example, it starts with basic integral problems and gradually increases the difficulty level. This allows learners to progress at their own pace.

[0812] The server provides solution support functions for students as they work on practice problems. Specifically, it generates hints and step-by-step explanations of solutions to support the problem-solving process. This helps students understand the problem-solving process more easily, improving their learning effectiveness.

[0813] As a concrete example, consider a user studying a mathematics unit on calculus. The server collects the user's progress data and inputs it into a generative AI model. The generative AI model detects that the user has insufficient understanding of a particular concept (e.g., the Fundamental Theorem of Integration). Based on this information, the server generates supplementary information about the Fundamental Theorem of Integration in the form of text and video and provides it to the user.

[0814] Furthermore, the server automatically generates exercises tailored to the user's level of understanding. For example, it starts with basic integral problems and provides problems of gradually increasing difficulty. As the user works on the problems, a solution assistance function is applied, providing hints and step-by-step explanations of the solutions.

[0815] Example prompt sentence:

[0816] Generate supplementary information about the Fundamental Theorem of Integration based on the user's learning progress. Create text and video content to explain the theory in a way that is easy for users to understand.

[0817] In this way, the server leverages generative AI models to provide each learner with an optimal learning experience.

[0818] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0819] Step 1:

[0820] The server collects learner progress data from learning management systems (LMS) and other educational platforms. Specifically, it obtains data such as study time, correct answer rate, and number of questions attempted through API. The input is the learner progress data, and the output is the collected progress data.

[0821] Step 2:

[0822] The server inputs the collected progress data into a generative AI model to evaluate the learner's level of understanding. The generative AI model analyzes the learner's response patterns and progress status to identify which concepts and skills are lacking. The input is the collected progress data, and the output is the comprehension assessment results.

[0823] Step 3:

[0824] The server generates feedback and supplementary information based on the results of comprehension assessment. For example, if a user's understanding of a particular concept is insufficient, it generates detailed explanations and examples of that concept in text or video format. The input is the comprehension assessment result, and the output is the generated feedback and supplementary information.

[0825] Step 4:

[0826] The server automatically generates exercises based on the learner's level of comprehension and interests. Using a generative AI model, it creates problems with the optimal difficulty and content for each learner. For example, it starts with basic integral problems and provides problems of gradually increasing difficulty. The input is the comprehension assessment results, and the output is the automatically generated exercises.

[0827] Step 5:

[0828] The server provides answer support functions when students work on exercises. Specifically, it generates hints and step-by-step explanations of solutions to support the process of solving the problems. The input is the automatically generated exercises, and the output is the provided answer support information.

[0829] In this way, the server leverages generative AI models to provide each learner with an optimal learning experience.

[0830] (Application example 1)

[0831] 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."

[0832] Conventional educational platforms have difficulty providing individual feedback and supplementary information based on learners' progress and level of understanding, which prevents them from fully improving learning outcomes. Furthermore, in the education and training of factory workers, there is a lack of individual progress management and feedback based on their level of understanding, making it difficult to improve work efficiency and quality. Furthermore, the automatic generation of exercises and answer support functions are inadequate, making it difficult to provide education that meets the individual needs of learners and workers.

[0833] 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.

[0834] In this invention, the server includes means for using generative AI to provide feedback and supplementary information according to the progress and level of understanding of learning, means for automatically generating practice problems suited to each worker, means for providing answer support, means for monitoring the worker's progress in real time, means for providing feedback according to the worker's level of understanding in the form of text, audio, and video, means for automatically generating practice problems based on the worker's level of understanding and interests, and means for supporting the worker when working on the problems. This enables education and training that meets the individual needs of learners and workers, improving learning effectiveness, work efficiency, and quality.

[0835] "Generative AI" is an artificial intelligence technology that generates new information and content based on data.

[0836] "Learning progress" is an indicator of how far a learner has progressed in the learning process.

[0837] "Level of understanding" is an indicator that shows how well a learner understands a particular content.

[0838] "Feedback" is evaluation or advice given to a learner regarding their behavior or performance.

[0839] "Supplementary information" is additional information provided to enhance the learner's understanding.

[0840] "Practice problems" are problems that learners solve to confirm what they have learned.

[0841] "Auto-generation" means that a system automatically creates content or information without human intervention.

[0842] "Solution support" refers to advice and hints provided to learners when solving problems.

[0843] "Workers" refers to people who perform work in factories or work sites.

[0844] "Real-time monitoring" means monitoring the ongoing situation immediately.

[0845] "Text" is information written in text.

[0846] "Sound" is audible information.

[0847] "Video" is a moving image that is visually displayed.

[0848] "Interests" are areas or content in which a learner or worker is particularly interested.

[0849] This invention is a system that applies an educational platform using generative AI to the education and training of factory workers. A specific embodiment of this system is shown below.

[0850] System configuration

[0851] The system consists of the following major components:

[0852] 1. Server: Runs the generative AI and provides feedback and supplementary information based on learning progress and level of understanding.

[0853] 2. Devices: Tablets or smartphones used by workers, allowing them to receive real-time feedback and practice questions.

[0854] 3. Network: Connects the server and the terminal to send and receive data.

[0855] Program processing

[0856] The server processes the data in the following steps:

[0857] 1. Collecting progress data: Collect progress data from the workers' devices, including the progress and understanding of the work.

[0858] 2. Feedback generation: Based on the progress data collected, generative AI is used to generate appropriate feedback, which can be in the form of text, audio, or video.

[0859] 3. Automatic generation of exercises: Generative AI is used to automatically generate exercises based on the worker's level of understanding and interests.

[0860] 4. Solution support: When workers tackle practice problems, generative AI is used to provide solution support.

[0861] Hardware and software used

[0862] Hardware: Factory robots, tablets and smartphones used by workers

[0863] Software: Python, OpenAI API

[0864] Specific examples

[0865] Progress data collection

[0866] The server collects the following progress data from the worker's terminal:

[0867] Worker ID: 12345

[0868] Learning progress: 50%

[0869] Comprehension: Medium

[0870] Interests: Machine operation

[0871] Generate feedback

[0872] The server generates the following feedback based on the progress data collected:

[0873] "You're making good progress. Now let's review the basics of machine operation."

[0874] Automatic generation of exercises

[0875] The server generates the following exercises based on the worker's level of understanding:

[0876] "Please answer the following questions regarding the basics of machine operation: 1. Explain the procedure for starting a machine."

[0877] Prompt Sentence Examples

[0878] Below are some examples of prompts used with generative AI:

[0879] Feedback generation prompt: "Generate appropriate feedback based on the following worker progress data: Worker ID: 12345, Learning Progress: 50%, Understanding: Medium, Interest: Machine Operation."

[0880] Exercise generation prompt: "Generate exercises suitable for workers with a medium level of understanding."

[0881] In this way, the server can use generative AI to support worker education and training, improving learning effectiveness, work efficiency, and quality.

[0882] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0883] Step 1:

[0884] The server collects progress data from the worker's device. Specifically, it acquires data such as the worker's ID, learning progress, level of understanding, and interest. The input is the progress data sent from the worker's device, and the output is the progress data saved on the server.

[0885] Step 2:

[0886] The server sends prompts to the generative AI based on the collected progress data and generates feedback. Specifically, it converts the progress data into prompts in text format and sends them to the generative AI. The inputs are the progress data and prompts, and the output is the generated feedback.

[0887] Step 3:

[0888] The server sends the generated feedback to the worker's terminal. Specifically, the feedback is delivered to the worker's terminal in the form of text, audio, or video. The input is the generated feedback, and the output is the feedback displayed on the worker's terminal.

[0889] Step 4:

[0890] The server automatically generates exercises based on the worker's level of understanding and interest. Specifically, it generates prompts based on the data on level of understanding and interest and sends them to the generative AI. The input is the data on level of understanding and interest and the prompts, and the output is the generated exercises.

[0891] Step 5:

[0892] The server sends the generated exercises to the worker's terminal. Specifically, the exercises are delivered to the worker's terminal in text format. The input is the generated exercise, and the output is the exercise displayed on the worker's terminal.

[0893] Step 6:

[0894] The server provides answer support when workers tackle practice problems. Specifically, it collects the workers' answer data and sends prompts to the generative AI to generate answer support. The input is the worker's answer data and prompts, and the output is the generated answer support.

[0895] Step 7:

[0896] The server sends the generated answer support to the worker's terminal. Specifically, the answer support is delivered to the worker's terminal in the form of text, audio, and video. The input is the generated answer support, and the output is the answer support displayed on the worker's terminal.

[0897] Example 2

[0898] 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."

[0899] Conventional learning support systems have had issues with the difficulty of providing appropriate questions according to the learner's level of understanding, and the lack of feedback on answers and supplementary information limits the learning effect. In particular, there is a need to automatically generate questions that meet individual learning needs and improve the quality of answer support.

[0900] 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.

[0901] In this invention, the server includes means for collecting data such as the learner's past answer history and answer time, and evaluating the learner's level of understanding using a generative AI model, means for inputting prompts into the generative AI model to automatically generate questions according to the learner's level of understanding, means for displaying questions to the learner and providing an interface for inputting answers, and means for evaluating the learner's answers and generating and providing answer hints and explanations as necessary. This makes it possible to provide appropriate questions according to the learner's level of understanding and provide high-quality answer support.

[0902] "Learner" refers to an individual receiving education or training.

[0903] "Answer history" refers to a record of questions that a learner has answered in the past and the results of those answers.

[0904] "Response time" refers to the time it takes a learner to complete a response to a particular question.

[0905] A "generative AI model" refers to an algorithm or system that uses artificial intelligence techniques to analyze data and perform a specific task (e.g., automatically generating questions or evaluating answers).

[0906] "Understanding" refers to an indicator of how well a learner understands specific knowledge or skills.

[0907] A "prompt" is an instruction given to a generative AI model to perform a specific task.

[0908] "Interface" refers to the means or screen through which a learner interacts with a system.

[0909] "Hints" refer to additional information or advice that can be helpful to learners when solving a problem.

[0910] "Explanation" refers to information that provides detailed explanations of how to answer a question and related concepts.

[0911] The present invention is a system for automatically generating questions according to a learner's level of understanding and providing support for answering the questions. A specific embodiment of this system will be described below.

[0912] First, the server collects data such as the learner's past answer history and answer time. This data is stored in a database and managed for each learner. The server then inputs the collected data into a generative AI model (e.g., a general natural language processing model) to evaluate the learner's level of understanding.

[0913] Next, the server inputs prompt sentences into the generative AI model to automatically generate questions according to the learner's level of understanding. For example, the following prompt sentences can be used:

[0914] To assess students' understanding of the basic concept of definite integrals, generate a problem like this: 'Find the definite integral of the following function: ∫(2x + 3)dx, from x = 0 to x = 2.'

[0915] The generated questions are sent from the server to the terminal. The terminal displays the questions to the learner and provides an interface for inputting answers. The learner inputs their answer to the displayed question. For example, if the learner inputs the answer "7", the terminal sends the answer to the server.

[0916] The server receives the learner's answer and uses the generative AI model to determine whether the answer is correct. For example, the generative AI model receives a prompt such as, "Please evaluate whether the learner's answer '7' is correct." The generative AI model determines that the answer is correct. The server then sends this evaluation result to the device, which then displays "That's correct!" to the learner.

[0917] Furthermore, if the learner enters an incorrect answer, the server inputs a prompt to the generative AI model saying, "Please generate a hint or explanation for the solution to this problem." The generative AI model generates a hint saying, "As a hint for the solution to this problem, first integrate the function, and then explain how to apply the range of the definite integral." The server sends this hint to the device, which then displays the hint to the learner.

[0918] In this way, a system is realized in which the server, terminal, and user work together to provide questions that correspond to the learner's level of understanding and support their learning. This system allows learners to tackle questions that are appropriate for their level of understanding and receive high-quality answer support.

[0919] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0920] Step 1:

[0921] The server collects data such as the learner's past answer history and answer time.

[0922] Input: Learner's answer history data, answer time data

[0923] Data processing: Obtain each learner's answer history and answer time from the database and format them into an analyzable format.

[0924] Output: Formatted answer history data and answer time data of the learner

[0925] Step 2:

[0926] The server inputs the collected data into a generative AI model to evaluate the learner's level of understanding.

[0927] Input: Formatted answer history data of learners, answer time data

[0928] Data computation: Input data into the generative AI model and evaluate comprehension using the prompt, "Please rate this learner's comprehension."

[0929] Output: Learner comprehension assessment results

[0930] Step 3:

[0931] The server inputs prompt sentences into the generative AI model and automatically generates questions based on the learner's level of understanding.

