System

Abstract

A system is provided.SOLUTION: This system includes a means for inputting the contents of a teaching material, a means for generating a AI teacher specialized in a specific teaching material on the basis of the inputted contents of the teaching material, a means for managing communication with a learner and providing an answer, an explanation and a hint through the AI teacher, and a means for instructing the reference place of the teaching material based on the answer and the explanation.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] When learners use commercially available learning materials, if they find it difficult to understand a section, they have to search for other materials or information online, resulting in reduced learning efficiency. In particular, additional resources are required to accommodate the diversity of learning materials and individual levels of understanding, which can increase learning time and reduce consistency of understanding. The present invention aims to solve these problems by providing an AI teacher specialized for specific learning materials, thereby providing an environment in which learners can study efficiently and effectively. [Means for solving the problem]

[0005] The present invention is a system that includes a means for inputting the content of learning materials, a means for generating an AI teacher specialized for specific learning materials based on the content of the input learning materials, a means for managing interactions with learners and providing answers, explanations, and hints via the AI ​​teacher, and a means for indicating reference locations in learning materials based on the answers and explanations. Furthermore, the system includes a means for managing a learning session after the AI ​​teacher is generated, and a means for responding to additional questions and requests from learners based on the answers and explanations generated by the AI ​​teacher, thereby enabling learners to smoothly progress in their studies while relying on specific learning materials.

[0006] "Teaching materials" are educational materials such as books, PDF files, and text files that learners can use in their learning activities.

[0007] "Input means" refers to a function or device that allows a user to upload or provide to the system the learning materials that the user wishes to use for learning.

[0008] An "AI teacher" is an artificial intelligence model that is generated based on the content of input teaching materials and poses questions to learners, providing answers, explanations, and hints.

[0009] A "learning session" is a time frame or process during which a learner engages in learning activities with an AI teacher.

[0010] "Means for managing interactions" refers to functions and devices that appropriately process and monitor interactions such as questions, answers, and hints submitted between learners and AI teachers.

[0011] "Means for providing answers and explanations" refers to the functions and devices that allow the AI ​​teacher to generate and provide appropriate answers and detailed explanations for questions and assignments submitted by learners.

[0012] "Means for providing hints" refers to functions or devices that provide hints or auxiliary information from an AI teacher to assist learners in the process of arriving at a solution when they are working on a problem.

[0013] "Means for indicating reference points" refers to functions or devices that show learners specific pages or sections of the relevant teaching materials when the AI ​​teacher provides answers or explanations.

[0014] "Means for responding to additional questions or requests" refers to functions or devices that allow the AI ​​teacher to provide additional answers or hints in response to a learner's request for further information. [Brief explanation of the drawings]

[0015] [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

[0016] 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.

[0017] First, the terms used in the following description will be explained.

[0018] 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).

[0019] 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.

[0020] 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.

[0021] 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.

[0022] 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."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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."

[0036] MODE FOR CARRYING OUT THE INVENTION

[0037] To specifically implement the present invention, the following system is constructed: The system is made up of user, terminal, and server components, each of which operates in cooperation with each other.

[0038] Teaching material input function

[0039] The user inputs the learning materials they wish to use into the system. Specifically, the user uploads the learning materials (PDF or text file) using the device's upload interface. The device then sends the uploaded learning material file to the server. During this process, the device acquires file information and sends it along with the learning material file itself to the server. The server saves the received file, extracts the text in the case of a PDF, and reads the content as is for text files. This preprocesses the content of the learning materials as text data and converts it into a format that can be input into the generation AI.

[0040] AI teacher generation function

[0041] The server sends the preprocessed text data to a generation AI. The generation AI (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for a specific learning material. Once this AI teacher model is generated, the server stores it in a database and associates it with a learning material ID. This process results in an AI teacher that provides educational support tailored to the content of the learning material.

[0042] Learning Session Management

[0043] When a user starts a learning session, they send a request through the device interface. The request includes the learning material ID of the learning material they want to use. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. This prepares the AI ​​model for the course to interact with the learner.

[0044] Interface Features

[0045] As the user progresses with their learning, they input questions and answers into the interface and press the send button. The device then sends the input information to the server. The server then inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. At this time, the AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints as needed.

[0046] This series of processes allows users to study efficiently and effectively based on specific learning materials. For example, if a user is using calculus learning materials and asks about how to use the differentiation formula, the AI ​​teacher will explain how to apply the formula and point them to the relevant part of the learning material. This allows learners to deepen their understanding through the learning materials and progress effectively.

[0047] As described above, the present invention provides a system that generates an AI teacher specialized in specific learning materials, allowing learners to progress through their studies efficiently.

[0048] The processing flow will be explained below.

[0049] Teaching material input function

[0050] Step 1:

[0051] The user selects the teaching material (PDF or text file) they want to use and uploads the selected teaching material through the upload interface on their device.

[0052] Step 2:

[0053] The device temporarily stores the uploaded learning material file in its storage, and simultaneously acquires file information (file name, file format, size, etc.).

[0054] Step 3:

[0055] The terminal sends the file itself to the server. The transmitted data also includes file information.

[0056] Step 4:

[0057] The server saves the received file in a designated folder. If the teaching material file is in PDF format, the server extracts the text from the PDF. If it is a text file, the server reads the content as is.

[0058] Step 5:

[0059] The server converts the extracted or read text data into a data format for input into the generative AI.

[0060] AI teacher generation function

[0061] Step 1:

[0062] The server sends the preprocessed text data to the API of the generative AI (e.g., a large-scale language model).

[0063] Step 2:

[0064] The generative AI analyzes the text data sent and generates an AI teacher model specialized for that teaching material.

[0065] Step 3:

[0066] The generation AI returns the generated AI teacher model to the server. The returned data includes the model file and related information.

[0067] Step 4:

[0068] The server stores the received AI teacher model in a database and associates it with the corresponding teaching material ID.

[0069] Learning Session Management

[0070] Step 1:

[0071] The user uses the learning session initiation interface of the device to send a request to start a learning session. This request includes the learning material ID of the learning material to be used.

[0072] Step 2:

[0073] The terminal sends a learning start request to the server.

[0074] Step 3:

[0075] The server receives the request and loads the AI ​​teacher model corresponding to the teaching material ID from the database.

[0076] Step 4:

[0077] The server loads the loaded AI teacher model into memory and initializes the learning session.

[0078] Interface Features

[0079] Step 1:

[0080] The user enters a specific question and answer into the terminal interface and presses the send button.

[0081] Step 2:

[0082] The terminal transmits the input text data to the server.

[0083] Step 3:

[0084] The server inputs the received questions and answers into the AI ​​teacher model and performs the process of generating appropriate answers and explanations.

[0085] Step 4:

[0086] The generated answers and explanations may also include references to the teaching material.

[0087] Step 5:

[0088] The server sends the generated answers and explanations to the terminal.

[0089] Step 6:

[0090] The device displays the received answers and explanations on the user's interface, and the user can request additional questions or hints again.

[0091] Step 7:

[0092] If the user wishes to submit an additional question or request, they provide new input to the terminal and resubmit, which again transmits the data to the server and repeats the process described above.

[0093] This allows users to efficiently study based on specific learning materials. Each processing step operates in conjunction with the other steps, providing a smooth learning environment for learners.

[0094] Example 1

[0095] 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."

[0096] Conventional learning support systems have difficulty providing appropriate answers and explanations specific to the content of the learning materials, making it difficult to meet the individual needs of learners. Furthermore, existing systems lack the functionality to directly point to reference locations in the learning materials, hindering efficient learning. Furthermore, the processes involved in generating and storing AI teacher models are cumbersome, often resulting in inconvenience when managing learning sessions. There is a need to address these issues.

[0097] 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.

[0098] In this invention, the server includes: means for inputting the content of the teaching material; means for generating an AI teacher specialized for a specific teaching material based on the content of the input teaching material; means for managing interactions with the learner and providing answers, explanations, and hints via the AI ​​teacher; means for indicating reference locations in the teaching material based on the answers and explanations; means for saving the generated AI teacher model in a database and associating it with a teaching material ID; means for transmitting preprocessed text data to a generation AI; means for the generation AI to generate a teacher model based on the specific teaching material; means for a user to input a question via a terminal and transmit the question to the server; and means for displaying the answers and explanations generated by the AI ​​teacher on the user's interface. This allows for efficient generation and management of an AI teacher model specialized for a teaching material, enabling learners to progress in their studies more efficiently and effectively.

[0099] The "means for inputting the contents of the teaching material" is a terminal interface that allows the user to upload teaching material files (PDF or text files) to the system and transmit the contents to the server.

[0100] "Means for generating AI teachers specialized for specific teaching materials" refers to the process of analyzing preprocessed text data and using generative AI to create an AI teacher model suitable for the content of specific teaching materials.

[0101] "Means for managing interactions with learners" refers to the process by which the server manages questions and answers sent by learners during a learning session and generates and provides appropriate answers from the AI ​​teacher model.

[0102] "Means for providing answers, explanations, and hints" refers to the process of utilizing an AI teacher model to generate answers to learners' questions, explanations about the content of the teaching materials, and additional hints, and provide them to learners.

[0103] "Means for indicating reference points in the teaching materials" refers to the process of pointing learners to specific parts of the teaching materials based on the answers and explanations generated by the AI ​​teacher model.

[0104] "Means for storing the generated AI teacher model in a database and associating it with a teaching material ID" refers to the process of storing the generated AI teacher model in a database and managing it in association with a specific teaching material ID.

[0105] "Means for sending preprocessed text data to the generation AI" refers to the process by which the server passes the preprocessed text data to the generation AI in order to analyze the content of the teaching materials.

[0106] "The means by which the generation AI generates a teacher model based on specific teaching materials" refers to the process by which the generation AI analyzes given text data and generates a teacher model corresponding to the content of specific teaching materials.

[0107] "Means for a user to input a question via a terminal and transmit the question to a server" refers to a process by which a learner inputs a question using a terminal interface and transmits it to a server.

[0108] "Means for displaying the answers and explanations generated by the AI ​​teacher on the user's interface" refers to the process of sending the answers and explanations generated by the AI ​​teacher model from the server to the terminal and displaying them on the user's interface.

[0109] MODE FOR CARRYING OUT THE INVENTION

[0110] This invention is a learning support system that generates an AI teacher model based on specific learning materials, allowing learners to study efficiently. This system is composed of user, terminal, and server components, which operate in cooperation with each other.

[0111] Hardware and Software Overview

[0112] The hardware used includes the devices used by users (PCs, tablets, smartphones, etc.) and a server for storing and processing data. A database management system runs on the server, storing the generated AI teacher model and teaching material data. Specific software used includes a text extraction engine for preprocessing (e.g., Tesseract OCR, PDFMiner) and generative AI (e.g., GPT-3, BERT).

[0113] Teaching material input function

[0114] The user selects and uploads a learning material file (PDF or text file) through the device's upload interface. The device then sends the selected file to the server via an HTTP POST request. The information sent includes the file's metadata (file name, size, format, etc.).

[0115] Preprocessing the files

[0116] The server saves the received learning material file in storage (e.g., Amazon S3, local file system). Next, the server checks the file format, and if it is a PDF, it uses a text extraction engine (e.g., Tesseract OCR or PDFMiner) to extract the text data. If it is a text file, it simply reads the content. This text data is preprocessed and converted into a format suitable for the generative AI model.

[0117] AI teacher generation function

[0118] The server sends the preprocessed text data to the generation AI. The generation AI (for example, GPT-3 or BERT) analyzes this text data and generates an AI teacher model specialized for a specific teaching material. The generated AI teacher model is associated with the teaching material ID and stored in a database. This process prepares an AI teacher that provides educational support tailored to the content of the teaching material.

[0119] Learning Session Management

[0120] When a user starts a learning session, they send a request from the device interface, including the learning material ID of the learning material they want to use. The server receives the request, reads the corresponding AI teacher model from the database, and loads it into memory. This makes the AI ​​teacher model corresponding to the selected learning material ready for interaction with the learner.

[0121] Interface Features

[0122] As the user progresses with their learning, they input questions and answers into the device's interface and press the send button. The device then sends the input information to the server. The server then inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. The AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints.

[0123] Specific examples

[0124] For example, if a user is using a calculus textbook and asks, "I don't know how to use the differential formula," the AI ​​teacher will explain how to apply the formula and point them to the relevant part of the textbook. This allows the learner to deepen their understanding through the textbook and progress effectively.

[0125] Prompt Sentence Examples

[0126] Use this differentiation formula to find the derivative of the following function:

[0127] The above is a specific embodiment for carrying out the present invention. This system enables a user to study efficiently and effectively based on specific learning materials.

[0128] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0129] Step 1: Upload your materials

[0130] The user selects a learning material file (PDF or text file) through the terminal interface and clicks the upload button. The terminal sends the selected file to the server via an HTTP POST request. The input is the learning material file and its metadata (file name, size, format, etc.). The output is the learning material file saved on the server.

[0131] Specific operation: When a user selects a file and presses the upload button, the device collects the file contents and metadata and sends them to the server. The server checks the received data and determines the directory to save the file.

[0132] Step 2: Preprocessing the files

[0133] The server saves the received learning material file in storage. Next, the server checks the file format. The input is the saved learning material file. If it is a PDF file, the server extracts the text data using a text extraction engine (e.g., Tesseract OCR, PDFMiner). If it is a text file, the server reads the content as is. The output is the preprocessed text data.

[0134] Specific operation: After the server saves the received file in storage, it checks the file format and, if it is a PDF, uses PDFMiner to extract the text from each page and combine the entire text data into a single string. Text files are read directly.

[0135] Step 3: Generate an AI teacher model

[0136] The server sends the preprocessed text data to the generation AI. The generation AI (e.g., GPT-3, BERT) analyzes this text data and generates an AI teacher model specialized for specific learning materials. The input is the preprocessed text data. The output is the generated AI teacher model.

[0137] Specific operation: The server sends the preprocessed text data to the generation AI. The generation AI analyzes the text data and generates a training model. The generated model is returned to the server and stored in a database.

[0138] Step 4: Save the AI ​​teacher model

[0139] The server saves the generated AI teacher model in a database and associates it with the learning material ID. The inputs are the AI ​​teacher model and the learning material ID. The output is the AI ​​teacher model saved in the database.

[0140] Specific operation: The server assigns a unique teaching material ID to the generated AI teacher model and saves it in the database. When saving, the model and teaching material ID are associated.

[0141] Step 5: Start your study session

[0142] When a user starts a learning session via their device, they select the ID of the learning material they want to study and send a request. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. The input is the user's request. The output is the AI ​​teacher model loaded into memory.

[0143] Specific operation: When a user sends a request including a teaching material ID, the server receives the request, retrieves the corresponding AI teacher model from the database, and loads it into memory.

[0144] Step 6: Enter and submit your question

[0145] As the user progresses with their learning, they input questions into the device interface and press the send button. The device then sends the questions to the server. The input is the user's question, and the output is the question sent to the server.

[0146] Specific operation: When a user inputs a question and presses the send button, the terminal generates a communication packet for sending the question to the server and sends it to the server.

[0147] Step 7: Parsing the question and generating an answer

[0148] The server inputs the received question into the AI ​​teacher model and generates an appropriate answer or explanation. The inputs are the user's question and the AI ​​teacher model. The output is the generated answer or explanation.

[0149] Specific operation: The server inputs the received question into the AI ​​teacher model, which analyzes the question and generates an appropriate answer or explanation, which is then returned to the server.

[0150] Step 8: View your answers

[0151] The server sends the generated answers and explanations to the terminal and displays them on the user's interface. The input is the generated answers and explanations. The output is the answers and explanations displayed on the user's interface.

[0152] Specific operation: The server generates a communication packet containing the generated answer and explanation, and sends it to the terminal. The terminal analyzes the received data and displays it on the user's interface.

[0153] (Application example 1)

[0154] 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."

[0155] Conventional learning support systems lack the ability to generate AI teachers specialized for learning materials and respond to learners' questions in real time, making efficient and effective learning difficult. Furthermore, even in cram schools and preparatory schools that provide education in brick-and-mortar locations, students have limited means of receiving immediate, detailed feedback during self-study, which hinders their learning progress.

[0156] 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.

[0157] In this invention, the server includes a means for inputting the content of the learning materials, a means for generating an AI teacher specialized for a specific learning material based on the content of the learning materials input, a means for managing interactions with the learner and providing answers, explanations, and hints via the AI ​​teacher, a means for indicating reference locations in the learning materials based on the answers and explanations, and a means for students to upload learning materials using a smart device in a brick-and-mortar store, generate an AI teacher based on the content of the uploaded learning materials, and provide learning support in real time. This enables students to study efficiently and effectively during self-study even in brick-and-mortar educational settings.

[0158] "Means for inputting the contents of the teaching materials" refers to a device or software function that allows users to upload teaching material data (such as PDF or text files) to the system.

[0159] The "means for generating an AI teacher specialized for a specific teaching material" refers to an algorithm or software function for analyzing the content of the uploaded teaching material and generating an artificial intelligence model specialized for that teaching material.

[0160] "Means for managing learner interactions" refers to a device or software function that receives questions or requests from learners, generates answers or explanations based on those questions, and returns them to the learner through the AI ​​teacher model.

[0161] "Means for providing answers, explanations, and hints" refers to devices or software functions that present information generated by the AI ​​teacher model to learners and support their learning.

[0162] "Means for indicating where to refer to teaching materials based on answers and explanations" refers to a device or software function that instructs learners on which specific parts of the teaching materials they should refer to based on information generated by the AI ​​teacher model.

[0163] "Means for students to upload learning materials using smart devices at a brick-and-mortar educational institution" means a device or software feature that allows students to upload learning materials to the system at a brick-and-mortar educational institution using a device such as a smartphone or tablet.

[0164] "Means for generating an AI teacher based on the content and providing learning support in real time" refers to a device or software function that instantly analyzes the content of uploaded teaching materials, generates an AI teacher model based on that, and then provides answers, explanations, and hints to learners in real time.

[0165] To specifically implement this invention, the following system is constructed. The system is mainly composed of a user terminal, a server, and a database. Each component operates in cooperation with the others.

[0166] Teaching material input function

[0167] A user uploads a learning material file (PDF or text file) using a device such as a smartphone or tablet. The learning material file is sent to the server through the device's upload interface. The server saves the received learning material file, extracts the text from PDF files, and reads the content directly from text files. During this process, the content of the learning material is preprocessed as text data and converted into a format that can be input into the generative AI model.

[0168] AI teacher generation function

[0169] The server sends the preprocessed text data to a generative AI model. The generative AI model (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for a specific teaching material. Once this AI teacher model is generated, the server stores it in a database and associates it with a teaching material ID. This creates an AI teacher that provides educational support tailored to the content of the teaching material.

[0170] Learning Session Management

[0171] When a user starts a learning session, they send a request through the device interface. The request includes the learning material ID of the learning material they want to use. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. This prepares the AI ​​model for the course to interact with the learner.

[0172] Interface Features

[0173] As the user progresses with their learning, they input questions and answers into the interface and press the send button. The device then sends the input information to the server. The server then inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. The AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints.

[0174] In-store learning support

[0175] At brick-and-mortar educational institutions (such as cram schools and preparatory schools), students upload study materials using smartphones or tablets, and an AI teacher is generated based on the content. This AI teacher can provide students with answers, explanations, and hints in real time. For example, if a student uploads calculus study materials and asks a question about a differentiation formula, the AI ​​teacher will explain how the formula can be applied and point them to the relevant section of the study material.

[0176] Hardware and software used

[0177] Smartphones and tablets: Used to upload teaching materials and as an interface.

[0178] Server: Storage of teaching material data, text extraction, generation and management of AI teaching models.

[0179] Database: Stores AI teacher models and teaching material data.

[0180] Generative AI models: Large-scale language models (e.g., OpenAI API).

[0181] Prompt Sentence Examples

[0182] Use the following as a prompt to generate an AI teacher based on the content of the teaching material:

[0183] "Generate an AI teacher specialized in the following materials: Starting with the basic formulas of calculus, and explaining examples of the application of various formulas. As a specific problem..."

[0184] Use the following as a prompt based on the student's question:

[0185] "Please tell me the basic formula for differentiation."

[0186] The system described above allows students to study efficiently and effectively during self-study, even in brick-and-mortar educational settings.

[0187] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0188] Step 1:

[0189] A user uploads a learning material file (PDF or text file) using a device such as a smartphone or tablet. At this time, the user specifies the learning material file in the device's upload interface and presses the "Upload" button. The device retrieves the contents of the learning material file and sends it to the server. The input is the learning material file, and the output is the learning material file sent to the server.

[0190] Step 2:

[0191] The server saves the received learning material file. Next, if the received file is in PDF format, the server performs text extraction processing. A PDF parser (e.g., PyPDF2) is used for text extraction, and the extracted text data is converted into a format that can be input into the generative AI model. The input is the learning material file, and the output is text data.

[0192] Step 3:

[0193] The server sends the preprocessed text data to a generative AI model. The generative AI model (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for specific learning materials. This process uses the OpenAI API. The input is text data, and the output is an AI teacher model.

[0194] Step 4:

[0195] The generated AI teacher model is stored in a database by the server and associated with the learning material ID. The server uses this association to call the appropriate AI teacher model in subsequent learning sessions. The input is the AI ​​teacher model and the learning material ID, and the output is storage in the database.

[0196] Step 5:

[0197] When a user starts a learning session, they send a request from the device interface. This request includes the learning material ID of the learning material they want to use. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. The input is the learning material ID, and the output is the AI ​​teacher model loaded into memory.

[0198] Step 6:

[0199] The user uses the interface to input questions and answers and presses the send button. The device sends this information to the server. The input is the user's question and answer, and the output is the transmission to the server.

[0200] Step 7:

[0201] The server inputs the received questions and answers into an AI teacher model to generate appropriate answers and explanations. This process uses a generative AI model. The input is the user's question and answer, and the output is the generated answer and explanation.

