System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting lesson data; means for anonymizing and generalizing the collected lesson data; means for performing machine learning using the anonymized and generalized data; means for automatically generating lesson materials based on lesson information; and means for providing the generated lesson materials.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In today's educational environment, teacher overwork is a serious problem, with many teachers suffering physical and mental strain from long working hours. Furthermore, many issues exist in school education, including bullying, school absenteeism, educational disparities, and declining academic ability. There is a particular need to ensure equal educational opportunities for children who are not attending school, and a solution is needed that both reduces teachers' workload and provides learning support for these children. Aiming to improve on-site efficiency, a system is needed that can more efficiently and effectively carry out lesson preparation, test creation and grading, and learning support for children who are not attending school. [Means for solving the problem]

[0005] The present invention provides a system including a means for collecting lesson data, a means for anonymizing and generalizing the collected lesson data, a means for performing machine learning using the anonymized and generalized data, a means for automatically generating lesson materials based on lesson information, and a means for providing the generated lesson materials.The system further includes a means for automatically generating a test based on the lesson information, a means for providing the automatically generated test, a means for inputting test results, a means for grading the input test results, a means for providing the grading results to a teacher, a means for requesting learning support for a child who is not attending school, a means for generating learning materials based on the request, a means for providing the generated learning materials, a means for monitoring learning progress, and a means for providing additional learning support as needed, thereby providing a system that can effectively provide learning support for children who are not attending school while reducing the workload of teachers.

[0006] "Class data" refers to a series of information related to classes, such as teaching materials used by teachers in classes, class content, test questions, and student feedback.

[0007] "Anonymization" refers to the process of removing personal information from collected data so that it is no longer possible to identify a specific individual.

[0008] "Generalization" refers to the process of abstracting specific data and converting it so that it can be applied broadly without being bound to a specific situation.

[0009] "Machine learning" refers to a technology in which a computer learns patterns and rules based on data and then applies the results of that learning to new data.

[0010] "Lesson information" refers to information necessary to carry out a specific lesson, and specifically includes the lesson theme, target grade, lesson time, etc.

[0011] "Class materials" refers to a series of materials and content used to conduct a class, and specifically includes PowerPoint, slides, audio files, etc.

[0012] "Test" refers to a question-based examination used to assess a student's academic ability and level of understanding.

[0013] "Request" means a request submitted by a User for a particular service or support.

[0014] "Study progress" refers to information that shows the progress and degree of achievement of a student according to their study plan.

[0015] "Learning support" refers to a set of activities, services, and tools provided to help students learn effectively. [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] This invention is an AI teaching system that aims to reduce the workload of teachers and support the learning of children who are not attending school. This system utilizes lesson data from all over the country and uses AI to automate and support lesson preparation, test creation, and grading.

[0038] Basic system configuration

[0039] The system consists of the following main components:

[0040] 1. Server

[0041] Collect, anonymize, and generalize class data.

[0042] Train and retrain AI models using machine learning.

[0043] Class materials and tests are automatically generated and provided based on class information.

[0044] Receives test result input and performs automatic scoring.

[0045] Generate learning materials based on requests for learning support for children who are not attending school and monitor their progress.

[0046] 2. Terminal

[0047] Provides an interface for teachers to enter lesson information.

[0048] Receive and download generated course materials and tests.

[0049] The teacher enters the test results and sends them to the server.

[0050] Children who are not attending school and their parents can submit requests for learning support.

[0051] 3. Users

[0052] The main users are expected to be teachers and school-refusing children (or their parents).

[0053] Teachers enter lesson information, download generated teaching materials and tests, and enter test results.

[0054] Children who are not attending school and their parents can submit requests for learning support and study using the provided teaching materials.

[0055] Program processing flow

[0056] Collection of lesson data

[0057] The server collects lesson data from each educational institution, including, for example, mathematics lesson content, teaching materials, test questions, and feedback.

[0058] The server anonymizes the collected data and removes any personal information.

[0059] The server generalizes the anonymized data, converting it so that it is not tied to any particular situation.

[0060] Training an AI model

[0061] The server trains the AI ​​model using anonymized and generalized data.

[0062] The server periodically retrains the AI ​​model using the latest lesson data stored in the database.

[0063] Automatic generation of teaching materials

[0064] The user (teacher) inputs information such as the lesson theme, target grade, and lesson time via the terminal.

[0065] The terminal transmits the input information to the server.

[0066] The server automatically generates appropriate lesson materials based on the lesson information entered, such as history lesson slides and audio files.

[0067] The server transmits the generated lesson materials to the teacher's terminal.

[0068] The user (teacher) receives the generated teaching materials and uses them in class.

[0069] Automatic test generation and scoring

[0070] The user (teacher) instructs the creation of a test on a terminal, inputting the scope, difficulty level, and format.

[0071] The terminal transmits the input information to the server.

[0072] The server automatically generates tests and provides them to teachers' terminals in PDF format, for example, generating multiple-choice and essay questions for a science test.

[0073] The user (teacher) distributes the generated test to students.

[0074] The user (teacher) enters the test results into the terminal.

[0075] The device sends the test results to the server, which then automatically scores them.

[0076] The server provides the grading results to the teacher.

[0077] Learning support for children who are not attending school

[0078] Users (children not attending school or their parents) can send requests for learning support through their devices, for example, requesting a review of math.

[0079] The terminal sends a request to the server.

[0080] The server generates appropriate learning materials based on the request, such as review videos and slides.

[0081] The server provides the generated learning materials to the user.

[0082] Users (children not attending school) study at home using the provided teaching materials.

[0083] The server monitors learning progress and provides additional learning materials and advice as needed.

[0084] The AI ​​teaching system of this invention can significantly reduce the workload of teachers and effectively support the learning of students who are not attending school. A specific example of its operation is a process in which a teacher generates PowerPoint presentations for a history lesson, distributes them to students, and then automatically generates and grades tests. Support for students who are not attending school includes providing study materials that can be used at home and tracking their learning progress.

[0085] The above is a specific embodiment for carrying out the present invention.

[0086] The processing flow will be explained below.

[0087] Program processing steps

[0088] Collection and learning of lesson data

[0089] server

[0090] Step 1:

[0091] The server collects lesson data from each educational institution, providing a mechanism for regularly uploading information such as the content of lessons taught by teachers, teaching materials used, test questions, and student feedback.

[0092] Step 2:

[0093] The server will anonymize the collected class data and remove personal information, for example by implementing algorithms that automatically detect and remove personally identifiable information such as names and student ID numbers.

[0094] Step 3:

[0095] The server then performs a generalization process on the anonymized data, specifically standardizing and storing the unique expressions of specific teachers and schools in an abstracted form.

[0096] Step 4:

[0097] The server uses anonymized and generalized data to train AI models, which use natural language processing (NLP) and machine learning (ML) algorithms to extract patterns in lesson content and effective teaching methods.

[0098] Step 5:

[0099] The server periodically uses lesson data stored in the database to retrain the AI ​​model to reflect the latest educational trends and data.

[0100] Automatic generation of teaching materials

[0101] Terminal

[0102] Step 1:

[0103] Users (teachers) use their terminals to input information such as lesson topic, target grade, lesson time, etc. A form is provided that allows users to easily input information through a dedicated interface.

[0104] Step 2:

[0105] The terminal transmits the input lesson information to the server.

[0106] server

[0107] Step 3:

[0108] Based on the lesson information received, the server searches the database for relevant lesson data and automatically generates optimal lesson materials using an AI model. For example, it creates new teaching materials by referencing past lesson slides, videos, audio files, etc. related to the same topic.

[0109] Step 4:

[0110] The server transmits the generated lesson materials to the teacher's terminal.

[0111] Terminal

[0112] Step 5:

[0113] Teachers receive the teaching materials generated on their devices, download them, edit or revise them as necessary, and use them in class.

[0114] Automated test creation and scoring

[0115] User

[0116] Step 1:

[0117] Teachers send a test creation request to the server via their terminal, specifying the scope, difficulty level, and question format (multiple choice, essay, etc.).

[0118] server

[0119] Step 2:

[0120] The server references the database based on the specified conditions and automatically generates optimal test questions from past data, using an AI model to extract questions of appropriate difficulty and content.

[0121] Step 3:

[0122] The server sends the automatically generated test to the teacher's device in PDF or other format.

[0123] Terminal

[0124] Step 4:

[0125] Teachers receive the tests generated on their devices, print them out as needed, and distribute them to students.

[0126] User

[0127] Step 5:

[0128] After students take the test, teachers enter the test results into a terminal.

[0129] server

[0130] Step 6:

[0131] The device sends the entered test results to a server, which then automatically scores them. Multiple-choice questions are instantly graded by AI, while essay questions are evaluated using NLP technology.

[0132] Step 7:

[0133] The server provides the grading results in the form of a report to the teacher, who can then check the results and make corrections as necessary.

[0134] Learning support for children who are not attending school

[0135] User

[0136] Step 1:

[0137] Children who are not attending school and their parents can send requests for learning support via their devices, specifically by entering information such as the subject, scope, and grade level they wish to study.

[0138] Terminal

[0139] Step 2:

[0140] The terminal transmits the request content to the server.

[0141] server

[0142] Step 3:

[0143] Based on the request, the server searches the database for relevant lesson data and automatically generates appropriate learning materials, such as review videos, slides, and practice questions.

[0144] Step 4:

[0145] The server provides the generated learning materials to the user, who then sends them to the user's device and makes them available for download.

[0146] Terminal

[0147] Step 5:

[0148] Users (children not attending school) download the provided teaching materials and use them for studying at home.

[0149] server

[0150] Step 6:

[0151] The server periodically monitors the user's learning progress, assesses their level of understanding, and provides additional learning materials and advice as needed.

[0152] The above are the specific processing steps of the system of the present invention. This system reduces the workload of teachers and effectively provides learning support to children who do not attend school.

[0153] Example 1

[0154] 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."

[0155] In today's educational environment, teachers' workloads are increasing, with much of their time being spent on lesson preparation and test creation / grading. This has resulted in situations where teachers are unable to concentrate on their primary educational activities. Supporting the learning of children who are not attending school is also an issue, with insufficient provision of appropriate teaching materials and monitoring of their learning progress. Effective methods are needed to resolve these issues and improve the quality of education.

[0156] 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.

[0157] In this invention, the server includes means for collecting lesson data, means for anonymizing and generalizing the collected lesson data, means for performing machine learning using the anonymized and generalized data, means for automatically generating lesson materials based on lesson information, means for providing the generated lesson materials, and means for monitoring learning progress and providing additional teaching materials and advice. This makes it possible to automate lesson preparation and test creation / grading, reducing the workload of teachers and providing effective learning support for children who are not attending school.

[0158] "Class data" refers to information including lesson content, teaching materials, test questions, feedback, etc. in educational settings.

[0159] "Anonymization" is the process of removing or obscuring personally identifiable information from collected data.

[0160] "Generalization" refers to standardizing data in a way that is independent of specific conditions or situations.

[0161] "Machine learning" is a technology that uses massive amounts of data to enable computers to automatically learn patterns and perform specific tasks.

[0162] "Classroom materials" are content such as documents, slides, audio files, and videos created to support educational activities.

[0163] "Auto-generation" is the process of using artificial intelligence or algorithms to generate content or data with minimal human intervention.

[0164] "Study progress" is an indicator of how far a student has progressed in their studies.

[0165] "Advice" is advice or guidance provided to students and teachers based on their learning progress.

[0166] "Tests" refer to question sets and exams used to assess students' understanding and learning status.

[0167] "Scoring" is the process of evaluating test responses and assigning a score or grade.

[0168] A "request" is an act by a user requesting a particular service or information.

[0169] MODE FOR CARRYING OUT THE INVENTION

[0170] This invention is an AI teacher system that aims to reduce the workload of teachers and support the learning of students who are not attending school. This system utilizes lesson data from educational institutions across the country, and AI automates lesson preparation, test creation and grading, as well as monitoring and supporting learning progress.

[0171] Basic system configuration

[0172] The system consists of the following main components:

[0173] 1. Server

[0174] Collect, anonymize, and generalize class data.

[0175] Train and retrain AI models using machine learning.

[0176] Class materials and tests are automatically generated and provided based on class information.

[0177] Receives test result input and performs automatic scoring.

[0178] Monitor your progress and provide additional learning materials and advice.

[0179] 2. Terminal

[0180] Provides an interface for teachers to enter lesson information.

[0181] Receive and download generated course materials and tests.

[0182] The teacher enters the test results and sends them to the server.

[0183] Provide an interface for accepting requests for learning support.

[0184] 3. Users

[0185] The main users are expected to be teachers and school-refusing children (or their parents).

[0186] Teachers enter lesson information, download generated teaching materials and tests, and enter test results.

[0187] Children who are not attending school and their parents can submit requests for learning support and study using the provided teaching materials.

[0188] Collection of lesson data

[0189] The server receives lesson data directly from educational institutions across the country, automatically retrieving it from school databases and online platforms, and can use existing APIs and data feeds.

[0190] Data anonymization and generalization

[0191] The server detects any personally identifiable information from the received lesson data and performs anonymization processing, which removes or conceals personal information such as student and teacher names. The anonymized data is generalized and not dependent on a specific region or school, and is standardized to be compatible with other datasets.

[0192] Training an AI model

[0193] The server trains the AI ​​model using anonymized and generalized data. Specifically, it builds a neural network and optimizes the model based on the collected data. This is done using specialized hardware such as high-performance GPUs and TPUs. The AI ​​model is periodically retrained using the latest data stored in the database to improve its accuracy.

[0194] Automatic generation of teaching materials

[0195] The user (teacher) inputs information such as the lesson topic, target grade, and lesson time via the device. This information is sent from the device to the server, and the AI ​​model uses generative AI technology to automatically generate lesson materials. For example, slides, audio files, and video materials for a history lesson can be generated. The generated materials are sent from the server to the device, where the teacher can download them and use them in class.

[0196] Automatic test generation and scoring

[0197] The user (teacher) issues instructions for creating a test via their device. The scope, difficulty level, and question format (multiple choice, essay, etc.) are entered and sent from the device to the server. The server automatically generates the test based on the entered information. The generated test is provided in PDF format or similar, which the teacher downloads and distributes to students. When students enter their test results, the device sends this data to the server, which then automatically grades them. The graded results are provided to the teacher, along with feedback.

[0198] Request learning support

[0199] The user (a child not attending school or a parent) sends a request for learning support via their device. For example, they may request a "math review." The request is sent from the device to the server, which automatically generates learning materials based on the request. For example, it may generate videos or slides for reviewing math. The generated learning materials are provided to the user, who studies them at home. The server monitors the learning progress and provides additional learning materials or advice as needed.

[0200] Examples of prompt statements

[0201] The following prompt sentences are used:

[0202] "I want to generate teaching materials for history classes."

[0203] "I want you to create a math test."

[0204] "Please provide math review materials."

[0205] The AI ​​teacher system of the present invention can significantly reduce the workload of teachers and provide effective learning support for children who are not attending school.

[0206] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0207] Program processing steps

[0208] Step 1: Collect lesson data

[0209] Specific behavior:

[0210] The server receives lesson data directly from educational institutions across the country, automatically retrieving data from school databases and online platforms via APIs.

[0211] Input: Data sent by educational institutions, such as course content, study materials, exam questions, and feedback.

[0212] Output: Raw lesson data collected.

[0213] Step 2: Anonymize and generalize data

[0214] Specific behavior:

[0215] The server then detects any personally identifiable information from the collected data and performs anonymization processing, for example, by using a filtering algorithm to remove personal information such as names and addresses.

[0216] The server generalizes the anonymized data, standardizing it to convert it into a format that is independent of specific schools or regions.

[0217] Input: Raw lesson data collected.

[0218] Output: Anonymized and generalized lesson data.

[0219] Step 3: Training the AI ​​model

[0220] Specific behavior:

[0221] The server uses anonymized and generalized data to train AI models, using neural networks and other machine learning techniques to train algorithms that generate educational content.

[0222] The server periodically retrains the AI ​​model with the latest data, thereby maintaining the model's accuracy and effectiveness.

[0223] Input: Anonymized and generalized lesson data.

[0224] Output: A trained AI model.

[0225] Step 4: Automatic generation of lesson materials

[0226] Specific behavior:

[0227] The user (teacher) uses a terminal to input information such as the lesson theme, target grade, lesson time, etc. This transmits specific lesson needs to the server.

[0228] The terminal transmits the input information to the server.

[0229] The server uses a generative AI model to automatically generate lesson materials based on the input lesson information, such as history lesson slides, audio files, and videos.

[0230] The server transmits the generated lesson materials to the teacher's terminal.

[0231] The user (teacher) checks the generated lesson materials and makes fine adjustments as necessary.

[0232] Input: Information such as lesson topic, target grade, lesson time, etc.

[0233] Output: Generated lesson materials (slides, audio files, videos, etc.).

[0234] Step 5: Auto-generate and grade tests

[0235] Specific behavior:

[0236] The user (teacher) uses a terminal to instruct the creation of a test, inputting the scope, difficulty level, and question format (multiple choice, essay, etc.).

[0237] The terminal transmits the input information to the server.

[0238] The server uses a generative AI model to automatically generate tests based on the input information, such as multiple-choice and essay questions for science, in PDF format.

[0239] The server provides the generated test to the teacher's terminal, who then distributes it to the students.

[0240] After the students take the test, the user (teacher) enters the results into the terminal.

[0241] The device sends the entered test results to the server, which then automatically scores them using a scoring algorithm for fast and accurate evaluation.

[0242] The server provides the grading results to the instructor and generates feedback.

[0243] Input: Test scope, difficulty level, question format. Test results.

[0244] Output: Auto-generated test, marking results and feedback.

[0245] Step 6: Request and provide learning support

[0246] Specific behavior:

[0247] A user (a child not attending school or a parent) uses a device to send a request for learning support, for example, a request for "math review."

[0248] The terminal sends a request to the server.

[0249] The server automatically generates learning materials based on the request using a generative AI model, such as review videos and slides.

[0250] The server provides the generated learning materials to the user.

[0251] Users (children not attending school) study at home using the provided teaching materials.

[0252] The server monitors learning progress and provides additional learning materials and advice as needed.

[0253] Enter: Request for study support.

[0254] Output: Generated learning materials, learning progress data, and further learning materials and advice.

[0255] This specific process will reduce the workload of teachers and provide effective learning support to children who are not attending school.

[0256] (Application example 1)

[0257] 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."

[0258] In the traditional education system, teachers are overwhelmed with the tasks of preparing lessons and creating and grading tests, making it difficult to provide adequate learning support for students who are not attending school. Furthermore, educational methods using virtual environments are limited, making it difficult to provide customized learning materials tailored to individual students' learning progress. This situation hinders the efficient use of educational resources and the improvement of educational quality.

[0259] 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.

[0260] In this invention, the server includes means for collecting lesson data, means for anonymizing and generalizing the collected lesson data, means for performing machine learning using the anonymized and generalized data, means for automatically generating lesson materials based on lesson information, means for providing the generated lesson materials, means for delivering lessons in a virtual environment, and means for tracking the learning progress of children who are not attending school and providing customized teaching materials. This reduces the workload of teachers and makes it possible to provide effective individual learning support to children who are not attending school.

[0261] "Class data" refers to information such as class content, teaching materials, test questions, and feedback collected from educational institutions.

[0262] "Anonymization" means processing data so that personal information is removed and individuals cannot be identified.

[0263] "Generalization" means transforming data so that it can be used widely and is not tied to a specific situation.

[0264] "Machine learning" is a technology that allows computers to automatically learn from data and make future predictions and classifications.

[0265] "Class information" refers to specific information about the implementation of a class, such as the class theme, target grade, and class time.

[0266] "Classroom teaching materials" are materials such as slides, audio files, and videos used in conducting lessons.

[0267] A "virtual environment" is a virtual space provided via the Internet, rather than a physical classroom or location.

[0268] A "school refusal student" is a student who is unable to attend school for an extended period of time for some reason.

[0269] "Learning progress" refers to a student's progress and achievement in learning.

[0270] "Customized learning materials" refer to learning resources that are optimized according to the learning situation and needs of each individual student.

[0271] This invention is an AI teaching system that aims to reduce the workload of teachers and support the learning of children who are not attending school. This system collects lesson data and uses AI to automate and support lesson preparation, test creation, and grading. It also delivers lessons through a virtual environment and provides customized teaching materials to children who are not attending school, thereby supporting their learning.

[0272] Basic system configuration

[0273] The system consists of the following main components:

[0274] 1. Server

[0275] Classroom data is collected, anonymized, and generalized.

[0276] Train and retrain AI models using machine learning.

[0277] Class materials are automatically generated and provided based on class information.

[0278] Deliver lessons in a virtual environment.

[0279] Automatically generate and grade tests.

[0280] Track the learning progress of children who are not attending school and generate and provide customized learning materials.

[0281] 2. Terminal

[0282] Provides an interface for teachers to enter lesson information.

[0283] Receive and download created course materials and tests.

[0284] Classes are delivered virtually and can be viewed live or recorded.

[0285] The teacher enters the test results and sends them to the server.

[0286] Provides an interface for school-refusing children and their parents to request learning support.

[0287] 3. Users

[0288] The main users are teachers, children who are not attending school, and their parents.

[0289] Teachers enter lesson information, download generated teaching materials and tests, and enter test results.

[0290] Children who are not attending school and their parents can enter requests for learning support and study using the provided teaching materials.

[0291] Program processing flow

[0292] Hardware and Software Use

[0293] Hardware: Smartphones and head-mounted displays are used, which allow for real-time streaming of classes and recordings for viewing.

[0294] Software used: Python's aiohttp library is used to perform asynchronous HTTP requests and communicate with the server.

[0295] Data processing and calculation

[0296] 1. Collection and anonymization of lesson data

[0297] The server collects lesson data from multiple educational institutions, anonymizes and generalizes the data, removing personal information and converting it into a common format.

[0298] 2. Training the AI ​​model

[0299] The server uses anonymized and generalized data to train AI models, which will automate the generation of future course materials and test creation.

[0300] 3. Automatic generation of teaching materials and tests

[0301] When teachers input information about the lesson theme and target grade through the terminal, the server automatically generates lesson materials based on that information and sends them back to the terminal.

[0302] Similarly, you can enter the scope and format of the test and the server will automatically generate the test and provide it in PDF format.

[0303] 4. Learning support

[0304] When a child who is not attending school or their parents request learning support, the server generates and provides customized teaching materials and tests based on the child's individual learning progress.

[0305] Examples and prompts

[0306] Examples:

[0307] Lesson theme: "Japanese History"

[0308] Target grade: 2nd year junior high school students

[0309] Lesson duration: 45 minutes

[0310] Example prompt sentence:

[0311] Lesson theme: Japanese history

[0312] Target grade: 2nd year junior high school students

[0313] Lesson duration: 45 minutes

[0314] Generated teaching materials: PowerPoint slides, lesson videos

[0315] The AI ​​teaching system of this invention reduces the workload of teachers and effectively supports the learning of students who are not attending school. In addition, by utilizing a virtual environment, it is possible to deliver real-time and recorded lessons, thereby improving the quality of education.

