System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving content to be learned from a user; means for analyzing content of the received content; means for automatically generating examination questions based on the analyzed content; and means for providing the generated examination questions to the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Today's learners use a variety of learning content for self-study, but there are often no commercially available question sets that fully correspond to that content. This makes it difficult to generate questions to check learning outcomes based on specific textbooks, websites, audio materials, or video materials. Furthermore, when there is a lack of question sets specific to the learning content, or when commercially available question sets are inappropriate, it can be difficult for learners to progress efficiently. [Means for solving the problem]

[0005] To solve the above-mentioned problems, the present invention provides a system that receives study content from a user, analyzes the received content, automatically generates test questions based on the analyzed content, and provides the generated test questions to the user. Specifically, the system can handle received content such as text data, web page data, audio data, and video data, and has a means for generating questions in the optimal format (fill-in-the-blank, multiple-choice, or written) based on the content. It also has a function for displaying the generated questions on the user's terminal and allowing them to check their answers, as well as a means for downloading and printing the questions in PDF format, allowing learners to efficiently progress through their studies in the way they want.

[0006] "Content to be studied" refers to information sources such as textbooks, websites, audio files, and video files that users use for self-study.

[0007] A "user" is a person who uses the system to generate test questions about the content they wish to study.

[0008] The "receiving means" is a part that has the function of transferring the content to be studied from the user terminal to the system server and storing it.

[0009] The "analysis means" is a part that has the function of understanding the content of the received content using technologies such as text extraction and voice recognition, and extracting important information.

[0010] The "means for automatically generating test questions" is a part that has the function of creating questions in an appropriate format (for example, fill-in-the-blank questions, multiple-choice questions, or essay questions) based on the content of the analyzed content.

[0011] The "means for providing" is a part that has the function of allowing users to view, answer, and grade the generated test questions.

[0012] A "fill-in-the-blank question" is a question in which some of the questions are left blank and the user is expected to fill in the blank with appropriate information.

[0013] A multiple choice question is a question in which multiple options are presented and the correct answer must be selected from the options.

[0014] An "essay question" is a question in which the user answers the question by freely writing in their own words.

[0015] "User terminal" means a device (e.g., PC, smartphone, tablet) used by a user to access the system.

[0016] A "database" is a storage system within the system that stores and manages generated test questions and analyzed content.

[0017] "Means for downloading and printing in PDF format" refers to the portion of the document that allows a user to download the generated problem set as a high-quality PDF file and make the PDF file printable. [Brief explanation of the drawings]

[0018] [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

[0019] 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.

[0020] First, the terms used in the following description will be explained.

[0021] 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).

[0022] 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.

[0023] 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.

[0024] 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.

[0025] 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."

[0026] [First embodiment]

[0027] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0028] 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.

[0029] 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).

[0030] 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.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0035] 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.

[0036] 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.

[0037] 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.

[0038] 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."

[0039] The present invention provides a system for efficiently generating and providing test questions from a learner's learning content. Specific embodiments of this system will be described below.

[0040] System Configuration

[0041] The system mainly consists of the following components:

[0042] User terminal

[0043] server

[0044] User terminal

[0045] A user terminal is a device on which a user uploads learning content, views generated test questions, and answers them. Examples of such devices include personal computers (PCs), smartphones, and tablets.

[0046] server

[0047] The server plays a central role in receiving learning content sent from user devices, analyzing it, generating test questions, and providing them to users. This server is equipped with powerful AI models, databases, text analysis, speech recognition, and other systems.

[0048] User operation (input)

[0049] Users upload the content they want to study (e.g., textbooks, websites, audio files, video files, etc.) from their terminals to the system. At this time, users can select the content and send it to the server in a specific format (e.g., PDF, URL, MP3, MP4, etc.).

[0050] Receiving and storing content

[0051] The server receives the learning content sent from the user's device and saves it in the designated storage. Within this storage, the format of the received file (e.g. PDF, audio, video, etc.) is automatically identified and the appropriate processing is performed.

[0052] Preparing and running content analysis

[0053] The server selects a different parsing algorithm for each file format received: for example, for a PDF file, the server uses a PDF text extraction algorithm to extract text, and for an audio file, it uses a speech recognition API to convert the audio to text.

[0054] Content Analysis

[0055] The AI ​​model on the server analyzes the extracted text to identify important keywords, themes, and key points. This analysis process uses natural language processing techniques to understand sentence structure and meaning.

[0056] Test question generation

[0057] Based on the analysis results, the server generates test questions in various formats (e.g., fill-in-the-blank, multiple-choice, essay, etc.). For example, questions using specific formulas and theorems are generated from mathematics textbooks. The content of the options and fill-in-the-blanks is also automatically generated by the AI ​​model.

[0058] Storage and provision of exam questions

[0059] The generated test questions are stored in a database on the server and made accessible to users. User devices can access and display these question sets through a web app or dedicated app.

[0060] Answers and feedback

[0061] The user answers the generated questions on their device. Once the answer is complete, the server automatically scores the answer and provides immediate feedback to the user. The user can then check the accuracy rate and explanations to assess their own understanding.

[0062] Generate print data

[0063] The system also provides a function for users to download and print the generated problem set in PDF format. When a user requests a printable PDF, the server generates a PDF with high-quality layout and provides it to the user.

[0064] Specific examples

[0065] For example, the following shows the process when using a high school mathematics textbook PDF as learning content:

[0066] 1. User operation: The user uploads a PDF file of a mathematics textbook from the user's terminal.

[0067] 2. Server processing: The server receives the PDF file and saves it in storage.

[0068] 3. Content analysis: Using a PDF text extraction algorithm, the text of the textbook is extracted, and the AI ​​model analyzes the content.

[0069] 4. Question Generation: Generate fill-in-the-blank and multiple-choice questions based on important mathematical formulas and theorems.

[0070] 5. Submit and Answer: The user answers the generated questions in the web app, and the server instantly scores and provides feedback.

[0071] 6. Printable data: When requested by the user, the server converts the problem set into PDF format and provides a download link.

[0072] In this way, this system is a powerful tool for efficiently generating test questions from the learning content desired by the user and for improving learning effectiveness.

[0073] The processing flow will be explained below.

[0074] Step 1:

[0075] The user accesses the learning content upload screen on their device (PC, smartphone, etc.). The user selects and uploads the content they want to study (e.g., textbook PDF, website URL, audio file, video file, etc.).

[0076] Step 2:

[0077] The server receives the content file sent from the user terminal, processes the received file securely, and saves it in the designated storage within the system.

[0078] Step 3:

[0079] The server determines the format of the received file (PDF, URL, audio, video, etc.) and prepares to select an analysis algorithm depending on the file format.

[0080] Step 4:

[0081] The server applies a parsing algorithm based on the file format.

[0082] For PDF: The server uses a PDF text extraction library (e.g., PyMuPDF) to extract text from the PDF.

[0083] For a URL: The server uses a web scraping tool (e.g., BeautifulSoup, Scrapy) to extract the text of the web page.

[0084] For audio files: The server uses a speech recognition API (e.g., Google Cloud Speech-to-Text) to convert the audio to text.

[0085] For video files: The server extracts the audio portion of the video and converts it to text using a speech recognition API.

[0086] Step 5:

[0087] The AI ​​model on the server analyzes the extracted text, using natural language processing techniques to identify important information in the content (keywords, themes, key points, etc.), and builds analytical data for generating questions.

[0088] Step 6:

[0089] The server automatically generates test questions based on the analyzed data. The generated questions are in the following formats:

[0090] Fill-in-the-blank questions: Questions that leave specific keywords or phrases blank and require users to fill them in.

[0091] Multiple choice questions: Questions that provide multiple options and require the user to select the correct answer.

[0092] Essay questions: Questions that require users to write freely.

[0093] Step 7:

[0094] The server stores the generated exam questions in a database, where they are prepared for user access.

[0095] Step 8:

[0096] Users access the generated exam questions on their device, answer them through a web or app interface, and submit their answers once they have completed the questions.

[0097] Step 9:

[0098] The server receives the user's answers and automatically scores them. The server immediately returns the results and feedback to the user, who can then check the answers and evaluate their own understanding.

[0099] Step 10:

[0100] If the user wishes, the server will make the generated test question set available for download in PDF format. When the user clicks the "Generate Printable PDF" button, the server will convert the question set into a high-quality PDF format and provide a download link.

[0101] Step 11:

[0102] Users can download the PDF file from the download link and print it out, allowing them to continue their studies offline.

[0103] The above are the specific processing steps for the system to automatically generate and provide test questions from learning content.

[0104] Example 1

[0105] 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."

[0106] Conventional learning systems require users to manually generate test questions from learning content, which is time-consuming and labor-intensive, and limits the quality and variety of questions generated. Furthermore, when content formats differ, it is difficult to select an appropriate analysis method for each, making efficient question generation impossible. The present invention aims to solve these problems and provide an environment in which users can study efficiently and effectively.

[0107] 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.

[0108] In this invention, the server includes means for receiving study content from a user, means for automatically determining the format of the received content and selecting an appropriate analysis algorithm, means for analyzing the content based on the selected analysis algorithm, means for automatically generating test questions based on the analyzed content, means for providing the generated test questions to the user, means for receiving the user's answers to the test questions and performing automatic grading and feedback, and means for providing the generated test questions in a printable PDF format. This allows users to efficiently generate and answer test questions from content in a variety of formats and have the results evaluated.

[0109] "Content to be studied" refers to information formats such as text data, web page data, audio data, and video data provided by users for study purposes.

[0110] "User" refers to a person who uses the learning system to upload learning content and answer generated exam questions.

[0111] "Means for receiving" refers to the function by which the system receives and records learning content sent by a user.

[0112] "Means for automatically identifying the format and selecting the appropriate analysis algorithm" refers to technology for identifying the format of received content and automatically selecting the analysis method that is optimal for that format.

[0113] "Means of analysis" refers to the function of analyzing the content of learning content using selected algorithms and extracting important information.

[0114] "Means for automatically generating test questions" refers to a function that automatically creates test questions such as fill-in-the-blank questions, multiple-choice questions, and essay questions based on the analyzed content.

[0115] "Means for providing to the user" refers to a function for presenting the generated test questions in a form that can be used by the user.

[0116] "Means for receiving answers and automatically scoring and providing feedback" refers to the function that allows the system to receive the answers provided by the user, determine whether they are correct, and immediately return evaluation results and advice.

[0117] "Means for providing in printable PDF format" refers to the ability to generate generated exam questions as high-quality PDF documents that users can download and print.

[0118] This invention is a system that allows learners to efficiently generate and provide test questions from any learning content. This system is mainly composed of a user terminal and a server.

[0119] User terminal

[0120] A user terminal is a device on which a user uploads learning content, views generated test questions, and answers them. This includes personal computers (PCs), smartphones, tablets, etc. Users access the system using a browser or dedicated application on these devices.

[0121] server

[0122] The server plays a central role in receiving learning content sent from user devices, analyzing it, and generating and providing test questions. The server incorporates powerful AI models, databases, text analysis, speech recognition, and other systems. Specifically, the following software and web services are used:

[0123] PDF Text Extraction Algorithm

[0124] Speech Recognition API

[0125] Natural language processing technology

[0126] Uploading learning content

[0127] Users upload the content they want to study (e.g., textbooks, websites, audio files, video files, etc.) from their terminals to the system. At this time, users can send this content to the server in a specific format (e.g., PDF, URL, MP3, MP4, etc.).

[0128] Receiving and storing content

[0129] The server receives the learning content sent from the user's device and stores it in storage. Within this storage, the format of the received file (e.g., PDF, audio, video, etc.) is automatically identified and the appropriate processing is performed.

[0130] Preparing and running content analysis

[0131] The server selects a different parsing algorithm for each file format received: for example, for a PDF file, the server uses a PDF text extraction algorithm to extract text, and for an audio file, it uses a speech recognition API to convert the audio to text.

[0132] Content Analysis

[0133] The AI ​​model on the server analyzes the extracted text to identify important keywords, themes, and key points. This analysis process uses natural language processing techniques to understand sentence structure and meaning.

[0134] Test question generation

[0135] Based on the analysis results, the server generates test questions in various formats (e.g., fill-in-the-blank, multiple-choice, essay, etc.). For example, questions using specific formulas and theorems are generated from mathematics textbooks. The content of the options and fill-in-the-blanks is also automatically generated by the AI ​​model.

[0136] Storage and provision of exam questions

[0137] The generated test questions are stored in a database on the server and made accessible to users. User devices can access and display these question sets through a web app or dedicated app.

[0138] Answers and feedback

[0139] The user answers the generated questions on their device. Once the answer is complete, the server automatically scores the answer and provides immediate feedback to the user. The user can then check the accuracy rate and explanations to assess their level of understanding.

[0140] Generate print data

[0141] The system also provides a function for users to download and print the generated problem set in PDF format. When a user requests a printable PDF, the server generates a PDF with high-quality layout and provides it to the user.

[0142] Specific examples

[0143] For example, the following shows the process when using a high school mathematics textbook PDF as learning content:

[0144] 1. User operation: The user uploads a PDF file of a mathematics textbook from the user's terminal.

[0145] 2. Server processing: The server receives the PDF file and saves it in storage.

[0146] 3. Content analysis: Using a PDF text extraction algorithm, the text of the textbook is extracted, and the AI ​​model analyzes the content.

[0147] 4. Question Generation: Generate fill-in-the-blank and multiple-choice questions based on important mathematical formulas and theorems. Prompt: "Generate exam questions based on the contents of this mathematics textbook."

[0148] 5. Submit and Answer: The user answers the generated questions in the web app, and the server instantly scores and provides feedback.

[0149] 6. Printable data: When requested by the user, the server converts the problem set into PDF format and provides a download link.

[0150] This system is a powerful tool for efficiently generating test questions from the learning content desired by the user, thereby enhancing learning effectiveness.

[0151] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0152] Step 1:

[0153] The user selects and uploads learning content from the user device. The learning content file (e.g., PDF, URL, MP3, MP4, etc.) is provided as input. The user device then sends the selected file to the server. Specifically, the user clicks the "Upload" button in a browser or dedicated application and selects the file.