[0202] Step 8:

[0203] The AI ​​teacher model can also provide references to teaching materials along with the generated answers and explanations. The generated answers and explanations are sent back to the device and displayed on the user's interface. The input is the generated data from the AI ​​teacher model, and the output is the information displayed on the user's interface.

[0204] Step 9:

[0205] Through this process, the user can ask the AI ​​teacher additional questions or hints. The user then enters the question again through the interface and the process is repeated. The input is a new question, and the output is a new answer or explanation.

[0206] The above processing steps allow users to study efficiently and effectively based on specific learning materials. For example, if a user is using calculus learning materials and asks how to use the differentiation formula, the AI ​​teacher will explain how to apply the formula and point them to the relevant part of the learning material.

[0207] 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.

[0208] MODE FOR CARRYING OUT THE INVENTION

[0209] To specifically implement this invention, the following system is constructed. The system is made up of components: a user, a terminal, a server, and an emotion engine, all of which operate in conjunction with one another.

[0210] Teaching material input function

[0211] The user inputs the learning materials they wish to use into the system. Specifically, the user uploads the learning materials (PDF or text file) using the device's upload interface. The device then sends the uploaded learning material file to the server. During this process, the device acquires file information and sends it along with the learning material file itself to the server. The server saves the received file, extracts the text in the case of a PDF, and reads the content as is for text files. This preprocesses the content of the learning materials as text data and converts it into a format that can be input into the generation AI.

[0212] AI teacher generation function

[0213] The server sends the preprocessed text data to a generation AI. The generation AI (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for a specific learning material. Once this AI teacher model is generated, the server stores it in a database and associates it with a learning material ID. This process results in an AI teacher that provides educational support tailored to the content of the learning material.

[0214] Learning Session Management

[0215] When a user starts a learning session, they send a request from the device's learning session start interface. The request includes the learning material ID of the learning material they want to use. The device then sends a learning start request to the server. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. This prepares the AI ​​model corresponding to the course for interaction with the learner.

[0216] Interface Features

[0217] As the user progresses with their learning, they input questions and answers into the interface and press the send button. The device then sends the input information to the server. The server then inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. At this time, the AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints as needed.

[0218] Incorporating an emotion engine

[0219] To further improve the user's learning experience, we use an emotion engine that analyzes emotions from the user's facial expressions, tone of voice, and input content.

[0220] Emotion data collection and analysis

[0221] The device uses devices such as sensors, microphones, and cameras to collect user emotional data. The data obtained from these devices is sent to the emotion engine in real time. The emotion engine analyzes this data and detects the user's current emotional state (e.g., stress, excitement, poor comprehension, etc.).

[0222] Emotion-based dialogue adjustment

[0223] The server adjusts the AI ​​teacher model's response based on the emotion data received from the emotion engine. For example, if the user is feeling stressed, the AI ​​teacher will provide gentle explanations and hints. If the user indicates difficulty in understanding, the AI ​​teacher will provide more detailed explanations and step-by-step hints.

[0224] Providing emotional feedback

[0225] As users progress through their studies, the emotion engine continuously monitors their emotional data, allowing the AI ​​teacher to provide feedback and suggest breaks at the appropriate times.

[0226] As a concrete example, consider a case where a user is using a calculus textbook. If the user asks about how to use the differential formula, the AI ​​teacher will explain an example of how the formula can be applied and point them to the relevant section of the textbook. At the same time, if the emotion engine detects the user's frustration, the AI ​​teacher will gently encourage the user by saying, "Shall I explain it again?" to help the user understand.

[0227] In this way, the present invention provides a system that allows learners to study more efficiently and comfortably by combining an emotion engine with an AI teacher specialized in specific learning materials.

[0228] The processing flow will be explained below.

[0229] Teaching material input function

[0230] Step 1:

[0231] The user selects the teaching material (PDF or text file) they want to use and uploads the selected teaching material through the upload interface on their device.

[0232] Step 2:

[0233] The device temporarily stores the uploaded teaching material file in storage and obtains file information (file name, file format, size, etc.).

[0234] Step 3:

[0235] The terminal sends the file itself to the server. The transmitted data also includes file information.

[0236] Step 4:

[0237] The server saves the received file in a designated folder. If the teaching material file is in PDF format, the server extracts the text from the PDF. If it is a text file, the server reads the content as is.

[0238] Step 5:

[0239] The server converts the extracted or read text data into a data format for input into the generative AI.

[0240] AI teacher generation function

[0241] Step 1:

[0242] The server sends the preprocessed text data to the API of the generative AI (e.g., a large-scale language model).

[0243] Step 2:

[0244] The generative AI analyzes the text data sent and generates an AI teacher model specialized for that teaching material.

[0245] Step 3:

[0246] The generation AI returns the generated AI teacher model to the server. The returned data includes the model file and related information.

[0247] Step 4:

[0248] The server stores the received AI teacher model in a database and associates it with the corresponding teaching material ID.

[0249] Learning Session Management

[0250] Step 1:

[0251] The user uses the learning session initiation interface of the device to send a request to start a learning session. This request includes the learning material ID of the learning material to be used.

[0252] Step 2:

[0253] The terminal sends a learning start request to the server.

[0254] Step 3:

[0255] The server receives the request and loads the AI ​​teacher model corresponding to the teaching material ID from the database.

[0256] Step 4:

[0257] The server loads the loaded AI teacher model into memory and initializes the learning session.

[0258] Interface Features

[0259] Step 1:

[0260] The user enters a specific question and answer into the terminal interface and presses the send button.

[0261] Step 2:

[0262] The terminal transmits the input text data to the server.

[0263] Step 3:

[0264] The server inputs the received questions and answers into the AI ​​teacher model and performs the process of generating appropriate answers and explanations.

[0265] Step 4:

[0266] The generated answers and explanations may also include references to the teaching material.

[0267] Step 5:

[0268] The server sends the generated answers and explanations to the terminal.

[0269] Step 6:

[0270] The device displays the received answers and explanations on the user's interface, and the user can request additional questions or hints again.

[0271] Step 7:

[0272] If the user wishes to submit an additional question or request, they provide new input to the terminal and resubmit, which again transmits the data to the server and repeats the process described above.

[0273] Incorporating an emotion engine

[0274] Step 1:

[0275] The device uses devices such as a camera and microphone to collect user emotional data, allowing it to obtain information such as facial expression analysis and voice tone analysis.

[0276] Step 2:

[0277] The device transmits the collected emotional data in real time to the emotion engine, which analyzes it and evaluates the user's emotional state.

[0278] Step 3:

[0279] The emotion engine's analysis results are sent to a server, which then uses the data to tailor the AI ​​teacher's response. For example, if the user is feeling stressed, the AI ​​teacher will provide a gentle explanation.

[0280] Step 4:

[0281] The emotion engine continuously monitors the user's emotions and provides feedback as needed, for example suggesting a break if the user is feeling frustrated.

[0282] Step 5:

[0283] The server sends appropriate feedback according to the emotional state to the terminal, which then displays it on the user's interface.

[0284] In this way, the present invention provides a system that allows learners to study more efficiently and comfortably by combining an emotion engine with an AI teacher specialized in specific learning materials. Each processing step works in conjunction with the other to improve the user's learning experience.

[0285] Example 2

[0286] 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."

[0287] Conventional educational systems have the problem of not providing sufficient individualized educational support tailored to each learner. They also lack a mechanism for understanding the learner's emotional state and providing appropriate feedback. This makes it difficult to provide efficient learning support, hindering the improvement of learners' understanding and motivation to learn.

[0288] 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 a user to input the contents of the learning material into the terminal, means for receiving the learning material file from the terminal and extracting text if the file is a PDF, means for sending the preprocessed text data to a generative AI model and generating an AI teacher specialized for the specific learning material, and means for saving the generated AI teacher model in a database. This allows the user to advance their learning using an AI teacher model specialized for the specific learning material, and further allows them to receive appropriate learning support based on emotion data.

[0289] "User" means a person who uses the system to upload learning materials, initiate learning sessions, and enter questions and answers.

[0290] A "terminal" is a device that a user operates to upload learning materials and input questions and answers, and has the functionality to communicate with the server.

[0291] The "server" is a central computing device that receives the teaching material files sent from the terminal, performs text extraction and preprocessing, and runs the generative AI model.

[0292] "Learning material file" refers to the learning material uploaded by the user to the system, and is in PDF or text file format.

[0293] A "generative AI model" is an artificial intelligence model that analyzes preprocessed text data and generates an AI teacher specialized in specific teaching materials.

[0294] An "AI teacher model" is a virtual teacher model generated by a generative AI model that provides educational support to learners based on specific teaching materials.

[0295] A "database" is a repository of information that stores and manages generated AI training models and other necessary data.

[0296] A "learning session" refers to the entire learning activity of a user using the system, including interactions with the AI ​​teacher model.

[0297] The "emotion engine" is a system component that analyzes the user's emotional state and adjusts the response of the AI ​​teacher model based on that.

[0298] "Emotion data" refers to data used to understand the user's emotional state, such as facial expressions and tone of voice.

[0299] "Text extraction" refers to the process of extracting textual information from non-text format educational material files such as PDFs.

[0300] "Loading into memory" means expanding the AI ​​teacher model read from the database into the server's working memory and preparing it for actual operation.

[0301] To specifically implement this invention, the system is made up of the following components: a user, a terminal, a server, and an emotion engine, all of which operate in conjunction with one another. Details of these components are explained below.

[0302] Teaching material input function

[0303] When a user uses the system, they first input the learning materials (PDF or text files) into the system via their terminal. The terminal provides an upload interface, allowing the user to select and upload the learning materials. In this process, the terminal obtains information about the learning material file (for example, file name and format) and sends this information and the learning material file itself to the server. The server saves the received file, and if it is in PDF format, it uses OCR technology to extract the text. If it is a text file, it simply reads the content as is. This preprocesses the content of the learning materials as text data and converts it into a form that can be input into the generative AI model.

[0304] AI teacher generation function

[0305] The server sends the preprocessed text data to a generative AI model. The generative AI model can be, for example, a large-scale language model (such as GPT-3). The generative AI model analyzes this text data and generates an AI teacher model specialized for a specific teaching material. The generated AI teacher model is stored in a database by the server and associated with the corresponding teaching material ID. This process prepares an AI teacher that provides educational support based on the content of the specific teaching material.

[0306] Learning Session Management

[0307] When a user starts a learning session, they send a request from the device's learning session start interface. The request includes the learning material ID of the learning material they want to use. The device then sends this request to the server, which then reads the corresponding AI teacher model from the database and loads it into memory. This prepares the AI ​​teacher model for interaction with the learner.

[0308] Interface Features

[0309] As the user progresses through their learning, they input questions and answers into the interface and press the send button. The device sends the input information to the server, which then inputs the received questions and answers into the AI ​​teacher model to generate appropriate answers and explanations. The AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints.

[0310] Incorporating an emotion engine

[0311] To further improve the user's learning experience, an emotion engine is used. The emotion engine analyzes emotions in real time from the user's facial expressions, tone of voice, input content, etc. The device is equipped with the necessary devices such as sensors, microphones, and cameras. The emotion engine analyzes the collected data and detects the user's current emotional state (e.g., stress, excitement, lack of comprehension, etc.). The server adjusts the response of the AI ​​teacher model based on the emotion data received from the emotion engine. For example, if the user is feeling stressed, the AI ​​teacher will provide gentle explanations and hints. If the user has difficulty understanding, it will provide more detailed explanations and step-by-step hints. The emotion engine also continuously monitors the user's emotion data and provides feedback and suggests breaks at appropriate times.

[0312] Specific examples

[0313] As a concrete example, consider a case where a user is using a calculus textbook. If the user asks about how to use the differential formula, the AI ​​teacher will explain an example of how the formula can be applied and point them to the relevant section of the textbook. At the same time, if the emotion engine detects the user's frustration, the AI ​​teacher will gently encourage the user by saying, "Shall I explain it again?" to help the user understand.

[0314] Example prompts to input to the generative AI model

[0315] Below is an example of a prompt sentence to input to the generative AI model.

[0316] "Enter the text data of a calculus textbook below. Based on that data, please generate an AI teacher model specialized for this textbook."

[0317] In this way, by combining an AI teacher model specialized for specific teaching materials with an emotion engine, the system provides an environment in which learners can study more efficiently and comfortably.

[0318] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0319] Step 1:

[0320] The user uploads the learning material file.

[0321] Specific operation: The user uses the upload interface on the terminal to select the teaching material file (PDF or text file) and presses the send button.

[0322] Input: A learning material file selected by the user.

[0323] Output: The teaching material file temporarily saved on the device.

[0324] Step 2:

[0325] The device obtains the uploaded file information.

[0326] Specific operation: The device obtains metadata such as file name, file format, and file size.

[0327] Input: The learning material file selected by the user.

[0328] Output: File information (metadata).

[0329] Step 3:

[0330] The terminal transmits the teaching material file and file information to the server.

[0331] Specific operation: The terminal sends the acquired file information and the main body of the teaching material file to the server.

[0332] Input: Teaching material files and file information.

[0333] Output: The teaching material file and file information sent to the server.

[0334] Step 4:

[0335] The server receives the educational material file, extracts the text if it is a PDF, or reads the content as is if it is a text file.

[0336] Specific operation: The server saves the received teaching material file and branches the processing based on the file format. If it is a PDF file, it uses OCR technology to extract the text, and if it is a text file, it reads the content as is.

[0337] Input: The teaching material file sent to the server.

[0338] Output: Preprocessed text data.

[0339] Step 5:

[0340] The server sends the preprocessed text data to the generation AI.

[0341] Specific operation: The server calls an API to send preprocessed text data to a generative AI model (e.g., GPT-3).

[0342] Input: Preprocessed text data.

[0343] Output: Text data received by the generation AI.

[0344] Step 6:

[0345] The generative AI analyzes the text data and generates an AI teacher model.

[0346] Specific operation: The generative AI model analyzes text data and generates an AI teacher model specialized for specific teaching materials.

[0347] Input: Text data received by the generation AI.

[0348] Output: The generated AI teacher model.

[0349] Step 7:

[0350] The server stores the generated AI teacher model in a database and associates it with the teaching material ID.

[0351] Specific operation: The server stores the generated AI teacher model in a database and associates the model with a specific teaching material ID.

[0352] Input: Generated AI teacher model.

[0353] Output: An AI teacher model and its associated information stored in a database.

[0354] Step 8:

[0355] The user requests to start a learning session.

[0356] Specific operation: The user sends a request including the learning material ID he / she wants to use from the learning session start interface of the terminal.

[0357] Input: A user request to start a learning session.

[0358] Output: The request sent to the terminal.

[0359] Step 9:

[0360] The device sends the request and the teaching material ID to the server.

[0361] Specific operation: The terminal sends a learning session start request and the learning material ID to the server.

[0362] Input: A learning session start request from the user and the learning material ID.

[0363] Output: The request sent to the server and the learning material ID.

[0364] Step 10:

[0365] The server receives the request, reads the AI ​​teacher model from the database, and loads it into memory.

[0366] Specific operation: The server reads the corresponding AI teacher model from the database and loads it into memory.

[0367] Input: The request sent to the server and the learning material ID.

[0368] Output: An AI teacher model loaded into memory.

[0369] Step 11:

[0370] The user enters questions and answers into the interface and submits them.

[0371] Specific operation: The user enters a question and answer into the interface and presses the submit button.

[0372] Input: The user's question and answer.

[0373] Output: Information typed into the terminal.

[0374] Step 12:

[0375] The terminal sends the input information to the server.

[0376] Specific operation: The device sends the entered questions and answers to the server.

[0377] Input: Questions and answers typed into the device.

[0378] Output: The question and answer sent to the server.

[0379] Step 13:

[0380] The server inputs questions and answers into an AI teacher model, which generates answers and explanations.

[0381] Specific operation: The server inputs the received questions and answers into the AI ​​teacher model and receives the generated answers and explanations.

[0382] Input: The question and answer sent to the server.

[0383] Output: Answers and explanations generated by the AI ​​teacher model.

[0384] Step 14:

[0385] The server sends the generated answers and explanations to the terminal.

[0386] Specific operation: The server sends the generated answers and explanations to the terminal, where they are displayed on the user's interface.

[0387] Input: Answers and explanations generated by an AI teacher model.

[0388] Output: Answers and explanations sent to your device.

[0389] Step 15:

[0390] The device uses a device to collect emotion data and transmits the data to a server.

[0391] Specific operation: The device uses sensors, microphones, cameras, etc. to collect emotional data such as the user's facial expressions and tone of voice, and sends it to the server.

[0392] Input: Collected emotion data.

[0393] Output: Emotion data sent to the server.

[0394] Step 16:

[0395] The emotion engine analyzes the data and detects the user's emotional state.

[0396] Specific operation: The emotion engine analyzes the received emotion data and detects the user's current emotional state.

[0397] Input: Emotion data sent to the server.

[0398] Output: Parsed emotional state.

[0399] Step 17:

[0400] The server adjusts the response of the AI ​​teacher model based on the emotional data.

[0401] Specific operation: The server adjusts the response of the AI ​​teacher model based on the emotion data sent from the emotion engine. For example, if the user is feeling stressed, it will provide a gentle commentary.

[0402] Input: Parsed emotional state.

[0403] Output: The adjusted response.

[0404] Step 18:

[0405] The emotion engine continuously monitors data and provides feedback and suggests breaks.

[0406] Specific behavior: The emotion engine continuously monitors the user's emotional state and provides feedback and suggests breaks at appropriate times.

[0407] Input: Continuously collected emotion data.

[0408] Output: Feedback and break suggestions.

[0409] (Application example 2)

[0410] 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."

[0411] Conventional educational support systems do not provide feedback that takes into account the learner's individual emotional state regarding their learning progress, making it difficult to provide effective learning support. Furthermore, particularly in new applications such as virtual stores, there is a lack of methods for analyzing users' purchasing motivation and interest in real time and adjusting the dialogue accordingly.

[0412] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for inputting the content of the learning material, a means for generating an AI teacher specialized for a specific learning material based on the content of the learning material input, a means for managing interactions with the learner and providing answers, explanations, and hints via the AI ​​teacher, a means for indicating reference locations in the learning material based on the answers and explanations, a means for analyzing the user's emotional state and adjusting responses based on the results, and a terminal for collecting emotional data. This provides effective learning support that takes into account the learner's individual emotional state, and enables dialogue adjustment based on the user's emotions even in virtual stores, etc.

[0413] The "means for inputting the contents of the teaching material" is an interface for users to upload the teaching material they wish to use to the system, and is used to send the teaching material data in PDF or text file format to the server.

[0414] The "means of generating an AI teacher specialized in specific teaching materials" refers to analyzing uploaded teaching material data and generating an AI model that provides educational support specialized in the content of the teaching materials.

[0415] "Means for managing interactions with learners and providing answers, explanations, and hints through the AI ​​teacher" refers to an AI teacher providing appropriate answers and explanations to questions and requests from learners and managing the progress of the interactions.

[0416] The "means for indicating the reference portion of the teaching material based on the answer or explanation" is a means for indicating to the learner the relevant portion of the teaching material related to the provided answer or explanation.

[0417] "Means for analyzing the user's emotional state and adjusting responses based on that" refers to technology that analyzes the user's emotions in real time from their facial expressions, tone of voice, etc., and adjusts the response content based on the results.

[0418] A "terminal for collecting emotional data" is a device that collects a user's audio and video data in real time and provides the data necessary for emotional analysis.

[0419] To specifically put this invention into practice, the following system is constructed. This system is made up of the following components: a user, a terminal, a server, and an emotion engine, all of which operate in conjunction with one another.

[0420] Teaching material input function

[0421] Users input the learning materials they want to use into the system. Specifically, they upload learning material files (PDF or text files) using the upload interface on their device. The device then sends the uploaded learning material files to the server. The server saves the received files, extracts the text from PDF files, and reads the content directly from text files. This preprocesses the content of the learning materials as text data and converts it into a format that can be input into the generative AI model.

[0422] AI teacher generation function

[0423] The server sends the preprocessed text data to a generative AI model. The generative AI model (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for a specific teaching material. This AI teacher model is stored in a database and associated with a teaching material ID. This allows an AI teacher to provide educational support tailored to the content of the teaching material.

[0424] Learning Session Management

[0425] When a user wants to start a learning session, they send a request from the device's learning session start interface. The request includes the learning material ID of the learning material they want to use. The device then sends a learning start request to the server. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. This prepares the AI ​​model corresponding to the course for interaction with the learner.

[0426] Interface Features

[0427] As the user progresses with their learning, they input questions and answers into the interface and press the send button. The device then sends the input information to the server. The server then inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. At this time, the AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints as needed.

[0428] Incorporating an emotion engine

[0429] To further improve the user's learning experience, we use an emotion engine that analyzes emotions from the user's facial expressions, tone of voice, and input content.

[0430] Emotion data collection and analysis

[0431] The device uses devices such as sensors, microphones, and cameras to collect user emotional data. The data obtained from these devices is sent to the emotion engine in real time. The emotion engine analyzes this data and detects the user's current emotional state (e.g., stress, excitement, poor comprehension, etc.).

[0432] Emotion-based dialogue adjustment

[0433] The server adjusts the AI ​​teacher model's response based on the emotion data received from the emotion engine. For example, if the user is feeling stressed, the AI ​​teacher will provide gentle explanations and hints. If the user indicates difficulty in understanding, the AI ​​teacher will provide more detailed explanations and step-by-step hints.

[0434] Providing emotional feedback

[0435] As users progress through their studies, the emotion engine continuously monitors their emotional data, allowing the AI ​​teacher to provide feedback and suggest breaks at the appropriate times.

[0436] Examples:

[0437] Consider a case where a user is using a calculus textbook.

[0438] When a user asks how to use a differential formula, the AI ​​teacher will explain an example of how the formula can be applied and point them to the relevant section of the teaching materials.

[0439] Example prompt sentence:

[0440] text

[0441] User Input: What are the features of this refrigerator?