[0316] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0317] Step 1:

[0318] The server collects lesson data from educational institutions. The collected lesson data includes lesson content, teaching materials, test questions, and feedback. It receives the lesson data as input, anonymizes and generalizes it to remove specific personal information and convert it into a common format. It generates the anonymized and generalized data as output.

[0319] Step 2:

[0320] The server performs machine learning using the anonymized and generalized data from step 1. Specifically, it trains a generative AI model based on the data and updates the AI ​​model. It receives the anonymized data as input and outputs the trained AI model.

[0321] Step 3:

[0322] The user (teacher) inputs lesson information such as lesson theme, target grade, lesson time, etc. through the terminal. The input lesson information is sent from the terminal to the server.

[0323] Step 4:

[0324] The server automatically generates lesson materials based on the lesson information entered in step 3. It uses a generative AI model to generate appropriate lesson slides and audio files. It receives lesson information as input and outputs lesson materials.

[0325] Step 5:

[0326] The server sends the generated lesson materials to the terminal. The terminal provides the lesson materials received from the server to the teacher. The teacher downloads the materials for use in the class. The terminal receives the lesson materials as input and outputs the materials sent to the terminal.

[0327] Step 6:

[0328] The user (teacher) uses a terminal to give instructions for creating a test. Specifically, they input the scope, difficulty, and format of the test. The input test information is sent from the terminal to the server.

[0329] Step 7:

[0330] The server automatically generates a test based on the test information entered in step 6. It uses a generative AI model to create a test including multiple choice and essay questions and generates it in PDF format. It receives test information as input and outputs the generated test.

[0331] Step 8:

[0332] The user (teacher) distributes the generated test to students and inputs the test results into the terminal, which then sends the input results to the server.

[0333] Step 9:

[0334] The server receives the test results entered in step 8 and automatically grades them using AI. It receives the test results as input and outputs the graded results. The graded results are sent to the terminal and provided to the teacher.

[0335] Step 10:

[0336] The user (a child not attending school or a parent) sends a request for learning support via the device. Specifically, they request the target subjects and learning content. The input request information is sent from the device to the server.

[0337] Step 11:

[0338] The server generates customized learning materials based on the request information entered in step 10. It uses a generative AI model to create review videos and slides. It receives the request information as input and outputs the generated learning materials. The output learning materials are sent to the terminal and provided to the child who is not attending school and their parents.

[0339] Step 12:

[0340] The server monitors the learning progress of students who are not attending school and provides additional learning materials or advice as needed. It is capable of receiving learning progress data as input and outputting appropriate learning support.

[0341] 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.

[0342] This invention is an AI teaching system designed to reduce the workload of teachers and support the learning of children who are not attending school. This system utilizes lesson data from across the country, and AI automates and supports lesson preparation, test creation, and grading. In addition, by combining it with an emotion engine that recognizes the user's emotions, it provides educational support that is more tailored to each individual user.

[0343] Basic system configuration

[0344] The system consists of the following main components:

[0345] 1. Server

[0346] Collect, anonymize, and generalize class data.

[0347] Train and retrain AI models using machine learning.

[0348] Class materials and tests are automatically generated and provided based on class information.

[0349] Receives test result input and performs automatic scoring.

[0350] Generate learning materials based on requests for learning support for children who are not attending school and monitor their progress.

[0351] It integrates an emotion engine that recognizes the user's emotions and adjusts lesson content and tests.

[0352] 2. Terminal

[0353] Provides an interface for teachers to enter lesson information.

[0354] Receive and download generated course materials and tests.

[0355] The teacher enters the test results and sends them to the server.

[0356] Children who are not attending school and their parents can submit requests for learning support.

[0357] It receives feedback from the emotion engine and transmits the user's state to the server through the emotion recognition function.

[0358] 3. Users

[0359] The main users are expected to be teachers and school-refusing children (or their parents).

[0360] Teachers enter lesson information, download generated teaching materials and tests, and enter test results.

[0361] Children who are not attending school and their parents can submit requests for learning support and study using the provided teaching materials.

[0362] While receiving feedback from the emotion engine, the user receives appropriate learning support based on their own emotional state.

[0363] Program processing flow

[0364] Collection of lesson data

[0365] The server collects lesson data from each educational institution, including, for example, mathematics lesson content, teaching materials, test questions, and feedback.

[0366] The server anonymizes the collected data and removes any personal information.

[0367] The server generalizes the anonymized data, converting it so that it is not tied to any particular situation.

[0368] Training an AI model

[0369] The server trains the AI ​​model using anonymized and generalized data.

[0370] The server periodically uses the latest lesson data stored in the database to retrain the AI ​​model, ensuring it reflects the latest educational trends and data.

[0371] Automatic generation of teaching materials

[0372] The user (teacher) inputs information such as the lesson theme, target grade, and lesson time via the terminal.

[0373] The terminal transmits the input information to the server.

[0374] Based on the lesson information entered, the server searches the database for relevant lesson data and uses an AI model to automatically generate optimal lesson materials, such as history lesson slides and audio files.

[0375] The server transmits the generated lesson materials to the teacher's terminal.

[0376] The user (teacher) receives the generated teaching materials and uses them in class.

[0377] Automatic test generation and scoring

[0378] The user (teacher) sends a test creation request to the server from their device, inputting the scope, difficulty level, and format.

[0379] The terminal transmits the input information to the server.

[0380] The server automatically generates tests and provides them to teachers' terminals in PDF format, for example, generating multiple-choice and essay questions for a science test.

[0381] The user (teacher) distributes the generated test to students.

[0382] The user (teacher) enters the test results into the terminal.

[0383] The device sends the test results to a server, which then automatically scores them. Multiple-choice questions are instantly scored by AI, while essay questions are evaluated using NLP technology.

[0384] The server provides the grading results to the teacher.

[0385] Learning support for children who are not attending school

[0386] Users (children not attending school or their parents) send a request for learning support via their device. Specifically, they input information such as the subject, scope, and grade they wish to study.

[0387] The terminal transmits the request content to the server.

[0388] The server searches for relevant lesson data based on the request and automatically generates appropriate learning materials, such as review videos, slides, and practice questions.

[0389] The server provides the generated learning materials to the user, who then sends them to the user's device and makes them available for download.

[0390] Users (children not attending school) download the provided teaching materials and use them for studying at home.

[0391] The server periodically monitors the user's learning progress, assesses their level of understanding, and provides additional learning materials and advice as needed.

[0392] Emotion engine integration

[0393] The terminal and server recognize the user's emotions through an emotion engine, for example, by using facial recognition technology and voice analysis to analyze the user's emotional state.

[0394] The server then adjusts course materials and test content based on feedback from the emotion engine, for example by providing less difficult questions to students who are nervous.

[0395] Similarly, learning support for users (children not attending school) is adjusted based on data from the emotion engine. If learning progress is falling behind or motivation is declining, additional support is provided.

[0396] As a concrete example, when a teacher is teaching a history class, the emotion engine first recognizes the emotional state of the teacher and students.The server then generates the optimal slides or audio materials for the lesson based on the recognized data and provides them to the devices.In addition, when conducting a test, measures are taken to increase motivation by providing relatively easy questions to students who are feeling nervous or stressed.

[0397] The collaboration between the AI ​​teacher system and emotion engine of this invention reduces the workload of teachers and enables more personalized educational support for each student. Learning support for children who are not attending school can also be provided optimally by understanding changes in their emotions, which will greatly contribute to reducing educational disparities.

[0398] The above is a specific embodiment for carrying out the present invention.

[0399] The processing flow will be explained below.

[0400] Program processing steps

[0401] Collection and learning of lesson data

[0402] server

[0403] Step 1:

[0404] The server collects lesson data from each educational institution, including the teaching materials used by teachers, lesson content, test questions, and student feedback.

[0405] Step 2:

[0406] The server anonymizes the collected lesson data and removes personal information, automatically detecting and deleting personally identifiable information such as names and student numbers.

[0407] Step 3:

[0408] The server generalizes the anonymized data by standardizing expressions specific to specific teachers and schools and storing them in a generic format.

[0409] Step 4:

[0410] The server trains the AI ​​model using anonymized and generalized data, and uses NLP and machine learning algorithms to extract features of the lesson content and teach the AI ​​model.

[0411] Step 5:

[0412] The server periodically retrains the AI ​​model using the latest lesson data stored in the database, ensuring that the model reflects the latest educational trends and data.

[0413] Automatic generation of teaching materials

[0414] Terminal

[0415] Step 1:

[0416] The user (teacher) inputs information such as the lesson theme, target grade, and lesson time via the terminal.

[0417] Step 2:

[0418] The terminal transmits the input lesson information to the server.

[0419] server

[0420] Step 3:

[0421] The server searches the database for relevant lesson data based on the lesson information received, and uses AI models to generate optimal lesson materials, such as history lesson slides and audio files.

[0422] Step 4:

[0423] The server transmits the generated lesson materials to the teacher's terminal.

[0424] Terminal

[0425] Step 5:

[0426] Teachers receive the teaching materials generated on their devices, download them, edit or revise them as necessary, and use them in class.

[0427] Automated test creation and scoring

[0428] User

[0429] Step 1:

[0430] Teachers can create tests through their terminals, inputting the scope, difficulty level, and format.

[0431] Terminal

[0432] Step 2:

[0433] The terminal transmits the input information to the server.

[0434] server

[0435] Step 3:

[0436] The server generates optimal test questions from a database based on the specified conditions. Multiple choice and essay questions are extracted and generated using an AI model.

[0437] Step 4:

[0438] The server provides automatically generated tests to teachers' terminals in PDF format or other formats.

[0439] Terminal

[0440] Step 5:

[0441] Teachers receive the tests generated on their devices, print them out as needed, and distribute them to students.

[0442] User

[0443] Step 6:

[0444] After students take the test, teachers enter the test results into a terminal.

[0445] server

[0446] Step 7:

[0447] The device sends the entered test results to a server, which then automatically scores them. Multiple-choice questions are instantly graded by AI, while essay questions are evaluated using NLP technology.

[0448] Step 8:

[0449] The server provides the grading results to the teacher, who can review the results and make corrections if necessary.

[0450] Learning support for children who are not attending school

[0451] User

[0452] Step 1:

[0453] Children who are not attending school and their parents can send requests for learning support via their devices, inputting the subject, scope, grade level, etc. they wish to study.

[0454] Terminal

[0455] Step 2:

[0456] The terminal sends the request contents to the server.

[0457] server

[0458] Step 3:

[0459] Based on the request, the server searches the database for relevant lesson data and generates appropriate learning materials, such as review videos and slides.

[0460] Step 4:

[0461] The server transmits the generated learning materials to the user's terminal.

[0462] Terminal

[0463] Step 5:

[0464] Users (children not attending school) receive the provided teaching materials on their devices, download them, and study at home.

[0465] server

[0466] Step 6:

[0467] The server periodically monitors the user's learning progress, assesses their level of understanding, and provides additional learning materials and advice as needed.

[0468] Emotion engine integration

[0469] Terminal

[0470] Step 1:

[0471] Users (teachers and students) activate the emotion engine through their devices during lessons and tests.

[0472] Step 2:

[0473] The emotion engine uses facial recognition technology and voice analysis to recognize the user's emotions, for example, whether the user is nervous or stressed.

[0474] server

[0475] Step 3:

[0476] The server receives feedback from the emotion engine and adjusts the difficulty of lessons and tests based on that data, for example providing students who are nervous with easier questions.

[0477] Step 4:

[0478] The server also adjusts learning support for students who are not attending school based on the emotion recognition results, for example, by providing additional support if their learning progress is falling behind or their motivation is declining.

[0479] Terminal

[0480] Step 5:

[0481] The user (a child not attending school) receives feedback from the emotion engine and receives appropriate learning support based on their own emotional state.

[0482] Through these processing steps, the system of the present invention significantly reduces the workload of teachers and provides optimal educational support tailored to the emotional state of each individual user. Specific operational examples include providing tests with adjusted difficulty to nervous students and providing additional support to motivate students who are not attending school. In this way, the quality and effectiveness of education can be improved.

[0483] Example 2

[0484] 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."

[0485] In today's educational environment, the increasing workload of teachers is a major problem. Repetitive tasks such as lesson preparation and test creation / grading place a particularly heavy burden on teachers. Furthermore, teachers' workloads are reaching their limits as they are required to provide learning support for children who are not attending school and individual attention to each student. A system is needed to effectively resolve these issues and improve the quality of education.

[0486] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting lesson data, means for anonymizing and generalizing the collected lesson data, means for performing machine learning using the anonymized and generalized data, means for automatically generating lesson materials based on lesson information, means for providing the generated lesson materials, means for analyzing user emotions using an emotion recognition function, and means for adjusting the content of lesson materials and tests based on the analyzed emotion data. This reduces the workload of teachers and enables personalized educational support for each student, including children who are not attending school.

[0487] "Class Data" is a general term for information including the content of classes conducted at educational institutions, teaching materials, test questions, and feedback.

[0488] "Anonymization" refers to the process of removing personal information from data and converting it into a state in which specific individuals cannot be identified.

[0489] "Generalization" refers to the process of converting data into a general-purpose format so that it is not tied to a specific situation.

[0490] "Machine learning" refers to a technology in which a computer automatically learns patterns using data and makes predictions and classifications.

[0491] "Class information" is a general term for information such as themes, target grades, and class times that teachers use to plan their lessons.

[0492] "Automatic generation" refers to the process of automatically creating data and information using AI or algorithms.

[0493] "Providing" refers to making the generated data or information available to users.

[0494] "Emotion recognition function" refers to technology that analyzes a user's emotional state using facial recognition technology and voice analysis.

[0495] "Analyzing" refers to the process of analyzing collected data and extracting meaningful information.

[0496] "Adjust" refers to the process of changing the content of data or information in response to specific conditions or circumstances.

[0497] This invention is an AI teaching system designed to reduce the workload of teachers and support the learning of children who are not attending school. This system utilizes lesson data from across the country, and AI automates and supports lesson preparation, test creation, and grading. In addition, by combining it with an emotion engine that recognizes the user's emotions, it provides educational support that is more tailored to each individual user.

[0498] Basic system configuration

[0499] The system consists of the following main components:

[0500] 1. Server

[0501] The server collects lesson data from educational institutions, anonymizes and generalizes the collected data, for example, using Python data processing libraries (Pandas, NumPy).

[0502] The server uses anonymized and generalized data to train AI models using machine learning libraries (TensorFlow, PyTorch).

[0503] Automatically generate teaching materials based on lesson information and provide them to students. For example, AI can be used to generate lesson slides (PDF format) and audio materials.

[0504] The emotion recognition function analyzes the user's emotions and adjusts the content of class materials and tests based on the analysis. This process uses facial recognition technology (OpenCV) and voice analysis (Librosa).

[0505] 2. Terminal

[0506] The terminal provides an interface for teachers to input lesson information, for example, using a web application (HTML, CSS, JavaScript).

[0507] Receive and download generated course materials and tests.

[0508] The teacher enters the test results and sends them to the server.

[0509] Children who are not attending school and their parents can submit requests for learning support.

[0510] It receives feedback from the emotion engine and sends the user's state to the server.

[0511] 3. Users

[0512] The main users are teachers and children who are not attending school (or their parents).

[0513] Teachers enter lesson information, download generated teaching materials and tests, and enter test results.

[0514] Children who are not attending school and their parents can submit requests for learning support and study using the provided teaching materials.

[0515] While receiving feedback from the emotion engine, appropriate learning support is provided based on the emotional state.

[0516] Specific examples

[0517] 1. Automatic generation of teaching materials

[0518] The user (teacher) enters lesson information for "World War II, 3rd year junior high school students, 50 minutes" through the terminal.

[0519] This information is sent to a server, which uses an AI model to search for relevant lesson data and automatically generate optimal lesson materials, including lesson slides (PDF format) and audio materials.

[0520] Teachers can download the generated teaching materials and use them in their classes.

[0521] 2. Automatic test generation and scoring

[0522] A teacher uses a terminal to send a request to create a test with the following scope: "World War II," difficulty level: "medium," and format: "multiple choice and essay questions."

[0523] This information is sent to a server, which uses an AI model to automatically generate tests.

[0524] The teacher distributes the generated test to the students and later enters the test results into the terminal.

[0525] The device sends the test results to a server, which then automatically grades them. Multiple-choice questions are graded immediately, while essay questions are graded using NLP technology.

[0526] The results of the grading will be provided to the instructor.

[0527] 3. Learning support for children who are not attending school

[0528] A child who is not attending school or their parents can send a request via their device for "Subject: Mathematics, Range: Quadratic Equations, Grade: 2nd Year of Junior High School."

[0529] The server searches for relevant lesson data based on the request and uses an AI model to automatically generate learning materials, including practice questions and instructional videos.

[0530] Parents can download the generated teaching materials, and their children who are not attending school can study at home.

[0531] The server monitors learning progress, assesses comprehension, and provides additional learning materials and study advice.

[0532] 4. Emotion engine integration

[0533] When a teacher is teaching a history lesson, the device uses the device's camera and microphone to recognize the emotional state of the teacher and students.

[0534] The recognized emotion data is sent to the server, which then generates optimal teaching materials for the lesson and provides them to the terminal.

[0535] When administering tests, adjustments are made based on emotional data as appropriate, such as providing less difficult questions to students who are feeling nervous or stressed.

[0536] As described above, the AI ​​teaching system of the present invention reduces the workload of teachers and provides individualized educational support to each student.

[0537] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0538] Step 1:

[0539] The server collects lesson data from each educational institution. Specifically, the server accesses the educational institution's database via API to obtain data such as lesson content, teaching materials, test questions, and feedback. The input data is educational data in JSON format, and the output data is raw data that is temporarily stored in the server's internal storage.

[0540] Step 2:

[0541] The server anonymizes the collected lesson data. Specifically, it uses Python data processing libraries (Pandas, NumPy) to run algorithms to remove personal information. For example, it converts student names into meaningless codes and replaces IDs with random numbers. The input data is raw lesson data, and the output data is anonymized lesson data.

[0542] Step 3:

[0543] The server generalizes the anonymized data. The server processes the data so that it is not dependent on a specific educational institution or situation. For example, it converts specific school names or place names into general expressions to increase the versatility of the data. The input data is anonymized lesson data, and the output data is generalized lesson data.

[0544] Step 4:

[0545] The server trains an AI model using the generalized data. Specifically, the server uses machine learning libraries (TensorFlow, PyTorch) to generate an AI model based on the dataset. For example, it uses math or history lesson data to train a model that predicts appropriate lesson content. The input data is the generalized lesson data, and the output data is the trained AI model.

[0546] Step 5:

[0547] The user (teacher) inputs lesson information into the terminal. For example, the teacher inputs lesson topic, target grade, lesson time, etc. into a web form. The input data is the lesson information, and the output data is a lesson request sent to the server.

[0548] Step 6:

[0549] The terminal sends the input lesson information to the server. The server receives information such as lesson theme, target grade, and lesson time, and searches the database for relevant lesson data based on that information. The input data is the lesson request, and the output data is the relevant lesson data.

[0550] Step 7:

[0551] The server uses an AI model to automatically generate optimal lesson materials based on the input lesson information. For example, it generates appropriate slides and audio materials based on the history lesson theme. The input data is the relevant lesson data, and the output data is the generated lesson materials (PDF slides and audio files).

[0552] Step 8:

[0553] The server sends the generated lesson materials to the teacher's device. For example, the generated slide data is provided to the teacher via email or cloud storage. The input data is the generated lesson materials, and the output data is the lesson materials in a format that the teacher can access.

[0554] Step 9:

[0555] The teacher sends a test creation request on the terminal. The teacher inputs the scope, difficulty level, and format. For example, the teacher might input "Scope: World War II, Difficulty: Medium, Format: Multiple choice and essay questions." The input data is the test creation request, and the output data is the request information sent to the server.

[0556] Step 10:

[0557] The device sends the input request information to the server. Based on the received information, the server uses an AI model to automatically generate a test. For example, it generates a PDF-format test that includes multiple-choice and essay questions. The input data is the request information, and the output data is the generated test.

[0558] Step 11:

[0559] The server provides the generated test to the teacher's terminal, where the input data is the generated test and the output data is the test materials sent to the teacher's terminal.

[0560] Step 12:

[0561] A teacher inputs test results into a terminal. For example, a teacher grades students' tests and enters the results into a web form. The input data is the student's test results, and the output data is the test result data sent to the server.

[0562] Step 13:

[0563] The device sends the entered test results to a server, which instantly scores multiple-choice questions using an AI model and automatically evaluates essay questions using NLP technology. The input data is the test result data, and the output data is the scoring results.

[0564] Step 14:

[0565] The server provides the grading results to the teacher. For example, it sends a report including the grading results to the teacher by email. The input data is the grading results, and the output data is a grading report to the teacher.

[0566] Step 15:

[0567] A child who is not attending school and their guardian input a request for learning support into the terminal. For example, they input "Subject: Mathematics, Area: Quadratic Equations, Grade: 8th Grade." The input data is the learning support request, and the output data is the request information sent to the server.

[0568] Step 16:

[0569] The device sends the request information to the server, which then generates appropriate learning materials based on the received information, such as a practice problem set or an instructional video. The input data is the request information, and the output data is the generated learning materials.

[0570] Step 17:

[0571] The server provides the generated learning materials to the user. For example, the server provides the generated learning materials to the user in a downloadable format via cloud storage. The input data is the generated learning materials, and the output data is the learning materials in a format that the user can access.

[0572] Step 18:

[0573] The user (a child not attending school) downloads the provided learning materials and studies at home. The input data are the learning materials, and the output data are the user's learning progress.

[0574] Step 19:

[0575] The server periodically monitors the user's learning progress, evaluates their level of understanding, and provides additional learning materials and advice as needed. The input data is learning progress data, and the output data is the level of understanding assessment and additional learning materials.

[0576] Step 20:

[0577] The device and server recognize the user's emotions through an emotion engine. For example, they use the device's camera and microphone to perform facial recognition and voice analysis. The input data is the user's facial expression and voice data, and the output data is the analysis result of the user's emotional state.

[0578] Step 21:

[0579] The server adjusts the content of the lesson materials and tests based on the feedback from the emotion engine, for example, providing easier questions to a nervous user. The input data is the analysis result of the emotional state, and the output data is the adjusted lesson materials and tests.

[0580] (Application example 2)

[0581] 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."

[0582] Conventional customer service in brick-and-mortar stores has had difficulty adequately responding to individual customer emotional states and preferences. Furthermore, privacy protection issues existed when collecting and utilizing customer information. The present invention aims to solve these issues, provide more attentive service to customers, and protect their privacy.