[0932] Input: Learner comprehension assessment results, prompt text

[0933] Data calculation: The generative AI model is given a prompt, "Please generate questions that correspond to the learner's level of understanding," and an appropriate question is generated.

[0934] Output: Generated problem

[0935] Step 4:

[0936] The server sends the generated questions to the terminal.

[0937] Input: Generated question

[0938] Data processing: Convert the generated questions into a format that can be sent to the terminal.

[0939] Output: Problems sent to terminal

[0940] Step 5:

[0941] The terminal displays questions to the learner and provides an interface for entering answers.

[0942] Input: Question sent to terminal

[0943] Specific operation: The device displays the questions on the screen and provides a form where the learner can enter their answers.

[0944] Output: The answer entered by the learner

[0945] Step 6:

[0946] The terminal transmits the learner's answers to the server.

[0947] Input: The answer entered by the learner

[0948] Data processing: Convert the learner's answers into a format that can be sent to the server.

[0949] Output: The learner's answer sent to the server

[0950] Step 7:

[0951] The server receives the learner's answer and uses the generative AI model to determine whether the answer is correct.

[0952] Input: Learner's answer

[0953] Data calculation: The generative AI model is given a prompt, "Please evaluate whether the learner's answer is correct," and the model determines whether the answer is correct or not.

[0954] Output: Evaluation result of the answer

[0955] Step 8:

[0956] The server transmits the evaluation result of the answer to the terminal.

[0957] Input: Answer evaluation result

[0958] Data processing: The evaluation results of the answers are converted into a format that can be sent to the terminal.

[0959] Output: Evaluation result of the answer sent to the device

[0960] Step 9:

[0961] The terminal displays the evaluation results of the answers to the learner.

[0962] Input: Evaluation result of the answer sent to the terminal

[0963] Specific operation: The device displays the evaluation results of the answers on the screen and provides feedback to the learner.

[0964] Output: The evaluation result of the answer displayed to the learner

[0965] Step 10:

[0966] If a learner enters an incorrect answer, the server inputs a prompt to the generative AI model saying, "Please generate hints and explanations for the answer to this question."

[0967] Input: Learner's incorrect answer, prompt

[0968] Data calculation: A prompt sentence is input into the generative AI model to generate hints and explanations for the answer.

[0969] Output: Generated hints and explanations

[0970] Step 11:

[0971] The server sends the generated hints and explanations to the terminal.

[0972] Input: Generated hints and explanations

[0973] Data processing: Convert the generated hints and explanations into a format that can be sent to the device.

[0974] Output: Hints and explanations sent to the terminal

[0975] Step 12:

[0976] The device displays hints and explanations to the learner.

[0977] Input: Hints and explanations sent to your device

[0978] Specific operation: The device displays hints and explanations on the screen to deepen the learner's understanding.

[0979] Output: Hints and explanations displayed to the learner

[0980] (Application example 2)

[0981] 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."

[0982] Conventional learning systems have the problem of not providing enough questions or feedback that correspond to the learner's level of understanding, limiting the learning effect. Furthermore, there is a lack of appropriate support for learners to progress through their studies at their own pace, making it difficult to improve the quality of their learning. Furthermore, in a learning environment using mobile devices such as smartphones, there is a need for a system that can generate questions and provide answer support that correspond to individual needs.

[0983] 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.

[0984] In this invention, the server includes means for providing feedback and supplementary information according to the learner's learning progress and level of understanding using generative AI, means for automatically generating practice problems suited to each individual, means for providing answer support, means for generating questions according to the learner's level of understanding and providing hints and explanations through an application installed on a smartphone, and means for inputting prompt sentences using a generative AI model and generating questions, hints, and explanations. This enables individualized support according to the learner's level of understanding, improving both the effectiveness and quality of learning.

[0985] "Generative AI" is an artificial intelligence technology that analyzes a learner's level of understanding and progress, and automatically generates appropriate feedback, supplementary information, questions, hints, and explanations based on that analysis.

[0986] "Feedback" means providing learners with evaluations and advice based on their level of understanding and progress in response to the problems and tasks they have tackled.

[0987] "Supplementary information" is information that provides additional knowledge or explanation that learners need to deepen their understanding.

[0988] "Exercises" are problems that learners work on to understand and master specific learning content.

[0989] "Automatic generation" refers to the use of artificial intelligence or algorithms to generate questions or information without manual intervention.

[0990] "Solution support" means providing hints and explanations that learners need to solve problems, helping them arrive at the answer.

[0991] A "smartphone" is a type of mobile phone, a multi-function device that can connect to the Internet and use applications.

[0992] An "application" is a software program designed to provide a particular function or service.

[0993] A "prompt" is an instruction given to a generative AI to generate a specific output.

[0994] "Problems, hints, and explanations" are practice problems that students need to solve in order to advance their studies, clues to help them find the answers, and detailed explanations of the answers to the problems.

[0995] The system for implementing this invention uses generative AI to provide feedback and supplemental information according to the progress and level of understanding of the learning, automatically generate practice problems suited to each individual, and provide answer support. A specific embodiment of this system is shown below.

[0996] The server uses a generative AI model to analyze the learner's level of understanding and provides appropriate feedback and supplementary information based on that analysis. Specifically, as the learner works on problems through an application installed on their smartphone, the server evaluates their answers in real time to determine their level of understanding. The server then automatically generates the next problem based on the learner's level of understanding and sends it to the application.

[0997] This system uses "text-davinci-003" as the generative AI model. The server receives a prompt as input and generates questions, hints, and explanations. For example, if the learner's level of comprehension is "intermediate," the following prompt is used:

[0998] Problem generation prompt: "Generate calculus problems appropriate for learners with an intermediate level of comprehension."

[0999] Hint prompt: "Please provide a hint for the following problem: {generated problem}"

[1000] Explanation prompt: "Please provide an explanation for the following problem: {generated problem}"

[1001] The server inputs these prompts into a generative AI model and sends the resulting output to a smartphone application, which displays the generated questions, hints, and explanations to the learner to support their learning.

[1002] As a concrete example, consider the case where a learner is working on a calculus problem. When the learner answers the problem, the answer is sent to the server, where a generative AI model evaluates the answer. Based on the evaluation results, the next problem is automatically generated and provided to the learner. Hints and detailed explanations for the answer are also provided at the same time, allowing the learner to deepen their understanding at their own pace.

[1003] In this way, systems using generative AI can respond individually to learners' levels of understanding, improving both the effectiveness and quality of learning.

[1004] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1005] Step 1:

[1006] The user launches the smartphone application and begins learning. The user selects the content to study (e.g., calculus) on the application. The input is the user's selected learning content, and the output is initial setting information based on that content. The application sends the initial setting information to the server based on the user's selection.

[1007] Step 2:

[1008] The server receives the user's initial setting information and inputs a prompt statement to the generative AI model. The prompt statement is an instruction statement for generating a question according to the user's level of understanding. The input is the initial setting information and the prompt statement, and the output is the generated question. The server generates a question using the generative AI model and sends the question to the application.

[1009] Step 3:

[1010] The application displays the questions received from the server to the user. The user inputs the answers to the displayed questions. The input is the user's answer, and the output is the answer data. The application sends the user's answer data to the server.

[1011] Step 4:

[1012] The server receives the user's answer data and inputs it into the generative AI model. The generative AI model evaluates the answer data and analyzes the user's level of understanding. The input is the user's answer data, and the output is the evaluation results and comprehension information. The server creates a prompt sentence to generate the next question based on the evaluation results and comprehension information.

[1013] Step 5:

[1014] The server inputs a new prompt into the generative AI model to generate the next problem. The input is the new prompt, and the output is the next problem. The server then sends the generated next problem to the application.

[1015] Step 6:

[1016] The application displays the next question received from the server to the user. The user again inputs the answer to the displayed question. The input is the user's answer, and the output is the answer data. The application sends the user's answer data to the server.

[1017] Step 7:

[1018] The server receives the user's answer data again and inputs it into the generative AI model. The generative AI model evaluates the answer data and further analyzes the user's level of understanding. The input is the user's answer data, and the output is the evaluation results and comprehension information. The server creates prompt sentences to generate hints and explanations based on the evaluation results and comprehension information.

[1019] Step 8:

[1020] The server inputs prompt sentences to generate hints and explanations into the generative AI model, and generates the hints and explanations. The input is the prompt sentence for the hint or explanation, and the output is the generated hint or explanation. The server sends the generated hint or explanation to the application.

[1021] Step 9:

[1022] The application displays hints and explanations received from the server to the user. The user then uses the displayed hints and explanations to tackle the next problem. The input is the hints and explanations, and the output is an improvement in the user's understanding. The application reports the user's learning progress to the server.

[1023] In this way, systems using generative AI can respond individually to the user's level of understanding, improving both the effectiveness and quality of learning.

[1024] Example 3

[1025] 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."

[1026] Conventional learning support systems have had difficulty providing optimal feedback and supplementary information tailored to each learner's level of understanding and learning style. Furthermore, they lacked sufficient support for automatically generating practice questions and answering them to meet individual needs, limiting the improvement of learning effectiveness. This led to problems such as a decline in learner motivation and a decline in learning efficiency.

[1027] 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.

[1028] In this invention, the server includes means for collecting learners' learning history and progress data from a learning management system or online learning platform, means for saving the collected data in a database, means for preprocessing the saved data, means for analyzing the learners' level of understanding and learning style based on the preprocessed data, means for generating prompt sentences using a generative AI model to provide learning support optimized for each learner, and means for providing the generated prompt sentences to the learner. This makes it possible to provide optimal feedback and supplemental information according to the individual needs of the learner, automatically generate practice questions, and provide answer support.

[1029] A "learning management system" is a software platform that manages learners' learning history and progress data and supports educational activities.

[1030] An "online learning platform" is a system that provides learning content via the Internet and allows learners to progress with their studies online.

[1031] "Learning history" is a record of a learner's past learning activities, and is data including test scores, study time, and percentage of correct answers to questions.

[1032] "Progress data" is data that indicates the status of a learner's current learning activities, and is information that includes the learner's learning progress and achievement level.

[1033] A "database" is a system for efficiently storing, managing, and searching collected data.

[1034] "Preprocessing" refers to operations such as cleaning and shaping data, filling in missing values, and handling outliers that are performed before data analysis or machine learning.

[1035] "Level of understanding" is an indicator that shows how well a learner understands a particular learning content.

[1036] "Learning style" is a characteristic that indicates the method or approach a learner takes to learn.

[1037] A "generative AI model" is a model that uses artificial intelligence technology to generate text and data, and examples include GPT-4 and BERT.

[1038] A "prompt" is a piece of text generated by a generative AI model that includes feedback to the learner, supplementary information, suggested practice questions, etc.

[1039] This invention is a system that collects learners' learning history and progress data from learning management systems and online learning platforms, and provides optimized learning support for each learner based on that data. Specific embodiments of this system are described below.

[1040] The server collects learners' learning history and progress data from learning management systems and online learning platforms (e.g., Moodle, Canvas). This includes obtaining data such as test scores, study time, and percentage of questions answered correctly through APIs. For example, the server uses the Moodle API to obtain Learner A's latest test results.

[1041] Next, the server stores the collected data in a database (e.g., MySQL, PostgreSQL). The database contains a table for each learner, storing each learner's progress data. For example, it executes an SQL query to insert the test results of learner A into the "test_results" table.

[1042] The server then uses Python to preprocess the data. Specifically, it uses the Pandas library to clean and format the data, and to impute missing values ​​and handle outliers. For example, if the study time for learner A is missing, it is imputed with the average study time.

[1043] Based on the preprocessed data, the server uses machine learning libraries such as Scikit-learn and TensorFlow to analyze the learner's level of understanding and learning style. Specifically, it performs clustering and regression analysis to evaluate the learner's performance. For example, it uses K-means clustering to determine which cluster Learner A belongs to.

[1044] Next, the server uses a generative AI model (e.g., GPT-4, BERT) to generate prompts to provide learning support optimized for each learner. The generated prompts include questions and explanations based on the learner's level of understanding, as well as suggested learning methods tailored to the learner's learning style. For example, the server might generate a prompt that reads, "Student A lacks understanding of differentiation, so we recommend that he or she relearn the basic concepts of differentiation. Also, visual explanations are effective, so please provide explanations using graphs."

[1045] Finally, the terminal displays the prompt received from the server to the learner. The user (learner) proceeds with their learning based on the presented learning support. For example, learner A re-learns the basic concept of differentiation according to the presented prompt and checks the explanation using graphs.