[0442] Emotion: Interest

[0443] The right response: This refrigerator is equipped with the latest cooling technology, is extremely energy efficient, and as a smart refrigerator, can be controlled remotely via a dedicated app.

[0444] Hardware and software used:

[0445] Camera and microphone in smart glasses

[0446] OpenCV (camera image acquisition)

[0447] DeepFace (emotional analysis)

[0448] GPT-3 (Response Generation)

[0449] pyttsx3 (audio output)

[0450] In this way, by combining an AI teacher specialized in learning materials with an emotion engine, we can provide a system that allows learners to learn more efficiently and comfortably. In specific applications such as virtual stores, it will also be possible to adjust dialogue based on the user's emotions, improving the customer experience.

[0451] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0452] Step 1:

[0453] The user uploads a learning material file (PDF or text file) using the upload interface of the terminal. The terminal then sends this file to the server. Specifically, the terminal receives the learning material file uploaded by the user, obtains the file information, and sends it to the server. The input is the learning material file selected by the user, and the output is the learning material file data sent to the server.

[0454] Step 2:

[0455] The server stores the received learning material files. If it is a PDF file, text extraction is performed, and if it is a text file, the content is read as is. The extracted text data is preprocessed and converted into a format that can be input into the generative AI model. The input is the learning material file, and the output is the preprocessed text data.

[0456] Step 3:

[0457] The server sends the preprocessed text data to a generative AI model. The generative AI model (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for a specific learning material. The server stores the generated AI teacher model in a database and associates it with a learning material ID. The input is the preprocessed text data, and the output is the AI ​​teacher model.

[0458] Step 4:

[0459] The user sends a request from the learning session start interface on the device. The request includes the learning material ID of the learning material they want to use. The device then sends a learning start request to the server. The input is the learning start request, and the output is the request data to the server.

[0460] Step 5:

[0461] The server receives a learning start request, reads the corresponding AI teacher model from the database, and loads it into memory. This prepares the AI ​​teacher model for interaction with the learner. The input is the learning start request, and the output is the AI ​​teacher model loaded into memory.

[0462] Step 6:

[0463] The user enters a question or answer into the interface and presses the send button. The terminal sends the entered information to the server. The input is the question or answer entered by the user, and the output is the data sent to the server.

[0464] Step 7:

[0465] The server inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. At this time, the AI ​​teacher model can also indicate references in the teaching materials. The generated answers and explanations are sent back to the device and displayed on the user's interface. The input is the questions and answers, and the output is the generated answers and explanations.

[0466] Step 8:

[0467] The device uses devices such as a camera and microphone to collect the user's emotional data. The data obtained from these devices is sent to the emotion engine in real time. The input is the user's audio and video data, and the output is data on the user's emotional state.

[0468] Step 9:

[0469] The emotion engine analyzes the user's current emotional state from their facial expressions and tone of voice. The analysis results are sent to the server. The input is audio and video data, and the output is analyzed emotional data.

[0470] Step 10:

[0471] The server adjusts the response of the AI ​​teacher model based on the emotional data received from the emotion engine. For example, if the user is feeling stressed, the AI ​​teacher will provide gentle explanations and hints. The input is the emotional data, and the output is the adjusted response.

[0472] Step 11:

[0473] The server continuously monitors the emotion data and provides feedback to the user or suggests taking a break at the appropriate time. The input is continuously collected emotion data, and the output is feedback to the user.

[0474] The above are the processing steps of this system, and the specific operations and data input and output are clearly stated for each step, allowing for a detailed understanding of the specific operation method and effects of the invention.

[0475] 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.

[0476] 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.

[0477] 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.

[0478] [Second embodiment]

[0479] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0480] 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.

[0481] 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).

[0482] 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.

[0483] 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.

[0484] 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).

[0485] 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.

[0486] 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.

[0487] 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.

[0488] 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.

[0489] 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.

[0490] 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."

[0491] MODE FOR CARRYING OUT THE INVENTION

[0492] To specifically implement the present invention, the following system is constructed: The system is made up of user, terminal, and server components, each of which operates in cooperation with each other.

[0493] Teaching material input function

[0494] The user inputs the learning materials they wish to use into the system. Specifically, the user uploads the learning materials (PDF or text file) using the device's upload interface. The device then sends the uploaded learning material file to the server. During this process, the device acquires file information and sends it along with the learning material file itself to the server. The server saves the received file, extracts the text in the case of a PDF, and reads the content as is for text files. This preprocesses the content of the learning materials as text data and converts it into a format that can be input into the generation AI.

[0495] AI teacher generation function

[0496] The server sends the preprocessed text data to a generation AI. The generation AI (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for a specific learning material. Once this AI teacher model is generated, the server stores it in a database and associates it with a learning material ID. This process results in an AI teacher that provides educational support tailored to the content of the learning material.

[0497] Learning Session Management

[0498] When a user starts a learning session, they send a request through the device interface. The request includes the learning material ID of the learning material they want to use. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. This prepares the AI ​​model for the course to interact with the learner.

[0499] Interface Features

[0500] As the user progresses with their learning, they input questions and answers into the interface and press the send button. The device then sends the input information to the server. The server then inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. At this time, the AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints as needed.

[0501] This series of processes allows users to study efficiently and effectively based on specific learning materials. For example, if a user is using calculus learning materials and asks about how to use the differentiation formula, the AI ​​teacher will explain how to apply the formula and point them to the relevant part of the learning material. This allows learners to deepen their understanding through the learning materials and progress effectively.

[0502] As described above, the present invention provides a system that generates an AI teacher specialized in specific learning materials, allowing learners to progress through their studies efficiently.

[0503] The processing flow will be explained below.

[0504] Teaching material input function

[0505] Step 1:

[0506] The user selects the teaching material (PDF or text file) they want to use and uploads the selected teaching material through the upload interface on their device.

[0507] Step 2:

[0508] The device temporarily stores the uploaded learning material file in its storage, and simultaneously acquires file information (file name, file format, size, etc.).

[0509] Step 3:

[0510] The terminal sends the file itself to the server. The transmitted data also includes file information.

[0511] Step 4:

[0512] The server saves the received file in a designated folder. If the teaching material file is in PDF format, the server extracts the text from the PDF. If it is a text file, the server reads the content as is.

[0513] Step 5:

[0514] The server converts the extracted or read text data into a data format for input into the generative AI.

[0515] AI teacher generation function

[0516] Step 1:

[0517] The server sends the preprocessed text data to the API of the generative AI (e.g., a large-scale language model).

[0518] Step 2:

[0519] The generative AI analyzes the text data sent and generates an AI teacher model specialized for that teaching material.

[0520] Step 3:

[0521] The generation AI returns the generated AI teacher model to the server. The returned data includes the model file and related information.

[0522] Step 4:

[0523] The server stores the received AI teacher model in a database and associates it with the corresponding teaching material ID.

[0524] Learning Session Management

[0525] Step 1:

[0526] The user uses the learning session initiation interface of the device to send a request to start a learning session. This request includes the learning material ID of the learning material to be used.

[0527] Step 2:

[0528] The terminal sends a learning start request to the server.

[0529] Step 3:

[0530] The server receives the request and loads the AI ​​teacher model corresponding to the teaching material ID from the database.

[0531] Step 4:

[0532] The server loads the loaded AI teacher model into memory and initializes the learning session.

[0533] Interface Features

[0534] Step 1:

[0535] The user enters a specific question and answer into the terminal interface and presses the send button.

[0536] Step 2:

[0537] The terminal transmits the input text data to the server.

[0538] Step 3:

[0539] The server inputs the received questions and answers into the AI ​​teacher model and performs the process of generating appropriate answers and explanations.

[0540] Step 4:

[0541] The generated answers and explanations may also include references to the teaching material.

[0542] Step 5:

[0543] The server sends the generated answers and explanations to the terminal.

[0544] Step 6:

[0545] The device displays the received answers and explanations on the user's interface, and the user can request additional questions or hints again.

[0546] Step 7:

[0547] If the user wishes to submit an additional question or request, they provide new input to the terminal and resubmit, which again transmits the data to the server and repeats the process described above.

[0548] This allows users to efficiently study based on specific learning materials. Each processing step operates in conjunction with the other steps, providing a smooth learning environment for learners.

[0549] Example 1

[0550] 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."

[0551] Conventional learning support systems have difficulty providing appropriate answers and explanations specific to the content of the learning materials, making it difficult to meet the individual needs of learners. Furthermore, existing systems lack the functionality to directly point to reference locations in the learning materials, hindering efficient learning. Furthermore, the processes involved in generating and storing AI teacher models are cumbersome, often resulting in inconvenience when managing learning sessions. There is a need to address these issues.

[0552] 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.

[0553] In this invention, the server includes: means for inputting the content of the teaching material; means for generating an AI teacher specialized for a specific teaching material based on the content of the input teaching material; means for managing interactions with the learner and providing answers, explanations, and hints via the AI ​​teacher; means for indicating reference locations in the teaching material based on the answers and explanations; means for saving the generated AI teacher model in a database and associating it with a teaching material ID; means for transmitting preprocessed text data to a generation AI; means for the generation AI to generate a teacher model based on the specific teaching material; means for a user to input a question via a terminal and transmit the question to the server; and means for displaying the answers and explanations generated by the AI ​​teacher on the user's interface. This allows for efficient generation and management of an AI teacher model specialized for a teaching material, enabling learners to progress in their studies more efficiently and effectively.

[0554] The "means for inputting the contents of the teaching material" is a terminal interface that allows the user to upload teaching material files (PDF or text files) to the system and transmit the contents to the server.

[0555] "Means for generating AI teachers specialized for specific teaching materials" refers to the process of analyzing preprocessed text data and using generative AI to create an AI teacher model suitable for the content of specific teaching materials.

[0556] "Means for managing interactions with learners" refers to the process by which the server manages questions and answers sent by learners during a learning session and generates and provides appropriate answers from the AI ​​teacher model.

[0557] "Means for providing answers, explanations, and hints" refers to the process of utilizing an AI teacher model to generate answers to learners' questions, explanations about the content of the teaching materials, and additional hints, and provide them to learners.

[0558] "Means for indicating reference points in the teaching materials" refers to the process of pointing learners to specific parts of the teaching materials based on the answers and explanations generated by the AI ​​teacher model.

[0559] "Means for storing the generated AI teacher model in a database and associating it with a teaching material ID" refers to the process of storing the generated AI teacher model in a database and managing it in association with a specific teaching material ID.

[0560] "Means for sending preprocessed text data to the generation AI" refers to the process by which the server passes the preprocessed text data to the generation AI in order to analyze the content of the teaching materials.

[0561] "The means by which the generation AI generates a teacher model based on specific teaching materials" refers to the process by which the generation AI analyzes given text data and generates a teacher model corresponding to the content of specific teaching materials.

[0562] "Means for a user to input a question via a terminal and transmit the question to a server" refers to a process by which a learner inputs a question using a terminal interface and transmits it to a server.

[0563] "Means for displaying the answers and explanations generated by the AI ​​teacher on the user's interface" refers to the process of sending the answers and explanations generated by the AI ​​teacher model from the server to the terminal and displaying them on the user's interface.

[0564] MODE FOR CARRYING OUT THE INVENTION

[0565] This invention is a learning support system that generates an AI teacher model based on specific learning materials, allowing learners to study efficiently. This system is composed of user, terminal, and server components, which operate in cooperation with each other.

[0566] Hardware and Software Overview

[0567] The hardware used includes the devices used by users (PCs, tablets, smartphones, etc.) and a server for storing and processing data. A database management system runs on the server, storing the generated AI teacher model and teaching material data. Specific software used includes a text extraction engine for preprocessing (e.g., Tesseract OCR, PDFMiner) and generative AI (e.g., GPT-3, BERT).

[0568] Teaching material input function

[0569] The user selects and uploads a learning material file (PDF or text file) through the device's upload interface. The device then sends the selected file to the server via an HTTP POST request. The information sent includes the file's metadata (file name, size, format, etc.).

[0570] Preprocessing the files

[0571] The server saves the received learning material file in storage (e.g., Amazon S3, local file system). Next, the server checks the file format, and if it is a PDF, it uses a text extraction engine (e.g., Tesseract OCR or PDFMiner) to extract the text data. If it is a text file, it simply reads the content. This text data is preprocessed and converted into a format suitable for the generative AI model.

[0572] AI teacher generation function

[0573] The server sends the preprocessed text data to the generation AI. The generation AI (for example, GPT-3 or BERT) analyzes this text data and generates an AI teacher model specialized for a specific teaching material. The generated AI teacher model is associated with the teaching material ID and stored in a database. This process prepares an AI teacher that provides educational support tailored to the content of the teaching material.

[0574] Learning Session Management

[0575] When a user starts a learning session, they send a request from the device interface, including the learning material ID of the learning material they want to use. The server receives the request, reads the corresponding AI teacher model from the database, and loads it into memory. This makes the AI ​​teacher model corresponding to the selected learning material ready for interaction with the learner.

[0576] Interface Features

[0577] As the user progresses with their learning, they input questions and answers into the device's interface and press the send button. The device then sends the input information to the server. The server then inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. The AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints.

[0578] Specific examples

[0579] For example, if a user is using a calculus textbook and asks, "I don't know how to use the differential formula," the AI ​​teacher will explain how to apply the formula and point them to the relevant part of the textbook. This allows the learner to deepen their understanding through the textbook and progress effectively.

[0580] Prompt Sentence Examples

[0581] Use this differentiation formula to find the derivative of the following function:

[0582] The above is a specific embodiment for carrying out the present invention. This system enables a user to study efficiently and effectively based on specific learning materials.

[0583] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0584] Step 1: Upload your materials

[0585] The user selects a learning material file (PDF or text file) through the terminal interface and clicks the upload button. The terminal sends the selected file to the server via an HTTP POST request. The input is the learning material file and its metadata (file name, size, format, etc.). The output is the learning material file saved on the server.

[0586] Specific operation: When a user selects a file and presses the upload button, the device collects the file contents and metadata and sends them to the server. The server checks the received data and determines the directory to save the file.

[0587] Step 2: Preprocessing the files

[0588] The server saves the received learning material file in storage. Next, the server checks the file format. The input is the saved learning material file. If it is a PDF file, the server extracts the text data using a text extraction engine (e.g., Tesseract OCR, PDFMiner). If it is a text file, the server reads the content as is. The output is the preprocessed text data.

[0589] Specific operation: After the server saves the received file in storage, it checks the file format and, if it is a PDF, uses PDFMiner to extract the text from each page and combine the entire text data into a single string. Text files are read directly.

[0590] Step 3: Generate an AI teacher model

[0591] The server sends the preprocessed text data to the generation AI. The generation AI (e.g., GPT-3, BERT) analyzes this text data and generates an AI teacher model specialized for specific learning materials. The input is the preprocessed text data. The output is the generated AI teacher model.

[0592] Specific operation: The server sends the preprocessed text data to the generation AI. The generation AI analyzes the text data and generates a training model. The generated model is returned to the server and stored in a database.

[0593] Step 4: Save the AI ​​teacher model

[0594] The server saves the generated AI teacher model in a database and associates it with the learning material ID. The inputs are the AI ​​teacher model and the learning material ID. The output is the AI ​​teacher model saved in the database.

[0595] Specific operation: The server assigns a unique teaching material ID to the generated AI teacher model and saves it in the database. When saving, the model and teaching material ID are associated.

[0596] Step 5: Start your study session

[0597] When a user starts a learning session via their device, they select the ID of the learning material they want to study and send a request. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. The input is the user's request. The output is the AI ​​teacher model loaded into memory.

[0598] Specific operation: When a user sends a request including a teaching material ID, the server receives the request, retrieves the corresponding AI teacher model from the database, and loads it into memory.

[0599] Step 6: Enter and submit your question

[0600] As the user progresses with their learning, they input questions into the device interface and press the send button. The device then sends the questions to the server. The input is the user's question, and the output is the question sent to the server.

[0601] Specific operation: When a user inputs a question and presses the send button, the terminal generates a communication packet for sending the question to the server and sends it to the server.

[0602] Step 7: Parsing the question and generating an answer

[0603] The server inputs the received question into the AI ​​teacher model and generates an appropriate answer or explanation. The inputs are the user's question and the AI ​​teacher model. The output is the generated answer or explanation.

[0604] Specific operation: The server inputs the received question into the AI ​​teacher model, which analyzes the question and generates an appropriate answer or explanation, which is then returned to the server.

[0605] Step 8: View your answers

[0606] The server sends the generated answers and explanations to the terminal and displays them on the user's interface. The input is the generated answers and explanations. The output is the answers and explanations displayed on the user's interface.

[0607] Specific operation: The server generates a communication packet containing the generated answer and explanation, and sends it to the terminal. The terminal analyzes the received data and displays it on the user's interface.

[0608] (Application example 1)

[0609] 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."

[0610] Conventional learning support systems lack the ability to generate AI teachers specialized for learning materials and respond to learners' questions in real time, making efficient and effective learning difficult. Furthermore, even in cram schools and preparatory schools that provide education in brick-and-mortar locations, students have limited means of receiving immediate, detailed feedback during self-study, which hinders their learning progress.

[0611] 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.

[0612] In this invention, the server includes a means for inputting the content of the learning materials, a means for generating an AI teacher specialized for a specific learning material based on the content of the learning materials input, a means for managing interactions with the learner and providing answers, explanations, and hints via the AI ​​teacher, a means for indicating reference locations in the learning materials based on the answers and explanations, and a means for students to upload learning materials using a smart device in a brick-and-mortar store, generate an AI teacher based on the content of the uploaded learning materials, and provide learning support in real time. This enables students to study efficiently and effectively during self-study even in brick-and-mortar educational settings.

[0613] "Means for inputting the contents of the teaching materials" refers to a device or software function that allows users to upload teaching material data (such as PDF or text files) to the system.

[0614] The "means for generating an AI teacher specialized for a specific teaching material" refers to an algorithm or software function for analyzing the content of the uploaded teaching material and generating an artificial intelligence model specialized for that teaching material.

[0615] "Means for managing learner interactions" refers to a device or software function that receives questions or requests from learners, generates answers or explanations based on those questions, and returns them to the learner through the AI ​​teacher model.

[0616] "Means for providing answers, explanations, and hints" refers to devices or software functions that present information generated by the AI ​​teacher model to learners and support their learning.

[0617] "Means for indicating where to refer to teaching materials based on answers and explanations" refers to a device or software function that instructs learners on which specific parts of the teaching materials they should refer to based on information generated by the AI ​​teacher model.

[0618] "Means for students to upload learning materials using smart devices at a brick-and-mortar educational institution" means a device or software feature that allows students to upload learning materials to the system at a brick-and-mortar educational institution using a device such as a smartphone or tablet.

[0619] "Means for generating an AI teacher based on the content and providing learning support in real time" refers to a device or software function that instantly analyzes the content of uploaded teaching materials, generates an AI teacher model based on that, and then provides answers, explanations, and hints to learners in real time.

[0620] To specifically implement this invention, the following system is constructed. The system is mainly composed of a user terminal, a server, and a database. Each component operates in cooperation with the others.

[0621] Teaching material input function

[0622] A user uploads a learning material file (PDF or text file) using a device such as a smartphone or tablet. The learning material file is sent to the server through the device's upload interface. The server saves the received learning material file, extracts the text from PDF files, and reads the content directly from text files. During this process, the content of the learning material is preprocessed as text data and converted into a format that can be input into the generative AI model.

[0623] AI teacher generation function

[0624] The server sends the preprocessed text data to a generative AI model. The generative AI model (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for a specific teaching material. Once this AI teacher model is generated, the server stores it in a database and associates it with a teaching material ID. This creates an AI teacher that provides educational support tailored to the content of the teaching material.

[0625] Learning Session Management

[0626] When a user starts a learning session, they send a request through the device interface. The request includes the learning material ID of the learning material they want to use. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. This prepares the AI ​​model for the course to interact with the learner.

[0627] Interface Features

[0628] As the user progresses with their learning, they input questions and answers into the interface and press the send button. The device then sends the input information to the server. The server then inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. The AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints.

[0629] In-store learning support

[0630] At brick-and-mortar educational institutions (such as cram schools and preparatory schools), students upload study materials using smartphones or tablets, and an AI teacher is generated based on the content. This AI teacher can provide students with answers, explanations, and hints in real time. For example, if a student uploads calculus study materials and asks a question about a differentiation formula, the AI ​​teacher will explain how the formula can be applied and point them to the relevant section of the study material.

[0631] Hardware and software used

[0632] Smartphones and tablets: Used to upload teaching materials and as an interface.

[0633] Server: Storage of teaching material data, text extraction, generation and management of AI teaching models.

[0634] Database: Stores AI teacher models and teaching material data.

[0635] Generative AI models: Large-scale language models (e.g., OpenAI API).

[0636] Prompt Sentence Examples

[0637] Use the following as a prompt to generate an AI teacher based on the content of the teaching material:

[0638] "Generate an AI teacher specialized in the following materials: Starting with the basic formulas of calculus, and explaining examples of the application of various formulas. As a specific problem..."

[0639] Use the following as a prompt based on the student's question:

[0640] "Please tell me the basic formula for differentiation."

[0641] The system described above allows students to study efficiently and effectively during self-study, even in brick-and-mortar educational settings.

[0642] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0643] Step 1:

[0644] A user uploads a learning material file (PDF or text file) using a device such as a smartphone or tablet. At this time, the user specifies the learning material file in the device's upload interface and presses the "Upload" button. The device retrieves the contents of the learning material file and sends it to the server. The input is the learning material file, and the output is the learning material file sent to the server.

[0645] Step 2:

[0646] The server saves the received learning material file. Next, if the received file is in PDF format, the server performs text extraction processing. A PDF parser (e.g., PyPDF2) is used for text extraction, and the extracted text data is converted into a format that can be input into the generative AI model. The input is the learning material file, and the output is text data.