[0583] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0584] In this invention, the server includes means for collecting lesson data, means for anonymizing and generalizing the collected lesson data, means for performing machine learning using the anonymized and generalized data, means for automatically generating lesson materials based on lesson information, means for providing the generated lesson materials, means for collecting customer information, means for anonymizing and generalizing the collected customer information, means for recognizing customer emotions, and means for adjusting response content based on the customer's emotional state. This enables individual responses according to the customer's emotional state and makes it possible to provide high-quality services while protecting the customer's privacy.

[0585] - "Class Data" refers to information related to education, such as class content, teaching materials, and test questions, collected from educational institutions.

[0586] "Classroom teaching materials" are educational materials such as slides, audio files, and handouts used in classes.

[0587] "Anonymization" is the process of removing personal information from collected data so that it is no longer possible to identify a specific individual.

[0588] "Generalization" is the process of transforming anonymized data so that it is not tied to a specific situation.

[0589] "Machine learning" is a technique that uses data to train AI models to perform specific tasks automatically.

[0590] "Customer information" refers to data related to customers at a store, such as customer identification information, purchase history, and behavioral data.

[0591] "Emotion recognition" is a technology that uses facial recognition technology and voice analysis to determine a customer's emotional state from their facial expressions and voice.

[0592] "Adjusting the response" means changing the content and delivery method of the service based on the customer's emotional state.

[0593] "Privacy protection" means properly managing customers' personal information so that it is not made known to third parties.

[0594] An "AI model" is a collection of algorithms that learn from data and perform tasks automatically.

[0595] This invention relates to an AI customer support system for brick-and-mortar stores. This system uses AI to adjust service content based on the individual emotional state of each customer, providing detailed and personalized service to customers.

[0596] Basic system configuration

[0597] The system consists of the following main components:

[0598] 1. Server

[0599] Collect, anonymize, and generalize class data and customer information.

[0600] Train and retrain AI models using machine learning.

[0601] Teaching materials and service content are automatically generated and provided based on lesson information and customer information.

[0602] Integrates an emotion engine that recognizes customer emotions and adjusts service content.

[0603] 2. Terminal

[0604] Provides an interface for store staff to input service information.

[0605] Receive and download generated educational materials and service content.

[0606] The staff enters the service information and sends it to the server.

[0607] Physical store customers improve their service experience through emotion engine feedback.

[0608] 3. Users

[0609] Store staff and customers are expected to be the main users.

[0610] Store staff input service information, download the generated educational materials and service content, and input service information again.

[0611] Customers in physical stores will receive a personalized service experience provided by the system.

[0612] Program processing flow

[0613] Collection of lesson data and customer information

[0614] The server collects learning data and customer information from each brick-and-mortar store, including customer purchase history, behavioral data, and emotional states captured through an emotion engine. This information is acquired using cameras (built into smartphones, smart glasses, and head-mounted displays (HMDs)). The collected data is anonymized and generalized to protect privacy.

[0615] Training an AI model

[0616] The server trains AI models using anonymized and generalized data. Supported software includes machine learning libraries (e.g., scikit-learn, TensorFlow, etc.). The AI ​​models are retrained using the latest data to provide always-updated services.

[0617] Automatic generation of service content

[0618] The user (store staff) inputs service information through a terminal. This includes information on the target customer demographic, desired service content, and target products. The terminal sends the input information to the server, which then uses an AI model to generate optimal service content. For example, it generates an introductory video or promotional information for a specific product. The generated service content is then sent to the store staff's terminal.

[0619] Emotion recognition and response adjustment

[0620] The device and server recognize the customer's emotions through an emotion engine. They analyze the customer's emotional state using facial recognition technology and voice analysis (e.g., DeepFace), and the server adjusts the service content based on the feedback from the emotion engine. For example, a customer who is feeling stressed may be recommended relaxation-related products.

[0621] Specific examples

[0622] When a customer picks up a product in a physical store, the smart glasses capture the action and recognize the customer's emotional state through the emotion engine. In this case, for example, if the smart glasses determine that the customer is interested in a product, they can display detailed descriptions and related promotions on the smart glasses.

[0623] Example prompt sentence:

[0624] "Create a voice guide that explains what products you would suggest to customers when they are stressed and why."

[0625] Example output:

[0626] Recommended product: Relaxation aroma oil

[0627] Reason: Relaxation aroma oils are effective in reducing stress and leaving you feeling refreshed.

[0628] Audio guide example:

[0629] "Hello, it looks like you're feeling stressed. This relaxation aroma oil is effective in reducing stress. Please give it a try."

[0630] In this way, by implementing the AI ​​customer support system of the present invention, it becomes possible to provide individual responses according to the customer's emotional state, thereby achieving both improved customer satisfaction and privacy protection.

[0631] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0632] Step 1:

[0633] The server collects lesson data and customer information. Specifically, it captures customer facial expressions and behavioral data using cameras and sensors in the physical store, and collects information such as shopping history and behavioral patterns. The raw data acquired as input is temporarily stored before being processed.

[0634] Step 2:

[0635] The server anonymizes and generalizes the collected data. Specifically, it removes personally identifiable information and standardizes and aggregates the data. This allows the data to be managed in a unified format and prevents the identification of specific individuals. Raw data is used as input, and anonymized and generalized data is output.

[0636] Step 3:

[0637] The server performs machine learning using the anonymized and generalized data. Specifically, it uses machine learning libraries (e.g., scikit-learn, TensorFlow) to train an AI model based on the data and develops an algorithm to predict customer behavior and sentiment. The anonymized data is used as input, and the trained AI model is obtained as output.

[0638] Step 4:

[0639] The server automatically generates service content based on lesson information and customer information. Specifically, it uses a generative AI model to generate prompts that recommend the most suitable products and services to the customer. The AI ​​model and customer information are used as input, and the generated service content (e.g., a list of recommended products and promotion information) is obtained as output.

[0640] Step 5:

[0641] The terminal receives the generated service content and provides it to the staff at the physical store. Specifically, the generated service content is displayed through an application installed on the terminal. The generated service content is used as input and is available to the staff as output.

[0642] Step 6:

[0643] The device and server recognize the customer's emotions through an emotion engine. Specifically, an emotion analysis library (e.g., DeepFace) is used to analyze the customer's facial expressions and voice to identify their emotional state. Real-time video and audio data is used as input, and the customer's emotional state is obtained as output.

[0644] Step 7:

[0645] The server adjusts the service content based on the feedback from the emotion engine. Specifically, it appropriately modifies the generated service content according to the obtained emotional state, providing personalized support. Using the emotional state and service content as input, the adjusted service content is obtained as output.

[0646] Step 8:

[0647] Users (store customers) receive personalized service experiences provided by the system. Specifically, they receive tailored service content through terminals or store staff. It is expected that users will receive tailored service content as input and achieve high levels of satisfaction as output.

[0648] 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.

[0649] 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.

[0650] 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.

[0651] [Second embodiment]

[0652] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0653] 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.

[0654] 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).

[0655] 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.

[0656] 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.

[0657] 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).

[0658] 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.

[0659] 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.

[0660] 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.

[0661] 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.

[0662] 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.

[0663] 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."

[0664] This invention is an AI teaching system that aims to reduce the workload of teachers and support the learning of children who are not attending school. This system utilizes lesson data from all over the country and uses AI to automate and support lesson preparation, test creation, and grading.

[0665] Basic system configuration

[0666] The system consists of the following main components:

[0667] 1. Server

[0668] Collect, anonymize, and generalize class data.

[0669] Train and retrain AI models using machine learning.

[0670] Class materials and tests are automatically generated and provided based on class information.

[0671] Receives test result input and performs automatic scoring.

[0672] Generate learning materials based on requests for learning support for children who are not attending school and monitor their progress.

[0673] 2. Terminal

[0674] Provides an interface for teachers to enter lesson information.

[0675] Receive and download generated course materials and tests.

[0676] The teacher enters the test results and sends them to the server.

[0677] Children who are not attending school and their parents can submit requests for learning support.

[0678] 3. Users

[0679] The main users are expected to be teachers and school-refusing children (or their parents).

[0680] Teachers enter lesson information, download generated teaching materials and tests, and enter test results.

[0681] Children who are not attending school and their parents can submit requests for learning support and study using the provided teaching materials.

[0682] Program processing flow

[0683] Collection of lesson data

[0684] The server collects lesson data from each educational institution, including, for example, mathematics lesson content, teaching materials, test questions, and feedback.

[0685] The server anonymizes the collected data and removes any personal information.

[0686] The server generalizes the anonymized data, converting it so that it is not tied to any particular situation.

[0687] Training an AI model

[0688] The server trains the AI ​​model using anonymized and generalized data.

[0689] The server periodically retrains the AI ​​model using the latest lesson data stored in the database.

[0690] Automatic generation of teaching materials

[0691] The user (teacher) inputs information such as the lesson theme, target grade, and lesson time via the terminal.

[0692] The terminal transmits the input information to the server.

[0693] The server automatically generates appropriate lesson materials based on the lesson information entered, such as history lesson slides and audio files.

[0694] The server transmits the generated lesson materials to the teacher's terminal.

[0695] The user (teacher) receives the generated teaching materials and uses them in class.

[0696] Automatic test generation and scoring

[0697] The user (teacher) instructs the creation of a test on a terminal, inputting the scope, difficulty level, and format.

[0698] The terminal transmits the input information to the server.

[0699] The server automatically generates tests and provides them to teachers' terminals in PDF format, for example, generating multiple-choice and essay questions for a science test.

[0700] The user (teacher) distributes the generated test to students.

[0701] The user (teacher) enters the test results into the terminal.

[0702] The device sends the test results to the server, which then automatically scores them.

[0703] The server provides the grading results to the teacher.

[0704] Learning support for children who are not attending school

[0705] Users (children not attending school or their parents) can send requests for learning support through their devices, for example, requesting a review of math.

[0706] The terminal sends a request to the server.

[0707] The server generates appropriate learning materials based on the request, such as review videos and slides.

[0708] The server provides the generated learning materials to the user.

[0709] Users (children not attending school) study at home using the provided teaching materials.

[0710] The server monitors learning progress and provides additional learning materials and advice as needed.

[0711] The AI ​​teaching system of this invention can significantly reduce the workload of teachers and effectively support the learning of students who are not attending school. A specific example of its operation is a process in which a teacher generates PowerPoint presentations for a history lesson, distributes them to students, and then automatically generates and grades tests. Support for students who are not attending school includes providing study materials that can be used at home and tracking their learning progress.

[0712] The above is a specific embodiment for carrying out the present invention.

[0713] The processing flow will be explained below.

[0714] Program processing steps

[0715] Collection and learning of lesson data

[0716] server

[0717] Step 1:

[0718] The server collects lesson data from each educational institution, providing a mechanism for regularly uploading information such as the content of lessons taught by teachers, teaching materials used, test questions, and student feedback.

[0719] Step 2:

[0720] The server will anonymize the collected class data and remove personal information, for example by implementing algorithms that automatically detect and remove personally identifiable information such as names and student ID numbers.

[0721] Step 3:

[0722] The server then performs a generalization process on the anonymized data, specifically standardizing and storing the unique expressions of specific teachers and schools in an abstracted form.

[0723] Step 4:

[0724] The server uses anonymized and generalized data to train AI models, which use natural language processing (NLP) and machine learning (ML) algorithms to extract patterns in lesson content and effective teaching methods.

[0725] Step 5:

[0726] The server periodically uses lesson data stored in the database to retrain the AI ​​model to reflect the latest educational trends and data.

[0727] Automatic generation of teaching materials

[0728] Terminal

[0729] Step 1:

[0730] Users (teachers) use their terminals to input information such as lesson topic, target grade, lesson time, etc. A form is provided that allows users to easily input information through a dedicated interface.

[0731] Step 2:

[0732] The terminal transmits the input lesson information to the server.

[0733] server

[0734] Step 3:

[0735] Based on the lesson information received, the server searches the database for relevant lesson data and automatically generates optimal lesson materials using an AI model. For example, it creates new teaching materials by referencing past lesson slides, videos, audio files, etc. related to the same topic.

[0736] Step 4:

[0737] The server transmits the generated lesson materials to the teacher's terminal.

[0738] Terminal

[0739] Step 5:

[0740] Teachers receive the teaching materials generated on their devices, download them, edit or revise them as necessary, and use them in class.

[0741] Automated test creation and scoring

[0742] User

[0743] Step 1:

[0744] Teachers send a test creation request to the server via their terminal, specifying the scope, difficulty level, and question format (multiple choice, essay, etc.).

[0745] server

[0746] Step 2:

[0747] The server references the database based on the specified conditions and automatically generates optimal test questions from past data, using an AI model to extract questions of appropriate difficulty and content.

[0748] Step 3:

[0749] The server sends the automatically generated test to the teacher's device in PDF or other format.

[0750] Terminal

[0751] Step 4:

[0752] Teachers receive the tests generated on their devices, print them out as needed, and distribute them to students.

[0753] User

[0754] Step 5:

[0755] After students take the test, teachers enter the test results into a terminal.

[0756] server

[0757] Step 6:

[0758] The device sends the entered test results to a server, which then automatically scores them. Multiple-choice questions are instantly graded by AI, while essay questions are evaluated using NLP technology.

[0759] Step 7:

[0760] The server provides the grading results in the form of a report to the teacher, who can then check the results and make corrections as necessary.

[0761] Learning support for children who are not attending school

[0762] User

[0763] Step 1:

[0764] Children who are not attending school and their parents can send requests for learning support via their devices, specifically by entering information such as the subject, scope, and grade level they wish to study.

[0765] Terminal

[0766] Step 2:

[0767] The terminal transmits the request content to the server.

[0768] server

[0769] Step 3:

[0770] Based on the request, the server searches the database for relevant lesson data and automatically generates appropriate learning materials, such as review videos, slides, and practice questions.

[0771] Step 4:

[0772] The server provides the generated learning materials to the user, who then sends them to the user's device and makes them available for download.

[0773] Terminal

[0774] Step 5:

[0775] Users (children not attending school) download the provided teaching materials and use them for studying at home.

[0776] server

[0777] Step 6:

[0778] The server periodically monitors the user's learning progress, assesses their level of understanding, and provides additional learning materials and advice as needed.

[0779] The above are the specific processing steps of the system of the present invention. This system reduces the workload of teachers and effectively provides learning support to children who do not attend school.

[0780] Example 1

[0781] 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."

[0782] In today's educational environment, teachers' workloads are increasing, with much of their time being spent on lesson preparation and test creation / grading. This has resulted in situations where teachers are unable to concentrate on their primary educational activities. Supporting the learning of children who are not attending school is also an issue, with insufficient provision of appropriate teaching materials and monitoring of their learning progress. Effective methods are needed to resolve these issues and improve the quality of education.

[0783] 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.

[0784] In this invention, the server includes means for collecting lesson data, means for anonymizing and generalizing the collected lesson data, means for performing machine learning using the anonymized and generalized data, means for automatically generating lesson materials based on lesson information, means for providing the generated lesson materials, and means for monitoring learning progress and providing additional teaching materials and advice. This makes it possible to automate lesson preparation and test creation / grading, reducing the workload of teachers and providing effective learning support for children who are not attending school.

[0785] "Class data" refers to information including lesson content, teaching materials, test questions, feedback, etc. in educational settings.

[0786] "Anonymization" is the process of removing or obscuring personally identifiable information from collected data.

[0787] "Generalization" refers to standardizing data in a way that is independent of specific conditions or situations.

[0788] "Machine learning" is a technology that uses massive amounts of data to enable computers to automatically learn patterns and perform specific tasks.

[0789] "Classroom materials" are content such as documents, slides, audio files, and videos created to support educational activities.

[0790] "Auto-generation" is the process of using artificial intelligence or algorithms to generate content or data with minimal human intervention.

[0791] "Study progress" is an indicator of how far a student has progressed in their studies.

[0792] "Advice" is advice or guidance provided to students and teachers based on their learning progress.

[0793] "Tests" refer to question sets and exams used to assess students' understanding and learning status.

[0794] "Scoring" is the process of evaluating test responses and assigning a score or grade.

[0795] A "request" is an act by a user requesting a particular service or information.

[0796] MODE FOR CARRYING OUT THE INVENTION

[0797] This invention is an AI teacher system that aims to reduce the workload of teachers and support the learning of students who are not attending school. This system utilizes lesson data from educational institutions across the country, and AI automates lesson preparation, test creation and grading, as well as monitoring and supporting learning progress.

[0798] Basic system configuration

[0799] The system consists of the following main components:

[0800] 1. Server

[0801] Collect, anonymize, and generalize class data.

[0802] Train and retrain AI models using machine learning.

[0803] Class materials and tests are automatically generated and provided based on class information.

[0804] Receives test result input and performs automatic scoring.

[0805] Monitor your progress and provide additional learning materials and advice.

[0806] 2. Terminal

[0807] Provides an interface for teachers to enter lesson information.

[0808] Receive and download generated course materials and tests.

[0809] The teacher enters the test results and sends them to the server.

[0810] Provide an interface for accepting requests for learning support.

[0811] 3. Users

[0812] The main users are expected to be teachers and school-refusing children (or their parents).

[0813] Teachers enter lesson information, download generated teaching materials and tests, and enter test results.

[0814] Children who are not attending school and their parents can submit requests for learning support and study using the provided teaching materials.

[0815] Collection of lesson data

[0816] The server receives lesson data directly from educational institutions across the country, automatically retrieving it from school databases and online platforms, and can use existing APIs and data feeds.

[0817] Data anonymization and generalization

[0818] The server detects any personally identifiable information from the received lesson data and performs anonymization processing, which removes or conceals personal information such as student and teacher names. The anonymized data is generalized and not dependent on a specific region or school, and is standardized to be compatible with other datasets.

[0819] Training an AI model

[0820] The server trains the AI ​​model using anonymized and generalized data. Specifically, it builds a neural network and optimizes the model based on the collected data. This is done using specialized hardware such as high-performance GPUs and TPUs. The AI ​​model is periodically retrained using the latest data stored in the database to improve its accuracy.

[0821] Automatic generation of teaching materials

[0822] The user (teacher) inputs information such as the lesson topic, target grade, and lesson time via the device. This information is sent from the device to the server, and the AI ​​model uses generative AI technology to automatically generate lesson materials. For example, slides, audio files, and video materials for a history lesson can be generated. The generated materials are sent from the server to the device, where the teacher can download them and use them in class.

[0823] Automatic test generation and scoring

[0824] The user (teacher) issues instructions for creating a test via their device. The scope, difficulty level, and question format (multiple choice, essay, etc.) are entered and sent from the device to the server. The server automatically generates the test based on the entered information. The generated test is provided in PDF format or similar, which the teacher downloads and distributes to students. When students enter their test results, the device sends this data to the server, which then automatically grades them. The graded results are provided to the teacher, along with feedback.

[0825] Request learning support

[0826] The user (a child not attending school or a parent) sends a request for learning support via their device. For example, they may request a "math review." The request is sent from the device to the server, which automatically generates learning materials based on the request. For example, it may generate videos or slides for reviewing math. The generated learning materials are provided to the user, who studies them at home. The server monitors the learning progress and provides additional learning materials or advice as needed.

[0827] Examples of prompt statements

[0828] The following prompt sentences are used:

[0829] "I want to generate teaching materials for history classes."

[0830] "I want you to create a math test."

[0831] "Please provide math review materials."

[0832] The AI ​​teacher system of the present invention can significantly reduce the workload of teachers and provide effective learning support for children who are not attending school.

[0833] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0834] Program processing steps

[0835] Step 1: Collect lesson data

[0836] Specific behavior:

[0837] The server receives lesson data directly from educational institutions across the country, automatically retrieving data from school databases and online platforms via APIs.

[0838] Input: Data sent by educational institutions, such as course content, study materials, exam questions, and feedback.

[0839] Output: Raw lesson data collected.

[0840] Step 2: Anonymize and generalize data

[0841] Specific behavior:

[0842] The server then detects any personally identifiable information from the collected data and performs anonymization processing, for example, by using a filtering algorithm to remove personal information such as names and addresses.

[0843] The server generalizes the anonymized data, standardizing it to convert it into a format that is independent of specific schools or regions.

[0844] Input: Raw lesson data collected.

[0845] Output: Anonymized and generalized lesson data.

[0846] Step 3: Training the AI ​​model

[0847] Specific behavior:

[0848] The server uses anonymized and generalized data to train AI models, using neural networks and other machine learning techniques to train algorithms that generate educational content.

[0849] The server periodically retrains the AI ​​model with the latest data, thereby maintaining the model's accuracy and effectiveness.

[0850] Input: Anonymized and generalized lesson data.

[0851] Output: A trained AI model.

[0852] Step 4: Automatic generation of lesson materials

[0853] Specific behavior:

[0854] The user (teacher) uses a terminal to input information such as the lesson theme, target grade, lesson time, etc. This transmits specific lesson needs to the server.

[0855] The terminal transmits the input information to the server.

[0856] The server uses a generative AI model to automatically generate lesson materials based on the input lesson information, such as history lesson slides, audio files, and videos.

[0857] The server transmits the generated lesson materials to the teacher's terminal.

[0858] The user (teacher) checks the generated lesson materials and makes fine adjustments as necessary.

[0859] Input: Information such as lesson topic, target grade, lesson time, etc.

[0860] Output: Generated lesson materials (slides, audio files, videos, etc.).

[0861] Step 5: Auto-generate and grade tests

[0862] Specific behavior:

[0863] The user (teacher) uses a terminal to instruct the creation of a test, inputting the scope, difficulty level, and question format (multiple choice, essay, etc.).

[0864] The terminal transmits the input information to the server.

[0865] The server uses a generative AI model to automatically generate tests based on the input information, such as multiple-choice and essay questions for science, in PDF format.

[0866] The server provides the generated test to the teacher's terminal, who then distributes it to the students.

[0867] After the students take the test, the user (teacher) enters the results into the terminal.

[0868] The device sends the entered test results to the server, which then automatically scores them using a scoring algorithm for fast and accurate evaluation.

[0869] The server provides the grading results to the instructor and generates feedback.

[0870] Input: Test scope, difficulty level, question format. Test results.

[0871] Output: Auto-generated test, marking results and feedback.

[0872] Step 6: Request and provide learning support

[0873] Specific behavior:

[0874] A user (a child not attending school or a parent) uses a device to send a request for learning support, for example, a request for "math review."

[0875] The terminal sends a request to the server.

[0876] The server automatically generates learning materials based on the request using a generative AI model, such as review videos and slides.

[0877] The server provides the generated learning materials to the user.

[0878] Users (children not attending school) study at home using the provided teaching materials.

[0879] The server monitors learning progress and provides additional learning materials and advice as needed.

[0880] Enter: Request for study support.

[0881] Output: Generated learning materials, learning progress data, and further learning materials and advice.