[0154] Step 2:

[0155] The server receives learning content sent from the user's device. The file sent by the user is provided as input. After receiving the file, the server saves it in storage and automatically determines the file format (e.g., PDF, URL, MP3, MP4). Specifically, it analyzes the received file name and format, and sorts and saves it in the specified folder.

[0156] Step 3:

[0157] The server selects the appropriate analysis algorithm based on the format of the saved file. The saved file format is provided as input. The server selects the PDF text extraction algorithm for PDF files and the speech recognition API for audio files. Specifically, it uses conditional branching to determine the file format and calls the corresponding analysis algorithm.

[0158] Step 4:

[0159] The server analyzes the content using the selected algorithm. The stored content and the selected analysis algorithm are provided as input. In the case of a PDF file, the server extracts text using a PDF text extraction algorithm, and in the case of an audio file, it converts the audio to text using a speech recognition API. Specifically, the analysis engine extracts information from the file and generates text data.

[0160] Step 5:

[0161] The AI ​​model on the server analyzes the extracted text and identifies important keywords and themes. The extracted text is provided as input. The server uses natural language processing techniques to extract keywords and identify key points. Specifically, the text analysis engine tokenizes the text and extracts themes and key information.

[0162] Step 6:

[0163] The server automatically generates test questions based on the analysis results. The key points and keywords of the analyzed text are provided as input. The server uses the generative AI model to create a prompt sentence and generates test questions based on that. Specifically, the server inputs the prompt sentence into the generative AI model and obtains the generated question sentence.

[0164] Step 7:

[0165] The server saves the generated test questions in a database so that they can be provided to users. The generated test questions are provided as input. The server inserts the generated question data into the database and provides a question set in response to a user request. Specifically, the server stores the question data in the database and provides it to the user's device via a WebAPI.

[0166] Step 8:

[0167] The user answers the generated questions using the user terminal. The test questions provided as input are displayed on the user terminal. After answering, the user sends the answers to the server. Specifically, the user answers the questions on the application and clicks the "Submit" button.

[0168] Step 9:

[0169] The server receives the answers submitted by the user, automatically scores them, and returns the feedback to the user. The user's answer data is provided as input. The server uses a scoring algorithm to determine whether the answer is correct and generates a result. Specifically, the server inputs the answer data into the scoring algorithm, generates a score result, and displays it to the user in real time.

[0170] Step 10:

[0171] When a user requests a printable PDF, the server generates a PDF with a high-quality layout and provides it to the user. The user's request is provided as input. The server converts the problem set into PDF format using a PDF generation tool and generates a download link. Specifically, the server starts the PDF generation process in response to the request and displays a link to the generated PDF file in the web application.

[0172] In this way, through a series of processing steps, the system provides the user with the functionality to efficiently generate test questions from their own learning content, answer them, and evaluate the results.

[0173] (Application example 1)

[0174] 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."

[0175] Conventional learning systems have had problems such as the difficulty for learners to efficiently generate test questions from any learning content, and the inability to properly analyze the format of uploaded content and automatically generate test questions, resulting in a significant lack of user convenience. Another issue is the difficulty in effectively providing the generated test questions to users and providing immediate feedback.

[0176] 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.

[0177] In this invention, the server includes means for receiving study content from a user, means for analyzing the received content, means for automatically generating test questions based on the analyzed content, means for providing the generated test questions to the user, means for extracting text according to the format of the received content, means for generating prompt sentences from the extracted text, means for using the prompt sentences to generate test questions using a generative AI model, and means for providing the generated test questions to the user via the web and accepting answers. This allows learners to efficiently generate test questions from any study content, automates analysis and text extraction, and provides the generated test questions instantly for rapid feedback.

[0178] "Content to be studied" refers to information such as text data, audio data, and video data that is uploaded by a user for study purposes.

[0179] "Means for receiving from users" refers to the interface and processing functions that allow users to upload learning content to the system.

[0180] "Content analysis means" refers to analytical techniques used to understand uploaded content and identify important keywords and themes.

[0181] "Means for automatically generating test questions" refers to algorithms or generative AI models that create appropriate test questions based on the analyzed content.

[0182] The "means for providing test questions to the user" refers to a function for displaying or transmitting the generated test questions to the user's device.

[0183] The "means for extracting text according to the format of received content" refers to a technology for extracting text information in a manner appropriate for the format of received content, such as text, audio, or video.

[0184] The "means for generating prompt sentences" refers to a technology that creates input sentences (prompt sentences) for the generative AI model to generate test questions based on the extracted text.

[0185] A "generative AI model" is an artificial intelligence model for generating test questions based on a given prompt.

[0186] "Means for providing and accepting answers via the web" refers to the interface and processing functions that allow users to access test questions and submit answers via a web browser or dedicated application.

[0187] To implement this invention, a system including the following elements is required: a user terminal, a server, a powerful AI model, text analysis, speech recognition, etc.

[0188] System Configuration

[0189] The system mainly consists of the following components:

[0190] User terminal: A device on which a user uploads learning content, views generated test questions, and answers them. Examples include smartphones, tablets, and personal computers (PCs).

[0191] Server: Plays a central role in receiving learning content sent from user devices, analyzing it, generating and providing test questions. This server is equipped with powerful AI models, databases, text analysis, speech recognition, and other systems.

[0192] User operations

[0193] Users can upload any learning content (e.g., textbooks, websites, audio files, video files, etc.) from their devices. At this time, users can select these contents and send them to the server in a specific format (e.g., PDF, URL, MP3, MP4, etc.).

[0194] Server Processing

[0195] 1. Receiving and storing content

[0196] The server receives the learning content sent from the user's device and saves it in the designated storage. Within this storage, the format of the received file is automatically identified and the appropriate processing is performed.

[0197] 2. Preparing and running content analysis

[0198] The server selects different parsing algorithms for each file format received: for example, for PDF files, the server performs PDF text extraction; for audio files, it uses speech recognition technology to convert speech to text.

[0199] 3. Content Analysis

[0200] The AI ​​model on the server analyzes the extracted text to identify important keywords and themes, using natural language processing techniques to understand sentence structure and meaning.

[0201] 4. Prompt generation

[0202] Based on the extracted text, the AI ​​model creates an input sentence (prompt sentence) for generating test questions, for example, in the format "Please create a test question from the following content: [extracted text]."

[0203] 5. Test Question Generation

[0204] Using the prompt, a generative AI model generates test questions, which can range from fill-in-the-blank questions, multiple-choice questions, and essay questions.

[0205] 6. Provision of test questions and answers

[0206] The generated test questions are stored in a database on the server and made accessible to users, who can then answer them via a web app or a dedicated app.

[0207] Specific examples

[0208] For example, if you use a high school mathematics textbook PDF as learning content, the process is as follows:

[0209] 1. User operation: The user uploads a PDF file of a mathematics textbook from the user terminal to the server.

[0210] 2. Server processing:

[0211] Receive PDF files and save them to storage.

[0212] Using PDF text extraction technology, textbook content is extracted.

[0213] The extracted text is analyzed by an AI model.

[0214] 3. Prompt generation: Based on the extracted text, a prompt is generated: "Please create test questions from the following content: [Textbook content]."

[0215] 4. Test question generation: Using a generative AI model, test questions are generated based on the analysis results.

[0216] 5. Submit and answer: The user answers the generated test questions in the web app, and the server immediately scores and provides feedback.

[0217] Example prompt sentence:

[0218] "Create an exam question from the following content: A description of the growth of morning glories. Morning glories are summer flowers and require watering and sunlight."

[0219] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0220] Step 1:

[0221] Users upload learning content

[0222] Users upload the content they want to study (e.g., PDF files, audio files, video files, etc.) from their own devices to the system. The input here is the study content selected by the user, and the output is the transmission of this content to the server.

[0223] Step 2:

[0224] The server receives and stores the learning content.

[0225] The server receives the learning content sent from the user's device and saves it in the specified storage. The input here is the content file sent by the user, and the output is the content file saved in the storage. At this time, the content format (e.g., PDF, MP3, MP4, etc.) is also automatically determined.

[0226] Step 3:

[0227] The server determines the content type and selects the appropriate parsing algorithm.

[0228] The server selects an appropriate parsing algorithm depending on the format of the received content, for example, a text extraction algorithm for a PDF file, or a speech recognition algorithm for an audio file. The input here is the saved content file and its format information, and the output is the selected parsing algorithm.

[0229] Step 4:

[0230] The server parses the content and extracts the text

[0231] The server analyzes the content using the selected analysis algorithm and extracts the required text data, for example extracting text from a PDF file or converting audio to text from an audio file. The input here is the content file and the selected analysis algorithm, and the output is the extracted text data.

[0232] Step 5:

[0233] The server generates a prompt

[0234] The server creates a prompt based on the extracted text data for the generative AI model to generate questions. An example of a prompt is: "Please create a test question from the following content: [extracted text]". Here, the input is the extracted text data, and the output is the generated prompt.

[0235] Step 6:

[0236] The server generates test questions using the generative AI model

[0237] The server generates test questions by inputting the generative AI model using the generated prompt sentences, where the inputs are the prompt sentences and the generative AI model, and the output is the generated test questions.

[0238] Step 7:

[0239] The server provides the test questions to the user.

[0240] The server stores the generated test questions in a database and makes them accessible to users. Users can access and answer these questions through a web application or a dedicated application. The input here is the generated test questions, and the output is the test questions provided via the web.

[0241] Step 8:

[0242] The user answers the questions and the server provides feedback

[0243] The user answers the provided test questions and sends the answers to the server, which instantly marks the answers and provides feedback to the user. The input here is the user's answer, and the output is the marking results and feedback.

[0244] 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.

[0245] This invention is a system that generates test questions from a user's study content and provides personalized questions according to the user's emotions by combining it with an emotion engine. Specific embodiments of this system are described below.

[0246] System Configuration

[0247] The system mainly consists of the following components:

[0248] User terminal

[0249] server

[0250] Emotion Engine

[0251] User terminal

[0252] A user device refers to the device on which a user uploads learning content, views generated test questions, and answers them. This includes PCs, smartphones, tablets, etc. The device is also equipped with a camera and microphone, and the emotion engine uses these devices to analyze the user's emotions.

[0253] server

[0254] The server plays a central role in receiving learning content sent from user devices, analyzing it, generating test questions, and providing them to users. This server is equipped with systems such as AI models, databases, text analysis, and speech recognition.

[0255] Emotion Engine

[0256] The emotion engine is a technology that recognizes emotions from the user's facial expressions and tone of voice, allowing it to analyze the user's emotions while answering test questions and provide appropriate feedback and adjust the questions in real time.

[0257] User operation (input)

[0258] Users upload the content they want to study (textbooks, websites, audio files, video files, etc.) from their devices. When uploading content, users also set up their devices to allow the use of the emotion engine.

[0259] Receiving and storing content

[0260] The server receives the content file sent from the user's device and saves it in the specified storage. The format of the received file (PDF, URL, audio, video, etc.) is automatically determined and the appropriate processing is performed.

[0261] Preparing and running content analysis

[0262] The server selects a different parsing algorithm for each file format received: for example, for PDF files, it uses a PDF text extraction algorithm to extract text, and for audio files, it uses a speech recognition API to convert speech to text.

[0263] Content Analysis

[0264] An AI model on the server analyzes the extracted text to identify important keywords, themes, and key points. This analysis process uses natural language processing techniques to understand sentence structure and meaning.

[0265] Test question generation

[0266] Based on the analysis results, the server generates test questions in various formats (fill-in-the-blank, multiple choice, essay, etc.), so that the generated questions are in a format that is appropriate for the user's learning content.

[0267] Storage and provision of exam questions

[0268] The server stores the generated test questions in a database and makes them available for users to access. User devices can access and display these question sets through a web app or a dedicated app.

[0269] Emotion Recognition and Feedback

[0270] As users answer questions, the emotion engine uses camera footage and audio data to recognize their emotions. Based on the recognized emotions, the server adjusts the difficulty of the questions and provides timely feedback and support messages.

[0271] Answers and feedback

[0272] When a user answers a question, the server automatically scores the answer and provides immediate feedback, taking into account emotional data recognized by the emotion engine, according to the user's level of understanding and state.

[0273] Generate print data

[0274] The system also provides users with the ability to download and print the generated problem sets in PDF format if they wish, allowing them to study offline.

[0275] Specific examples

[0276] The steps for using a high school mathematics textbook PDF as learning content are as follows:

[0277] 1. User operation: The user uploads a PDF file of a mathematics textbook. The user allows the emotion engine to be used.

[0278] 2. Server processing: Receive the PDF file, save it in storage, extract the text, and analyze it using the AI ​​model.

[0279] 3. Question generation: Based on the analyzed content, fill-in-the-blank and multiple-choice questions are generated.

[0280] 4. Submit and Answer: The user answers the generated questions in the web app.

[0281] 5. Emotion recognition and feedback: The emotion engine analyzes the user's facial expressions and voice and adjusts feedback and problem difficulty in real time.

[0282] 6. Answer result: Once the answer is completed, the server immediately scores it and provides feedback that takes into account emotional data.

[0283] 7. Printable data: Upon user request, the problem set will be converted into PDF format and a download link will be provided.

[0284] As described above, the present invention efficiently generates test questions from a user's learning content and combines them with an emotion engine to provide a personalized learning experience that responds to the user's emotions.

[0285] The processing flow will be explained below.

[0286] Step 1:

[0287] The user accesses the learning content upload screen on their device (PC, smartphone, etc.). The user selects and uploads the content they want to study (textbook PDF, website URL, audio file, video file, etc.). They also set up permission to use the emotion engine.

[0288] Step 2:

[0289] The server receives the content file sent from the user's device. The received file is safely processed and saved in the designated storage within the system. The emotion engine usage settings are also saved at the same time.

[0290] Step 3:

[0291] The server determines the format of the received file (PDF, URL, audio, video, etc.) and prepares to select an analysis algorithm depending on the file format.

[0292] Step 4:

[0293] The server applies a parsing algorithm based on the file format.

[0294] For PDF: The server uses a PDF text extraction library (e.g., PyMuPDF) to extract text from the PDF.

[0295] For a URL: The server uses a web scraping tool (e.g., BeautifulSoup, Scrapy) to extract the text of the web page.