[0647] Step 3:

[0648] The server sends the preprocessed text data to a generative AI model. The generative AI model (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for specific learning materials. This process uses the OpenAI API. The input is text data, and the output is an AI teacher model.

[0649] Step 4:

[0650] The generated AI teacher model is stored in a database by the server and associated with the learning material ID. The server uses this association to call the appropriate AI teacher model in subsequent learning sessions. The input is the AI ​​teacher model and the learning material ID, and the output is storage in the database.

[0651] Step 5:

[0652] When a user starts a learning session, they send a request from the device interface. This request includes the learning material ID of the learning material they want to use. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. The input is the learning material ID, and the output is the AI ​​teacher model loaded into memory.

[0653] Step 6:

[0654] The user uses the interface to input questions and answers and presses the send button. The device sends this information to the server. The input is the user's question and answer, and the output is the transmission to the server.

[0655] Step 7:

[0656] The server inputs the received questions and answers into an AI teacher model to generate appropriate answers and explanations. This process uses a generative AI model. The input is the user's question and answer, and the output is the generated answer and explanation.

[0657] Step 8:

[0658] The AI ​​teacher model can also provide references to teaching materials along with the generated answers and explanations. The generated answers and explanations are sent back to the device and displayed on the user's interface. The input is the generated data from the AI ​​teacher model, and the output is the information displayed on the user's interface.

[0659] Step 9:

[0660] Through this process, the user can ask the AI ​​teacher additional questions or hints. The user then enters the question again through the interface and the process is repeated. The input is a new question, and the output is a new answer or explanation.

[0661] The above processing steps allow users to study efficiently and effectively based on specific learning materials. For example, if a user is using calculus learning materials and asks how to use the differentiation formula, the AI ​​teacher will explain how to apply the formula and point them to the relevant part of the learning material.

[0662] 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.

[0663] MODE FOR CARRYING OUT THE INVENTION

[0664] To specifically implement this invention, the following system is constructed. The system is made up of components: a user, a terminal, a server, and an emotion engine, all of which operate in conjunction with one another.

[0665] Teaching material input function

[0666] The user inputs the learning materials they wish to use into the system. Specifically, the user uploads the learning materials (PDF or text file) using the device's upload interface. The device then sends the uploaded learning material file to the server. During this process, the device acquires file information and sends it along with the learning material file itself to the server. The server saves the received file, extracts the text in the case of a PDF, and reads the content as is for text files. This preprocesses the content of the learning materials as text data and converts it into a format that can be input into the generation AI.

[0667] AI teacher generation function

[0668] The server sends the preprocessed text data to a generation AI. The generation AI (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for a specific learning material. Once this AI teacher model is generated, the server stores it in a database and associates it with a learning material ID. This process results in an AI teacher that provides educational support tailored to the content of the learning material.

[0669] Learning Session Management

[0670] When a user starts a learning session, they send a request from the device's learning session start interface. The request includes the learning material ID of the learning material they want to use. The device then sends a learning start request to the server. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. This prepares the AI ​​model corresponding to the course for interaction with the learner.

[0671] Interface Features

[0672] As the user progresses with their learning, they input questions and answers into the interface and press the send button. The device then sends the input information to the server. The server then inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. At this time, the AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints as needed.

[0673] Incorporating an emotion engine

[0674] To further improve the user's learning experience, we use an emotion engine that analyzes emotions from the user's facial expressions, tone of voice, and input content.

[0675] Emotion data collection and analysis

[0676] The device uses devices such as sensors, microphones, and cameras to collect user emotional data. The data obtained from these devices is sent to the emotion engine in real time. The emotion engine analyzes this data and detects the user's current emotional state (e.g., stress, excitement, poor comprehension, etc.).

[0677] Emotion-based dialogue adjustment

[0678] The server adjusts the AI ​​teacher model's response based on the emotion data received from the emotion engine. For example, if the user is feeling stressed, the AI ​​teacher will provide gentle explanations and hints. If the user indicates difficulty in understanding, the AI ​​teacher will provide more detailed explanations and step-by-step hints.

[0679] Providing emotional feedback

[0680] As users progress through their studies, the emotion engine continuously monitors their emotional data, allowing the AI ​​teacher to provide feedback and suggest breaks at the appropriate times.

[0681] As a concrete example, consider a case where a user is using a calculus textbook. If the user asks about how to use the differential formula, the AI ​​teacher will explain an example of how the formula can be applied and point them to the relevant section of the textbook. At the same time, if the emotion engine detects the user's frustration, the AI ​​teacher will gently encourage the user by saying, "Shall I explain it again?" to help the user understand.

[0682] In this way, the present invention provides a system that allows learners to study more efficiently and comfortably by combining an emotion engine with an AI teacher specialized in specific learning materials.

[0683] The processing flow will be explained below.

[0684] Teaching material input function

[0685] Step 1:

[0686] The user selects the teaching material (PDF or text file) they want to use and uploads the selected teaching material through the upload interface on their device.

[0687] Step 2:

[0688] The device temporarily stores the uploaded teaching material file in storage and obtains file information (file name, file format, size, etc.).

[0689] Step 3:

[0690] The terminal sends the file itself to the server. The transmitted data also includes file information.

[0691] Step 4:

[0692] The server saves the received file in a designated folder. If the teaching material file is in PDF format, the server extracts the text from the PDF. If it is a text file, the server reads the content as is.

[0693] Step 5:

[0694] The server converts the extracted or read text data into a data format for input into the generative AI.

[0695] AI teacher generation function

[0696] Step 1:

[0697] The server sends the preprocessed text data to the API of the generative AI (e.g., a large-scale language model).

[0698] Step 2:

[0699] The generative AI analyzes the text data sent and generates an AI teacher model specialized for that teaching material.

[0700] Step 3:

[0701] The generation AI returns the generated AI teacher model to the server. The returned data includes the model file and related information.

[0702] Step 4:

[0703] The server stores the received AI teacher model in a database and associates it with the corresponding teaching material ID.

[0704] Learning Session Management

[0705] Step 1:

[0706] The user uses the learning session initiation interface of the device to send a request to start a learning session. This request includes the learning material ID of the learning material to be used.

[0707] Step 2:

[0708] The terminal sends a learning start request to the server.

[0709] Step 3:

[0710] The server receives the request and loads the AI ​​teacher model corresponding to the teaching material ID from the database.

[0711] Step 4:

[0712] The server loads the loaded AI teacher model into memory and initializes the learning session.

[0713] Interface Features

[0714] Step 1:

[0715] The user enters a specific question and answer into the terminal interface and presses the send button.

[0716] Step 2:

[0717] The terminal transmits the input text data to the server.

[0718] Step 3:

[0719] The server inputs the received questions and answers into the AI ​​teacher model and performs the process of generating appropriate answers and explanations.

[0720] Step 4:

[0721] The generated answers and explanations may also include references to the teaching material.

[0722] Step 5:

[0723] The server sends the generated answers and explanations to the terminal.

[0724] Step 6:

[0725] The device displays the received answers and explanations on the user's interface, and the user can request additional questions or hints again.

[0726] Step 7:

[0727] If the user wishes to submit an additional question or request, they provide new input to the terminal and resubmit, which again transmits the data to the server and repeats the process described above.

[0728] Incorporating an emotion engine

[0729] Step 1:

[0730] The device uses devices such as a camera and microphone to collect user emotional data, allowing it to obtain information such as facial expression analysis and voice tone analysis.

[0731] Step 2:

[0732] The device transmits the collected emotional data in real time to the emotion engine, which analyzes it and evaluates the user's emotional state.

[0733] Step 3:

[0734] The emotion engine's analysis results are sent to a server, which then uses the data to tailor the AI ​​teacher's response. For example, if the user is feeling stressed, the AI ​​teacher will provide a gentle explanation.

[0735] Step 4:

[0736] The emotion engine continuously monitors the user's emotions and provides feedback as needed, for example suggesting a break if the user is feeling frustrated.

[0737] Step 5:

[0738] The server sends appropriate feedback according to the emotional state to the terminal, which then displays it on the user's interface.

[0739] In this way, the present invention provides a system that allows learners to study more efficiently and comfortably by combining an emotion engine with an AI teacher specialized in specific learning materials. Each processing step works in conjunction with the other to improve the user's learning experience.

[0740] Example 2

[0741] 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."

[0742] Conventional educational systems have the problem of not providing sufficient individualized educational support tailored to each learner. They also lack a mechanism for understanding the learner's emotional state and providing appropriate feedback. This makes it difficult to provide efficient learning support, hindering the improvement of learners' understanding and motivation to learn.

[0743] 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 a user to input the contents of the learning material into the terminal, means for receiving the learning material file from the terminal and extracting text if the file is a PDF, means for sending the preprocessed text data to a generative AI model and generating an AI teacher specialized for the specific learning material, and means for saving the generated AI teacher model in a database. This allows the user to advance their learning using an AI teacher model specialized for the specific learning material, and further allows them to receive appropriate learning support based on emotion data.

[0744] "User" means a person who uses the system to upload learning materials, initiate learning sessions, and enter questions and answers.

[0745] A "terminal" is a device that a user operates to upload learning materials and input questions and answers, and has the functionality to communicate with the server.

[0746] The "server" is a central computing device that receives the teaching material files sent from the terminal, performs text extraction and preprocessing, and runs the generative AI model.

[0747] "Learning material file" refers to the learning material uploaded by the user to the system, and is in PDF or text file format.

[0748] A "generative AI model" is an artificial intelligence model that analyzes preprocessed text data and generates an AI teacher specialized in specific teaching materials.

[0749] An "AI teacher model" is a virtual teacher model generated by a generative AI model that provides educational support to learners based on specific teaching materials.

[0750] A "database" is a repository of information that stores and manages generated AI training models and other necessary data.

[0751] A "learning session" refers to the entire learning activity of a user using the system, including interactions with the AI ​​teacher model.

[0752] The "emotion engine" is a system component that analyzes the user's emotional state and adjusts the response of the AI ​​teacher model based on that.

[0753] "Emotion data" refers to data used to understand the user's emotional state, such as facial expressions and tone of voice.

[0754] "Text extraction" refers to the process of extracting textual information from non-text format educational material files such as PDFs.

[0755] "Loading into memory" means expanding the AI ​​teacher model read from the database into the server's working memory and preparing it for actual operation.

[0756] To specifically implement this invention, the system is made up of the following components: a user, a terminal, a server, and an emotion engine, all of which operate in conjunction with one another. Details of these components are explained below.

[0757] Teaching material input function

[0758] When a user uses the system, they first input the learning materials (PDF or text files) into the system via their terminal. The terminal provides an upload interface, allowing the user to select and upload the learning materials. In this process, the terminal obtains information about the learning material file (for example, file name and format) and sends this information and the learning material file itself to the server. The server saves the received file, and if it is in PDF format, it uses OCR technology to extract the text. If it is a text file, it simply reads the content as is. This preprocesses the content of the learning materials as text data and converts it into a form that can be input into the generative AI model.

[0759] AI teacher generation function

[0760] The server sends the preprocessed text data to a generative AI model. The generative AI model can be, for example, a large-scale language model (such as GPT-3). The generative AI model analyzes this text data and generates an AI teacher model specialized for a specific teaching material. The generated AI teacher model is stored in a database by the server and associated with the corresponding teaching material ID. This process prepares an AI teacher that provides educational support based on the content of the specific teaching material.

[0761] Learning Session Management

[0762] When a user starts a learning session, they send a request from the device's learning session start interface. The request includes the learning material ID of the learning material they want to use. The device then sends this request to the server, which then reads the corresponding AI teacher model from the database and loads it into memory. This prepares the AI ​​teacher model for interaction with the learner.

[0763] Interface Features

[0764] As the user progresses through their learning, they input questions and answers into the interface and press the send button. The device sends the input information to the server, which then inputs the received questions and answers into the AI ​​teacher model to generate appropriate answers and explanations. The AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints.

[0765] Incorporating an emotion engine

[0766] To further improve the user's learning experience, an emotion engine is used. The emotion engine analyzes emotions in real time from the user's facial expressions, tone of voice, input content, etc. The device is equipped with the necessary devices such as sensors, microphones, and cameras. The emotion engine analyzes the collected data and detects the user's current emotional state (e.g., stress, excitement, lack of comprehension, etc.). The server adjusts the response of the AI ​​teacher model based on the emotion data received from the emotion engine. For example, if the user is feeling stressed, the AI ​​teacher will provide gentle explanations and hints. If the user has difficulty understanding, it will provide more detailed explanations and step-by-step hints. The emotion engine also continuously monitors the user's emotion data and provides feedback and suggests breaks at appropriate times.

[0767] Specific examples

[0768] As a concrete example, consider a case where a user is using a calculus textbook. If the user asks about how to use the differential formula, the AI ​​teacher will explain an example of how the formula can be applied and point them to the relevant section of the textbook. At the same time, if the emotion engine detects the user's frustration, the AI ​​teacher will gently encourage the user by saying, "Shall I explain it again?" to help the user understand.

[0769] Example prompts to input to the generative AI model

[0770] Below is an example of a prompt sentence to input to the generative AI model.

[0771] "Enter the text data of a calculus textbook below. Based on that data, please generate an AI teacher model specialized for this textbook."

[0772] In this way, by combining an AI teacher model specialized for specific teaching materials with an emotion engine, the system provides an environment in which learners can study more efficiently and comfortably.

[0773] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0774] Step 1:

[0775] The user uploads the learning material file.

[0776] Specific operation: The user uses the upload interface on the terminal to select the teaching material file (PDF or text file) and presses the send button.

[0777] Input: A learning material file selected by the user.

[0778] Output: The teaching material file temporarily saved on the device.

[0779] Step 2:

[0780] The device obtains the uploaded file information.

[0781] Specific operation: The device obtains metadata such as file name, file format, and file size.

[0782] Input: The learning material file selected by the user.

[0783] Output: File information (metadata).

[0784] Step 3:

[0785] The terminal transmits the teaching material file and file information to the server.

[0786] Specific operation: The terminal sends the acquired file information and the main body of the teaching material file to the server.

[0787] Input: Teaching material files and file information.

[0788] Output: The teaching material file and file information sent to the server.

[0789] Step 4:

[0790] The server receives the educational material file, extracts the text if it is a PDF, or reads the content as is if it is a text file.

[0791] Specific operation: The server saves the received teaching material file and branches the processing based on the file format. If it is a PDF file, it uses OCR technology to extract the text, and if it is a text file, it reads the content as is.

[0792] Input: The teaching material file sent to the server.

[0793] Output: Preprocessed text data.

[0794] Step 5:

[0795] The server sends the preprocessed text data to the generation AI.

[0796] Specific operation: The server calls an API to send preprocessed text data to a generative AI model (e.g., GPT-3).

[0797] Input: Preprocessed text data.

[0798] Output: Text data received by the generation AI.

[0799] Step 6:

[0800] The generative AI analyzes the text data and generates an AI teacher model.

[0801] Specific operation: The generative AI model analyzes text data and generates an AI teacher model specialized for specific teaching materials.

[0802] Input: Text data received by the generation AI.

[0803] Output: The generated AI teacher model.

[0804] Step 7:

[0805] The server stores the generated AI teacher model in a database and associates it with the teaching material ID.

[0806] Specific operation: The server stores the generated AI teacher model in a database and associates the model with a specific teaching material ID.

[0807] Input: Generated AI teacher model.

[0808] Output: An AI teacher model and its associated information stored in a database.

[0809] Step 8:

[0810] The user requests to start a learning session.

[0811] Specific operation: The user sends a request including the learning material ID he / she wants to use from the learning session start interface of the terminal.

[0812] Input: A user request to start a learning session.

[0813] Output: The request sent to the terminal.

[0814] Step 9:

[0815] The device sends the request and the teaching material ID to the server.

[0816] Specific operation: The terminal sends a learning session start request and the learning material ID to the server.

[0817] Input: A learning session start request from the user and the learning material ID.

[0818] Output: The request sent to the server and the learning material ID.

[0819] Step 10:

[0820] The server receives the request, reads the AI ​​teacher model from the database, and loads it into memory.

[0821] Specific operation: The server reads the corresponding AI teacher model from the database and loads it into memory.

[0822] Input: The request sent to the server and the learning material ID.

[0823] Output: An AI teacher model loaded into memory.

[0824] Step 11:

[0825] The user enters questions and answers into the interface and submits them.

[0826] Specific operation: The user enters a question and answer into the interface and presses the submit button.

[0827] Input: The user's question and answer.

[0828] Output: Information typed into the terminal.

[0829] Step 12:

[0830] The terminal sends the input information to the server.

[0831] Specific operation: The device sends the entered questions and answers to the server.

[0832] Input: Questions and answers typed into the device.

[0833] Output: The question and answer sent to the server.

[0834] Step 13:

[0835] The server inputs questions and answers into an AI teacher model, which generates answers and explanations.

[0836] Specific operation: The server inputs the received questions and answers into the AI ​​teacher model and receives the generated answers and explanations.

[0837] Input: The question and answer sent to the server.

[0838] Output: Answers and explanations generated by the AI ​​teacher model.

[0839] Step 14:

[0840] The server sends the generated answers and explanations to the terminal.

[0841] Specific operation: The server sends the generated answers and explanations to the terminal, where they are displayed on the user's interface.

[0842] Input: Answers and explanations generated by an AI teacher model.

[0843] Output: Answers and explanations sent to your device.

[0844] Step 15:

[0845] The device uses a device to collect emotion data and transmits the data to a server.

[0846] Specific operation: The device uses sensors, microphones, cameras, etc. to collect emotional data such as the user's facial expressions and tone of voice, and sends it to the server.

[0847] Input: Collected emotion data.

[0848] Output: Emotion data sent to the server.

[0849] Step 16:

[0850] The emotion engine analyzes the data and detects the user's emotional state.

[0851] Specific operation: The emotion engine analyzes the received emotion data and detects the user's current emotional state.

[0852] Input: Emotion data sent to the server.

[0853] Output: Parsed emotional state.

[0854] Step 17:

[0855] The server adjusts the response of the AI ​​teacher model based on the emotional data.

[0856] Specific operation: The server adjusts the response of the AI ​​teacher model based on the emotion data sent from the emotion engine. For example, if the user is feeling stressed, it will provide a gentle commentary.

[0857] Input: Parsed emotional state.

[0858] Output: The adjusted response.

[0859] Step 18:

[0860] The emotion engine continuously monitors data and provides feedback and suggests breaks.

[0861] Specific behavior: The emotion engine continuously monitors the user's emotional state and provides feedback and suggests breaks at appropriate times.

[0862] Input: Continuously collected emotion data.

[0863] Output: Feedback and break suggestions.

[0864] (Application example 2)

[0865] 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."

[0866] Conventional educational support systems do not provide feedback that takes into account the learner's individual emotional state regarding their learning progress, making it difficult to provide effective learning support. Furthermore, particularly in new applications such as virtual stores, there is a lack of methods for analyzing users' purchasing motivation and interest in real time and adjusting the dialogue accordingly.

[0867] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for inputting the content of the learning material, a means for generating an AI teacher specialized for a specific learning material based on the content of the learning material input, a means for managing interactions with the learner and providing answers, explanations, and hints via the AI ​​teacher, a means for indicating reference locations in the learning material based on the answers and explanations, a means for analyzing the user's emotional state and adjusting responses based on the results, and a terminal for collecting emotional data. This provides effective learning support that takes into account the learner's individual emotional state, and enables dialogue adjustment based on the user's emotions even in virtual stores, etc.

[0868] The "means for inputting the contents of the teaching material" is an interface for users to upload the teaching material they wish to use to the system, and is used to send the teaching material data in PDF or text file format to the server.

[0869] The "means of generating an AI teacher specialized in specific teaching materials" refers to analyzing uploaded teaching material data and generating an AI model that provides educational support specialized in the content of the teaching materials.

[0870] "Means for managing interactions with learners and providing answers, explanations, and hints through the AI ​​teacher" refers to an AI teacher providing appropriate answers and explanations to questions and requests from learners and managing the progress of the interactions.

[0871] The "means for indicating the reference portion of the teaching material based on the answer or explanation" is a means for indicating to the learner the relevant portion of the teaching material related to the provided answer or explanation.

[0872] "Means for analyzing the user's emotional state and adjusting responses based on that" refers to technology that analyzes the user's emotions in real time from their facial expressions, tone of voice, etc., and adjusts the response content based on the results.

[0873] A "terminal for collecting emotional data" is a device that collects a user's audio and video data in real time and provides the data necessary for emotional analysis.

[0874] To specifically put this invention into practice, the following system is constructed. This system is made up of the following components: a user, a terminal, a server, and an emotion engine, all of which operate in conjunction with one another.

[0875] Teaching material input function

[0876] Users input the learning materials they want to use into the system. Specifically, they upload learning material files (PDF or text files) using the upload interface on their device. The device then sends the uploaded learning material files to the server. The server saves the received files, extracts the text from PDF files, and reads the content directly from text files. This preprocesses the content of the learning materials as text data and converts it into a format that can be input into the generative AI model.

[0877] AI teacher generation function

[0878] The server sends the preprocessed text data to a generative AI model. The generative AI model (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for a specific teaching material. This AI teacher model is stored in a database and associated with a teaching material ID. This allows an AI teacher to provide educational support tailored to the content of the teaching material.

[0879] Learning Session Management

[0880] When a user wants to start a learning session, they send a request from the device's learning session start interface. The request includes the learning material ID of the learning material they want to use. The device then sends a learning start request to the server. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. This prepares the AI ​​model corresponding to the course for interaction with the learner.

[0881] Interface Features

[0882] As the user progresses with their learning, they input questions and answers into the interface and press the send button. The device then sends the input information to the server. The server then inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. At this time, the AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints as needed.

[0883] Incorporating an emotion engine

[0884] To further improve the user's learning experience, we use an emotion engine that analyzes emotions from the user's facial expressions, tone of voice, and input content.