[0882] This specific process will reduce the workload of teachers and provide effective learning support to children who are not attending school.

[0883] (Application example 1)

[0884] 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."

[0885] In the traditional education system, teachers are overwhelmed with the tasks of preparing lessons and creating and grading tests, making it difficult to provide adequate learning support for students who are not attending school. Furthermore, educational methods using virtual environments are limited, making it difficult to provide customized learning materials tailored to individual students' learning progress. This situation hinders the efficient use of educational resources and the improvement of educational quality.

[0886] 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.

[0887] In this invention, the server includes means for collecting lesson data, means for anonymizing and generalizing the collected lesson data, means for performing machine learning using the anonymized and generalized data, means for automatically generating lesson materials based on lesson information, means for providing the generated lesson materials, means for delivering lessons in a virtual environment, and means for tracking the learning progress of children who are not attending school and providing customized teaching materials. This reduces the workload of teachers and makes it possible to provide effective individual learning support to children who are not attending school.

[0888] "Class data" refers to information such as class content, teaching materials, test questions, and feedback collected from educational institutions.

[0889] "Anonymization" means processing data so that personal information is removed and individuals cannot be identified.

[0890] "Generalization" means transforming data so that it can be used widely and is not tied to a specific situation.

[0891] "Machine learning" is a technology that allows computers to automatically learn from data and make future predictions and classifications.

[0892] "Class information" refers to specific information about the implementation of a class, such as the class theme, target grade, and class time.

[0893] "Classroom teaching materials" are materials such as slides, audio files, and videos used in conducting lessons.

[0894] A "virtual environment" is a virtual space provided via the Internet, rather than a physical classroom or location.

[0895] A "school refusal student" is a student who is unable to attend school for an extended period of time for some reason.

[0896] "Learning progress" refers to a student's progress and achievement in learning.

[0897] "Customized learning materials" refer to learning resources that are optimized according to the learning situation and needs of each individual student.

[0898] This invention is an AI teaching system that aims to reduce the workload of teachers and support the learning of children who are not attending school. This system collects lesson data and uses AI to automate and support lesson preparation, test creation, and grading. It also delivers lessons through a virtual environment and provides customized teaching materials to children who are not attending school, thereby supporting their learning.

[0899] Basic system configuration

[0900] The system consists of the following main components:

[0901] 1. Server

[0902] Classroom data is collected, anonymized, and generalized.

[0903] Train and retrain AI models using machine learning.

[0904] Class materials are automatically generated and provided based on class information.

[0905] Deliver lessons in a virtual environment.

[0906] Automatically generate and grade tests.

[0907] Track the learning progress of children who are not attending school and generate and provide customized learning materials.

[0908] 2. Terminal

[0909] Provides an interface for teachers to enter lesson information.

[0910] Receive and download created course materials and tests.

[0911] Classes are delivered virtually and can be viewed live or recorded.

[0912] The teacher enters the test results and sends them to the server.

[0913] Provides an interface for school-refusing children and their parents to request learning support.

[0914] 3. Users

[0915] The main users are teachers, children who are not attending school, and their parents.

[0916] Teachers enter lesson information, download generated teaching materials and tests, and enter test results.

[0917] Children who are not attending school and their parents can enter requests for learning support and study using the provided teaching materials.

[0918] Program processing flow

[0919] Hardware and Software Use

[0920] Hardware: Smartphones and head-mounted displays are used, which allow for real-time streaming of classes and recordings for viewing.

[0921] Software used: Python's aiohttp library is used to perform asynchronous HTTP requests and communicate with the server.

[0922] Data processing and calculation

[0923] 1. Collection and anonymization of lesson data

[0924] The server collects lesson data from multiple educational institutions, anonymizes and generalizes the data, removing personal information and converting it into a common format.

[0925] 2. Training the AI ​​model

[0926] The server uses anonymized and generalized data to train AI models, which will automate the generation of future course materials and test creation.

[0927] 3. Automatic generation of teaching materials and tests

[0928] When teachers input information about the lesson theme and target grade through the terminal, the server automatically generates lesson materials based on that information and sends them back to the terminal.

[0929] Similarly, you can enter the scope and format of the test and the server will automatically generate the test and provide it in PDF format.

[0930] 4. Learning support

[0931] When a child who is not attending school or their parents request learning support, the server generates and provides customized teaching materials and tests based on the child's individual learning progress.

[0932] Examples and prompts

[0933] Examples:

[0934] Lesson theme: "Japanese History"

[0935] Target grade: 2nd year junior high school students

[0936] Lesson duration: 45 minutes

[0937] Example prompt sentence:

[0938] Lesson theme: Japanese history

[0939] Target grade: 2nd year junior high school students

[0940] Lesson duration: 45 minutes

[0941] Generated teaching materials: PowerPoint slides, lesson videos

[0942] The AI ​​teaching system of this invention reduces the workload of teachers and effectively supports the learning of students who are not attending school. In addition, by utilizing a virtual environment, it is possible to deliver real-time and recorded lessons, thereby improving the quality of education.

[0943] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0944] Step 1:

[0945] The server collects lesson data from educational institutions. The collected lesson data includes lesson content, teaching materials, test questions, and feedback. It receives the lesson data as input, anonymizes and generalizes it to remove specific personal information and convert it into a common format. It generates the anonymized and generalized data as output.

[0946] Step 2:

[0947] The server performs machine learning using the anonymized and generalized data from step 1. Specifically, it trains a generative AI model based on the data and updates the AI ​​model. It receives the anonymized data as input and outputs the trained AI model.

[0948] Step 3:

[0949] The user (teacher) inputs lesson information such as lesson theme, target grade, lesson time, etc. through the terminal. The input lesson information is sent from the terminal to the server.

[0950] Step 4:

[0951] The server automatically generates lesson materials based on the lesson information entered in step 3. It uses a generative AI model to generate appropriate lesson slides and audio files. It receives lesson information as input and outputs lesson materials.

[0952] Step 5:

[0953] The server sends the generated lesson materials to the terminal. The terminal provides the lesson materials received from the server to the teacher. The teacher downloads the materials for use in the class. The terminal receives the lesson materials as input and outputs the materials sent to the terminal.

[0954] Step 6:

[0955] The user (teacher) uses a terminal to give instructions for creating a test. Specifically, they input the scope, difficulty, and format of the test. The input test information is sent from the terminal to the server.

[0956] Step 7:

[0957] The server automatically generates a test based on the test information entered in step 6. It uses a generative AI model to create a test including multiple choice and essay questions and generates it in PDF format. It receives test information as input and outputs the generated test.

[0958] Step 8:

[0959] The user (teacher) distributes the generated test to students and inputs the test results into the terminal, which then sends the input results to the server.

[0960] Step 9:

[0961] The server receives the test results entered in step 8 and automatically grades them using AI. It receives the test results as input and outputs the graded results. The graded results are sent to the terminal and provided to the teacher.

[0962] Step 10:

[0963] The user (a child not attending school or a parent) sends a request for learning support via the device. Specifically, they request the target subjects and learning content. The input request information is sent from the device to the server.

[0964] Step 11:

[0965] The server generates customized learning materials based on the request information entered in step 10. It uses a generative AI model to create review videos and slides. It receives the request information as input and outputs the generated learning materials. The output learning materials are sent to the terminal and provided to the child who is not attending school and their parents.

[0966] Step 12:

[0967] The server monitors the learning progress of students who are not attending school and provides additional learning materials or advice as needed. It is capable of receiving learning progress data as input and outputting appropriate learning support.

[0968] 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.

[0969] This invention is an AI teaching system designed to reduce the workload of teachers and support the learning of children who are not attending school. This system utilizes lesson data from across the country, and AI automates and supports lesson preparation, test creation, and grading. In addition, by combining it with an emotion engine that recognizes the user's emotions, it provides educational support that is more tailored to each individual user.

[0970] Basic system configuration

[0971] The system consists of the following main components:

[0972] 1. Server

[0973] Collect, anonymize, and generalize class data.

[0974] Train and retrain AI models using machine learning.

[0975] Class materials and tests are automatically generated and provided based on class information.

[0976] Receives test result input and performs automatic scoring.

[0977] Generate learning materials based on requests for learning support for children who are not attending school and monitor their progress.

[0978] It integrates an emotion engine that recognizes the user's emotions and adjusts lesson content and tests.

[0979] 2. Terminal

[0980] Provides an interface for teachers to enter lesson information.

[0981] Receive and download generated course materials and tests.

[0982] The teacher enters the test results and sends them to the server.

[0983] Children who are not attending school and their parents can submit requests for learning support.

[0984] It receives feedback from the emotion engine and transmits the user's state to the server through the emotion recognition function.

[0985] 3. Users

[0986] The main users are expected to be teachers and school-refusing children (or their parents).

[0987] Teachers enter lesson information, download generated teaching materials and tests, and enter test results.

[0988] Children who are not attending school and their parents can submit requests for learning support and study using the provided teaching materials.

[0989] While receiving feedback from the emotion engine, the user receives appropriate learning support based on their own emotional state.

[0990] Program processing flow

[0991] Collection of lesson data

[0992] The server collects lesson data from each educational institution, including, for example, mathematics lesson content, teaching materials, test questions, and feedback.

[0993] The server anonymizes the collected data and removes any personal information.

[0994] The server generalizes the anonymized data, converting it so that it is not tied to any particular situation.

[0995] Training an AI model

[0996] The server trains the AI ​​model using anonymized and generalized data.

[0997] The server periodically uses the latest lesson data stored in the database to retrain the AI ​​model, ensuring it reflects the latest educational trends and data.

[0998] Automatic generation of teaching materials

[0999] The user (teacher) inputs information such as the lesson theme, target grade, and lesson time via the terminal.

[1000] The terminal transmits the input information to the server.

[1001] Based on the lesson information entered, the server searches the database for relevant lesson data and uses an AI model to automatically generate optimal lesson materials, such as history lesson slides and audio files.

[1002] The server transmits the generated lesson materials to the teacher's terminal.

[1003] The user (teacher) receives the generated teaching materials and uses them in class.

[1004] Automatic test generation and scoring

[1005] The user (teacher) sends a test creation request to the server from their device, inputting the scope, difficulty level, and format.

[1006] The terminal transmits the input information to the server.

[1007] The server automatically generates tests and provides them to teachers' terminals in PDF format, for example, generating multiple-choice and essay questions for a science test.

[1008] The user (teacher) distributes the generated test to students.

[1009] The user (teacher) enters the test results into the terminal.

[1010] The device sends the test results to a server, which then automatically scores them. Multiple-choice questions are instantly scored by AI, while essay questions are evaluated using NLP technology.

[1011] The server provides the grading results to the teacher.

[1012] Learning support for children who are not attending school

[1013] Users (children not attending school or their parents) send a request for learning support via their device. Specifically, they input information such as the subject, scope, and grade they wish to study.

[1014] The terminal transmits the request content to the server.

[1015] The server searches for relevant lesson data based on the request and automatically generates appropriate learning materials, such as review videos, slides, and practice questions.

[1016] The server provides the generated learning materials to the user, who then sends them to the user's device and makes them available for download.

[1017] Users (children not attending school) download the provided teaching materials and use them for studying at home.

[1018] The server periodically monitors the user's learning progress, assesses their level of understanding, and provides additional learning materials and advice as needed.

[1019] Emotion engine integration

[1020] The terminal and server recognize the user's emotions through an emotion engine, for example, by using facial recognition technology and voice analysis to analyze the user's emotional state.

[1021] The server then adjusts course materials and test content based on feedback from the emotion engine, for example by providing less difficult questions to students who are nervous.

[1022] Similarly, learning support for users (children not attending school) is adjusted based on data from the emotion engine. If learning progress is falling behind or motivation is declining, additional support is provided.

[1023] As a concrete example, when a teacher is teaching a history class, the emotion engine first recognizes the emotional state of the teacher and students.The server then generates the optimal slides or audio materials for the lesson based on the recognized data and provides them to the devices.In addition, when conducting a test, measures are taken to increase motivation by providing relatively easy questions to students who are feeling nervous or stressed.

[1024] The collaboration between the AI ​​teacher system and emotion engine of this invention reduces the workload of teachers and enables more personalized educational support for each student. Learning support for children who are not attending school can also be provided optimally by understanding changes in their emotions, which will greatly contribute to reducing educational disparities.

[1025] The above is a specific embodiment for carrying out the present invention.

[1026] The processing flow will be explained below.

[1027] Program processing steps

[1028] Collection and learning of lesson data

[1029] server

[1030] Step 1:

[1031] The server collects lesson data from each educational institution, including the teaching materials used by teachers, lesson content, test questions, and student feedback.

[1032] Step 2:

[1033] The server anonymizes the collected lesson data and removes personal information, automatically detecting and deleting personally identifiable information such as names and student numbers.

[1034] Step 3:

[1035] The server generalizes the anonymized data by standardizing expressions specific to specific teachers and schools and storing them in a generic format.

[1036] Step 4:

[1037] The server trains the AI ​​model using anonymized and generalized data, and uses NLP and machine learning algorithms to extract features of the lesson content and teach the AI ​​model.

[1038] Step 5:

[1039] The server periodically retrains the AI ​​model using the latest lesson data stored in the database, ensuring that the model reflects the latest educational trends and data.

[1040] Automatic generation of teaching materials

[1041] Terminal

[1042] Step 1:

[1043] The user (teacher) inputs information such as the lesson theme, target grade, and lesson time via the terminal.

[1044] Step 2:

[1045] The terminal transmits the input lesson information to the server.

[1046] server

[1047] Step 3:

[1048] The server searches the database for relevant lesson data based on the lesson information received, and uses AI models to generate optimal lesson materials, such as history lesson slides and audio files.

[1049] Step 4:

[1050] The server transmits the generated lesson materials to the teacher's terminal.

[1051] Terminal

[1052] Step 5:

[1053] Teachers receive the teaching materials generated on their devices, download them, edit or revise them as necessary, and use them in class.

[1054] Automated test creation and scoring

[1055] User

[1056] Step 1:

[1057] Teachers can create tests through their terminals, inputting the scope, difficulty level, and format.

[1058] Terminal

[1059] Step 2:

[1060] The terminal transmits the input information to the server.

[1061] server

[1062] Step 3:

[1063] The server generates optimal test questions from a database based on the specified conditions. Multiple choice and essay questions are extracted and generated using an AI model.

[1064] Step 4:

[1065] The server provides automatically generated tests to teachers' terminals in PDF format or other formats.

[1066] Terminal

[1067] Step 5:

[1068] Teachers receive the tests generated on their devices, print them out as needed, and distribute them to students.

[1069] User

[1070] Step 6:

[1071] After students take the test, teachers enter the test results into a terminal.

[1072] server

[1073] Step 7:

[1074] The device sends the entered test results to a server, which then automatically scores them. Multiple-choice questions are instantly graded by AI, while essay questions are evaluated using NLP technology.

[1075] Step 8:

[1076] The server provides the grading results to the teacher, who can review the results and make corrections if necessary.

[1077] Learning support for children who are not attending school

[1078] User

[1079] Step 1:

[1080] Children who are not attending school and their parents can send requests for learning support via their devices, inputting the subject, scope, grade level, etc. they wish to study.

[1081] Terminal

[1082] Step 2:

[1083] The terminal sends the request contents to the server.

[1084] server

[1085] Step 3:

[1086] Based on the request, the server searches the database for relevant lesson data and generates appropriate learning materials, such as review videos and slides.

[1087] Step 4:

[1088] The server transmits the generated learning materials to the user's terminal.

[1089] Terminal

[1090] Step 5:

[1091] Users (children not attending school) receive the provided teaching materials on their devices, download them, and study at home.

[1092] server

[1093] Step 6:

[1094] The server periodically monitors the user's learning progress, assesses their level of understanding, and provides additional learning materials and advice as needed.

[1095] Emotion engine integration

[1096] Terminal

[1097] Step 1:

[1098] Users (teachers and students) activate the emotion engine through their devices during lessons and tests.

[1099] Step 2:

[1100] The emotion engine uses facial recognition technology and voice analysis to recognize the user's emotions, for example, whether the user is nervous or stressed.

[1101] server

[1102] Step 3:

[1103] The server receives feedback from the emotion engine and adjusts the difficulty of lessons and tests based on that data, for example providing students who are nervous with easier questions.

[1104] Step 4:

[1105] The server also adjusts learning support for students who are not attending school based on the emotion recognition results, for example, by providing additional support if their learning progress is falling behind or their motivation is declining.

[1106] Terminal

[1107] Step 5:

[1108] The user (a child not attending school) receives feedback from the emotion engine and receives appropriate learning support based on their own emotional state.

[1109] Through these processing steps, the system of the present invention significantly reduces the workload of teachers and provides optimal educational support tailored to the emotional state of each individual user. Specific operational examples include providing tests with adjusted difficulty to nervous students and providing additional support to motivate students who are not attending school. In this way, the quality and effectiveness of education can be improved.

[1110] Example 2

[1111] 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."

[1112] In today's educational environment, the increasing workload of teachers is a major problem. Repetitive tasks such as lesson preparation and test creation / grading place a particularly heavy burden on teachers. Furthermore, teachers' workloads are reaching their limits as they are required to provide learning support for children who are not attending school and individual attention to each student. A system is needed to effectively resolve these issues and improve the quality of education.

[1113] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting lesson data, means for anonymizing and generalizing the collected lesson data, means for performing machine learning using the anonymized and generalized data, means for automatically generating lesson materials based on lesson information, means for providing the generated lesson materials, means for analyzing user emotions using an emotion recognition function, and means for adjusting the content of lesson materials and tests based on the analyzed emotion data. This reduces the workload of teachers and enables personalized educational support for each student, including children who are not attending school.

[1114] "Class Data" is a general term for information including the content of classes conducted at educational institutions, teaching materials, test questions, and feedback.

[1115] "Anonymization" refers to the process of removing personal information from data and converting it into a state in which specific individuals cannot be identified.

[1116] "Generalization" refers to the process of converting data into a general-purpose format so that it is not tied to a specific situation.

[1117] "Machine learning" refers to a technology in which a computer automatically learns patterns using data and makes predictions and classifications.

[1118] "Class information" is a general term for information such as themes, target grades, and class times that teachers use to plan their lessons.

[1119] "Automatic generation" refers to the process of automatically creating data and information using AI or algorithms.

[1120] "Providing" refers to making the generated data or information available to users.

[1121] "Emotion recognition function" refers to technology that analyzes a user's emotional state using facial recognition technology and voice analysis.

[1122] "Analyzing" refers to the process of analyzing collected data and extracting meaningful information.

[1123] "Adjust" refers to the process of changing the content of data or information in response to specific conditions or circumstances.

[1124] This invention is an AI teaching system designed to reduce the workload of teachers and support the learning of children who are not attending school. This system utilizes lesson data from across the country, and AI automates and supports lesson preparation, test creation, and grading. In addition, by combining it with an emotion engine that recognizes the user's emotions, it provides educational support that is more tailored to each individual user.

[1125] Basic system configuration

[1126] The system consists of the following main components:

[1127] 1. Server

[1128] The server collects lesson data from educational institutions, anonymizes and generalizes the collected data, for example, using Python data processing libraries (Pandas, NumPy).

[1129] The server uses anonymized and generalized data to train AI models using machine learning libraries (TensorFlow, PyTorch).

[1130] Automatically generate teaching materials based on lesson information and provide them to students. For example, AI can be used to generate lesson slides (PDF format) and audio materials.

[1131] The emotion recognition function analyzes the user's emotions and adjusts the content of class materials and tests based on the analysis. This process uses facial recognition technology (OpenCV) and voice analysis (Librosa).

[1132] 2. Terminal

[1133] The terminal provides an interface for teachers to input lesson information, for example, using a web application (HTML, CSS, JavaScript).

[1134] Receive and download generated course materials and tests.

[1135] The teacher enters the test results and sends them to the server.

[1136] Children who are not attending school and their parents can submit requests for learning support.

[1137] It receives feedback from the emotion engine and sends the user's state to the server.

[1138] 3. Users

[1139] The main users are teachers and children who are not attending school (or their parents).

[1140] Teachers enter lesson information, download generated teaching materials and tests, and enter test results.

[1141] Children who are not attending school and their parents can submit requests for learning support and study using the provided teaching materials.

[1142] While receiving feedback from the emotion engine, appropriate learning support is provided based on the emotional state.

[1143] Specific examples

[1144] 1. Automatic generation of teaching materials

[1145] The user (teacher) enters lesson information for "World War II, 3rd year junior high school students, 50 minutes" through the terminal.

[1146] This information is sent to a server, which uses an AI model to search for relevant lesson data and automatically generate optimal lesson materials, including lesson slides (PDF format) and audio materials.

[1147] Teachers can download the generated teaching materials and use them in their classes.

[1148] 2. Automatic test generation and scoring

[1149] A teacher uses a terminal to send a request to create a test with the following scope: "World War II," difficulty level: "medium," and format: "multiple choice and essay questions."

[1150] This information is sent to a server, which uses an AI model to automatically generate tests.

[1151] The teacher distributes the generated test to the students and later enters the test results into the terminal.

[1152] The device sends the test results to a server, which then automatically grades them. Multiple-choice questions are graded immediately, while essay questions are graded using NLP technology.

[1153] The results of the grading will be provided to the instructor.

[1154] 3. Learning support for children who are not attending school

[1155] A child who is not attending school or their parents can send a request via their device for "Subject: Mathematics, Range: Quadratic Equations, Grade: 2nd Year of Junior High School."

[1156] The server searches for relevant lesson data based on the request and uses an AI model to automatically generate learning materials, including practice questions and instructional videos.

[1157] Parents can download the generated teaching materials, and their children who are not attending school can study at home.

[1158] The server monitors learning progress, assesses comprehension, and provides additional learning materials and study advice.

[1159] 4. Emotion engine integration

[1160] When a teacher is teaching a history lesson, the device uses the device's camera and microphone to recognize the emotional state of the teacher and students.

[1161] The recognized emotion data is sent to the server, which then generates optimal teaching materials for the lesson and provides them to the terminal.

[1162] When administering tests, adjustments are made based on emotional data as appropriate, such as providing less difficult questions to students who are feeling nervous or stressed.

[1163] As described above, the AI ​​teaching system of the present invention reduces the workload of teachers and provides individualized educational support to each student.

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

[1165] Step 1:

[1166] The server collects lesson data from each educational institution. Specifically, the server accesses the educational institution's database via API to obtain data such as lesson content, teaching materials, test questions, and feedback. The input data is educational data in JSON format, and the output data is raw data that is temporarily stored in the server's internal storage.

[1167] Step 2:

[1168] The server anonymizes the collected lesson data. Specifically, it uses Python data processing libraries (Pandas, NumPy) to run algorithms to remove personal information. For example, it converts student names into meaningless codes and replaces IDs with random numbers. The input data is raw lesson data, and the output data is anonymized lesson data.

