System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting test results and homework results of a learner; means for identifying a weak area of the learner based on the collected test results and homework results; means for generating questions related to the weak area of the learner; means for presenting the generated questions to the learner; means for analyzing the learner's answers and providing appropriate feedback; means for using a AI avatar for explanation; and means for periodically checking the learner's progress and adjusting a learning plan.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] Traditional learning methods have the problem that it is difficult to efficiently identify weak areas of each learner, which limits the learning effect. Furthermore, one-on-one tutoring is often burdensome due to its high cost and time constraints. Furthermore, traditional learning methods provide uniform content, making it difficult to provide an optimized learning experience for each learner. To solve these issues, a method was needed to provide efficient and effective learning support for individual learners. [Means for solving the problem]

[0005] The present invention provides a system that includes a means for collecting learners' test results and homework results, a means for identifying the learners' weak areas based on the collected data, a means for generating questions related to the learners' weak areas, a means for presenting the generated questions to the learners, a means for analyzing the learners' answers and providing appropriate feedback, a means for using an AI avatar to provide explanations, and a means for periodically checking the learners' progress and adjusting their learning plans. This system efficiently identifies the learners' weak areas and provides appropriate questions and explanations, thereby improving the learners' understanding. Furthermore, explanations by the AI ​​avatar attract the learners' interest and improve the learning effect through an approach different from conventional learning methods.

[0006] "Student" refers to an individual who uses the system to learn.

[0007] "Test results" refers to the grades and answers given to exams and mock tests taken by learners.

[0008] "Homework results" refers to the content and grades of the homework assigned to the learner.

[0009] "Means of collection" refers to the functions and processes for capturing and storing learners' test results and homework results as data.

[0010] "Means for identifying learners' weak areas based on data" refers to a function that analyzes collected data and, based on the results, determines the areas in which learners are having difficulty understanding or mastering.

[0011] "Means for generating questions" refers to a function that automatically creates questions related to areas in which the learner is weak.

[0012] "Means for presenting questions to a learner" refers to a function for displaying or presenting the generated questions to a learner.

[0013] "Means for analyzing answers and providing appropriate feedback" refers to the function of evaluating the content of a learner's answers and providing the learner with appropriate guidance and areas for improvement based on the results.

[0014] "Means using an AI avatar to provide explanations" refers to a function in which an avatar equipped with artificial intelligence provides explanations and guidance on problems using voice and text.

[0015] "Means to check progress and adjust learning plans" refers to the function of regularly evaluating a learner's learning progress and reviewing and optimizing the learning plan based on the results. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11]FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0017] 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.

[0018] First, the terms used in the following description will be explained.

[0019] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0020] 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.

[0021] 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.

[0022] 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.

[0023] 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."

[0024] [First embodiment]

[0025] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0026] 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.

[0027] 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).

[0028] 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.

[0029] 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.

[0030] 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.

[0031] 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.

[0032] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0037] System Overview

[0038] This invention is an AI tutoring system that provides learning support tailored to the individual needs of each learner. This system is primarily composed of a server, a terminal, and a learner (user). The server performs central data analysis and problem generation, while the terminal provides an interface for the user.

[0039] Data collection

[0040] Users input their daily homework and past test results through the device, which then sends the data to a server, which stores the data in a database for later analysis.

[0041] Identifying weak areas

[0042] Based on the collected data, the server uses AI algorithms to identify areas where learners struggle, such as analyzing test scores and homework answers to identify areas where learners frequently make mistakes and spend time.

[0043] Problem Generation

[0044] The server generates a set of problems of increasing difficulty for each identified weak area, allowing learners to tackle problems ranging from basic to advanced. The generated problems are then sent to the device.

[0045] Problem Statement

[0046] The terminal presents the questions sent from the server to the user. The user answers the questions and enters the answers into the terminal. The terminal then sends the answers to the server.

[0047] Answer analysis and feedback

[0048] The server analyzes the user's answers and determines whether they are correct or incorrect. Based on the analysis results, it generates appropriate feedback, which may include specific explanations or advice.

[0049] AI Avatar Explained

[0050] Users can select their preferred AI avatar on their device, which will then use voice and text to explain how to solve the problem and any areas they don't understand, making it easier for users to understand the information visually and audibly.

[0051] Constant progress monitoring and adjustment of study plans

[0052] The server periodically checks the user's learning progress and adjusts the learning plan as needed. For example, if a particular weak area is overcome, the server will provide new questions that focus on that area. This allows the learner to continuously improve their academic ability.

[0053] Specific examples

[0054] As a concrete example, a user sends the results of an arithmetic test to a server, which determines that the user has difficulty calculating fractions. The server then generates a simple fraction addition problem (e.g., 1 / 2 + 1 / 4) and sends it to the device. The user solves the problem and enters the answer on the device, and the server determines whether the answer is correct or incorrect. If the answer is incorrect, an AI avatar selected by the user will provide an explanation, such as, "When adding fractions, you first need to make the denominators the same."

[0055] This allows users to gain a deeper understanding and learn more efficiently than with conventional one-way learning methods.

[0056] The processing flow will be explained below.

[0057] Step 1: Collect data

[0058] The user uses the terminal to input daily homework results and past test results.

[0059] The terminal transmits the input data to the server.

[0060] The server stores the received data in a database.

[0061] Step 2: Identify your weaknesses

[0062] The server analyzes the stored data using AI algorithms.

[0063] Based on the analysis results, the server identifies areas in which the learner is weak.

[0064] For example, the server analyzes test results and homework answer trends to extract areas where learners frequently make mistakes.

[0065] Step 3: Generate the problem

[0066] The server generates questions related to the identified weaknesses.

[0067] The server uses an AI algorithm to create sets of questions that gradually increase in difficulty, starting with easy questions.

[0068] The server sends the generated questions to the terminal.

[0069] Step 4: State the problem

[0070] The terminal presents the questions sent from the server to the user.

[0071] The user answers the questions presented on the terminal.

[0072] The terminal transmits the user's answer results to the server.

[0073] Step 5: Analysis and feedback of answers

[0074] The server analyzes the user's answers and determines whether they are correct or incorrect.

[0075] The server generates appropriate feedback based on the accuracy of the answer and the time it took to answer.

[0076] The server transmits the feedback content to the terminal.

[0077] Step 6: AI Avatar Explained

[0078] Users select their preferred AI avatar on their device.

[0079] The device displays the commentary sent from the server in the form of an AI avatar.

[0080] The AI ​​avatar uses voice and text to explain how to solve the problem and key points.

[0081] Step 7: Regularly monitor your progress and adjust your study plan

[0082] The server periodically checks the user's learning progress.

[0083] The server adjusts the study plan based on the study progress.

[0084] For example, if a particular weak area is overcome, the server generates new problems for that weak area.

[0085] Example 1

[0086] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0087] Conventional learning support systems do not provide sufficient, effective learning support that meets the individual needs of learners. Furthermore, they have difficulty identifying areas in which learners are weak, providing questions that address those areas, and providing appropriate feedback. This creates the problem of ineffectively improving learners' academic abilities.

[0088] 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.

[0089] In this invention, the server includes means for collecting data on the learner's grades and assignments, means for identifying the learner's weak points in academic ability based on the collected data, and means for automatically generating questions related to the learner's weak points in academic ability, thereby enabling effective learning support and feedback to be provided to the learner according to their individual learning needs.

[0090] "Academic performance data" refers to the scores and evaluation results of tests and assignments taken by learners.

[0091] "Assignment data" refers to the answers and progress of homework and practice problems that a learner has completed.

[0092] "Academic weaknesses" refer to areas of study where a learner is particularly weak or areas where they frequently make mistakes.

[0093] "Automatically generated" refers to software or algorithms creating problems or challenges without human intervention.

[0094] "Virtual character" means a digital avatar generated by a computer program that can communicate through voice and text.

[0095] "Instructional information" refers to educational information such as explanations, advice, and feedback provided to learners.

[0096] "Machine learning algorithms" refer to computational methods and models that allow computers to learn from data and automatically improve their performance.

[0097] A "study plan" refers to the tasks and study schedule that a learner must complete each day.

[0098] This invention provides a system that provides learning support tailored to the individual needs of learners. This system is primarily composed of a server, a terminal, and a learner (user). The server performs central data analysis and question generation, while the terminal functions as an interface for the user.

[0099] System Overview

[0100] This system helps learners improve their academic ability through the following processes:

[0101] Hardware and software used

[0102] The hardware used includes servers (e.g., cloud-based servers) and user devices (e.g., tablets, PCs, etc.), while the software includes database management systems (e.g., MySQL or PostgreSQL), AI algorithms (e.g., Python's scikit-learn or TensorFlow), and generative AI models (e.g., GPT-4).

[0103] Data collection

[0104] Users use the device to input their daily homework and past test results. When the user enters this data through a dedicated application on the device or a web interface, the device sends it to the server, which stores the received data in a database.

[0105] Identifying weak areas

[0106] The server uses machine learning algorithms to analyze the stored data, such as Python's scikit-learn or TensorFlow, to identify areas where learners frequently make mistakes and spend time, allowing the server to identify weak areas in the learner's academic performance.

[0107] Problem Generation

[0108] The server generates questions using a generative AI model based on the identified weak areas. The generative AI model uses GPT-4 or similar. The server creates a prompt and sends it to the generative AI model. For example, the following prompt can be used:

[0109] Example prompt sentence:

[0110] "Your learner has difficulty with fractions. Generate a simple fraction addition problem like this: Problem: 1 / 2 + 1 / 4"

[0111] Problem Statement

[0112] The terminal receives the questions sent from the server and presents them to the user. The user answers the questions displayed on the terminal and enters the answers into the terminal.

[0113] Answer analysis and feedback

[0114] The device sends the user's answers to the server, which analyzes them and identifies correct and incorrect answers. The server then generates feedback based on the analysis results, including specific explanations and advice.

[0115] AI Avatar Explained

[0116] Users select their preferred virtual character on their device, who then uses voice (e.g., text-to-speech API) and text to explain how to solve the problem or any areas of confusion, helping users understand the information visually and audibly.

[0117] Constant progress monitoring and adjustment of study plans

[0118] The server periodically checks the learner's progress and adjusts the learning plan as needed. For example, if a particular academic weakness is overcome, the server will provide new questions that focus on that weakness. This iterative process allows the learner to continuously improve their academic performance.

[0119] In this way, the learning support system of the present invention responds to the individual needs of each learner and provides effective learning support, thereby enabling learners to improve their academic ability efficiently and continuously.

[0120] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0121] Step 1:

[0122] Users input their daily homework results and past test results through their devices. Specifically, they input this data using a dedicated application on their device or a form provided in the web interface. The input data includes information such as the content of the questions the user answered, their answers, and the time it took to complete them. The device then sends this input data to the server.

[0123] Input: Homework and test result data entered by the user into the device

[0124] Output: Data sent from the device to the server

[0125] Step 2:

[0126] The server stores the received data in a database. By using a database management system (e.g., MySQL or PostgreSQL), all training data can be managed efficiently.

[0127] Input: Data sent from the terminal

[0128] Output: Data stored in the database

[0129] Step 3:

[0130] The server uses machine learning algorithms (e.g., Python's scikit-learn or TensorFlow) to analyze the stored data. The server applies the algorithms to identify areas where learners frequently make mistakes and spend time on grades and assignments. This allows the server to identify the learner's academic weaknesses.

[0131] Input: Grade data and assignment data stored in the database

[0132] Output: Identified academic weaknesses

[0133] Step 4:

[0134] The server generates problems using a generative AI model (e.g., GPT-4) based on the identified weaknesses. The server creates a prompt like the following and sends it to the generative AI model: "The learner has difficulty with fraction calculations. Please generate a simple fraction addition problem like this: Problem: 1 / 2 + 1 / 4."

[0135] Input: Identified academic weaknesses

[0136] Output: The problem returned by the generative AI model

[0137] Step 5:

[0138] The device receives the questions sent from the server and presents them to the user. Specifically, the user can view the questions through a dedicated app on the device or a web interface. The user answers the presented questions.

[0139] Input: Question sent by server

[0140] Output: The problem presented to the user

[0141] Step 6:

[0142] The user inputs the answer into the terminal, which then transmits the answer to the server.

[0143] Input: Answer result entered by the user into the terminal

[0144] Output: Answer results sent from the device to the server

[0145] Step 7:

[0146] The server analyzes the answers and determines whether they are correct or incorrect. The server uses a Python script to compare the answers with the correct answers. The server then generates specific feedback based on the analysis results. The generated feedback includes specific explanations and advice.

[0147] Input: Answer results sent from the device

[0148] Output: Feedback information

[0149] Step 8:

[0150] Users select their preferred virtual character on their device, and the virtual character uses voice and text to explain how to solve the problem or provide explanations for any unclear points. The voice is generated using a text-to-speech API, and the information is provided visually using a 3D model created in Unity.

[0151] Input: Feedback information

[0152] Output: Explanation by virtual character

[0153] Step 9:

[0154] The server periodically checks the student's progress and adjusts the learning plan as necessary. The server periodically (e.g., weekly or monthly) analyzes the performance data and generates new questions and updates the learning plan if weak areas in the student's academic ability are overcome.

[0155] Input: Learning progress data

[0156] Output: Updated lesson plan

[0157] The above are the specific processing steps of the program of this system.

[0158] (Application example 1)

[0159] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0160] Conventional AI tutoring systems provide individual learning support based on learners' test results and homework results, but they do not provide immediate learning support in physical stores. Furthermore, the lack of an intuitive visual or audio interface makes it difficult for learners to engage and for effective learning to occur.

[0161] 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.

[0162] In this invention, the server includes means for collecting test results and homework results of learners, means for identifying areas of weakness of learners based on the collected data, and means for generating questions related to the areas of weakness of learners. This enables learners to use a visual device to send book and teaching material information to the server in real time, receive questions through the visual device, collect answers through voice input, and receive explanations from an AI avatar.

[0163] A "learner" is an individual seeking to acquire knowledge or skills.

[0164] A "visual device" is a device that provides information visually, and generally refers to smart glasses or head-mounted displays.

[0165] A "server" is a computer system that performs central processing such as collecting, analyzing, and storing data.

[0166] "Test results" refers to the results and answers given by the learner on the test.

[0167] "Homework results" refers to the grades and answers to homework submitted by the learner.

[0168] A "weak area" is an area in which a learner has particular difficulty understanding or mastering.

[0169] "Questions" refer to learning tasks or exercises that learners must answer.

[0170] "Means of collection" refers to the methods and techniques used to collect the necessary data.

[0171] "Analytical means" refers to the methods and techniques used to analyze data and derive meaningful information.

[0172] "Feedback" refers to evaluation and advice given to a learner based on their answers and behavior.

[0173] An "AI avatar" refers to a virtual character that uses artificial intelligence to provide learning assistance.

[0174] "Voice input" refers to a method in which learners input information by voice.

[0175] "Progress" refers to a learner's learning progress and achievements.

[0176] A "learning plan" refers to the learning content and schedule planned to help a learner achieve their learning goals.

[0177] This invention is an AI tutoring system that provides learning support tailored to the individual needs of each learner. This system is primarily composed of a server, a terminal, and a learner, and utilizes visual devices and voice input functions to provide instant learning support in a brick-and-mortar store.

[0178] System Overview

[0179] The server collects the learner's test results and homework results, analyzes them, and identifies the learner's weak areas. It then generates questions that address the identified weak areas and sends the questions to the learner's device or visual device. The learner receives the questions through the visual device and answers them using voice input. The answers are sent to the server, and appropriate feedback is provided based on the analysis results.

[0180] Hardware and Software

[0181] Hardware used

[0182] Visual devices: smart glasses or head-mounted displays (e.g., smart glasses)

[0183] Device: Smartphone or tablet

[0184] Server: A computer system that analyzes and stores data (e.g., cloud services).

[0185] Software used

[0186] Data collection and analysis: Python scripts

[0187] Server communication: RESTful API

[0188] Audio Description: Text-to-Speech (TTS) engine (e.g., Google Cloud Text-to-Speech)

[0189] Specific examples of explanation

[0190] Let's say a student chooses a book on "fraction calculation" at a physical store. When the student puts on the smart glasses and reads the book information, that information is sent to the server in real time. The server combines this with past learning data and determines that the student has difficulty with "fraction addition." The server then generates a simple "fraction addition" problem (e.g., 1 / 2 + 1 / 4) and sends it to the smart glasses. The student answers via voice input, and the answer is sent to the server. The server analyzes the answer, determines whether it is correct or incorrect, and sends an explanation to the smart glasses via voice and text, such as "When adding fractions, you first need to make the denominators the same," with an AI avatar providing the explanation. This allows the student to deepen their understanding on the spot.

[0191] Prompt Sentence Examples

[0192] The generative AI model can generate explanations for specific learning items by inputting the following prompts:

[0193] "setting"

[0194] Generate detailed explanations to help students understand fraction addition.

[0195] "prompt"

[0196] When adding fractions, you first need to make the denominators the same. To do this, convert the fractions so that the denominators are the lowest common multiple. For example, to calculate 1 / 2 + 1 / 4, convert 1 / 2 to 2 / 4, then calculate 2 / 4 + 1 / 4... (Detailed steps continue.)

[0197] The system allows learners to receive personalized learning support in real time within the store, providing an intuitive learning environment using visual and audio.

[0198] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0199] Step 1:

[0200] A user uses a visual device to select a book or educational material in a physical store and reads its information. The visual device then uses a camera to scan the barcode or QR code on the book or educational material to obtain the relevant information. The obtained information is sent to the server in real time. The input is the information of the book or educational material read by the visual device, and the output is the information sent to the server.

[0201] Step 2:

[0202] The server compares the received information on books and teaching materials with the learner's past test results and homework results to identify the learner's weak areas. Past learning data for each learner is stored in the database, and an AI algorithm is used to analyze the learner's weak areas. The input is information on books and teaching materials and past learning data, and the output is the identified learner's weak areas.

[0203] Step 3:

[0204] The server generates problems in stages that address the identified weak areas. An AI model is used to generate problems with increasing difficulty, from simple to advanced. The generated problems are sent to a visual device. The input is the identified weak areas and a problem generation prompt, and the output is the generated problem set.

[0205] Step 4:

[0206] The visual device presents the questions received from the server to the user. The questions are displayed on the visual device's display, and the user confirms them. Using the visual device's voice input function, the user inputs the answer to the question by voice. The input is the question sent from the server, and the output is the user's voice answer.

[0207] Step 5:

[0208] The visual device converts the user's voice response into text data and sends it to the server. The voice response is converted into text data using a Speech-to-Text (STT) engine. The input is the user's voice response, and the output is the response as text data.

[0209] Step 6:

[0210] The server analyzes the user's answers and determines whether they are correct or incorrect. An AI algorithm is used for the analysis to evaluate the accuracy of the answers. The input is the user's answer text, and the output is the analysis result of the answer.

[0211] Step 7:

[0212] The server generates appropriate feedback based on the analysis results. The appropriate feedback includes points of mistakes and explanations. The feedback content is sent to the visual device in audio and text format. The input is the analysis result of the answer, and the output is the feedback information.

[0213] Step 8:

[0214] The visual device presents the feedback received from the server to the user, and the AI ​​avatar provides explanations in voice and text. For example, an explanation such as "When adding fractions, you must first make the denominators the same." The input is the feedback information sent from the server, and the output is the explanation to the user.

[0215] 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.

[0216] System Overview

[0217] This invention is an AI tutoring system that provides learning support tailored to the individual needs of each learner, and also combines it with an emotion engine that recognizes the user's emotions and provides an optimal learning environment. This system is primarily composed of a server, a terminal, a learner (user), and an emotion engine. The server performs central data analysis and question generation, the terminal provides an interface for the user, and the emotion engine recognizes the user's emotions.

[0218] Data collection

[0219] Users input their daily homework results and past test results through the device. The device then sends this data to the server, which stores the received data in a database for later analysis. The device is also equipped with a camera and microphone, which allow the emotion engine to analyze the user's facial expressions and voice.

[0220] Identifying weak areas

[0221] Based on the stored data, the server uses AI algorithms to identify areas where learners struggle, analyzing test scores and homework answers to identify areas where learners frequently make mistakes and where they spend a lot of time on specific questions.