[0296] For audio files: The server uses a speech recognition API (e.g., Google Cloud Speech-to-Text) to convert the audio to text.

[0297] For video files: The server extracts the audio portion of the video and converts it to text using a speech recognition API.

[0298] Step 5:

[0299] The AI ​​model on the server analyzes the extracted text, using natural language processing techniques to identify important information in the content (keywords, themes, key points, etc.), and builds analytical data for generating questions.

[0300] Step 6:

[0301] The server automatically generates test questions based on the analyzed data. The generated questions are in the following formats:

[0302] Fill-in-the-blank questions: Questions that leave specific keywords or phrases blank and require users to fill them in.

[0303] Multiple choice questions: Questions that provide multiple options and require the user to select the correct answer.

[0304] Essay questions: Questions that require users to write freely.

[0305] Step 7:

[0306] The server stores the generated exam questions in a database, where they are prepared for user access.

[0307] Step 8:

[0308] Users access the test questions generated on their own devices and answer them through a web or app interface. As they begin answering questions, the emotion engine uses camera footage and audio data to recognize the user's emotions in real time.

[0309] Step 9:

[0310] The server receives and analyzes the user's emotional data, and based on that data, adjusts the difficulty of the test questions provided and provides feedback and support messages at appropriate times.

[0311] Step 10:

[0312] Once the user has completed the questions, the server automatically scores the answers and provides immediate feedback, taking into account the emotional data recognized by the emotion engine, providing detailed feedback tailored to the user's level of understanding and state.

[0313] Step 11:

[0314] If the user wishes, the server will make the generated test question set available for download in PDF format. When the user clicks the "Generate Printable PDF" button, the server will convert the question set into a high-quality PDF format and provide a download link.

[0315] Step 12:

[0316] Users can download the PDF file from the download link and print it out, allowing them to continue their studies offline.

[0317] As a concrete example, let's consider the case where a user uses a high school mathematics textbook PDF as learning content:

[0318] 1. User operation: The user uploads a PDF file of a mathematics textbook. The user allows the emotion engine to be used.

[0319] 2. Server processing: The PDF file is received and stored in storage. The text is extracted and analyzed using the AI ​​model.

[0320] 3. Question generation: Based on the analyzed content, fill-in-the-blank and multiple-choice questions are generated.

[0321] 4. Providing and answering: The user answers the generated questions in the web app. The emotion engine analyzes the user's facial expressions and voice.

[0322] 5. Emotion recognition and feedback: The emotion engine recognizes the user's emotions in real time, and the server provides appropriate feedback and problem adjustments.

[0323] 6. Answer result: Once the answer is completed, the server immediately scores it and provides feedback that takes into account emotional data.

[0324] 7. Printable data: Upon user request, the problem set will be converted into PDF format and a download link will be provided.

[0325] As described above, the present invention efficiently generates test questions from learning content and combines them with an emotion engine to provide a personalized learning experience that responds to the user's emotions.

[0326] Example 2

[0327] 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."

[0328] While conventional learning systems can generate test questions from users' learning content, they have the problem of being unable to provide personalized feedback or adjust questions based on the user's emotions. Furthermore, because they do not take into account the user's level of understanding or emotional state, efficient and effective learning is difficult.

[0329] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving study content from a user, means for analyzing the received content, means for automatically generating test questions based on the analyzed content, means for providing the generated test questions to the user, means for recognizing the user's emotions, and means for providing feedback and adjusting the questions in real time based on the recognized emotions. This enables a personalized learning experience that takes into account the user's emotional state.

[0330] "Content to be studied" refers to information provided by a user for study purposes, and includes formats such as text data, web page data, audio data, and video data.

[0331] "Means for receiving from users" refers to interfaces and functions that allow users to upload content to be studied to the system.

[0332] "Means for analyzing the content" refers to technology that mechanically analyzes the received learning content and identifies important keywords, themes, key points, etc.

[0333] "Means for automatically generating test questions" refers to a function that automatically creates fill-in-the-blank questions, multiple-choice questions, essay questions, etc. based on the content of the analyzed content.

[0334] "Means for providing test questions to users" refers to an interface or function that allows users to access and answer the generated test questions.

[0335] "Means for recognizing user emotions" refers to technology that automatically determines emotions from a user's facial expressions and tone of voice.

[0336] "Means for adjusting feedback and questions in real time" refers to a mechanism that instantly and adaptively changes the learning content, difficulty of questions, and feedback based on the recognized user emotions.

[0337] MODE FOR CARRYING OUT THE INVENTION

[0338] This invention is a system that generates test questions from learning content provided by a user and provides feedback according to the user's emotions using an emotion engine. Specific embodiments of this system are described below.

[0339] System Configuration

[0340] The system mainly consists of the following components:

[0341] 1. User Device

[0342] 2. Server

[0343] 3. Emotion Engine

[0344] User terminal

[0345] A user device refers to the device on which a user uploads learning content and views and answers generated test questions. User devices include PCs, smartphones, tablets, etc. These devices are equipped with cameras and microphones, which the emotion engine uses to analyze the user's emotions.

[0346] server

[0347] The server receives learning content sent from the user's device, analyzes the content, generates test questions, and provides them. This server incorporates the following systems:

[0348] AI models (e.g., generative AI models)

[0349] Database (e.g. PostgreSQL)

[0350] Text Analysis Algorithms

[0351] Speech recognition API (e.g., Google Cloud's speech recognition API)

[0352] The server automatically determines the format of the uploaded learning content (PDF, URL, audio, video, etc.) and processes it appropriately.

[0353] Emotion Engine

[0354] The emotion engine is a technology that recognizes emotions from the user's facial expressions and tone of voice. Specifically, it collects camera footage and audio data while the user is answering test questions and analyzes it using technologies such as Microsoft's Emotion API.

[0355] Specific examples

[0356] The steps for using a high school mathematics textbook PDF as learning content are as follows:

[0357] 1. User operation: A user uploads a PDF file of a mathematics textbook. When uploading, the user selects the option to allow the use of the emotion engine.

[0358] 2. Server processing: The server receives the PDF file, stores it in the storage system, extracts the text using a PDF text extraction algorithm, and analyzes it using an AI model.

[0359] 3. Question generation: Based on the analyzed content, fill-in-the-blank questions, multiple-choice questions, etc. are generated.

[0360] 4. Submit and answer: Users answer questions generated through a web app or dedicated app.

[0361] 5. Emotion recognition and feedback: The emotion engine analyzes the user's facial expressions and voice and provides real-time feedback and adjusts the difficulty of the questions.

[0362] 6. Answer result: After the answer is given, the server automatically scores it and provides feedback that takes into account emotional data.

[0363] 7. Print Data Generation: If the user wishes, convert the problem set into PDF format and provide a download link.

[0364] Prompt Sentence Examples

[0365] "I have uploaded a PDF of a high school mathematics textbook. Please extract important keywords and themes from this content and generate fill-in-the-blank and multiple-choice questions based on them. Also, please adjust the difficulty of the questions by analyzing the user's emotions when answering them."

[0366] The above is a specific embodiment of this system, which efficiently generates test questions from the user's learning content and combines it with an emotion engine to provide a personalized learning experience that responds to the user's emotions.

[0367] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0368] Step 1:

[0369] The user device uploads learning content. The user clicks the "Upload Content" button on the web app, and a file selection dialog appears. The user selects the file, checks the option to allow the use of the emotion engine, and clicks the upload button.

[0370] Input: User's learning content file, emotion engine permission settings

[0371] Output: Learning content is sent from the user device to the server.

[0372] Step 2:

[0373] The server receives the learning content and stores it in storage. The server receives a file upload request from the user device and stores the file in temporary storage. The server then moves the stored file to a specific location in the storage system (e.g., Amazon S3).

[0374] Input: Learning content sent from the user's device

[0375] Output: Learning content files saved to storage

[0376] Step 3:

[0377] The server analyzes the file format and selects the appropriate analysis algorithm. The server analyzes the metadata of the uploaded file and determines the file format (PDF, URL, audio, video, etc.).

[0378] Input: Learning content files saved in storage

[0379] Output: Analysis algorithms depending on the file format

[0380] Step 4:

[0381] The server analyzes the content and runs the analysis algorithm it has selected. For PDF files, it uses the PDF text extraction algorithm, and for audio files, it uses a speech recognition API (e.g., Google Cloud's speech recognition API).

[0382] Input: Parsing algorithm depending on the file format

[0383] Output: Extracted text data

[0384] Step 5:

[0385] The AI ​​model on the server analyzes the extracted text data, using natural language processing techniques to identify important keywords, themes, and key points. This analysis is performed using tools such as Amazon Comprehend.

[0386] Input: Extracted text data

[0387] Output: Analyzed keywords, themes, and key points data

[0388] Step 6:

[0389] The server generates test questions based on the analysis results. The server calls up a generative AI model and automatically creates fill-in-the-blank, multiple-choice, and essay questions based on the analyzed keywords and themes.

[0390] Input: Analyzed keywords, themes, and key points data

[0391] Output: Generated test question data

[0392] Step 7:

[0393] The server saves the generated test questions in a database and prepares them for serving to users.The server saves the generated questions in a database (e.g., PostgreSQL) and generates a URL for the question set via an API so that it can be accessed from a web app or dedicated app.

[0394] Input: Generated test question data

[0395] Output: Question data stored in the database, URL for accessing it

[0396] Step 8:

[0397] The user's device displays the test questions, and the user answers them. The user accesses the provided URL and the questions are displayed on the web app. The user answers the questions in the browser.

[0398] Input: The URL the user accessed in the web app

[0399] Output: User's answers to the questions

[0400] Step 9:

[0401] The emotion engine recognizes the user's emotions while they are answering. While the user is entering their answer, the emotion engine periodically collects camera footage and microphone audio, and calls the Emotion API to obtain the analysis results.

[0402] Input: User's camera video and audio data

[0403] Output: Parsed emotion data

[0404] Step 10:

[0405] The server provides real-time feedback and adjusts the difficulty level. The server recalculates the difficulty of the questions based on the emotional data sent from the emotion engine, and displays feedback messages or easy questions as needed.

[0406] Input: Parsed emotion data and user response data

[0407] Output: Adjusted difficulty of the problem, feedback message

[0408] Step 11:

[0409] The server scores the user's answers and provides feedback. When the user submits their answer, the server automatically scores it and provides immediate feedback that takes into account emotional data.

[0410] Input: User's answer data

[0411] Output: Marking results and feedback

[0412] Step 12:

[0413] The server generates the printable data and provides it to the user. If the user wishes, the server converts the generated problem set into PDF format and provides a download link.

[0414] Input: User request

[0415] Output: PDF problem set and download link

[0416] The above are the specific processing steps of this system.

[0417] (Application example 2)

[0418] 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."

[0419] Conventional learning systems have struggled to provide a personalized learning experience that responds to the user's emotions. This has resulted in a lack of appropriate feedback and support based on emotions such as stress and excitement felt during learning, resulting in reduced learning efficiency. Furthermore, conventional learning systems often lacked sufficient support for factory employees to efficiently master specific skills and operations.

[0420] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0421] In this invention, the server

[0422] means for receiving content to be studied from a user;

[0423] means for analyzing the content of the received content;

[0424] means for automatically generating test questions based on the analyzed content;

[0425] means for providing the generated test questions to a user;

[0426] A sentiment analysis means;

[0427] a means for adjusting the difficulty and format of test questions based on user sentiment;

[0428] Includes.

[0429] This allows for a personalized learning experience by understanding employees' emotions in real time while they are studying and providing appropriate feedback. Furthermore, the generated test questions are dynamically adjusted according to the user's emotions, maximizing the effectiveness of learning. Furthermore, using emotion analysis technology can help factory employees efficiently master operations and skills, improving overall production efficiency.

[0430] "Content to be studied" refers to information or learning materials that a user wishes to study, and includes text data, web page data, audio data, video data, and the like.

[0431] "Means for analysis" refers to the technologies and algorithms used to understand the content of received content and extract important parts, including natural language processing and speech recognition.

[0432] "Means for automatically generating test questions" refers to a system that generates questions in a format appropriate for the user's learning content based on analyzed content, and uses an AI model or a generative AI model.

[0433] "Means for providing" refers to devices or applications for providing the generated test questions in a form that users can access, including smartphone apps and web apps.

[0434] "Emotion analysis means" refers to technology or devices for analyzing emotions from a user's facial expressions and voice, and includes image analysis and voice analysis using a camera or microphone.

[0435] "Feedback" refers to instructions or comments provided based on a user's answers and feelings, which can help improve the effectiveness of learning.

[0436] "Means for adjusting difficulty and format" refers to a system that changes the difficulty and format of test questions according to the user's emotions and learning situation, and is dynamically adjusted in real time.

[0437] A "personalized learning experience" aims to maximize learning effectiveness by providing learning methods and content optimized for each user's individual learning situation and emotions.

[0438] "Real-time" means reacting and responding immediately to changes in the user's learning and emotions.

[0439] "Downloadable in PDF format" means that the generated exam questions are saved as electronic files and are available for users to download via the Internet.

[0440] The system necessary to implement this invention mainly comprises the following elements: a server, a user terminal, an emotion analysis means, and an AI model.

[0441] server

[0442] The server plays a central role in analyzing learning content received from users and automatically generating test questions. Specifically, the server receives content uploaded by users, such as PDFs, web pages, audio, and video, and extracts text using appropriate analysis algorithms. The extracted text is analyzed using natural language processing (NLP) techniques to identify important keywords and themes. Test questions are then automatically generated based on the analysis results using an AI model. These test questions are then made available to users for access, viewing, and answering via a web app or dedicated app.

[0443] User terminal

[0444] User devices include PCs, smartphones, tablets, etc., and are used by users to upload learning content and answer generated test questions. User devices are equipped with cameras and microphones, which are used as emotion analysis tools. This makes it possible to analyze emotions in real time from the user's facial expressions and tone of voice, and send the results to the server.

[0445] Emotion analysis means

[0446] Emotion analysis is a technology that uses a camera and microphone to analyze the user's facial expressions and voice. Specifically, it collects camera footage and microphone audio from the user's device and analyzes the user's emotions in real time using the EmotionRecognizer library. This emotional data is sent to a server and used to dynamically adjust the difficulty and format of test questions according to the user's emotions.