[1169] Step 3:

[1170] The server generalizes the anonymized data. The server processes the data so that it is not dependent on a specific educational institution or situation. For example, it converts specific school names or place names into general expressions to increase the versatility of the data. The input data is anonymized lesson data, and the output data is generalized lesson data.

[1171] Step 4:

[1172] The server trains an AI model using the generalized data. Specifically, the server uses machine learning libraries (TensorFlow, PyTorch) to generate an AI model based on the dataset. For example, it uses math or history lesson data to train a model that predicts appropriate lesson content. The input data is the generalized lesson data, and the output data is the trained AI model.

[1173] Step 5:

[1174] The user (teacher) inputs lesson information into the terminal. For example, the teacher inputs lesson topic, target grade, lesson time, etc. into a web form. The input data is the lesson information, and the output data is a lesson request sent to the server.

[1175] Step 6:

[1176] The terminal sends the input lesson information to the server. The server receives information such as lesson theme, target grade, and lesson time, and searches the database for relevant lesson data based on that information. The input data is the lesson request, and the output data is the relevant lesson data.

[1177] Step 7:

[1178] The server uses an AI model to automatically generate optimal lesson materials based on the input lesson information. For example, it generates appropriate slides and audio materials based on the history lesson theme. The input data is the relevant lesson data, and the output data is the generated lesson materials (PDF slides and audio files).

[1179] Step 8:

[1180] The server sends the generated lesson materials to the teacher's device. For example, the generated slide data is provided to the teacher via email or cloud storage. The input data is the generated lesson materials, and the output data is the lesson materials in a format that the teacher can access.

[1181] Step 9:

[1182] The teacher sends a test creation request on the terminal. The teacher inputs the scope, difficulty level, and format. For example, the teacher might input "Scope: World War II, Difficulty: Medium, Format: Multiple choice and essay questions." The input data is the test creation request, and the output data is the request information sent to the server.

[1183] Step 10:

[1184] The device sends the input request information to the server. Based on the received information, the server uses an AI model to automatically generate a test. For example, it generates a PDF-format test that includes multiple-choice and essay questions. The input data is the request information, and the output data is the generated test.

[1185] Step 11:

[1186] The server provides the generated test to the teacher's terminal, where the input data is the generated test and the output data is the test materials sent to the teacher's terminal.

[1187] Step 12:

[1188] A teacher inputs test results into a terminal. For example, a teacher grades students' tests and enters the results into a web form. The input data is the student's test results, and the output data is the test result data sent to the server.

[1189] Step 13:

[1190] The device sends the entered test results to a server, which instantly scores multiple-choice questions using an AI model and automatically evaluates essay questions using NLP technology. The input data is the test result data, and the output data is the scoring results.

[1191] Step 14:

[1192] The server provides the grading results to the teacher. For example, it sends a report including the grading results to the teacher by email. The input data is the grading results, and the output data is a grading report to the teacher.

[1193] Step 15:

[1194] A child who is not attending school and their guardian input a request for learning support into the terminal. For example, they input "Subject: Mathematics, Area: Quadratic Equations, Grade: 8th Grade." The input data is the learning support request, and the output data is the request information sent to the server.

[1195] Step 16:

[1196] The device sends the request information to the server, which then generates appropriate learning materials based on the received information, such as a practice problem set or an instructional video. The input data is the request information, and the output data is the generated learning materials.

[1197] Step 17:

[1198] The server provides the generated learning materials to the user. For example, the server provides the generated learning materials to the user in a downloadable format via cloud storage. The input data is the generated learning materials, and the output data is the learning materials in a format that the user can access.

[1199] Step 18:

[1200] The user (a child not attending school) downloads the provided learning materials and studies at home. The input data are the learning materials, and the output data are the user's learning progress.

[1201] Step 19:

[1202] The server periodically monitors the user's learning progress, evaluates their level of understanding, and provides additional learning materials and advice as needed. The input data is learning progress data, and the output data is the level of understanding assessment and additional learning materials.

[1203] Step 20:

[1204] The device and server recognize the user's emotions through an emotion engine. For example, they use the device's camera and microphone to perform facial recognition and voice analysis. The input data is the user's facial expression and voice data, and the output data is the analysis result of the user's emotional state.

[1205] Step 21:

[1206] The server adjusts the content of the lesson materials and tests based on the feedback from the emotion engine, for example, providing easier questions to a nervous user. The input data is the analysis result of the emotional state, and the output data is the adjusted lesson materials and tests.

[1207] (Application example 2)

[1208] 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."

[1209] Conventional customer service in brick-and-mortar stores has had difficulty adequately responding to individual customer emotional states and preferences. Furthermore, privacy protection issues existed when collecting and utilizing customer information. The present invention aims to solve these issues, provide more attentive service to customers, and protect their privacy.

[1210] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1211] In this invention, the server includes means for collecting lesson data, means for anonymizing and generalizing the collected lesson data, means for performing machine learning using the anonymized and generalized data, means for automatically generating lesson materials based on lesson information, means for providing the generated lesson materials, means for collecting customer information, means for anonymizing and generalizing the collected customer information, means for recognizing customer emotions, and means for adjusting response content based on the customer's emotional state. This enables individual responses according to the customer's emotional state and makes it possible to provide high-quality services while protecting the customer's privacy.

[1212] - "Class Data" refers to information related to education, such as class content, teaching materials, and test questions, collected from educational institutions.

[1213] "Classroom teaching materials" are educational materials such as slides, audio files, and handouts used in classes.

[1214] "Anonymization" is the process of removing personal information from collected data so that it is no longer possible to identify a specific individual.

[1215] "Generalization" is the process of transforming anonymized data so that it is not tied to a specific situation.

[1216] "Machine learning" is a technique that uses data to train AI models to perform specific tasks automatically.

[1217] "Customer information" refers to data related to customers at a store, such as customer identification information, purchase history, and behavioral data.

[1218] "Emotion recognition" is a technology that uses facial recognition technology and voice analysis to determine a customer's emotional state from their facial expressions and voice.

[1219] "Adjusting the response" means changing the content and delivery method of the service based on the customer's emotional state.

[1220] "Privacy protection" means properly managing customers' personal information so that it is not made known to third parties.

[1221] An "AI model" is a collection of algorithms that learn from data and perform tasks automatically.

[1222] This invention relates to an AI customer support system for brick-and-mortar stores. This system uses AI to adjust service content based on the individual emotional state of each customer, providing detailed and personalized service to customers.

[1223] Basic system configuration

[1224] The system consists of the following main components:

[1225] 1. Server

[1226] Collect, anonymize, and generalize class data and customer information.

[1227] Train and retrain AI models using machine learning.

[1228] Teaching materials and service content are automatically generated and provided based on lesson information and customer information.

[1229] Integrates an emotion engine that recognizes customer emotions and adjusts service content.

[1230] 2. Terminal

[1231] Provides an interface for store staff to input service information.

[1232] Receive and download generated educational materials and service content.

[1233] The staff enters the service information and sends it to the server.

[1234] Physical store customers improve their service experience through emotion engine feedback.

[1235] 3. Users

[1236] Store staff and customers are expected to be the main users.

[1237] Store staff input service information, download the generated educational materials and service content, and input service information again.

[1238] Customers in physical stores will receive a personalized service experience provided by the system.

[1239] Program processing flow

[1240] Collection of lesson data and customer information

[1241] The server collects learning data and customer information from each brick-and-mortar store, including customer purchase history, behavioral data, and emotional states captured through an emotion engine. This information is acquired using cameras (built into smartphones, smart glasses, and head-mounted displays (HMDs)). The collected data is anonymized and generalized to protect privacy.

[1242] Training an AI model

[1243] The server trains AI models using anonymized and generalized data. Supported software includes machine learning libraries (e.g., scikit-learn, TensorFlow, etc.). The AI ​​models are retrained using the latest data to provide always-updated services.

[1244] Automatic generation of service content

[1245] The user (store staff) inputs service information through a terminal. This includes information on the target customer demographic, desired service content, and target products. The terminal sends the input information to the server, which then uses an AI model to generate optimal service content. For example, it generates an introductory video or promotional information for a specific product. The generated service content is then sent to the store staff's terminal.

[1246] Emotion recognition and response adjustment

[1247] The device and server recognize the customer's emotions through an emotion engine. They analyze the customer's emotional state using facial recognition technology and voice analysis (e.g., DeepFace), and the server adjusts the service content based on the feedback from the emotion engine. For example, a customer who is feeling stressed may be recommended relaxation-related products.

[1248] Specific examples

[1249] When a customer picks up a product in a physical store, the smart glasses capture the action and recognize the customer's emotional state through the emotion engine. In this case, for example, if the smart glasses determine that the customer is interested in a product, they can display detailed descriptions and related promotions on the smart glasses.

[1250] Example prompt sentence:

[1251] "Create a voice guide that explains what products you would suggest to customers when they are stressed and why."

[1252] Example output:

[1253] Recommended product: Relaxation aroma oil

[1254] Reason: Relaxation aroma oils are effective in reducing stress and leaving you feeling refreshed.

[1255] Audio guide example:

[1256] "Hello, it looks like you're feeling stressed. This relaxation aroma oil is effective in reducing stress. Please give it a try."

[1257] In this way, by implementing the AI ​​customer support system of the present invention, it becomes possible to provide individual responses according to the customer's emotional state, thereby achieving both improved customer satisfaction and privacy protection.

[1258] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1259] Step 1:

[1260] The server collects lesson data and customer information. Specifically, it captures customer facial expressions and behavioral data using cameras and sensors in the physical store, and collects information such as shopping history and behavioral patterns. The raw data acquired as input is temporarily stored before being processed.

[1261] Step 2:

[1262] The server anonymizes and generalizes the collected data. Specifically, it removes personally identifiable information and standardizes and aggregates the data. This allows the data to be managed in a unified format and prevents the identification of specific individuals. Raw data is used as input, and anonymized and generalized data is output.

[1263] Step 3:

[1264] The server performs machine learning using the anonymized and generalized data. Specifically, it uses machine learning libraries (e.g., scikit-learn, TensorFlow) to train an AI model based on the data and develops an algorithm to predict customer behavior and sentiment. The anonymized data is used as input, and the trained AI model is obtained as output.

[1265] Step 4:

[1266] The server automatically generates service content based on lesson information and customer information. Specifically, it uses a generative AI model to generate prompts that recommend the most suitable products and services to the customer. The AI ​​model and customer information are used as input, and the generated service content (e.g., a list of recommended products and promotion information) is obtained as output.

[1267] Step 5:

[1268] The terminal receives the generated service content and provides it to the staff at the physical store. Specifically, the generated service content is displayed through an application installed on the terminal. The generated service content is used as input and is available to the staff as output.

[1269] Step 6:

[1270] The device and server recognize the customer's emotions through an emotion engine. Specifically, an emotion analysis library (e.g., DeepFace) is used to analyze the customer's facial expressions and voice to identify their emotional state. Real-time video and audio data is used as input, and the customer's emotional state is obtained as output.

[1271] Step 7:

[1272] The server adjusts the service content based on the feedback from the emotion engine. Specifically, it appropriately modifies the generated service content according to the obtained emotional state, providing personalized support. Using the emotional state and service content as input, the adjusted service content is obtained as output.

[1273] Step 8:

[1274] Users (store customers) receive personalized service experiences provided by the system. Specifically, they receive tailored service content through terminals or store staff. It is expected that users will receive tailored service content as input and achieve high levels of satisfaction as output.

[1275] 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.

[1276] 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.

[1277] 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.

[1278] [Third embodiment]

[1279] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1280] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[1281] 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).

[1282] 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.

[1283] 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.

[1284] 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).

[1285] 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.

[1286] 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.

[1287] 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.

[1288] 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.

[1289] 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.

[1290] 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."

[1291] This invention is an AI teaching system that aims to reduce the workload of teachers and support the learning of children who are not attending school. This system utilizes lesson data from all over the country and uses AI to automate and support lesson preparation, test creation, and grading.

[1292] Basic system configuration

[1293] The system consists of the following main components:

[1294] 1. Server

[1295] Collect, anonymize, and generalize class data.

[1296] Train and retrain AI models using machine learning.

[1297] Class materials and tests are automatically generated and provided based on class information.

[1298] Receives test result input and performs automatic scoring.

[1299] Generate learning materials based on requests for learning support for children who are not attending school and monitor their progress.

[1300] 2. Terminal

[1301] Provides an interface for teachers to enter lesson information.

[1302] Receive and download generated course materials and tests.

[1303] The teacher enters the test results and sends them to the server.

[1304] Children who are not attending school and their parents can submit requests for learning support.

[1305] 3. Users

[1306] The main users are expected to be teachers and school-refusing children (or their parents).

[1307] Teachers enter lesson information, download generated teaching materials and tests, and enter test results.

[1308] Children who are not attending school and their parents can submit requests for learning support and study using the provided teaching materials.

[1309] Program processing flow

[1310] Collection of lesson data

[1311] The server collects lesson data from each educational institution, including, for example, mathematics lesson content, teaching materials, test questions, and feedback.

[1312] The server anonymizes the collected data and removes any personal information.

[1313] The server generalizes the anonymized data, converting it so that it is not tied to any particular situation.

[1314] Training an AI model

[1315] The server trains the AI ​​model using anonymized and generalized data.

[1316] The server periodically retrains the AI ​​model using the latest lesson data stored in the database.

[1317] Automatic generation of teaching materials

[1318] The user (teacher) inputs information such as the lesson theme, target grade, and lesson time via the terminal.

[1319] The terminal transmits the input information to the server.

[1320] The server automatically generates appropriate lesson materials based on the lesson information entered, such as history lesson slides and audio files.

[1321] The server transmits the generated lesson materials to the teacher's terminal.

[1322] The user (teacher) receives the generated teaching materials and uses them in class.

[1323] Automatic test generation and scoring

[1324] The user (teacher) instructs the creation of a test on a terminal, inputting the scope, difficulty level, and format.

[1325] The terminal transmits the input information to the server.

[1326] The server automatically generates tests and provides them to teachers' terminals in PDF format, for example, generating multiple-choice and essay questions for a science test.

[1327] The user (teacher) distributes the generated test to students.

[1328] The user (teacher) enters the test results into the terminal.

[1329] The device sends the test results to the server, which then automatically scores them.

[1330] The server provides the grading results to the teacher.

[1331] Learning support for children who are not attending school

[1332] Users (children not attending school or their parents) can send requests for learning support through their devices, for example, requesting a review of math.

[1333] The terminal sends a request to the server.

[1334] The server generates appropriate learning materials based on the request, such as review videos and slides.

[1335] The server provides the generated learning materials to the user.

[1336] Users (children not attending school) study at home using the provided teaching materials.

[1337] The server monitors learning progress and provides additional learning materials and advice as needed.

[1338] The AI ​​teaching system of this invention can significantly reduce the workload of teachers and effectively support the learning of students who are not attending school. A specific example of its operation is a process in which a teacher generates PowerPoint presentations for a history lesson, distributes them to students, and then automatically generates and grades tests. Support for students who are not attending school includes providing study materials that can be used at home and tracking their learning progress.

[1339] The above is a specific embodiment for carrying out the present invention.

[1340] The processing flow will be explained below.

[1341] Program processing steps

[1342] Collection and learning of lesson data

[1343] server

[1344] Step 1:

[1345] The server collects lesson data from each educational institution, providing a mechanism for regularly uploading information such as the content of lessons taught by teachers, teaching materials used, test questions, and student feedback.

[1346] Step 2:

[1347] The server will anonymize the collected class data and remove personal information, for example by implementing algorithms that automatically detect and remove personally identifiable information such as names and student ID numbers.

[1348] Step 3:

[1349] The server then performs a generalization process on the anonymized data, specifically standardizing and storing the unique expressions of specific teachers and schools in an abstracted form.

[1350] Step 4:

[1351] The server uses anonymized and generalized data to train AI models, which use natural language processing (NLP) and machine learning (ML) algorithms to extract patterns in lesson content and effective teaching methods.

[1352] Step 5:

[1353] The server periodically uses lesson data stored in the database to retrain the AI ​​model to reflect the latest educational trends and data.

[1354] Automatic generation of teaching materials

[1355] Terminal

[1356] Step 1:

[1357] Users (teachers) use their terminals to input information such as lesson topic, target grade, lesson time, etc. A form is provided that allows users to easily input information through a dedicated interface.

[1358] Step 2:

[1359] The terminal transmits the input lesson information to the server.

[1360] server

[1361] Step 3:

[1362] Based on the lesson information received, the server searches the database for relevant lesson data and automatically generates optimal lesson materials using an AI model. For example, it creates new teaching materials by referencing past lesson slides, videos, audio files, etc. related to the same topic.

[1363] Step 4:

[1364] The server transmits the generated lesson materials to the teacher's terminal.

[1365] Terminal

[1366] Step 5:

[1367] Teachers receive the teaching materials generated on their devices, download them, edit or revise them as necessary, and use them in class.

[1368] Automated test creation and scoring

[1369] User

[1370] Step 1:

[1371] Teachers send a test creation request to the server via their terminal, specifying the scope, difficulty level, and question format (multiple choice, essay, etc.).

[1372] server

[1373] Step 2:

[1374] The server references the database based on the specified conditions and automatically generates optimal test questions from past data, using an AI model to extract questions of appropriate difficulty and content.

[1375] Step 3:

[1376] The server sends the automatically generated test to the teacher's device in PDF or other format.

[1377] Terminal

[1378] Step 4:

[1379] Teachers receive the tests generated on their devices, print them out as needed, and distribute them to students.

[1380] User

[1381] Step 5:

[1382] After students take the test, teachers enter the test results into a terminal.

[1383] server

[1384] Step 6:

[1385] The device sends the entered test results to a server, which then automatically scores them. Multiple-choice questions are instantly graded by AI, while essay questions are evaluated using NLP technology.

[1386] Step 7:

[1387] The server provides the grading results in the form of a report to the teacher, who can then check the results and make corrections as necessary.

[1388] Learning support for children who are not attending school

[1389] User

[1390] Step 1:

[1391] Children who are not attending school and their parents can send requests for learning support via their devices, specifically by entering information such as the subject, scope, and grade level they wish to study.

[1392] Terminal

[1393] Step 2:

[1394] The terminal transmits the request content to the server.

[1395] server

[1396] Step 3:

[1397] Based on the request, the server searches the database for relevant lesson data and automatically generates appropriate learning materials, such as review videos, slides, and practice questions.

[1398] Step 4:

[1399] The server provides the generated learning materials to the user, who then sends them to the user's device and makes them available for download.

[1400] Terminal

[1401] Step 5:

[1402] Users (children not attending school) download the provided teaching materials and use them for studying at home.

[1403] server

[1404] Step 6:

[1405] The server periodically monitors the user's learning progress, assesses their level of understanding, and provides additional learning materials and advice as needed.

[1406] The above are the specific processing steps of the system of the present invention. This system reduces the workload of teachers and effectively provides learning support to children who do not attend school.

[1407] Example 1

[1408] 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."

[1409] In today's educational environment, teachers' workloads are increasing, with much of their time being spent on lesson preparation and test creation / grading. This has resulted in situations where teachers are unable to concentrate on their primary educational activities. Supporting the learning of children who are not attending school is also an issue, with insufficient provision of appropriate teaching materials and monitoring of their learning progress. Effective methods are needed to resolve these issues and improve the quality of education.

[1410] 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.

[1411] In this invention, the server includes means for collecting lesson data, means for anonymizing and generalizing the collected lesson data, means for performing machine learning using the anonymized and generalized data, means for automatically generating lesson materials based on lesson information, means for providing the generated lesson materials, and means for monitoring learning progress and providing additional teaching materials and advice. This makes it possible to automate lesson preparation and test creation / grading, reducing the workload of teachers and providing effective learning support for children who are not attending school.

[1412] "Class data" refers to information including lesson content, teaching materials, test questions, feedback, etc. in educational settings.

[1413] "Anonymization" is the process of removing or obscuring personally identifiable information from collected data.

[1414] "Generalization" refers to standardizing data in a way that is independent of specific conditions or situations.

[1415] "Machine learning" is a technology that uses massive amounts of data to enable computers to automatically learn patterns and perform specific tasks.

[1416] "Classroom materials" are content such as documents, slides, audio files, and videos created to support educational activities.

[1417] "Auto-generation" is the process of using artificial intelligence or algorithms to generate content or data with minimal human intervention.

[1418] "Study progress" is an indicator of how far a student has progressed in their studies.

[1419] "Advice" is advice or guidance provided to students and teachers based on their learning progress.

[1420] "Tests" refer to question sets and exams used to assess students' understanding and learning status.

[1421] "Scoring" is the process of evaluating test responses and assigning a score or grade.

[1422] A "request" is an act by a user requesting a particular service or information.

[1423] MODE FOR CARRYING OUT THE INVENTION

[1424] This invention is an AI teacher system that aims to reduce the workload of teachers and support the learning of students who are not attending school. This system utilizes lesson data from educational institutions across the country, and AI automates lesson preparation, test creation and grading, as well as monitoring and supporting learning progress.

[1425] Basic system configuration

[1426] The system consists of the following main components:

[1427] 1. Server

[1428] Collect, anonymize, and generalize class data.

[1429] Train and retrain AI models using machine learning.

[1430] Class materials and tests are automatically generated and provided based on class information.

[1431] Receives test result input and performs automatic scoring.

[1432] Monitor your progress and provide additional learning materials and advice.

[1433] 2. Terminal

[1434] Provides an interface for teachers to enter lesson information.

[1435] Receive and download generated course materials and tests.

[1436] The teacher enters the test results and sends them to the server.

[1437] Provide an interface for accepting requests for learning support.

[1438] 3. Users

[1439] The main users are expected to be teachers and school-refusing children (or their parents).

[1440] Teachers enter lesson information, download generated teaching materials and tests, and enter test results.

[1441] Children who are not attending school and their parents can submit requests for learning support and study using the provided teaching materials.

[1442] Collection of lesson data

[1443] The server receives lesson data directly from educational institutions across the country, automatically retrieving it from school databases and online platforms, and can use existing APIs and data feeds.

[1444] Data anonymization and generalization

[1445] The server detects any personally identifiable information from the received lesson data and performs anonymization processing, which removes or conceals personal information such as student and teacher names. The anonymized data is generalized and not dependent on a specific region or school, and is standardized to be compatible with other datasets.

[1446] Training an AI model

[1447] The server trains the AI ​​model using anonymized and generalized data. Specifically, it builds a neural network and optimizes the model based on the collected data. This is done using specialized hardware such as high-performance GPUs and TPUs. The AI ​​model is periodically retrained using the latest data stored in the database to improve its accuracy.

[1448] Automatic generation of teaching materials

[1449] The user (teacher) inputs information such as the lesson topic, target grade, and lesson time via the device. This information is sent from the device to the server, and the AI ​​model uses generative AI technology to automatically generate lesson materials. For example, slides, audio files, and video materials for a history lesson can be generated. The generated materials are sent from the server to the device, where the teacher can download them and use them in class.