[0222] Problem Generation

[0223] The server generates a set of problems of increasing difficulty for the identified weak areas, starting with easy problems and gradually increasing in difficulty. The generated problems are sent to the device.

[0224] Problem Statement

[0225] The terminal presents the questions sent from the server to the user. The user answers the questions and enters the answers into the terminal. The terminal then transmits the answers to the server.

[0226] Answer analysis and feedback

[0227] The server analyzes the user's answers and determines whether they are correct or incorrect. Based on the analysis results, it generates appropriate feedback and sends it to the device. This feedback includes specific explanations and advice.

[0228] AI Avatar Explained

[0229] Users select their preferred AI avatar on their device. The emotion engine analyzes the user's facial expressions and voice to recognize their emotional state. The AI ​​avatar uses voice and text to provide explanations tailored to the user's level of understanding or detailed explanations of specific issues. It also provides reassuring and encouraging comments based on the emotion engine's results.

[0230] Constant progress monitoring and adjustment of study plans

[0231] The server periodically checks the user's learning progress and adjusts the learning plan as needed. For example, if a particular weakness is overcome, it will provide new questions that focus on that weakness. It also takes into account the learner's emotional state and creates a learning plan that is tailored to that state.

[0232] Specific examples

[0233] For example, if a user submits the results of a math test to a server, the server will determine that the user has difficulty calculating fractions. The server then generates a simple fraction addition problem (e.g., 1 / 2 + 1 / 4) and sends it to the device. While the user is solving the problem, the emotion engine analyzes the user's facial expressions and voice and determines that the user is nervous. At this point, an AI avatar will provide encouragement and explanation, such as, "Relax and think about it. First, you need to make the denominators the same."

[0234] This allows users to have an optimal learning experience that is tailored to their individual feelings and level of understanding, compared to conventional one-way learning methods.

[0235] The processing flow will be explained below.

[0236] Step 1: Collect data

[0237] Users input their daily homework results and past test results into the terminal.

[0238] The terminal transmits the input data to the server.

[0239] The server stores the received data in a database.

[0240] In addition, the device's camera and microphone are used to capture the user's facial expressions and voice and send them to the emotion engine.

[0241] Step 2: Recognize emotions

[0242] The emotion engine analyzes the user's facial expressions and voice data to recognize the user's emotional state.

[0243] The recognized emotion data is sent to the server and reflected in future feedback and commentary.

[0244] Step 3: Identify your weaknesses

[0245] The server analyzes the saved homework and test results using AI algorithms.

[0246] Based on the analysis results, the server identifies areas in which the learner is weak.

[0247] For example, the server will extract areas where you frequently make mistakes or where you spend a lot of time on specific problems.

[0248] Step 4: Generate the problem

[0249] The server generates questions related to the identified weaknesses.

[0250] The server uses an AI algorithm to create sets of questions that gradually increase in difficulty, starting with easy questions.

[0251] The server sends the generated questions to the terminal.

[0252] Step 5: State the problem

[0253] The terminal presents the questions sent from the server to the user.

[0254] The user answers the questions presented to them.

[0255] The terminal collects the answer results and sends them to the server.

[0256] Step 6: Analysis and feedback of answers

[0257] The server analyzes the user's answers and determines whether they are correct or incorrect.

[0258] The server generates appropriate feedback based on the accuracy of the answer and the time it took to answer.

[0259] The server transmits the generated feedback content to the terminal.

[0260] Step 7: AI Avatar Explained

[0261] Users select their preferred AI avatar on their device.

[0262] The terminal receives the result of the emotion engine and obtains commentary content based on the user's emotional state from the server.

[0263] The AI ​​avatar uses voice and text to explain how to solve the problem and key points.

[0264] For example, if the user is nervous, the AI ​​avatar will offer encouraging comments such as, "Relax and think about it. First, we need to make the denominators the same."

[0265] Step 8: Check your progress and adjust your study plan

[0266] The server periodically checks the user's learning progress.

[0267] The server adjusts the learning plan based on the learning progress.

[0268] Once a particular weakness has been overcome, the server provides new problems that focus on that weakness.

[0269] The server also takes into account the recognized emotion data to provide the user with an optimal learning environment.

[0270] Example 2

[0271] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0272] Conventional learning support systems rely on one-way information transmission and are unable to take into account the individual learner's level of understanding or emotional state in a timely manner. As a result, appropriate feedback based on the learner's weak areas or level of understanding is not provided, making it difficult to provide effective learning support. In addition, it is difficult to provide a learning environment that takes into account the learner's emotional state, which is likely to reduce motivation to learn.

[0273] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for collecting the learner's assessment results, means for identifying the learner's learning difficulties based on the collected data, means for generating assignments related to the learner's learning difficulties, means for presenting the generated assignments to the learner, means for analyzing the learner's answers and providing appropriate feedback, means for using an AI character, means for periodically checking the learner's progress and adjusting the learning plan, means for analyzing the learner's emotional state, and means for adjusting the feedback based on the analysis results. This makes it possible to provide an optimal learning environment according to the individual learner's level of understanding and emotional state, thereby enabling effective and motivating learning support.

[0274] "Learner's assessment results" refers to information that indicates the results of learning and performance, such as test results and homework results that the learner has completed.

[0275] "Means of collection" refers to the technical methods, devices, and software used to collect learner assessment results and other necessary data.

[0276] "Learning difficulties" refer to areas where a learner has a lack of understanding or makes many mistakes in a particular academic subject or problem.

[0277] "Means of identification" refers to technical methods, devices, and software that analyze and extract learners' learning difficulties based on collected data.

[0278] "Assignments" are learning items such as questions and exercises that are generated based on the learner's learning difficulties.

[0279] "Generation means" refers to technical methods, devices, or software that automatically generate tasks related to learning difficulties.

[0280] "Presentation means" refers to the technical methods, devices, and software used to visually and audibly present the generated assignment to the learner.

[0281] "Means for analyzing answers" refers to technical methods, devices, and software for analyzing answers provided by learners and determining whether they are correct or incorrect.

[0282] "Feedback" refers to information such as explanations, advice, and instructions for retrying that are provided to learners based on the analyzed answers.

[0283] An "artificial intelligence character" is a computer-generated character used to provide feedback and commentary through interaction with a user.

[0284] "Means for regularly checking learner progress" refers to technological methods, devices, or software that periodically evaluate and analyze learner progress.

[0285] "Means for adjusting the learning plan" refers to technological methods, devices, and software for adjusting the learning content and pace to suit the learner based on the results of regular progress checks.

[0286] "Emotional state" indicates the user's psychological and emotional state and is determined based on data such as facial expressions and voice.

[0287] "Means of analysis" refers to the technical methods, devices, and software that analyze the data necessary to determine emotional state.

[0288] "Means for adjusting feedback based on the analysis results" refers to technical methods, devices, or software that adjust the content and format of the feedback provided based on the results of sentiment analysis.

[0289] This invention is an AI tutoring system that provides learning support tailored to the individual needs of learners, and also combines it with an emotion engine that recognizes the user's emotions and provides an optimal learning environment. This system is primarily composed of a server, a terminal, a learner (user), and an emotion engine. Each component is described in detail below.

[0290] Data collection

[0291] Users use their devices to input learning data, including daily homework results and past test results. The devices first store this data locally and then send it to the server. The devices used include, for example, personal computers and tablets. The devices are also equipped with cameras and microphones, which capture the user's facial expressions and voice. The captured facial and voice data is analyzed in real time by the emotion engine.

[0292] Data storage and analysis

[0293] The server stores the received learning data in a database, which makes it available for later analysis. Based on this data, the server uses an AI algorithm to identify the learner's learning difficulties. For example, from math test data, the server can identify that the learner has difficulty calculating fractions.

[0294] Issue generation and presentation

[0295] The server generates tasks that address the identified learning difficulties. These tasks are designed to increase in difficulty in stages, adjusting to include beginner level problems and gradually more difficult problems. The generated tasks are sent to the terminal, which then presents them to the user. The user answers the presented tasks and re-enters the answers into the terminal. The entered answers are then sent from the terminal to the server.

[0296] Answer analysis and feedback

[0297] The server analyzes the user's answers and determines whether they are correct or incorrect. Based on the analysis results, it generates appropriate feedback and sends it to the device. The feedback includes specific explanations and advice. For example, if the answer is incorrect, the server provides advice such as, "You made a mistake, but you need to make the denominators the same."

[0298] Emotion Engine and AI Avatar

[0299] Users select their preferred AI character on the device. The emotion engine analyzes the user's facial expressions and voice to recognize their emotional state in real time. The AI ​​character uses voice and text to provide explanations and feedback based on the user's level of understanding. For example, if the character determines that the user is nervous, it will make encouraging comments such as, "Relax and try again."

[0300] Check your progress and adjust your study plan

[0301] The server periodically monitors the user's learning progress and adjusts the learning plan as needed. It generates new assignments based on the user's progress and provides feedback that focuses on new learning difficulties or areas that need improvement. It also takes into account the learner's emotional state and builds a learning plan that is tailored to that state.

[0302] Specific examples

[0303] For example, if a user sends the results of an arithmetic test to a server, the server analyzes the results and determines that the user has difficulty calculating fractions. The server then generates a simple fraction addition problem (e.g., 1 / 2 + 1 / 4) and sends it to the device. While the user solves the problem and the answer is sent to the server, the emotion engine analyzes the user's facial expressions and voice and determines that the user is nervous. At this point, an AI character provides encouragement and explanation, such as, "Relax and think about it. First, you need to make the denominators the same."

[0304] Examples of input prompts for a generative AI model include:

[0305] Describe a learning support system in which a user inputs past test data, a server extracts learning difficulties from that data, and generates tasks. An emotion engine analyzes the user's facial expressions and voice to provide real-time feedback.

[0306] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0307] Step 1: Data entry

[0308] Users use a device to input homework results and past test results. This input data may include answers in text format or multiple choice format. Specifically, the user enters data into an input box on the device screen and presses the "Submit" button. The input data includes homework scores, test scores, answers, etc., and this becomes the basic data for the system.

[0309] Input: Homework results, test results

[0310] Output: Save to local database, move data to next step

[0311] Step 2: Send data

[0312] The device sends the homework and test result data entered by the user to the server. This operation is performed using the device's communication function and is executed automatically by the user's operation. For example, when the send button is pressed, the device packs the data and sends it to the server's API endpoint.

[0313] Input: Data from the local database

[0314] Output: Upload data to the server

[0315] Step 3: Saving to the database

[0316] The server analyzes the received homework and test results data and stores them in a database. The data stored includes the score for each question, the answer content, and whether the answer was correct or incorrect. This operation is basically performed by issuing an insert query to the database.

[0317] Input: Received data

[0318] Output: Save data to database

[0319] Step 4: Collecting emotion data

[0320] The device's camera and microphone are used to collect the user's facial expressions and voice data in real time, and the collected data is sent to the emotion engine.

[0321] Input: User's facial expression and voice data

[0322] Output: Sending data to the emotion engine

[0323] Step 5: Analyze the sentiment data

[0324] The emotion engine analyzes the collected facial and voice data to determine the user's emotional state. It uses facial recognition and voice analysis algorithms. Specifically, if the user is smiling, it identifies the emotion as "relaxed," while if the user is smiling, it identifies the emotion as "tense."

[0325] Input: facial expression and voice data

[0326] Output: Emotion analysis results

[0327] Step 6: Identifying learning difficulties

[0328] The server analyzes the stored learning data using an AI algorithm to identify the learner's learning difficulties. For example, based on test data, it identifies that the learner has difficulty calculating fractions. During this process, the correct answer rate and answer time for each problem are also used as part of the analysis.

[0329] Input: Training data stored in a database

[0330] Output: Identification of learning difficulties

[0331] Step 7: Generate an issue

[0332] The server generates a set of increasingly difficult problems for each identified learning difficulty, starting with easy problems (e.g., basic addition of fractions) and gradually increasing in difficulty (e.g., subtraction of mixed fractions). This problem generation is done using an AI problem generation algorithm.

[0333] Input: Results of identifying learning difficulties

[0334] Output: The generated set of issues

[0335] Step 8: State the problem

[0336] The terminal presents the task sent from the server to the user. The displayed task consists of a text question and an answer input box. The terminal provides an interface for the user to perform the operations necessary to solve the task.

[0337] Input: Issue data from the server

[0338] Output: Present the task to the user

[0339] Step 9: Enter and submit your answers

[0340] The user inputs the answer to the task presented on the terminal and presses the send button to send the answer to the server. This data includes the user's answer content, answer time, etc.

[0341] Input: User's answer

[0342] Output: Send answer data to the server

[0343] Step 10: Analyze the solution

[0344] The server analyzes the received answer data and determines whether the answer is correct or incorrect. It checks the validity of the answer using an AI algorithm, and if the answer is correct, it will state that, and if it is incorrect, it will explain the reason.

[0345] Input: User's answer data

[0346] Output: Analysis results (correct / incorrect)

[0347] Step 11: Generate feedback

[0348] The server generates appropriate feedback based on the analysis results, including specific advice such as "Well done!" if the answer is correct, or "Sorry, try calculating it again with the same denominator" if the answer is incorrect.

[0349] Input: Answer analysis result

[0350] Output: Generated feedback

[0351] Step 12: AI Avatar Commentary

[0352] Users select their preferred AI avatar on their device, and the emotion engine analyzes the user's facial expressions and voice. Based on this, the AI ​​avatar provides the user with commentary and encouragement in voice and text. For example, if it determines that the user is nervous, it will say, "Relax and think about it."

[0353] Input: Sentiment analysis results, feedback content

[0354] Output: Audio and text commentary by an AI avatar

[0355] Step 13: Check and adjust your progress

[0356] The server periodically checks the user's learning progress and creates a new learning plan, taking into account past learning data, feedback, and emotional state. Based on the progress data, the server provides new assignments and feedback, creating an optimal learning environment for the user.

[0357] Input: Learning data and emotion data stored in the database

[0358] Output: New lesson plan, next set of assignments

[0359] (Application example 2)

[0360] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0361] Conventional learning support systems provide feedback based on individual learners' progress and level of understanding, but they are unable to take into account the learner's emotional state. As a result, if a learner feels nervous or stressed, appropriate support may not be provided, which could result in a decrease in learning effectiveness. To address this issue, a system is needed that provides optimal feedback according to the learner's emotional state.

[0362] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting the learner's test results and homework results, means for identifying the learner's weak areas based on the collected data, means for generating questions related to the learner's weak areas, means for presenting the generated questions to the learner, means for analyzing the learner's answers and providing appropriate feedback, means for using an AI avatar to provide explanations, means for periodically checking the learner's progress and adjusting the learning plan, means for determining the user's emotional state using an emotion recognition engine, and means for adjusting the learning content and feedback based on the user's emotional state. This enables optimal learning support tailored to the individual learning needs of the learner while taking the learner's emotional state into consideration.

[0363] "Student" refers to an individual who is engaged in learning activities using a learning support system.

[0364] "Means for collecting test results and homework results" refers to devices or software that input learners' daily learning results into the system for recording and storage.

[0365] A "means for identifying areas of difficulty" is a device or software that analyzes the collected data to detect areas or fields in which the learner has difficulty understanding.

[0366] A "means for generating questions" is a device or software that automatically generates appropriate study questions based on identified areas of weakness.

[0367] A "means for presenting questions" is a device or software that presents the generated questions to the learner visually or audibly.

[0368] "Means for analyzing answers and providing appropriate feedback" refers to a device or software that evaluates the learner's answers, takes the results into consideration, and returns useful advice or explanations.

[0369] "Means using AI avatars" refers to devices or software that use artificial intelligence technology to utilize virtual characters that provide answers and explanations to learners.

[0370] A "means for checking learning progress and adjusting learning plans" is a device or software that periodically evaluates a learner's learning progress and modifies or improves the learning plan based on that progress.

[0371] An "emotion recognition engine" is a device or software equipped with artificial intelligence technology that analyzes a learner's facial expressions and voice to determine their emotional state.

[0372] The "means for adjusting learning content and feedback based on emotional state" refers to a device or software that provides appropriate learning content and feedback according to the learner's emotional state based on information obtained from an emotion recognition engine.

[0373] This invention is an AI tutoring system to support employee training in brick-and-mortar stores, recognizing user emotions to provide an optimal learning environment. The system is primarily composed of a server, a terminal, users, and an emotion recognition engine. The server performs central data analysis and generates questions, the terminal provides an interface for users, and the emotion recognition engine recognizes user emotions.

[0374] Hardware and software configuration:

[0375] 1. Server: Mainly responsible for data analysis, question generation, and learning progress management. Software used includes AI algorithms, database management systems, and web servers (e.g., Apache).

[0376] 2. Device: A smartphone is used to provide the user interface, present learning content, and display feedback. The device is equipped with a camera and microphone, which are used to collect the user's facial expressions and voice. The recommended hardware is a smart device with a camera (e.g., iPhone, Android smartphone). The software used includes the UI framework used by the application (e.g., React Native, Flutter, etc.).

[0377] 3. Emotion Recognition Engine: Analyzes the learner's facial expressions and voice to determine their emotional state in real time. The software used includes machine learning libraries such as TensorFlow and OpenCV.

[0378] Explanation of program operation:

[0379] First, a user (such as a new employee) uses a device to log in to a training application. The application retrieves the user's learning history from the server and presents the next learning content to be tackled. This learning content is generated by an AI algorithm on the server based on the collected data. While the user is learning, the device's built-in camera and microphone collect the user's facial expressions and voice in real time. This data is analyzed by an emotion recognition engine to determine the user's emotional state.

[0380] Once the user's facial and voice data is collected, it is passed to an emotion recognition engine. This engine uses TensorFlow and OpenCV to analyze the user's emotional state in real time. For example, if the application determines that the user is nervous, it will display a message such as, "Relax, and try to smile first." The learning content and feedback are then dynamically adjusted to provide appropriate support tailored to the user's emotional state.

[0381] Examples:

[0382] Consider a scenario where a new employee is undergoing customer service skill training. When the user launches the application and starts the training module, the server generates the following scenario based on the user's progress data:

[0383] Example prompt sentence:

[0384] "Think about the ideal way to serve customers in the following situation: When there are many customers in the store during busy hours."

[0385] While the user is learning based on the prompts, the emotion recognition engine analyzes the user's emotions and provides relaxing messages such as "Relax, try to smile first" if the user is nervous. In this way, it is possible to provide support tailored to the needs of each individual user while maximizing the learning effect.

[0386] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0387] Step 1:

[0388] The user logs in to the training application using a device (smartphone). The user's ID and password are required as input, and are verified by the user authentication server. The user's learning history and current progress are obtained as output. If authentication is successful, the application's home screen is displayed, showing the next learning content to be tackled (e.g., customer service skills training).

[0389] Step 2:

[0390] The server receives the user's login information and learning history and generates the next learning content. The user's past learning history stored in the learning database is used as input. Using data analysis and AI algorithms, questions and practice scenarios are generated based on the individual's learning progress. The generated learning content is sent to the device as output.

[0391] Step 3:

[0392] The terminal receives the learning content sent from the server and presents it to the user. The learning content data from the server is used as input. When the user logs in to their account and starts a training session, the learning content (e.g., a customer service scenario) is displayed on the screen. As output, the learning content is presented on the display screen.

[0393] Step 4:

[0394] The user works through exercises and scenarios according to the learning content presented. The user's answers and reactions are entered as input. The device's built-in camera and microphone also collect the user's facial expressions and voice data. This data is sent to the emotion recognition engine in real time.

[0395] Step 5:

[0396] The emotion recognition engine analyzes the user's facial expressions and voice data in real time to determine the user's emotional state. Image data and voice data collected from the camera and microphone are used as input. The emotional state (e.g., nervous, relaxed, etc.) is identified using TensorFlow and OpenCV. The determined emotional state is sent to the device as output.

[0397] Step 6:

[0398] The server provides appropriate feedback to the user based on the determined emotional state. As input, the results of the emotion recognition engine and the user's learning progress data are used. Using an AI algorithm, advice, explanations, and encouraging messages are generated according to the user's emotional state. As output, the generated feedback is sent to the device.