[0885] Emotion data collection and analysis

[0886] The device uses devices such as sensors, microphones, and cameras to collect user emotional data. The data obtained from these devices is sent to the emotion engine in real time. The emotion engine analyzes this data and detects the user's current emotional state (e.g., stress, excitement, poor comprehension, etc.).

[0887] Emotion-based dialogue adjustment

[0888] The server adjusts the AI ​​teacher model's response based on the emotion data received from the emotion engine. For example, if the user is feeling stressed, the AI ​​teacher will provide gentle explanations and hints. If the user indicates difficulty in understanding, the AI ​​teacher will provide more detailed explanations and step-by-step hints.

[0889] Providing emotional feedback

[0890] As users progress through their studies, the emotion engine continuously monitors their emotional data, allowing the AI ​​teacher to provide feedback and suggest breaks at the appropriate times.

[0891] Examples:

[0892] Consider a case where a user is using a calculus textbook.

[0893] When a user asks how to use a differential formula, the AI ​​teacher will explain an example of how the formula can be applied and point them to the relevant section of the teaching materials.

[0894] Example prompt sentence:

[0895] text

[0896] User Input: What are the features of this refrigerator?

[0897] Emotion: Interest

[0898] The right response: This refrigerator is equipped with the latest cooling technology, is extremely energy efficient, and as a smart refrigerator, can be controlled remotely via a dedicated app.

[0899] Hardware and software used:

[0900] Camera and microphone in smart glasses

[0901] OpenCV (camera image acquisition)

[0902] DeepFace (emotional analysis)

[0903] GPT-3 (Response Generation)

[0904] pyttsx3 (audio output)

[0905] In this way, by combining an AI teacher specialized in learning materials with an emotion engine, we can provide a system that allows learners to learn more efficiently and comfortably. In specific applications such as virtual stores, it will also be possible to adjust dialogue based on the user's emotions, improving the customer experience.

[0906] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0907] Step 1:

[0908] The user uploads a learning material file (PDF or text file) using the upload interface of the terminal. The terminal then sends this file to the server. Specifically, the terminal receives the learning material file uploaded by the user, obtains the file information, and sends it to the server. The input is the learning material file selected by the user, and the output is the learning material file data sent to the server.

[0909] Step 2:

[0910] The server stores the received learning material files. If it is a PDF file, text extraction is performed, and if it is a text file, the content is read as is. The extracted text data is preprocessed and converted into a format that can be input into the generative AI model. The input is the learning material file, and the output is the preprocessed text data.

[0911] Step 3:

[0912] The server sends the preprocessed text data to a generative AI model. The generative AI model (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for a specific learning material. The server stores the generated AI teacher model in a database and associates it with a learning material ID. The input is the preprocessed text data, and the output is the AI ​​teacher model.

[0913] Step 4:

[0914] The user sends a request from the learning session start interface on the device. The request includes the learning material ID of the learning material they want to use. The device then sends a learning start request to the server. The input is the learning start request, and the output is the request data to the server.

[0915] Step 5:

[0916] The server receives a learning start request, reads the corresponding AI teacher model from the database, and loads it into memory. This prepares the AI ​​teacher model for interaction with the learner. The input is the learning start request, and the output is the AI ​​teacher model loaded into memory.

[0917] Step 6:

[0918] The user enters a question or answer into the interface and presses the send button. The terminal sends the entered information to the server. The input is the question or answer entered by the user, and the output is the data sent to the server.

[0919] Step 7:

[0920] The server inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. At this time, the AI ​​teacher model can also indicate references in the teaching materials. The generated answers and explanations are sent back to the device and displayed on the user's interface. The input is the questions and answers, and the output is the generated answers and explanations.

[0921] Step 8:

[0922] The device uses devices such as a camera and microphone to collect the user's emotional data. The data obtained from these devices is sent to the emotion engine in real time. The input is the user's audio and video data, and the output is data on the user's emotional state.

[0923] Step 9:

[0924] The emotion engine analyzes the user's current emotional state from their facial expressions and tone of voice. The analysis results are sent to the server. The input is audio and video data, and the output is analyzed emotional data.

[0925] Step 10:

[0926] The server adjusts the response of the AI ​​teacher model based on the emotional data received from the emotion engine. For example, if the user is feeling stressed, the AI ​​teacher will provide gentle explanations and hints. The input is the emotional data, and the output is the adjusted response.

[0927] Step 11:

[0928] The server continuously monitors the emotion data and provides feedback to the user or suggests taking a break at the appropriate time. The input is continuously collected emotion data, and the output is feedback to the user.

[0929] The above are the processing steps of this system, and the specific operations and data input and output are clearly stated for each step, allowing for a detailed understanding of the specific operation method and effects of the invention.

[0930] 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.

[0931] 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.

[0932] 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.

[0933] [Third embodiment]

[0934] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0935] 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.

[0936] 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).

[0937] 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.

[0938] 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.

[0939] 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).

[0940] 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.

[0941] 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.

[0942] 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.

[0943] 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.

[0944] 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.

[0945] 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."

[0946] MODE FOR CARRYING OUT THE INVENTION

[0947] To specifically implement the present invention, the following system is constructed: The system is made up of user, terminal, and server components, each of which operates in cooperation with each other.

[0948] Teaching material input function

[0949] The user inputs the learning materials they wish to use into the system. Specifically, the user uploads the learning materials (PDF or text file) using the device's upload interface. The device then sends the uploaded learning material file to the server. During this process, the device acquires file information and sends it along with the learning material file itself to the server. The server saves the received file, extracts the text in the case of a PDF, and reads the content as is for text files. This preprocesses the content of the learning materials as text data and converts it into a format that can be input into the generation AI.

[0950] AI teacher generation function

[0951] The server sends the preprocessed text data to a generation AI. The generation AI (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for a specific learning material. Once this AI teacher model is generated, the server stores it in a database and associates it with a learning material ID. This process results in an AI teacher that provides educational support tailored to the content of the learning material.

[0952] Learning Session Management

[0953] When a user starts a learning session, they send a request through the device interface. The request includes the learning material ID of the learning material they want to use. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. This prepares the AI ​​model for the course to interact with the learner.

[0954] Interface Features

[0955] As the user progresses with their learning, they input questions and answers into the interface and press the send button. The device then sends the input information to the server. The server then inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. At this time, the AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints as needed.

[0956] This series of processes allows users to study efficiently and effectively based on specific learning materials. For example, if a user is using calculus learning materials and asks about how to use the differentiation formula, the AI ​​teacher will explain how to apply the formula and point them to the relevant part of the learning material. This allows learners to deepen their understanding through the learning materials and progress effectively.

[0957] As described above, the present invention provides a system that generates an AI teacher specialized in specific learning materials, allowing learners to progress through their studies efficiently.

[0958] The processing flow will be explained below.

[0959] Teaching material input function

[0960] Step 1:

[0961] The user selects the teaching material (PDF or text file) they want to use and uploads the selected teaching material through the upload interface on their device.

[0962] Step 2:

[0963] The device temporarily stores the uploaded learning material file in its storage, and simultaneously acquires file information (file name, file format, size, etc.).

[0964] Step 3:

[0965] The terminal sends the file itself to the server. The transmitted data also includes file information.

[0966] Step 4:

[0967] The server saves the received file in a designated folder. If the teaching material file is in PDF format, the server extracts the text from the PDF. If it is a text file, the server reads the content as is.

[0968] Step 5:

[0969] The server converts the extracted or read text data into a data format for input into the generative AI.

[0970] AI teacher generation function

[0971] Step 1:

[0972] The server sends the preprocessed text data to the API of the generative AI (e.g., a large-scale language model).

[0973] Step 2:

[0974] The generative AI analyzes the text data sent and generates an AI teacher model specialized for that teaching material.

[0975] Step 3:

[0976] The generation AI returns the generated AI teacher model to the server. The returned data includes the model file and related information.

[0977] Step 4:

[0978] The server stores the received AI teacher model in a database and associates it with the corresponding teaching material ID.

[0979] Learning Session Management

[0980] Step 1:

[0981] The user uses the learning session initiation interface of the device to send a request to start a learning session. This request includes the learning material ID of the learning material to be used.

[0982] Step 2:

[0983] The terminal sends a learning start request to the server.

[0984] Step 3:

[0985] The server receives the request and loads the AI ​​teacher model corresponding to the teaching material ID from the database.

[0986] Step 4:

[0987] The server loads the loaded AI teacher model into memory and initializes the learning session.

[0988] Interface Features

[0989] Step 1:

[0990] The user enters a specific question and answer into the terminal interface and presses the send button.

[0991] Step 2:

[0992] The terminal transmits the input text data to the server.

[0993] Step 3:

[0994] The server inputs the received questions and answers into the AI ​​teacher model and performs the process of generating appropriate answers and explanations.

[0995] Step 4:

[0996] The generated answers and explanations may also include references to the teaching material.

[0997] Step 5:

[0998] The server sends the generated answers and explanations to the terminal.

[0999] Step 6:

[1000] The device displays the received answers and explanations on the user's interface, and the user can request additional questions or hints again.

[1001] Step 7:

[1002] If the user wishes to submit an additional question or request, they provide new input to the terminal and resubmit, which again transmits the data to the server and repeats the process described above.

[1003] This allows users to efficiently study based on specific learning materials. Each processing step operates in conjunction with the other steps, providing a smooth learning environment for learners.

[1004] Example 1

[1005] 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."

[1006] Conventional learning support systems have difficulty providing appropriate answers and explanations specific to the content of the learning materials, making it difficult to meet the individual needs of learners. Furthermore, existing systems lack the functionality to directly point to reference locations in the learning materials, hindering efficient learning. Furthermore, the processes involved in generating and storing AI teacher models are cumbersome, often resulting in inconvenience when managing learning sessions. There is a need to address these issues.

[1007] 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.

[1008] In this invention, the server includes: means for inputting the content of the teaching material; means for generating an AI teacher specialized for a specific teaching material based on the content of the input teaching material; means for managing interactions with the learner and providing answers, explanations, and hints via the AI ​​teacher; means for indicating reference locations in the teaching material based on the answers and explanations; means for saving the generated AI teacher model in a database and associating it with a teaching material ID; means for transmitting preprocessed text data to a generation AI; means for the generation AI to generate a teacher model based on the specific teaching material; means for a user to input a question via a terminal and transmit the question to the server; and means for displaying the answers and explanations generated by the AI ​​teacher on the user's interface. This allows for efficient generation and management of an AI teacher model specialized for a teaching material, enabling learners to progress in their studies more efficiently and effectively.

[1009] The "means for inputting the contents of the teaching material" is a terminal interface that allows the user to upload teaching material files (PDF or text files) to the system and transmit the contents to the server.

[1010] "Means for generating AI teachers specialized for specific teaching materials" refers to the process of analyzing preprocessed text data and using generative AI to create an AI teacher model suitable for the content of specific teaching materials.

[1011] "Means for managing interactions with learners" refers to the process by which the server manages questions and answers sent by learners during a learning session and generates and provides appropriate answers from the AI ​​teacher model.

[1012] "Means for providing answers, explanations, and hints" refers to the process of utilizing an AI teacher model to generate answers to learners' questions, explanations about the content of the teaching materials, and additional hints, and provide them to learners.

[1013] "Means for indicating reference points in the teaching materials" refers to the process of pointing learners to specific parts of the teaching materials based on the answers and explanations generated by the AI ​​teacher model.

[1014] "Means for storing the generated AI teacher model in a database and associating it with a teaching material ID" refers to the process of storing the generated AI teacher model in a database and managing it in association with a specific teaching material ID.

[1015] "Means for sending preprocessed text data to the generation AI" refers to the process by which the server passes the preprocessed text data to the generation AI in order to analyze the content of the teaching materials.

[1016] "The means by which the generation AI generates a teacher model based on specific teaching materials" refers to the process by which the generation AI analyzes given text data and generates a teacher model corresponding to the content of specific teaching materials.

[1017] "Means for a user to input a question via a terminal and transmit the question to a server" refers to a process by which a learner inputs a question using a terminal interface and transmits it to a server.

[1018] "Means for displaying the answers and explanations generated by the AI ​​teacher on the user's interface" refers to the process of sending the answers and explanations generated by the AI ​​teacher model from the server to the terminal and displaying them on the user's interface.

[1019] MODE FOR CARRYING OUT THE INVENTION

[1020] This invention is a learning support system that generates an AI teacher model based on specific learning materials, allowing learners to study efficiently. This system is composed of user, terminal, and server components, which operate in cooperation with each other.

[1021] Hardware and Software Overview

[1022] The hardware used includes the devices used by users (PCs, tablets, smartphones, etc.) and a server for storing and processing data. A database management system runs on the server, storing the generated AI teacher model and teaching material data. Specific software used includes a text extraction engine for preprocessing (e.g., Tesseract OCR, PDFMiner) and generative AI (e.g., GPT-3, BERT).

[1023] Teaching material input function

[1024] The user selects and uploads a learning material file (PDF or text file) through the device's upload interface. The device then sends the selected file to the server via an HTTP POST request. The information sent includes the file's metadata (file name, size, format, etc.).

[1025] Preprocessing the files

[1026] The server saves the received learning material file in storage (e.g., Amazon S3, local file system). Next, the server checks the file format, and if it is a PDF, it uses a text extraction engine (e.g., Tesseract OCR or PDFMiner) to extract the text data. If it is a text file, it simply reads the content. This text data is preprocessed and converted into a format suitable for the generative AI model.

[1027] AI teacher generation function

[1028] The server sends the preprocessed text data to the generation AI. The generation AI (for example, GPT-3 or BERT) analyzes this text data and generates an AI teacher model specialized for a specific teaching material. The generated AI teacher model is associated with the teaching material ID and stored in a database. This process prepares an AI teacher that provides educational support tailored to the content of the teaching material.

[1029] Learning Session Management

[1030] When a user starts a learning session, they send a request from the device interface, including the learning material ID of the learning material they want to use. The server receives the request, reads the corresponding AI teacher model from the database, and loads it into memory. This makes the AI ​​teacher model corresponding to the selected learning material ready for interaction with the learner.

[1031] Interface Features

[1032] As the user progresses with their learning, they input questions and answers into the device's interface and press the send button. The device then sends the input information to the server. The server then inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. The AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints.

[1033] Specific examples

[1034] For example, if a user is using a calculus textbook and asks, "I don't know how to use the differential formula," the AI ​​teacher will explain how to apply the formula and point them to the relevant part of the textbook. This allows the learner to deepen their understanding through the textbook and progress effectively.

[1035] Prompt Sentence Examples

[1036] Use this differentiation formula to find the derivative of the following function:

[1037] The above is a specific embodiment for carrying out the present invention. This system enables a user to study efficiently and effectively based on specific learning materials.

[1038] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1039] Step 1: Upload your materials

[1040] The user selects a learning material file (PDF or text file) through the terminal interface and clicks the upload button. The terminal sends the selected file to the server via an HTTP POST request. The input is the learning material file and its metadata (file name, size, format, etc.). The output is the learning material file saved on the server.

[1041] Specific operation: When a user selects a file and presses the upload button, the device collects the file contents and metadata and sends them to the server. The server checks the received data and determines the directory to save the file.

[1042] Step 2: Preprocessing the files

[1043] The server saves the received learning material file in storage. Next, the server checks the file format. The input is the saved learning material file. If it is a PDF file, the server extracts the text data using a text extraction engine (e.g., Tesseract OCR, PDFMiner). If it is a text file, the server reads the content as is. The output is the preprocessed text data.

[1044] Specific operation: After the server saves the received file in storage, it checks the file format and, if it is a PDF, uses PDFMiner to extract the text from each page and combine the entire text data into a single string. Text files are read directly.

[1045] Step 3: Generate an AI teacher model

[1046] The server sends the preprocessed text data to the generation AI. The generation AI (e.g., GPT-3, BERT) analyzes this text data and generates an AI teacher model specialized for specific learning materials. The input is the preprocessed text data. The output is the generated AI teacher model.

[1047] Specific operation: The server sends the preprocessed text data to the generation AI. The generation AI analyzes the text data and generates a training model. The generated model is returned to the server and stored in a database.

[1048] Step 4: Save the AI ​​teacher model

[1049] The server saves the generated AI teacher model in a database and associates it with the learning material ID. The inputs are the AI ​​teacher model and the learning material ID. The output is the AI ​​teacher model saved in the database.

[1050] Specific operation: The server assigns a unique teaching material ID to the generated AI teacher model and saves it in the database. When saving, the model and teaching material ID are associated.

[1051] Step 5: Start your study session

[1052] When a user starts a learning session via their device, they select the ID of the learning material they want to study and send a request. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. The input is the user's request. The output is the AI ​​teacher model loaded into memory.

[1053] Specific operation: When a user sends a request including a teaching material ID, the server receives the request, retrieves the corresponding AI teacher model from the database, and loads it into memory.

[1054] Step 6: Enter and submit your question

[1055] As the user progresses with their learning, they input questions into the device interface and press the send button. The device then sends the questions to the server. The input is the user's question, and the output is the question sent to the server.

[1056] Specific operation: When a user inputs a question and presses the send button, the terminal generates a communication packet for sending the question to the server and sends it to the server.

[1057] Step 7: Parsing the question and generating an answer

[1058] The server inputs the received question into the AI ​​teacher model and generates an appropriate answer or explanation. The inputs are the user's question and the AI ​​teacher model. The output is the generated answer or explanation.

[1059] Specific operation: The server inputs the received question into the AI ​​teacher model, which analyzes the question and generates an appropriate answer or explanation, which is then returned to the server.

[1060] Step 8: View your answers

[1061] The server sends the generated answers and explanations to the terminal and displays them on the user's interface. The input is the generated answers and explanations. The output is the answers and explanations displayed on the user's interface.

[1062] Specific operation: The server generates a communication packet containing the generated answer and explanation, and sends it to the terminal. The terminal analyzes the received data and displays it on the user's interface.

[1063] (Application example 1)

[1064] 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."

[1065] Conventional learning support systems lack the ability to generate AI teachers specialized for learning materials and respond to learners' questions in real time, making efficient and effective learning difficult. Furthermore, even in cram schools and preparatory schools that provide education in brick-and-mortar locations, students have limited means of receiving immediate, detailed feedback during self-study, which hinders their learning progress.

[1066] 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.

[1067] In this invention, the server includes a means for inputting the content of the learning materials, a means for generating an AI teacher specialized for a specific learning material based on the content of the learning materials input, a means for managing interactions with the learner and providing answers, explanations, and hints via the AI ​​teacher, a means for indicating reference locations in the learning materials based on the answers and explanations, and a means for students to upload learning materials using a smart device in a brick-and-mortar store, generate an AI teacher based on the content of the uploaded learning materials, and provide learning support in real time. This enables students to study efficiently and effectively during self-study even in brick-and-mortar educational settings.

[1068] "Means for inputting the contents of the teaching materials" refers to a device or software function that allows users to upload teaching material data (such as PDF or text files) to the system.

[1069] The "means for generating an AI teacher specialized for a specific teaching material" refers to an algorithm or software function for analyzing the content of the uploaded teaching material and generating an artificial intelligence model specialized for that teaching material.

[1070] "Means for managing learner interactions" refers to a device or software function that receives questions or requests from learners, generates answers or explanations based on those questions, and returns them to the learner through the AI ​​teacher model.

[1071] "Means for providing answers, explanations, and hints" refers to devices or software functions that present information generated by the AI ​​teacher model to learners and support their learning.

[1072] "Means for indicating where to refer to teaching materials based on answers and explanations" refers to a device or software function that instructs learners on which specific parts of the teaching materials they should refer to based on information generated by the AI ​​teacher model.

[1073] "Means for students to upload learning materials using smart devices at a brick-and-mortar educational institution" means a device or software feature that allows students to upload learning materials to the system at a brick-and-mortar educational institution using a device such as a smartphone or tablet.

[1074] "Means for generating an AI teacher based on the content and providing learning support in real time" refers to a device or software function that instantly analyzes the content of uploaded teaching materials, generates an AI teacher model based on that, and then provides answers, explanations, and hints to learners in real time.

[1075] To specifically implement this invention, the following system is constructed. The system is mainly composed of a user terminal, a server, and a database. Each component operates in cooperation with the others.

[1076] Teaching material input function

[1077] A user uploads a learning material file (PDF or text file) using a device such as a smartphone or tablet. The learning material file is sent to the server through the device's upload interface. The server saves the received learning material file, extracts the text from PDF files, and reads the content directly from text files. During this process, the content of the learning material is preprocessed as text data and converted into a format that can be input into the generative AI model.

[1078] AI teacher generation function

[1079] The server sends the preprocessed text data to a generative AI model. The generative AI model (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for a specific teaching material. Once this AI teacher model is generated, the server stores it in a database and associates it with a teaching material ID. This creates an AI teacher that provides educational support tailored to the content of the teaching material.

[1080] Learning Session Management

[1081] When a user starts a learning session, they send a request through the device interface. The request includes the learning material ID of the learning material they want to use. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. This prepares the AI ​​model for the course to interact with the learner.

[1082] Interface Features

[1083] As the user progresses with their learning, they input questions and answers into the interface and press the send button. The device then sends the input information to the server. The server then inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. The AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints.

[1084] In-store learning support

[1085] At brick-and-mortar educational institutions (such as cram schools and preparatory schools), students upload study materials using smartphones or tablets, and an AI teacher is generated based on the content. This AI teacher can provide students with answers, explanations, and hints in real time. For example, if a student uploads calculus study materials and asks a question about a differentiation formula, the AI ​​teacher will explain how the formula can be applied and point them to the relevant section of the study material.

[1086] Hardware and software used

[1087] Smartphones and tablets: Used to upload teaching materials and as an interface.

[1088] Server: Storage of teaching material data, text extraction, generation and management of AI teaching models.

[1089] Database: Stores AI teacher models and teaching material data.

[1090] Generative AI models: Large-scale language models (e.g., OpenAI API).