[1450] Automatic test generation and scoring

[1451] The user (teacher) issues instructions for creating a test via their device. The scope, difficulty level, and question format (multiple choice, essay, etc.) are entered and sent from the device to the server. The server automatically generates the test based on the entered information. The generated test is provided in PDF format or similar, which the teacher downloads and distributes to students. When students enter their test results, the device sends this data to the server, which then automatically grades them. The graded results are provided to the teacher, along with feedback.

[1452] Request learning support

[1453] The user (a child not attending school or a parent) sends a request for learning support via their device. For example, they may request a "math review." The request is sent from the device to the server, which automatically generates learning materials based on the request. For example, it may generate videos or slides for reviewing math. The generated learning materials are provided to the user, who studies them at home. The server monitors the learning progress and provides additional learning materials or advice as needed.

[1454] Examples of prompt statements

[1455] The following prompt sentences are used:

[1456] "I want to generate teaching materials for history classes."

[1457] "I want you to create a math test."

[1458] "Please provide math review materials."

[1459] The AI ​​teacher system of the present invention can significantly reduce the workload of teachers and provide effective learning support for children who are not attending school.

[1460] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1461] Program processing steps

[1462] Step 1: Collect lesson data

[1463] Specific behavior:

[1464] The server receives lesson data directly from educational institutions across the country, automatically retrieving data from school databases and online platforms via APIs.

[1465] Input: Data sent by educational institutions, such as course content, study materials, exam questions, and feedback.

[1466] Output: Raw lesson data collected.

[1467] Step 2: Anonymize and generalize data

[1468] Specific behavior:

[1469] The server then detects any personally identifiable information from the collected data and performs anonymization processing, for example, by using a filtering algorithm to remove personal information such as names and addresses.

[1470] The server generalizes the anonymized data, standardizing it to convert it into a format that is independent of specific schools or regions.

[1471] Input: Raw lesson data collected.

[1472] Output: Anonymized and generalized lesson data.

[1473] Step 3: Training the AI ​​model

[1474] Specific behavior:

[1475] The server uses anonymized and generalized data to train AI models, using neural networks and other machine learning techniques to train algorithms that generate educational content.

[1476] The server periodically retrains the AI ​​model with the latest data, thereby maintaining the model's accuracy and effectiveness.

[1477] Input: Anonymized and generalized lesson data.

[1478] Output: A trained AI model.

[1479] Step 4: Automatic generation of lesson materials

[1480] Specific behavior:

[1481] The user (teacher) uses a terminal to input information such as the lesson theme, target grade, lesson time, etc. This transmits specific lesson needs to the server.

[1482] The terminal transmits the input information to the server.

[1483] The server uses a generative AI model to automatically generate lesson materials based on the input lesson information, such as history lesson slides, audio files, and videos.

[1484] The server transmits the generated lesson materials to the teacher's terminal.

[1485] The user (teacher) checks the generated lesson materials and makes fine adjustments as necessary.

[1486] Input: Information such as lesson topic, target grade, lesson time, etc.

[1487] Output: Generated lesson materials (slides, audio files, videos, etc.).

[1488] Step 5: Auto-generate and grade tests

[1489] Specific behavior:

[1490] The user (teacher) uses a terminal to instruct the creation of a test, inputting the scope, difficulty level, and question format (multiple choice, essay, etc.).

[1491] The terminal transmits the input information to the server.

[1492] The server uses a generative AI model to automatically generate tests based on the input information, such as multiple-choice and essay questions for science, in PDF format.

[1493] The server provides the generated test to the teacher's terminal, who then distributes it to the students.

[1494] After the students take the test, the user (teacher) enters the results into the terminal.

[1495] The device sends the entered test results to the server, which then automatically scores them using a scoring algorithm for fast and accurate evaluation.

[1496] The server provides the grading results to the instructor and generates feedback.

[1497] Input: Test scope, difficulty level, question format. Test results.

[1498] Output: Auto-generated test, marking results and feedback.

[1499] Step 6: Request and provide learning support

[1500] Specific behavior:

[1501] A user (a child not attending school or a parent) uses a device to send a request for learning support, for example, a request for "math review."

[1502] The terminal sends a request to the server.

[1503] The server automatically generates learning materials based on the request using a generative AI model, such as review videos and slides.

[1504] The server provides the generated learning materials to the user.

[1505] Users (children not attending school) study at home using the provided teaching materials.

[1506] The server monitors learning progress and provides additional learning materials and advice as needed.

[1507] Enter: Request for study support.

[1508] Output: Generated learning materials, learning progress data, and further learning materials and advice.

[1509] This specific process will reduce the workload of teachers and provide effective learning support to children who are not attending school.

[1510] (Application example 1)

[1511] 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."

[1512] In the traditional education system, teachers are overwhelmed with the tasks of preparing lessons and creating and grading tests, making it difficult to provide adequate learning support for students who are not attending school. Furthermore, educational methods using virtual environments are limited, making it difficult to provide customized learning materials tailored to individual students' learning progress. This situation hinders the efficient use of educational resources and the improvement of educational quality.

[1513] 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.

[1514] In this invention, the server includes means for collecting lesson data, means for anonymizing and generalizing the collected lesson data, means for performing machine learning using the anonymized and generalized data, means for automatically generating lesson materials based on lesson information, means for providing the generated lesson materials, means for delivering lessons in a virtual environment, and means for tracking the learning progress of children who are not attending school and providing customized teaching materials. This reduces the workload of teachers and makes it possible to provide effective individual learning support to children who are not attending school.

[1515] "Class data" refers to information such as class content, teaching materials, test questions, and feedback collected from educational institutions.

[1516] "Anonymization" means processing data so that personal information is removed and individuals cannot be identified.

[1517] "Generalization" means transforming data so that it can be used widely and is not tied to a specific situation.

[1518] "Machine learning" is a technology that allows computers to automatically learn from data and make future predictions and classifications.

[1519] "Class information" refers to specific information about the implementation of a class, such as the class theme, target grade, and class time.

[1520] "Classroom teaching materials" are materials such as slides, audio files, and videos used in conducting lessons.

[1521] A "virtual environment" is a virtual space provided via the Internet, rather than a physical classroom or location.

[1522] A "school refusal student" is a student who is unable to attend school for an extended period of time for some reason.

[1523] "Learning progress" refers to a student's progress and achievement in learning.

[1524] "Customized learning materials" refer to learning resources that are optimized according to the learning situation and needs of each individual student.

[1525] This invention is an AI teaching system that aims to reduce the workload of teachers and support the learning of children who are not attending school. This system collects lesson data and uses AI to automate and support lesson preparation, test creation, and grading. It also delivers lessons through a virtual environment and provides customized teaching materials to children who are not attending school, thereby supporting their learning.

[1526] Basic system configuration

[1527] The system consists of the following main components:

[1528] 1. Server

[1529] Classroom data is collected, anonymized, and generalized.

[1530] Train and retrain AI models using machine learning.

[1531] Class materials are automatically generated and provided based on class information.

[1532] Deliver lessons in a virtual environment.

[1533] Automatically generate and grade tests.

[1534] Track the learning progress of children who are not attending school and generate and provide customized learning materials.

[1535] 2. Terminal

[1536] Provides an interface for teachers to enter lesson information.

[1537] Receive and download created course materials and tests.

[1538] Classes are delivered virtually and can be viewed live or recorded.

[1539] The teacher enters the test results and sends them to the server.

[1540] Provides an interface for school-refusing children and their parents to request learning support.

[1541] 3. Users

[1542] The main users are teachers, children who are not attending school, and their parents.

[1543] Teachers enter lesson information, download generated teaching materials and tests, and enter test results.

[1544] Children who are not attending school and their parents can enter requests for learning support and study using the provided teaching materials.

[1545] Program processing flow

[1546] Hardware and Software Use

[1547] Hardware: Smartphones and head-mounted displays are used, which allow for real-time streaming of classes and recordings for viewing.

[1548] Software used: Python's aiohttp library is used to perform asynchronous HTTP requests and communicate with the server.

[1549] Data processing and calculation

[1550] 1. Collection and anonymization of lesson data

[1551] The server collects lesson data from multiple educational institutions, anonymizes and generalizes the data, removing personal information and converting it into a common format.

[1552] 2. Training the AI ​​model

[1553] The server uses anonymized and generalized data to train AI models, which will automate the generation of future course materials and test creation.

[1554] 3. Automatic generation of teaching materials and tests

[1555] When teachers input information about the lesson theme and target grade through the terminal, the server automatically generates lesson materials based on that information and sends them back to the terminal.

[1556] Similarly, you can enter the scope and format of the test and the server will automatically generate the test and provide it in PDF format.

[1557] 4. Learning support

[1558] When a child who is not attending school or their parents request learning support, the server generates and provides customized teaching materials and tests based on the child's individual learning progress.

[1559] Examples and prompts

[1560] Examples:

[1561] Lesson theme: "Japanese History"

[1562] Target grade: 2nd year junior high school students

[1563] Lesson duration: 45 minutes

[1564] Example prompt sentence:

[1565] Lesson theme: Japanese history

[1566] Target grade: 2nd year junior high school students

[1567] Lesson duration: 45 minutes

[1568] Generated teaching materials: PowerPoint slides, lesson videos

[1569] The AI ​​teaching system of this invention reduces the workload of teachers and effectively supports the learning of students who are not attending school. In addition, by utilizing a virtual environment, it is possible to deliver real-time and recorded lessons, thereby improving the quality of education.

[1570] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1571] Step 1:

[1572] The server collects lesson data from educational institutions. The collected lesson data includes lesson content, teaching materials, test questions, and feedback. It receives the lesson data as input, anonymizes and generalizes it to remove specific personal information and convert it into a common format. It generates the anonymized and generalized data as output.

[1573] Step 2:

[1574] The server performs machine learning using the anonymized and generalized data from step 1. Specifically, it trains a generative AI model based on the data and updates the AI ​​model. It receives the anonymized data as input and outputs the trained AI model.

[1575] Step 3:

[1576] The user (teacher) inputs lesson information such as lesson theme, target grade, lesson time, etc. through the terminal. The input lesson information is sent from the terminal to the server.

[1577] Step 4:

[1578] The server automatically generates lesson materials based on the lesson information entered in step 3. It uses a generative AI model to generate appropriate lesson slides and audio files. It receives lesson information as input and outputs lesson materials.

[1579] Step 5:

[1580] The server sends the generated lesson materials to the terminal. The terminal provides the lesson materials received from the server to the teacher. The teacher downloads the materials for use in the class. The terminal receives the lesson materials as input and outputs the materials sent to the terminal.

[1581] Step 6:

[1582] The user (teacher) uses a terminal to give instructions for creating a test. Specifically, they input the scope, difficulty, and format of the test. The input test information is sent from the terminal to the server.

[1583] Step 7:

[1584] The server automatically generates a test based on the test information entered in step 6. It uses a generative AI model to create a test including multiple choice and essay questions and generates it in PDF format. It receives test information as input and outputs the generated test.

[1585] Step 8:

[1586] The user (teacher) distributes the generated test to students and inputs the test results into the terminal, which then sends the input results to the server.

[1587] Step 9:

[1588] The server receives the test results entered in step 8 and automatically grades them using AI. It receives the test results as input and outputs the graded results. The graded results are sent to the terminal and provided to the teacher.

[1589] Step 10:

[1590] The user (a child not attending school or a parent) sends a request for learning support via the device. Specifically, they request the target subjects and learning content. The input request information is sent from the device to the server.

[1591] Step 11:

[1592] The server generates customized learning materials based on the request information entered in step 10. It uses a generative AI model to create review videos and slides. It receives the request information as input and outputs the generated learning materials. The output learning materials are sent to the terminal and provided to the child who is not attending school and their parents.

[1593] Step 12:

[1594] The server monitors the learning progress of students who are not attending school and provides additional learning materials or advice as needed. It is capable of receiving learning progress data as input and outputting appropriate learning support.

[1595] 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.

[1596] This invention is an AI teaching system designed to reduce the workload of teachers and support the learning of children who are not attending school. This system utilizes lesson data from across the country, and AI automates and supports lesson preparation, test creation, and grading. In addition, by combining it with an emotion engine that recognizes the user's emotions, it provides educational support that is more tailored to each individual user.

[1597] Basic system configuration

[1598] The system consists of the following main components:

[1599] 1. Server

[1600] Collect, anonymize, and generalize class data.

[1601] Train and retrain AI models using machine learning.

[1602] Class materials and tests are automatically generated and provided based on class information.

[1603] Receives test result input and performs automatic scoring.

[1604] Generate learning materials based on requests for learning support for children who are not attending school and monitor their progress.

[1605] It integrates an emotion engine that recognizes the user's emotions and adjusts lesson content and tests.

[1606] 2. Terminal

[1607] Provides an interface for teachers to enter lesson information.

[1608] Receive and download generated course materials and tests.

[1609] The teacher enters the test results and sends them to the server.

[1610] Children who are not attending school and their parents can submit requests for learning support.

[1611] It receives feedback from the emotion engine and transmits the user's state to the server through the emotion recognition function.

[1612] 3. Users

[1613] The main users are expected to be teachers and school-refusing children (or their parents).

[1614] Teachers enter lesson information, download generated teaching materials and tests, and enter test results.

[1615] Children who are not attending school and their parents can submit requests for learning support and study using the provided teaching materials.

[1616] While receiving feedback from the emotion engine, the user receives appropriate learning support based on their own emotional state.

[1617] Program processing flow

[1618] Collection of lesson data

[1619] The server collects lesson data from each educational institution, including, for example, mathematics lesson content, teaching materials, test questions, and feedback.

[1620] The server anonymizes the collected data and removes any personal information.

[1621] The server generalizes the anonymized data, converting it so that it is not tied to any particular situation.

[1622] Training an AI model

[1623] The server trains the AI ​​model using anonymized and generalized data.

[1624] The server periodically uses the latest lesson data stored in the database to retrain the AI ​​model, ensuring it reflects the latest educational trends and data.

[1625] Automatic generation of teaching materials

[1626] The user (teacher) inputs information such as the lesson theme, target grade, and lesson time via the terminal.

[1627] The terminal transmits the input information to the server.

[1628] Based on the lesson information entered, the server searches the database for relevant lesson data and uses an AI model to automatically generate optimal lesson materials, such as history lesson slides and audio files.

[1629] The server transmits the generated lesson materials to the teacher's terminal.

[1630] The user (teacher) receives the generated teaching materials and uses them in class.

[1631] Automatic test generation and scoring

[1632] The user (teacher) sends a test creation request to the server from their device, inputting the scope, difficulty level, and format.

[1633] The terminal transmits the input information to the server.

[1634] The server automatically generates tests and provides them to teachers' terminals in PDF format, for example, generating multiple-choice and essay questions for a science test.

[1635] The user (teacher) distributes the generated test to students.

[1636] The user (teacher) enters the test results into the terminal.

[1637] The device sends the test results to a server, which then automatically scores them. Multiple-choice questions are instantly scored by AI, while essay questions are evaluated using NLP technology.

[1638] The server provides the grading results to the teacher.

[1639] Learning support for children who are not attending school

[1640] Users (children not attending school or their parents) send a request for learning support via their device. Specifically, they input information such as the subject, scope, and grade they wish to study.

[1641] The terminal transmits the request content to the server.

[1642] The server searches for relevant lesson data based on the request and automatically generates appropriate learning materials, such as review videos, slides, and practice questions.

[1643] The server provides the generated learning materials to the user, who then sends them to the user's device and makes them available for download.

[1644] Users (children not attending school) download the provided teaching materials and use them for studying at home.

[1645] The server periodically monitors the user's learning progress, assesses their level of understanding, and provides additional learning materials and advice as needed.

[1646] Emotion engine integration

[1647] The terminal and server recognize the user's emotions through an emotion engine, for example, by using facial recognition technology and voice analysis to analyze the user's emotional state.

[1648] The server then adjusts course materials and test content based on feedback from the emotion engine, for example by providing less difficult questions to students who are nervous.

[1649] Similarly, learning support for users (children not attending school) is adjusted based on data from the emotion engine. If learning progress is falling behind or motivation is declining, additional support is provided.

[1650] As a concrete example, when a teacher is teaching a history class, the emotion engine first recognizes the emotional state of the teacher and students.The server then generates the optimal slides or audio materials for the lesson based on the recognized data and provides them to the devices.In addition, when conducting a test, measures are taken to increase motivation by providing relatively easy questions to students who are feeling nervous or stressed.

[1651] The collaboration between the AI ​​teacher system and emotion engine of this invention reduces the workload of teachers and enables more personalized educational support for each student. Learning support for children who are not attending school can also be provided optimally by understanding changes in their emotions, which will greatly contribute to reducing educational disparities.

[1652] The above is a specific embodiment for carrying out the present invention.

[1653] The processing flow will be explained below.

[1654] Program processing steps

[1655] Collection and learning of lesson data

[1656] server

[1657] Step 1:

[1658] The server collects lesson data from each educational institution, including the teaching materials used by teachers, lesson content, test questions, and student feedback.

[1659] Step 2:

[1660] The server anonymizes the collected lesson data and removes personal information, automatically detecting and deleting personally identifiable information such as names and student numbers.

[1661] Step 3:

[1662] The server generalizes the anonymized data by standardizing expressions specific to specific teachers and schools and storing them in a generic format.

[1663] Step 4:

[1664] The server trains the AI ​​model using anonymized and generalized data, and uses NLP and machine learning algorithms to extract features of the lesson content and teach the AI ​​model.

[1665] Step 5:

[1666] The server periodically retrains the AI ​​model using the latest lesson data stored in the database, ensuring that the model reflects the latest educational trends and data.

[1667] Automatic generation of teaching materials

[1668] Terminal

[1669] Step 1:

[1670] The user (teacher) inputs information such as the lesson theme, target grade, and lesson time via the terminal.

[1671] Step 2:

[1672] The terminal transmits the input lesson information to the server.

[1673] server

[1674] Step 3:

[1675] The server searches the database for relevant lesson data based on the lesson information received, and uses AI models to generate optimal lesson materials, such as history lesson slides and audio files.

[1676] Step 4:

[1677] The server transmits the generated lesson materials to the teacher's terminal.

[1678] Terminal

[1679] Step 5:

[1680] Teachers receive the teaching materials generated on their devices, download them, edit or revise them as necessary, and use them in class.

[1681] Automated test creation and scoring

[1682] User

[1683] Step 1:

[1684] Teachers can create tests through their terminals, inputting the scope, difficulty level, and format.

[1685] Terminal

[1686] Step 2:

[1687] The terminal transmits the input information to the server.

[1688] server

[1689] Step 3:

[1690] The server generates optimal test questions from a database based on the specified conditions. Multiple choice and essay questions are extracted and generated using an AI model.

[1691] Step 4:

[1692] The server provides automatically generated tests to teachers' terminals in PDF format or other formats.

[1693] Terminal

[1694] Step 5:

[1695] Teachers receive the tests generated on their devices, print them out as needed, and distribute them to students.

[1696] User

[1697] Step 6:

[1698] After students take the test, teachers enter the test results into a terminal.

[1699] server

[1700] Step 7:

[1701] The device sends the entered test results to a server, which then automatically scores them. Multiple-choice questions are instantly graded by AI, while essay questions are evaluated using NLP technology.

[1702] Step 8:

[1703] The server provides the grading results to the teacher, who can review the results and make corrections if necessary.

[1704] Learning support for children who are not attending school

[1705] User

[1706] Step 1:

[1707] Children who are not attending school and their parents can send requests for learning support via their devices, inputting the subject, scope, grade level, etc. they wish to study.

[1708] Terminal

[1709] Step 2:

[1710] The terminal sends the request contents to the server.

[1711] server

[1712] Step 3:

[1713] Based on the request, the server searches the database for relevant lesson data and generates appropriate learning materials, such as review videos and slides.

[1714] Step 4:

[1715] The server transmits the generated learning materials to the user's terminal.

[1716] Terminal

[1717] Step 5:

[1718] Users (children not attending school) receive the provided teaching materials on their devices, download them, and study at home.

[1719] server

[1720] Step 6:

[1721] The server periodically monitors the user's learning progress, assesses their level of understanding, and provides additional learning materials and advice as needed.

[1722] Emotion engine integration

[1723] Terminal

[1724] Step 1:

[1725] Users (teachers and students) activate the emotion engine through their devices during lessons and tests.

[1726] Step 2:

[1727] The emotion engine uses facial recognition technology and voice analysis to recognize the user's emotions, for example, whether the user is nervous or stressed.

[1728] server

[1729] Step 3:

[1730] The server receives feedback from the emotion engine and adjusts the difficulty of lessons and tests based on that data, for example providing students who are nervous with easier questions.

[1731] Step 4:

[1732] The server also adjusts learning support for students who are not attending school based on the emotion recognition results, for example, by providing additional support if their learning progress is falling behind or their motivation is declining.

[1733] Terminal

[1734] Step 5:

[1735] The user (a child not attending school) receives feedback from the emotion engine and receives appropriate learning support based on their own emotional state.

[1736] Through these processing steps, the system of the present invention significantly reduces the workload of teachers and provides optimal educational support tailored to the emotional state of each individual user. Specific operational examples include providing tests with adjusted difficulty to nervous students and providing additional support to motivate students who are not attending school. In this way, the quality and effectiveness of education can be improved.

[1737] Example 2

[1738] 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."

[1739] In today's educational environment, the increasing workload of teachers is a major problem. Repetitive tasks such as lesson preparation and test creation / grading place a particularly heavy burden on teachers. Furthermore, teachers' workloads are reaching their limits as they are required to provide learning support for children who are not attending school and individual attention to each student. A system is needed to effectively resolve these issues and improve the quality of education.

[1740] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting lesson data, means for anonymizing and generalizing the collected lesson data, means for performing machine learning using the anonymized and generalized data, means for automatically generating lesson materials based on lesson information, means for providing the generated lesson materials, means for analyzing user emotions using an emotion recognition function, and means for adjusting the content of lesson materials and tests based on the analyzed emotion data. This reduces the workload of teachers and enables personalized educational support for each student, including children who are not attending school.

[1741] "Class Data" is a general term for information including the content of classes conducted at educational institutions, teaching materials, test questions, and feedback.

[1742] "Anonymization" refers to the process of removing personal information from data and converting it into a state in which specific individuals cannot be identified.

[1743] "Generalization" refers to the process of converting data into a general-purpose format so that it is not tied to a specific situation.

[1744] "Machine learning" refers to a technology in which a computer automatically learns patterns using data and makes predictions and classifications.

[1745] "Class information" is a general term for information such as themes, target grades, and class times that teachers use to plan their lessons.

[1746] "Automatic generation" refers to the process of automatically creating data and information using AI or algorithms.

[1747] "Providing" refers to making the generated data or information available to users.