[0399] Step 7:

[0400] The terminal displays the feedback sent from the server to the user. As input, the feedback data from the server is used. A text message or a voice message according to the emotional state is displayed on the user's screen. As output, the user receives feedback through voice and text.

[0401] Step 8:

[0402] The user continues learning based on the feedback and takes on new challenges. The answers and reactions after receiving the feedback are used as inputs, which dynamically adjust the next learning content. The output is the learning progress sent back to the server, and the whole system repeats the learning cycle.

[0403] 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.

[0404] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0405] 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.

[0406] [Second embodiment]

[0407] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0408] 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.

[0409] 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).

[0410] 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.

[0411] 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.

[0412] 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).

[0413] 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.

[0414] 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.

[0415] 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.

[0416] 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.

[0417] 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.

[0418] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0419] System Overview

[0420] This invention is an AI tutoring system that provides learning support tailored to the individual needs of each learner. This system is primarily composed of a server, a terminal, and a learner (user). The server performs central data analysis and problem generation, while the terminal provides an interface for the user.

[0421] Data collection

[0422] Users input their daily homework and past test results through the device, which then sends the data to a server, which stores the data in a database for later analysis.

[0423] Identifying weak areas

[0424] Based on the collected data, the server uses AI algorithms to identify areas where learners struggle, such as analyzing test scores and homework answers to identify areas where learners frequently make mistakes and spend time.

[0425] Problem Generation

[0426] The server generates a set of problems of increasing difficulty for each identified weak area, allowing learners to tackle problems ranging from basic to advanced. The generated problems are then sent to the device.

[0427] Problem Statement

[0428] The terminal presents the questions sent from the server to the user. The user answers the questions and enters the answers into the terminal. The terminal then sends the answers to the server.

[0429] Answer analysis and feedback

[0430] The server analyzes the user's answers and determines whether they are correct or incorrect. Based on the analysis results, it generates appropriate feedback, which may include specific explanations or advice.

[0431] AI Avatar Explained

[0432] Users can select their preferred AI avatar on their device, which will then use voice and text to explain how to solve the problem and any areas they don't understand, making it easier for users to understand the information visually and audibly.

[0433] Constant progress monitoring and adjustment of study plans

[0434] The server periodically checks the user's learning progress and adjusts the learning plan as needed. For example, if a particular weak area is overcome, the server will provide new questions that focus on that area. This allows the learner to continuously improve their academic ability.

[0435] Specific examples

[0436] As a concrete example, a user sends the results of an arithmetic test to a server, which determines that the user has difficulty calculating fractions. The server then generates a simple fraction addition problem (e.g., 1 / 2 + 1 / 4) and sends it to the device. The user solves the problem and enters the answer on the device, and the server determines whether the answer is correct or incorrect. If the answer is incorrect, an AI avatar selected by the user will provide an explanation, such as, "When adding fractions, you first need to make the denominators the same."

[0437] This allows users to gain a deeper understanding and learn more efficiently than with conventional one-way learning methods.

[0438] The processing flow will be explained below.

[0439] Step 1: Collect data

[0440] The user uses the terminal to input daily homework results and past test results.

[0441] The terminal transmits the input data to the server.

[0442] The server stores the received data in a database.

[0443] Step 2: Identify your weaknesses

[0444] The server analyzes the stored data using AI algorithms.

[0445] Based on the analysis results, the server identifies areas in which the learner is weak.

[0446] For example, the server analyzes test results and homework answer trends to extract areas where learners frequently make mistakes.

[0447] Step 3: Generate the problem

[0448] The server generates questions related to the identified weaknesses.

[0449] The server uses an AI algorithm to create sets of questions that gradually increase in difficulty, starting with easy questions.

[0450] The server sends the generated questions to the terminal.

[0451] Step 4: State the problem

[0452] The terminal presents the questions sent from the server to the user.

[0453] The user answers the questions presented on the terminal.

[0454] The terminal transmits the user's answer results to the server.

[0455] Step 5: Analysis and feedback of answers

[0456] The server analyzes the user's answers and determines whether they are correct or incorrect.

[0457] The server generates appropriate feedback based on the accuracy of the answer and the time it took to answer.

[0458] The server transmits the feedback content to the terminal.

[0459] Step 6: AI Avatar Explained

[0460] Users select their preferred AI avatar on their device.

[0461] The device displays the commentary sent from the server in the form of an AI avatar.

[0462] The AI ​​avatar uses voice and text to explain how to solve the problem and key points.

[0463] Step 7: Regularly monitor your progress and adjust your study plan

[0464] The server periodically checks the user's learning progress.

[0465] The server adjusts the study plan based on the study progress.

[0466] For example, if a particular weak area is overcome, the server generates new problems for that weak area.

[0467] Example 1

[0468] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0469] Conventional learning support systems do not provide sufficient, effective learning support that meets the individual needs of learners. Furthermore, they have difficulty identifying areas in which learners are weak, providing questions that address those areas, and providing appropriate feedback. This creates the problem of ineffectively improving learners' academic abilities.

[0470] 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.

[0471] In this invention, the server includes means for collecting data on the learner's grades and assignments, means for identifying the learner's weak points in academic ability based on the collected data, and means for automatically generating questions related to the learner's weak points in academic ability, thereby enabling effective learning support and feedback to be provided to the learner according to their individual learning needs.

[0472] "Academic performance data" refers to the scores and evaluation results of tests and assignments taken by learners.

[0473] "Assignment data" refers to the answers and progress of homework and practice problems that a learner has completed.

[0474] "Academic weaknesses" refer to areas of study where a learner is particularly weak or areas where they frequently make mistakes.

[0475] "Automatically generated" refers to software or algorithms creating problems or challenges without human intervention.

[0476] "Virtual character" means a digital avatar generated by a computer program that can communicate through voice and text.

[0477] "Instructional information" refers to educational information such as explanations, advice, and feedback provided to learners.

[0478] "Machine learning algorithms" refer to computational methods and models that allow computers to learn from data and automatically improve their performance.

[0479] A "study plan" refers to the tasks and study schedule that a learner must complete each day.

[0480] This invention provides a system that provides learning support tailored to the individual needs of learners. This system is primarily composed of a server, a terminal, and a learner (user). The server performs central data analysis and question generation, while the terminal functions as an interface for the user.

[0481] System Overview

[0482] This system helps learners improve their academic ability through the following processes:

[0483] Hardware and software used

[0484] The hardware used includes servers (e.g., cloud-based servers) and user devices (e.g., tablets, PCs, etc.), while the software includes database management systems (e.g., MySQL or PostgreSQL), AI algorithms (e.g., Python's scikit-learn or TensorFlow), and generative AI models (e.g., GPT-4).

[0485] Data collection

[0486] Users use the device to input their daily homework and past test results. When the user enters this data through a dedicated application on the device or a web interface, the device sends it to the server, which stores the received data in a database.

[0487] Identifying weak areas

[0488] The server uses machine learning algorithms to analyze the stored data, such as Python's scikit-learn or TensorFlow, to identify areas where learners frequently make mistakes and spend time, allowing the server to identify weak areas in the learner's academic performance.

[0489] Problem Generation

[0490] The server generates questions using a generative AI model based on the identified weak areas. The generative AI model uses GPT-4 or similar. The server creates a prompt and sends it to the generative AI model. For example, the following prompt can be used:

[0491] Example prompt sentence:

[0492] "Your learner has difficulty with fractions. Generate a simple fraction addition problem like this: Problem: 1 / 2 + 1 / 4"

[0493] Problem Statement

[0494] The terminal receives the questions sent from the server and presents them to the user. The user answers the questions displayed on the terminal and enters the answers into the terminal.

[0495] Answer analysis and feedback

[0496] The device sends the user's answers to the server, which analyzes them and identifies correct and incorrect answers. The server then generates feedback based on the analysis results, including specific explanations and advice.

[0497] AI Avatar Explained

[0498] Users select their preferred virtual character on their device, who then uses voice (e.g., text-to-speech API) and text to explain how to solve the problem or any areas of confusion, helping users understand the information visually and audibly.

[0499] Constant progress monitoring and adjustment of study plans

[0500] The server periodically checks the learner's progress and adjusts the learning plan as needed. For example, if a particular academic weakness is overcome, the server will provide new questions that focus on that weakness. This iterative process allows the learner to continuously improve their academic performance.

[0501] In this way, the learning support system of the present invention responds to the individual needs of each learner and provides effective learning support, thereby enabling each learner to improve their academic ability efficiently and continuously.

[0502] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0503] Step 1:

[0504] Users input their daily homework results and past test results through their devices. Specifically, they input this data using a dedicated application on their device or a form provided in the web interface. The input data includes information such as the content of the questions the user answered, their answers, and the time it took to complete them. The device then sends this input data to the server.

[0505] Input: Homework and test result data entered by the user into the device

[0506] Output: Data sent from the device to the server

[0507] Step 2:

[0508] The server stores the received data in a database. By using a database management system (e.g., MySQL or PostgreSQL), all training data can be managed efficiently.

[0509] Input: Data sent from the terminal

[0510] Output: Data stored in the database

[0511] Step 3:

[0512] The server uses machine learning algorithms (e.g., Python's scikit-learn or TensorFlow) to analyze the stored data. The server applies the algorithms to identify areas where learners frequently make mistakes and spend time on grades and assignments. This allows the server to identify the learner's academic weaknesses.

[0513] Input: Grade data and assignment data stored in the database

[0514] Output: Identified academic weaknesses

[0515] Step 4:

[0516] The server uses a generative AI model (e.g., GPT-4) to generate problems based on the identified weaknesses. The server creates a prompt like the following and sends it to the generative AI model: "The learner has difficulty with fraction calculations. Please generate a simple fraction addition problem like this: Problem: 1 / 2 + 1 / 4."

[0517] Input: Identified academic weaknesses

[0518] Output: The problem returned by the generative AI model

[0519] Step 5:

[0520] The device receives the questions sent from the server and presents them to the user. Specifically, the user can view the questions through a dedicated app on the device or a web interface. The user answers the presented questions.

[0521] Input: Question sent by server

[0522] Output: The problem presented to the user

[0523] Step 6:

[0524] The user inputs the answer into the terminal, which then transmits the answer to the server.

[0525] Input: Answer result entered by the user into the terminal

[0526] Output: Answer results sent from the device to the server

[0527] Step 7:

[0528] The server analyzes the answers and determines whether they are correct or incorrect. The server uses a Python script to compare the answers with the correct answers. The server then generates specific feedback based on the analysis results. The generated feedback includes specific explanations and advice.

[0529] Input: Answer results sent from the device

[0530] Output: Feedback information

[0531] Step 8:

[0532] Users select their preferred virtual character on their device, and the virtual character uses voice and text to explain how to solve the problem or provide explanations for any unclear points. The voice is generated using a text-to-speech API, and the information is provided visually using a 3D model created in Unity.

[0533] Input: Feedback information

[0534] Output: Explanation by virtual character

[0535] Step 9:

[0536] The server periodically checks the student's progress and adjusts the learning plan as necessary. The server periodically (e.g., weekly or monthly) analyzes the performance data and generates new questions and updates the learning plan if the student's academic weaknesses are overcome.

[0537] Input: Learning progress data

[0538] Output: Updated lesson plan

[0539] The above are the specific processing steps of the program of this system.

[0540] (Application example 1)

[0541] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0542] Conventional AI tutoring systems provide individual learning support based on learners' test results and homework results, but they do not provide immediate learning support in physical stores. Furthermore, the lack of an intuitive visual or audio interface makes it difficult for learners to engage and for effective learning to occur.

[0543] 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.

[0544] In this invention, the server includes means for collecting test results and homework results of learners, means for identifying areas of weakness of learners based on the collected data, and means for generating questions related to the areas of weakness of learners. This enables learners to use a visual device to send book and teaching material information to the server in real time, receive questions through the visual device, collect answers through voice input, and receive explanations from an AI avatar.

[0545] A "learner" is an individual seeking to acquire knowledge or skills.

[0546] A "visual device" is a device that provides information visually, and generally refers to smart glasses or head-mounted displays.

[0547] A "server" is a computer system that performs central processing such as collecting, analyzing, and storing data.

[0548] "Test results" refers to the results and answers given by the learner on the test.

[0549] "Homework results" refers to the grades and answers to homework submitted by the learner.

[0550] A "weak area" is an area in which a learner has particular difficulty understanding or mastering.

[0551] "Questions" refer to learning tasks or exercises that learners must answer.

[0552] "Means of collection" refers to the methods and techniques used to collect the necessary data.

[0553] "Analytical means" refers to the methods and techniques used to analyze data and derive meaningful information.

[0554] "Feedback" refers to evaluation and advice given to a learner based on their answers and behavior.

[0555] An "AI avatar" refers to a virtual character that uses artificial intelligence to provide learning assistance.

[0556] "Voice input" refers to a method in which learners input information by voice.

[0557] "Progress" refers to a learner's learning progress and achievements.

[0558] A "learning plan" refers to the learning content and schedule planned to help a learner achieve their learning goals.

[0559] This invention is an AI tutoring system that provides learning support tailored to the individual needs of each learner. This system is primarily composed of a server, a terminal, and a learner, and utilizes visual devices and voice input functions to provide instant learning support in a brick-and-mortar store.

[0560] System Overview

[0561] The server collects the learner's test results and homework results, analyzes them, and identifies the learner's weak areas. It then generates questions that address the identified weak areas and sends the questions to the learner's device or visual device. The learner receives the questions through the visual device and answers them using voice input. The answers are sent to the server, and appropriate feedback is provided based on the analysis results.

[0562] Hardware and Software

[0563] Hardware used

[0564] Visual devices: smart glasses or head-mounted displays (e.g., smart glasses)

[0565] Device: Smartphone or tablet

[0566] Server: A computer system that analyzes and stores data (e.g., cloud services).

[0567] Software used

[0568] Data collection and analysis: Python scripts

[0569] Server communication: RESTful API

[0570] Audio Description: Text-to-Speech (TTS) engine (e.g., Google Cloud Text-to-Speech)

[0571] Specific examples of explanation

[0572] Let's say a student chooses a book on "fraction calculation" at a physical store. When the student puts on the smart glasses and reads the book information, that information is sent to the server in real time. The server combines this with past learning data and determines that the student has difficulty with "fraction addition." The server then generates a simple "fraction addition" problem (e.g., 1 / 2 + 1 / 4) and sends it to the smart glasses. The student answers via voice input, and the answer is sent to the server. The server analyzes the answer, determines whether it is correct or incorrect, and sends an explanation to the smart glasses via voice and text, such as "When adding fractions, you first need to make the denominators the same," with an AI avatar providing the explanation. This allows the student to deepen their understanding on the spot.

[0573] Prompt Sentence Examples

[0574] The generative AI model can generate explanations for specific learning items by inputting the following prompts:

[0575] "setting"

[0576] Generate detailed explanations to help students understand fraction addition.

[0577] "prompt"

[0578] When adding fractions, you first need to make the denominators the same. To do this, convert the fractions so that the denominators are the lowest common multiple. For example, to calculate 1 / 2 + 1 / 4, convert 1 / 2 to 2 / 4, then calculate 2 / 4 + 1 / 4... (Detailed steps continue.)

[0579] The system allows learners to receive personalized learning support in real time within the store, providing an intuitive learning environment using visual and audio.

[0580] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0581] Step 1:

[0582] A user uses a visual device to select a book or educational material in a physical store and reads its information. The visual device then uses a camera to scan the barcode or QR code on the book or educational material to obtain the relevant information. The obtained information is sent to the server in real time. The input is the information of the book or educational material read by the visual device, and the output is the information sent to the server.

[0583] Step 2:

[0584] The server compares the received information on books and teaching materials with the learner's past test results and homework results to identify the learner's weak areas. Past learning data for each learner is stored in the database, and an AI algorithm is used to analyze the learner's weak areas. The input is information on books and teaching materials and past learning data, and the output is the identified learner's weak areas.

[0585] Step 3:

[0586] The server generates problems in stages that address the identified weak areas. An AI model is used to generate problems with increasing difficulty, from simple to advanced. The generated problems are sent to a visual device. The input is the identified weak areas and a problem generation prompt, and the output is the generated problem set.

[0587] Step 4:

[0588] The visual device presents the questions received from the server to the user. The questions are displayed on the visual device's display, and the user confirms them. Using the visual device's voice input function, the user inputs the answer to the question by voice. The input is the question sent from the server, and the output is the user's voice answer.

[0589] Step 5:

[0590] The visual device converts the user's voice response into text data and sends it to the server. The voice response is converted into text data using a Speech-to-Text (STT) engine. The input is the user's voice response, and the output is the response as text data.

[0591] Step 6:

[0592] The server analyzes the user's answers and determines whether they are correct or incorrect. An AI algorithm is used for the analysis to evaluate the accuracy of the answers. The input is the user's answer text, and the output is the analysis result of the answer.

[0593] Step 7:

[0594] The server generates appropriate feedback based on the analysis results. The appropriate feedback includes points of mistakes and explanations. The feedback content is sent to the visual device in audio and text format. The input is the analysis result of the answer, and the output is the feedback information.

[0595] Step 8:

[0596] The visual device presents the feedback received from the server to the user, and the AI ​​avatar provides explanations in voice and text. For example, an explanation such as "When adding fractions, you must first make the denominators the same." The input is the feedback information sent from the server, and the output is the explanation to the user.

[0597] 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.

[0598] System Overview

[0599] This invention is an AI tutoring system that provides learning support tailored to the individual needs of each learner, and also combines it with an emotion engine that recognizes the user's emotions and provides an optimal learning environment. This system is primarily composed of a server, a terminal, a learner (user), and an emotion engine. The server performs central data analysis and question generation, the terminal provides an interface for the user, and the emotion engine recognizes the user's emotions.

[0600] Data collection

[0601] Users input their daily homework results and past test results through the device. The device then sends this data to the server, which stores the received data in a database for later analysis. The device is also equipped with a camera and microphone, which allow the emotion engine to analyze the user's facial expressions and voice.

[0602] Identifying weak areas

[0603] Based on the stored data, the server uses AI algorithms to identify areas where learners struggle, analyzing test scores and homework answers to identify areas where learners frequently make mistakes and where they spend a lot of time on specific questions.

[0604] Problem Generation

[0605] The server generates a set of problems of increasing difficulty for the identified weak areas, starting with easy problems and gradually increasing in difficulty. The generated problems are sent to the device.

[0606] Problem Statement

[0607] The terminal presents the questions sent from the server to the user. The user answers the questions and enters the answers into the terminal. The terminal then sends the answers to the server.

[0608] Answer analysis and feedback

[0609] The server analyzes the user's answers and determines whether they are correct or incorrect. Based on the analysis results, it generates appropriate feedback and sends it to the device. This feedback includes specific explanations and advice.

[0610] AI Avatar Explained

[0611] Users select their preferred AI avatar on their device. The emotion engine analyzes the user's facial expressions and voice to recognize their emotional state. The AI ​​avatar uses voice and text to provide explanations tailored to the user's level of understanding or detailed explanations of specific issues. It also provides reassuring and encouraging comments based on the emotion engine's results.

[0612] Constant progress monitoring and adjustment of study plans

[0613] The server periodically checks the user's learning progress and adjusts the learning plan as needed. For example, if a particular weakness is overcome, it will provide new questions that focus on that weakness. It also takes into account the learner's emotional state and creates a learning plan that is tailored to that state.

[0614] Specific examples

[0615] For example, if a user submits the results of a math test to a server, the server will determine that the user has difficulty calculating fractions. The server then generates a simple fraction addition problem (e.g., 1 / 2 + 1 / 4) and sends it to the device. While the user is solving the problem, the emotion engine analyzes the user's facial expressions and voice and determines that the user is nervous. At this point, an AI avatar will provide encouragement and explanation, such as, "Relax and think about it. First, you need to make the denominators the same."

[0616] This allows users to have an optimal learning experience that is tailored to their individual feelings and level of understanding, compared to conventional one-way learning methods.

[0617] The processing flow will be explained below.