[1046] In this way, the present invention can provide learning support that is optimized for each learner, thereby maximizing the learning effect.The flow of the identification process in the third embodiment will be described with reference to FIG.

[1047] Step 1:

[1048] The server collects learners' learning history and progress data from learning management systems and online learning platforms. Specifically, it obtains data such as test scores, study time, and percentage of correct answers via API. For example, it uses the Moodle API to obtain Learner A's latest test results.

[1049] Input: Learning history and progress data from your learning management system or online learning platform

[1050] Output: Collected learning history and progress data

[1051] Step 2:

[1052] The server saves the collected data in a database, such as MySQL or PostgreSQL. For example, it executes an SQL query to insert the test results of learner A into the "test_results" table.

[1053] Input: Collected learning history and progress data

[1054] Output: Data stored in the database

[1055] Step 3:

[1056] The server uses Python to preprocess the data. Specifically, it uses the Pandas library to clean and format the data, and to fill in missing values ​​and handle outliers. For example, if the study time for learner A is missing, it is filled in with the average study time.

[1057] Input: Data stored in a database

[1058] Output: Preprocessed data

[1059] Step 4:

[1060] The server analyzes the learner's level of understanding and learning style based on the preprocessed data. Specifically, it uses machine learning libraries such as Scikit-learn and TensorFlow to perform clustering and regression analysis to evaluate the learner's performance. For example, it uses K-means clustering to determine which cluster Learner A belongs to.

[1061] Input: Preprocessed data

[1062] Output: Analysis of learner's comprehension and learning style

[1063] Step 5:

[1064] The server uses a generative AI model to generate prompts to provide learning support optimized for each learner. Specifically, it uses generative AI models such as GPT-4 and BERT to generate prompts that include questions and explanations based on the learner's level of understanding and suggestions for learning methods tailored to the learner's learning style. For example, it generates a prompt that reads, "Student A lacks understanding of differentiation, so we recommend that he / she relearn the basic concepts of differentiation. Also, visual explanations are effective, so please provide explanations using graphs."

[1065] Input: Analysis results of learner comprehension and learning style

[1066] Output: Generated prompt statement

[1067] Step 6:

[1068] The terminal displays the prompt received from the server to the learner. The user (learner) proceeds with their learning based on the presented learning support. For example, learner A re-learns the basic concept of differentiation according to the presented prompt and checks the explanation using graphs.

[1069] Input: Generated prompt text

[1070] Output: The prompt displayed to the learner

[1071] (Application example 3)

[1072] 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."

[1073] Conventional learning support systems are limited to providing feedback and supplementary information according to the learner's progress and level of understanding, and also have limitations in automatically generating exercises and providing answer support that meet the individual needs of the learner. Furthermore, there is a lack of means to optimize the work efficiency and work style of robots working in factories, making it difficult to improve productivity throughout the factory.

[1074] 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 using generative AI to provide feedback and supplemental information according to the learning progress and level of understanding, means for automatically generating practice problems suited to each learner, means for providing answer support, means for accumulating work history data and analyzing work efficiency and work style, and means for generating optimized work instructions. This makes it possible to provide optimized learning support to each learner and maximize learning effectiveness, as well as improve the work efficiency of robots in factories and overall productivity.

[1075] "Generative AI" refers to artificial intelligence that generates new information and content based on data.

[1076] "Learning progress" is an indicator of how far a learner has progressed in the learning process.

[1077] "Level of understanding" is an indicator of how well a learner understands the learning content.

[1078] "Feedback" refers to evaluation and advice provided to learners regarding their behavior and performance.

[1079] "Supplemental information" is additional information provided to complement the learning content.

[1080] "Practice problems" are problems that learners solve to gain a practical understanding of the learning content.

[1081] "Solution support" refers to advice and hints provided to learners when solving practice problems.

[1082] "Work history data" is a record of work that robots and machines have done in the past.

[1083] "Work efficiency" is an index that indicates how much work can be completed within a certain time.

[1084] A "work style" is a specific method or pattern used by a robot or machine to perform a task.

[1085] "Optimized work instructions" are instructions that take into account work efficiency and work style to perform work most effectively.

[1086] A system for implementing this invention includes means for using generative AI to provide feedback and supplementary information according to learning progress and level of understanding, means for automatically generating practice problems suited to each individual, means for providing answer support, means for accumulating work history data and analyzing work efficiency and work style, and means for generating optimized work instructions.

[1087] The server collects learners' learning history and progress data and analyzes this data using generative AI. Specifically, it uses software such as Python, Pandas, and Scikit-learn to preprocess the data and train machine learning models. This allows it to understand the learner's level of understanding and learning style and provide optimal feedback and supplementary information.

[1088] The server also automatically generates exercises tailored to the individual needs of each learner and provides support in answering them, allowing learners to tackle exercises that are optimized for them and maximizing their learning effectiveness.

[1089] Furthermore, the server accumulates work history data of the robots working in the factory and analyzes their work efficiency and work style. Specifically, it uses machine learning algorithms such as KMeans clustering to classify the robots' work styles and generate optimized work instructions. This improves the work efficiency of the robots in the factory and increases overall productivity.

[1090] For example, based on the work history data of a factory robot over the past month, the robot's working style can be analyzed, and based on the results, optimized work instructions such as a "high-speed work mode" or a "precision work mode" can be provided.

[1091] Examples of prompts to input to a generative AI model include:

[1092] "Based on the work history data of factory robots over the past month, analyze the robot's work style and generate optimized work instructions."

[1093] In this way, the server can provide optimized learning support to each learner, maximizing the learning effect, while also improving the work efficiency of robots in the factory and increasing overall productivity.

[1094] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1095] Step 1:

[1096] The server collects learners' learning history and progress data. Specifically, it acquires log data of the learners' learning activities on the online platform and stores it in a database. The input is the learner's log data, and the output is the saved learning history data.

[1097] Step 2:

[1098] The server preprocesses the collected learning history data. Specifically, it cleans and formats the data using the Python Pandas library. The input is the saved learning history data, and the output is the preprocessed data.

[1099] Step 3:

[1100] The server trains a generative AI model based on the preprocessed data. Specifically, it uses the Scikit-learn library to build a machine learning model and analyzes the learner's level of understanding and learning style. The input is the preprocessed data, and the output is the trained generative AI model.

[1101] Step 4:

[1102] The server uses a trained generative AI model to generate optimal feedback and supplemental information for the learner. Specifically, it inputs a prompt sentence into the generative AI model and generates appropriate feedback and supplemental information. The input is the prompt sentence and the trained generative AI model, and the output is the generated feedback and supplemental information.

[1103] Step 5:

[1104] The server automatically generates exercises that meet the individual needs of each learner. Specifically, it uses a generative AI model to generate exercises that correspond to the learner's level of understanding. The input is the learner's comprehension data and the generative AI model, and the output is the generated exercises.

[1105] Step 6:

[1106] The server provides answer support for the generated exercises. Specifically, it uses a generative AI model to generate hints and advice for learners when solving the exercises. The inputs are the generated exercises and the generative AI model, and the output is answer support information.

[1107] Step 7:

[1108] The server collects work history data from the robots working in the factory. Specifically, it stores data acquired from the robots' sensors in a database. The input is the robot's sensor data, and the output is the stored work history data.

[1109] Step 8:

[1110] The server preprocesses the collected work history data. Specifically, it cleans and formats the data using the Python Pandas library. The input is the saved work history data, and the output is the preprocessed data.

[1111] Step 9:

[1112] The server trains a generative AI model based on the preprocessed data. Specifically, it uses the Scikit-learn library to perform KMeans clustering to analyze the robot's working style. The input is the preprocessed data, and the output is the trained generative AI model.

[1113] Step 10:

[1114] The server uses a trained generative AI model to generate optimized work instructions. Specifically, a prompt sentence is input into the generative AI model to generate optimal work instructions for the robot. The input is the prompt sentence and the trained generative AI model, and the output is the generated work instructions.

[1115] Step 11:

[1116] The server sends the generated work instructions to the robot. Specifically, the work instructions are sent to the robot via the network, and the robot performs the work according to the instructions. The input is the generated work instructions, and the output is the robot's work actions.

[1117] 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.

[1118] "Example 1"

[1119] One embodiment of the present invention is an educational support system that combines generative AI and an emotion engine. This system uses the emotion engine to grasp the learner's emotional state in addition to their learning progress and level of understanding. Specifically, the system estimates the learner's emotional state from their facial expressions, tone of voice, and behavioral patterns during learning. The system then provides feedback and supplemental information according to the learner's emotional state, improving the learner's learning effectiveness.

[1120] "Example 2"

[1121] As a concrete example, imagine a learner is working on a math problem and the emotion engine detects frustration. In this case, the generative AI can use that information to provide the learner with easier problems or hints on how to solve them, thereby reducing frustration and improving learning outcomes.

[1122] "Example 3"

[1123] The emotion engine can also detect a learner's excitement. For example, if it detects a learner's joy after solving a problem, the generative AI can use that information to provide problems of similar difficulty, thereby maintaining the learner's excitement and increasing their motivation to learn.

[1124] The processing flow of each embodiment will be described below.

[1125] "Example 1"

[1126] Step 1: The learner activates the learning system and engages with the learning materials.

[1127] Step 2: The emotion engine estimates the learner's emotional state from their facial expressions, tone of voice, and behavioral patterns during learning.

[1128] Step 3: The generative AI receives information from the emotion engine and provides feedback and supplemental information based on the learner's emotional state.

[1129] "Example 2"

[1130] Step 1: Learners tackle math problems.

[1131] Step 2: The emotion engine detects learner frustration.

[1132] Step 3: The generative AI uses that information to provide the learner with easier questions or hints on how to solve them.

[1133] "Example 3"

[1134] Step 1: Learners solve the problem.

[1135] Step 2: The emotion engine detects the learner's joy.

[1136] Step 3: The generative AI uses that information to provide problems of similar difficulty.

[1137] Example 1

[1138] 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."

[1139] Conventional educational support systems provide insufficient feedback and supplementary information based on learners' learning progress and level of understanding, and the automatic generation of practice problems and answer support tailored to individual needs are also limited. Furthermore, feedback does not take into account the learner's emotional state, making it difficult to maximize learning effectiveness. This makes it difficult to provide optimal learning support for each learner.

[1140] 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.

[1141] In this invention, the server includes means for grasping a learner's learning progress and level of understanding, means for analyzing the learner's emotional state, means for providing feedback and supplemental information based on the learner's learning progress, level of understanding, and emotional state using generative AI, means for automatically generating practice problems tailored to each individual, and means for providing answer support. This makes it possible to provide optimal feedback and supplemental information based on the learner's learning progress, level of understanding, and emotional state, and also realizes the automatic generation of practice problems and answer support tailored to individual needs. This makes it possible to provide optimal learning support to each learner and maximize learning effectiveness.

[1142] "Learning progress" is a measure of how far a learner has progressed through a learning activity.

[1143] "Level of understanding" is an indicator that shows how well a learner understands a particular learning content.

[1144] "Emotional state" refers to the learner's emotional and psychological state during learning, and can be estimated from facial expressions, tone of voice, behavioral patterns, etc.

[1145] "Generative AI" is a system that uses artificial intelligence technology to analyze data and automatically generate feedback, supplementary information, practice questions, and more.

[1146] "Feedback" refers to evaluations, advice, supplementary information, etc. provided to learners regarding their learning activities.

[1147] "Supplementary information" refers to additional information or materials provided to help learners deepen their understanding.

[1148] "Exercises" refer to problems and tasks that learners tackle to gain a practical understanding of the learning content.

[1149] "Solution support" refers to support such as hints and solution procedures provided to learners when they are working on practice problems.

[1150] "Learning effectiveness" is an indicator that shows the degree to which learners improve their knowledge and skills through learning activities.

[1151] "Individual needs" refers to the specific demands and requirements of each learner based on their learning style, level of understanding, interests, etc.

[1152] This invention relates to an educational support system that combines generative AI and an emotion engine. Specifically, it is a system that grasps a learner's learning progress, level of understanding, and emotional state, and provides feedback and supplementary information accordingly. The system aims to maximize learning effectiveness by automatically generating exercises that are optimal for each learner and providing answer support.

[1153] Hardware and software used

[1154] The server monitors learners' activities on the learning platform and collects data, such as the questions they answer, the learning materials they view, and the amount of time they spend studying. This allows the server to understand the learners' learning progress and level of understanding.

[1155] The device uses a camera and microphone to collect the learner's facial expressions and tone of voice, and the server sends this data to an emotion engine to analyze the learner's emotional state.