[1748] "Emotion recognition function" refers to technology that analyzes a user's emotional state using facial recognition technology and voice analysis.

[1749] "Analyzing" refers to the process of analyzing collected data and extracting meaningful information.

[1750] "Adjust" refers to the process of changing the content of data or information in response to specific conditions or circumstances.

[1751] This invention is an AI teaching system designed to reduce the workload of teachers and support the learning of children who are not attending school. This system utilizes lesson data from across the country, and AI automates and supports lesson preparation, test creation, and grading. In addition, by combining it with an emotion engine that recognizes the user's emotions, it provides educational support that is more tailored to each individual user.

[1752] Basic system configuration

[1753] The system consists of the following main components:

[1754] 1. Server

[1755] The server collects lesson data from educational institutions, anonymizes and generalizes the collected data, for example, using Python data processing libraries (Pandas, NumPy).

[1756] The server uses anonymized and generalized data to train AI models using machine learning libraries (TensorFlow, PyTorch).

[1757] Automatically generate teaching materials based on lesson information and provide them to students. For example, AI can be used to generate lesson slides (PDF format) and audio materials.

[1758] The emotion recognition function analyzes the user's emotions and adjusts the content of class materials and tests based on the analysis. This process uses facial recognition technology (OpenCV) and voice analysis (Librosa).

[1759] 2. Terminal

[1760] The terminal provides an interface for teachers to input lesson information, for example, using a web application (HTML, CSS, JavaScript).

[1761] Receive and download generated course materials and tests.

[1762] The teacher enters the test results and sends them to the server.

[1763] Children who are not attending school and their parents can submit requests for learning support.

[1764] It receives feedback from the emotion engine and sends the user's state to the server.

[1765] 3. Users

[1766] The main users are teachers and children who are not attending school (or their parents).

[1767] Teachers enter lesson information, download generated teaching materials and tests, and enter test results.

[1768] Children who are not attending school and their parents can submit requests for learning support and study using the provided teaching materials.

[1769] While receiving feedback from the emotion engine, appropriate learning support is provided based on the emotional state.

[1770] Specific examples

[1771] 1. Automatic generation of teaching materials

[1772] The user (teacher) enters lesson information for "World War II, 3rd year junior high school students, 50 minutes" through the terminal.

[1773] This information is sent to a server, which uses an AI model to search for relevant lesson data and automatically generate optimal lesson materials, including lesson slides (PDF format) and audio materials.

[1774] Teachers can download the generated teaching materials and use them in their classes.

[1775] 2. Automatic test generation and scoring

[1776] A teacher uses a terminal to send a request to create a test with the following scope: "World War II," difficulty level: "medium," and format: "multiple choice and essay questions."

[1777] This information is sent to a server, which uses an AI model to automatically generate tests.

[1778] The teacher distributes the generated test to the students and later enters the test results into the terminal.

[1779] The device sends the test results to a server, which then automatically grades them. Multiple-choice questions are graded immediately, while essay questions are graded using NLP technology.

[1780] The results of the grading will be provided to the instructor.

[1781] 3. Learning support for children who are not attending school

[1782] A child who is not attending school or their parents can send a request via their device for "Subject: Mathematics, Range: Quadratic Equations, Grade: 2nd Year of Junior High School."

[1783] The server searches for relevant lesson data based on the request and uses an AI model to automatically generate learning materials, including practice questions and instructional videos.

[1784] Parents can download the generated teaching materials, and their children who are not attending school can study at home.

[1785] The server monitors learning progress, assesses comprehension, and provides additional learning materials and study advice.

[1786] 4. Emotion engine integration

[1787] When a teacher is teaching a history lesson, the device uses the device's camera and microphone to recognize the emotional state of the teacher and students.

[1788] The recognized emotion data is sent to the server, which then generates optimal teaching materials for the lesson and provides them to the terminal.

[1789] When administering tests, adjustments are made based on emotional data as appropriate, such as providing less difficult questions to students who are feeling nervous or stressed.

[1790] As described above, the AI ​​teaching system of the present invention reduces the workload of teachers and provides individualized educational support to each student.

[1791] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1792] Step 1:

[1793] The server collects lesson data from each educational institution. Specifically, the server accesses the educational institution's database via API to obtain data such as lesson content, teaching materials, test questions, and feedback. The input data is educational data in JSON format, and the output data is raw data that is temporarily stored in the server's internal storage.

[1794] Step 2:

[1795] The server anonymizes the collected lesson data. Specifically, it uses Python data processing libraries (Pandas, NumPy) to run algorithms to remove personal information. For example, it converts student names into meaningless codes and replaces IDs with random numbers. The input data is raw lesson data, and the output data is anonymized lesson data.

[1796] Step 3:

[1797] The server generalizes the anonymized data. The server processes the data so that it is not dependent on a specific educational institution or situation. For example, it converts specific school names or place names into general expressions to increase the versatility of the data. The input data is anonymized lesson data, and the output data is generalized lesson data.

[1798] Step 4:

[1799] The server trains an AI model using the generalized data. Specifically, the server uses machine learning libraries (TensorFlow, PyTorch) to generate an AI model based on the dataset. For example, it uses math or history lesson data to train a model that predicts appropriate lesson content. The input data is the generalized lesson data, and the output data is the trained AI model.

[1800] Step 5:

[1801] The user (teacher) inputs lesson information into the terminal. For example, the teacher inputs lesson topic, target grade, lesson time, etc. into a web form. The input data is the lesson information, and the output data is a lesson request sent to the server.

[1802] Step 6:

[1803] The terminal sends the input lesson information to the server. The server receives information such as lesson theme, target grade, and lesson time, and searches the database for relevant lesson data based on that information. The input data is the lesson request, and the output data is the relevant lesson data.

[1804] Step 7:

[1805] The server uses an AI model to automatically generate optimal lesson materials based on the input lesson information. For example, it generates appropriate slides and audio materials based on the history lesson theme. The input data is the relevant lesson data, and the output data is the generated lesson materials (PDF slides and audio files).

[1806] Step 8:

[1807] The server sends the generated lesson materials to the teacher's device. For example, the generated slide data is provided to the teacher via email or cloud storage. The input data is the generated lesson materials, and the output data is the lesson materials in a format that the teacher can access.

[1808] Step 9:

[1809] The teacher sends a test creation request on the terminal. The teacher inputs the scope, difficulty level, and format. For example, the teacher might input "Scope: World War II, Difficulty: Medium, Format: Multiple choice and essay questions." The input data is the test creation request, and the output data is the request information sent to the server.

[1810] Step 10:

[1811] The device sends the input request information to the server. Based on the received information, the server uses an AI model to automatically generate a test. For example, it generates a PDF-format test that includes multiple-choice and essay questions. The input data is the request information, and the output data is the generated test.

[1812] Step 11:

[1813] The server provides the generated test to the teacher's terminal, where the input data is the generated test and the output data is the test materials sent to the teacher's terminal.

[1814] Step 12:

[1815] A teacher inputs test results into a terminal. For example, a teacher grades students' tests and enters the results into a web form. The input data is the student's test results, and the output data is the test result data sent to the server.

[1816] Step 13:

[1817] The device sends the entered test results to a server, which instantly scores multiple-choice questions using an AI model and automatically evaluates essay questions using NLP technology. The input data is the test result data, and the output data is the scoring results.

[1818] Step 14:

[1819] The server provides the grading results to the teacher. For example, it sends a report including the grading results to the teacher by email. The input data is the grading results, and the output data is a grading report to the teacher.

[1820] Step 15:

[1821] A child who is not attending school and their guardian input a request for learning support into the terminal. For example, they input "Subject: Mathematics, Area: Quadratic Equations, Grade: 8th Grade." The input data is the learning support request, and the output data is the request information sent to the server.

[1822] Step 16:

[1823] The device sends the request information to the server, which then generates appropriate learning materials based on the received information, such as a practice problem set or an instructional video. The input data is the request information, and the output data is the generated learning materials.

[1824] Step 17:

[1825] The server provides the generated learning materials to the user. For example, the server provides the generated learning materials to the user in a downloadable format via cloud storage. The input data is the generated learning materials, and the output data is the learning materials in a format that the user can access.

[1826] Step 18:

[1827] The user (a child not attending school) downloads the provided learning materials and studies at home. The input data are the learning materials, and the output data are the user's learning progress.

[1828] Step 19:

[1829] The server periodically monitors the user's learning progress, evaluates their level of understanding, and provides additional learning materials and advice as needed. The input data is learning progress data, and the output data is the level of understanding assessment and additional learning materials.

[1830] Step 20:

[1831] The device and server recognize the user's emotions through an emotion engine. For example, they use the device's camera and microphone to perform facial recognition and voice analysis. The input data is the user's facial expression and voice data, and the output data is the analysis result of the user's emotional state.

[1832] Step 21:

[1833] The server adjusts the content of the lesson materials and tests based on the feedback from the emotion engine, for example, providing easier questions to a nervous user. The input data is the analysis result of the emotional state, and the output data is the adjusted lesson materials and tests.

[1834] (Application example 2)

[1835] 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."

[1836] Conventional customer service in brick-and-mortar stores has had difficulty adequately responding to individual customer emotional states and preferences. Furthermore, privacy protection issues existed when collecting and utilizing customer information. The present invention aims to solve these issues, provide more attentive service to customers, and protect their privacy.

[1837] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1838] In this invention, the server includes means for collecting lesson data, means for anonymizing and generalizing the collected lesson data, means for performing machine learning using the anonymized and generalized data, means for automatically generating lesson materials based on lesson information, means for providing the generated lesson materials, means for collecting customer information, means for anonymizing and generalizing the collected customer information, means for recognizing customer emotions, and means for adjusting response content based on the customer's emotional state. This enables individual responses according to the customer's emotional state and makes it possible to provide high-quality services while protecting the customer's privacy.

[1839] - "Class Data" refers to information related to education, such as class content, teaching materials, and test questions, collected from educational institutions.

[1840] "Classroom teaching materials" are educational materials such as slides, audio files, and handouts used in classes.

[1841] "Anonymization" is the process of removing personal information from collected data so that it is no longer possible to identify a specific individual.

[1842] "Generalization" is the process of transforming anonymized data so that it is not tied to a specific situation.

[1843] "Machine learning" is a technique that uses data to train AI models to perform specific tasks automatically.

[1844] "Customer information" refers to data related to customers at a store, such as customer identification information, purchase history, and behavioral data.

[1845] "Emotion recognition" is a technology that uses facial recognition technology and voice analysis to determine a customer's emotional state from their facial expressions and voice.

[1846] "Adjusting the response" means changing the content and delivery method of the service based on the customer's emotional state.

[1847] "Privacy protection" means properly managing customers' personal information so that it is not made known to third parties.

[1848] An "AI model" is a collection of algorithms that learn from data and perform tasks automatically.

[1849] This invention relates to an AI customer support system for brick-and-mortar stores. This system uses AI to adjust service content based on the individual emotional state of each customer, providing detailed and personalized service to customers.

[1850] Basic system configuration

[1851] The system consists of the following main components:

[1852] 1. Server

[1853] Collect, anonymize, and generalize class data and customer information.

[1854] Train and retrain AI models using machine learning.

[1855] Teaching materials and service content are automatically generated and provided based on lesson information and customer information.

[1856] Integrates an emotion engine that recognizes customer emotions and adjusts service content.

[1857] 2. Terminal

[1858] Provides an interface for store staff to input service information.

[1859] Receive and download generated educational materials and service content.

[1860] The staff enters the service information and sends it to the server.

[1861] Physical store customers improve their service experience through emotion engine feedback.

[1862] 3. Users

[1863] Store staff and customers are expected to be the main users.

[1864] Store staff input service information, download the generated educational materials and service content, and input service information again.

[1865] Customers in physical stores will receive a personalized service experience provided by the system.

[1866] Program processing flow

[1867] Collection of lesson data and customer information

[1868] The server collects learning data and customer information from each brick-and-mortar store, including customer purchase history, behavioral data, and emotional states captured through an emotion engine. This information is acquired using cameras (built into smartphones, smart glasses, and head-mounted displays (HMDs)). The collected data is anonymized and generalized to protect privacy.

[1869] Training an AI model

[1870] The server trains AI models using anonymized and generalized data. Supported software includes machine learning libraries (e.g., scikit-learn, TensorFlow, etc.). The AI ​​models are retrained using the latest data to provide always-updated services.

[1871] Automatic generation of service content

[1872] The user (store staff) inputs service information through a terminal. This includes information on the target customer demographic, desired service content, and target products. The terminal sends the input information to the server, which then uses an AI model to generate optimal service content. For example, it generates an introductory video or promotional information for a specific product. The generated service content is then sent to the store staff's terminal.

[1873] Emotion recognition and response adjustment

[1874] The device and server recognize the customer's emotions through an emotion engine. They analyze the customer's emotional state using facial recognition technology and voice analysis (e.g., DeepFace), and the server adjusts the service content based on the feedback from the emotion engine. For example, a customer who is feeling stressed may be recommended relaxation-related products.

[1875] Specific examples

[1876] When a customer picks up a product in a physical store, the smart glasses capture the action and recognize the customer's emotional state through the emotion engine. In this case, for example, if the smart glasses determine that the customer is interested in a product, they can display detailed descriptions and related promotions on the smart glasses.

[1877] Example prompt sentence:

[1878] "Create a voice guide that explains what products you would suggest to customers when they are stressed and why."

[1879] Example output:

[1880] Recommended product: Relaxation aroma oil

[1881] Reason: Relaxation aroma oils are effective in reducing stress and leaving you feeling refreshed.

[1882] Audio guide example:

[1883] "Hello, it looks like you're feeling stressed. This relaxation aroma oil is effective in reducing stress. Please give it a try."

[1884] In this way, by implementing the AI ​​customer support system of the present invention, it becomes possible to provide individual responses according to the customer's emotional state, thereby achieving both improved customer satisfaction and privacy protection.

[1885] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1886] Step 1:

[1887] The server collects lesson data and customer information. Specifically, it captures customer facial expressions and behavioral data using cameras and sensors in the physical store, and collects information such as shopping history and behavioral patterns. The raw data acquired as input is temporarily stored before being processed.

[1888] Step 2:

[1889] The server anonymizes and generalizes the collected data. Specifically, it removes personally identifiable information and standardizes and aggregates the data. This allows the data to be managed in a unified format and prevents the identification of specific individuals. Raw data is used as input, and anonymized and generalized data is output.

[1890] Step 3:

[1891] The server performs machine learning using the anonymized and generalized data. Specifically, it uses machine learning libraries (e.g., scikit-learn, TensorFlow) to train an AI model based on the data and develops an algorithm to predict customer behavior and sentiment. The anonymized data is used as input, and the trained AI model is obtained as output.

[1892] Step 4:

[1893] The server automatically generates service content based on lesson information and customer information. Specifically, it uses a generative AI model to generate prompts that recommend the most suitable products and services to the customer. The AI ​​model and customer information are used as input, and the generated service content (e.g., a list of recommended products and promotion information) is obtained as output.

[1894] Step 5:

[1895] The terminal receives the generated service content and provides it to the staff at the physical store. Specifically, the generated service content is displayed through an application installed on the terminal. The generated service content is used as input and is available to the staff as output.

[1896] Step 6:

[1897] The device and server recognize the customer's emotions through an emotion engine. Specifically, an emotion analysis library (e.g., DeepFace) is used to analyze the customer's facial expressions and voice to identify their emotional state. Real-time video and audio data is used as input, and the customer's emotional state is obtained as output.

[1898] Step 7:

[1899] The server adjusts the service content based on the feedback from the emotion engine. Specifically, it appropriately modifies the generated service content according to the obtained emotional state, providing personalized support. Using the emotional state and service content as input, the adjusted service content is obtained as output.

[1900] Step 8:

[1901] Users (store customers) receive personalized service experiences provided by the system. Specifically, they receive tailored service content through terminals or store staff. It is expected that users will receive tailored service content as input and achieve high levels of satisfaction as output.

[1902] 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.

[1903] 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.

[1904] 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.

[1905] [Fourth embodiment]

[1906] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1907] 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.

[1908] 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).

[1909] 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.

[1910] 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.

[1911] 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).

[1912] 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.

[1913] 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.

[1914] 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.

[1915] 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.

[1916] 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.

[1917] 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.

[1918] 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."

[1919] This invention is an AI teaching system that aims to reduce the workload of teachers and support the learning of children who are not attending school. This system utilizes lesson data from all over the country and uses AI to automate and support lesson preparation, test creation, and grading.

[1920] Basic system configuration

[1921] The system consists of the following main components:

[1922] 1. Server

[1923] Collect, anonymize, and generalize class data.

[1924] Train and retrain AI models using machine learning.

[1925] Class materials and tests are automatically generated and provided based on class information.

[1926] Receives test result input and performs automatic scoring.

[1927] Generate learning materials based on requests for learning support for children who are not attending school and monitor their progress.

[1928] 2. Terminal

[1929] Provides an interface for teachers to enter lesson information.

[1930] Receive and download generated course materials and tests.

[1931] The teacher enters the test results and sends them to the server.

[1932] Children who are not attending school and their parents can submit requests for learning support.

[1933] 3. Users

[1934] The main users are expected to be teachers and school-refusing children (or their parents).

[1935] Teachers enter lesson information, download generated teaching materials and tests, and enter test results.

[1936] Children who are not attending school and their parents can submit requests for learning support and study using the provided teaching materials.

[1937] Program processing flow

[1938] Collection of lesson data

[1939] The server collects lesson data from each educational institution, including, for example, mathematics lesson content, teaching materials, test questions, and feedback.

[1940] The server anonymizes the collected data and removes any personal information.

[1941] The server generalizes the anonymized data, converting it so that it is not tied to any particular situation.

[1942] Training an AI model

[1943] The server trains the AI ​​model using anonymized and generalized data.

[1944] The server periodically retrains the AI ​​model using the latest lesson data stored in the database.

[1945] Automatic generation of teaching materials

[1946] The user (teacher) inputs information such as the lesson theme, target grade, and lesson time via the terminal.

[1947] The terminal transmits the input information to the server.

[1948] The server automatically generates appropriate lesson materials based on the lesson information entered, such as history lesson slides and audio files.

[1949] The server transmits the generated lesson materials to the teacher's terminal.

[1950] The user (teacher) receives the generated teaching materials and uses them in class.

[1951] Automatic test generation and scoring

[1952] The user (teacher) instructs the creation of a test on a terminal, inputting the scope, difficulty level, and format.

[1953] The terminal transmits the input information to the server.

[1954] The server automatically generates tests and provides them to teachers' terminals in PDF format, for example, generating multiple-choice and essay questions for a science test.

[1955] The user (teacher) distributes the generated test to students.

[1956] The user (teacher) enters the test results into the terminal.

[1957] The device sends the test results to the server, which then automatically scores them.

[1958] The server provides the grading results to the teacher.

[1959] Learning support for children who are not attending school

[1960] Users (children not attending school or their parents) can send requests for learning support through their devices, for example, requesting a review of math.

[1961] The terminal sends a request to the server.

[1962] The server generates appropriate learning materials based on the request, such as review videos and slides.

[1963] The server provides the generated learning materials to the user.

[1964] Users (children not attending school) study at home using the provided teaching materials.

[1965] The server monitors learning progress and provides additional learning materials and advice as needed.

[1966] The AI ​​teaching system of this invention can significantly reduce the workload of teachers and effectively support the learning of students who are not attending school. A specific example of its operation is a process in which a teacher generates PowerPoint presentations for a history lesson, distributes them to students, and then automatically generates and grades tests. Support for students who are not attending school includes providing study materials that can be used at home and tracking their learning progress.

[1967] The above is a specific embodiment for carrying out the present invention.

[1968] The processing flow will be explained below.

[1969] Program processing steps

[1970] Collection and learning of lesson data

[1971] server

[1972] Step 1:

[1973] The server collects lesson data from each educational institution, providing a mechanism for regularly uploading information such as the content of lessons taught by teachers, teaching materials used, test questions, and student feedback.

[1974] Step 2:

[1975] The server will anonymize the collected class data and remove personal information, for example by implementing algorithms that automatically detect and remove personally identifiable information such as names and student ID numbers.

[1976] Step 3:

[1977] The server then performs a generalization process on the anonymized data, specifically standardizing and storing the unique expressions of specific teachers and schools in an abstracted form.

[1978] Step 4:

[1979] The server uses anonymized and generalized data to train AI models, which use natural language processing (NLP) and machine learning (ML) algorithms to extract patterns in lesson content and effective teaching methods.

[1980] Step 5:

[1981] The server periodically uses lesson data stored in the database to retrain the AI ​​model to reflect the latest educational trends and data.

[1982] Automatic generation of teaching materials

[1983] Terminal

[1984] Step 1:

[1985] Users (teachers) use their terminals to input information such as lesson topic, target grade, lesson time, etc. A form is provided that allows users to easily input information through a dedicated interface.

[1986] Step 2:

[1987] The terminal transmits the input lesson information to the server.

[1988] server

[1989] Step 3:

[1990] Based on the lesson information received, the server searches the database for relevant lesson data and automatically generates optimal lesson materials using an AI model. For example, it creates new teaching materials by referencing past lesson slides, videos, audio files, etc. related to the same topic.

[1991] Step 4:

[1992] The server transmits the generated lesson materials to the teacher's terminal.

[1993] Terminal

[1994] Step 5:

[1995] Teachers receive the teaching materials generated on their devices, download them, edit or revise them as necessary, and use them in class.

[1996] Automated test creation and scoring

[1997] User

[1998] Step 1:

[1999] Teachers send a test creation request to the server via their terminal, specifying the scope, difficulty level, and question format (multiple choice, essay, etc.).

[2000] server

[2001] Step 2:

[2002] The server references the database based on the specified conditions and automatically generates optimal test questions from past data, using an AI model to extract questions of appropriate difficulty and content.

[2003] Step 3:

[2004] The server sends the automatically generated test to the teacher's device in PDF or other format.

[2005] Terminal

[2006] Step 4:

[2007] Teachers receive the tests generated on their devices, print them out as needed, and distribute them to students.

[2008] User

[2009] Step 5:

[2010] After students take the test, teachers enter the test results into a terminal.

[2011] server

[2012] Step 6:

[2013] The device sends the entered test results to a server, which then automatically scores them. Multiple-choice questions are instantly graded by AI, while essay questions are evaluated using NLP technology.

[2014] Step 7:

[2015] The server provides the grading results in the form of a report to the teacher, who can then check the results and make corrections as necessary.

[2016] Learning support for children who are not attending school

[2017] User

[2018] Step 1:

[2019] Children who are not attending school and their parents can send requests for learning support via their devices, specifically by entering information such as the subject, scope, and grade level they wish to study.

[2020] Terminal

[2021] Step 2:

[2022] The terminal transmits the request content to the server.