[0447] Feedback and Adjustments

[0448] The generated test questions are provided to the user, and as the user answers them, the server monitors the user's state through emotional analysis. For example, if the user is feeling "frustrated," the server will lower the difficulty of the questions based on the emotional data, reducing the user's stress and improving learning efficiency. Similarly, if the user is determined to be "confident," the server will provide questions of increased difficulty. This ensures that the user always has the optimal learning experience.

[0449] Specific examples

[0450] For example, suppose a factory employee uploads "Robot Operation Manual.pdf" to the system as learning content. The system extracts text from the PDF file and analyzes important operating procedures and troubleshooting methods. Based on this analysis, test questions are generated to deepen the employee's understanding and are provided through the smart glasses. When the employee answers the test questions, the smart glasses' camera and microphone are used to analyze their emotions. If the system detects "frustration," it automatically adjusts the difficulty of the questions and provides appropriate feedback.

[0451] Prompt Sentence Examples

[0452] An example of a prompt sentence to input to the generative AI model is:

[0453] "You have an employee learning how to operate a robot. If the analysis reveals that his state is 'frustrated,' create questions that are easy to understand. Conversely, if the analysis reveals that he is 'confident,' create questions that are difficult to understand."

[0454] Instructions include:

[0455] In this way, the present invention is a system that maximizes learning effectiveness by providing a personalized learning experience according to the user's emotions.

[0456] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0457] Step 1:

[0458] Users upload learning content.

[0459] The user uploads the content they want to learn (e.g., "Robot Operation Manual.pdf") from their device to the system. The device receives the file and sends it to the server. The input is the learning content file, and the output is the content saved on the server.

[0460] Step 2:

[0461] The server stores the received content and prepares it for analysis.

[0462] The server saves the uploaded file in storage and automatically detects the file format. For example, if it is a PDF file, the server uses a PDF text extraction algorithm to extract the text. The input is the received content and the output is the data ready to be parsed.

[0463] Step 3:

[0464] The server analyzes the learning content.

[0465] The server uses natural language processing (NLP) techniques to analyze the text and identify important keywords and themes. The specific software used here is an NLP library. The input is prepared text data, and the output is the analyzed keywords and themes.

[0466] Step 4:

[0467] The server generates test questions based on the analysis results.

[0468] The server uses a generative AI model to automatically generate test questions based on the analyzed keywords and themes. The generated questions can be fill-in-the-blank, multiple choice, or essay questions. The input is the analysis results, and the output is the generated test questions.

[0469] Step 5:

[0470] The server provides the generated test questions to the user terminal.

[0471] The server stores the generated test questions in a database and makes them accessible to users. Users can view the questions through a web app or a dedicated app. The input is the generated test question data, and the output is the questions displayed on the user's device.

[0472] Step 6:

[0473] The user answers the test questions.

[0474] The user answers the provided test questions and sends the answers from the terminal to the server. The input is the user's answer data, and the output is the answer data saved on the server.

[0475] Step 7:

[0476] The emotion analysis means analyzes the emotion of the user.

[0477] The system uses the camera and microphone on the user's device to collect facial expressions and voice while the user is answering test questions. It then uses an emotion analysis library (e.g., EmotionRecognizer) to analyze the user's emotions in real time. The input is camera video and audio data, and the output is analyzed emotion data.

[0478] Step 8:

[0479] The server adjusts the difficulty and format of the test questions based on the user's feelings.

[0480] The server adjusts the difficulty of the test questions based on the analyzed emotional data. For example, if the user is judged to be "frustrated," the server sets the difficulty of the questions low, and if the user is judged to be "confident," the server increases the difficulty. The input is emotional data, and the output is the adjusted test questions.

[0481] Step 9:

[0482] The server provides feedback to the user.

[0483] The system provides users with adjusted test questions and appropriate learning feedback in real time, allowing them to receive feedback that is tailored to their learning situation. The input is the adjusted test questions and feedback data, and the output is the feedback displayed on the user's device.

[0484] 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.

[0485] 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.

[0486] 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.

[0487] [Second embodiment]

[0488] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0489] 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.

[0490] 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).

[0491] 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.

[0492] 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.

[0493] 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).

[0494] 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.

[0495] 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.

[0496] 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.

[0497] 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.

[0498] 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.

[0499] 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."

[0500] The present invention provides a system for efficiently generating and providing test questions from a learner's learning content. Specific embodiments of this system will be described below.

[0501] System Configuration

[0502] The system mainly consists of the following components:

[0503] User terminal

[0504] server

[0505] User terminal

[0506] A user terminal is a device on which a user uploads learning content, views generated test questions, and answers them. Examples of such devices include personal computers (PCs), smartphones, and tablets.

[0507] server

[0508] The server plays a central role in receiving learning content sent from user devices, analyzing it, generating test questions, and providing them to users. This server is equipped with powerful AI models, databases, text analysis, speech recognition, and other systems.

[0509] User operation (input)

[0510] Users upload the content they want to study (e.g., textbooks, websites, audio files, video files, etc.) from their terminals to the system. At this time, users can select the content and send it to the server in a specific format (e.g., PDF, URL, MP3, MP4, etc.).

[0511] Receiving and storing content

[0512] The server receives the learning content sent from the user's device and saves it in the designated storage. Within this storage, the format of the received file (e.g. PDF, audio, video, etc.) is automatically identified and the appropriate processing is performed.

[0513] Preparing and running content analysis

[0514] The server selects a different parsing algorithm for each file format received: for example, for a PDF file, the server uses a PDF text extraction algorithm to extract text, and for an audio file, it uses a speech recognition API to convert the audio to text.

[0515] Content Analysis

[0516] The AI ​​model on the server analyzes the extracted text to identify important keywords, themes, and key points. This analysis process uses natural language processing techniques to understand sentence structure and meaning.

[0517] Test question generation

[0518] Based on the analysis results, the server generates test questions in various formats (e.g., fill-in-the-blank, multiple-choice, essay, etc.). For example, questions using specific formulas and theorems are generated from mathematics textbooks. The content of the options and fill-in-the-blanks is also automatically generated by the AI ​​model.

[0519] Storage and provision of exam questions

[0520] The generated test questions are stored in a database on the server and made accessible to users. User devices can access and display these question sets through a web app or dedicated app.

[0521] Answers and feedback

[0522] The user answers the generated questions on their device. Once the answer is complete, the server automatically scores the answer and provides immediate feedback to the user. The user can then check the accuracy rate and explanations to assess their own understanding.

[0523] Generate print data

[0524] The system also provides a function for users to download and print the generated problem set in PDF format. When a user requests a printable PDF, the server generates a PDF with high-quality layout and provides it to the user.

[0525] Specific examples

[0526] For example, the following shows the process when using a high school mathematics textbook PDF as learning content:

[0527] 1. User operation: The user uploads a PDF file of a mathematics textbook from the user's terminal.

[0528] 2. Server processing: The server receives the PDF file and saves it in storage.

[0529] 3. Content analysis: Using a PDF text extraction algorithm, the text of the textbook is extracted, and the AI ​​model analyzes the content.

[0530] 4. Question Generation: Generate fill-in-the-blank and multiple-choice questions based on important mathematical formulas and theorems.

[0531] 5. Submit and Answer: The user answers the generated questions in the web app, and the server instantly scores and provides feedback.

[0532] 6. Printable data: When requested by the user, the server converts the problem set into PDF format and provides a download link.

[0533] In this way, this system is a powerful tool for efficiently generating test questions from the learning content desired by the user and for improving learning effectiveness.

[0534] The processing flow will be explained below.

[0535] Step 1:

[0536] The user accesses the learning content upload screen on their device (PC, smartphone, etc.). The user selects and uploads the content they want to study (e.g., textbook PDF, website URL, audio file, video file, etc.).

[0537] Step 2:

[0538] The server receives the content file sent from the user terminal, processes the received file securely, and saves it in the designated storage within the system.

[0539] Step 3:

[0540] The server determines the format of the received file (PDF, URL, audio, video, etc.) and prepares to select an analysis algorithm depending on the file format.

[0541] Step 4:

[0542] The server applies a parsing algorithm based on the file format.

[0543] For PDF: The server uses a PDF text extraction library (e.g., PyMuPDF) to extract text from the PDF.

[0544] For a URL: The server uses a web scraping tool (e.g., BeautifulSoup, Scrapy) to extract the text of the web page.

[0545] For audio files: The server uses a speech recognition API (e.g., Google Cloud Speech-to-Text) to convert the audio to text.

[0546] For video files: The server extracts the audio portion of the video and converts it to text using a speech recognition API.

[0547] Step 5:

[0548] The AI ​​model on the server analyzes the extracted text, using natural language processing techniques to identify important information in the content (keywords, themes, key points, etc.), and builds analytical data for generating questions.

[0549] Step 6:

[0550] The server automatically generates test questions based on the analyzed data. The generated questions are in the following formats:

[0551] Fill-in-the-blank questions: Questions that leave specific keywords or phrases blank and require users to fill them in.

[0552] Multiple choice questions: Questions that provide multiple options and require the user to select the correct answer.

[0553] Essay questions: Questions that require users to write freely.

[0554] Step 7:

[0555] The server stores the generated exam questions in a database, where they are prepared for user access.

[0556] Step 8:

[0557] Users access the generated exam questions on their device, answer them through a web or app interface, and submit their answers once they have completed the questions.

[0558] Step 9:

[0559] The server receives the user's answers and automatically scores them. The server immediately returns the results and feedback to the user, who can then check the answers and evaluate their own understanding.

[0560] Step 10:

[0561] If the user wishes, the server will make the generated test question set available for download in PDF format. When the user clicks the "Generate Printable PDF" button, the server will convert the question set into a high-quality PDF format and provide a download link.

[0562] Step 11:

[0563] Users can download the PDF file from the download link and print it out, allowing them to continue their studies offline.

[0564] The above are the specific processing steps for the system to automatically generate and provide test questions from learning content.

[0565] Example 1

[0566] 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."

[0567] Conventional learning systems require users to manually generate test questions from learning content, which is time-consuming and labor-intensive, and limits the quality and variety of questions generated. Furthermore, when content formats differ, it is difficult to select an appropriate analysis method for each, making efficient question generation impossible. The present invention aims to solve these problems and provide an environment in which users can study efficiently and effectively.

[0568] 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.

[0569] In this invention, the server includes means for receiving study content from a user, means for automatically determining the format of the received content and selecting an appropriate analysis algorithm, means for analyzing the content based on the selected analysis algorithm, means for automatically generating test questions based on the analyzed content, means for providing the generated test questions to the user, means for receiving the user's answers to the test questions and performing automatic grading and feedback, and means for providing the generated test questions in a printable PDF format. This allows users to efficiently generate and answer test questions from content in a variety of formats and have the results evaluated.

[0570] "Content to be studied" refers to information formats such as text data, web page data, audio data, and video data provided by users for study purposes.

[0571] "User" refers to a person who uses the learning system to upload learning content and answer generated exam questions.

[0572] "Means for receiving" refers to the function by which the system receives and records learning content sent by a user.

[0573] "Means for automatically identifying the format and selecting the appropriate analysis algorithm" refers to technology for identifying the format of received content and automatically selecting the analysis method that is optimal for that format.

[0574] "Means of analysis" refers to the function of analyzing the content of learning content using selected algorithms and extracting important information.

[0575] "Means for automatically generating test questions" refers to a function that automatically creates test questions such as fill-in-the-blank questions, multiple-choice questions, and essay questions based on the analyzed content.

[0576] "Means for providing to the user" refers to a function for presenting the generated test questions in a form that can be used by the user.

[0577] "Means for receiving answers and automatically scoring and providing feedback" refers to the function that allows the system to receive the answers provided by the user, determine whether they are correct, and immediately return evaluation results and advice.

[0578] "Means for providing in printable PDF format" refers to the ability to generate generated exam questions as high-quality PDF documents that users can download and print.

[0579] This invention is a system that allows learners to efficiently generate and provide test questions from any learning content. This system is mainly composed of a user terminal and a server.

[0580] User terminal

[0581] A user terminal is a device on which a user uploads learning content, views generated test questions, and answers them. This includes personal computers (PCs), smartphones, tablets, etc. Users access the system using a browser or dedicated application on these devices.

[0582] server

[0583] The server plays a central role in receiving learning content sent from user devices, analyzing it, and generating and providing test questions. The server incorporates powerful AI models, databases, text analysis, speech recognition, and other systems. Specifically, the following software and web services are used:

[0584] PDF Text Extraction Algorithm

[0585] Speech Recognition API

[0586] Natural language processing technology

[0587] Uploading learning content

[0588] Users upload the content they want to study (e.g., textbooks, websites, audio files, video files, etc.) from their terminals to the system. At this time, users can send this content to the server in a specific format (e.g., PDF, URL, MP3, MP4, etc.).

[0589] Receiving and storing content

[0590] The server receives the learning content sent from the user's device and stores it in storage. Within this storage, the format of the received file (e.g., PDF, audio, video, etc.) is automatically identified and the appropriate processing is performed.

[0591] Preparing and running content analysis

[0592] The server selects a different parsing algorithm for each file format received: for example, for a PDF file, the server uses a PDF text extraction algorithm to extract text, and for an audio file, it uses a speech recognition API to convert the audio to text.

[0593] Content Analysis

[0594] The AI ​​model on the server analyzes the extracted text to identify important keywords, themes, and key points. This analysis process uses natural language processing techniques to understand sentence structure and meaning.

[0595] Test question generation

[0596] Based on the analysis results, the server generates test questions in various formats (e.g., fill-in-the-blank, multiple-choice, essay, etc.). For example, questions using specific formulas and theorems are generated from mathematics textbooks. The content of the options and fill-in-the-blanks is also automatically generated by the AI ​​model.

[0597] Storage and provision of exam questions

[0598] The generated test questions are stored in a database on the server and made accessible to users. User devices can access and display these question sets through a web app or dedicated app.

[0599] Answers and feedback

[0600] The user answers the generated questions on their device. Once the answer is complete, the server automatically scores the answer and provides immediate feedback to the user. The user can then check the accuracy rate and explanations to assess their level of understanding.