[1156] The server uses a generative AI model to generate feedback and supplemental information based on the collected learning data and the analysis of the learner's emotional state, and provides the learner with the feedback and supplemental information in the form of text, audio, video, etc.

[1157] Furthermore, the server automatically generates exercises tailored to the learner's level of understanding, interests, and learning style, providing each learner with the most appropriate questions.An answer support function is also provided to assist learners as they work through the questions.

[1158] Specific examples

[1159] For example, suppose a student is solving a math problem. The server collects the student's correct answer rate and answer time to grasp the student's learning progress. At the same time, the device uses a camera and microphone to analyze the student's facial expressions and tone of voice to estimate their emotional state.

[1160] If a learner is struggling with a problem, the server uses a generative AI model to generate a text message asking, "Can you explain how to solve this problem again?" and provides it to the learner. It also automatically generates related problems that the learner might be interested in and presents them as the next practice problem.

[1161] Prompt Sentence Examples

[1162] "When a learner is solving a math problem, what kind of feedback should be provided if their answer time is increasing and their facial expression shows confusion?"

[1163] In this way, the server helps the learner to maximize the learning effect.

[1164] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1165] Step 1:

[1166] The server monitors learners' activities on the learning platform and collects data, such as the questions they answer, the learning materials they view, and the amount of time they spend studying. This allows the server to understand the learners' learning progress and level of understanding.

[1167] Input: Learner's answer history, viewed materials, study time

[1168] Data processing: Analysis of log data

[1169] Output: Learning progress data, comprehension data

[1170] Specific operation: The server analyzes the log data of the learning platform to obtain the learner's answer history, and also records the type and time of the learning materials viewed by the learner.

[1171] Step 2:

[1172] The device uses a camera and microphone to collect the learner's facial expressions and tone of voice, and the server sends this data to an emotion engine to analyze the learner's emotional state.

[1173] Input: Learner's facial expression data, voice tone data

[1174] Data processing: Analysis using emotion engine

[1175] Output: Emotional state data

[1176] Specific operation: While the learner is solving the problem, the device captures facial expressions with a camera and records the tone of voice with a microphone. The server sends the collected data to an emotion engine to estimate the learner's emotional state (e.g., confusion, excitement, fatigue).

[1177] Step 3:

[1178] The server uses a generative AI model to generate feedback and supplemental information based on the collected learning data and the analysis of the learner's emotional state, and provides the learner with the feedback and supplemental information in the form of text, audio, video, etc.

[1179] Input: learning progress data, comprehension data, emotional state data

[1180] Data processing: Feedback generation using generative AI models

[1181] Output: Feedback, supplementary information

[1182] Specific behavior: If the learner is confused, the server generates a text message saying, "Can you explain how to solve this problem again?" It also recommends related videos that the learner may be interested in.

[1183] Step 4:

[1184] The server automatically generates exercises tailored to the learner's level of understanding, interests, and learning style, providing each learner with the most appropriate exercises.

[1185] Input: learning progress data, comprehension data, interest data

[1186] Data processing: Generative AI model for generating exercises

[1187] Output: Auto-generated exercises

[1188] Specific operation: The server analyzes the learner's past answer history and generates new questions based on their level of understanding. Using a generative AI model, it creates questions based on the learner's interests.

[1189] Step 5:

[1190] The server assists the learner in working through the problems, providing hints and step-by-step guides to the solutions.

[1191] Input: automatically generated exercises, learner's answer status

[1192] Data processing: Answer support generation using generative AI models

[1193] Output: Hints, solution guide

[1194] Specific operation: When a learner gets stuck on a problem, the server displays a hint such as "Here's the next step." It also provides a video that explains in detail the steps to solve the problem.

[1195] (Application example 1)

[1196] 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."

[1197] Conventional educational platforms provide insufficient feedback and supplementary information according to the learner's learning progress and level of understanding, and the automatic generation of practice problems and answer support tailored to individual needs are also limited. Furthermore, there is a problem in that the feedback provided does not take into account the learner's emotional state, which means that the learning effect is not maximized. Furthermore, when it comes to educational support for factory workers, there is a lack of support when learning how to operate and maintain robots.

[1198] 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.

[1199] In this invention, the server includes means for using generative AI to provide feedback and supplemental information according to the progress and level of understanding of the learner, means for automatically generating practice problems suited to each individual, means for providing answer support, means for grasping the learner's emotional state using an emotion engine and providing feedback and supplemental information according to that, and means for providing educational support for employees learning how to operate and maintain robots in factories. This not only maximizes the learner's learning effect and enables the provision of practice problems and answer support tailored to individual needs, but also makes it possible to maintain the learner's motivation by providing appropriate feedback according to the learner's emotional state, thereby improving the quality of employee training in factories.

[1200] "Generative AI" is an artificial intelligence technology that generates new information and content based on data.

[1201] "Learning progress" is an indicator of how far a learner has progressed in the learning process.

[1202] "Level of understanding" is an index that indicates how well a learner understands the learning content.

[1203] "Feedback" refers to evaluations and advice provided to learners based on their learning progress and level of understanding.

[1204] "Supplementary information" is additional information provided to help learners gain a deeper understanding of the content they are learning.

[1205] "Practice problems" are problems that learners solve to gain a practical understanding of the learning content.

[1206] "Auto-generation" refers to the use of artificial intelligence and algorithms to create new information and content without human intervention.

[1207] "Solution support" refers to advice and hints provided to learners when solving practice problems.

[1208] The "emotion engine" is a technology that analyzes the learner's emotional state and provides appropriate feedback and supplementary information based on that.

[1209] "Emotional state" refers to a state that indicates a learner's current feelings or mood.

[1210] "Robot operation" refers to the control and operation of robots used in factories.

[1211] "Maintenance" refers to the maintenance work carried out to keep machines and equipment operating normally.

[1212] "Employee education support" refers to education and training provided to employees working in factories to help them acquire the necessary knowledge and skills.

[1213] As an embodiment of the present invention, a smart factory training support system will be described as an example. This system combines generative AI and an emotion engine to provide training support for employees working in a factory.

[1214] The server includes a means for providing feedback and supplemental information based on learning progress and understanding using generative AI. Specifically, the server collects employee learning data and evaluates progress using a generative AI model. Based on the evaluation results, the server provides appropriate feedback and supplemental information.

[1215] The server also includes a means for automatically generating exercises tailored to each individual. Using a generative AI model, the server automatically generates and provides exercises tailored to each employee's learning content and progress. This allows employees to tackle problems tailored to their individual needs.

[1216] Furthermore, the server includes a means for providing solution support. When employees work on practice problems, the server uses a generative AI model to provide solution support. Specifically, the server provides hints and directions for solving problems.

[1217] The system also includes a means for grasping the learner's emotional state using an emotion engine and providing appropriate feedback and supplementary information. The server uses a camera to capture the employee's facial expressions and analyzes their emotional state using EmotionRecognizer. Based on the analysis results, the system provides appropriate feedback and supplementary information.

[1218] This also includes a means of providing educational support to employees learning how to operate and maintain the robots that work in factories. The server provides the necessary information and practice questions when employees learn how to operate and maintain the robots, maximizing the effectiveness of their learning.

[1219] For example, if an employee is learning how to operate a robot and their progress is slow or their emotional state is estimated to be "frustrated," the server will provide feedback such as, "Your progress is slow. Please check the additional information below," or "Take a short break to refresh yourself."

[1220] An example of a prompt is as follows:

[1221] "Enter employee learning progress data and generate feedback if progress is lagging."

[1222] In this way, the smart factory training support system can maximize employee learning effectiveness and improve efficiency within the factory.

[1223] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1224] Step 1:

[1225] The server collects employee learning data. Specifically, it obtains data on progress and understanding that employees enter into the learning platform. The input data includes study time, the percentage of correct answers to questions, and self-evaluation of the learning content. Based on this data, the server prepares basic data for evaluating learning progress.

[1226] Step 2:

[1227] The server uses a generative AI model to evaluate learning progress. Using collected learning data as input, the generative AI model analyzes progress and outputs evaluation results. Specifically, it quantifies the degree of learning progress and level of understanding and saves them as evaluation results. These evaluation results are used to provide feedback in the next step.

[1228] Step 3:

[1229] The server generates feedback and supplemental information based on the evaluation results. Using a generative AI model, it creates feedback based on the employee's progress and level of understanding. For example, if progress is behind, it generates feedback such as "Your progress is behind. Please check the supplemental information below." The generated feedback is sent to the employee's device.

[1230] Step 4:

[1231] The server automatically generates exercises suited to each individual. Using a generative AI model, exercises are generated based on the employee's learning content and progress. The employee's learning history and evaluation results are used as input data. The generated exercises are sent to the employee's device, where they can then work on them.

[1232] Step 5:

[1233] The server provides answer support. When employees work on practice problems, it uses a generative AI model to provide answer support. Specifically, it generates hints for the problems and directions for the solutions, and sends them to the employee's device. This allows the employee to receive help in solving the problems.

[1234] Step 6:

[1235] The server uses an emotion engine to understand the employee's emotional state. It uses a camera to capture the employee's facial expression and analyzes the emotional state using EmotionRecognizer. Image data of the facial expression is used as input data. The analysis results are output as the employee's emotional state.

[1236] Step 7:

[1237] The server provides feedback and supplementary information according to the employee's emotional state. Based on the analysis results of the emotion engine, the server generates appropriate feedback and supplementary information. For example, if the employee is estimated to be "frustrated," the server generates feedback such as "Take a short break to refresh yourself." The generated feedback is sent to the employee's device.

[1238] Step 8:

[1239] The server provides educational support to employees who are learning how to operate and maintain the robots that work in the factory. It provides the information and practice questions that employees need when learning how to operate and maintain the robots. Employee learning history and evaluation results are used as input data. The provided information and practice questions are sent to the employee's device, allowing the employee to continue their learning.

[1240] Example 2

[1241] 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."

[1242] Conventional learning support systems have had the problem of being unable to sufficiently improve learning effectiveness because it is difficult to provide appropriate feedback and questions based on the learner's level of understanding and emotional state. In addition, they do not adequately respond when the learner becomes frustrated, which can lead to a decrease in motivation to learn.

[1243] 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.

[1244] In this invention, the server includes means for collecting learner data, means for evaluating the learner's level of understanding based on the collected data, means for automatically generating exercises according to the learner's level of understanding using a generative AI model, means for presenting the generated exercises to the learner, means for recording and evaluating the learner's answers, means for providing hints and explanations based on the results of the answer evaluation, and means for detecting the learner's emotions and taking measures to reduce frustration. This makes it possible to provide appropriate feedback and questions according to the learner's level of understanding and emotional state, thereby improving learning effectiveness and maintaining motivation to learn.

[1245] "Means for collecting learner data" refers to a device or program for collecting questions that learners have answered in the past, their answer data, answer time, answer process, etc.

[1246] A "means for assessing a learner's level of understanding" is a device or program that analyzes and scores a learner's level of understanding based on collected data.

[1247] "Means for automatically generating exercises appropriate to a learner's level of understanding using a generative AI model" refers to a device or program that uses a generative AI model to automatically create exercises appropriate to a learner's level of understanding.

[1248] The "means for presenting the generated exercises to the learner" is a device or program for displaying the generated exercises to the learner.

[1249] "Means for recording and evaluating learner's answers" refers to a device or program for recording the answers given by learners and determining whether they are correct or incorrect.

[1250] The "means for providing hints and explanations based on the evaluation results of the answers" refers to a device or program for generating and providing appropriate hints and explanations to the learner based on the evaluation results of the answers.

[1251] "Means for detecting learners' emotions and taking measures to reduce frustration" refers to a device or program that detects emotions from a learner's facial expressions and voice, and takes appropriate measures if the learner is feeling frustrated.

[1252] The present invention is a system for automatically generating exercises according to a learner's level of understanding and providing support for solving the exercises. A specific embodiment of this system will be described below.

[1253] The server uses a database to collect learner data. Specifically, it uses a database management system such as MySQL or PostgreSQL to store questions that learners have previously answered, their answer data, answer times, answering processes, and so on.

[1254] The server then evaluates the learner's level of understanding based on the collected data. It preprocesses the data using Python's pandas library and models the learner's level of understanding using a machine learning library such as scikit-learn. For example, it can score the learner's level of understanding based on their past accuracy rate and answer time.

[1255] The server uses a generative AI model (e.g., GPT-4) to automatically generate exercises appropriate for the learner's level of comprehension. Specifically, the server incorporates the comprehension score into a prompt, and inputs a prompt such as "Please generate calculus problems suitable for a learner with a comprehension score of 70" into the generative AI model.