[1091] Prompt Sentence Examples

[1092] Use the following as a prompt to generate an AI teacher based on the content of the teaching material:

[1093] "Generate an AI teacher specialized in the following materials: Starting with the basic formulas of calculus, and explaining examples of the application of various formulas. As a specific problem..."

[1094] Use the following as a prompt based on the student's question:

[1095] "Please tell me the basic formula for differentiation."

[1096] The system described above allows students to study efficiently and effectively during self-study, even in brick-and-mortar educational settings.

[1097] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1098] Step 1:

[1099] A user uploads a learning material file (PDF or text file) using a device such as a smartphone or tablet. At this time, the user specifies the learning material file in the device's upload interface and presses the "Upload" button. The device retrieves the contents of the learning material file and sends it to the server. The input is the learning material file, and the output is the learning material file sent to the server.

[1100] Step 2:

[1101] The server saves the received learning material file. Next, if the received file is in PDF format, the server performs text extraction processing. A PDF parser (e.g., PyPDF2) is used for text extraction, and the extracted text data is converted into a format that can be input into the generative AI model. The input is the learning material file, and the output is text data.

[1102] Step 3:

[1103] The server sends the preprocessed text data to a generative AI model. The generative AI model (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for specific learning materials. This process uses the OpenAI API. The input is text data, and the output is an AI teacher model.

[1104] Step 4:

[1105] The generated AI teacher model is stored in a database by the server and associated with the learning material ID. The server uses this association to call the appropriate AI teacher model in subsequent learning sessions. The input is the AI ​​teacher model and the learning material ID, and the output is storage in the database.

[1106] Step 5:

[1107] When a user starts a learning session, they send a request from the device interface. This request includes the learning material ID of the learning material they want to use. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. The input is the learning material ID, and the output is the AI ​​teacher model loaded into memory.

[1108] Step 6:

[1109] The user uses the interface to input questions and answers and presses the send button. The device sends this information to the server. The input is the user's question and answer, and the output is the transmission to the server.

[1110] Step 7:

[1111] The server inputs the received questions and answers into an AI teacher model to generate appropriate answers and explanations. This process uses a generative AI model. The input is the user's question and answer, and the output is the generated answer and explanation.

[1112] Step 8:

[1113] The AI ​​teacher model can also provide references to teaching materials along with the generated answers and explanations. The generated answers and explanations are sent back to the device and displayed on the user's interface. The input is the generated data from the AI ​​teacher model, and the output is the information displayed on the user's interface.

[1114] Step 9:

[1115] Through this process, the user can ask the AI ​​teacher additional questions or hints. The user then enters the question again through the interface and the process is repeated. The input is a new question, and the output is a new answer or explanation.

[1116] The above processing steps allow users to study efficiently and effectively based on specific learning materials. For example, if a user is using calculus learning materials and asks how to use the differentiation formula, the AI ​​teacher will explain how to apply the formula and point them to the relevant part of the learning material.

[1117] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1118] MODE FOR CARRYING OUT THE INVENTION

[1119] To specifically implement this invention, the following system is constructed. The system is made up of components: a user, a terminal, a server, and an emotion engine, all of which operate in conjunction with one another.

[1120] Teaching material input function

[1121] The user inputs the learning materials they wish to use into the system. Specifically, the user uploads the learning materials (PDF or text file) using the device's upload interface. The device then sends the uploaded learning material file to the server. During this process, the device acquires file information and sends it along with the learning material file itself to the server. The server saves the received file, extracts the text in the case of a PDF, and reads the content as is for text files. This preprocesses the content of the learning materials as text data and converts it into a format that can be input into the generation AI.

[1122] AI teacher generation function

[1123] The server sends the preprocessed text data to a generation AI. The generation AI (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for a specific learning material. Once this AI teacher model is generated, the server stores it in a database and associates it with a learning material ID. This process results in an AI teacher that provides educational support tailored to the content of the learning material.

[1124] Learning Session Management

[1125] When a user starts a learning session, they send a request from the device's learning session start interface. The request includes the learning material ID of the learning material they want to use. The device then sends a learning start request to the server. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. This prepares the AI ​​model corresponding to the course for interaction with the learner.

[1126] Interface Features

[1127] As the user progresses with their learning, they input questions and answers into the interface and press the send button. The device then sends the input information to the server. The server then inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. At this time, the AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints as needed.

[1128] Incorporating an emotion engine

[1129] To further improve the user's learning experience, we use an emotion engine that analyzes emotions from the user's facial expressions, tone of voice, and input content.

[1130] Emotion data collection and analysis

[1131] The device uses devices such as sensors, microphones, and cameras to collect user emotional data. The data obtained from these devices is sent to the emotion engine in real time. The emotion engine analyzes this data and detects the user's current emotional state (e.g., stress, excitement, poor comprehension, etc.).

[1132] Emotion-based dialogue adjustment

[1133] The server adjusts the AI ​​teacher model's response based on the emotion data received from the emotion engine. For example, if the user is feeling stressed, the AI ​​teacher will provide gentle explanations and hints. If the user indicates difficulty in understanding, the AI ​​teacher will provide more detailed explanations and step-by-step hints.

[1134] Providing emotional feedback

[1135] As users progress through their studies, the emotion engine continuously monitors their emotional data, allowing the AI ​​teacher to provide feedback and suggest breaks at the appropriate times.

[1136] As a concrete example, consider a case where a user is using a calculus textbook. If the user asks about how to use the differential formula, the AI ​​teacher will explain an example of how the formula can be applied and point them to the relevant section of the textbook. At the same time, if the emotion engine detects the user's frustration, the AI ​​teacher will gently encourage the user by saying, "Shall I explain it again?" to help the user understand.

[1137] In this way, the present invention provides a system that allows learners to study more efficiently and comfortably by combining an emotion engine with an AI teacher specialized in specific learning materials.

[1138] The processing flow will be explained below.

[1139] Teaching material input function

[1140] Step 1:

[1141] The user selects the teaching material (PDF or text file) they want to use and uploads the selected teaching material through the upload interface on their device.

[1142] Step 2:

[1143] The device temporarily stores the uploaded teaching material file in storage and obtains file information (file name, file format, size, etc.).

[1144] Step 3:

[1145] The terminal sends the file itself to the server. The transmitted data also includes file information.

[1146] Step 4:

[1147] The server saves the received file in a designated folder. If the teaching material file is in PDF format, the server extracts the text from the PDF. If it is a text file, the server reads the content as is.

[1148] Step 5:

[1149] The server converts the extracted or read text data into a data format for input into the generative AI.

[1150] AI teacher generation function

[1151] Step 1:

[1152] The server sends the preprocessed text data to the API of the generative AI (e.g., a large-scale language model).

[1153] Step 2:

[1154] The generative AI analyzes the text data sent and generates an AI teacher model specialized for that teaching material.

[1155] Step 3:

[1156] The generation AI returns the generated AI teacher model to the server. The returned data includes the model file and related information.

[1157] Step 4:

[1158] The server stores the received AI teacher model in a database and associates it with the corresponding teaching material ID.

[1159] Learning Session Management

[1160] Step 1:

[1161] The user uses the learning session initiation interface of the device to send a request to start a learning session. This request includes the learning material ID of the learning material to be used.

[1162] Step 2:

[1163] The terminal sends a learning start request to the server.

[1164] Step 3:

[1165] The server receives the request and loads the AI ​​teacher model corresponding to the teaching material ID from the database.

[1166] Step 4:

[1167] The server loads the loaded AI teacher model into memory and initializes the learning session.

[1168] Interface Features

[1169] Step 1:

[1170] The user enters a specific question and answer into the terminal interface and presses the send button.

[1171] Step 2:

[1172] The terminal transmits the input text data to the server.

[1173] Step 3:

[1174] The server inputs the received questions and answers into the AI ​​teacher model and performs the process of generating appropriate answers and explanations.

[1175] Step 4:

[1176] The generated answers and explanations may also include references to the teaching material.

[1177] Step 5:

[1178] The server sends the generated answers and explanations to the terminal.

[1179] Step 6:

[1180] The device displays the received answers and explanations on the user's interface, and the user can request additional questions or hints again.

[1181] Step 7:

[1182] If the user wishes to submit an additional question or request, they provide new input to the terminal and resubmit, which again transmits the data to the server and repeats the process described above.

[1183] Incorporating an emotion engine

[1184] Step 1:

[1185] The device uses devices such as a camera and microphone to collect user emotional data, allowing it to obtain information such as facial expression analysis and voice tone analysis.

[1186] Step 2:

[1187] The device transmits the collected emotional data in real time to the emotion engine, which analyzes it and evaluates the user's emotional state.

[1188] Step 3:

[1189] The emotion engine's analysis results are sent to a server, which then uses the data to tailor the AI ​​teacher's response. For example, if the user is feeling stressed, the AI ​​teacher will provide a gentle explanation.

[1190] Step 4:

[1191] The emotion engine continuously monitors the user's emotions and provides feedback as needed, for example suggesting a break if the user is feeling frustrated.

[1192] Step 5:

[1193] The server sends appropriate feedback according to the emotional state to the terminal, which then displays it on the user's interface.

[1194] In this way, the present invention provides a system that allows learners to study more efficiently and comfortably by combining an emotion engine with an AI teacher specialized in specific learning materials. Each processing step works in conjunction with the other to improve the user's learning experience.

[1195] Example 2

[1196] 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."

[1197] Conventional educational systems have the problem of not providing sufficient individualized educational support tailored to each learner. They also lack a mechanism for understanding the learner's emotional state and providing appropriate feedback. This makes it difficult to provide efficient learning support, hindering the improvement of learners' understanding and motivation to learn.

[1198] 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 a user to input the contents of the learning material into the terminal, means for receiving the learning material file from the terminal and extracting text if the file is a PDF, means for sending the preprocessed text data to a generative AI model and generating an AI teacher specialized for the specific learning material, and means for saving the generated AI teacher model in a database. This allows the user to advance their learning using an AI teacher model specialized for the specific learning material, and further allows them to receive appropriate learning support based on emotion data.

[1199] "User" means a person who uses the system to upload learning materials, initiate learning sessions, and enter questions and answers.

[1200] A "terminal" is a device that a user operates to upload learning materials and input questions and answers, and has the functionality to communicate with the server.

[1201] The "server" is a central computing device that receives the teaching material files sent from the terminal, performs text extraction and preprocessing, and runs the generative AI model.

[1202] "Learning material file" refers to the learning material uploaded by the user to the system, and is in PDF or text file format.

[1203] A "generative AI model" is an artificial intelligence model that analyzes preprocessed text data and generates an AI teacher specialized in specific teaching materials.

[1204] An "AI teacher model" is a virtual teacher model generated by a generative AI model that provides educational support to learners based on specific teaching materials.

[1205] A "database" is a repository of information that stores and manages generated AI training models and other necessary data.

[1206] A "learning session" refers to the entire learning activity of a user using the system, including interactions with the AI ​​teacher model.

[1207] The "emotion engine" is a system component that analyzes the user's emotional state and adjusts the response of the AI ​​teacher model based on that.

[1208] "Emotion data" refers to data used to understand the user's emotional state, such as facial expressions and tone of voice.

[1209] "Text extraction" refers to the process of extracting textual information from non-text format educational material files such as PDFs.

[1210] "Loading into memory" means expanding the AI ​​teacher model read from the database into the server's working memory and preparing it for actual operation.

[1211] To specifically implement this invention, the system is made up of the following components: a user, a terminal, a server, and an emotion engine, all of which operate in conjunction with one another. Details of these components are explained below.

[1212] Teaching material input function

[1213] When a user uses the system, they first input the learning materials (PDF or text files) into the system via their terminal. The terminal provides an upload interface, allowing the user to select and upload the learning materials. In this process, the terminal obtains information about the learning material file (for example, file name and format) and sends this information and the learning material file itself to the server. The server saves the received file, and if it is in PDF format, it uses OCR technology to extract the text. If it is a text file, it simply reads the content as is. This preprocesses the content of the learning materials as text data and converts it into a form that can be input into the generative AI model.

[1214] AI teacher generation function

[1215] The server sends the preprocessed text data to a generative AI model. The generative AI model can be, for example, a large-scale language model (such as GPT-3). The generative AI model analyzes this text data and generates an AI teacher model specialized for a specific teaching material. The generated AI teacher model is stored in a database by the server and associated with the corresponding teaching material ID. This process prepares an AI teacher that provides educational support based on the content of the specific teaching material.

[1216] Learning Session Management

[1217] When a user starts a learning session, they send a request from the device's learning session start interface. The request includes the learning material ID of the learning material they want to use. The device then sends this request to the server, which then reads the corresponding AI teacher model from the database and loads it into memory. This prepares the AI ​​teacher model for interaction with the learner.

[1218] Interface Features

[1219] As the user progresses through their learning, they input questions and answers into the interface and press the send button. The device sends the input information to the server, which then inputs the received questions and answers into the AI ​​teacher model to generate appropriate answers and explanations. The AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints.

[1220] Incorporating an emotion engine

[1221] To further improve the user's learning experience, an emotion engine is used. The emotion engine analyzes emotions in real time from the user's facial expressions, tone of voice, input content, etc. The device is equipped with the necessary devices such as sensors, microphones, and cameras. The emotion engine analyzes the collected data and detects the user's current emotional state (e.g., stress, excitement, lack of comprehension, etc.). The server adjusts the response of the AI ​​teacher model based on the emotion data received from the emotion engine. For example, if the user is feeling stressed, the AI ​​teacher will provide gentle explanations and hints. If the user has difficulty understanding, it will provide more detailed explanations and step-by-step hints. The emotion engine also continuously monitors the user's emotion data and provides feedback and suggests breaks at appropriate times.

[1222] Specific examples

[1223] As a concrete example, consider a case where a user is using a calculus textbook. If the user asks about how to use the differential formula, the AI ​​teacher will explain an example of how the formula can be applied and point them to the relevant section of the textbook. At the same time, if the emotion engine detects the user's frustration, the AI ​​teacher will gently encourage the user by saying, "Shall I explain it again?" to help the user understand.

[1224] Example prompts to input to the generative AI model

[1225] Below is an example of a prompt sentence to input to the generative AI model.

[1226] "Enter the text data of a calculus textbook below. Based on that data, please generate an AI teacher model specialized for this textbook."

[1227] In this way, by combining an AI teacher model specialized for specific teaching materials with an emotion engine, the system provides an environment in which learners can study more efficiently and comfortably.

[1228] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1229] Step 1:

[1230] The user uploads the learning material file.

[1231] Specific operation: The user uses the upload interface on the terminal to select the teaching material file (PDF or text file) and presses the send button.

[1232] Input: A learning material file selected by the user.

[1233] Output: The teaching material file temporarily saved on the device.

[1234] Step 2:

[1235] The device obtains the uploaded file information.

[1236] Specific operation: The device obtains metadata such as file name, file format, and file size.

[1237] Input: The learning material file selected by the user.

[1238] Output: File information (metadata).

[1239] Step 3:

[1240] The terminal transmits the teaching material file and file information to the server.

[1241] Specific operation: The terminal sends the acquired file information and the main body of the teaching material file to the server.

[1242] Input: Teaching material files and file information.

[1243] Output: The teaching material file and file information sent to the server.

[1244] Step 4:

[1245] The server receives the educational material file, extracts the text if it is a PDF, or reads the content as is if it is a text file.

[1246] Specific operation: The server saves the received teaching material file and branches the processing based on the file format. If it is a PDF file, it uses OCR technology to extract the text, and if it is a text file, it reads the content as is.

[1247] Input: The teaching material file sent to the server.

[1248] Output: Preprocessed text data.

[1249] Step 5:

[1250] The server sends the preprocessed text data to the generation AI.

[1251] Specific operation: The server calls an API to send preprocessed text data to a generative AI model (e.g., GPT-3).

[1252] Input: Preprocessed text data.

[1253] Output: Text data received by the generation AI.

[1254] Step 6:

[1255] The generative AI analyzes the text data and generates an AI teacher model.

[1256] Specific operation: The generative AI model analyzes text data and generates an AI teacher model specialized for specific teaching materials.

[1257] Input: Text data received by the generation AI.

[1258] Output: The generated AI teacher model.

[1259] Step 7:

[1260] The server stores the generated AI teacher model in a database and associates it with the teaching material ID.

[1261] Specific operation: The server stores the generated AI teacher model in a database and associates the model with a specific teaching material ID.

[1262] Input: Generated AI teacher model.

[1263] Output: An AI teacher model and its associated information stored in a database.

[1264] Step 8:

[1265] The user requests to start a learning session.

[1266] Specific operation: The user sends a request including the learning material ID he / she wants to use from the learning session start interface of the terminal.

[1267] Input: A user request to start a learning session.

[1268] Output: The request sent to the terminal.

[1269] Step 9:

[1270] The device sends the request and the teaching material ID to the server.

[1271] Specific operation: The terminal sends a learning session start request and the learning material ID to the server.

[1272] Input: A learning session start request from the user and the learning material ID.

[1273] Output: The request sent to the server and the learning material ID.

[1274] Step 10:

[1275] The server receives the request, reads the AI ​​teacher model from the database, and loads it into memory.

[1276] Specific operation: The server reads the corresponding AI teacher model from the database and loads it into memory.

[1277] Input: The request sent to the server and the learning material ID.

[1278] Output: An AI teacher model loaded into memory.

[1279] Step 11:

[1280] The user enters questions and answers into the interface and submits them.

[1281] Specific operation: The user enters a question and answer into the interface and presses the submit button.

[1282] Input: The user's question and answer.

[1283] Output: Information typed into the terminal.

[1284] Step 12:

[1285] The terminal sends the input information to the server.

[1286] Specific operation: The device sends the entered questions and answers to the server.

[1287] Input: Questions and answers typed into the device.

[1288] Output: The question and answer sent to the server.

[1289] Step 13:

[1290] The server inputs questions and answers into an AI teacher model, which generates answers and explanations.

[1291] Specific operation: The server inputs the received questions and answers into the AI ​​teacher model and receives the generated answers and explanations.

[1292] Input: The question and answer sent to the server.

[1293] Output: Answers and explanations generated by the AI ​​teacher model.

[1294] Step 14:

[1295] The server sends the generated answers and explanations to the terminal.

[1296] Specific operation: The server sends the generated answers and explanations to the terminal, where they are displayed on the user's interface.

[1297] Input: Answers and explanations generated by an AI teacher model.

[1298] Output: Answers and explanations sent to your device.

[1299] Step 15:

[1300] The device uses a device to collect emotion data and transmits the data to a server.

[1301] Specific operation: The device uses sensors, microphones, cameras, etc. to collect emotional data such as the user's facial expressions and tone of voice, and sends it to the server.

[1302] Input: Collected emotion data.

[1303] Output: Emotion data sent to the server.

[1304] Step 16:

[1305] The emotion engine analyzes the data and detects the user's emotional state.

[1306] Specific operation: The emotion engine analyzes the received emotion data and detects the user's current emotional state.

[1307] Input: Emotion data sent to the server.

[1308] Output: Parsed emotional state.

[1309] Step 17:

[1310] The server adjusts the response of the AI ​​teacher model based on the emotional data.

[1311] Specific operation: The server adjusts the response of the AI ​​teacher model based on the emotion data sent from the emotion engine. For example, if the user is feeling stressed, it will provide a gentle commentary.

[1312] Input: Parsed emotional state.

[1313] Output: The adjusted response.

[1314] Step 18:

[1315] The emotion engine continuously monitors data and provides feedback and suggests breaks.

[1316] Specific behavior: The emotion engine continuously monitors the user's emotional state and provides feedback and suggests breaks at appropriate times.

[1317] Input: Continuously collected emotion data.

[1318] Output: Feedback and break suggestions.

[1319] (Application example 2)

[1320] 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."

[1321] Conventional educational support systems do not provide feedback that takes into account the learner's individual emotional state regarding their learning progress, making it difficult to provide effective learning support. Furthermore, particularly in new applications such as virtual stores, there is a lack of methods for analyzing users' purchasing motivation and interest in real time and adjusting the dialogue accordingly.

[1322] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for inputting the content of the learning material, a means for generating an AI teacher specialized for a specific learning material based on the content of the learning material input, a means for managing interactions with the learner and providing answers, explanations, and hints via the AI ​​teacher, a means for indicating reference locations in the learning material based on the answers and explanations, a means for analyzing the user's emotional state and adjusting responses based on the results, and a terminal for collecting emotional data. This provides effective learning support that takes into account the learner's individual emotional state, and enables dialogue adjustment based on the user's emotions even in virtual stores, etc.

[1323] The "means for inputting the contents of the teaching material" is an interface for users to upload the teaching material they wish to use to the system, and is used to send the teaching material data in PDF or text file format to the server.

[1324] The "means of generating an AI teacher specialized in specific teaching materials" refers to analyzing uploaded teaching material data and generating an AI model that provides educational support specialized in the content of the teaching materials.

[1325] "Means for managing interactions with learners and providing answers, explanations, and hints through the AI ​​teacher" refers to an AI teacher providing appropriate answers and explanations to questions and requests from learners and managing the progress of the interactions.

[1326] The "means for indicating the reference portion of the teaching material based on the answer or explanation" is a means for indicating to the learner the relevant portion of the teaching material related to the provided answer or explanation.

[1327] "Means for analyzing the user's emotional state and adjusting responses based on that" refers to technology that analyzes the user's emotions in real time from their facial expressions, tone of voice, etc., and adjusts the response content based on the results.

[1328] A "terminal for collecting emotional data" is a device that collects a user's audio and video data in real time and provides the data necessary for emotional analysis.

[1329] To specifically put this invention into practice, the following system is constructed. This system is made up of the following components: a user, a terminal, a server, and an emotion engine, all of which operate in conjunction with one another.

[1330] Teaching material input function

[1331] Users input the learning materials they want to use into the system. Specifically, they upload learning material files (PDF or text files) using the upload interface on their device. The device then sends the uploaded learning material files to the server. The server saves the received files, extracts the text from PDF files, and reads the content directly from text files. This preprocesses the content of the learning materials as text data and converts it into a format that can be input into the generative AI model.