[0601] Generate print data

[0602] The system also provides a function for users to download and print the generated problem set in PDF format. When a user requests a printable PDF, the server generates a PDF with high-quality layout and provides it to the user.

[0603] Specific examples

[0604] For example, the following shows the process when using a high school mathematics textbook PDF as learning content:

[0605] 1. User operation: The user uploads a PDF file of a mathematics textbook from the user's terminal.

[0606] 2. Server processing: The server receives the PDF file and saves it in storage.

[0607] 3. Content analysis: Using a PDF text extraction algorithm, the text of the textbook is extracted, and the AI ​​model analyzes the content.

[0608] 4. Question Generation: Generate fill-in-the-blank and multiple-choice questions based on important mathematical formulas and theorems. Prompt: "Generate exam questions based on the contents of this mathematics textbook."

[0609] 5. Submit and Answer: The user answers the generated questions in the web app, and the server instantly scores and provides feedback.

[0610] 6. Printable data: When requested by the user, the server converts the problem set into PDF format and provides a download link.

[0611] This system is a powerful tool for efficiently generating test questions from the learning content desired by the user, thereby enhancing learning effectiveness.

[0612] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0613] Step 1:

[0614] The user selects and uploads learning content from the user device. The learning content file (e.g., PDF, URL, MP3, MP4, etc.) is provided as input. The user device then sends the selected file to the server. Specifically, the user clicks the "Upload" button in a browser or dedicated application and selects the file.

[0615] Step 2:

[0616] The server receives learning content sent from the user's device. The file sent by the user is provided as input. After receiving the file, the server saves it in storage and automatically determines the file format (e.g., PDF, URL, MP3, MP4). Specifically, it analyzes the received file name and format, and sorts and saves it in the specified folder.

[0617] Step 3:

[0618] The server selects the appropriate analysis algorithm based on the format of the saved file. The saved file format is provided as input. The server selects the PDF text extraction algorithm for PDF files and the speech recognition API for audio files. Specifically, it uses conditional branching to determine the file format and calls the corresponding analysis algorithm.

[0619] Step 4:

[0620] The server analyzes the content using the selected algorithm. The stored content and the selected analysis algorithm are provided as input. In the case of a PDF file, the server extracts text using a PDF text extraction algorithm, and in the case of an audio file, it converts the audio to text using a speech recognition API. Specifically, the analysis engine extracts information from the file and generates text data.

[0621] Step 5:

[0622] The AI ​​model on the server analyzes the extracted text and identifies important keywords and themes. The extracted text is provided as input. The server uses natural language processing techniques to extract keywords and identify key points. Specifically, the text analysis engine tokenizes the text and extracts themes and key information.

[0623] Step 6:

[0624] The server automatically generates test questions based on the analysis results. The key points and keywords of the analyzed text are provided as input. The server uses the generative AI model to create a prompt sentence and generates test questions based on that. Specifically, the server inputs the prompt sentence into the generative AI model and obtains the generated question sentence.

[0625] Step 7:

[0626] The server saves the generated test questions in a database so that they can be provided to users. The generated test questions are provided as input. The server inserts the generated question data into the database and provides a question set in response to a user request. Specifically, the server stores the question data in the database and provides it to the user's device via a WebAPI.

[0627] Step 8:

[0628] The user answers the generated questions using the user terminal. The test questions provided as input are displayed on the user terminal. After answering, the user sends the answers to the server. Specifically, the user answers the questions on the application and clicks the "Submit" button.

[0629] Step 9:

[0630] The server receives the answers submitted by the user, automatically scores them, and returns the feedback to the user. The user's answer data is provided as input. The server uses a scoring algorithm to determine whether the answer is correct and generates a result. Specifically, the server inputs the answer data into the scoring algorithm, generates a score result, and displays it to the user in real time.

[0631] Step 10:

[0632] When a user requests a printable PDF, the server generates a PDF with a high-quality layout and provides it to the user. The user's request is provided as input. The server converts the problem set into PDF format using a PDF generation tool and generates a download link. Specifically, the server starts the PDF generation process in response to the request and displays a link to the generated PDF file in the web application.

[0633] In this way, through a series of processing steps, the system provides the user with the functionality to efficiently generate test questions from their own learning content, answer them, and evaluate the results.

[0634] (Application example 1)

[0635] 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."

[0636] Conventional learning systems have had problems such as the difficulty for learners to efficiently generate test questions from any learning content, and the inability to properly analyze the format of uploaded content and automatically generate test questions, resulting in a significant lack of user convenience. Another issue is the difficulty in effectively providing the generated test questions to users and providing immediate feedback.

[0637] 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.

[0638] In this invention, the server includes means for receiving study content from a user, means for analyzing the received content, means for automatically generating test questions based on the analyzed content, means for providing the generated test questions to the user, means for extracting text according to the format of the received content, means for generating prompt sentences from the extracted text, means for using the prompt sentences to generate test questions using a generative AI model, and means for providing the generated test questions to the user via the web and accepting answers. This allows learners to efficiently generate test questions from any study content, automates analysis and text extraction, and provides the generated test questions instantly for rapid feedback.

[0639] "Content to be studied" refers to information such as text data, audio data, and video data that is uploaded by a user for study purposes.

[0640] "Means for receiving from users" refers to the interface and processing functions that allow users to upload learning content to the system.

[0641] "Content analysis means" refers to analytical techniques used to understand uploaded content and identify important keywords and themes.

[0642] "Means for automatically generating test questions" refers to algorithms or generative AI models that create appropriate test questions based on the analyzed content.

[0643] The "means for providing test questions to the user" refers to a function for displaying or transmitting the generated test questions to the user's device.

[0644] The "means for extracting text according to the format of received content" refers to a technology for extracting text information in a manner appropriate for the format of received content, such as text, audio, or video.

[0645] The "means for generating prompt sentences" refers to a technology that creates input sentences (prompt sentences) for the generative AI model to generate test questions based on the extracted text.

[0646] A "generative AI model" is an artificial intelligence model for generating test questions based on a given prompt.

[0647] "Means for providing and accepting answers via the web" refers to the interface and processing functions that allow users to access test questions and submit answers via a web browser or dedicated application.

[0648] To implement this invention, a system including the following elements is required: a user terminal, a server, a powerful AI model, text analysis, speech recognition, etc.

[0649] System Configuration

[0650] The system mainly consists of the following components:

[0651] User terminal: A device on which a user uploads learning content, views generated test questions, and answers them. Examples include smartphones, tablets, and personal computers (PCs).

[0652] Server: Plays a central role in receiving learning content sent from user devices, analyzing it, generating and providing test questions. This server is equipped with powerful AI models, databases, text analysis, speech recognition, and other systems.

[0653] User operations

[0654] Users can upload any learning content (e.g., textbooks, websites, audio files, video files, etc.) from their devices. At this time, users can select these contents and send them to the server in a specific format (e.g., PDF, URL, MP3, MP4, etc.).

[0655] Server Processing

[0656] 1. Receiving and storing content

[0657] The server receives the learning content sent from the user's device and saves it in the designated storage. Within this storage, the format of the received file is automatically identified and the appropriate processing is performed.

[0658] 2. Preparing and running content analysis

[0659] The server selects different parsing algorithms for each file format received: for example, for PDF files, the server performs PDF text extraction; for audio files, it uses speech recognition technology to convert speech to text.

[0660] 3. Content Analysis

[0661] The AI ​​model on the server analyzes the extracted text to identify important keywords and themes, using natural language processing techniques to understand sentence structure and meaning.

[0662] 4. Prompt generation

[0663] Based on the extracted text, the AI ​​model creates an input sentence (prompt sentence) for generating test questions, for example, in the format "Please create a test question from the following content: [extracted text]."

[0664] 5. Test Question Generation

[0665] Using the prompt, a generative AI model generates test questions, which can range from fill-in-the-blank questions, multiple-choice questions, and essay questions.

[0666] 6. Provision of test questions and answers

[0667] The generated test questions are stored in a database on the server and made accessible to users, who can then answer them via a web app or a dedicated app.

[0668] Specific examples

[0669] For example, if you use a high school mathematics textbook PDF as learning content, the process is as follows:

[0670] 1. User operation: The user uploads a PDF file of a mathematics textbook from the user terminal to the server.

[0671] 2. Server processing:

[0672] Receive PDF files and save them to storage.

[0673] Using PDF text extraction technology, textbook content is extracted.

[0674] The extracted text is analyzed by an AI model.

[0675] 3. Prompt generation: Based on the extracted text, a prompt is generated: "Please create test questions from the following content: [Textbook content]."

[0676] 4. Test question generation: Using a generative AI model, test questions are generated based on the analysis results.

[0677] 5. Submit and answer: The user answers the generated test questions in the web app, and the server immediately scores and provides feedback.

[0678] Example prompt sentence:

[0679] "Create an exam question from the following content: A description of the growth of morning glories. Morning glories are summer flowers and require watering and sunlight."

[0680] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0681] Step 1:

[0682] Users upload learning content

[0683] Users upload the content they want to study (e.g., PDF files, audio files, video files, etc.) from their own devices to the system. The input here is the study content selected by the user, and the output is the transmission of this content to the server.

[0684] Step 2:

[0685] The server receives and stores the learning content.

[0686] The server receives the learning content sent from the user's device and saves it in the specified storage. The input here is the content file sent by the user, and the output is the content file saved in the storage. At this time, the content format (e.g., PDF, MP3, MP4, etc.) is also automatically determined.

[0687] Step 3:

[0688] The server determines the content type and selects the appropriate parsing algorithm.

[0689] The server selects an appropriate parsing algorithm depending on the format of the received content, for example, a text extraction algorithm for a PDF file, or a speech recognition algorithm for an audio file. The input here is the saved content file and its format information, and the output is the selected parsing algorithm.

[0690] Step 4:

[0691] The server parses the content and extracts the text

[0692] The server analyzes the content using the selected analysis algorithm and extracts the required text data, for example extracting text from a PDF file or converting audio to text from an audio file. The input here is the content file and the selected analysis algorithm, and the output is the extracted text data.

[0693] Step 5:

[0694] The server generates a prompt

[0695] The server creates a prompt based on the extracted text data for the generative AI model to generate questions. An example of a prompt is: "Please create a test question from the following content: [extracted text]". Here, the input is the extracted text data, and the output is the generated prompt.

[0696] Step 6:

[0697] The server generates test questions using the generative AI model

[0698] The server generates test questions by inputting the generative AI model using the generated prompt sentences, where the inputs are the prompt sentences and the generative AI model, and the output is the generated test questions.

[0699] Step 7:

[0700] The server provides the test questions to the user.

[0701] The server stores the generated test questions in a database and makes them accessible to users. Users can access and answer these questions through a web application or a dedicated application. The input here is the generated test questions, and the output is the test questions provided via the web.

[0702] Step 8:

[0703] The user answers the questions and the server provides feedback

[0704] The user answers the provided test questions and sends the answers to the server, which instantly marks the answers and provides feedback to the user. The input here is the user's answer, and the output is the marking results and feedback.

[0705] 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.

[0706] This invention is a system that generates test questions from a user's study content and provides personalized questions according to the user's emotions by combining it with an emotion engine. Specific embodiments of this system are described below.

[0707] System Configuration

[0708] The system mainly consists of the following components:

[0709] User terminal

[0710] server

[0711] Emotion Engine

[0712] User terminal

[0713] A user device refers to the device on which a user uploads learning content, views generated test questions, and answers them. This includes PCs, smartphones, tablets, etc. The device is also equipped with a camera and microphone, and the emotion engine uses these devices to analyze the user's emotions.

[0714] server

[0715] The server plays a central role in receiving learning content sent from user devices, analyzing it, generating test questions, and providing them to users. This server is equipped with systems such as AI models, databases, text analysis, and speech recognition.

[0716] Emotion Engine

[0717] The emotion engine is a technology that recognizes emotions from the user's facial expressions and tone of voice, allowing it to analyze the user's emotions while answering test questions and provide appropriate feedback and adjust the questions in real time.

[0718] User operation (input)

[0719] Users upload the content they want to study (textbooks, websites, audio files, video files, etc.) from their devices. When uploading content, users also set up their devices to allow the use of the emotion engine.

[0720] Receiving and storing content

[0721] The server receives the content file sent from the user's device and saves it in the specified storage. The format of the received file (PDF, URL, audio, video, etc.) is automatically determined and the appropriate processing is performed.

[0722] Preparing and running content analysis

[0723] The server selects a different parsing algorithm for each file format received: for example, for PDF files, it uses a PDF text extraction algorithm to extract text, and for audio files, it uses a speech recognition API to convert speech to text.

[0724] Content Analysis

[0725] An AI model on the server analyzes the extracted text to identify important keywords, themes, and key points. This analysis process uses natural language processing techniques to understand sentence structure and meaning.

[0726] Test question generation

[0727] Based on the analysis results, the server generates test questions in various formats (fill-in-the-blank, multiple choice, essay, etc.), so that the generated questions are in a format that is appropriate for the user's learning content.

[0728] Storage and provision of exam questions

[0729] The server stores the generated test questions in a database and makes them available for users to access. User devices can access and display these question sets through a web app or a dedicated app.

[0730] Emotion Recognition and Feedback

[0731] As users answer questions, the emotion engine uses camera footage and audio data to recognize their emotions. Based on the recognized emotions, the server adjusts the difficulty of the questions and provides timely feedback and support messages.

[0732] Answers and feedback

[0733] When a user answers a question, the server automatically scores the answer and provides immediate feedback, taking into account emotional data recognized by the emotion engine, according to the user's level of understanding and state.

[0734] Generate print data

[0735] The system also provides users with the ability to download and print the generated problem sets in PDF format if they wish, allowing them to study offline.

[0736] Specific examples

[0737] The steps for using a high school mathematics textbook PDF as learning content are as follows:

[0738] 1. User operation: The user uploads a PDF file of a mathematics textbook. The user allows the emotion engine to be used.

[0739] 2. Server processing: Receive the PDF file, save it in storage, extract the text, and analyze it using the AI ​​model.

[0740] 3. Question generation: Based on the analyzed content, fill-in-the-blank and multiple-choice questions are generated.

[0741] 4. Submit and Answer: The user answers the generated questions in the web app.