[1256] The device presents the questions received from the server to the learner. Specifically, the questions are displayed on the screen through a web application (for example, a front end using React or Vue.js).

[1257] The user (learner) answers the questions presented to them. The device records the learner's answers in real time and sends them to the server. The server evaluates the received answers and determines whether they are correct. Specifically, the server uses a Python mathematical processing library (e.g., SymPy) to determine whether the answers are correct.

[1258] The server generates hints and explanations as needed based on the evaluation results of the answers. Using the generative AI model, it generates prompts such as "Please give me a hint for the answer to this problem" and inputs them into the AI ​​model. The generated hints and explanations are sent to the device and presented to the learner.

[1259] Furthermore, the device monitors the learner's facial expressions and voice in real time and detects frustration using an emotion engine (e.g., Emotion API). The server receives data from the emotion engine and provides easier questions or additional hints if the learner is frustrated.

[1260] As a concrete example, suppose a learner is working on the problem "Find the derivative of f(x) = x^2." If the device detects that the learner is struggling to find the answer and the emotion engine detects frustration, the server inputs a prompt message to the generative AI model saying, "The learner is feeling frustrated. Please provide an easier problem." The generative AI model then generates a hint such as, "Please tell me the basic steps to find the derivative of f(x) = x^2," and presents it to the learner via the device.

[1261] In this way, the server, terminal, and user work together to provide questions and support according to the learner's level of understanding, thereby improving learning effectiveness.

[1262] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1263] Step 1: Collect learner data

[1264] The server stores in a database the questions that learners have answered in the past, their answer data, the time it took to answer, the process of answering, etc. Specifically, every time a learner submits an answer, the data is recorded in a database management system such as MySQL or PostgreSQL. The input is the learner's answer data, and the output is the learning history data stored in the database.

[1265] Step 2: Assess learner comprehension

[1266] The server evaluates the learner's level of understanding based on the collected data. It preprocesses the data using Python's pandas library and models the level of understanding using machine learning libraries such as scikit-learn. The input is learning history data obtained from the database, and the output is the learner's comprehension score. Specifically, it scores the learner's level of understanding based on past correct answer rates and answer times.

[1267] Step 3: Automatic question generation

[1268] The server uses a generative AI model to automatically generate exercises that correspond to the learner's level of comprehension. Specifically, the server incorporates the comprehension score into a prompt, and inputs a prompt such as "Please generate calculus problems suitable for a learner with a comprehension score of 70" into the generative AI model. The input is the learner's comprehension score and the prompt, and the output is the generated exercises.

[1269] Step 4: State the problem

[1270] The terminal presents the questions received from the server to the learner. Specifically, the questions are displayed on the screen via a web application. The input is the exercise questions sent from the server, and the output is the questions presented to the learner.

[1271] Step 5: Record and evaluate your answers

[1272] The user (learner) answers the questions presented to them. The device records the learner's answers in real time and sends them to the server. The server evaluates the received answers and determines whether they are correct. Specifically, it uses a Python mathematical processing library (e.g., SymPy) to determine whether the answers are correct. The input is the learner's answer data, and the output is the evaluation result of the answer.

[1273] Step 6: Providing hints and explanations

[1274] The server generates hints and explanations as needed based on the evaluation results of the answers. Using the generative AI model, it generates prompts such as "Please give me a hint for the answer to this problem" and inputs them into the AI ​​model. The generated hints and explanations are sent to the device and presented to the learner. The inputs are the evaluation results of the answers and the prompt, and the output is the generated hints and explanations.

[1275] Step 7: Detect and respond to emotions

[1276] The device monitors the learner's facial expressions and voice in real time and detects frustration using an emotion engine. The server receives data from the emotion engine and, if the learner is feeling frustrated, provides easier questions or additional hints. The input is the learner's facial and voice data, and the output is the emotion detection results and countermeasures. Specifically, when the emotion engine detects frustration, the server inputs a prompt statement into the generative AI model saying, "The learner is feeling frustrated. Please provide easier questions," and presents the generated countermeasures to the learner via the device.

[1277] (Application example 2)

[1278] 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."

[1279] Conventional learning support systems and customer service support systems have had difficulty understanding users' levels of understanding and emotional states in real time and providing appropriate feedback and suggestions accordingly. Furthermore, they lacked the means to detect users' frustrations and respond appropriately, which meant that learning effects and customer satisfaction could not be fully improved.

[1280] 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.

[1281] In this invention, the server includes means for providing feedback and supplementary information according to the learning progress and level of understanding using generative AI, means for automatically generating practice problems suited to each user, means for providing answer support, means for detecting the user's emotional state using an emotion engine and taking appropriate action, and means for automatically generating and presenting suggestions according to the user's needs. This enables appropriate feedback and suggestions according to the user's level of understanding and emotional state, thereby improving learning effectiveness and customer satisfaction.

[1282] "Generative AI" is artificial intelligence that automatically generates appropriate feedback and suggestions based on user input and the situation.

[1283] "Feedback" refers to evaluations and advice provided based on a user's behavior and understanding.

[1284] "Supplementary information" is additional information or explanation provided to enhance the user's understanding.

[1285] "Exercises" are problems that learners work on to acquire specific knowledge or skills.

[1286] "Solution support" refers to hints and explanations provided to users when solving problems.

[1287] The "emotion engine" is a technology that analyzes a user's facial expressions and behavior to detect their emotional state.

[1288] "Suggestions" are recommendations of products or services provided based on the user's needs and circumstances.

[1289] "User" means an individual or customer who uses the system.

[1290] "Understanding" refers to the degree to which a user has understood a particular piece of knowledge or skill.

[1291] "Frustration" refers to the dissatisfaction and stress felt by users.

[1292] "Needs" refers to the demands and requirements of users.

[1293] A system for implementing the present invention includes a generative AI, an emotion engine, smart glasses, a camera, and a server. A specific embodiment of this system will be described below.

[1294] The server includes means for using generative AI to provide feedback and supplementary information according to the progress and level of understanding of the learning, means for automatically generating practice problems suited to each individual, means for providing answer support, means for using an emotion engine to detect the user's emotional state and take appropriate action, and means for automatically generating and presenting suggestions according to the user's needs.

[1295] The smart glasses have a built-in camera that captures the user's facial expressions and behavior in real time. The video data acquired from the camera is sent to a server and analyzed by an emotion engine. The emotion engine analyzes the user's facial expressions and behavior to detect their emotional state. For example, if the user is feeling frustrated, the emotion engine sends that information to the generative AI.

[1296] The generative AI automatically generates suggestions based on the user's needs based on the information received from the emotion engine. For example, if the user is feeling frustrated, the generative AI inputs a prompt such as, "The customer is feeling frustrated. Please make an appropriate product suggestion." This suggestion is then displayed on the smart glasses' display and presented to the user.

[1297] As a concrete example, the following prompt sentence can be input to a generative AI model:

[1298] "The customer is frustrated. Please provide appropriate product suggestions."

[1299] This system enables appropriate feedback and suggestions based on the user's level of understanding and emotional state, improving learning effectiveness and customer satisfaction.

[1300] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1301] Step 1:

[1302] The user wears the smart glasses, and the camera captures the user's facial expressions and actions in real time.

[1303] Input: User's facial expressions and actions

[1304] Output: Captured video data

[1305] Specific operation: The camera in the smart glasses detects the user's face and captures video data.

[1306] Step 2:

[1307] The terminal transmits the captured video data to the server.

[1308] Input: Captured video data

[1309] Output: Video data sent to the server

[1310] Specific operation: The smart glasses send video data to the server via Wi-Fi or Bluetooth.

[1311] Step 3:

[1312] The video data received by the server is analyzed using an emotion engine to detect the user's emotional state.

[1313] Input: Video data sent to the server

[1314] Output: User's emotional state (e.g., frustration)

[1315] Specific operation: The server inputs video data into the emotion engine and detects the emotional state using a facial expression analysis algorithm.

[1316] Step 4:

[1317] The server sends the emotional state obtained from the emotion engine to the generative AI.

[1318] Input: User's emotional state

[1319] Output: Emotional state sent to the generative AI

[1320] Specific operation: The server passes the output of the emotion engine to the generative AI.

[1321] Step 5:

[1322] Generative AI automatically generates appropriate suggestions based on your emotional state.

[1323] Input: User's emotional state

[1324] Output: Generated suggestions (e.g. product suggestions)

[1325] Specific operation: The generative AI inputs the prompt sentence, "The customer is frustrated. Please provide an appropriate product suggestion." and generates an appropriate suggestion.

[1326] Step 6:

[1327] The server sends the generated proposal to the smart glasses.

[1328] Input: Generated proposals

[1329] Output: Suggestions sent to the smart glasses

[1330] Specific operation: The server sends the output of the generative AI to the smart glasses.

[1331] Step 7:

[1332] The smart glasses display the suggestions to the user.

[1333] Input: Suggestion sent to smart glasses

[1334] Output: The proposal displayed to the user

[1335] Specific operation: The smart glasses display shows the suggestion and presents it to the user.

[1336] Example 3

[1337] 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."

[1338] Conventional learning support systems have difficulty providing optimal feedback and supplementary information according to each learner's individual level of understanding and learning style, making it impossible to maximize learning effectiveness. They also do not adequately provide appropriate questions to maintain learners' motivation. Furthermore, learning support does not take into account the learner's emotional state, which can lead to a decline in the quality of learning.

[1339] 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.

[1340] In this invention, the server includes means for providing feedback and supplementary information according to the progress and level of understanding of the learner using generative AI, means for automatically generating practice problems suited to each learner, means for providing answer support, means for acquiring the learner's learning history and progress data, means for analyzing the learner's level of understanding and learning style, means for monitoring the learner's facial expressions and behavior and detecting their excitement level, and means for generating new questions based on the detected excitement level. This allows the provision of learning support optimized for each learner, maximizing learning effectiveness and maintaining motivation to learn.

[1341] "Generative AI" is an artificial intelligence technology that provides optimal feedback and supplementary information based on the learner's progress and level of understanding, and automatically generates practice problems that meet individual needs.

[1342] "Feedback" refers to evaluations and advice provided to learners regarding their learning activities, and is information used to improve learning outcomes.

[1343] "Supplementary information" refers to additional knowledge or explanation provided to deepen a learner's understanding.

[1344] "Practice problems" are problems that students should solve in order to gain a practical understanding of the learning content.

[1345] "Solution support" refers to hints and explanations provided to learners when solving practice problems.

[1346] A "learning history" is a record of the learning activities that a learner has undertaken up to now.

[1347] "Progress data" refers to data that indicates how far a learner has progressed in their studies.

[1348] "Level of understanding" is an indicator that shows how well a learner understands a particular learning content.

[1349] "Learning style" refers to the characteristics of a learner that indicate the method and environment in which they learn most effectively.

[1350] "Facial expressions" refer to the learner's facial expressions, which provide clues to their emotional state.

[1351] "Behavior" refers to the actions and reactions that learners make while learning.

[1352] "Excitement" refers to the joy and sense of accomplishment that learners feel when they solve a problem.

[1353] This invention is a system that uses generative AI to provide feedback and supplementary information according to learning progress and level of understanding, automatically generates practice problems suited to each individual, and provides answer support. Furthermore, by acquiring the learner's learning history and progress data and analyzing their level of understanding and learning style, it provides learning support optimized for each learner. It also generates new problems to maintain motivation by monitoring the learner's facial expressions and behavior and detecting their excitement.

[1354] Hardware and software used

[1355] Hardware: Servers, devices (PCs, tablets, smartphones)

[1356] Software: Generative AI models (e.g., GPT-4), emotion engines, database management systems (e.g., MySQL)

[1357] Program processing

[1358] 1. The user logs in to the learning platform. The user accesses the learning platform using a device (PC, tablet, smartphone) and enters their login information.

[1359] 2. The server retrieves the user's learning history and progress data from the database. The server connects to a database management system (e.g., MySQL) and queries the user's learning history and progress data.

[1360] 3. The server sends learning history and progress data to the generative AI model. The server then calls an API to send the acquired data to the generative AI model.

[1361] 4. The generative AI model analyzes the user's level of understanding and learning style. The generative AI model (e.g., GPT-4) analyzes the received data and evaluates the user's level of understanding and learning style.

[1362] 5. The generative AI model generates optimized learning content based on the analysis results. The generative AI model generates learning content (e.g., new questions and explanations) that is optimal for the user.

[1363] 6. The server sends the generated learning content to the device. The server sends the learning content received from the generative AI model to the user's device.