[0618] Step 1: Collect data

[0619] Users input their daily homework results and past test results into the terminal.

[0620] The terminal transmits the input data to the server.

[0621] The server stores the received data in a database.

[0622] In addition, the device's camera and microphone are used to capture the user's facial expressions and voice and send them to the emotion engine.

[0623] Step 2: Recognize emotions

[0624] The emotion engine analyzes the user's facial expressions and voice data to recognize the user's emotional state.

[0625] The recognized emotion data is sent to the server and reflected in future feedback and commentary.

[0626] Step 3: Identify your weaknesses

[0627] The server analyzes the saved homework and test results using AI algorithms.

[0628] Based on the analysis results, the server identifies areas in which the learner is weak.

[0629] For example, the server will extract areas where you frequently make mistakes or where you spend a lot of time on specific problems.

[0630] Step 4: Generate the problem

[0631] The server generates questions related to the identified weaknesses.

[0632] The server uses an AI algorithm to create sets of questions that gradually increase in difficulty, starting with easy questions.

[0633] The server sends the generated questions to the terminal.

[0634] Step 5: State the problem

[0635] The terminal presents the questions sent from the server to the user.

[0636] The user answers the questions presented to them.

[0637] The terminal collects the answer results and sends them to the server.

[0638] Step 6: Analysis and feedback of answers

[0639] The server analyzes the user's answers and determines whether they are correct or incorrect.

[0640] The server generates appropriate feedback based on the accuracy of the answer and the time it took to answer.

[0641] The server transmits the generated feedback content to the terminal.

[0642] Step 7: AI Avatar Explained

[0643] Users select their preferred AI avatar on their device.

[0644] The terminal receives the result of the emotion engine and obtains commentary content based on the user's emotional state from the server.

[0645] The AI ​​avatar uses voice and text to explain how to solve the problem and key points.

[0646] For example, if the user is nervous, the AI ​​avatar will offer encouraging comments such as, "Relax and think about it. First, we need to make the denominators the same."

[0647] Step 8: Check your progress and adjust your study plan

[0648] The server periodically checks the user's learning progress.

[0649] The server adjusts the learning plan based on the learning progress.

[0650] Once a particular weakness has been overcome, the server provides new problems that focus on that weakness.

[0651] The server also takes into account the recognized emotion data to provide the user with an optimal learning environment.

[0652] Example 2

[0653] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0654] Conventional learning support systems rely on one-way information transmission and are unable to take into account the individual learner's level of understanding or emotional state in a timely manner. As a result, appropriate feedback based on the learner's weak areas or level of understanding is not provided, making it difficult to provide effective learning support. In addition, it is difficult to provide a learning environment that takes into account the learner's emotional state, which is likely to reduce motivation to learn.

[0655] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for collecting the learner's assessment results, means for identifying the learner's learning difficulties based on the collected data, means for generating assignments related to the learner's learning difficulties, means for presenting the generated assignments to the learner, means for analyzing the learner's answers and providing appropriate feedback, means for using an AI character, means for periodically checking the learner's progress and adjusting the learning plan, means for analyzing the learner's emotional state, and means for adjusting the feedback based on the analysis results. This makes it possible to provide an optimal learning environment according to the individual learner's level of understanding and emotional state, thereby enabling effective and motivating learning support.

[0656] "Learner's assessment results" refers to information that indicates the results of learning and performance, such as test results and homework results that the learner has completed.

[0657] "Means of collection" refers to the technical methods, devices, and software used to collect learner assessment results and other necessary data.

[0658] "Learning difficulties" refer to areas where a learner has a lack of understanding or makes many mistakes in a particular academic subject or problem.

[0659] "Means of identification" refers to technical methods, devices, and software that analyze and extract learners' learning difficulties based on collected data.

[0660] "Assignments" are learning items such as questions and exercises that are generated based on the learner's learning difficulties.

[0661] "Generation means" refers to technical methods, devices, or software that automatically generate tasks related to learning difficulties.

[0662] "Presentation means" refers to the technical methods, devices, and software used to visually and audibly present the generated assignment to the learner.

[0663] "Means for analyzing answers" refers to technical methods, devices, and software for analyzing answers provided by learners and determining whether they are correct or incorrect.

[0664] "Feedback" refers to information such as explanations, advice, and instructions for retrying that are provided to learners based on the analyzed answers.

[0665] An "artificial intelligence character" is a computer-generated character used to provide feedback and commentary through interaction with a user.

[0666] "Means for regularly checking learner progress" refers to technological methods, devices, or software that periodically evaluate and analyze learner progress.

[0667] "Means for adjusting the learning plan" refers to technological methods, devices, and software for adjusting the learning content and pace to suit the learner based on the results of regular progress checks.

[0668] "Emotional state" indicates the user's psychological and emotional state and is determined based on data such as facial expressions and voice.

[0669] "Means of analysis" refers to the technical methods, devices, and software that analyze the data necessary to determine emotional state.

[0670] "Means for adjusting feedback based on the analysis results" refers to technical methods, devices, or software that adjust the content and format of the feedback provided based on the results of sentiment analysis.

[0671] This invention is an AI tutoring system that provides learning support tailored to the individual needs of learners, and also combines it with an emotion engine that recognizes the user's emotions and provides an optimal learning environment. This system is primarily composed of a server, a terminal, a learner (user), and an emotion engine. Each component is described in detail below.

[0672] Data collection

[0673] Users use their devices to input learning data, including daily homework results and past test results. The devices first store this data locally and then send it to the server. The devices used include, for example, personal computers and tablets. The devices are also equipped with cameras and microphones, which capture the user's facial expressions and voice. The captured facial and voice data is analyzed in real time by the emotion engine.

[0674] Data storage and analysis

[0675] The server stores the received learning data in a database, which makes it available for later analysis. Based on this data, the server uses an AI algorithm to identify the learner's learning difficulties. For example, from math test data, the server can identify that the learner has difficulty calculating fractions.

[0676] Assignment generation and presentation

[0677] The server generates tasks that address the identified learning difficulties. These tasks are designed to increase in difficulty in stages, adjusting to include beginner level problems and gradually more difficult problems. The generated tasks are sent to the terminal, which then presents them to the user. The user answers the presented tasks and re-enters the answers into the terminal. The entered answers are then sent from the terminal to the server.

[0678] Answer analysis and feedback

[0679] The server analyzes the user's answers and determines whether they are correct or incorrect. Based on the analysis results, it generates appropriate feedback and sends it to the device. The feedback includes specific explanations and advice. For example, if the answer is incorrect, the server provides advice such as, "You made a mistake, but you need to make the denominators the same."

[0680] Emotion Engine and AI Avatar

[0681] Users select their preferred AI character on the device. The emotion engine analyzes the user's facial expressions and voice to recognize their emotional state in real time. The AI ​​character uses voice and text to provide explanations and feedback based on the user's level of understanding. For example, if the character determines that the user is nervous, it will make encouraging comments such as, "Relax and try again."

[0682] Check your progress and adjust your study plan

[0683] The server periodically monitors the user's learning progress and adjusts the learning plan as needed. It generates new assignments based on the user's progress and provides feedback that focuses on new learning difficulties or areas that need improvement. It also takes into account the learner's emotional state and builds a learning plan that is tailored to that state.

[0684] Specific examples

[0685] For example, if a user sends the results of an arithmetic test to a server, the server analyzes the results and determines that the user has difficulty calculating fractions. The server then generates a simple fraction addition problem (e.g., 1 / 2 + 1 / 4) and sends it to the device. While the user solves the problem and the answer is sent to the server, the emotion engine analyzes the user's facial expressions and voice and determines that the user is nervous. At this point, an AI character provides encouragement and explanation, such as, "Relax and think about it. First, you need to make the denominators the same."

[0686] Examples of input prompts for a generative AI model include:

[0687] Describe a learning support system in which a user inputs past test data, a server extracts learning difficulties from that data, and generates tasks. An emotion engine analyzes the user's facial expressions and voice to provide real-time feedback.

[0688] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0689] Step 1: Data entry

[0690] Users use a device to input homework results and past test results. This input data may include answers in text format or multiple choice format. Specifically, the user enters data into an input box on the device screen and presses the "Submit" button. The input data includes homework scores, test scores, answers, etc., and this becomes the basic data for the system.

[0691] Input: Homework results, test results

[0692] Output: Save to local database, move data to next step

[0693] Step 2: Send data

[0694] The device sends the homework and test result data entered by the user to the server. This operation is performed using the device's communication function and is executed automatically by the user's operation. For example, when the send button is pressed, the device packs the data and sends it to the server's API endpoint.

[0695] Input: Data from the local database

[0696] Output: Upload data to the server

[0697] Step 3: Saving to the database

[0698] The server analyzes the received homework and test results data and stores them in a database. The data stored includes the score for each question, the answer content, and whether the answer was correct or incorrect. This operation is basically performed by issuing an insert query to the database.

[0699] Input: Received data

[0700] Output: Save data to database

[0701] Step 4: Collecting emotion data

[0702] The device's camera and microphone are used to collect the user's facial expressions and voice data in real time, and the collected data is sent to the emotion engine.

[0703] Input: User's facial expression and voice data

[0704] Output: Sending data to the emotion engine

[0705] Step 5: Analyze the sentiment data

[0706] The emotion engine analyzes the collected facial and voice data to determine the user's emotional state. It uses facial recognition and voice analysis algorithms. Specifically, if the user is smiling, it identifies the emotion as "relaxed," while if the user is smiling, it identifies the emotion as "tense."

[0707] Input: facial expression and voice data

[0708] Output: Emotion analysis results

[0709] Step 6: Identifying learning difficulties

[0710] The server analyzes the stored learning data using an AI algorithm to identify the learner's learning difficulties. For example, based on test data, it identifies that the learner has difficulty calculating fractions. During this process, the correct answer rate and answer time for each problem are also used as part of the analysis.

[0711] Input: Training data stored in a database

[0712] Output: Identification of learning difficulties

[0713] Step 7: Generate an issue

[0714] The server generates a set of increasingly difficult problems for each identified learning difficulty, starting with easy problems (e.g., basic addition of fractions) and gradually increasing in difficulty (e.g., subtraction of mixed fractions). This problem generation is done using an AI problem generation algorithm.

[0715] Input: Results of identifying learning difficulties

[0716] Output: The generated set of issues

[0717] Step 8: State the problem

[0718] The terminal presents the task sent from the server to the user. The displayed task consists of a text question and an answer input box. The terminal provides an interface for the user to perform the operations necessary to solve the task.

[0719] Input: Issue data from the server

[0720] Output: Present the task to the user

[0721] Step 9: Enter and submit your answers

[0722] The user inputs the answer to the task presented on the terminal and presses the send button to send the answer to the server. This data includes the user's answer content, answer time, etc.

[0723] Input: User's answer

[0724] Output: Send answer data to the server

[0725] Step 10: Analyze the solution

[0726] The server analyzes the received answer data and determines whether the answer is correct or incorrect. It checks the validity of the answer using an AI algorithm, and if the answer is correct, it will state that, and if it is incorrect, it will explain the reason.

[0727] Input: User's answer data

[0728] Output: Analysis results (correct / incorrect)

[0729] Step 11: Generate feedback

[0730] The server generates appropriate feedback based on the analysis results, including specific advice such as "Well done!" if the answer is correct, or "Sorry, try calculating it again with the same denominator" if the answer is incorrect.

[0731] Input: Answer analysis result

[0732] Output: Generated feedback

[0733] Step 12: AI Avatar Commentary

[0734] Users select their preferred AI avatar on their device, and the emotion engine analyzes the user's facial expressions and voice. Based on this, the AI ​​avatar provides the user with commentary and encouragement in voice and text. For example, if it determines that the user is nervous, it will say, "Relax and think about it."

[0735] Input: Sentiment analysis results, feedback content

[0736] Output: Audio and text commentary by an AI avatar

[0737] Step 13: Check and adjust your progress

[0738] The server periodically checks the user's learning progress and creates a new learning plan, taking into account past learning data, feedback, and emotional state. Based on the progress data, the server provides new assignments and feedback, creating an optimal learning environment for the user.

[0739] Input: Learning data and emotion data stored in the database

[0740] Output: New lesson plan, next set of assignments

[0741] (Application example 2)

[0742] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0743] Conventional learning support systems provide feedback based on individual learners' progress and level of understanding, but they are unable to take into account the learner's emotional state. As a result, if a learner feels nervous or stressed, appropriate support may not be provided, which could result in a decrease in learning effectiveness. To address this issue, a system is needed that provides optimal feedback according to the learner's emotional state.

[0744] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting the learner's test results and homework results, means for identifying the learner's weak areas based on the collected data, means for generating questions related to the learner's weak areas, means for presenting the generated questions to the learner, means for analyzing the learner's answers and providing appropriate feedback, means for using an AI avatar to provide explanations, means for periodically checking the learner's progress and adjusting the learning plan, means for determining the user's emotional state using an emotion recognition engine, and means for adjusting the learning content and feedback based on the user's emotional state. This enables optimal learning support tailored to the individual learning needs of the learner while taking the learner's emotional state into consideration.

[0745] "Student" refers to an individual who is engaged in learning activities using a learning support system.

[0746] "Means for collecting test results and homework results" refers to devices or software that input, record, and store learners' daily learning results into the system.

[0747] A "means for identifying areas of difficulty" is a device or software that analyzes the collected data to detect areas or fields in which the learner has difficulty understanding.

[0748] A "means for generating questions" is a device or software that automatically generates appropriate study questions based on identified areas of weakness.

[0749] A "means for presenting questions" is a device or software that presents the generated questions to the learner visually or audibly.

[0750] "Means for analyzing answers and providing appropriate feedback" refers to a device or software that evaluates the learner's answers, takes the results into consideration, and returns useful advice or explanations.

[0751] "Means using AI avatars" refers to devices or software that use artificial intelligence technology to utilize virtual characters that provide answers and explanations to learners.

[0752] A "means for checking learning progress and adjusting learning plans" is a device or software that periodically evaluates a learner's learning progress and modifies or improves the learning plan based on that progress.

[0753] An "emotion recognition engine" is a device or software equipped with artificial intelligence technology that analyzes a learner's facial expressions and voice to determine their emotional state.

[0754] The "means for adjusting learning content and feedback based on emotional state" refers to a device or software that provides appropriate learning content and feedback according to the learner's emotional state based on information obtained from an emotion recognition engine.

[0755] This invention is an AI tutoring system to support employee training in brick-and-mortar stores, recognizing users' emotions to provide an optimal learning environment. The system is primarily composed of a server, a terminal, users, and an emotion recognition engine. The server performs central data analysis and generates questions, the terminal provides an interface for users, and the emotion recognition engine recognizes users' emotions.

[0756] Hardware and software configuration:

[0757] 1. Server: Mainly responsible for data analysis, question generation, and learning progress management. Software used includes AI algorithms, database management systems, and web servers (e.g., Apache).

[0758] 2. Device: A smartphone is used to provide the user interface, present learning content, and display feedback. The device is equipped with a camera and microphone, which are used to collect the user's facial expressions and voice. The recommended hardware is a smart device with a camera (e.g., iPhone, Android smartphone). The software used includes the UI framework used by the application (e.g., React Native, Flutter, etc.).

[0759] 3. Emotion Recognition Engine: Analyzes the learner's facial expressions and voice to determine their emotional state in real time. The software used includes machine learning libraries such as TensorFlow and OpenCV.

[0760] Explanation of program operation:

[0761] First, a user (such as a new employee) uses a device to log in to a training application. The application retrieves the user's learning history from the server and presents the next learning content to be tackled. This learning content is generated by an AI algorithm on the server based on the collected data. While the user is learning, the device's built-in camera and microphone collect the user's facial expressions and voice in real time. This data is analyzed by an emotion recognition engine to determine the user's emotional state.

[0762] Once the user's facial and voice data is collected, it is passed to an emotion recognition engine. This engine uses TensorFlow and OpenCV to analyze the user's emotional state in real time. For example, if the application determines that the user is nervous, it will display a message such as, "Relax, and try to smile first." The learning content and feedback are then dynamically adjusted to provide appropriate support tailored to the user's emotional state.

[0763] Examples:

[0764] Consider a scenario where a new employee is undergoing customer service skill training. When the user launches the application and starts the training module, the server generates the following scenario based on the user's progress data:

[0765] Example prompt sentence:

[0766] "Think about the ideal way to serve customers in the following situation: When there are many customers in the store during busy hours."

[0767] While the user is learning based on the prompts, the emotion recognition engine analyzes the user's emotions and provides relaxing messages such as "Relax, try to smile first" if the user is nervous. In this way, it is possible to provide support tailored to the needs of each individual user while maximizing the learning effect.

[0768] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0769] Step 1:

[0770] The user logs in to the training application using a device (smartphone). The user's ID and password are required as input, and are verified by the user authentication server. The user's learning history and current progress are obtained as output. If authentication is successful, the application's home screen is displayed, showing the next learning content to be tackled (e.g., customer service skills training).

[0771] Step 2:

[0772] The server receives the user's login information and learning history and generates the next learning content. The user's past learning history stored in the learning database is used as input. Using data analysis and AI algorithms, questions and practice scenarios are generated based on the individual's learning progress. The generated learning content is sent to the device as output.

[0773] Step 3:

[0774] The terminal receives the learning content sent from the server and presents it to the user. The learning content data from the server is used as input. When the user logs in to their account and starts a training session, the learning content (e.g., a customer service scenario) is displayed on the screen. As output, the learning content is presented on the display screen.

[0775] Step 4:

[0776] The user works through exercises and scenarios according to the learning content presented. The user's answers and reactions are entered as input. The device's built-in camera and microphone also collect the user's facial expressions and voice data. This data is sent to the emotion recognition engine in real time.

[0777] Step 5:

[0778] The emotion recognition engine analyzes the user's facial expressions and voice data in real time to determine the user's emotional state. Image data and voice data collected from the camera and microphone are used as input. The emotional state (e.g., nervous, relaxed, etc.) is identified using TensorFlow and OpenCV. The determined emotional state is sent to the device as output.

[0779] Step 6:

[0780] The server provides appropriate feedback to the user based on the determined emotional state. As input, the results of the emotion recognition engine and the user's learning progress data are used. Using an AI algorithm, advice, explanations, and encouraging messages are generated according to the user's emotional state. As output, the generated feedback is sent to the device.

[0781] Step 7:

[0782] The terminal displays the feedback sent from the server to the user. As input, the feedback data from the server is used. A text message or a voice message according to the emotional state is displayed on the user's screen. As output, the user receives feedback through voice and text.

[0783] Step 8:

[0784] The user continues learning based on the feedback and takes on new challenges. The answers and reactions after receiving the feedback are used as inputs, which dynamically adjust the next learning content. The output is the learning progress sent back to the server, and the whole system repeats the learning cycle.

[0785] 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.

[0786] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0787] 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.

[0788] [Third embodiment]

[0789] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0790] 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.

[0791] 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).

[0792] 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.

[0793] 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.

[0794] 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).

[0795] 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.

[0796] 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.

[0797] 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.

[0798] 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.

[0799] 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.

[0800] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0801] System Overview

[0802] This invention is an AI tutoring system that provides learning support tailored to the individual needs of each learner. This system is primarily composed of a server, a terminal, and a learner (user). The server performs central data analysis and problem generation, while the terminal provides an interface for the user.

[0803] Data collection

[0804] Users input their daily homework and past test results through the device, which then sends the data to a server, which stores the data in a database for later analysis.

[0805] Identifying weak areas

[0806] Based on the collected data, the server uses AI algorithms to identify areas where learners struggle, such as analyzing test scores and homework answers to identify areas where learners frequently make mistakes and spend time.

[0807] Problem Generation

[0808] The server generates a set of problems of increasing difficulty for each identified weak area, allowing learners to tackle problems ranging from basic to advanced. The generated problems are then sent to the device.

[0809] Problem Statement

[0810] The terminal presents the questions sent from the server to the user. The user answers the questions and enters the answers into the terminal. The terminal then sends the answers to the server.

[0811] Answer analysis and feedback

[0812] The server analyzes the user's answers and determines whether they are correct or incorrect. Based on the analysis results, it generates appropriate feedback, which may include specific explanations or advice.