[0742] 5. Emotion recognition and feedback: The emotion engine analyzes the user's facial expressions and voice and adjusts feedback and problem difficulty in real time.

[0743] 6. Answer result: Once the answer is completed, the server immediately scores it and provides feedback that takes into account emotional data.

[0744] 7. Printable data: Upon user request, the problem set will be converted into PDF format and a download link will be provided.

[0745] As described above, the present invention efficiently generates test questions from a user's learning content and combines them with an emotion engine to provide a personalized learning experience that responds to the user's emotions.

[0746] The processing flow will be explained below.

[0747] Step 1:

[0748] The user accesses the learning content upload screen on their device (PC, smartphone, etc.). The user selects and uploads the content they want to study (textbook PDF, website URL, audio file, video file, etc.). They also set up permission to use the emotion engine.

[0749] Step 2:

[0750] The server receives the content file sent from the user's device. The received file is safely processed and saved in the designated storage within the system. The emotion engine usage settings are also saved at the same time.

[0751] Step 3:

[0752] The server determines the format of the received file (PDF, URL, audio, video, etc.) and prepares to select an analysis algorithm depending on the file format.

[0753] Step 4:

[0754] The server applies a parsing algorithm based on the file format.

[0755] For PDF: The server uses a PDF text extraction library (e.g., PyMuPDF) to extract text from the PDF.

[0756] For a URL: The server uses a web scraping tool (e.g., BeautifulSoup, Scrapy) to extract the text of the web page.

[0757] For audio files: The server uses a speech recognition API (e.g., Google Cloud Speech-to-Text) to convert the audio to text.

[0758] For video files: The server extracts the audio portion of the video and converts it to text using a speech recognition API.

[0759] Step 5:

[0760] The AI ​​model on the server analyzes the extracted text, using natural language processing techniques to identify important information in the content (keywords, themes, key points, etc.), and builds analytical data for generating questions.

[0761] Step 6:

[0762] The server automatically generates test questions based on the analyzed data. The generated questions are in the following formats:

[0763] Fill-in-the-blank questions: Questions that leave specific keywords or phrases blank and require users to fill them in.

[0764] Multiple choice questions: Questions that provide multiple options and require the user to select the correct answer.

[0765] Essay questions: Questions that require users to write freely.

[0766] Step 7:

[0767] The server stores the generated exam questions in a database, where they are prepared for user access.

[0768] Step 8:

[0769] Users access the test questions generated on their own devices and answer them through a web or app interface. As they begin answering questions, the emotion engine uses camera footage and audio data to recognize the user's emotions in real time.

[0770] Step 9:

[0771] The server receives and analyzes the user's emotional data, and based on that data, adjusts the difficulty of the test questions provided and provides feedback and support messages at appropriate times.

[0772] Step 10:

[0773] Once the user has completed the questions, the server automatically scores the answers and provides immediate feedback, taking into account the emotional data recognized by the emotion engine, providing detailed feedback tailored to the user's level of understanding and state.

[0774] Step 11:

[0775] If the user wishes, the server will make the generated test question set available for download in PDF format. When the user clicks the "Generate Printable PDF" button, the server will convert the question set into a high-quality PDF format and provide a download link.

[0776] Step 12:

[0777] Users can download the PDF file from the download link and print it out, allowing them to continue their studies offline.

[0778] As a concrete example, let's consider the case where a user uses a high school mathematics textbook PDF as learning content:

[0779] 1. User operation: The user uploads a PDF file of a mathematics textbook. The user allows the emotion engine to be used.

[0780] 2. Server processing: The PDF file is received and stored in storage. The text is extracted and analyzed using the AI ​​model.

[0781] 3. Question generation: Based on the analyzed content, fill-in-the-blank and multiple-choice questions are generated.

[0782] 4. Providing and answering: The user answers the generated questions in the web app. The emotion engine analyzes the user's facial expressions and voice.

[0783] 5. Emotion recognition and feedback: The emotion engine recognizes the user's emotions in real time, and the server provides appropriate feedback and problem adjustments.

[0784] 6. Answer result: Once the answer is completed, the server immediately scores it and provides feedback that takes into account emotional data.

[0785] 7. Printable data: Upon user request, the problem set will be converted into PDF format and a download link will be provided.

[0786] As described above, the present invention efficiently generates test questions from learning content and combines them with an emotion engine to provide a personalized learning experience that responds to the user's emotions.

[0787] Example 2

[0788] 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."

[0789] While conventional learning systems can generate test questions from users' learning content, they have the problem of being unable to provide personalized feedback or adjust questions based on the user's emotions. Furthermore, because they do not take into account the user's level of understanding or emotional state, efficient and effective learning is difficult.

[0790] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving study content from a user, means for analyzing the received content, means for automatically generating test questions based on the analyzed content, means for providing the generated test questions to the user, means for recognizing the user's emotions, and means for providing feedback and adjusting the questions in real time based on the recognized emotions. This enables a personalized learning experience that takes into account the user's emotional state.

[0791] "Content to be studied" refers to information provided by a user for study purposes, and includes formats such as text data, web page data, audio data, and video data.

[0792] "Means for receiving from users" refers to interfaces and functions that allow users to upload content to be studied to the system.

[0793] "Means for analyzing the content" refers to technology that mechanically analyzes the received learning content and identifies important keywords, themes, key points, etc.

[0794] "Means for automatically generating test questions" refers to a function that automatically creates fill-in-the-blank questions, multiple-choice questions, essay questions, etc. based on the content of the analyzed content.

[0795] "Means for providing test questions to users" refers to an interface or function that allows users to access and answer the generated test questions.

[0796] "Means for recognizing user emotions" refers to technology that automatically determines emotions from a user's facial expressions and tone of voice.

[0797] "Means for adjusting feedback and questions in real time" refers to a mechanism that instantly and adaptively changes the learning content, difficulty of questions, and feedback based on the recognized user emotions.

[0798] MODE FOR CARRYING OUT THE INVENTION

[0799] This invention is a system that generates test questions from learning content provided by a user and provides feedback according to the user's emotions using an emotion engine. Specific embodiments of this system are described below.

[0800] System Configuration

[0801] The system mainly consists of the following components:

[0802] 1. User Device

[0803] 2. Server

[0804] 3. Emotion Engine

[0805] User terminal

[0806] A user device refers to the device on which a user uploads learning content and views and answers generated test questions. User devices include PCs, smartphones, tablets, etc. These devices are equipped with cameras and microphones, which the emotion engine uses to analyze the user's emotions.

[0807] server

[0808] The server receives learning content sent from the user's device, analyzes the content, generates test questions, and provides them. This server incorporates the following systems:

[0809] AI models (e.g., generative AI models)

[0810] Database (e.g. PostgreSQL)

[0811] Text Analysis Algorithms

[0812] Speech recognition API (e.g., Google Cloud's speech recognition API)

[0813] The server automatically determines the format of the uploaded learning content (PDF, URL, audio, video, etc.) and processes it appropriately.

[0814] Emotion Engine

[0815] The emotion engine is a technology that recognizes emotions from the user's facial expressions and tone of voice. Specifically, it collects camera footage and audio data while the user is answering test questions and analyzes it using technologies such as Microsoft's Emotion API.

[0816] Specific examples

[0817] The steps for using a high school mathematics textbook PDF as learning content are as follows:

[0818] 1. User operation: A user uploads a PDF file of a mathematics textbook. When uploading, the user selects the option to allow the use of the emotion engine.

[0819] 2. Server processing: The server receives the PDF file, stores it in the storage system, extracts the text using a PDF text extraction algorithm, and analyzes it using an AI model.

[0820] 3. Question generation: Based on the analyzed content, fill-in-the-blank questions, multiple-choice questions, etc. are generated.

[0821] 4. Submit and answer: Users answer questions generated through a web app or dedicated app.

[0822] 5. Emotion recognition and feedback: The emotion engine analyzes the user's facial expressions and voice and provides real-time feedback and adjusts the difficulty of the questions.

[0823] 6. Answer result: After the answer is given, the server automatically scores it and provides feedback that takes into account emotional data.

[0824] 7. Print Data Generation: If the user wishes, convert the problem set into PDF format and provide a download link.

[0825] Prompt Sentence Examples

[0826] "I have uploaded a PDF of a high school mathematics textbook. Please extract important keywords and themes from this content and generate fill-in-the-blank and multiple-choice questions based on them. Also, please adjust the difficulty of the questions by analyzing the user's emotions when answering them."

[0827] The above is a specific embodiment of this system, which efficiently generates test questions from the user's learning content and combines it with an emotion engine to provide a personalized learning experience that responds to the user's emotions.

[0828] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0829] Step 1:

[0830] The user device uploads learning content. The user clicks the "Upload Content" button on the web app, and a file selection dialog appears. The user selects the file, checks the option to allow the use of the emotion engine, and clicks the upload button.

[0831] Input: User's learning content file, emotion engine permission settings

[0832] Output: Learning content is sent from the user device to the server.

[0833] Step 2:

[0834] The server receives the learning content and stores it in storage. The server receives a file upload request from the user device and stores the file in temporary storage. The server then moves the stored file to a specific location in the storage system (e.g., Amazon S3).

[0835] Input: Learning content sent from the user's device

[0836] Output: Learning content files saved to storage

[0837] Step 3:

[0838] The server analyzes the file format and selects the appropriate analysis algorithm. The server analyzes the metadata of the uploaded file and determines the file format (PDF, URL, audio, video, etc.).

[0839] Input: Learning content files saved in storage

[0840] Output: Analysis algorithms depending on the file format

[0841] Step 4:

[0842] The server analyzes the content and runs the analysis algorithm it has selected. For PDF files, it uses the PDF text extraction algorithm, and for audio files, it uses a speech recognition API (e.g., Google Cloud's speech recognition API).

[0843] Input: Parsing algorithm depending on the file format

[0844] Output: Extracted text data

[0845] Step 5:

[0846] The AI ​​model on the server analyzes the extracted text data, using natural language processing techniques to identify important keywords, themes, and key points. This analysis is performed using tools such as Amazon Comprehend.

[0847] Input: Extracted text data

[0848] Output: Analyzed keywords, themes, and key points data

[0849] Step 6:

[0850] The server generates test questions based on the analysis results. The server calls up a generative AI model and automatically creates fill-in-the-blank, multiple-choice, and essay questions based on the analyzed keywords and themes.

[0851] Input: Analyzed keywords, themes, and key points data

[0852] Output: Generated test question data

[0853] Step 7:

[0854] The server saves the generated test questions in a database and prepares them for serving to users.The server saves the generated questions in a database (e.g., PostgreSQL) and generates a URL for the question set via an API so that it can be accessed from a web app or dedicated app.

[0855] Input: Generated test question data

[0856] Output: Question data stored in the database, URL for accessing it

[0857] Step 8:

[0858] The user's device displays the test questions, and the user answers them. The user accesses the provided URL and the questions are displayed on the web app. The user answers the questions in the browser.

[0859] Input: The URL the user accessed in the web app

[0860] Output: User's answers to the questions

[0861] Step 9:

[0862] The emotion engine recognizes the user's emotions while they are answering. While the user is entering their answer, the emotion engine periodically collects camera footage and microphone audio, and calls the Emotion API to obtain the analysis results.

[0863] Input: User's camera video and audio data

[0864] Output: Parsed emotion data

[0865] Step 10:

[0866] The server provides real-time feedback and adjusts the difficulty level. The server recalculates the difficulty of the questions based on the emotional data sent from the emotion engine, and displays feedback messages or easy questions as needed.

[0867] Input: Parsed emotion data and user response data

[0868] Output: Adjusted difficulty of the problem, feedback message

[0869] Step 11:

[0870] The server scores the user's answers and provides feedback. When the user submits their answer, the server automatically scores it and provides immediate feedback that takes into account emotional data.

[0871] Input: User's answer data

[0872] Output: Marking results and feedback

[0873] Step 12:

[0874] The server generates the printable data and provides it to the user. If the user wishes, the server converts the generated problem set into PDF format and provides a download link.

[0875] Input: User request

[0876] Output: PDF problem set and download link

[0877] The above are the specific processing steps of this system.

[0878] (Application example 2)

[0879] 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."

[0880] Conventional learning systems have struggled to provide a personalized learning experience that responds to the user's emotions. This has resulted in a lack of appropriate feedback and support based on emotions such as stress and excitement felt during learning, resulting in reduced learning efficiency. Furthermore, conventional learning systems often lacked sufficient support for factory employees to efficiently master specific skills and operations.

[0881] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0882] In this invention, the server

[0883] means for receiving content to be studied from a user;

[0884] means for analyzing the content of the received content;

[0885] means for automatically generating test questions based on the analyzed content;

[0886] means for providing the generated test questions to a user;

[0887] A sentiment analysis means;

[0888] a means for adjusting the difficulty and format of test questions based on user sentiment;

[0889] Includes.

[0890] This allows for a personalized learning experience by understanding employees' emotions in real time while they are studying and providing appropriate feedback. Furthermore, the generated test questions are dynamically adjusted according to the user's emotions, maximizing the effectiveness of learning. Furthermore, using emotion analysis technology can help factory employees efficiently master operations and skills, improving overall production efficiency.

[0891] "Content to be studied" refers to information or learning materials that a user wishes to study, and includes text data, web page data, audio data, video data, and the like.

[0892] "Means for analysis" refers to the technologies and algorithms used to understand the content of received content and extract important parts, including natural language processing and speech recognition.

[0893] "Means for automatically generating test questions" refers to a system that generates questions in a format appropriate for the user's learning content based on analyzed content, and uses an AI model or a generative AI model.

[0894] "Means for providing" refers to devices or applications for providing the generated test questions in a form that users can access, including smartphone apps and web apps.

[0895] "Emotion analysis means" refers to technology or devices for analyzing emotions from a user's facial expressions and voice, and includes image analysis and voice analysis using a camera or microphone.

[0896] "Feedback" refers to instructions or comments provided based on a user's answers and feelings, which can help improve the effectiveness of learning.

[0897] "Means for adjusting difficulty and format" refers to a system that changes the difficulty and format of test questions according to the user's emotions and learning situation, and is dynamically adjusted in real time.

[0898] A "personalized learning experience" aims to maximize learning effectiveness by providing learning methods and content optimized for each user's individual learning situation and emotions.