[1364] 7. The device displays the learning content to the user. The device displays the received learning content to the user.

[1365] 8. The user uses the learning content to solve the problem. The user uses the displayed learning content to solve the problem.

[1366] 9. The emotion engine monitors the user's facial expressions and behavior to detect excitement. The emotion engine analyzes the user's facial expressions and tone of voice via the user's camera and microphone to detect excitement.

[1367] 10. The emotion engine sends the excited state information detected to the generative AI model. The emotion engine sends the excited state data detected to the generative AI model.

[1368] 11. The generative AI model generates new questions based on the excitement state information. The generative AI model generates new questions based on the excitement state information to maintain the user's motivation.

[1369] 12. The server sends the new problem to the device. The server sends the new problem received from the generative AI model to the user's device.

[1370] 13. The device presents a new problem to the user. The device displays a new problem to the user and prompts them to take the next learning step.

[1371] Specific examples

[1372] For example, suppose a user is solving a math problem. When the user solves the problem, the emotion engine detects the user's joy. Based on this information, the generative AI model generates the next problem to be solved and displays it on the device.

[1373] Prompt Sentence Examples

[1374] "If you detect the joy the user feels after solving a math problem, generate the next problem for them to solve."

[1375] This system provides learning support that is optimized for each learner, maximizing the learning effect. The flow of the specific processing in the third embodiment will be described with reference to FIG.

[1376] Step 1:

[1377] A user logs into the learning platform.

[1378] Input: User login information (user ID, password)

[1379] Specific operation: The user accesses the learning platform using a device (PC, tablet, smartphone) and enters login information.

[1380] Output: Login success message, user session information

[1381] Step 2:

[1382] The server retrieves the user's learning history and progress data from the database.

[1383] Input: User session information

[1384] Specific operation: The server connects to a database management system (e.g., MySQL) and queries the user's learning history and progress data.

[1385] Output: User learning history data, progress data

[1386] Step 3:

[1387] The server sends learning history and progress data to the generated AI model.

[1388] Input: User learning history data, progress data

[1389] Specific operation: The server calls an API to send the acquired data to the generative AI model.

[1390] Output: Message that data has been sent to the generative AI model

[1391] Step 4:

[1392] The generative AI model analyzes the user's level of understanding and learning style.

[1393] Input: User learning history data, progress data

[1394] What it does: A generative AI model (e.g., GPT-4) analyzes the data it receives and evaluates the user's level of understanding and learning style.

[1395] Output: User comprehension assessment results, learning style analysis results

[1396] Step 5:

[1397] The generative AI model generates optimized learning content based on the analysis results.

[1398] Input: User comprehension assessment results, learning style analysis results

[1399] Specific operation: The generative AI model generates optimal learning content (e.g., new questions and explanations) for the user.

[1400] Output: Optimized learning content

[1401] Step 6:

[1402] The server transmits the generated learning content to the terminal.

[1403] Input: Optimized learning content

[1404] Specific operation: The server sends the learning content received from the generative AI model to the user's device.

[1405] Output: Learning content submission completion message

[1406] Step 7:

[1407] The terminal displays the learning content to the user.

[1408] Input: Optimized learning content

[1409] Specific operation: The device displays the received learning content to the user.

[1410] Output: User views learning content

[1411] Step 8:

[1412] The user uses the learning content to solve the problem.

[1413] Input: Optimized learning content

[1414] Specific Action: The user solves a problem using the displayed learning content.

[1415] Output: Problem-solving results

[1416] Step 9:

[1417] The emotion engine monitors the user's facial expressions and behavior to detect excitement.

[1418] Input: User facial expression data, behavior data

[1419] Specific operation: The emotion engine analyzes the user's facial expressions and tone of voice via the camera and microphone to detect excitement.

[1420] Output: Excitation state detection result

[1421] Step 10:

[1422] The emotion engine sends the detected excitement information to the generative AI model.

[1423] Input: Excitation state detection result

[1424] Specific operation: The emotion engine sends the excited state data detected to the generative AI model.

[1425] Output: Message that data has been sent to the generative AI model

[1426] Step 11:

[1427] The generative AI model generates new problems based on the excited state information.

[1428] Input: Excitation state detection result

[1429] Specific operation: The generative AI model generates new problems based on the excitement state information to maintain the user's motivation.

[1430] Output: New problem

[1431] Step 12:

[1432] The server sends a new question to the device.

[1433] Input: New problem

[1434] Specific operation: The server sends the new problem received from the generative AI model to the user's device.

[1435] Output: New issue submission successful message

[1436] Step 13:

[1437] The terminal presents the user with a new problem.

[1438] Input: New problem

[1439] Specific operation: The device displays a new problem to the user and prompts them for the next learning step.

[1440] Output: User views new question

[1441] (Application example 3)

[1442] 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."

[1443] Conventional learning support systems are insufficient in providing feedback and supplementary information according to the learner's progress and level of understanding, and the automatic generation of practice problems and answer support tailored to individual needs are also limited. Furthermore, when it comes to learning support for workers in factories, there is a lack of optimized training programs based on work history and progress, and a lack of means to maintain motivation that takes emotional state into account. This makes it difficult to maximize learning effectiveness and work efficiency.

[1444] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[1445] In this invention, the server includes means for using generative AI to provide feedback and supplementary information according to the learning progress and level of understanding, means for automatically generating practice problems suited to each individual, means for providing answer support, means for recording the worker's work history and progress and storing it in a database, means for providing the worker with an optimized training program based on the stored data, and means for detecting the worker's emotional state using an emotion engine and adjusting the next task content based on that information. This makes it possible to provide learning support and training programs optimized for each learner and worker, maximizing learning effectiveness and work efficiency.

[1446] "Generative AI" is an artificial intelligence technology that provides optimal feedback and supplementary information based on the progress and level of understanding of learners and workers, and automatically generates exercises and training programs that meet individual needs.

[1447] "Feedback" refers to information and advice provided to learners and workers based on their progress and level of understanding, with the aim of improving learning effectiveness and work efficiency.

[1448] "Supplementary information" refers to additional information or materials provided to enhance a learner's or worker's understanding.

[1449] "Practice problems" are problems that learners must solve to acquire specific knowledge or skills, and are automatically generated by generative AI.

[1450] "Solution support" refers to hints and explanations provided to learners when they are solving practice problems, and is intended to enhance learning effectiveness.

[1451] "Work history" refers to a record of work that a worker has done in the past, and is stored in a database.

[1452] "Progress" is a measure of how far a learner or worker has progressed toward a particular task or goal.

[1453] A "database" is a system for efficiently storing, managing, and searching collected data.

[1454] "Training Program" means a series of learning activities or exercises designed to enable a worker to acquire specific skills or knowledge.

[1455] An "emotion engine" is a technology that detects the emotional state of learners and workers and uses that information to provide appropriate feedback and adjust the next task.

[1456] An "emotional state" is a psychological state such as joy, excitement, or stress that a learner or worker feels in a particular situation.

[1457] A system for implementing this invention has the following configuration. First, the server uses generative AI to provide feedback and supplementary information according to the learning progress and level of understanding. Specifically, the server collects progress data of learners and workers and stores it in a database. Next, the generative AI generates a training program optimized for the learner or worker based on the stored data.

[1458] The server uses the following hardware and software:

[1459] Hardware:

[1460] Factory robots (e.g. general-purpose robotic arms)

[1461] Emotion detection sensors (e.g. emotion recognition sensors)

[1462] Database server (e.g. MySQL server)

[1463] software:

[1464] Generative AI models (e.g., GPT-4)

[1465] Emotion engine (e.g. emotion recognition SDK)

[1466] Data analysis tools (e.g., Python's Pandas library)

[1467] The server first uses the factory robot to collect the worker's work history and progress in real time and stores it in a MySQL database. It then uses Python's Pandas library to analyze the collected data and evaluate the worker's level of understanding and work style. It then uses a generative AI model (GPT-4) to generate a training program optimized for the worker and provides it to the worker through the factory robot's interface.

[1468] Furthermore, an emotion engine (emotion recognition SDK) is used to detect the emotional state of the worker in real time. It detects the joy of the worker when they have a successful experience and adjusts the next task based on that information. This helps maintain the worker's motivation and maximizes learning effects and work efficiency.

[1469] As a concrete example, if Worker A is learning a new work procedure and has difficulty with a particular step, the server analyzes Worker A's progress data and uses a generative AI model to generate training content specific to the step he or she is having difficulty with. The emotion engine detects the joy Worker A feels when he or she successfully completes that step and provides the next step of the same difficulty level.

[1470] An example of a prompt sentence is as follows:

[1471] Generate training content tailored to specific procedures based on the progress data of Worker A. Also consider the emotional data of Worker A when he had a successful experience, and adjust the difficulty of the next procedure to be provided.

[1472] In this way, the server provides learning support and training programs optimized for each learner and worker, maximizing learning effectiveness and work efficiency.

[1473] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1474] Step 1:

[1475] The server uses factory robots to collect workers' work history and progress data in real time.

[1476] Input: Work history and progress data of workers

[1477] Output: Collected work history and progress data

[1478] Specific operation: Using sensors and cameras installed on factory robots, workers' movements and work content are recorded and stored in a database.

[1479] Step 2:

[1480] The server stores the collected work history and progress data in a database.

[1481] Input: Collected work history and progress data

[1482] Output: Data stored in the database

[1483] What it does: Connects to a MySQL database and inserts collected data into the appropriate tables.

[1484] Step 3:

[1485] The server uses Python's Pandas library to analyze the accumulated data and evaluate the worker's level of understanding and working style.

[1486] Input: Work history and progress data stored in the database

[1487] Output: Evaluation results of worker's understanding and work style

[1488] Specific operations: Use the Pandas library to read data and apply statistical methods and machine learning algorithms to evaluate worker performance.

[1489] Step 4:

[1490] The server uses a generative AI model (GPT-4) to generate a training program optimized for the worker.

[1491] Input: Evaluation results of worker's understanding and work style

[1492] Output: Optimized training program

[1493] Specific operation: The evaluation results are input as prompts into GPT-4 to obtain the generated training program.

[1494] Step 5:

[1495] The server provides the generated training program to the worker through the interface of the factory robot.

[1496] Input: Optimized training program

[1497] Output: Training program provided to the worker

[1498] Specific operation: The contents of the training program are presented to workers using the factory robot's display and audio output.

[1499] Step 6:

[1500] The server uses an emotion engine (emotion recognition SDK) to detect the worker's emotional state in real time.

[1501] Input: Real-time video and audio data of the worker

[1502] Output: Emotional state data of the worker

[1503] Specific operation: Using the emotion recognition SDK, the worker's facial expressions and tone of voice are analyzed to detect their emotional state.

[1504] Step 7:

[1505] The server adjusts the next task based on the emotional state data.

[1506] Input: Worker emotional state data

[1507] Output: Adjusted next steps

[1508] Specific operation: Analyzes emotional state data, detects the joy felt when a worker has a successful experience, and adjusts the difficulty of the next task provided.

[1509] In this way, the server provides learning support and training programs optimized for each learner and worker, maximizing learning effectiveness and work efficiency.

[1510] 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.

[1511] 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.

[1512] Another example of generative AI is Gemini (internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[1513] 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.

[1514] [Third embodiment]

[1515] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1516] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1517] 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).

[1518] 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.

[1519] 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.

[1520] 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).

[1521] 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.

[1522] 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.

[1523] 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.

[1524] 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.

[1525] 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.

[1526] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1527] "Example 1"

[1528] The present invention is an educational platform that uses generative AI. Specifically, it grasps the learner's learning progress and level of understanding and provides feedback and supplementary information accordingly. This feedback and supplementary information is provided in the form of text, audio, video, etc., and helps deepen the learner's understanding. It also automatically generates exercises based on the learner's level of understanding, interests, and learning style. This provides questions that are tailored to each learner, maximizing learning effectiveness. It also provides an answer support function to support the learner as they work through the questions.

[1529] "Example 2"

[1530] As a concrete example, let's imagine a mathematics seminar class. Suppose a student is studying calculus. The generative AI grasps the student's level of understanding and automatically generates problems that correspond to the student's level of understanding, from basic concepts to applications of calculus. It also provides hints and explanations for the answers as the student works on the problems, deepening the student's understanding.

[1531] "Example 3"

[1532] Furthermore, generative AI accumulates learners' learning history and progress, and uses this information to analyze their level of understanding and learning style, thereby providing optimal learning support for each learner and maximizing learning effectiveness.

[1533] The processing flow of each embodiment will be described below.