[0813] AI Avatar Explained

[0814] Users can select their preferred AI avatar on their device, which will then use voice and text to explain how to solve the problem and any areas they don't understand, making it easier for users to understand the information visually and audibly.

[0815] Constant progress monitoring and adjustment of study plans

[0816] The server periodically checks the user's learning progress and adjusts the learning plan as needed. For example, if a particular weak area is overcome, the server will provide new questions that focus on that area. This allows the learner to continuously improve their academic ability.

[0817] Specific examples

[0818] As a concrete example, a user sends the results of an arithmetic test to a server, which determines that the user has difficulty calculating fractions. The server then generates a simple fraction addition problem (e.g., 1 / 2 + 1 / 4) and sends it to the device. The user solves the problem and enters the answer on the device, and the server determines whether the answer is correct or incorrect. If the answer is incorrect, an AI avatar selected by the user will provide an explanation, such as, "When adding fractions, you first need to make the denominators the same."

[0819] This allows users to gain a deeper understanding and learn more efficiently than with conventional one-way learning methods.

[0820] The processing flow will be explained below.

[0821] Step 1: Collect data

[0822] The user uses the terminal to input daily homework results and past test results.

[0823] The terminal transmits the input data to the server.

[0824] The server stores the received data in a database.

[0825] Step 2: Identify your weaknesses

[0826] The server analyzes the stored data using AI algorithms.

[0827] Based on the analysis results, the server identifies areas in which the learner is weak.

[0828] For example, the server analyzes test results and homework answer trends to extract areas where learners frequently make mistakes.

[0829] Step 3: Generate the problem

[0830] The server generates questions related to the identified weaknesses.

[0831] The server uses an AI algorithm to create sets of questions that gradually increase in difficulty, starting with easy questions.

[0832] The server sends the generated questions to the terminal.

[0833] Step 4: State the problem

[0834] The terminal presents the questions sent from the server to the user.

[0835] The user answers the questions presented on the terminal.

[0836] The terminal transmits the user's answer results to the server.

[0837] Step 5: Analysis and feedback of answers

[0838] The server analyzes the user's answers and determines whether they are correct or incorrect.

[0839] The server generates appropriate feedback based on the accuracy of the answer and the time it took to answer.

[0840] The server transmits the feedback content to the terminal.

[0841] Step 6: AI Avatar Explained

[0842] Users select their preferred AI avatar on their device.

[0843] The device displays the commentary sent from the server in the form of an AI avatar.

[0844] The AI ​​avatar uses voice and text to explain how to solve the problem and key points.

[0845] Step 7: Regularly monitor your progress and adjust your study plan

[0846] The server periodically checks the user's learning progress.

[0847] The server adjusts the study plan based on the study progress.

[0848] For example, if a particular weak area is overcome, the server generates new problems for that weak area.

[0849] Example 1

[0850] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0851] Conventional learning support systems do not provide sufficient, effective learning support that meets the individual needs of learners. Furthermore, they have difficulty identifying areas in which learners are weak, providing questions that address those areas, and providing appropriate feedback. This creates the problem of ineffectively improving learners' academic abilities.

[0852] 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.

[0853] In this invention, the server includes means for collecting data on the learner's grades and assignments, means for identifying the learner's weak points in academic ability based on the collected data, and means for automatically generating questions related to the learner's weak points in academic ability, thereby enabling effective learning support and feedback to be provided to the learner according to their individual learning needs.

[0854] "Academic performance data" refers to the scores and evaluation results of tests and assignments taken by learners.

[0855] "Assignment data" refers to the answers and progress of homework and practice problems that a learner has completed.

[0856] "Academic weaknesses" refer to areas of study where a learner is particularly weak or areas where they frequently make mistakes.

[0857] "Automatically generated" refers to software or algorithms creating problems or challenges without human intervention.

[0858] "Virtual character" means a digital avatar generated by a computer program that can communicate through voice and text.

[0859] "Instructional information" refers to educational information such as explanations, advice, and feedback provided to learners.

[0860] "Machine learning algorithms" refer to computational methods and models that allow computers to learn from data and automatically improve their performance.

[0861] A "study plan" refers to the tasks and study schedule that a learner must complete each day.

[0862] This invention provides a system that provides learning support tailored to the individual needs of learners. This system is primarily composed of a server, a terminal, and a learner (user). The server performs central data analysis and question generation, while the terminal functions as an interface for the user.

[0863] System Overview

[0864] This system helps learners improve their academic ability through the following processes:

[0865] Hardware and software used

[0866] The hardware used includes servers (e.g., cloud-based servers) and user devices (e.g., tablets, PCs, etc.), while the software includes database management systems (e.g., MySQL or PostgreSQL), AI algorithms (e.g., Python's scikit-learn or TensorFlow), and generative AI models (e.g., GPT-4).

[0867] Data collection

[0868] Users use the device to input their daily homework and past test results. When the user enters this data through a dedicated application on the device or a web interface, the device sends it to the server, which stores the received data in a database.

[0869] Identifying weak areas

[0870] The server uses machine learning algorithms to analyze the stored data, such as Python's scikit-learn or TensorFlow, to identify areas where learners frequently make mistakes and spend time, allowing the server to identify weak areas in the learner's academic performance.

[0871] Problem Generation

[0872] The server generates questions using a generative AI model based on the identified weak areas. The generative AI model uses GPT-4 or similar. The server creates a prompt and sends it to the generative AI model. For example, the following prompt can be used:

[0873] Example prompt sentence:

[0874] "Your learner has difficulty with fractions. Generate a simple fraction addition problem like this: Problem: 1 / 2 + 1 / 4"

[0875] Problem Statement

[0876] The terminal receives the questions sent from the server and presents them to the user. The user answers the questions displayed on the terminal and enters the answers into the terminal.

[0877] Answer analysis and feedback

[0878] The device sends the user's answers to the server, which analyzes them and identifies correct and incorrect answers. The server then generates feedback based on the analysis results, including specific explanations and advice.

[0879] AI Avatar Explained

[0880] Users select their preferred virtual character on their device, who then uses voice (e.g., text-to-speech API) and text to explain how to solve the problem or any areas of confusion, helping users understand the information visually and audibly.

[0881] Constant progress monitoring and adjustment of study plans

[0882] The server periodically checks the learner's progress and adjusts the learning plan as needed. For example, if a particular academic weakness is overcome, the server will provide new questions that focus on that weakness. This iterative process allows the learner to continuously improve their academic performance.

[0883] In this way, the learning support system of the present invention responds to the individual needs of each learner and provides effective learning support, thereby enabling each learner to improve their academic ability efficiently and continuously.

[0884] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0885] Step 1:

[0886] Users input their daily homework results and past test results through their devices. Specifically, they input this data using a dedicated application on their device or a form provided in the web interface. The input data includes information such as the content of the questions the user answered, their answers, and the time it took to complete them. The device then sends this input data to the server.

[0887] Input: Homework and test result data entered by the user into the device

[0888] Output: Data sent from the device to the server

[0889] Step 2:

[0890] The server stores the received data in a database. By using a database management system (e.g., MySQL or PostgreSQL), all training data can be managed efficiently.

[0891] Input: Data sent from the terminal

[0892] Output: Data stored in the database

[0893] Step 3:

[0894] The server uses machine learning algorithms (e.g., Python's scikit-learn or TensorFlow) to analyze the stored data. The server applies the algorithms to identify areas where learners frequently make mistakes and spend time on grades and assignments. This allows the server to identify the learner's academic weaknesses.

[0895] Input: Grade data and assignment data stored in the database

[0896] Output: Identified academic weaknesses

[0897] Step 4:

[0898] The server uses a generative AI model (e.g., GPT-4) to generate problems based on the identified weaknesses. The server creates a prompt like the following and sends it to the generative AI model: "The learner has difficulty with fraction calculations. Please generate a simple fraction addition problem like this: Problem: 1 / 2 + 1 / 4."

[0899] Input: Identified academic weaknesses

[0900] Output: The problem returned by the generative AI model

[0901] Step 5:

[0902] The device receives the questions sent from the server and presents them to the user. Specifically, the user can view the questions through a dedicated app on the device or a web interface. The user answers the presented questions.

[0903] Input: Question sent by server

[0904] Output: The problem presented to the user

[0905] Step 6:

[0906] The user inputs the answer into the terminal, which then transmits the answer to the server.

[0907] Input: Answer result entered by the user into the terminal

[0908] Output: Answer results sent from the device to the server

[0909] Step 7:

[0910] The server analyzes the answers and determines whether they are correct or incorrect. The server uses a Python script to compare the answers with the correct answers. The server then generates specific feedback based on the analysis results. The generated feedback includes specific explanations and advice.

[0911] Input: Answer results sent from the device

[0912] Output: Feedback information

[0913] Step 8:

[0914] Users select their preferred virtual character on their device, and the virtual character uses voice and text to explain how to solve the problem or provide explanations for any unclear points. The voice is generated using a text-to-speech API, and the information is provided visually using a 3D model created in Unity.

[0915] Input: Feedback information

[0916] Output: Explanation by virtual character

[0917] Step 9:

[0918] The server periodically checks the student's progress and adjusts the learning plan as necessary. The server periodically (e.g., weekly or monthly) analyzes the performance data and generates new questions and updates the learning plan if the student's academic weaknesses are overcome.

[0919] Input: Learning progress data

[0920] Output: Updated lesson plan

[0921] The above are the specific processing steps of the program of this system.

[0922] (Application example 1)

[0923] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0924] Conventional AI tutoring systems provide individual learning support based on learners' test results and homework results, but they do not provide immediate learning support in physical stores. Furthermore, the lack of an intuitive visual or audio interface makes it difficult for learners to engage and for effective learning to occur.

[0925] 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.

[0926] In this invention, the server includes means for collecting test results and homework results of learners, means for identifying areas of weakness of learners based on the collected data, and means for generating questions related to the areas of weakness of learners. This enables learners to use a visual device to send book and teaching material information to the server in real time, receive questions through the visual device, collect answers through voice input, and receive explanations from an AI avatar.

[0927] A "learner" is an individual seeking to acquire knowledge or skills.

[0928] A "visual device" is a device that provides information visually, and generally refers to smart glasses or head-mounted displays.

[0929] A "server" is a computer system that performs central processing such as collecting, analyzing, and storing data.

[0930] "Test results" refers to the results and answers given by the learner on the test.

[0931] "Homework results" refers to the grades and answers to homework submitted by the learner.

[0932] A "weak area" is an area in which a learner has particular difficulty understanding or mastering.

[0933] "Questions" refer to learning tasks or exercises that learners must answer.

[0934] "Means of collection" refers to the methods and techniques used to collect the necessary data.

[0935] "Analytical means" refers to the methods and techniques used to analyze data and derive meaningful information.

[0936] "Feedback" refers to evaluation and advice given to a learner based on their answers and behavior.

[0937] An "AI avatar" refers to a virtual character that uses artificial intelligence to provide learning assistance.

[0938] "Voice input" refers to a method in which learners input information by voice.

[0939] "Progress" refers to a learner's learning progress and achievements.

[0940] A "learning plan" refers to the learning content and schedule planned to help a learner achieve their learning goals.

[0941] This invention is an AI tutoring system that provides learning support tailored to the individual needs of each learner. This system is primarily composed of a server, a terminal, and a learner, and utilizes visual devices and voice input functions to provide instant learning support in a brick-and-mortar store.

[0942] System Overview

[0943] The server collects the learner's test results and homework results, analyzes them, and identifies the learner's weak areas. It then generates questions that address the identified weak areas and sends the questions to the learner's device or visual device. The learner receives the questions through the visual device and answers them using voice input. The answers are sent to the server, and appropriate feedback is provided based on the analysis results.

[0944] Hardware and Software

[0945] Hardware used

[0946] Visual devices: smart glasses or head-mounted displays (e.g., smart glasses)

[0947] Device: Smartphone or tablet

[0948] Server: A computer system that analyzes and stores data (e.g., cloud services).

[0949] Software used

[0950] Data collection and analysis: Python scripts

[0951] Server communication: RESTful API

[0952] Audio Description: Text-to-Speech (TTS) engine (e.g., Google Cloud Text-to-Speech)

[0953] Specific examples of explanation

[0954] Let's say a student chooses a book on "fraction calculation" at a physical store. When the student puts on the smart glasses and reads the book information, that information is sent to the server in real time. The server combines this with past learning data and determines that the student has difficulty with "fraction addition." The server then generates a simple "fraction addition" problem (e.g., 1 / 2 + 1 / 4) and sends it to the smart glasses. The student answers via voice input, and the answer is sent to the server. The server analyzes the answer, determines whether it is correct or incorrect, and sends an explanation to the smart glasses via voice and text, such as "When adding fractions, you first need to make the denominators the same," with an AI avatar providing the explanation. This allows the student to deepen their understanding on the spot.

[0955] Prompt Sentence Examples

[0956] The generative AI model can generate explanations for specific learning items by inputting the following prompts:

[0957] "setting"

[0958] Generate detailed explanations to help students understand fraction addition.

[0959] "prompt"

[0960] When adding fractions, you first need to make the denominators the same. To do this, convert the fractions so that the denominators are the lowest common multiple. For example, to calculate 1 / 2 + 1 / 4, convert 1 / 2 to 2 / 4, then calculate 2 / 4 + 1 / 4... (Detailed steps continue.)

[0961] The system allows learners to receive personalized learning support in real time within the store, providing an intuitive learning environment using visual and audio.

[0962] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0963] Step 1:

[0964] A user uses a visual device to select a book or educational material in a physical store and reads its information. The visual device then uses a camera to scan the barcode or QR code on the book or educational material to obtain the relevant information. The obtained information is sent to the server in real time. The input is the information of the book or educational material read by the visual device, and the output is the information sent to the server.

[0965] Step 2:

[0966] The server compares the received information on books and teaching materials with the learner's past test results and homework results to identify the learner's weak areas. Past learning data for each learner is stored in the database, and an AI algorithm is used to analyze the learner's weak areas. The input is information on books and teaching materials and past learning data, and the output is the identified learner's weak areas.

[0967] Step 3:

[0968] The server generates problems in stages that address the identified weak areas. An AI model is used to generate problems with increasing difficulty, from simple to advanced. The generated problems are sent to a visual device. The input is the identified weak areas and a problem generation prompt, and the output is the generated problem set.

[0969] Step 4:

[0970] The visual device presents the questions received from the server to the user. The questions are displayed on the visual device's display, and the user confirms them. Using the visual device's voice input function, the user inputs the answer to the question by voice. The input is the question sent from the server, and the output is the user's voice answer.

[0971] Step 5:

[0972] The visual device converts the user's voice response into text data and sends it to the server. The voice response is converted into text data using a Speech-to-Text (STT) engine. The input is the user's voice response, and the output is the response as text data.

[0973] Step 6:

[0974] The server analyzes the user's answers and determines whether they are correct or incorrect. An AI algorithm is used for the analysis to evaluate the accuracy of the answers. The input is the user's answer text, and the output is the analysis result of the answer.

[0975] Step 7:

[0976] The server generates appropriate feedback based on the analysis results. The appropriate feedback includes points of mistakes and explanations. The feedback content is sent to the visual device in audio and text format. The input is the analysis result of the answer, and the output is the feedback information.

[0977] Step 8:

[0978] The visual device presents the feedback received from the server to the user, and the AI ​​avatar provides explanations in voice and text. For example, an explanation such as "When adding fractions, you must first make the denominators the same." The input is the feedback information sent from the server, and the output is the explanation to the user.

[0979] 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.

[0980] System Overview

[0981] This invention is an AI tutoring system that provides learning support tailored to the individual needs of each learner, and also combines it with an emotion engine that recognizes the user's emotions and provides an optimal learning environment. This system is primarily composed of a server, a terminal, a learner (user), and an emotion engine. The server performs central data analysis and question generation, the terminal provides an interface for the user, and the emotion engine recognizes the user's emotions.

[0982] Data collection

[0983] Users input their daily homework results and past test results through the device. The device then sends this data to the server, which stores the received data in a database for later analysis. The device is also equipped with a camera and microphone, which allow the emotion engine to analyze the user's facial expressions and voice.

[0984] Identifying weak areas

[0985] Based on the stored data, the server uses AI algorithms to identify areas where learners struggle, analyzing test scores and homework answers to identify areas where learners frequently make mistakes and where they spend a lot of time on specific questions.

[0986] Problem Generation

[0987] The server generates a set of problems of increasing difficulty for the identified weak areas, starting with easy problems and gradually increasing in difficulty. The generated problems are sent to the device.

[0988] Problem Statement

[0989] The terminal presents the questions sent from the server to the user. The user answers the questions and enters the answers into the terminal. The terminal then sends the answers to the server.

[0990] Answer analysis and feedback

[0991] The server analyzes the user's answers and determines whether they are correct or incorrect. Based on the analysis results, it generates appropriate feedback and sends it to the device. This feedback includes specific explanations and advice.

[0992] AI Avatar Explained

[0993] Users select their preferred AI avatar on their device. The emotion engine analyzes the user's facial expressions and voice to recognize their emotional state. The AI ​​avatar uses voice and text to provide explanations tailored to the user's level of understanding or detailed explanations of specific issues. It also provides reassuring and encouraging comments based on the emotion engine's results.

[0994] Constant progress monitoring and adjustment of study plans

[0995] The server periodically checks the user's learning progress and adjusts the learning plan as needed. For example, if a particular weakness is overcome, it will provide new questions that focus on that weakness. It also takes into account the learner's emotional state and creates a learning plan that is tailored to that state.

[0996] Specific examples

[0997] For example, if a user submits the results of a math test to a server, the server will determine that the user has difficulty calculating fractions. The server then generates a simple fraction addition problem (e.g., 1 / 2 + 1 / 4) and sends it to the device. While the user is solving the problem, the emotion engine analyzes the user's facial expressions and voice and determines that the user is nervous. At this point, an AI avatar will provide encouragement and explanation, such as, "Relax and think about it. First, you need to make the denominators the same."

[0998] This allows users to have an optimal learning experience that is tailored to their individual feelings and level of understanding, compared to conventional one-way learning methods.

[0999] The processing flow will be explained below.

[1000] Step 1: Collect data

[1001] Users input their daily homework results and past test results into the terminal.

[1002] The terminal transmits the input data to the server.

[1003] The server stores the received data in a database.

[1004] In addition, the device's camera and microphone are used to capture the user's facial expressions and voice and send them to the emotion engine.

[1005] Step 2: Recognize emotions

[1006] The emotion engine analyzes the user's facial expressions and voice data to recognize the user's emotional state.

[1007] The recognized emotion data is sent to the server and reflected in future feedback and commentary.

[1008] Step 3: Identify your weaknesses

[1009] The server analyzes the saved homework and test results using AI algorithms.

[1010] Based on the analysis results, the server identifies areas in which the learner is weak.

[1011] For example, the server will extract areas where you frequently make mistakes or where you spend a lot of time on specific problems.

[1012] Step 4: Generate the problem

[1013] The server generates questions related to the identified weaknesses.

[1014] The server uses an AI algorithm to create sets of questions that gradually increase in difficulty, starting with easy questions.

[1015] The server sends the generated questions to the terminal.

[1016] Step 5: State the problem

[1017] The terminal presents the questions sent from the server to the user.

[1018] The user answers the questions presented to them.

[1019] The terminal collects the answer results and sends them to the server.

[1020] Step 6: Analysis and feedback of answers

[1021] The server analyzes the user's answers and determines whether they are correct or incorrect.

[1022] The server generates appropriate feedback based on the accuracy of the answer and the time it took to answer.

[1023] The server transmits the generated feedback content to the terminal.

[1024] Step 7: AI Avatar Explained

[1025] Users select their preferred AI avatar on their device.

[1026] The terminal receives the result of the emotion engine and obtains commentary content based on the user's emotional state from the server.

[1027] The AI ​​avatar uses voice and text to explain how to solve the problem and key points.

[1028] For example, if the user is nervous, the AI ​​avatar will offer encouraging comments such as, "Relax and think about it. First, we need to make the denominators the same."