[0899] "Real-time" means reacting and responding immediately to changes in the user's learning and emotions.

[0900] "Downloadable in PDF format" means that the generated exam questions are saved as electronic files and are available for users to download via the Internet.

[0901] The system necessary to implement this invention mainly comprises the following elements: a server, a user terminal, an emotion analysis means, and an AI model.

[0902] server

[0903] The server plays a central role in analyzing learning content received from users and automatically generating test questions. Specifically, the server receives content uploaded by users, such as PDFs, web pages, audio, and video, and extracts text using appropriate analysis algorithms. The extracted text is analyzed using natural language processing (NLP) techniques to identify important keywords and themes. Test questions are then automatically generated based on the analysis results using an AI model. These test questions are then made available to users for access, viewing, and answering via a web app or dedicated app.

[0904] User terminal

[0905] User devices include PCs, smartphones, tablets, etc., and are used by users to upload learning content and answer generated test questions. User devices are equipped with cameras and microphones, which are used as emotion analysis tools. This makes it possible to analyze emotions in real time from the user's facial expressions and tone of voice, and send the results to the server.

[0906] Emotion analysis means

[0907] Emotion analysis is a technology that uses a camera and microphone to analyze the user's facial expressions and voice. Specifically, it collects camera footage and microphone audio from the user's device and analyzes the user's emotions in real time using the EmotionRecognizer library. This emotional data is sent to a server and used to dynamically adjust the difficulty and format of test questions according to the user's emotions.

[0908] Feedback and Adjustments

[0909] The generated test questions are provided to the user, and as the user answers them, the server monitors the user's state through emotional analysis. For example, if the user is feeling "frustrated," the server will lower the difficulty of the questions based on the emotional data, reducing the user's stress and improving learning efficiency. Similarly, if the user is determined to be "confident," the server will provide questions of increased difficulty. This ensures that the user always has the optimal learning experience.

[0910] Specific examples

[0911] For example, suppose a factory employee uploads "Robot Operation Manual.pdf" to the system as learning content. The system extracts text from the PDF file and analyzes important operating procedures and troubleshooting methods. Based on this analysis, test questions are generated to deepen the employee's understanding and are provided through the smart glasses. When the employee answers the test questions, the smart glasses' camera and microphone are used to analyze their emotions. If the system detects "frustration," it automatically adjusts the difficulty of the questions and provides appropriate feedback.

[0912] Prompt Sentence Examples

[0913] An example of a prompt sentence to input to the generative AI model is:

[0914] "You have an employee learning how to operate a robot. If the analysis reveals that his state is 'frustrated,' create questions that are easy to understand. Conversely, if the analysis reveals that he is 'confident,' create questions that are difficult to understand."

[0915] Instructions include:

[0916] In this way, the present invention is a system that maximizes learning effectiveness by providing a personalized learning experience according to the user's emotions.

[0917] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0918] Step 1:

[0919] Users upload learning content.

[0920] The user uploads the content they want to learn (e.g., "Robot Operation Manual.pdf") from their device to the system. The device receives the file and sends it to the server. The input is the learning content file, and the output is the content saved on the server.

[0921] Step 2:

[0922] The server stores the received content and prepares it for analysis.

[0923] The server saves the uploaded file in storage and automatically detects the file format. For example, if it is a PDF file, the server uses a PDF text extraction algorithm to extract the text. The input is the received content and the output is the data ready to be parsed.

[0924] Step 3:

[0925] The server analyzes the learning content.

[0926] The server uses natural language processing (NLP) techniques to analyze the text and identify important keywords and themes. The specific software used here is an NLP library. The input is prepared text data, and the output is the analyzed keywords and themes.

[0927] Step 4:

[0928] The server generates test questions based on the analysis results.

[0929] The server uses a generative AI model to automatically generate test questions based on the analyzed keywords and themes. The generated questions can be fill-in-the-blank, multiple choice, or essay questions. The input is the analysis results, and the output is the generated test questions.

[0930] Step 5:

[0931] The server provides the generated test questions to the user terminal.

[0932] The server stores the generated test questions in a database and makes them accessible to users. Users can view the questions through a web app or a dedicated app. The input is the generated test question data, and the output is the questions displayed on the user's device.

[0933] Step 6:

[0934] The user answers the test questions.

[0935] The user answers the provided test questions and sends the answers from the terminal to the server. The input is the user's answer data, and the output is the answer data saved on the server.

[0936] Step 7:

[0937] The emotion analysis means analyzes the emotion of the user.

[0938] The system uses the camera and microphone on the user's device to collect facial expressions and voice while the user is answering test questions. It then uses an emotion analysis library (e.g., EmotionRecognizer) to analyze the user's emotions in real time. The input is camera video and audio data, and the output is analyzed emotion data.

[0939] Step 8:

[0940] The server adjusts the difficulty and format of the test questions based on the user's feelings.

[0941] The server adjusts the difficulty of the test questions based on the analyzed emotional data. For example, if the user is judged to be "frustrated," the server sets the difficulty of the questions low, and if the user is judged to be "confident," the server increases the difficulty. The input is emotional data, and the output is the adjusted test questions.

[0942] Step 9:

[0943] The server provides feedback to the user.

[0944] The system provides users with adjusted test questions and appropriate learning feedback in real time, allowing them to receive feedback that is tailored to their learning situation. The input is the adjusted test questions and feedback data, and the output is the feedback displayed on the user's device.

[0945] 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.

[0946] 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.

[0947] 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.

[0948] [Third embodiment]

[0949] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0950] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0951] 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).

[0952] 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.

[0953] 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.

[0954] 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).

[0955] 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.

[0956] 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.

[0957] 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.

[0958] 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.

[0959] 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.

[0960] 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."

[0961] The present invention provides a system for efficiently generating and providing test questions from a learner's learning content. Specific embodiments of this system will be described below.

[0962] System Configuration

[0963] The system mainly consists of the following components:

[0964] User terminal

[0965] server

[0966] User terminal

[0967] A user terminal is a device on which a user uploads learning content, views generated test questions, and answers them. Examples of such devices include personal computers (PCs), smartphones, and tablets.

[0968] server

[0969] The server plays a central role in receiving learning content sent from user devices, analyzing it, generating test questions, and providing them to users. This server is equipped with powerful AI models, databases, text analysis, speech recognition, and other systems.

[0970] User operation (input)

[0971] Users upload the content they want to study (e.g., textbooks, websites, audio files, video files, etc.) from their terminals to the system. At this time, users can select the content and send it to the server in a specific format (e.g., PDF, URL, MP3, MP4, etc.).

[0972] Receiving and storing content

[0973] The server receives the learning content sent from the user's device and saves it in the designated storage. Within this storage, the format of the received file (e.g. PDF, audio, video, etc.) is automatically identified and the appropriate processing is performed.

[0974] Preparing and running content analysis

[0975] The server selects a different parsing algorithm for each file format received: for example, for a PDF file, the server uses a PDF text extraction algorithm to extract text, and for an audio file, it uses a speech recognition API to convert the audio to text.

[0976] Content Analysis

[0977] The AI ​​model on the server analyzes the extracted text to identify important keywords, themes, and key points. This analysis process uses natural language processing techniques to understand sentence structure and meaning.

[0978] Test question generation

[0979] Based on the analysis results, the server generates test questions in various formats (e.g., fill-in-the-blank, multiple-choice, essay, etc.). For example, questions using specific formulas and theorems are generated from mathematics textbooks. The content of the options and fill-in-the-blanks is also automatically generated by the AI ​​model.

[0980] Storage and provision of exam questions

[0981] The generated test questions are stored in a database on the server and made accessible to users. User devices can access and display these question sets through a web app or dedicated app.

[0982] Answers and feedback

[0983] The user answers the generated questions on their device. Once the answer is complete, the server automatically scores the answer and provides immediate feedback to the user. The user can then check the accuracy rate and explanations to assess their own understanding.

[0984] Generate print data

[0985] The system also provides a function for users to download and print the generated problem set in PDF format. When a user requests a printable PDF, the server generates a PDF with high-quality layout and provides it to the user.

[0986] Specific examples

[0987] For example, the following shows the process when using a high school mathematics textbook PDF as learning content:

[0988] 1. User operation: The user uploads a PDF file of a mathematics textbook from the user's terminal.

[0989] 2. Server processing: The server receives the PDF file and saves it in storage.

[0990] 3. Content analysis: Using a PDF text extraction algorithm, the text of the textbook is extracted, and the AI ​​model analyzes the content.

[0991] 4. Question Generation: Generate fill-in-the-blank and multiple-choice questions based on important mathematical formulas and theorems.

[0992] 5. Submit and Answer: The user answers the generated questions in the web app, and the server instantly scores and provides feedback.

[0993] 6. Printable data: When requested by the user, the server converts the problem set into PDF format and provides a download link.

[0994] In this way, this system is a powerful tool for efficiently generating test questions from the learning content desired by the user and for improving learning effectiveness.

[0995] The processing flow will be explained below.

[0996] Step 1:

[0997] The user accesses the learning content upload screen on their device (PC, smartphone, etc.). The user selects and uploads the content they want to study (e.g., textbook PDF, website URL, audio file, video file, etc.).

[0998] Step 2:

[0999] The server receives the content file sent from the user terminal, processes the received file securely, and saves it in the designated storage within the system.

[1000] Step 3:

[1001] The server determines the format of the received file (PDF, URL, audio, video, etc.) and prepares to select an analysis algorithm depending on the file format.

[1002] Step 4:

[1003] The server applies a parsing algorithm based on the file format.

[1004] For PDF: The server uses a PDF text extraction library (e.g., PyMuPDF) to extract text from the PDF.

[1005] For a URL: The server uses a web scraping tool (e.g., BeautifulSoup, Scrapy) to extract the text of the web page.

[1006] For audio files: The server uses a speech recognition API (e.g., Google Cloud Speech-to-Text) to convert the audio to text.

[1007] For video files: The server extracts the audio portion of the video and converts it to text using a speech recognition API.

[1008] Step 5:

[1009] The AI ​​model on the server analyzes the extracted text, using natural language processing techniques to identify important information in the content (keywords, themes, key points, etc.), and builds analytical data for generating questions.

[1010] Step 6:

[1011] The server automatically generates test questions based on the analyzed data. The generated questions are in the following formats:

[1012] Fill-in-the-blank questions: Questions that leave specific keywords or phrases blank and require users to fill them in.

[1013] Multiple choice questions: Questions that provide multiple options and require the user to select the correct answer.

[1014] Essay questions: Questions that require users to write freely.

[1015] Step 7:

[1016] The server stores the generated exam questions in a database, where they are prepared for user access.

[1017] Step 8:

[1018] Users access the generated exam questions on their device, answer them through a web or app interface, and submit their answers once they have completed the questions.

[1019] Step 9:

[1020] The server receives the user's answers and automatically scores them. The server immediately returns the results and feedback to the user, who can then check the answers and evaluate their own understanding.

[1021] Step 10:

[1022] If the user wishes, the server will make the generated test question set available for download in PDF format. When the user clicks the "Generate Printable PDF" button, the server will convert the question set into a high-quality PDF format and provide a download link.

[1023] Step 11:

[1024] Users can download the PDF file from the download link and print it out, allowing them to continue their studies offline.

[1025] The above are the specific processing steps for the system to automatically generate and provide test questions from learning content.

[1026] Example 1

[1027] 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."

[1028] Conventional learning systems require users to manually generate test questions from learning content, which is time-consuming and labor-intensive, and limits the quality and variety of questions generated. Furthermore, when content formats differ, it is difficult to select an appropriate analysis method for each, making efficient question generation impossible. The present invention aims to solve these problems and provide an environment in which users can study efficiently and effectively.

[1029] 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.

[1030] In this invention, the server includes means for receiving study content from a user, means for automatically determining the format of the received content and selecting an appropriate analysis algorithm, means for analyzing the content based on the selected analysis algorithm, means for automatically generating test questions based on the analyzed content, means for providing the generated test questions to the user, means for receiving the user's answers to the test questions and performing automatic grading and feedback, and means for providing the generated test questions in a printable PDF format. This allows users to efficiently generate and answer test questions from content in a variety of formats and have the results evaluated.

[1031] "Content to be studied" refers to information formats such as text data, web page data, audio data, and video data provided by users for study purposes.

[1032] "User" refers to a person who uses the learning system to upload learning content and answer generated exam questions.

[1033] "Means for receiving" refers to the function by which the system receives and records learning content sent by a user.

[1034] "Means for automatically identifying the format and selecting the appropriate analysis algorithm" refers to technology for identifying the format of received content and automatically selecting the analysis method that is optimal for that format.

[1035] "Means of analysis" refers to the function of analyzing the content of learning content using selected algorithms and extracting important information.

[1036] "Means for automatically generating test questions" refers to a function that automatically creates test questions such as fill-in-the-blank questions, multiple-choice questions, and essay questions based on the analyzed content.

[1037] "Means for providing to the user" refers to a function for presenting the generated test questions in a form that can be used by the user.

[1038] "Means for receiving answers and automatically scoring and providing feedback" refers to the function that allows the system to receive the answers provided by the user, determine whether they are correct, and immediately return evaluation results and advice.

[1039] "Means for providing in printable PDF format" refers to the ability to generate generated exam questions as high-quality PDF documents that users can download and print.

[1040] This invention is a system that allows learners to efficiently generate and provide test questions from any learning content. This system is mainly composed of a user terminal and a server.

[1041] User terminal

[1042] A user terminal is a device on which a user uploads learning content, views generated test questions, and answers them. This includes personal computers (PCs), smartphones, tablets, etc. Users access the system using a browser or dedicated application on these devices.

[1043] server

[1044] The server plays a central role in receiving learning content sent from user devices, analyzing it, and generating and providing test questions. The server incorporates powerful AI models, databases, text analysis, speech recognition, and other systems. Specifically, the following software and web services are used:

[1045] PDF Text Extraction Algorithm

[1046] Speech Recognition API

[1047] Natural language processing technology

[1048] Uploading learning content

[1049] Users upload the content they want to study (e.g., textbooks, websites, audio files, video files, etc.) from their terminals to the system. At this time, users can send this content to the server in a specific format (e.g., PDF, URL, MP3, MP4, etc.).