[1534] "Example 1"

[1535] Step 1: The generative AI understands the learner's learning progress and level of understanding based on the learner's activities on the platform and test results.

[1536] Step 2: Generate and provide feedback and supplementary information based on the learner's level of understanding. This can be in the form of text, audio, or video to deepen the learner's understanding.

[1537] Step 3: Automatically generate exercises tailored to the learner's level of understanding, interests, and learning style. This provides questions that are tailored to each individual learner, maximizing learning effectiveness.

[1538] Step 4: As students work through the problems, provide them with hints and explanations to deepen their understanding.

[1539] "Example 2"

[1540] Step 1: The student begins learning calculus.

[1541] Step 2: The generative AI grasps the learner's level of understanding and automatically generates questions that correspond to the learner's level of understanding, from basic concepts of calculus to applications.

[1542] Step 3: As students work through the problems, provide them with hints and explanations to deepen their understanding.

[1543] "Example 3"

[1544] Step 1: The generative AI accumulates the learner's learning history and progress.

[1545] Step 2: The generative AI analyzes the learner's level of understanding and learning style based on the accumulated data.

[1546] Step 3: Based on the analysis results, provide learning support optimized for each learner to maximize learning effectiveness.

[1547] Example 1

[1548] 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."

[1549] With conventional educational platforms, it is difficult to provide appropriate feedback and supplementary information according to each learner's progress and level of understanding, and there is a lack of automatic generation of exercises and answer support that meet individual needs. This makes it difficult to maximize learning effectiveness, and the challenge is to provide an optimal learning experience that matches the learner's level of understanding and interests.

[1550] 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.

[1551] In this invention, the server includes means for collecting learner progress data, means for analyzing the collected data and evaluating the learner's level of understanding, means for generating feedback and supplementary information based on the evaluation results, means for automatically generating exercises according to the learner's level of understanding and interests, and means for providing support for answering the exercises. This makes it possible to provide optimal feedback and supplementary information to each learner, automatically generate exercises that meet the learner's individual needs, and provide support for answering the exercises.

[1552] "Student progress data" refers to a record of the learning activities that a learner has performed on the educational platform, and includes information such as study time, percentage of correct answers, and number of problems attempted.

[1553] A "means for collecting" is a method or device for obtaining learner progress data from a learning management system or other educational platform.

[1554] "Means for analyzing and assessing comprehension" refers to methods and devices for assessing learners' comprehension based on collected data, and involves analyzing the data using a generative AI model.

[1555] The "means for generating feedback and supplementary information" refers to a method or device for generating appropriate feedback and supplementary information based on the results of the learner's comprehension assessment.

[1556] "Means for automatically generating exercises" refers to a method or device for automatically creating exercises that correspond to the learner's level of understanding and interests.

[1557] A "means for providing solution support" is a method or device for providing hints and step-by-step explanations of solutions to learners as they work through practice problems.

[1558] This invention relates to an educational platform that uses a generative AI model. Specifically, it is a system that grasps a learner's learning progress and level of understanding and provides feedback and supplementary information accordingly. The system aims to maximize learning effectiveness by automatically generating exercises tailored to each learner and providing answer support functions.

[1559] The server collects learner progress data from learning management systems (LMS) and other educational platforms. Collected data includes study time, correct answer rate, and number of problems attempted. The server inputs this data into a generative AI model (e.g., OpenAI's GPT-4) to assess the learner's level of understanding. The generative AI model analyzes the learner's response patterns and progress to identify which concepts and skills the learner is lacking.

[1560] Based on the analysis results, the server generates appropriate feedback and supplementary information for the learner. For example, if the learner's understanding of a particular concept is lacking, the server generates detailed explanations and examples in the form of text or video. This allows the learner to supplement their missing knowledge and deepen their understanding.

[1561] Furthermore, the server automatically generates exercises tailored to the learner's level of understanding, interests, and learning style. Using a generative AI model, it creates exercises with the optimal difficulty and content for each learner. For example, it starts with basic integral problems and gradually increases the difficulty level. This allows learners to progress at their own pace.

[1562] The server provides solution support functions for students as they work on practice problems. Specifically, it generates hints and step-by-step explanations of solutions to support the problem-solving process. This helps students understand the problem-solving process more easily, improving their learning effectiveness.

[1563] As a concrete example, consider a user studying a mathematics unit on calculus. The server collects the user's progress data and inputs it into a generative AI model. The generative AI model detects that the user has insufficient understanding of a particular concept (e.g., the Fundamental Theorem of Integration). Based on this information, the server generates supplementary information about the Fundamental Theorem of Integration in the form of text and video and provides it to the user.

[1564] Furthermore, the server automatically generates exercises tailored to the user's level of understanding. For example, it starts with basic integral problems and provides problems of gradually increasing difficulty. As the user works on the problems, a solution assistance function is applied, providing hints and step-by-step explanations of the solutions.

[1565] Example prompt sentence:

[1566] Generate supplementary information about the Fundamental Theorem of Integration based on the user's learning progress. Create text and video content to explain the theory in a way that is easy for users to understand.

[1567] In this way, the server leverages generative AI models to provide each learner with an optimal learning experience.

[1568] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1569] Step 1:

[1570] The server collects learner progress data from learning management systems (LMS) and other educational platforms. Specifically, it obtains data such as study time, correct answer rate, and number of questions attempted through API. The input is the learner progress data, and the output is the collected progress data.

[1571] Step 2:

[1572] The server inputs the collected progress data into a generative AI model to evaluate the learner's level of understanding. The generative AI model analyzes the learner's response patterns and progress status to identify which concepts and skills are lacking. The input is the collected progress data, and the output is the comprehension assessment results.

[1573] Step 3:

[1574] The server generates feedback and supplementary information based on the results of comprehension assessment. For example, if a user's understanding of a particular concept is insufficient, it generates detailed explanations and examples of that concept in text or video format. The input is the comprehension assessment result, and the output is the generated feedback and supplementary information.

[1575] Step 4:

[1576] The server automatically generates exercises based on the learner's level of comprehension and interests. Using a generative AI model, it creates problems with the optimal difficulty and content for each learner. For example, it starts with basic integral problems and provides problems of gradually increasing difficulty. The input is the comprehension assessment results, and the output is the automatically generated exercises.

[1577] Step 5:

[1578] The server provides answer support functions when students work on exercises. Specifically, it generates hints and step-by-step explanations of solutions to support the process of solving the problems. The input is the automatically generated exercises, and the output is the provided answer support information.

[1579] In this way, the server leverages generative AI models to provide each learner with an optimal learning experience.

[1580] (Application example 1)

[1581] 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."

[1582] Conventional educational platforms have difficulty providing individual feedback and supplementary information based on learners' progress and level of understanding, which prevents them from fully improving learning outcomes. Furthermore, in the education and training of factory workers, there is a lack of individual progress management and feedback based on their level of understanding, making it difficult to improve work efficiency and quality. Furthermore, the automatic generation of exercises and answer support functions are inadequate, making it difficult to provide education that meets the individual needs of learners and workers.

[1583] 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.

[1584] In this invention, the server includes means for using generative AI to provide feedback and supplementary information according to the progress and level of understanding of learning, means for automatically generating practice problems suited to each worker, means for providing answer support, means for monitoring the worker's progress in real time, means for providing feedback according to the worker's level of understanding in the form of text, audio, and video, means for automatically generating practice problems based on the worker's level of understanding and interests, and means for supporting the worker when working on the problems. This enables education and training that meets the individual needs of learners and workers, improving learning effectiveness, work efficiency, and quality.

[1585] "Generative AI" is an artificial intelligence technology that generates new information and content based on data.

[1586] "Learning progress" is an indicator of how far a learner has progressed in the learning process.

[1587] "Level of understanding" is an indicator that shows how well a learner understands a particular content.

[1588] "Feedback" is evaluation or advice given to a learner regarding their behavior or performance.

[1589] "Supplementary information" is additional information provided to enhance the learner's understanding.

[1590] "Practice problems" are problems that learners solve to confirm what they have learned.

[1591] "Auto-generation" means that a system automatically creates content or information without human intervention.

[1592] "Solution support" refers to advice and hints provided to learners when solving problems.

[1593] "Workers" refers to people who perform work in factories or work sites.

[1594] "Real-time monitoring" means monitoring the ongoing situation immediately.

[1595] "Text" is information written in text.

[1596] "Sound" is audible information.

[1597] "Video" is a moving image that is visually displayed.

[1598] "Interests" are areas or content in which a learner or worker is particularly interested.

[1599] This invention is a system that applies an educational platform using generative AI to the education and training of factory workers. A specific embodiment of this system is shown below.

[1600] System configuration

[1601] The system consists of the following major components:

[1602] 1. Server: Runs the generative AI and provides feedback and supplementary information based on learning progress and level of understanding.

[1603] 2. Devices: Tablets or smartphones used by workers, allowing them to receive real-time feedback and practice questions.

[1604] 3. Network: Connects the server and the terminal to send and receive data.

[1605] Program processing

[1606] The server processes the data in the following steps:

[1607] 1. Collecting progress data: Collect progress data from the workers' devices, including the progress and understanding of the work.

[1608] 2. Feedback generation: Based on the progress data collected, generative AI is used to generate appropriate feedback, which can be in the form of text, audio, or video.

[1609] 3. Automatic generation of exercises: Generative AI is used to automatically generate exercises based on the worker's level of understanding and interests.

[1610] 4. Solution support: When workers tackle practice problems, generative AI is used to provide solution support.

[1611] Hardware and software used

[1612] Hardware: Factory robots, tablets and smartphones used by workers

[1613] Software: Python, OpenAI API

[1614] Specific examples

[1615] Progress data collection

[1616] The server collects the following progress data from the worker's terminal:

[1617] Worker ID: 12345

[1618] Learning progress: 50%

[1619] Comprehension: Medium

[1620] Interests: Machine operation

[1621] Generate feedback

[1622] The server generates the following feedback based on the progress data collected:

[1623] "You're making good progress. Now let's review the basics of machine operation."

[1624] Automatic generation of exercises

[1625] The server generates the following exercises based on the worker's level of understanding:

[1626] "Please answer the following questions regarding the basics of machine operation: 1. Explain the procedure for starting a machine."

[1627] Prompt Sentence Examples

[1628] Below are some examples of prompts used with generative AI:

[1629] Feedback generation prompt: "Generate appropriate feedback based on the following worker progress data: Worker ID: 12345, Learning Progress: 50%, Understanding: Medium, Interest: Machine Operation."

[1630] Exercise generation prompt: "Generate exercises suitable for workers with a medium level of understanding."

[1631] In this way, the server can use generative AI to support worker education and training, improving learning effectiveness, work efficiency, and quality.

[1632] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1633] Step 1:

[1634] The server collects progress data from the worker's device. Specifically, it acquires data such as the worker's ID, learning progress, level of understanding, and interest. The input is the progress data sent from the worker's device, and the output is the progress data saved on the server.

[1635] Step 2:

[1636] The server sends prompts to the generative AI based on the collected progress data and generates feedback. Specifically, it converts the progress data into prompts in text format and sends them to the generative AI. The inputs are the progress data and prompts, and the output is the generated feedback.

[1637] Step 3:

[1638] The server sends the generated feedback to the worker's terminal. Specifically, the feedback is delivered to the worker's terminal in the form of text, audio, or video. The input is the generated feedback, and the output is the feedback displayed on the worker's terminal.

[1639] Step 4:

[1640] The server automatically generates exercises based on the worker's level of understanding and interest. Specifically, it generates prompts based on the data on level of understanding and interest and sends them to the generative AI. The input is the data on level of understanding and interest and the prompts, and the output is the generated exercises.

[1641] Step 5:

[1642] The server sends the generated exercises to the worker's terminal. Specifically, the exercises are delivered to the worker's terminal in text format. The input is the generated exercise, and the output is the exercise displayed on the worker's terminal.

[1643] Step 6:

[1644] The server provides answer support when workers tackle practice problems. Specifically, it collects the workers' answer data and sends prompts to the generative AI to generate answer support. The input is the worker's answer data and prompts, and the output is the generated answer support.

[1645] Step 7:

[1646] The server sends the generated answer support to the worker's terminal. Specifically, the answer support is delivered to the worker's terminal in the form of text, audio, and video. The input is the generated answer support, and the output is the answer support displayed on the worker's terminal.

[1647] Example 2

[1648] 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."

[1649] Conventional learning support systems have had issues with the difficulty of providing appropriate questions according to the learner's level of understanding, and the lack of feedback on answers and supplementary information limits the learning effect. In particular, there is a need to automatically generate questions that meet individual learning needs and improve the quality of answer support.