[1029] Step 8: Check your progress and adjust your study plan

[1030] The server periodically checks the user's learning progress.

[1031] The server adjusts the learning plan based on the learning progress.

[1032] Once a particular weakness has been overcome, the server provides new problems that focus on that weakness.

[1033] The server also takes into account the recognized emotion data to provide the user with an optimal learning environment.

[1034] Example 2

[1035] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1036] Conventional learning support systems rely on one-way information transmission and are unable to take into account the individual learner's level of understanding or emotional state in a timely manner. As a result, appropriate feedback based on the learner's weak areas or level of understanding is not provided, making it difficult to provide effective learning support. In addition, it is difficult to provide a learning environment that takes into account the learner's emotional state, which is likely to reduce motivation to learn.

[1037] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for collecting the learner's assessment results, means for identifying the learner's learning difficulties based on the collected data, means for generating assignments related to the learner's learning difficulties, means for presenting the generated assignments to the learner, means for analyzing the learner's answers and providing appropriate feedback, means for using an AI character, means for periodically checking the learner's progress and adjusting the learning plan, means for analyzing the learner's emotional state, and means for adjusting the feedback based on the analysis results. This makes it possible to provide an optimal learning environment according to the individual learner's level of understanding and emotional state, thereby enabling effective and motivating learning support.

[1038] "Learner's assessment results" refers to information that indicates the results of learning and performance, such as test results and homework results that the learner has completed.

[1039] "Means of collection" refers to the technical methods, devices, and software used to collect learner assessment results and other necessary data.

[1040] "Learning difficulties" refer to areas where a learner has a lack of understanding or makes many mistakes in a particular academic subject or problem.

[1041] "Means of identification" refers to technical methods, devices, and software that analyze and extract learners' learning difficulties based on collected data.

[1042] "Assignments" are learning items such as questions and exercises that are generated based on the learner's learning difficulties.

[1043] "Generation means" refers to technical methods, devices, or software that automatically generate tasks related to learning difficulties.

[1044] "Presentation means" refers to the technical methods, devices, and software used to visually and audibly present the generated assignment to the learner.

[1045] "Means for analyzing answers" refers to technical methods, devices, and software for analyzing answers provided by learners and determining whether they are correct or incorrect.

[1046] "Feedback" refers to information such as explanations, advice, and instructions for retrying that are provided to learners based on the analyzed answers.

[1047] An "artificial intelligence character" is a computer-generated character used to provide feedback and commentary through interaction with a user.

[1048] "Means for regularly checking learner progress" refers to technological methods, devices, or software that periodically evaluate and analyze learner progress.

[1049] "Means for adjusting the learning plan" refers to technological methods, devices, and software for adjusting the learning content and pace to suit the learner based on the results of regular progress checks.

[1050] "Emotional state" indicates the user's psychological and emotional state and is determined based on data such as facial expressions and voice.

[1051] "Means of analysis" refers to the technical methods, devices, and software that analyze the data necessary to determine emotional state.

[1052] "Means for adjusting feedback based on the analysis results" refers to technical methods, devices, or software that adjust the content and format of the feedback provided based on the results of sentiment analysis.

[1053] This invention is an AI tutoring system that provides learning support tailored to the individual needs of learners, and also combines it with an emotion engine that recognizes the user's emotions and provides an optimal learning environment. This system is primarily composed of a server, a terminal, a learner (user), and an emotion engine. Each component is described in detail below.

[1054] Data collection

[1055] Users use their devices to input learning data, including daily homework results and past test results. The devices first store this data locally and then send it to the server. The devices used include, for example, personal computers and tablets. The devices are also equipped with cameras and microphones, which capture the user's facial expressions and voice. The captured facial and voice data is analyzed in real time by the emotion engine.

[1056] Data storage and analysis

[1057] The server stores the received learning data in a database, which makes it available for later analysis. Based on this data, the server uses an AI algorithm to identify the learner's learning difficulties. For example, from math test data, the server can identify that the learner has difficulty calculating fractions.

[1058] Assignment generation and presentation

[1059] The server generates tasks that address the identified learning difficulties. These tasks are designed to increase in difficulty in stages, adjusting to include beginner level problems and gradually more difficult problems. The generated tasks are sent to the terminal, which then presents them to the user. The user answers the presented tasks and re-enters the answers into the terminal. The entered answers are then sent from the terminal to the server.

[1060] Answer analysis and feedback

[1061] The server analyzes the user's answers and determines whether they are correct or incorrect. Based on the analysis results, it generates appropriate feedback and sends it to the device. The feedback includes specific explanations and advice. For example, if the answer is incorrect, the server provides advice such as, "You made a mistake, but you need to make the denominators the same."

[1062] Emotion Engine and AI Avatar

[1063] Users select their preferred AI character on the device. The emotion engine analyzes the user's facial expressions and voice to recognize their emotional state in real time. The AI ​​character uses voice and text to provide explanations and feedback based on the user's level of understanding. For example, if the character determines that the user is nervous, it will make encouraging comments such as, "Relax and try again."

[1064] Check your progress and adjust your study plan

[1065] The server periodically monitors the user's learning progress and adjusts the learning plan as needed. It generates new assignments based on the user's progress and provides feedback that focuses on new learning difficulties or areas that need improvement. It also takes into account the learner's emotional state and builds a learning plan that is tailored to that state.

[1066] Specific examples

[1067] For example, if a user sends the results of an arithmetic test to a server, the server analyzes the results and determines that the user has difficulty calculating fractions. The server then generates a simple fraction addition problem (e.g., 1 / 2 + 1 / 4) and sends it to the device. While the user solves the problem and the answer is sent to the server, the emotion engine analyzes the user's facial expressions and voice and determines that the user is nervous. At this point, an AI character provides encouragement and explanation, such as, "Relax and think about it. First, you need to make the denominators the same."

[1068] Examples of input prompts for a generative AI model include:

[1069] Describe a learning support system in which a user inputs past test data, a server extracts learning difficulties from that data, and generates tasks. An emotion engine analyzes the user's facial expressions and voice to provide real-time feedback.

[1070] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1071] Step 1: Data entry

[1072] Users use a device to input homework results and past test results. This input data may include answers in text format or multiple choice format. Specifically, the user enters data into an input box on the device screen and presses the "Submit" button. The input data includes homework scores, test scores, answers, etc., and this becomes the basic data for the system.

[1073] Input: Homework results, test results

[1074] Output: Save to local database, move data to next step

[1075] Step 2: Send data

[1076] The device sends the homework and test result data entered by the user to the server. This operation is performed using the device's communication function and is executed automatically by the user's operation. For example, when the send button is pressed, the device packs the data and sends it to the server's API endpoint.

[1077] Input: Data from the local database

[1078] Output: Upload data to the server

[1079] Step 3: Saving to the database

[1080] The server analyzes the received homework and test results data and stores them in a database. The data stored includes the score for each question, the answer content, and whether the answer was correct or incorrect. This operation is basically performed by issuing an insert query to the database.

[1081] Input: Received data

[1082] Output: Save data to database

[1083] Step 4: Collecting emotion data

[1084] The device's camera and microphone are used to collect the user's facial expressions and voice data in real time, and the collected data is sent to the emotion engine.

[1085] Input: User's facial expression and voice data

[1086] Output: Sending data to the emotion engine

[1087] Step 5: Analyze the sentiment data

[1088] The emotion engine analyzes the collected facial and voice data to determine the user's emotional state. It uses facial recognition and voice analysis algorithms. Specifically, if the user is smiling, it identifies the emotion as "relaxed," while if the user is smiling, it identifies the emotion as "tense."

[1089] Input: facial expression and voice data

[1090] Output: Emotion analysis results

[1091] Step 6: Identifying learning difficulties

[1092] The server analyzes the stored learning data using an AI algorithm to identify the learner's learning difficulties. For example, based on test data, it identifies that the learner has difficulty calculating fractions. During this process, the correct answer rate and answer time for each problem are also used as part of the analysis.

[1093] Input: Training data stored in a database

[1094] Output: Identification of learning difficulties

[1095] Step 7: Generate an issue

[1096] The server generates a set of increasingly difficult problems for each identified learning difficulty, starting with easy problems (e.g., basic addition of fractions) and gradually increasing in difficulty (e.g., subtraction of mixed fractions). This problem generation is done using an AI problem generation algorithm.

[1097] Input: Results of identifying learning difficulties

[1098] Output: The generated set of issues

[1099] Step 8: State the problem

[1100] The terminal presents the task sent from the server to the user. The displayed task consists of a text question and an answer input box. The terminal provides an interface for the user to perform the operations necessary to solve the task.

[1101] Input: Issue data from the server

[1102] Output: Present the task to the user

[1103] Step 9: Enter and submit your answers

[1104] The user inputs the answer to the task presented on the terminal and presses the send button to send the answer to the server. This data includes the user's answer content, answer time, etc.

[1105] Input: User's answer

[1106] Output: Send answer data to the server

[1107] Step 10: Analyze the solution

[1108] The server analyzes the received answer data and determines whether the answer is correct or incorrect. It checks the validity of the answer using an AI algorithm, and if the answer is correct, it will state that, and if it is incorrect, it will explain the reason.

[1109] Input: User's answer data

[1110] Output: Analysis results (correct / incorrect)

[1111] Step 11: Generate feedback

[1112] The server generates appropriate feedback based on the analysis results, including specific advice such as "Well done!" if the answer is correct, or "Sorry, try calculating it again with the same denominator" if the answer is incorrect.

[1113] Input: Answer analysis result

[1114] Output: Generated feedback

[1115] Step 12: AI Avatar Commentary

[1116] Users select their preferred AI avatar on their device, and the emotion engine analyzes the user's facial expressions and voice. Based on this, the AI ​​avatar provides the user with commentary and encouragement in voice and text. For example, if it determines that the user is nervous, it will say, "Relax and think about it."

[1117] Input: Sentiment analysis results, feedback content

[1118] Output: Audio and text commentary by an AI avatar

[1119] Step 13: Check and adjust your progress

[1120] The server periodically checks the user's learning progress and creates a new learning plan, taking into account past learning data, feedback, and emotional state. Based on the progress data, the server provides new assignments and feedback, creating an optimal learning environment for the user.

[1121] Input: Learning data and emotion data stored in the database

[1122] Output: New lesson plan, next set of assignments

[1123] (Application example 2)

[1124] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1125] Conventional learning support systems provide feedback based on individual learners' progress and level of understanding, but they are unable to take into account the learner's emotional state. As a result, if a learner feels nervous or stressed, appropriate support may not be provided, which could result in a decrease in learning effectiveness. To address this issue, a system is needed that provides optimal feedback according to the learner's emotional state.

[1126] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting the learner's test results and homework results, means for identifying the learner's weak areas based on the collected data, means for generating questions related to the learner's weak areas, means for presenting the generated questions to the learner, means for analyzing the learner's answers and providing appropriate feedback, means for using an AI avatar to provide explanations, means for periodically checking the learner's progress and adjusting the learning plan, means for determining the user's emotional state using an emotion recognition engine, and means for adjusting the learning content and feedback based on the user's emotional state. This enables optimal learning support tailored to the individual learning needs of the learner while taking the learner's emotional state into consideration.

[1127] "Student" refers to an individual who is engaged in learning activities using a learning support system.

[1128] "Means for collecting test results and homework results" refers to devices or software that input, record, and store learners' daily learning results into the system.

[1129] A "means for identifying areas of difficulty" is a device or software that analyzes the collected data to detect areas or fields in which the learner has difficulty understanding.

[1130] A "means for generating questions" is a device or software that automatically generates appropriate study questions based on identified areas of weakness.

[1131] A "means for presenting questions" is a device or software that presents the generated questions to the learner visually or audibly.

[1132] "Means for analyzing answers and providing appropriate feedback" refers to a device or software that evaluates the learner's answers, takes the results into consideration, and returns useful advice or explanations.

[1133] "Means using AI avatars" refers to devices or software that use artificial intelligence technology to utilize virtual characters that provide answers and explanations to learners.

[1134] A "means for checking learning progress and adjusting learning plans" is a device or software that periodically evaluates a learner's learning progress and modifies or improves the learning plan based on that progress.

[1135] An "emotion recognition engine" is a device or software equipped with artificial intelligence technology that analyzes a learner's facial expressions and voice to determine their emotional state.

[1136] The "means for adjusting learning content and feedback based on emotional state" refers to a device or software that provides appropriate learning content and feedback according to the learner's emotional state based on information obtained from an emotion recognition engine.

[1137] This invention is an AI tutoring system to support employee training in brick-and-mortar stores, recognizing users' emotions to provide an optimal learning environment. The system is primarily composed of a server, a terminal, users, and an emotion recognition engine. The server performs central data analysis and generates questions, the terminal provides an interface for users, and the emotion recognition engine recognizes users' emotions.

[1138] Hardware and software configuration:

[1139] 1. Server: Mainly responsible for data analysis, question generation, and learning progress management. Software used includes AI algorithms, database management systems, and web servers (e.g., Apache).

[1140] 2. Device: A smartphone is used to provide the user interface, present learning content, and display feedback. The device is equipped with a camera and microphone, which are used to collect the user's facial expressions and voice. The recommended hardware is a smart device with a camera (e.g., iPhone, Android smartphone). The software used includes the UI framework used by the application (e.g., React Native, Flutter, etc.).

[1141] 3. Emotion Recognition Engine: Analyzes the learner's facial expressions and voice to determine their emotional state in real time. The software used includes machine learning libraries such as TensorFlow and OpenCV.

[1142] Explanation of program operation:

[1143] First, a user (such as a new employee) uses a device to log in to a training application. The application retrieves the user's learning history from the server and presents the next learning content to be tackled. This learning content is generated by an AI algorithm on the server based on the collected data. While the user is learning, the device's built-in camera and microphone collect the user's facial expressions and voice in real time. This data is analyzed by an emotion recognition engine to determine the user's emotional state.

[1144] Once the user's facial and voice data is collected, it is passed to an emotion recognition engine. This engine uses TensorFlow and OpenCV to analyze the user's emotional state in real time. For example, if the application determines that the user is nervous, it will display a message such as, "Relax, and try to smile first." The learning content and feedback are then dynamically adjusted to provide appropriate support tailored to the user's emotional state.

[1145] Examples:

[1146] Consider a scenario where a new employee is undergoing customer service skill training. When the user launches the application and starts the training module, the server generates the following scenario based on the user's progress data:

[1147] Example prompt sentence:

[1148] "Think about the ideal way to serve customers in the following situation: When there are many customers in the store during busy hours."

[1149] While the user is learning based on the prompts, the emotion recognition engine analyzes the user's emotions and provides relaxing messages such as "Relax, try to smile first" if the user is nervous. In this way, it is possible to provide support tailored to the needs of each individual user while maximizing the learning effect.

[1150] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1151] Step 1:

[1152] The user logs in to the training application using a device (smartphone). The user's ID and password are required as input, and are verified by the user authentication server. The user's learning history and current progress are obtained as output. If authentication is successful, the application's home screen is displayed, showing the next learning content to be tackled (e.g., customer service skills training).

[1153] Step 2:

[1154] The server receives the user's login information and learning history and generates the next learning content. The user's past learning history stored in the learning database is used as input. Using data analysis and AI algorithms, questions and practice scenarios are generated based on the individual's learning progress. The generated learning content is sent to the device as output.

[1155] Step 3:

[1156] The terminal receives the learning content sent from the server and presents it to the user. The learning content data from the server is used as input. When the user logs in to their account and starts a training session, the learning content (e.g., a customer service scenario) is displayed on the screen. As output, the learning content is presented on the display screen.

[1157] Step 4:

[1158] The user works through exercises and scenarios according to the learning content presented. The user's answers and reactions are entered as input. The device's built-in camera and microphone also collect the user's facial expressions and voice data. This data is sent to the emotion recognition engine in real time.

[1159] Step 5:

[1160] The emotion recognition engine analyzes the user's facial expressions and voice data in real time to determine the user's emotional state. Image data and voice data collected from the camera and microphone are used as input. The emotional state (e.g., nervous, relaxed, etc.) is identified using TensorFlow and OpenCV. The determined emotional state is sent to the device as output.

[1161] Step 6:

[1162] The server provides appropriate feedback to the user based on the determined emotional state. As input, the results of the emotion recognition engine and the user's learning progress data are used. Using an AI algorithm, advice, explanations, and encouraging messages are generated according to the user's emotional state. As output, the generated feedback is sent to the device.

[1163] Step 7:

[1164] The terminal displays the feedback sent from the server to the user. As input, the feedback data from the server is used. A text message or a voice message according to the emotional state is displayed on the user's screen. As output, the user receives feedback through voice and text.

[1165] Step 8:

[1166] The user continues learning based on the feedback and takes on new challenges. The answers and reactions after receiving the feedback are used as inputs, which dynamically adjust the next learning content. The output is the learning progress sent back to the server, and the whole system repeats the learning cycle.

[1167] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1168] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1169] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1170] [Fourth embodiment]

[1171] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1172] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1173] 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).

[1174] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1175] 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.

[1176] 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).

[1177] 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.

[1178] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1179] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1180] 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.

[1181] 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.

[1182] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1183] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1184] System Overview

[1185] This invention is an AI tutoring system that provides learning support tailored to the individual needs of each learner. This system is primarily composed of a server, a terminal, and a learner (user). The server performs central data analysis and problem generation, while the terminal provides an interface for the user.

[1186] Data collection

[1187] Users input their daily homework and past test results through the device, which then sends the data to a server, which stores the data in a database for later analysis.

[1188] Identifying weak areas

[1189] Based on the collected data, the server uses AI algorithms to identify areas where learners struggle, such as analyzing test scores and homework answers to identify areas where learners frequently make mistakes and spend time.

[1190] Problem Generation

[1191] The server generates a set of problems of increasing difficulty for each identified weak area, allowing learners to tackle problems ranging from basic to advanced. The generated problems are then sent to the device.

[1192] Problem Statement

[1193] The terminal presents the questions sent from the server to the user. The user answers the questions and enters the answers into the terminal. The terminal then sends the answers to the server.

[1194] Answer analysis and feedback

[1195] The server analyzes the user's answers and determines whether they are correct or incorrect. Based on the analysis results, it generates appropriate feedback, which may include specific explanations or advice.

[1196] AI Avatar Explained

[1197] Users can select their preferred AI avatar on their device, which will then use voice and text to explain how to solve the problem and any areas they don't understand, making it easier for users to understand the information visually and audibly.

[1198] Constant progress monitoring and adjustment of study plans

[1199] The server periodically checks the user's learning progress and adjusts the learning plan as needed. For example, if a particular weak area is overcome, the server will provide new questions that focus on that area. This allows the learner to continuously improve their academic ability.

[1200] Specific examples

[1201] As a concrete example, a user sends the results of an arithmetic test to a server, which determines that the user has difficulty calculating fractions. The server then generates a simple fraction addition problem (e.g., 1 / 2 + 1 / 4) and sends it to the device. The user solves the problem and enters the answer on the device, and the server determines whether the answer is correct or incorrect. If the answer is incorrect, an AI avatar selected by the user will provide an explanation, such as, "When adding fractions, you first need to make the denominators the same."

[1202] This allows users to gain a deeper understanding and learn more efficiently than with conventional one-way learning methods.

[1203] The processing flow will be explained below.

[1204] Step 1: Collect data

[1205] The user uses the terminal to input daily homework results and past test results.

[1206] The terminal transmits the input data to the server.

[1207] The server stores the received data in a database.

[1208] Step 2: Identify your weaknesses

[1209] The server analyzes the stored data using AI algorithms.

[1210] Based on the analysis results, the server identifies areas in which the learner is weak.

[1211] For example, the server analyzes test results and homework answer trends to extract areas where learners frequently make mistakes.

[1212] Step 3: Generate the problem

[1213] The server generates questions related to the identified weaknesses.

[1214] The server uses an AI algorithm to create sets of questions that gradually increase in difficulty, starting with easy questions.

[1215] The server sends the generated questions to the terminal.

[1216] Step 4: State the problem

[1217] The terminal presents the questions sent from the server to the user.

[1218] The user answers the questions presented on the terminal.

[1219] The terminal transmits the user's answer results to the server.