[1332] AI teacher generation function

[1333] The server sends the preprocessed text data to a generative AI model. The generative AI model (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for a specific teaching material. This AI teacher model is stored in a database and associated with a teaching material ID. This allows an AI teacher to provide educational support tailored to the content of the teaching material.

[1334] Learning Session Management

[1335] When a user wants to start a learning session, they send a request from the device's learning session start interface. The request includes the learning material ID of the learning material they want to use. The device then sends a learning start request to the server. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. This prepares the AI ​​model corresponding to the course for interaction with the learner.

[1336] Interface Features

[1337] As the user progresses with their learning, they input questions and answers into the interface and press the send button. The device then sends the input information to the server. The server then inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. At this time, the AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints as needed.

[1338] Incorporating an emotion engine

[1339] To further improve the user's learning experience, we use an emotion engine that analyzes emotions from the user's facial expressions, tone of voice, and input content.

[1340] Emotion data collection and analysis

[1341] The device uses devices such as sensors, microphones, and cameras to collect user emotional data. The data obtained from these devices is sent to the emotion engine in real time. The emotion engine analyzes this data and detects the user's current emotional state (e.g., stress, excitement, poor comprehension, etc.).

[1342] Emotion-based dialogue adjustment

[1343] The server adjusts the AI ​​teacher model's response based on the emotion data received from the emotion engine. For example, if the user is feeling stressed, the AI ​​teacher will provide gentle explanations and hints. If the user indicates difficulty in understanding, the AI ​​teacher will provide more detailed explanations and step-by-step hints.

[1344] Providing emotional feedback

[1345] As users progress through their studies, the emotion engine continuously monitors their emotional data, allowing the AI ​​teacher to provide feedback and suggest breaks at the appropriate times.

[1346] Examples:

[1347] Consider a case where a user is using a calculus textbook.

[1348] When a user asks how to use a differential formula, the AI ​​teacher will explain an example of how the formula can be applied and point them to the relevant section of the teaching materials.

[1349] Example prompt sentence:

[1350] text

[1351] User Input: What are the features of this refrigerator?

[1352] Emotion: Interest

[1353] The right response: This refrigerator is equipped with the latest cooling technology, is extremely energy efficient, and as a smart refrigerator, can be controlled remotely via a dedicated app.

[1354] Hardware and software used:

[1355] Camera and microphone in smart glasses

[1356] OpenCV (camera image acquisition)

[1357] DeepFace (emotional analysis)

[1358] GPT-3 (Response Generation)

[1359] pyttsx3 (audio output)

[1360] In this way, by combining an AI teacher specialized in learning materials with an emotion engine, we can provide a system that allows learners to learn more efficiently and comfortably. In specific applications such as virtual stores, it will also be possible to adjust dialogue based on the user's emotions, improving the customer experience.

[1361] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1362] Step 1:

[1363] The user uploads a learning material file (PDF or text file) using the upload interface of the terminal. The terminal then sends this file to the server. Specifically, the terminal receives the learning material file uploaded by the user, obtains the file information, and sends it to the server. The input is the learning material file selected by the user, and the output is the learning material file data sent to the server.

[1364] Step 2:

[1365] The server stores the received learning material files. If it is a PDF file, text extraction is performed, and if it is a text file, the content is read as is. The extracted text data is preprocessed and converted into a format that can be input into the generative AI model. The input is the learning material file, and the output is the preprocessed text data.

[1366] Step 3:

[1367] The server sends the preprocessed text data to a generative AI model. The generative AI model (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for a specific learning material. The server stores the generated AI teacher model in a database and associates it with a learning material ID. The input is the preprocessed text data, and the output is the AI ​​teacher model.

[1368] Step 4:

[1369] The user sends a request from the learning session start interface on the device. The request includes the learning material ID of the learning material they want to use. The device then sends a learning start request to the server. The input is the learning start request, and the output is the request data to the server.

[1370] Step 5:

[1371] The server receives a learning start request, reads the corresponding AI teacher model from the database, and loads it into memory. This prepares the AI ​​teacher model for interaction with the learner. The input is the learning start request, and the output is the AI ​​teacher model loaded into memory.

[1372] Step 6:

[1373] The user enters a question or answer into the interface and presses the send button. The terminal sends the entered information to the server. The input is the question or answer entered by the user, and the output is the data sent to the server.

[1374] Step 7:

[1375] The server inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. At this time, the AI ​​teacher model can also indicate references in the teaching materials. The generated answers and explanations are sent back to the device and displayed on the user's interface. The input is the questions and answers, and the output is the generated answers and explanations.

[1376] Step 8:

[1377] The device uses devices such as a camera and microphone to collect the user's emotional data. The data obtained from these devices is sent to the emotion engine in real time. The input is the user's audio and video data, and the output is data on the user's emotional state.

[1378] Step 9:

[1379] The emotion engine analyzes the user's current emotional state from their facial expressions and tone of voice. The analysis results are sent to the server. The input is audio and video data, and the output is analyzed emotional data.

[1380] Step 10:

[1381] The server adjusts the response of the AI ​​teacher model based on the emotional data received from the emotion engine. For example, if the user is feeling stressed, the AI ​​teacher will provide gentle explanations and hints. The input is the emotional data, and the output is the adjusted response.

[1382] Step 11:

[1383] The server continuously monitors the emotion data and provides feedback to the user or suggests taking a break at the appropriate time. The input is continuously collected emotion data, and the output is feedback to the user.

[1384] The above are the processing steps of this system, and the specific operations and data input and output are clearly stated for each step, allowing for a detailed understanding of the specific operation method and effects of the invention.

[1385] 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.

[1386] 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.

[1387] 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.

[1388] [Fourth embodiment]

[1389] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1390] 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.

[1391] 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).

[1392] 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.

[1393] 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.

[1394] 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).

[1395] 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.

[1396] 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.

[1397] 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.

[1398] 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.

[1399] 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.

[1400] 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.

[1401] 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."

[1402] MODE FOR CARRYING OUT THE INVENTION

[1403] To specifically implement the present invention, the following system is constructed: The system is made up of user, terminal, and server components, each of which operates in cooperation with each other.

[1404] Teaching material input function

[1405] The user inputs the learning materials they wish to use into the system. Specifically, the user uploads the learning materials (PDF or text file) using the device's upload interface. The device then sends the uploaded learning material file to the server. During this process, the device acquires file information and sends it along with the learning material file itself to the server. The server saves the received file, extracts the text in the case of a PDF, and reads the content as is for text files. This preprocesses the content of the learning materials as text data and converts it into a format that can be input into the generation AI.

[1406] AI teacher generation function

[1407] The server sends the preprocessed text data to a generation AI. The generation AI (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for a specific learning material. Once this AI teacher model is generated, the server stores it in a database and associates it with a learning material ID. This process results in an AI teacher that provides educational support tailored to the content of the learning material.

[1408] Learning Session Management

[1409] When a user starts a learning session, they send a request through the device interface. The request includes the learning material ID of the learning material they want to use. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. This prepares the AI ​​model for the course to interact with the learner.

[1410] Interface Features

[1411] As the user progresses with their learning, they input questions and answers into the interface and press the send button. The device then sends the input information to the server. The server then inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. At this time, the AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints as needed.

[1412] This series of processes allows users to study efficiently and effectively based on specific learning materials. For example, if a user is using calculus learning materials and asks about how to use the differentiation formula, the AI ​​teacher will explain how to apply the formula and point them to the relevant part of the learning material. This allows learners to deepen their understanding through the learning materials and progress effectively.

[1413] As described above, the present invention provides a system that generates an AI teacher specialized in specific learning materials, allowing learners to progress through their studies efficiently.

[1414] The processing flow will be explained below.

[1415] Teaching material input function

[1416] Step 1:

[1417] The user selects the teaching material (PDF or text file) they want to use and uploads the selected teaching material through the upload interface on their device.

[1418] Step 2:

[1419] The device temporarily stores the uploaded learning material file in its storage, and simultaneously acquires file information (file name, file format, size, etc.).

[1420] Step 3:

[1421] The terminal sends the file itself to the server. The transmitted data also includes file information.

[1422] Step 4:

[1423] The server saves the received file in a designated folder. If the teaching material file is in PDF format, the server extracts the text from the PDF. If it is a text file, the server reads the content as is.

[1424] Step 5:

[1425] The server converts the extracted or read text data into a data format for input into the generative AI.

[1426] AI teacher generation function

[1427] Step 1:

[1428] The server sends the preprocessed text data to the API of the generative AI (e.g., a large-scale language model).

[1429] Step 2:

[1430] The generative AI analyzes the text data sent and generates an AI teacher model specialized for that teaching material.

[1431] Step 3:

[1432] The generation AI returns the generated AI teacher model to the server. The returned data includes the model file and related information.

[1433] Step 4:

[1434] The server stores the received AI teacher model in a database and associates it with the corresponding teaching material ID.

[1435] Learning Session Management

[1436] Step 1:

[1437] The user uses the learning session initiation interface of the device to send a request to start a learning session. This request includes the learning material ID of the learning material to be used.

[1438] Step 2:

[1439] The terminal sends a learning start request to the server.

[1440] Step 3:

[1441] The server receives the request and loads the AI ​​teacher model corresponding to the teaching material ID from the database.

[1442] Step 4:

[1443] The server loads the loaded AI teacher model into memory and initializes the learning session.

[1444] Interface Features

[1445] Step 1:

[1446] The user enters a specific question and answer into the terminal interface and presses the send button.

[1447] Step 2:

[1448] The terminal transmits the input text data to the server.

[1449] Step 3:

[1450] The server inputs the received questions and answers into the AI ​​teacher model and performs the process of generating appropriate answers and explanations.

[1451] Step 4:

[1452] The generated answers and explanations may also include references to the teaching material.

[1453] Step 5:

[1454] The server sends the generated answers and explanations to the terminal.

[1455] Step 6:

[1456] The device displays the received answers and explanations on the user's interface, and the user can request additional questions or hints again.

[1457] Step 7:

[1458] If the user wishes to submit an additional question or request, they provide new input to the terminal and resubmit, which again transmits the data to the server and repeats the process described above.

[1459] This allows users to efficiently study based on specific learning materials. Each processing step operates in conjunction with the other steps, providing a smooth learning environment for learners.

[1460] Example 1

[1461] 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."

[1462] Conventional learning support systems have difficulty providing appropriate answers and explanations specific to the content of the learning materials, making it difficult to meet the individual needs of learners. Furthermore, existing systems lack the functionality to directly point to reference locations in the learning materials, hindering efficient learning. Furthermore, the processes involved in generating and storing AI teacher models are cumbersome, often resulting in inconvenience when managing learning sessions. There is a need to address these issues.

[1463] 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.

[1464] In this invention, the server includes: means for inputting the content of the teaching material; means for generating an AI teacher specialized for a specific teaching material based on the content of the input teaching material; means for managing interactions with the learner and providing answers, explanations, and hints via the AI ​​teacher; means for indicating reference locations in the teaching material based on the answers and explanations; means for saving the generated AI teacher model in a database and associating it with a teaching material ID; means for transmitting preprocessed text data to a generation AI; means for the generation AI to generate a teacher model based on the specific teaching material; means for a user to input a question via a terminal and transmit the question to the server; and means for displaying the answers and explanations generated by the AI ​​teacher on the user's interface. This allows for efficient generation and management of an AI teacher model specialized for a teaching material, enabling learners to progress in their studies more efficiently and effectively.

[1465] The "means for inputting the contents of the teaching material" is a terminal interface that allows the user to upload teaching material files (PDF or text files) to the system and transmit the contents to the server.

[1466] "Means for generating AI teachers specialized for specific teaching materials" refers to the process of analyzing preprocessed text data and using generative AI to create an AI teacher model suitable for the content of specific teaching materials.

[1467] "Means for managing interactions with learners" refers to the process by which the server manages questions and answers sent by learners during a learning session and generates and provides appropriate answers from the AI ​​teacher model.

[1468] "Means for providing answers, explanations, and hints" refers to the process of utilizing an AI teacher model to generate answers to learners' questions, explanations about the content of the teaching materials, and additional hints, and provide them to learners.

[1469] "Means for indicating reference points in the teaching materials" refers to the process of pointing learners to specific parts of the teaching materials based on the answers and explanations generated by the AI ​​teacher model.

[1470] "Means for storing the generated AI teacher model in a database and associating it with a teaching material ID" refers to the process of storing the generated AI teacher model in a database and managing it in association with a specific teaching material ID.

[1471] "Means for sending preprocessed text data to the generation AI" refers to the process by which the server passes the preprocessed text data to the generation AI in order to analyze the content of the teaching materials.

[1472] "The means by which the generation AI generates a teacher model based on specific teaching materials" refers to the process by which the generation AI analyzes given text data and generates a teacher model corresponding to the content of specific teaching materials.

[1473] "Means for a user to input a question via a terminal and transmit the question to a server" refers to a process by which a learner inputs a question using a terminal interface and transmits it to a server.

[1474] "Means for displaying the answers and explanations generated by the AI ​​teacher on the user's interface" refers to the process of sending the answers and explanations generated by the AI ​​teacher model from the server to the terminal and displaying them on the user's interface.

[1475] MODE FOR CARRYING OUT THE INVENTION

[1476] This invention is a learning support system that generates an AI teacher model based on specific learning materials, allowing learners to study efficiently. This system is composed of user, terminal, and server components, which operate in cooperation with each other.

[1477] Hardware and Software Overview

[1478] The hardware used includes the devices used by users (PCs, tablets, smartphones, etc.) and a server for storing and processing data. A database management system runs on the server, storing the generated AI teacher model and teaching material data. Specific software used includes a text extraction engine for preprocessing (e.g., Tesseract OCR, PDFMiner) and generative AI (e.g., GPT-3, BERT).

[1479] Teaching material input function

[1480] The user selects and uploads a learning material file (PDF or text file) through the device's upload interface. The device then sends the selected file to the server via an HTTP POST request. The information sent includes the file's metadata (file name, size, format, etc.).

[1481] Preprocessing the files

[1482] The server saves the received learning material file in storage (e.g., Amazon S3, local file system). Next, the server checks the file format, and if it is a PDF, it uses a text extraction engine (e.g., Tesseract OCR or PDFMiner) to extract the text data. If it is a text file, it simply reads the content. This text data is preprocessed and converted into a format suitable for the generative AI model.

[1483] AI teacher generation function

[1484] The server sends the preprocessed text data to the generation AI. The generation AI (for example, GPT-3 or BERT) analyzes this text data and generates an AI teacher model specialized for a specific teaching material. The generated AI teacher model is associated with the teaching material ID and stored in a database. This process prepares an AI teacher that provides educational support tailored to the content of the teaching material.

[1485] Learning Session Management

[1486] When a user starts a learning session, they send a request from the device interface, including the learning material ID of the learning material they want to use. The server receives the request, reads the corresponding AI teacher model from the database, and loads it into memory. This makes the AI ​​teacher model corresponding to the selected learning material ready for interaction with the learner.

[1487] Interface Features

[1488] As the user progresses with their learning, they input questions and answers into the device's interface and press the send button. The device then sends the input information to the server. The server then inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. The AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints.

[1489] Specific examples

[1490] For example, if a user is using a calculus textbook and asks, "I don't know how to use the differential formula," the AI ​​teacher will explain how to apply the formula and point them to the relevant part of the textbook. This allows the learner to deepen their understanding through the textbook and progress effectively.

[1491] Prompt Sentence Examples

[1492] Use this differentiation formula to find the derivative of the following function:

[1493] The above is a specific embodiment for carrying out the present invention. This system enables a user to study efficiently and effectively based on specific learning materials.

[1494] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1495] Step 1: Upload your materials

[1496] The user selects a learning material file (PDF or text file) through the terminal interface and clicks the upload button. The terminal sends the selected file to the server via an HTTP POST request. The input is the learning material file and its metadata (file name, size, format, etc.). The output is the learning material file saved on the server.

[1497] Specific operation: When a user selects a file and presses the upload button, the device collects the file contents and metadata and sends them to the server. The server checks the received data and determines the directory to save the file.

[1498] Step 2: Preprocessing the files

[1499] The server saves the received learning material file in storage. Next, the server checks the file format. The input is the saved learning material file. If it is a PDF file, the server extracts the text data using a text extraction engine (e.g., Tesseract OCR, PDFMiner). If it is a text file, the server reads the content as is. The output is the preprocessed text data.

[1500] Specific operation: After the server saves the received file in storage, it checks the file format and, if it is a PDF, uses PDFMiner to extract the text from each page and combine the entire text data into a single string. Text files are read directly.

[1501] Step 3: Generate an AI teacher model

[1502] The server sends the preprocessed text data to the generation AI. The generation AI (e.g., GPT-3, BERT) analyzes this text data and generates an AI teacher model specialized for specific learning materials. The input is the preprocessed text data. The output is the generated AI teacher model.

[1503] Specific operation: The server sends the preprocessed text data to the generation AI. The generation AI analyzes the text data and generates a training model. The generated model is returned to the server and stored in a database.

[1504] Step 4: Save the AI ​​teacher model

[1505] The server saves the generated AI teacher model in a database and associates it with the learning material ID. The inputs are the AI ​​teacher model and the learning material ID. The output is the AI ​​teacher model saved in the database.

[1506] Specific operation: The server assigns a unique teaching material ID to the generated AI teacher model and saves it in the database. When saving, the model and teaching material ID are associated.

[1507] Step 5: Start your study session

[1508] When a user starts a learning session via their device, they select the ID of the learning material they want to study and send a request. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. The input is the user's request. The output is the AI ​​teacher model loaded into memory.

[1509] Specific operation: When a user sends a request including a teaching material ID, the server receives the request, retrieves the corresponding AI teacher model from the database, and loads it into memory.

[1510] Step 6: Enter and submit your question

[1511] As the user progresses with their learning, they input questions into the device interface and press the send button. The device then sends the questions to the server. The input is the user's question, and the output is the question sent to the server.

[1512] Specific operation: When a user inputs a question and presses the send button, the terminal generates a communication packet for sending the question to the server and sends it to the server.

[1513] Step 7: Parsing the question and generating an answer

[1514] The server inputs the received question into the AI ​​teacher model and generates an appropriate answer or explanation. The inputs are the user's question and the AI ​​teacher model. The output is the generated answer or explanation.

[1515] Specific operation: The server inputs the received question into the AI ​​teacher model, which analyzes the question and generates an appropriate answer or explanation, which is then returned to the server.

[1516] Step 8: View your answers

[1517] The server sends the generated answers and explanations to the terminal and displays them on the user's interface. The input is the generated answers and explanations. The output is the answers and explanations displayed on the user's interface.

[1518] Specific operation: The server generates a communication packet containing the generated answer and explanation, and sends it to the terminal. The terminal analyzes the received data and displays it on the user's interface.

[1519] (Application example 1)

[1520] 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."

[1521] Conventional learning support systems lack the ability to generate AI teachers specialized for learning materials and respond to learners' questions in real time, making efficient and effective learning difficult. Furthermore, even in cram schools and preparatory schools that provide education in brick-and-mortar locations, students have limited means of receiving immediate, detailed feedback during self-study, which hinders their learning progress.

[1522] 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.

[1523] In this invention, the server includes a means for inputting the content of the learning materials, a means for generating an AI teacher specialized for a specific learning material based on the content of the learning materials input, a means for managing interactions with the learner and providing answers, explanations, and hints via the AI ​​teacher, a means for indicating reference locations in the learning materials based on the answers and explanations, and a means for students to upload learning materials using a smart device in a brick-and-mortar store, generate an AI teacher based on the content of the uploaded learning materials, and provide learning support in real time. This enables students to study efficiently and effectively during self-study even in brick-and-mortar educational settings.

[1524] "Means for inputting the contents of the teaching materials" refers to a device or software function that allows users to upload teaching material data (such as PDF or text files) to the system.

[1525] The "means for generating an AI teacher specialized for a specific teaching material" refers to an algorithm or software function for analyzing the content of the uploaded teaching material and generating an artificial intelligence model specialized for that teaching material.

[1526] "Means for managing learner interactions" refers to a device or software function that receives questions or requests from learners, generates answers or explanations based on those questions, and returns them to the learner through the AI ​​teacher model.

[1527] "Means for providing answers, explanations, and hints" refers to devices or software functions that present information generated by the AI ​​teacher model to learners and support their learning.

[1528] "Means for indicating where to refer to teaching materials based on answers and explanations" refers to a device or software function that instructs learners on which specific parts of the teaching materials they should refer to based on information generated by the AI ​​teacher model.

[1529] "Means for students to upload learning materials using smart devices at a brick-and-mortar educational institution" means a device or software feature that allows students to upload learning materials to the system at a brick-and-mortar educational institution using a device such as a smartphone or tablet.

[1530] "Means for generating an AI teacher based on the content and providing learning support in real time" refers to a device or software function that instantly analyzes the content of uploaded teaching materials, generates an AI teacher model based on that, and then provides answers, explanations, and hints to learners in real time.

[1531] To specifically implement this invention, the following system is constructed. The system is mainly composed of a user terminal, a server, and a database. Each component operates in cooperation with the others.

[1532] Teaching material input function

[1533] A user uploads a learning material file (PDF or text file) using a device such as a smartphone or tablet. The learning material file is sent to the server through the device's upload interface. The server saves the received learning material file, extracts the text from PDF files, and reads the content directly from text files. During this process, the content of the learning material is preprocessed as text data and converted into a format that can be input into the generative AI model.

[1534] AI teacher generation function

[1535] The server sends the preprocessed text data to a generative AI model. The generative AI model (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for a specific teaching material. Once this AI teacher model is generated, the server stores it in a database and associates it with a teaching material ID. This creates an AI teacher that provides educational support tailored to the content of the teaching material.

[1536] Learning Session Management

[1537] When a user starts a learning session, they send a request through the device interface. The request includes the learning material ID of the learning material they want to use. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. This prepares the AI ​​model for the course to interact with the learner.