[2023] server

[2024] Step 3:

[2025] Based on the request, the server searches the database for relevant lesson data and automatically generates appropriate learning materials, such as review videos, slides, and practice questions.

[2026] Step 4:

[2027] The server provides the generated learning materials to the user, who then sends them to the user's device and makes them available for download.

[2028] Terminal

[2029] Step 5:

[2030] Users (children not attending school) download the provided teaching materials and use them for studying at home.

[2031] server

[2032] Step 6:

[2033] The server periodically monitors the user's learning progress, assesses their level of understanding, and provides additional learning materials and advice as needed.

[2034] The above are the specific processing steps of the system of the present invention. This system reduces the workload of teachers and effectively provides learning support to children who do not attend school.

[2035] Example 1

[2036] 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."

[2037] In today's educational environment, teachers' workloads are increasing, with much of their time being spent on lesson preparation and test creation / grading. This has resulted in situations where teachers are unable to concentrate on their primary educational activities. Supporting the learning of children who are not attending school is also an issue, with insufficient provision of appropriate teaching materials and monitoring of their learning progress. Effective methods are needed to resolve these issues and improve the quality of education.

[2038] 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.

[2039] In this invention, the server includes means for collecting lesson data, means for anonymizing and generalizing the collected lesson data, means for performing machine learning using the anonymized and generalized data, means for automatically generating lesson materials based on lesson information, means for providing the generated lesson materials, and means for monitoring learning progress and providing additional teaching materials and advice. This makes it possible to automate lesson preparation and test creation / grading, reducing the workload of teachers and providing effective learning support for children who are not attending school.

[2040] "Class data" refers to information including lesson content, teaching materials, test questions, feedback, etc. in educational settings.

[2041] "Anonymization" is the process of removing or obscuring personally identifiable information from collected data.

[2042] "Generalization" refers to standardizing data in a way that is independent of specific conditions or situations.

[2043] "Machine learning" is a technology that uses massive amounts of data to enable computers to automatically learn patterns and perform specific tasks.

[2044] "Classroom materials" are content such as documents, slides, audio files, and videos created to support educational activities.

[2045] "Auto-generation" is the process of using artificial intelligence or algorithms to generate content or data with minimal human intervention.

[2046] "Study progress" is an indicator of how far a student has progressed in their studies.

[2047] "Advice" is advice or guidance provided to students and teachers based on their learning progress.

[2048] "Tests" refer to question sets and exams used to assess students' understanding and learning status.

[2049] "Scoring" is the process of evaluating test responses and assigning a score or grade.

[2050] A "request" is an act by a user requesting a particular service or information.

[2051] MODE FOR CARRYING OUT THE INVENTION

[2052] This invention is an AI teacher system that aims to reduce the workload of teachers and support the learning of students who are not attending school. This system utilizes lesson data from educational institutions across the country, and AI automates lesson preparation, test creation and grading, as well as monitoring and supporting learning progress.

[2053] Basic system configuration

[2054] The system consists of the following main components:

[2055] 1. Server

[2056] Collect, anonymize, and generalize class data.

[2057] Train and retrain AI models using machine learning.

[2058] Class materials and tests are automatically generated and provided based on class information.

[2059] Receives test result input and performs automatic scoring.

[2060] Monitor your progress and provide additional learning materials and advice.

[2061] 2. Terminal

[2062] Provides an interface for teachers to enter lesson information.

[2063] Receive and download generated course materials and tests.

[2064] The teacher enters the test results and sends them to the server.

[2065] Provide an interface for accepting requests for learning support.

[2066] 3. Users

[2067] The main users are expected to be teachers and school-refusing children (or their parents).

[2068] Teachers enter lesson information, download generated teaching materials and tests, and enter test results.

[2069] Children who are not attending school and their parents can submit requests for learning support and study using the provided teaching materials.

[2070] Collection of lesson data

[2071] The server receives lesson data directly from educational institutions across the country, automatically retrieving it from school databases and online platforms, and can use existing APIs and data feeds.

[2072] Data anonymization and generalization

[2073] The server detects any personally identifiable information from the received lesson data and performs anonymization processing, which removes or conceals personal information such as student and teacher names. The anonymized data is generalized and not dependent on a specific region or school, and is standardized to be compatible with other datasets.

[2074] Training an AI model

[2075] The server trains the AI ​​model using anonymized and generalized data. Specifically, it builds a neural network and optimizes the model based on the collected data. This is done using specialized hardware such as high-performance GPUs and TPUs. The AI ​​model is periodically retrained using the latest data stored in the database to improve its accuracy.

[2076] Automatic generation of teaching materials

[2077] The user (teacher) inputs information such as the lesson topic, target grade, and lesson time via the device. This information is sent from the device to the server, and the AI ​​model uses generative AI technology to automatically generate lesson materials. For example, slides, audio files, and video materials for a history lesson can be generated. The generated materials are sent from the server to the device, where the teacher can download them and use them in class.

[2078] Automatic test generation and scoring

[2079] The user (teacher) issues instructions for creating a test via their device. The scope, difficulty level, and question format (multiple choice, essay, etc.) are entered and sent from the device to the server. The server automatically generates the test based on the entered information. The generated test is provided in PDF format or similar, which the teacher downloads and distributes to students. When students enter their test results, the device sends this data to the server, which then automatically grades them. The graded results are provided to the teacher, along with feedback.

[2080] Request learning support

[2081] The user (a child not attending school or a parent) sends a request for learning support via their device. For example, they may request a "math review." The request is sent from the device to the server, which automatically generates learning materials based on the request. For example, it may generate videos or slides for reviewing math. The generated learning materials are provided to the user, who studies them at home. The server monitors the learning progress and provides additional learning materials or advice as needed.

[2082] Examples of prompt statements

[2083] The following prompt sentences are used:

[2084] "I want to generate teaching materials for history classes."

[2085] "I want you to create a math test."

[2086] "Please provide math review materials."

[2087] The AI ​​teacher system of the present invention can significantly reduce the workload of teachers and provide effective learning support for children who are not attending school.

[2088] The flow of the identification process in the first embodiment will be described with reference to FIG.

[2089] Program processing steps

[2090] Step 1: Collect lesson data

[2091] Specific behavior:

[2092] The server receives lesson data directly from educational institutions across the country, automatically retrieving data from school databases and online platforms via APIs.

[2093] Input: Data sent by educational institutions, such as course content, study materials, exam questions, and feedback.

[2094] Output: Raw lesson data collected.

[2095] Step 2: Anonymize and generalize data

[2096] Specific behavior:

[2097] The server then detects any personally identifiable information from the collected data and performs anonymization processing, for example, by using a filtering algorithm to remove personal information such as names and addresses.

[2098] The server generalizes the anonymized data, standardizing it to convert it into a format that is independent of specific schools or regions.

[2099] Input: Raw lesson data collected.

[2100] Output: Anonymized and generalized lesson data.

[2101] Step 3: Training the AI ​​model

[2102] Specific behavior:

[2103] The server uses anonymized and generalized data to train AI models, using neural networks and other machine learning techniques to train algorithms that generate educational content.

[2104] The server periodically retrains the AI ​​model with the latest data, thereby maintaining the model's accuracy and effectiveness.

[2105] Input: Anonymized and generalized lesson data.

[2106] Output: A trained AI model.

[2107] Step 4: Automatic generation of lesson materials

[2108] Specific behavior:

[2109] The user (teacher) uses a terminal to input information such as the lesson theme, target grade, lesson time, etc. This transmits specific lesson needs to the server.

[2110] The terminal transmits the input information to the server.

[2111] The server uses a generative AI model to automatically generate lesson materials based on the input lesson information, such as history lesson slides, audio files, and videos.

[2112] The server transmits the generated lesson materials to the teacher's terminal.

[2113] The user (teacher) checks the generated lesson materials and makes fine adjustments as necessary.

[2114] Input: Information such as lesson topic, target grade, lesson time, etc.

[2115] Output: Generated lesson materials (slides, audio files, videos, etc.).

[2116] Step 5: Auto-generate and grade tests

[2117] Specific behavior:

[2118] The user (teacher) uses a terminal to instruct the creation of a test, inputting the scope, difficulty level, and question format (multiple choice, essay, etc.).

[2119] The terminal transmits the input information to the server.

[2120] The server uses a generative AI model to automatically generate tests based on the input information, such as multiple-choice and essay questions for science, in PDF format.

[2121] The server provides the generated test to the teacher's terminal, who then distributes it to the students.

[2122] After the students take the test, the user (teacher) enters the results into the terminal.

[2123] The device sends the entered test results to the server, which then automatically scores them using a scoring algorithm for fast and accurate evaluation.

[2124] The server provides the grading results to the instructor and generates feedback.

[2125] Input: Test scope, difficulty level, question format. Test results.

[2126] Output: Auto-generated test, marking results and feedback.

[2127] Step 6: Request and provide learning support

[2128] Specific behavior:

[2129] A user (a child not attending school or a parent) uses a device to send a request for learning support, for example, a request for "math review."

[2130] The terminal sends a request to the server.

[2131] The server automatically generates learning materials based on the request using a generative AI model, such as review videos and slides.

[2132] The server provides the generated learning materials to the user.

[2133] Users (children not attending school) study at home using the provided teaching materials.

[2134] The server monitors learning progress and provides additional learning materials and advice as needed.

[2135] Enter: Request for study support.

[2136] Output: Generated learning materials, learning progress data, and further learning materials and advice.

[2137] This specific process will reduce the workload of teachers and provide effective learning support to children who are not attending school.

[2138] (Application example 1)

[2139] 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."

[2140] In the traditional education system, teachers are overwhelmed with the tasks of preparing lessons and creating and grading tests, making it difficult to provide adequate learning support for students who are not attending school. Furthermore, educational methods using virtual environments are limited, making it difficult to provide customized learning materials tailored to individual students' learning progress. This situation hinders the efficient use of educational resources and the improvement of educational quality.

[2141] 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.

[2142] In this invention, the server includes means for collecting lesson data, means for anonymizing and generalizing the collected lesson data, means for performing machine learning using the anonymized and generalized data, means for automatically generating lesson materials based on lesson information, means for providing the generated lesson materials, means for delivering lessons in a virtual environment, and means for tracking the learning progress of children who are not attending school and providing customized teaching materials. This reduces the workload of teachers and makes it possible to provide effective individual learning support to children who are not attending school.

[2143] "Class data" refers to information such as class content, teaching materials, test questions, and feedback collected from educational institutions.

[2144] "Anonymization" means processing data so that personal information is removed and individuals cannot be identified.

[2145] "Generalization" means transforming data so that it can be used widely and is not tied to a specific situation.

[2146] "Machine learning" is a technology that allows computers to automatically learn from data and make future predictions and classifications.

[2147] "Class information" refers to specific information about the implementation of a class, such as the class theme, target grade, and class time.

[2148] "Classroom teaching materials" are materials such as slides, audio files, and videos used in conducting lessons.

[2149] A "virtual environment" is a virtual space provided via the Internet, rather than a physical classroom or location.

[2150] A "school refusal student" is a student who is unable to attend school for an extended period of time for some reason.

[2151] "Learning progress" refers to a student's progress and achievement in learning.

[2152] "Customized learning materials" refer to learning resources that are optimized according to the learning situation and needs of each individual student.

[2153] This invention is an AI teaching system that aims to reduce the workload of teachers and support the learning of children who are not attending school. This system collects lesson data and uses AI to automate and support lesson preparation, test creation, and grading. It also delivers lessons through a virtual environment and provides customized teaching materials to children who are not attending school, thereby supporting their learning.

[2154] Basic system configuration

[2155] The system consists of the following main components:

[2156] 1. Server

[2157] Classroom data is collected, anonymized, and generalized.

[2158] Train and retrain AI models using machine learning.

[2159] Class materials are automatically generated and provided based on class information.

[2160] Deliver lessons in a virtual environment.

[2161] Automatically generate and grade tests.

[2162] Track the learning progress of children who are not attending school and generate and provide customized learning materials.

[2163] 2. Terminal

[2164] Provides an interface for teachers to enter lesson information.

[2165] Receive and download created course materials and tests.

[2166] Classes are delivered virtually and can be viewed live or recorded.

[2167] The teacher enters the test results and sends them to the server.

[2168] Provides an interface for school-refusing children and their parents to request learning support.

[2169] 3. Users

[2170] The main users are teachers, children who are not attending school, and their parents.

[2171] Teachers enter lesson information, download generated teaching materials and tests, and enter test results.

[2172] Children who are not attending school and their parents can enter requests for learning support and study using the provided teaching materials.

[2173] Program processing flow

[2174] Hardware and Software Use

[2175] Hardware: Smartphones and head-mounted displays are used, which allow for real-time streaming of classes and recordings for viewing.

[2176] Software used: Python's aiohttp library is used to perform asynchronous HTTP requests and communicate with the server.

[2177] Data processing and calculation

[2178] 1. Collection and anonymization of lesson data

[2179] The server collects lesson data from multiple educational institutions, anonymizes and generalizes the data, removing personal information and converting it into a common format.

[2180] 2. Training the AI ​​model

[2181] The server uses anonymized and generalized data to train AI models, which will automate the generation of future course materials and test creation.

[2182] 3. Automatic generation of teaching materials and tests

[2183] When teachers input information about the lesson theme and target grade through the terminal, the server automatically generates lesson materials based on that information and sends them back to the terminal.

[2184] Similarly, you can enter the scope and format of the test and the server will automatically generate the test and provide it in PDF format.

[2185] 4. Learning support

[2186] When a child who is not attending school or their parents request learning support, the server generates and provides customized teaching materials and tests based on the child's individual learning progress.

[2187] Examples and prompts

[2188] Examples:

[2189] Lesson theme: "Japanese History"

[2190] Target grade: 2nd year junior high school students

[2191] Lesson duration: 45 minutes

[2192] Example prompt sentence:

[2193] Lesson theme: Japanese history

[2194] Target grade: 2nd year junior high school students

[2195] Lesson duration: 45 minutes

[2196] Generated teaching materials: PowerPoint slides, lesson videos

[2197] The AI ​​teaching system of this invention reduces the workload of teachers and effectively supports the learning of students who are not attending school. In addition, by utilizing a virtual environment, it is possible to deliver real-time and recorded lessons, thereby improving the quality of education.

[2198] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[2199] Step 1:

[2200] The server collects lesson data from educational institutions. The collected lesson data includes lesson content, teaching materials, test questions, and feedback. It receives the lesson data as input, anonymizes and generalizes it to remove specific personal information and convert it into a common format. It generates the anonymized and generalized data as output.

[2201] Step 2:

[2202] The server performs machine learning using the anonymized and generalized data from step 1. Specifically, it trains a generative AI model based on the data and updates the AI ​​model. It receives the anonymized data as input and outputs the trained AI model.

[2203] Step 3:

[2204] The user (teacher) inputs lesson information such as lesson theme, target grade, lesson time, etc. through the terminal. The input lesson information is sent from the terminal to the server.

[2205] Step 4:

[2206] The server automatically generates lesson materials based on the lesson information entered in step 3. It uses a generative AI model to generate appropriate lesson slides and audio files. It receives lesson information as input and outputs lesson materials.

[2207] Step 5:

[2208] The server sends the generated lesson materials to the terminal. The terminal provides the lesson materials received from the server to the teacher. The teacher downloads the materials for use in the class. The terminal receives the lesson materials as input and outputs the materials sent to the terminal.

[2209] Step 6:

[2210] The user (teacher) uses a terminal to give instructions for creating a test. Specifically, they input the scope, difficulty, and format of the test. The input test information is sent from the terminal to the server.

[2211] Step 7:

[2212] The server automatically generates a test based on the test information entered in step 6. It uses a generative AI model to create a test including multiple choice and essay questions and generates it in PDF format. It receives test information as input and outputs the generated test.

[2213] Step 8:

[2214] The user (teacher) distributes the generated test to students and inputs the test results into the terminal, which then sends the input results to the server.

[2215] Step 9:

[2216] The server receives the test results entered in step 8 and automatically grades them using AI. It receives the test results as input and outputs the graded results. The graded results are sent to the terminal and provided to the teacher.

[2217] Step 10:

[2218] The user (a child not attending school or a parent) sends a request for learning support via the device. Specifically, they request the target subjects and learning content. The input request information is sent from the device to the server.

[2219] Step 11:

[2220] The server generates customized learning materials based on the request information entered in step 10. It uses a generative AI model to create review videos and slides. It receives the request information as input and outputs the generated learning materials. The output learning materials are sent to the terminal and provided to the child who is not attending school and their parents.

[2221] Step 12:

[2222] The server monitors the learning progress of students who are not attending school and provides additional learning materials or advice as needed. It is capable of receiving learning progress data as input and outputting appropriate learning support.

[2223] 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.

[2224] This invention is an AI teaching system designed to reduce the workload of teachers and support the learning of children who are not attending school. This system utilizes lesson data from across the country, and AI automates and supports lesson preparation, test creation, and grading. In addition, by combining it with an emotion engine that recognizes the user's emotions, it provides educational support that is more tailored to each individual user.

[2225] Basic system configuration

[2226] The system consists of the following main components:

[2227] 1. Server

[2228] Collect, anonymize, and generalize class data.

[2229] Train and retrain AI models using machine learning.

[2230] Class materials and tests are automatically generated and provided based on class information.

[2231] Receives test result input and performs automatic scoring.

[2232] Generate learning materials based on requests for learning support for children who are not attending school and monitor their progress.

[2233] It integrates an emotion engine that recognizes the user's emotions and adjusts lesson content and tests.

[2234] 2. Terminal

[2235] Provides an interface for teachers to enter lesson information.

[2236] Receive and download generated course materials and tests.

[2237] The teacher enters the test results and sends them to the server.

[2238] Children who are not attending school and their parents can submit requests for learning support.

[2239] It receives feedback from the emotion engine and transmits the user's state to the server through the emotion recognition function.

[2240] 3. Users

[2241] The main users are expected to be teachers and school-refusing children (or their parents).

[2242] Teachers enter lesson information, download generated teaching materials and tests, and enter test results.

[2243] Children who are not attending school and their parents can submit requests for learning support and study using the provided teaching materials.

[2244] While receiving feedback from the emotion engine, the user receives appropriate learning support based on their own emotional state.

[2245] Program processing flow

[2246] Collection of lesson data

[2247] The server collects lesson data from each educational institution, including, for example, mathematics lesson content, teaching materials, test questions, and feedback.

[2248] The server anonymizes the collected data and removes any personal information.

[2249] The server generalizes the anonymized data, converting it so that it is not tied to any particular situation.

[2250] Training an AI model

[2251] The server trains the AI ​​model using anonymized and generalized data.

[2252] The server periodically uses the latest lesson data stored in the database to retrain the AI ​​model, ensuring it reflects the latest educational trends and data.

[2253] Automatic generation of teaching materials

[2254] The user (teacher) inputs information such as the lesson theme, target grade, and lesson time via the terminal.

[2255] The terminal transmits the input information to the server.

[2256] Based on the lesson information entered, the server searches the database for relevant lesson data and uses an AI model to automatically generate optimal lesson materials, such as history lesson slides and audio files.

[2257] The server transmits the generated lesson materials to the teacher's terminal.

[2258] The user (teacher) receives the generated teaching materials and uses them in class.

[2259] Automatic test generation and scoring

[2260] The user (teacher) sends a test creation request to the server from their device, inputting the scope, difficulty level, and format.

[2261] The terminal transmits the input information to the server.

[2262] The server automatically generates tests and provides them to teachers' terminals in PDF format, for example, generating multiple-choice and essay questions for a science test.

[2263] The user (teacher) distributes the generated test to students.

[2264] The user (teacher) enters the test results into the terminal.

[2265] The device sends the test results to a server, which then automatically scores them. Multiple-choice questions are instantly scored by AI, while essay questions are evaluated using NLP technology.

[2266] The server provides the grading results to the teacher.

[2267] Learning support for children who are not attending school

[2268] Users (children not attending school or their parents) send a request for learning support via their device. Specifically, they input information such as the subject, scope, and grade they wish to study.

[2269] The terminal transmits the request content to the server.

[2270] The server searches for relevant lesson data based on the request and automatically generates appropriate learning materials, such as review videos, slides, and practice questions.

[2271] The server provides the generated learning materials to the user, who then sends them to the user's device and makes them available for download.

[2272] Users (children not attending school) download the provided teaching materials and use them for studying at home.

[2273] The server periodically monitors the user's learning progress, assesses their level of understanding, and provides additional learning materials and advice as needed.

[2274] Emotion engine integration

[2275] The terminal and server recognize the user's emotions through an emotion engine, for example, by using facial recognition technology and voice analysis to analyze the user's emotional state.

[2276] The server then adjusts course materials and test content based on feedback from the emotion engine, for example by providing less difficult questions to students who are nervous.

[2277] Similarly, learning support for users (children not attending school) is adjusted based on data from the emotion engine. If learning progress is falling behind or motivation is declining, additional support is provided.

[2278] As a concrete example, when a teacher is teaching a history class, the emotion engine first recognizes the emotional state of the teacher and students.The server then generates the optimal slides or audio materials for the lesson based on the recognized data and provides them to the devices.In addition, when conducting a test, measures are taken to increase motivation by providing relatively easy questions to students who are feeling nervous or stressed.

[2279] The collaboration between the AI ​​teacher system and emotion engine of this invention reduces the workload of teachers and enables more personalized educational support for each student. Learning support for children who are not attending school can also be provided optimally by understanding changes in their emotions, which will greatly contribute to reducing educational disparities.

[2280] The above is a specific embodiment for carrying out the present invention.

[2281] The processing flow will be explained below.

[2282] Program processing steps

[2283] Collection and learning of lesson data

[2284] server

[2285] Step 1:

[2286] The server collects lesson data from each educational institution, including the teaching materials used by teachers, lesson content, test questions, and student feedback.

[2287] Step 2:

[2288] The server anonymizes the collected lesson data and removes personal information, automatically detecting and deleting personally identifiable information such as names and student numbers.

[2289] Step 3:

[2290] The server generalizes the anonymized data by standardizing expressions specific to specific teachers and schools and storing them in a generic format.

[2291] Step 4:

[2292] The server trains the AI ​​model using anonymized and generalized data, and uses NLP and machi...

Claims

1. A means of collecting lesson data; A means of anonymizing and generalizing the collected lesson data; a means for performing machine learning using the anonymized and generalized data; A means for automatically generating lesson materials based on lesson information; A means for providing the generated teaching materials; A system including:

2. A means for automatically generating tests based on lesson information; further comprising means for providing automatically generated tests; The system of claim 1 .

3. a means for inputting test results; a means for scoring the entered test results; further including a means for providing the results of the grading to the teacher; The system of claim 2.

4. A means to request learning support for children who are not attending school, and means for generating learning materials based on the request; means for providing the generated learning materials; a means of monitoring learning progress; further including measures to provide additional learning support as needed; The system of claim 1 .

Images