[1220] Step 5: Analysis and feedback of answers

[1221] The server analyzes the user's answers and determines whether they are correct or incorrect.

[1222] The server generates appropriate feedback based on the accuracy of the answer and the time it took to answer.

[1223] The server transmits the feedback content to the terminal.

[1224] Step 6: AI Avatar Explained

[1225] Users select their preferred AI avatar on their device.

[1226] The device displays the commentary sent from the server in the form of an AI avatar.

[1227] The AI ​​avatar uses voice and text to explain how to solve the problem and key points.

[1228] Step 7: Regularly monitor your progress and adjust your study plan

[1229] The server periodically checks the user's learning progress.

[1230] The server adjusts the study plan based on the study progress.

[1231] For example, if a particular weak area is overcome, the server generates new problems for that weak area.

[1232] Example 1

[1233] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1234] Conventional learning support systems do not provide sufficient, effective learning support that meets the individual needs of learners. Furthermore, they have difficulty identifying areas in which learners are weak, providing questions that address those areas, and providing appropriate feedback. This creates the problem of ineffectively improving learners' academic abilities.

[1235] 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.

[1236] In this invention, the server includes means for collecting data on the learner's grades and assignments, means for identifying the learner's weak points in academic ability based on the collected data, and means for automatically generating questions related to the learner's weak points in academic ability, thereby enabling effective learning support and feedback to be provided to the learner according to their individual learning needs.

[1237] "Academic performance data" refers to the scores and evaluation results of tests and assignments taken by learners.

[1238] "Assignment data" refers to the answers and progress of homework and practice problems that a learner has completed.

[1239] "Academic weaknesses" refer to areas of study where a learner is particularly weak or areas where they frequently make mistakes.

[1240] "Automatically generated" refers to software or algorithms creating problems or challenges without human intervention.

[1241] "Virtual character" means a digital avatar generated by a computer program that can communicate through voice and text.

[1242] "Instructional information" refers to educational information such as explanations, advice, and feedback provided to learners.

[1243] "Machine learning algorithms" refer to computational methods and models that allow computers to learn from data and automatically improve their performance.

[1244] A "study plan" refers to the tasks and study schedule that a learner must complete each day.

[1245] This invention provides a system that provides learning support tailored to the individual needs of learners. This system is primarily composed of a server, a terminal, and learners (users). The server performs central data analysis and question generation, while the terminal functions as an interface for the user.

[1246] System Overview

[1247] This system helps learners improve their academic ability through the following processes:

[1248] Hardware and software used

[1249] The hardware used includes servers (e.g., cloud-based servers) and user devices (e.g., tablets, PCs, etc.), while the software includes database management systems (e.g., MySQL or PostgreSQL), AI algorithms (e.g., Python's scikit-learn or TensorFlow), and generative AI models (e.g., GPT-4).

[1250] Data collection

[1251] Users use the device to input their daily homework and past test results. When the user enters this data through a dedicated application on the device or a web interface, the device sends it to the server, which stores the received data in a database.

[1252] Identifying weak areas

[1253] The server uses machine learning algorithms to analyze the stored data, such as Python's scikit-learn or TensorFlow, to identify areas where learners frequently make mistakes and spend time, allowing the server to identify weak areas in the learner's academic performance.

[1254] Problem Generation

[1255] The server generates questions using a generative AI model based on the identified weak areas. The generative AI model uses GPT-4 or similar. The server creates a prompt and sends it to the generative AI model. For example, the following prompt can be used:

[1256] Example prompt sentence:

[1257] "Your learner has difficulty with fractions. Generate a simple fraction addition problem like this: Problem: 1 / 2 + 1 / 4"

[1258] Problem Statement

[1259] The terminal receives the questions sent from the server and presents them to the user. The user answers the questions displayed on the terminal and enters the answers into the terminal.

[1260] Answer analysis and feedback

[1261] The device sends the user's answers to the server, which analyzes them and identifies correct and incorrect answers. The server then generates feedback based on the analysis results, including specific explanations and advice.

[1262] AI Avatar Explained

[1263] Users select their preferred virtual character on their device, who then uses voice (e.g., text-to-speech API) and text to explain how to solve the problem or any areas of confusion, helping users understand the information visually and audibly.

[1264] Constant progress monitoring and adjustment of study plans

[1265] The server periodically checks the learner's progress and adjusts the learning plan as needed. For example, if a particular academic weakness is overcome, the server will provide new questions that focus on that weakness. This iterative process allows the learner to continuously improve their academic performance.

[1266] In this way, the learning support system of the present invention responds to the individual needs of each learner and provides effective learning support, thereby enabling each learner to improve their academic ability efficiently and continuously.

[1267] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1268] Step 1:

[1269] Users input their daily homework results and past test results through their devices. Specifically, they input this data using a dedicated application on their device or a form provided in the web interface. The input data includes information such as the content of the questions the user answered, their answers, and the time it took to complete them. The device then sends this input data to the server.

[1270] Input: Homework and test result data entered by the user into the device

[1271] Output: Data sent from the device to the server

[1272] Step 2:

[1273] The server stores the received data in a database. By using a database management system (e.g., MySQL or PostgreSQL), all training data can be managed efficiently.

[1274] Input: Data sent from the terminal

[1275] Output: Data stored in the database

[1276] Step 3:

[1277] The server uses machine learning algorithms (e.g., Python's scikit-learn or TensorFlow) to analyze the stored data. The server applies the algorithms to identify areas where learners frequently make mistakes and spend time on grades and assignments. This allows the server to identify the learner's academic weaknesses.

[1278] Input: Grade data and assignment data stored in the database

[1279] Output: Identified academic weaknesses

[1280] Step 4:

[1281] The server uses a generative AI model (e.g., GPT-4) to generate problems based on the identified weaknesses. The server creates a prompt like the following and sends it to the generative AI model: "The learner has difficulty with fraction calculations. Please generate a simple fraction addition problem like this: Problem: 1 / 2 + 1 / 4."

[1282] Input: Identified academic weaknesses

[1283] Output: The problem returned by the generative AI model

[1284] Step 5:

[1285] The device receives the questions sent from the server and presents them to the user. Specifically, the user can view the questions through a dedicated app on the device or a web interface. The user answers the presented questions.

[1286] Input: Question sent by server

[1287] Output: The problem presented to the user

[1288] Step 6:

[1289] The user inputs the answer into the terminal, which then transmits the answer to the server.

[1290] Input: Answer result entered by the user into the terminal

[1291] Output: Answer results sent from the device to the server

[1292] Step 7:

[1293] The server analyzes the answers and determines whether they are correct or incorrect. The server uses a Python script to compare the answers with the correct answers. The server then generates specific feedback based on the analysis results. The generated feedback includes specific explanations and advice.

[1294] Input: Answer results sent from the device

[1295] Output: Feedback information

[1296] Step 8:

[1297] Users select their preferred virtual character on their device, and the virtual character uses voice and text to explain how to solve the problem or provide explanations for any unclear points. The voice is generated using a text-to-speech API, and the information is provided visually using a 3D model created in Unity.

[1298] Input: Feedback information

[1299] Output: Explanation by virtual character

[1300] Step 9:

[1301] The server periodically checks the student's progress and adjusts the learning plan as necessary. The server periodically (e.g., weekly or monthly) analyzes the performance data and generates new questions and updates the learning plan if the student's academic weaknesses are overcome.

[1302] Input: Learning progress data

[1303] Output: Updated lesson plan

[1304] The above are the specific processing steps of the program of this system.

[1305] (Application example 1)

[1306] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1307] Conventional AI tutoring systems provide individual learning support based on learners' test results and homework results, but they do not provide immediate learning support in physical stores. Furthermore, the lack of an intuitive visual or audio interface makes it difficult for learners to engage and for effective learning to occur.

[1308] 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.

[1309] In this invention, the server includes means for collecting test results and homework results of learners, means for identifying areas of weakness of learners based on the collected data, and means for generating questions related to the areas of weakness of learners. This enables learners to use a visual device to send book and teaching material information to the server in real time, receive questions through the visual device, collect answers through voice input, and receive explanations from an AI avatar.

[1310] A "learner" is an individual seeking to acquire knowledge or skills.

[1311] A "visual device" is a device that provides information visually, and generally refers to smart glasses or head-mounted displays.

[1312] A "server" is a computer system that performs central processing such as collecting, analyzing, and storing data.

[1313] "Test results" refers to the results and answers given by the learner on the test.

[1314] "Homework results" refers to the grades and answers to homework submitted by the learner.

[1315] A "weak area" is an area in which a learner has particular difficulty understanding or mastering.

[1316] "Questions" refer to learning tasks or exercises that learners must answer.

[1317] "Means of collection" refers to the methods and techniques used to collect the necessary data.

[1318] "Analytical means" refers to the methods and techniques used to analyze data and derive meaningful information.

[1319] "Feedback" refers to evaluation and advice given to a learner based on their answers and behavior.

[1320] An "AI avatar" refers to a virtual character that uses artificial intelligence to provide learning assistance.

[1321] "Voice input" refers to a method in which learners input information by voice.

[1322] "Progress" refers to a learner's learning progress and achievements.

[1323] A "learning plan" refers to the learning content and schedule planned to help a learner achieve their learning goals.

[1324] This invention is an AI tutoring system that provides learning support tailored to the individual needs of each learner. This system is primarily composed of a server, a terminal, and a learner, and utilizes visual devices and voice input functions to provide instant learning support in a brick-and-mortar store.

[1325] System Overview

[1326] The server collects the learner's test results and homework results, analyzes them, and identifies the learner's weak areas. It then generates questions that address the identified weak areas and sends the questions to the learner's device or visual device. The learner receives the questions through the visual device and answers them using voice input. The answers are sent to the server, which provides appropriate feedback based on the analysis results.

[1327] Hardware and Software

[1328] Hardware used

[1329] Visual devices: smart glasses or head-mounted displays (e.g., smart glasses)

[1330] Device: Smartphone or tablet

[1331] Server: A computer system that analyzes and stores data (e.g., cloud services).

[1332] Software used

[1333] Data collection and analysis: Python scripts

[1334] Server communication: RESTful API

[1335] Audio Description: Text-to-Speech (TTS) engine (e.g., Google Cloud Text-to-Speech)

[1336] Specific examples of explanation

[1337] Let's say a student chooses a book on "fraction calculation" at a physical store. When the student puts on the smart glasses and reads the book information, that information is sent to the server in real time. The server combines this with past learning data and determines that the student has difficulty with "fraction addition." The server then generates a simple "fraction addition" problem (e.g., 1 / 2 + 1 / 4) and sends it to the smart glasses. The student answers via voice input, and the answer is sent to the server. The server analyzes the answer, determines whether it is correct or incorrect, and sends an explanation to the smart glasses via voice and text, such as "When adding fractions, you first need to make the denominators the same," with an AI avatar providing the explanation. This allows the student to deepen their understanding on the spot.

[1338] Prompt Sentence Examples

[1339] The generative AI model can generate explanations for specific learning items by inputting the following prompts:

[1340] "setting"

[1341] Generate detailed explanations to help students understand fraction addition.

[1342] "prompt"

[1343] When adding fractions, you first need to make the denominators the same. To do this, convert the fractions so that the denominators are the lowest common multiple. For example, to calculate 1 / 2 + 1 / 4, convert 1 / 2 to 2 / 4, then calculate 2 / 4 + 1 / 4... (Detailed steps continue.)

[1344] The system allows learners to receive personalized learning support in real time within the store, providing an intuitive learning environment using visual and audio.

[1345] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1346] Step 1:

[1347] A user uses a visual device to select a book or educational material in a physical store and reads its information. The visual device then uses a camera to scan the barcode or QR code on the book or educational material to obtain the relevant information. The obtained information is sent to the server in real time. The input is the information of the book or educational material read by the visual device, and the output is the information sent to the server.

[1348] Step 2:

[1349] The server compares the received information on books and teaching materials with the learner's past test results and homework results to identify the learner's weak areas. Past learning data for each learner is stored in the database, and an AI algorithm is used to analyze the learner's weak areas. The input is information on books and teaching materials and past learning data, and the output is the identified learner's weak areas.

[1350] Step 3:

[1351] The server generates problems in stages that address the identified weak areas. An AI model is used to generate problems with increasing difficulty, from simple to advanced. The generated problems are sent to a visual device. The input is the identified weak areas and a problem generation prompt, and the output is the generated problem set.

[1352] Step 4:

[1353] The visual device presents the questions received from the server to the user. The questions are displayed on the visual device's display, and the user confirms them. Using the visual device's voice input function, the user inputs the answer to the question by voice. The input is the question sent from the server, and the output is the user's voice answer.

[1354] Step 5:

[1355] The visual device converts the user's voice response into text data and sends it to the server. The voice response is converted into text data using a Speech-to-Text (STT) engine. The input is the user's voice response, and the output is the response as text data.

[1356] Step 6:

[1357] The server analyzes the user's answers and determines whether they are correct or incorrect. An AI algorithm is used for the analysis to evaluate the accuracy of the answers. The input is the user's answer text, and the output is the analysis result of the answer.

[1358] Step 7:

[1359] The server generates appropriate feedback based on the analysis results. The appropriate feedback includes points of mistakes and explanations. The feedback content is sent to the visual device in audio and text format. The input is the analysis result of the answer, and the output is the feedback information.

[1360] Step 8:

[1361] The visual device presents the feedback received from the server to the user, and the AI ​​avatar provides explanations in voice and text. For example, an explanation such as "When adding fractions, you must first make the denominators the same." The input is the feedback information sent from the server, and the output is the explanation to the user.

[1362] 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.

[1363] System Overview

[1364] This invention is an AI tutoring system that provides learning support tailored to the individual needs of each learner, and also combines it with an emotion engine that recognizes the user's emotions and provides an optimal learning environment. This system is primarily composed of a server, a terminal, a learner (user), and an emotion engine. The server performs central data analysis and question generation, the terminal provides an interface for the user, and the emotion engine recognizes the user's emotions.

[1365] Data collection

[1366] Users input their daily homework results and past test results through the device. The device then sends this data to the server, which stores the received data in a database for later analysis. The device is also equipped with a camera and microphone, which allow the emotion engine to analyze the user's facial expressions and voice.

[1367] Identifying weak areas

[1368] Based on the stored data, the server uses AI algorithms to identify areas where learners struggle, analyzing test scores and homework answers to identify areas where learners frequently make mistakes and where they spend a lot of time on specific questions.

[1369] Problem Generation

[1370] The server generates a set of problems of increasing difficulty for the identified weak areas, starting with easy problems and gradually increasing in difficulty. The generated problems are sent to the device.

[1371] Problem Statement

[1372] The terminal presents the questions sent from the server to the user. The user answers the questions and enters the answers into the terminal. The terminal then transmits the answers to the server.

[1373] Answer analysis and feedback

[1374] The server analyzes the user's answers and determines whether they are correct or incorrect. Based on the analysis results, it generates appropriate feedback and sends it to the device. This feedback includes specific explanations and advice.

[1375] AI Avatar Explained

[1376] Users select their preferred AI avatar on their device. The emotion engine analyzes the user's facial expressions and voice to recognize their emotional state. The AI ​​avatar uses voice and text to provide explanations tailored to the user's level of understanding or detailed explanations of specific issues. It also provides reassuring and encouraging comments based on the emotion engine's results.

[1377] Constant progress monitoring and adjustment of study plans

[1378] The server periodically checks the user's learning progress and adjusts the learning plan as needed. For example, if a particular weakness is overcome, it will provide new questions that focus on that weakness. It also takes into account the learner's emotional state and creates a learning plan that is tailored to that state.

[1379] Specific examples

[1380] For example, if a user submits the results of a math test to a server, the server will determine that the user has difficulty calculating fractions. The server then generates a simple fraction addition problem (e.g., 1 / 2 + 1 / 4) and sends it to the device. While the user is solving the problem, the emotion engine analyzes the user's facial expressions and voice and determines that the user is nervous. At this point, an AI avatar will provide encouragement and explanation, such as, "Relax and think about it. First, you need to make the denominators the same."

[1381] This allows users to have an optimal learning experience that is tailored to their individual feelings and level of understanding, compared to conventional one-way learning methods.

[1382] The processing flow will be explained below.

[1383] Step 1: Collect data

[1384] Users input their daily homework results and past test results into the terminal.

[1385] The terminal transmits the input data to the server.

[1386] The server stores the received data in a database.

[1387] In addition, the device's camera and microphone are used to capture the user's facial expressions and voice and send them to the emotion engine.

[1388] Step 2: Recognize emotions

[1389] The emotion engine analyzes the user's facial expressions and voice data to recognize the user's emotional state.

[1390] The recognized emotion data is sent to the server and reflected in future feedback and commentary.

[1391] Step 3: Identify your weaknesses

[1392] The server analyzes the saved homework and test results using AI algorithms.

[1393] Based on the analysis results, the server identifies areas in which the learner is weak.

[1394] For example, the server will extract areas where you frequently make mistakes or where you spend a lot of time on specific problems.

[1395] Step 4: Generate the problem

[1396] The server generates questions related to the identified weaknesses.

[1397] The server uses an AI algorithm to create sets of questions that gradually increase in difficulty, starting with easy questions.

[1398] The server sends the generated questions to the terminal.

[1399] Step 5: State the problem

[1400] The terminal presents the questions sent from the server to the user.

[1401] The user answers the questions presented to them.

[1402] The terminal collects the answer results and sends them to the server.

[1403] Step 6: Analysis and feedback of answers

[1404] The server analyzes the user's answers and determines whether they are correct or incorrect.

[1405] The server generates appropriate feedback based on the accuracy of the answer and the time it took to answer.

[1406] The server transmits the generated feedback content to the terminal.

[1407] Step 7: AI Avatar Explained

[1408] Users select their preferred AI avatar on their device.

[1409] The terminal receives the result of the emotion engine and obtains commentary content based on the user's emotional state from the server.

[1410] The AI ​​avatar uses voice and text to explain how to solve the problem and key points.

[1411] For example, if the user is nervous, the AI ​​avatar will offer encouraging comments such as, "Relax and think about it. First, we need to make the denominators the same."

[1412] Step 8: Check your progress and adjust your study plan

[1413] The server periodically checks the user's learning progress.

[1414] The server adjusts the learning plan based on the learning progress.

[1415] Once a particular weakness has been overcome, the server provides new problems that focus on that weakness.

[1416] The server also takes into account the recognized emotion data to provide the user with an optimal learning environment.

[1417] Example 2

[1418] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1419] Conventional learning support systems rely on one-way information transmission and are unable to take into account the individual learner's level of understanding or emotional state in a timely manner. As a result, appropriate feedback based on the learner's weak areas or level of understanding is not provided, making it difficult to provide effective learning support. In addition, it is difficult to provide a learning environment that takes into account the learner's emotional state, which is likely to reduce motivation to learn.

[1420] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for collecting the learner's assessment results, means for identifying the learner's learning difficulties based on the collected data, means for generating assignments related to the learner's learning difficulties, means for presenting the generated assignments to the learner, means for analyzing the learner's answers and providing appropriate feedback, means for using an AI character, means for periodically checking the learner's progress and adjusting the learning plan, means for analyzing the learner's emotional state, and means for adjusting the feedback based on the analysis results. This makes it possible to provide an optimal learning environment according to the individual learner's level of understanding and emotional state, thereby enabling effective and motivating learning support.

[1421] "Learner's assessment results" refers to information that indicates the results of learning and performance, such as test results and homework results that the learner has completed.

[1422] "Means of collection" refers to the technical methods, devices, and software used to collect learner assessment results and other necessary data.

[1423] "Learning difficulties" refer to areas where a learner has a lack of understanding or makes many mistakes in a particular academic subject or problem.

[1424] "Means of identification" refers to technical methods, devices, and software that analyze and extract learners' learning difficulties based on collected data.

[1425] "Assignments" are learning items such as questions and exercises that are generated based on the learner's learning difficulties.

[1426] "Generation means" refers to technical methods, devices, or software that automatically generate tasks related to learning difficulties.

[1427] "Presentation means" refers to the technical methods, devices, and software used to visually and audibly present the generated assignment to the learner.