[1538] Interface Features

[1539] As the user progresses with their learning, they input questions and answers into the interface and press the send button. The device then sends the input information to the server. The server then inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. The AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints.

[1540] In-store learning support

[1541] At brick-and-mortar educational institutions (such as cram schools and preparatory schools), students upload study materials using smartphones or tablets, and an AI teacher is generated based on the content. This AI teacher can provide students with answers, explanations, and hints in real time. For example, if a student uploads calculus study materials and asks a question about a differentiation formula, the AI ​​teacher will explain how the formula can be applied and point them to the relevant section of the study material.

[1542] Hardware and software used

[1543] Smartphones and tablets: Used to upload teaching materials and as an interface.

[1544] Server: Storage of teaching material data, text extraction, generation and management of AI teaching models.

[1545] Database: Stores AI teacher models and teaching material data.

[1546] Generative AI models: Large-scale language models (e.g., OpenAI API).

[1547] Prompt Sentence Examples

[1548] Use the following as a prompt to generate an AI teacher based on the content of the teaching material:

[1549] "Generate an AI teacher specialized in the following materials: Starting with the basic formulas of calculus, and explaining examples of the application of various formulas. As a specific problem..."

[1550] Use the following as a prompt based on the student's question:

[1551] "Please tell me the basic formula for differentiation."

[1552] The system described above allows students to study efficiently and effectively during self-study, even in brick-and-mortar educational settings.

[1553] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1554] Step 1:

[1555] A user uploads a learning material file (PDF or text file) using a device such as a smartphone or tablet. At this time, the user specifies the learning material file in the device's upload interface and presses the "Upload" button. The device retrieves the contents of the learning material file and sends it to the server. The input is the learning material file, and the output is the learning material file sent to the server.

[1556] Step 2:

[1557] The server saves the received learning material file. Next, if the received file is in PDF format, the server performs text extraction processing. A PDF parser (e.g., PyPDF2) is used for text extraction, and the extracted text data is converted into a format that can be input into the generative AI model. The input is the learning material file, and the output is text data.

[1558] Step 3:

[1559] The server sends the preprocessed text data to a generative AI model. The generative AI model (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for specific learning materials. This process uses the OpenAI API. The input is text data, and the output is an AI teacher model.

[1560] Step 4:

[1561] The generated AI teacher model is stored in a database by the server and associated with the learning material ID. The server uses this association to call the appropriate AI teacher model in subsequent learning sessions. The input is the AI ​​teacher model and the learning material ID, and the output is storage in the database.

[1562] Step 5:

[1563] When a user starts a learning session, they send a request from the device interface. This request includes the learning material ID of the learning material they want to use. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. The input is the learning material ID, and the output is the AI ​​teacher model loaded into memory.

[1564] Step 6:

[1565] The user uses the interface to input questions and answers and presses the send button. The device sends this information to the server. The input is the user's question and answer, and the output is the transmission to the server.

[1566] Step 7:

[1567] The server inputs the received questions and answers into an AI teacher model to generate appropriate answers and explanations. This process uses a generative AI model. The input is the user's question and answer, and the output is the generated answer and explanation.

[1568] Step 8:

[1569] The AI ​​teacher model can also provide references to teaching materials along with the generated answers and explanations. The generated answers and explanations are sent back to the device and displayed on the user's interface. The input is the generated data from the AI ​​teacher model, and the output is the information displayed on the user's interface.

[1570] Step 9:

[1571] Through this process, the user can ask the AI ​​teacher additional questions or hints. The user then enters the question again through the interface and the process is repeated. The input is a new question, and the output is a new answer or explanation.

[1572] The above processing steps allow users to study efficiently and effectively based on specific learning materials. For example, if a user is using calculus learning materials and asks how to use the differentiation formula, the AI ​​teacher will explain how to apply the formula and point them to the relevant part of the learning material.

[1573] 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.

[1574] MODE FOR CARRYING OUT THE INVENTION

[1575] To specifically implement this invention, the following system is constructed. The system is made up of components: a user, a terminal, a server, and an emotion engine, all of which operate in conjunction with one another.

[1576] Teaching material input function

[1577] The user inputs the learning materials they wish to use into the system. Specifically, the user uploads the learning materials (PDF or text file) using the device's upload interface. The device then sends the uploaded learning material file to the server. During this process, the device acquires file information and sends it along with the learning material file itself to the server. The server saves the received file, extracts the text in the case of a PDF, and reads the content as is for text files. This preprocesses the content of the learning materials as text data and converts it into a format that can be input into the generation AI.

[1578] AI teacher generation function

[1579] The server sends the preprocessed text data to a generation AI. The generation AI (e.g., a large-scale language model) analyzes this text data and generates an AI teacher model specialized for a specific learning material. Once this AI teacher model is generated, the server stores it in a database and associates it with a learning material ID. This process results in an AI teacher that provides educational support tailored to the content of the learning material.

[1580] Learning Session Management

[1581] When a user starts a learning session, they send a request from the device's learning session start interface. The request includes the learning material ID of the learning material they want to use. The device then sends a learning start request to the server. The server receives this request, reads the corresponding AI teacher model from the database, and loads it into memory. This prepares the AI ​​model corresponding to the course for interaction with the learner.

[1582] Interface Features

[1583] As the user progresses with their learning, they input questions and answers into the interface and press the send button. The device then sends the input information to the server. The server then inputs the received questions and answers into the AI ​​teacher model, which generates appropriate answers and explanations. At this time, the AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints as needed.

[1584] Incorporating an emotion engine

[1585] To further improve the user's learning experience, we use an emotion engine that analyzes emotions from the user's facial expressions, tone of voice, and input content.

[1586] Emotion data collection and analysis

[1587] The device uses devices such as sensors, microphones, and cameras to collect user emotional data. The data obtained from these devices is sent to the emotion engine in real time. The emotion engine analyzes this data and detects the user's current emotional state (e.g., stress, excitement, poor comprehension, etc.).

[1588] Emotion-based dialogue adjustment

[1589] The server adjusts the AI ​​teacher model's response based on the emotion data received from the emotion engine. For example, if the user is feeling stressed, the AI ​​teacher will provide gentle explanations and hints. If the user indicates difficulty in understanding, the AI ​​teacher will provide more detailed explanations and step-by-step hints.

[1590] Providing emotional feedback

[1591] As users progress through their studies, the emotion engine continuously monitors their emotional data, allowing the AI ​​teacher to provide feedback and suggest breaks at the appropriate times.

[1592] As a concrete example, consider a case where a user is using a calculus textbook. If the user asks about how to use the differential formula, the AI ​​teacher will explain an example of how the formula can be applied and point them to the relevant section of the textbook. At the same time, if the emotion engine detects the user's frustration, the AI ​​teacher will gently encourage the user by saying, "Shall I explain it again?" to help the user understand.

[1593] In this way, the present invention provides a system that allows learners to study more efficiently and comfortably by combining an emotion engine with an AI teacher specialized in specific learning materials.

[1594] The processing flow will be explained below.

[1595] Teaching material input function

[1596] Step 1:

[1597] The user selects the teaching material (PDF or text file) they want to use and uploads the selected teaching material through the upload interface on their device.

[1598] Step 2:

[1599] The device temporarily stores the uploaded teaching material file in storage and obtains file information (file name, file format, size, etc.).

[1600] Step 3:

[1601] The terminal sends the file itself to the server. The transmitted data also includes file information.

[1602] Step 4:

[1603] The server saves the received file in a designated folder. If the teaching material file is in PDF format, the server extracts the text from the PDF. If it is a text file, the server reads the content as is.

[1604] Step 5:

[1605] The server converts the extracted or read text data into a data format for input into the generative AI.

[1606] AI teacher generation function

[1607] Step 1:

[1608] The server sends the preprocessed text data to the API of the generative AI (e.g., a large-scale language model).

[1609] Step 2:

[1610] The generative AI analyzes the text data sent and generates an AI teacher model specialized for that teaching material.

[1611] Step 3:

[1612] The generation AI returns the generated AI teacher model to the server. The returned data includes the model file and related information.

[1613] Step 4:

[1614] The server stores the received AI teacher model in a database and associates it with the corresponding teaching material ID.

[1615] Learning Session Management

[1616] Step 1:

[1617] The user uses the learning session initiation interface of the device to send a request to start a learning session. This request includes the learning material ID of the learning material to be used.

[1618] Step 2:

[1619] The terminal sends a learning start request to the server.

[1620] Step 3:

[1621] The server receives the request and loads the AI ​​teacher model corresponding to the teaching material ID from the database.

[1622] Step 4:

[1623] The server loads the loaded AI teacher model into memory and initializes the learning session.

[1624] Interface Features

[1625] Step 1:

[1626] The user enters a specific question and answer into the terminal interface and presses the send button.

[1627] Step 2:

[1628] The terminal transmits the input text data to the server.

[1629] Step 3:

[1630] The server inputs the received questions and answers into the AI ​​teacher model and performs the process of generating appropriate answers and explanations.

[1631] Step 4:

[1632] The generated answers and explanations may also include references to the teaching material.

[1633] Step 5:

[1634] The server sends the generated answers and explanations to the terminal.

[1635] Step 6:

[1636] The device displays the received answers and explanations on the user's interface, and the user can request additional questions or hints again.

[1637] Step 7:

[1638] If the user wishes to submit an additional question or request, they provide new input to the terminal and resubmit, which again transmits the data to the server and repeats the process described above.

[1639] Incorporating an emotion engine

[1640] Step 1:

[1641] The device uses devices such as a camera and microphone to collect user emotional data, allowing it to obtain information such as facial expression analysis and voice tone analysis.

[1642] Step 2:

[1643] The device transmits the collected emotional data in real time to the emotion engine, which analyzes it and evaluates the user's emotional state.

[1644] Step 3:

[1645] The emotion engine's analysis results are sent to a server, which then uses the data to tailor the AI ​​teacher's response. For example, if the user is feeling stressed, the AI ​​teacher will provide a gentle explanation.

[1646] Step 4:

[1647] The emotion engine continuously monitors the user's emotions and provides feedback as needed, for example suggesting a break if the user is feeling frustrated.

[1648] Step 5:

[1649] The server sends appropriate feedback according to the emotional state to the terminal, which then displays it on the user's interface.

[1650] In this way, the present invention provides a system that allows learners to study more efficiently and comfortably by combining an emotion engine with an AI teacher specialized in specific learning materials. Each processing step works in conjunction with the other to improve the user's learning experience.

[1651] Example 2

[1652] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1653] Conventional educational systems have the problem of not providing sufficient individualized educational support tailored to each learner. They also lack a mechanism for understanding the learner's emotional state and providing appropriate feedback. This makes it difficult to provide efficient learning support, hindering the improvement of learners' understanding and motivation to learn.

[1654] 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 a user to input the contents of the learning material into the terminal, means for receiving the learning material file from the terminal and extracting text if the file is a PDF, means for sending the preprocessed text data to a generative AI model and generating an AI teacher specialized for the specific learning material, and means for saving the generated AI teacher model in a database. This allows the user to advance their learning using an AI teacher model specialized for the specific learning material, and further allows them to receive appropriate learning support based on emotion data.

[1655] "User" means a person who uses the system to upload learning materials, initiate learning sessions, and enter questions and answers.

[1656] A "terminal" is a device that a user operates to upload learning materials and input questions and answers, and has the functionality to communicate with the server.

[1657] The "server" is a central computing device that receives the teaching material files sent from the terminal, performs text extraction and preprocessing, and runs the generative AI model.

[1658] "Learning material file" refers to the learning material uploaded by the user to the system, and is in PDF or text file format.

[1659] A "generative AI model" is an artificial intelligence model that analyzes preprocessed text data and generates an AI teacher specialized in specific teaching materials.

[1660] An "AI teacher model" is a virtual teacher model generated by a generative AI model that provides educational support to learners based on specific teaching materials.

[1661] A "database" is a repository of information that stores and manages generated AI training models and other necessary data.

[1662] A "learning session" refers to the entire learning activity of a user using the system, including interactions with the AI ​​teacher model.

[1663] The "emotion engine" is a system component that analyzes the user's emotional state and adjusts the response of the AI ​​teacher model based on that.

[1664] "Emotion data" refers to data used to understand the user's emotional state, such as facial expressions and tone of voice.

[1665] "Text extraction" refers to the process of extracting textual information from non-text format educational material files such as PDFs.

[1666] "Loading into memory" means expanding the AI ​​teacher model read from the database into the server's working memory and preparing it for actual operation.

[1667] To specifically implement this invention, the system is made up of the following components: a user, a terminal, a server, and an emotion engine, all of which operate in conjunction with one another. Details of these components are explained below.

[1668] Teaching material input function

[1669] When a user uses the system, they first input the learning materials (PDF or text files) into the system via their terminal. The terminal provides an upload interface, allowing the user to select and upload the learning materials. In this process, the terminal obtains information about the learning material file (for example, file name and format) and sends this information and the learning material file itself to the server. The server saves the received file, and if it is in PDF format, it uses OCR technology to extract the text. If it is a text file, it simply reads the content as is. This preprocesses the content of the learning materials as text data and converts it into a form that can be input into the generative AI model.

[1670] AI teacher generation function

[1671] The server sends the preprocessed text data to a generative AI model. The generative AI model can be, for example, a large-scale language model (such as GPT-3). The generative AI model analyzes this text data and generates an AI teacher model specialized for a specific teaching material. The generated AI teacher model is stored in a database by the server and associated with the corresponding teaching material ID. This process prepares an AI teacher that provides educational support based on the content of the specific teaching material.

[1672] Learning Session Management

[1673] When a user starts a learning session, they send a request from the device's learning session start interface. The request includes the learning material ID of the learning material they want to use. The device then sends this request to the server, which then reads the corresponding AI teacher model from the database and loads it into memory. This prepares the AI ​​teacher model for interaction with the learner.

[1674] Interface Features

[1675] As the user progresses through their learning, they input questions and answers into the interface and press the send button. The device sends the input information to the server, which then inputs the received questions and answers into the AI ​​teacher model to generate appropriate answers and explanations. The AI ​​teacher model can also provide references to learning materials. The generated answers and explanations are then sent back to the device and displayed on the user's interface. Through this process, the user can ask the AI ​​teacher for additional questions or hints.

[1676] Incorporating an emotion engine

[1677] To further improve the user's learning experience, an emotion engine is used. The emotion engine analyzes emotions in real time from the user's facial expressions, tone of voice, input content, etc. The device is equipped with the necessary devices such as sensors, microphones, and cameras. The emotion engine analyzes the collected data and detects the user's current emotional state (e.g., stress, excitement, lack of comprehension, etc.). The server adjusts the response of the AI ​​teacher model based on the emotion data received from the emotion engine. For example, if the user is feeling stressed, the AI ​​teacher will provide gentle explanations and hints. If the user has difficulty understanding, it will provide more detailed explanations and step-by-step hints. The emotion engine also continuously monitors the user's emotion data and provides feedback and suggests breaks at appropriate times.

[1678] Specific examples

[1679] As a concrete example, consider a case where a user is using a calculus textbook. If the user asks about how to use the differential formula, the AI ​​teacher will explain an example of how the formula can be applied and point them to the relevant section of the textbook. At the same time, if the emotion engine detects the user's frustration, the AI ​​teacher will gently encourage the user by saying, "Shall I explain it again?" to help the user understand.

[1680] Example prompts to input to the generative AI model

[1681] Below is an example of a prompt sentence to input to the generative AI model.

[1682] "Enter the text data of a calculus textbook below. Based on that data, please generate an AI teacher model specialized for this textbook."

[1683] In this way, by combining an AI teacher model specialized for specific teaching materials with an emotion engine, the system provides an environment in which learners can study more efficiently and comfortably.

[1684] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1685] Step 1:

[1686] The user uploads the learning material file.

[1687] Specific operation: The user uses the upload interface on the terminal to select the teaching material file (PDF or text file) and presses the send button.

[1688] Input: A learning material file selected by the user.

[1689] Output: The teaching material file temporarily saved on the device.

[1690] Step 2:

[1691] The device obtains the uploaded file information.

[1692] Specific operation: The device obtains metadata such as file name, file format, and file size.

[1693] Input: The learning material file selected by the user.

[1694] Output: File information (metadata).

[1695] Step 3:

[1696] The terminal transmits the teaching material file and file information to the server.

[1697] Specific operation: The terminal sends the acquired file information and the main body of the teaching material file to the server.

[1698] Input: Teaching material files and file information.

[1699] Output: The teaching material file and file information sent to the server.

[1700] Step 4:

[1701] The server receives the educational material file, extracts the text if it is a PDF, or reads the content as is if it is a text file.

[1702] Specific operation: The server saves the received teaching material file and branches the processing based on the file format. If it is a PDF file, it uses OCR technology to extract the text, and if it is a text file, it reads the content as is.

[1703] Input: The teaching material file sent to the server.

[1704] Output: Preprocessed text data.

[1705] Step 5:

[1706] The server sends the preprocessed text data to the generation AI.

[1707] Specific operation: The server calls an API to send preprocessed text data to a generative AI model (e.g., GPT-3).

[1708] Input: Preprocessed text data.

[1709] Output: Text data received by the generation AI.

[1710] Step 6:

[1711] The generative AI analyzes the text data and generates an AI teacher model.

[1712] Specific operation: The generative AI model analyzes text data and generates an AI teacher model specialized for specific teaching materials.

[1713] Input: Text data received by the generation AI.

[1714] Output: The generated AI teacher model.

[1715] Step 7:

[1716] The server stores the generated AI teacher model in a database and associates it with the teaching material ID.

[1717] Specific operation: The server stores the generated AI teacher model in a database and associates the model with a specific teaching material ID.

[1718] Input: Generated AI teacher model.

[1719] Output: An AI teacher model and its associated information stored in a database.

[1720] Step 8:

[1721] The user requests to start a learning session.

[1722] Specific operation: The user sends a request including the learning material ID he / she wants to use from the learning session start interface of the terminal.

[1723] Input: A user request to start a learning session.

[1724] Output: The request sent to the terminal.

[1725] Step 9:

[1726] The device sends the request and the teaching material ID to the server.

[1727] Specific operation: The terminal sends a learning session start request and the learning material ID to the server.

[1728] Input: A learning session start request from the user and the learning material ID.

[1729] Output: The request sent to the server and the learning material ID.

[1730] Step 10:

[1731] The server receives the request, reads the AI ​​teacher model from the database, and loads it into memory.

[1732] Specific operation: The server reads the corresponding AI teacher model from the database and loads it into memory.

[1733] Input: The request sent to the server and the learning material ID.

[1734] Output: An AI teacher model loaded into memory.

[1735] Step 11:

[1736] The user enters questions and answers into the interface and submits them.

[1737] Specific operation: The user enters a question and answer into the interface and presses the submit button.

[1738] Input: The user's question and answer.

[1739] Output: Information typed into the terminal.

[1740] Step 12:

[1741] The terminal sends the input information to the server.

[1742] Specific operation: The device sends the entered questions and answers to the server.

[1743] Input: Questions and answers typed into the device.

[1744] Output: The question and answer sent to the server.

[1745] Step 13:

[1746] The server inputs questions and answers into an AI teacher model, which generates answers and explanations.

[1747] Specific operation: The server inputs the received questions and answers into the AI ​​teacher model and receives the generated answers and explanations.

[1748] Input: The question and answer sent to the server.

[1749] Output: Answers and explanations generated by the AI ​​teacher model.

[1750] Step 14:

[1751] The server sends the generated answers and explanations to the terminal.

[1752] Specific operation: The server sends the generated answers and explanations to the terminal, where they are displayed on the user's interface.

[1753] Input: Answers and explanations generated by an AI teacher model.

[1754] Output: Answers and explanations sent to your device.

[1755] Step 15:

[1756] The device uses a device to collect emotion data and transmits the data to a server.

[1757] Specific operation: The device uses sensors, microphones, cameras, etc. to collect emotional data such as the user's facial expressions and tone of voice, and sends it to the server.

[1758] Input: Collected emotion data.

[1759] Output: Emotion data sent to the server.

[1760] Step 16:

[1761] The emotion engine analyzes the data and detects the user's emotional state.

[1762] Specific operation: The emotion engine analyzes the received emotion data and detects the user's current emotional state.

[1763] Input: Emotion data sent to the server.

[1764] Output: Parsed emotional state.

[1765] Step 17:

[1766] The server adjusts the response of the AI ​​teacher model based on the emotional data.

[1767] Specific operation: The server adjusts the response of the AI ​​teacher model based on the emotion data sent from the emotion engine. For example, if the user is feeling stressed, it will provide a gentle commentary.

[1768] Input: Parsed emotional state.

[1769] Output: The adjusted response.

[1770] Step 18:

[1771] The emotion engine continuously monitors data and provides feedback and suggests breaks.

[1772] Specific behavior: The emotion engine continuously monitors the user's emotional state and provides feedback and suggests breaks at appropriate times.

[1773] Input: Continuously collected emotion data.

[1774] Output: Feedback and break suggestions.

[1775] (Application example 2)

[1776] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1777] Conventional educational support systems do not provide feedback that takes into account the learner's individual emotional state regarding their learning progress, making it difficult to provide effective learning support. Furthermore, particularly in new applications such as virtual stores, there is a lack of methods for analyzing users' purchasing motivation and interest in real time and adjusting the dialogue accordingly.

[1778] The specific processing by the specific processing unit 290 of the da...

Claims

1. a means for inputting the contents of the teaching material; A means for generating an AI teacher specialized in a specific teaching material based on the content of the input teaching material; A means for managing interactions with learners and providing answers, explanations, and hints through said AI teacher; A system including a means for indicating reference locations in teaching materials based on the answers and explanations.

2. The system of claim 1 , further comprising means for managing a learning session after the AI ​​teacher is created.

3. The system of claim 1 further comprising means for responding to additional questions or requests from the learner based on the answers and explanations generated by the AI ​​teacher.

Images