[1650] 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.

[1651] In this invention, the server includes means for collecting data such as the learner's past answer history and answer time, and evaluating the learner's level of understanding using a generative AI model, means for inputting prompts into the generative AI model to automatically generate questions according to the learner's level of understanding, means for displaying questions to the learner and providing an interface for inputting answers, and means for evaluating the learner's answers and generating and providing answer hints and explanations as necessary. This makes it possible to provide appropriate questions according to the learner's level of understanding and provide high-quality answer support.

[1652] "Learner" refers to an individual receiving education or training.

[1653] "Answer history" refers to a record of questions that a learner has answered in the past and the results of those answers.

[1654] "Response time" refers to the time it takes a learner to complete a response to a particular question.

[1655] A "generative AI model" refers to an algorithm or system that uses artificial intelligence techniques to analyze data and perform a specific task (e.g., automatically generating questions or evaluating answers).

[1656] "Understanding" refers to an indicator of how well a learner understands specific knowledge or skills.

[1657] A "prompt" is an instruction given to a generative AI model to perform a specific task.

[1658] "Interface" refers to the means or screen through which a learner interacts with a system.

[1659] "Hints" refer to additional information or advice that can be helpful to learners when solving a problem.

[1660] "Explanation" refers to information that provides detailed explanations of how to answer a question and related concepts.

[1661] The present invention is a system for automatically generating questions according to a learner's level of understanding and providing support for answering the questions. A specific embodiment of this system will be described below.

[1662] First, the server collects data such as the learner's past answer history and answer time. This data is stored in a database and managed for each learner. The server then inputs the collected data into a generative AI model (e.g., a general natural language processing model) to evaluate the learner's level of understanding.

[1663] Next, the server inputs prompt sentences into the generative AI model to automatically generate questions according to the learner's level of understanding. For example, the following prompt sentences can be used:

[1664] To assess students' understanding of the basic concept of definite integrals, generate a problem like this: 'Find the definite integral of the following function: ∫(2x + 3)dx, from x = 0 to x = 2.'

[1665] The generated questions are sent from the server to the terminal. The terminal displays the questions to the learner and provides an interface for inputting answers. The learner inputs their answer to the displayed question. For example, if the learner inputs the answer "7", the terminal sends the answer to the server.

[1666] The server receives the learner's answer and uses the generative AI model to determine whether the answer is correct. For example, the generative AI model receives a prompt such as, "Please evaluate whether the learner's answer '7' is correct." The generative AI model determines that the answer is correct. The server then sends this evaluation result to the device, which then displays "That's correct!" to the learner.

[1667] Furthermore, if the learner enters an incorrect answer, the server inputs a prompt to the generative AI model saying, "Please generate a hint or explanation for the solution to this problem." The generative AI model generates a hint saying, "As a hint for the solution to this problem, first integrate the function, and then explain how to apply the range of the definite integral." The server sends this hint to the device, which then displays the hint to the learner.

[1668] In this way, a system is realized in which the server, terminal, and user work together to provide questions that correspond to the learner's level of understanding and support their learning. This system allows learners to tackle questions that are appropriate for their level of understanding and receive high-quality answer support.

[1669] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1670] Step 1:

[1671] The server collects data such as the learner's past answer history and answer time.

[1672] Input: Learner's answer history data, answer time data

[1673] Data processing: Obtain each learner's answer history and answer time from the database and format them into an analyzable format.

[1674] Output: Formatted answer history data and answer time data of the learner

[1675] Step 2:

[1676] The server inputs the collected data into a generative AI model to evaluate the learner's level of understanding.

[1677] Input: Formatted answer history data of learners, answer time data

[1678] Data computation: Input data into the generative AI model and evaluate comprehension using the prompt, "Please rate this learner's comprehension."

[1679] Output: Learner comprehension assessment results

[1680] Step 3:

[1681] The server inputs prompt sentences into the generative AI model and automatically generates questions based on the learner's level of understanding.

[1682] Input: Learner comprehension assessment results, prompt text

[1683] Data calculation: The generative AI model is given a prompt, "Please generate questions that correspond to the learner's level of understanding," and an appropriate question is generated.

[1684] Output: Generated problem

[1685] Step 4:

[1686] The server sends the generated questions to the terminal.

[1687] Input: Generated question

[1688] Data processing: Convert the generated questions into a format that can be sent to the terminal.

[1689] Output: Problems sent to terminal

[1690] Step 5:

[1691] The terminal displays questions to the learner and provides an interface for entering answers.

[1692] Input: Question sent to terminal

[1693] Specific operation: The device displays the questions on the screen and provides a form where the learner can enter their answers.

[1694] Output: The answer entered by the learner

[1695] Step 6:

[1696] The terminal transmits the learner's answers to the server.

[1697] Input: The answer entered by the learner

[1698] Data processing: Convert the learner's answers into a format that can be sent to the server.

[1699] Output: The learner's answer sent to the server

[1700] Step 7:

[1701] The server receives the learner's answer and uses the generative AI model to determine whether the answer is correct.

[1702] Input: Learner's answer

[1703] Data calculation: The generative AI model is given a prompt, "Please evaluate whether the learner's answer is correct," and the model determines whether the answer is correct or not.

[1704] Output: Evaluation result of the answer

[1705] Step 8:

[1706] The server transmits the evaluation result of the answer to the terminal.

[1707] Input: Answer evaluation result

[1708] Data processing: The evaluation results of the answers are converted into a format that can be sent to the terminal.

[1709] Output: Evaluation result of the answer sent to the device

[1710] Step 9:

[1711] The terminal displays the evaluation results of the answers to the learner.

[1712] Input: Evaluation result of the answer sent to the terminal

[1713] Specific operation: The device displays the evaluation results of the answers on the screen and provides feedback to the learner.

[1714] Output: The evaluation result of the answer displayed to the learner

[1715] Step 10:

[1716] If a learner enters an incorrect answer, the server inputs a prompt to the generative AI model saying, "Please generate hints and explanations for the answer to this question."

[1717] Input: Learner's incorrect answer, prompt

[1718] Data calculation: A prompt sentence is input into the generative AI model to generate hints and explanations for the answer.

[1719] Output: Generated hints and explanations

[1720] Step 11:

[1721] The server sends the generated hints and explanations to the terminal.

[1722] Input: Generated hints and explanations

[1723] Data processing: Convert the generated hints and explanations into a format that can be sent to the device.

[1724] Output: Hints and explanations sent to the terminal

[1725] Step 12:

[1726] The device displays hints and explanations to the learner.

[1727] Input: Hints and explanations sent to your device

[1728] Specific operation: The device displays hints and explanations on the screen to deepen the learner's understanding.

[1729] Output: Hints and explanations displayed to the learner

[1730] (Application example 2)

[1731] 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."

[1732] Conventional learning systems have the problem of not providing enough questions or feedback that correspond to the learner's level of understanding, limiting the learning effect. Furthermore, there is a lack of appropriate support for learners to progress through their studies at their own pace, making it difficult to improve the quality of their learning. Furthermore, in a learning environment using mobile devices such as smartphones, there is a need for a system that can generate questions and provide answer support that correspond to individual needs.

[1733] 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.

[1734] In this invention, the server includes means for providing feedback and supplementary information according to the learner's learning progress and level of understanding using generative AI, means for automatically generating practice problems suited to each individual, means for providing answer support, means for generating questions according to the learner's level of understanding and providing hints and explanations through an application installed on a smartphone, and means for inputting prompt sentences using a generative AI model and generating questions, hints, and explanations. This enables individualized support according to the learner's level of understanding, improving both the effectiveness and quality of learning.

[1735] "Generative AI" is an artificial intelligence technology that analyzes a learner's level of understanding and progress, and automatically generates appropriate feedback, supplementary information, questions, hints, and explanations based on that analysis.

[1736] "Feedback" means providing learners with evaluations and advice based on their level of understanding and progress in response to the problems and tasks they have tackled.

[1737] "Supplementary information" is information that provides additional knowledge or explanation that learners need to deepen their understanding.

[1738] "Exercises" are problems that learners work on to understand and master specific learning content.

[1739] "Automatic generation" refers to the use of artificial intelligence or algorithms to generate questions or information without manual intervention.

[1740] "Solution support" means providing hints and explanations that learners need to solve problems, helping them arrive at the answer.

[1741] A "smartphone" is a type of mobile phone, a multi-function device that can connect to the Internet and use applications.

[1742] An "application" is a software program designed to provide a particular function or service.

[1743] A "prompt" is an instruction given to a generative AI to generate a specific output.

[1744] "Problems, hints, and explanations" are practice problems that students need to solve in order to advance their studies, clues to help them find the answers, and detailed explanations of the answers to the problems.

[1745] The system for implementing this invention uses generative AI to provide feedback and supplemental information according to the progress and level of understanding of the learning, automatically generate practice problems suited to each individual, and provide answer support. A specific embodiment of this system is shown below.

[1746] The server uses a generative AI model to analyze the learner's level of understanding and provides appropriate feedback and supplementary information based on that analysis. Specifically, as the learner works on problems through an application installed on their smartphone, the server evaluates their answers in real time to determine their level of understanding. The server then automatically generates the next problem based on the learner's level of understanding and sends it to the application.

[1747] This system uses "text-davinci-003" as the generative AI model. The server receives a prompt as input and generates questions, hints, and explanations. For example, if the learner's level of comprehension is "intermediate," the following prompt is used:

[1748] Problem generation prompt: "Generate calculus problems appropriate for learners with an intermediate level of comprehension."

[1749] Hint prompt: "Please provide a hint for the following problem: {generated problem}"

[1750] Explanation prompt: "Please provide an explanation for the following problem: {generated problem}"

[1751] The server inputs these prompts into a generative AI model and sends the resulting output to a smartphone application, which displays the generated questions, hints, and explanations to the learner to support their learning.

[1752] As a concrete example, consider the case where a learner is working on a calculus problem. When the learner answers the problem, the answer is sent to the server, where a generative AI model evaluates the answer. Based on the evaluation results, the next problem is automatically generated and provided to the learner. Hints and detailed explanations for the answer are also provided at the same time, allowing the learner to deepen their understanding at their own pace.

[1753] In this way, systems using generative AI can respond individually to learners' levels of understanding, improving both the effectiveness and quality of learning.

[1754] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1755] Step 1:

[1756] The user launches the smartphone application and begins learning. The user selects the content to study (e.g., calculus) on the application. The input is the user's selected learning content, and the output is initial setting information based on that content. The application sends the initial setting information to the server based on the user's selection.

[1757] Step 2:

[1758] The server receives the user's initial setting information and inputs a prompt statement to the generative AI model. The prompt statement is an instruction statement for generating a question according to the user's level of understanding. The input is the initial setting information and the prompt statement, and the output is the generated question. The server generates a question using the generative AI model and sends the question to the application.

[1759] Step 3:

[1760] The application displays the questions received from the server to the user. The user inputs the answers to the displayed questions. The input is the user's answer, and the output is the answer data. The application sends the user's answer data to the server.

[1761] Step 4:

[1762] The server receives the user's answer data and inputs it into the gene...

Claims

1. a first means for transmitting initial setting information based on learning content selected by a user; a second means for generating a predetermined first prompt sentence for generating a question based on the received initial setting information, the first prompt sentence being used to output a question according to the user's level of understanding; a third means for inputting the generated first prompt sentence into a generative AI model to generate a question according to the user's level of understanding; a fourth means for displaying the generated questions corresponding to the user's level of understanding on the user's terminal; a fifth means for collecting facial expressions and tones of voice when the user is working on the generated problem and for estimating an emotional state using an emotion identification model; Including, The second means adjusts a predetermined second prompt sentence for generating a next question according to the estimated emotional state, adjusting the difficulty level to be lowered at least when the emotional state is frustration, and adjusting the difficulty level to be the same when the emotional state is joy, The third means inputs the adjusted second prompt sentence into a generative AI model to generate a next question.

2. The first means acquires answer history data and answer time data indicating answers input by the user to questions corresponding to the user's level of understanding, The second means inputs the acquired answer history data and answer time data into a generation AI model to acquire an evaluation result of the user's understanding level, the second means generates a third prompt sentence for providing a predetermined hint and explanation for the problem, based on the evaluation result and comprehension level information indicating the user's comprehension level, when the time it takes to solve the problem that the user is working on is equal to or longer than a predetermined time and the estimated emotional state is frustration or confusion; The third means inputs the generated third prompt sentence into a generative AI model to generate hints and explanations for the problem; The fourth means displays the generated hints and explanations for the problem on the user's terminal. The system of claim 1 .

Images