[1428] "Means for analyzing answers" refers to technical methods, devices, and software for analyzing answers provided by learners and determining whether they are correct or incorrect.

[1429] "Feedback" refers to information such as explanations, advice, and instructions for retrying that are provided to learners based on the analyzed answers.

[1430] An "artificial intelligence character" is a computer-generated character used to provide feedback and commentary through interaction with a user.

[1431] "Means for regularly checking learner progress" refers to technological methods, devices, or software that periodically evaluate and analyze learner progress.

[1432] "Means for adjusting the learning plan" refers to technological methods, devices, and software for adjusting the learning content and pace to suit the learner based on the results of regular progress checks.

[1433] "Emotional state" indicates the user's psychological and emotional state and is determined based on data such as facial expressions and voice.

[1434] "Means of analysis" refers to the technical methods, devices, and software that analyze the data necessary to determine emotional state.

[1435] "Means for adjusting feedback based on the analysis results" refers to technical methods, devices, or software that adjust the content and format of the feedback provided based on the results of sentiment analysis.

[1436] This invention is an AI tutoring system that provides learning support tailored to the individual needs of learners, and also combines it with an emotion engine that recognizes the user's emotions and provides an optimal learning environment. This system is primarily composed of a server, a terminal, a learner (user), and an emotion engine. Each component is described in detail below.

[1437] Data collection

[1438] Users use their devices to input learning data, including daily homework results and past test results. The devices first store this data locally and then send it to the server. The devices used include, for example, personal computers and tablets. The devices are also equipped with cameras and microphones, which capture the user's facial expressions and voice. The captured facial and voice data is analyzed in real time by the emotion engine.

[1439] Data storage and analysis

[1440] The server stores the received learning data in a database, which makes it available for later analysis. Based on this data, the server uses an AI algorithm to identify the learner's learning difficulties. For example, from math test data, the server can identify that the learner has difficulty calculating fractions.

[1441] Assignment generation and presentation

[1442] The server generates tasks that address the identified learning difficulties. These tasks are designed to increase in difficulty in stages, adjusting to include beginner level problems and gradually more difficult problems. The generated tasks are sent to the terminal, which then presents them to the user. The user answers the presented tasks and re-enters the answers into the terminal. The entered answers are then sent from the terminal to the server.

[1443] Answer analysis and feedback

[1444] The server analyzes the user's answers and determines whether they are correct or incorrect. Based on the analysis results, it generates appropriate feedback and sends it to the device. The feedback includes specific explanations and advice. For example, if the answer is incorrect, the server provides advice such as, "You made a mistake, but you need to make the denominators the same."

[1445] Emotion Engine and AI Avatar

[1446] Users select their preferred AI character on the device. The emotion engine analyzes the user's facial expressions and voice to recognize their emotional state in real time. The AI ​​character uses voice and text to provide explanations and feedback based on the user's level of understanding. For example, if the character determines that the user is nervous, it will make encouraging comments such as, "Relax and try again."

[1447] Check your progress and adjust your study plan

[1448] The server periodically monitors the user's learning progress and adjusts the learning plan as needed. It generates new assignments based on the user's progress and provides feedback that focuses on new learning difficulties or areas that need improvement. It also takes into account the learner's emotional state and builds a learning plan that is tailored to that state.

[1449] Specific examples

[1450] For example, if a user sends the results of an arithmetic test to a server, the server analyzes the results and determines that the user has difficulty calculating fractions. The server then generates a simple fraction addition problem (e.g., 1 / 2 + 1 / 4) and sends it to the device. While the user solves the problem and the answer is sent to the server, the emotion engine analyzes the user's facial expressions and voice and determines that the user is nervous. At this point, an AI character provides encouragement and explanation, such as, "Relax and think about it. First, you need to make the denominators the same."

[1451] Examples of input prompts for a generative AI model include:

[1452] Describe a learning support system in which a user inputs past test data, a server extracts learning difficulties from that data, and generates tasks. An emotion engine analyzes the user's facial expressions and voice to provide real-time feedback.

[1453] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1454] Step 1: Data entry

[1455] Users use a device to input homework results and past test results. This input data may include answers in text format or multiple choice format. Specifically, the user enters data into an input box on the device screen and presses the "Submit" button. The input data includes homework scores, test scores, answers, etc., and this becomes the basic data for the system.

[1456] Input: Homework results, test results

[1457] Output: Save to local database, move data to next step

[1458] Step 2: Send data

[1459] The device sends the homework and test result data entered by the user to the server. This operation is performed using the device's communication function and is executed automatically by the user. For example, when the send button is pressed, the device packs the data and sends it to the server's API endpoint.

[1460] Input: Data from the local database

[1461] Output: Upload data to the server

[1462] Step 3: Saving to the database

[1463] The server analyzes the received homework and test results data and stores them in a database. The data stored includes the score for each question, the answer content, and whether the answer was correct or incorrect. This operation is basically performed by issuing an insert query to the database.

[1464] Input: Received data

[1465] Output: Save data to database

[1466] Step 4: Collecting emotion data

[1467] The device's camera and microphone are used to collect the user's facial expressions and voice data in real time, and the collected data is sent to the emotion engine.

[1468] Input: User's facial expression and voice data

[1469] Output: Sending data to the emotion engine

[1470] Step 5: Analyze the sentiment data

[1471] The emotion engine analyzes the collected facial and voice data to determine the user's emotional state. It uses facial recognition and voice analysis algorithms. Specifically, if the user is smiling, it identifies the emotion as "relaxed," while if the user is smiling, it identifies the emotion as "tense."

[1472] Input: facial expression and voice data

[1473] Output: Emotion analysis results

[1474] Step 6: Identifying learning difficulties

[1475] The server analyzes the stored learning data using an AI algorithm to identify the learner's learning difficulties. For example, based on test data, it identifies that the learner has difficulty calculating fractions. During this process, the correct answer rate and answer time for each problem are also used as part of the analysis.

[1476] Input: Training data stored in a database

[1477] Output: Identification of learning difficulties

[1478] Step 7: Generate an issue

[1479] The server generates a set of increasingly difficult problems for each identified learning difficulty, starting with easy problems (e.g., basic addition of fractions) and gradually increasing in difficulty (e.g., subtraction of mixed fractions). This problem generation is done using an AI problem generation algorithm.

[1480] Input: Results of identifying learning difficulties

[1481] Output: The generated set of issues

[1482] Step 8: State the problem

[1483] The terminal presents the task sent from the server to the user. The displayed task consists of a text question and an answer input box. The terminal provides an interface for the user to perform the operations necessary to solve the task.

[1484] Input: Issue data from the server

[1485] Output: Present the task to the user

[1486] Step 9: Enter and submit your answers

[1487] The user inputs the answer to the task presented on the terminal and presses the send button to send the answer to the server. This data includes the user's answer content, answer time, etc.

[1488] Input: User's answer

[1489] Output: Send answer data to the server

[1490] Step 10: Analyze the solution

[1491] The server analyzes the received answer data and determines whether the answer is correct or incorrect. It checks the validity of the answer using an AI algorithm, and if the answer is correct, it will state that, and if it is incorrect, it will explain the reason.

[1492] Input: User's answer data

[1493] Output: Analysis results (correct / incorrect)

[1494] Step 11: Generate feedback

[1495] The server generates appropriate feedback based on the analysis results, including specific advice such as "Well done!" if the answer is correct, or "Sorry, try calculating it again with the same denominator" if the answer is incorrect.

[1496] Input: Answer analysis result

[1497] Output: Generated feedback

[1498] Step 12: AI Avatar Commentary

[1499] Users select their preferred AI avatar on their device, and the emotion engine analyzes the user's facial expressions and voice. Based on this, the AI ​​avatar provides the user with commentary and encouragement in voice and text. For example, if it determines that the user is nervous, it will say, "Relax and think about it."

[1500] Input: Sentiment analysis results, feedback content

[1501] Output: Audio and text commentary by an AI avatar

[1502] Step 13: Check and adjust your progress

[1503] The server periodically checks the user's learning progress and creates a new learning plan, taking into account past learning data, feedback, and emotional state. Based on the progress data, the server provides new assignments and feedback, creating an optimal learning environment for the user.

[1504] Input: Learning data and emotion data stored in the database

[1505] Output: New lesson plan, next set of assignments

[1506] (Application example 2)

[1507] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1508] Conventional learning support systems provide feedback based on individual learners' progress and level of understanding, but they are unable to take into account the learner's emotional state. As a result, if a learner feels nervous or stressed, appropriate support may not be provided, which could result in a decrease in learning effectiveness. To address this issue, a system is needed that provides optimal feedback according to the learner's emotional state.

[1509] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting the learner's test results and homework results, means for identifying the learner's weak areas based on the collected data, means for generating questions related to the learner's weak areas, means for presenting the generated questions to the learner, means for analyzing the learner's answers and providing appropriate feedback, means for using an AI avatar to provide explanations, means for periodically checking the learner's progress and adjusting the learning plan, means for determining the user's emotional state using an emotion recognition engine, and means for adjusting the learning content and feedback based on the user's emotional state. This enables optimal learning support tailored to the individual learning needs of the learner while taking the learner's emotional state into consideration.

[1510] "Student" refers to an individual who is engaged in learning activities using a learning support system.

[1511] "Means for collecting test results and homework results" refers to devices or software that input learners' daily learning results into the system for recording and storage.

[1512] A "means for identifying areas of difficulty" is a device or software that analyzes the collected data to detect areas or fields in which the learner has difficulty understanding.

[1513] A "means for generating questions" is a device or software that automatically generates appropriate study questions based on identified areas of weakness.

[1514] A "means for presenting questions" is a device or software that presents the generated questions to the learner visually or audibly.

[1515] "Means for analyzing answers and providing appropriate feedback" refers to a device or software that evaluates the learner's answers, takes the results into consideration, and returns useful advice or explanations.

[1516] "Means using AI avatars" refers to devices or software that use artificial intelligence technology to utilize virtual characters that provide answers and explanations to learners.

[1517] A "means for checking learning progress and adjusting learning plans" is a device or software that periodically evaluates a learner's learning progress and modifies or improves the learning plan based on that progress.

[1518] An "emotion recognition engine" is a device or software equipped with artificial intelligence technology that analyzes a learner's facial expressions and voice to determine their emotional state.

[1519] The "means for adjusting learning content and feedback based on emotional state" refers to a device or software that provides appropriate learning content and feedback according to the learner's emotional state based on information obtained from an emotion recognition engine.

[1520] This invention is an AI tutoring system to support employee training in brick-and-mortar stores, recognizing user emotions to provide an optimal learning environment. The system is primarily composed of a server, a terminal, users, and an emotion recognition engine. The server performs central data analysis and generates questions, the terminal provides an interface for users, and the emotion recognition engine recognizes user emotions.

[1521] Hardware and software configuration:

[1522] 1. Server: Mainly responsible for data analysis, question generation, and learning progress management. Software used includes AI algorithms, database management systems, and web servers (e.g., Apache).

[1523] 2. Device: A smartphone is used to provide the user interface, present learning content, and display feedback. The device is equipped with a camera and microphone, which are used to collect the user's facial expressions and voice. The recommended hardware is a smart device with a camera (e.g., iPhone, Android smartphone). The software used includes the UI framework used by the application (e.g., React Native, Flutter, etc.).

[1524] 3. Emotion Recognition Engine: Analyzes the learner's facial expressions and voice to determine their emotional state in real time. The software used includes machine learning libraries such as TensorFlow and OpenCV.

[1525] Explanation of program operation:

[1526] First, a user (such as a new employee) uses a device to log in to a training application. The application retrieves the user's learning history from the server and presents the next learning content to be tackled. This learning content is generated by an AI algorithm on the server based on the collected data. While the user is learning, the device's built-in camera and microphone collect the user's facial expressions and voice in real time. This data is analyzed by an emotion recognition engine to determine the user's emotional state.

[1527] Once the user's facial and voice data is collected, it is passed to an emotion recognition engine. This engine uses TensorFlow and OpenCV to analyze the user's emotional state in real time. For example, if the application determines that the user is nervous, it will display a message such as, "Relax, and try to smile first." The learning content and feedback are then dynamically adjusted to provide appropriate support tailored to the user's emotional state.

[1528] Examples:

[1529] Consider a scenario where a new employee is undergoing customer service skill training. When the user launches the application and starts the training module, the server generates the following scenario based on the user's progress data:

[1530] Example prompt sentence:

[1531] "Think about the ideal way to serve customers in the following situation: When there are many customers in the store during busy hours."

[1532] While the user is learning based on the prompts, the emotion recognition engine analyzes the user's emotions and provides relaxing messages such as "Relax, try to smile first" if the user is nervous. In this way, it is possible to provide support tailored to the needs of each individual user while maximizing the learning effect.

[1533] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1534] Step 1:

[1535] The user logs in to the training application using a device (smartphone). The user's ID and password are required as input, and are verified by the user authentication server. The user's learning history and current progress are obtained as output. If authentication is successful, the application's home screen is displayed, showing the next learning content to be tackled (e.g., customer service skills training).

[1536] Step 2:

[1537] The server receives the user's login information and learning history and generates the next learning content. The user's past learning history stored in the learning database is used as input. Using data analysis and AI algorithms, questions and practice scenarios are generated based on the individual's learning progress. The generated learning content is sent to the device as output.

[1538] Step 3:

[1539] The terminal receives the learning content sent from the server and presents it to the user. The learning content data from the server is used as input. When the user logs in to their account and starts a training session, the learning content (e.g., a customer service scenario) is displayed on the screen. As output, the learning content is presented on the display screen.

[1540] Step 4:

[1541] The user works through exercises and scenarios according to the learning content presented. The user's answers and reactions are entered as input. The device's built-in camera and microphone also collect the user's facial expressions and voice data. This data is sent to the emotion recognition engine in real time.

[1542] Step 5:

[1543] The emotion recognition engine analyzes the user's facial expressions and voice data in real time to determine the user's emotional state. Image data and voice data collected from the camera and microphone are used as input. The emotional state (e.g., nervous, relaxed, etc.) is identified using TensorFlow and OpenCV. The determined emotional state is sent to the device as output.

[1544] Step 6:

[1545] The server provides appropriate feedback to the user based on the determined emotional state. As input, the results of the emotion recognition engine and the user's learning progress data are used. Using an AI algorithm, advice, explanations, and encouraging messages are generated according to the user's emotional state. As output, the generated feedback is sent to the device.

[1546] Step 7:

[1547] The terminal displays the feedback sent from the server to the user. As input, the feedback data from the server is used. A text message or a voice message according to the emotional state is displayed on the user's screen. As output, the user receives feedback through voice and text.

[1548] Step 8:

[1549] The user continues learning based on the feedback and takes on new challenges. The answers and reactions after receiving the feedback are used as inputs, which dynamically adjust the next learning content. The output is the learning progress sent back to the server, and the whole system repeats the learning cycle.

[1550] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice 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 voice data.

[1551] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1552] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1553] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1554] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1555] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1556] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1557] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1558] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1559] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1560] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1561] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1562] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1563] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1564] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1565] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1566] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1567] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1568] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1569] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1570] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1571] The following is further disclosed regarding the above embodiment.

[1572] (Claim 1)

[1573] A means of collecting learners' test results and homework results;

[1574] A means of identifying areas of weakness of learners based on the collected data;

[1575] a means for generating questions related to the learner's weak areas;

[1576] a means for presenting the generated questions to the learner;

[1577] A means of analyzing learners' answers and providing appropriate feedback;

[1578] Using an AI avatar to provide commentary,

[1579] A means to regularly check learner progress and adjust learning plans;

[1580] A system including:

[1581] (Claim 2)

[1582] 10. The system of claim 1, further comprising means for the AI ​​avatar to provide audio and text commentary when providing learning feedback.

[1583] (Claim 3)

[1584] 10. The system of claim 1, further comprising means for using an AI algorithm to identify areas of weakness of the learner.

[1585] "Example 1"

[1586] (Claim 1)

[1587] a means for collecting learner performance data and assignment data;

[1588] A means of identifying learners' academic weaknesses based on the collected data;

[1589] A means for automatically generating questions related to a learner's academic weaknesses;

[1590] a means for displaying the generated questions to the learner;

[1591] A means for analyzing the learner's answers and providing appropriate instructional information;

[1592] A means using a virtual character to provide commentary;

[1593] a means of regularly monitoring learners' progress and adjusting their learning plans;

[1594] A system including:

[1595] (Claim 2)

[1596] 10. The system of claim 1, further comprising means for the virtual character to provide audio and text commentary when providing feedback.

[1597] (Claim 3)

[1598] 10. The system of claim 1, further comprising means for using a machine learning algorithm to identify academic weaknesses of the learner.

[1599] "Application Example 1"

[1600] (Claim 1)

[1601] A means of collecting learners' test results and homework results;

[1602] A means of identifying areas of weakness of learners based on the collected data;

[1603] a means for generating questions related to the learner's weak areas;

[1604] a means for presenting the generated questions to the learner;

[1605] A means of analyzing learners' answers and providing appropriate feedback;

[1606] Using an AI avatar to provide commentary,

[1607] A means to regularly check learner progress and adjust learning plans;

[1608] A means for a learner to receive learning support using a visual device;

[1609] means for transmitting book and teaching material information to a server in real time through a visual device;

[1610] a means for presenting questions using a visual device and collecting answers by voice input;

[1611] A system including:

[1612] (Claim 2)

[1613] 10. The system of claim 1, further comprising means for the AI ​​avatar to provide audio and text commentary when providing learning feedback.

[1614] (Claim 3)

[1615] 10. The system of claim 1, further comprising means for using an AI algorithm to identify areas of weakness of the learner.

[1616] "Example 2: Combining Emotion Engines"

[1617] (Claim 1)

[1618] a means of collecting learner assessment results;

[1619] A means of identifying learners' learning difficulties based on the collected data;

[1620] a means for generating tasks related to the learner's learning difficulties;

[1621] a means for presenting the generated assignment to the learner;

[1622] A means of analyzing learners' answers and providing appropriate feedback;

[1623] a means for using an artificial intelligence character;

[1624] A means to regularly review learners' progress and adjust their learning plans;

[1625] a means for analyzing the learner's emotional state;

[1626] a means for adjusting the feedback based on the analysis results;

[1627] A system including:

[1628] (Claim 2)

[1629] 2. The system according to claim 1, wherein the artificial intelligence character provides commentary by voice and text.

[1630] (Claim 3)

[1631] 10. The system of claim 1, wherein an artificial intelligence algorithm is used to identify the learner's learning difficulties.

[1632] "Application example 2 when combining emotion engines"

[1633] (Claim 1)

[1634] A means of collecting learners' test results and homework results;

[1635] A means of identifying areas of weakness of learners based on the collected data;

[1636] a means for generating questions related to the learner's weak areas;

[1637] a means for presenting the generated questions to the learner;

[1638] A means of analyzing learners' answers and providing appropriate feedback;

[1639] Using an AI avatar to provide commentary,

[1640] A means to regularly check learner progress and adjust learning plans;

[1641] means for determining the emotional state of a user using an emotion recognition engine;

[1642] means for adjusting learning content and feedback based on the user's emotional state;

[1643] A system including:

[1644] (Claim 2)

[1645] The system of claim 1, wherein when providing learning feedback, the AI ​​avatar provides explanations in voice and text, and further includes means for providing advice and encouragement according to the user's emotional state.

[1646] (Claim 3)

[1647] 10. The system of claim 1, further comprising means for using an AI algorithm and means for using an emotion recognition engine to identify areas of weakness of the learner. [Explanation of symbols]

[1648] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of collecting learners' test results and homework results; A means of identifying areas of weakness of learners based on the collected data; a means for generating questions related to the learner's weak areas; a means for presenting the generated questions to the learner; A means of analyzing learners' answers and providing appropriate feedback; Using an AI avatar to provide commentary, A means to regularly check learner progress and adjust learning plans; A system including:

2. 10. The system of claim 1, further comprising means for the AI ​​avatar to provide audio and text commentary when providing learning feedback.

3. 10. The system of claim 1, further comprising means for using an AI algorithm to identify areas of weakness of the learner.

Images