[1050] Receiving and storing content

[1051] The server receives the learning content sent from the user's device and stores it in storage. Within this storage, the format of the received file (e.g., PDF, audio, video, etc.) is automatically identified and the appropriate processing is performed.

[1052] Preparing and running content analysis

[1053] The server selects a different parsing algorithm for each file format received: for example, for a PDF file, the server uses a PDF text extraction algorithm to extract text, and for an audio file, it uses a speech recognition API to convert the audio to text.

[1054] Content Analysis

[1055] The AI ​​model on the server analyzes the extracted text to identify important keywords, themes, and key points. This analysis process uses natural language processing techniques to understand sentence structure and meaning.

[1056] Test question generation

[1057] Based on the analysis results, the server generates test questions in various formats (e.g., fill-in-the-blank, multiple-choice, essay, etc.). For example, questions using specific formulas and theorems are generated from mathematics textbooks. The content of the options and fill-in-the-blanks is also automatically generated by the AI ​​model.

[1058] Storage and provision of exam questions

[1059] The generated test questions are stored in a database on the server and made accessible to users. User devices can access and display these question sets through a web app or dedicated app.

[1060] Answers and feedback

[1061] The user answers the generated questions on their device. Once the answer is complete, the server automatically scores the answer and provides immediate feedback to the user. The user can then check the accuracy rate and explanations to assess their level of understanding.

[1062] Generate print data

[1063] The system also provides a function for users to download and print the generated problem set in PDF format. When a user requests a printable PDF, the server generates a PDF with high-quality layout and provides it to the user.

[1064] Specific examples

[1065] For example, the following shows the process when using a high school mathematics textbook PDF as learning content:

[1066] 1. User operation: The user uploads a PDF file of a mathematics textbook from the user's terminal.

[1067] 2. Server processing: The server receives the PDF file and saves it in storage.

[1068] 3. Content analysis: Using a PDF text extraction algorithm, the text of the textbook is extracted, and the AI ​​model analyzes the content.

[1069] 4. Question Generation: Generate fill-in-the-blank and multiple-choice questions based on important mathematical formulas and theorems. Prompt: "Generate exam questions based on the contents of this mathematics textbook."

[1070] 5. Submit and Answer: The user answers the generated questions in the web app, and the server instantly scores and provides feedback.

[1071] 6. Printable data: When requested by the user, the server converts the problem set into PDF format and provides a download link.

[1072] This system is a powerful tool for efficiently generating test questions from the learning content desired by the user, thereby enhancing learning effectiveness.

[1073] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1074] Step 1:

[1075] The user selects and uploads learning content from the user device. The learning content file (e.g., PDF, URL, MP3, MP4, etc.) is provided as input. The user device then sends the selected file to the server. Specifically, the user clicks the "Upload" button in a browser or dedicated application and selects the file.

[1076] Step 2:

[1077] The server receives learning content sent from the user's device. The file sent by the user is provided as input. After receiving the file, the server saves it in storage and automatically determines the file format (e.g., PDF, URL, MP3, MP4). Specifically, it analyzes the received file name and format, and sorts and saves it in the specified folder.

[1078] Step 3:

[1079] The server selects the appropriate analysis algorithm based on the format of the saved file. The saved file format is provided as input. The server selects the PDF text extraction algorithm for PDF files and the speech recognition API for audio files. Specifically, it uses conditional branching to determine the file format and calls the corresponding analysis algorithm.

[1080] Step 4:

[1081] The server analyzes the content using the selected algorithm. The stored content and the selected analysis algorithm are provided as input. In the case of a PDF file, the server extracts text using a PDF text extraction algorithm, and in the case of an audio file, it converts the audio to text using a speech recognition API. Specifically, the analysis engine extracts information from the file and generates text data.

[1082] Step 5:

[1083] The AI ​​model on the server analyzes the extracted text and identifies important keywords and themes. The extracted text is provided as input. The server uses natural language processing techniques to extract keywords and identify key points. Specifically, the text analysis engine tokenizes the text and extracts themes and key information.

[1084] Step 6:

[1085] The server automatically generates test questions based on the analysis results. The key points and keywords of the analyzed text are provided as input. The server uses the generative AI model to create a prompt sentence and generates test questions based on that. Specifically, the server inputs the prompt sentence into the generative AI model and obtains the generated question sentence.

[1086] Step 7:

[1087] The server saves the generated test questions in a database so that they can be provided to users. The generated test questions are provided as input. The server inserts the generated question data into the database and provides a question set in response to a user request. Specifically, the server stores the question data in the database and provides it to the user's device via a WebAPI.

[1088] Step 8:

[1089] The user answers the generated questions using the user terminal. The test questions provided as input are displayed on the user terminal. After answering, the user sends the answers to the server. Specifically, the user answers the questions on the application and clicks the "Submit" button.

[1090] Step 9:

[1091] The server receives the answers submitted by the user, automatically scores them, and returns the feedback to the user. The user's answer data is provided as input. The server uses a scoring algorithm to determine whether the answer is correct and generates a result. Specifically, the server inputs the answer data into the scoring algorithm, generates a score result, and displays it to the user in real time.

[1092] Step 10:

[1093] When a user requests a printable PDF, the server generates a PDF with a high-quality layout and provides it to the user. The user's request is provided as input. The server converts the problem set into PDF format using a PDF generation tool and generates a download link. Specifically, the server starts the PDF generation process in response to the request and displays a link to the generated PDF file in the web application.

[1094] In this way, through a series of processing steps, the system provides the user with the functionality to efficiently generate test questions from their own learning content, answer them, and evaluate the results.

[1095] (Application example 1)

[1096] 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."

[1097] Conventional learning systems have had problems such as the difficulty for learners to efficiently generate test questions from any learning content, and the inability to properly analyze the format of uploaded content and automatically generate test questions, resulting in a significant lack of user convenience. Another issue is the difficulty in effectively providing the generated test questions to users and providing immediate feedback.

[1098] 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.

[1099] In this invention, the server includes means for receiving study content from a user, means for analyzing the received content, means for automatically generating test questions based on the analyzed content, means for providing the generated test questions to the user, means for extracting text according to the format of the received content, means for generating prompt sentences from the extracted text, means for using the prompt sentences to generate test questions using a generative AI model, and means for providing the generated test questions to the user via the web and accepting answers. This allows learners to efficiently generate test questions from any study content, automates analysis and text extraction, and provides the generated test questions instantly for rapid feedback.

[1100] "Content to be studied" refers to information such as text data, audio data, and video data that is uploaded by a user for study purposes.

[1101] "Means for receiving from users" refers to the interface and processing functions that allow users to upload learning content to the system.

[1102] "Content analysis means" refers to analytical techniques used to understand uploaded content and identify important keywords and themes.

[1103] "Means for automatically generating test questions" refers to algorithms or generative AI models that create appropriate test questions based on the analyzed content.

[1104] The "means for providing test questions to the user" refers to a function for displaying or transmitting the generated test questions to the user's device.

[1105] The "means for extracting text according to the format of received content" refers to a technology for extracting text information in a manner appropriate for the format of received content, such as text, audio, or video.

[1106] The "means for generating prompt sentences" refers to a technology that creates input sentences (prompt sentences) for the generative AI model to generate test questions based on the extracted text.

[1107] A "generative AI model" is an artificial intelligence model for generating test questions based on a given prompt.

[1108] "Means for providing and accepting answers via the web" refers to the interface and processing functions that allow users to access test questions and submit answers via a web browser or dedicated application.

[1109] To implement this invention, a system including the following elements is required: a user terminal, a server, a powerful AI model, text analysis, speech recognition, etc.

[1110] System Configuration

[1111] The system mainly consists of the following components:

[1112] User terminal: A device on which a user uploads learning content, views generated test questions, and answers them. Examples include smartphones, tablets, and personal computers (PCs).

[1113] Server: Plays a central role in receiving learning content sent from user devices, analyzing it, generating and providing test questions. This server is equipped with powerful AI models, databases, text analysis, speech recognition, and other systems.

[1114] User operations

[1115] Users can upload any learning content (e.g., textbooks, websites, audio files, video files, etc.) from their devices. At this time, users can select these contents and send them to the server in a specific format (e.g., PDF, URL, MP3, MP4, etc.).

[1116] Server Processing

[1117] 1. Receiving and storing content

[1118] The server receives the learning content sent from the user's device and saves it in the designated storage. Within this storage, the format of the received file is automatically identified and the appropriate processing is performed.

[1119] 2. Preparing and running content analysis

[1120] The server selects different parsing algorithms for each file format received: for example, for PDF files, the server performs PDF text extraction; for audio files, it uses speech recognition technology to convert speech to text.

[1121] 3. Content Analysis

[1122] The AI ​​model on the server analyzes the extracted text to identify important keywords and themes, using natural language processing techniques to understand sentence structure and meaning.

[1123] 4. Prompt generation

[1124] Based on the extracted text, the AI ​​model creates an input sentence (prompt sentence) for generating test questions, for example, in the format "Please create a test question from the following content: [extracted text]."

[1125] 5. Test Question Generation

[1126] Using the prompt, a generative AI model generates test questions, which can range from fill-in-the-blank questions, multiple-choice questions, and essay questions.

[1127] 6. Provision of test questions and answers

[1128] The generated test questions are stored in a database on the server and made accessible to users, who can then answer them via a web app or a dedicated app.

[1129] Specific examples

[1130] For example, if you use a high school mathematics textbook PDF as learning content, the process is as follows:

[1131] 1. User operation: The user uploads a PDF file of a mathematics textbook from the user terminal to the server.

[1132] 2. Server processing:

[1133] Receive PDF files and save them to storage.

[1134] Using PDF text extraction technology, textbook content is extracted.

[1135] The extracted text is analyzed by an AI model.

[1136] 3. Prompt generation: Based on the extracted text, a prompt is generated: "Please create test questions from the following content: [Textbook content]."

[1137] 4. Test question generation: Using a generative AI model, test questions are generated based on the analysis results.

[1138] 5. Submit and answer: The user answers the generated test questions in the web app, and the server immediately scores and provides feedback.

[1139] Example prompt sentence:

[1140] "Create an exam question from the following content: A description of the growth of morning glories. Morning glories are summer flowers and require watering and sunlight."

[1141] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1142] Step 1:

[1143] Users upload learning content

[1144] Users upload the content they want to study (e.g., PDF files, audio files, video files, etc.) from their own devices to the system. The input here is the study content selected by the user, and the output is the transmission of this content to the server.

[1145] Step 2:

[1146] The server receives and stores the learning content.

[1147] The server receives the learning content sent from the user's device and saves it in the specified storage. The input here is the content file sent by the user, and the output is the content file saved in the storage. At this time, the content format (e.g., PDF, MP3, MP4, etc.) is also automatically determined.

[1148] Step 3:

[1149] The server determines the content type and selects the appropriate parsing algorithm.

[1150] The server selects an appropriate parsing algorithm depending on the format of the received content, for example, a text extraction algorithm for a PDF file, or a speech recognition algorithm for an audio file. The input here is the saved content file and its format information, and the output is the selected parsing algorithm.

[1151] Step 4:

[1152] The server parses the content and extracts the text

[1153] The server analyzes the content using the selected analysis algorithm and extracts the required text data, for example extracting text from a PDF file or converting audio to text from an audio file. The input here is the content file and the selected analysis algorithm, and the output is the extracted text data.

[1154] Step 5:

[1155] The server generates a prompt

[1156] The server creates a prompt based on the extracted text data for the generative AI model to generate questions. An example of a prompt is: "Please create a test question from the following content: [extracted text]". Here, the input is the extracted text data, and the output is the generated prompt.

[1157] Step 6:

[1158] The server generates test questions using the generative AI model

[1159] The server generates test questions by inputting the generative AI model using the generated prompt sentences, where the inputs are the prompt sentences and the generative AI model, and the output is the generated test questions.

[1160] Step 7:

[1161] The server provides the test questions to the user.

[1162] The server stores the generated test questions in a database and makes them accessible to users. Users can access and answer these questions through a web application or a dedicated application. The input here is the generated test questions, and the output is the test questions provided via the web.

[1163] Step 8:

[1164] The user answers the questions and the server provides feedback

[1165] The user answers the provided test questions and sends the answers to the server, which instantly marks the answers and provides feedback to the user. The input here is the user's answer, and the output is the marking results and feedback.

[1166] 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.

[1167] This invention is a system that generates test questions from a user's study content and provides personalized questions according to the user's emotions by combining it with an emotion engine. Specific embodiments of this system are described below.

[1168] System Configuration

[1169] The system mainly consists of the following components:

[1170] User terminal

[1171] server

[1172] Emotion Engine

[1173] User terminal

[1174] A user device refers to the device on which a user uploads learning content, views generated test questions, and answers them. This includes PCs, smartphones, tablets, etc. The device is also equipped with a camera and microphone, and the emotion engine uses these devices to analyze the user's emotions.

[1175] server

[1176] The server plays a central role in receiving learning content sent from user devices, analyzing it, generating test questions, and providing them to users. This server is equipped with systems such as AI models, databases, text analysis, and speech recognition.

[1177] Emotion Engine

[1178] The emotion engine is a technology that recognizes emotions from the user's facial expressions and tone of voice, allowing it to analyze the user's emotions while answering test questions and provide appropriate feedback and adjust the questions in real time.

[1179] User operation (input)

[1180] Users upload the content they want to study (textbooks, websites, audio files, video files, etc.) from their devices. When uploading content, users also set up their devices to allow the use of the emotion engine.

[1181] Receiving and storing content

[1182] The server receives the content file sent from the user's device and saves it in the specified storage. The format of the received file (PDF, URL, audio, video, etc.) is automatically determined and the appropriate processing is performed.

[1183] Preparing and running content analysis

[1184] The server selects a different parsing algorithm for each file format received: for example, for PDF files, it uses a PDF text extraction algorithm to extract text, and for audio files, it uses a speech recognition API to convert speech to text.

[1185] Content Analysis

[1186] An AI model on the server analyzes the extracted text to identify important keywords, themes, and key points. This analysis process uses natural language processing techniques to understand sentence structure and meaning.

[1187] Test question generation

[1188] Based on the analysis results, the server generates test questions in various formats (fill-in-the-blank, multiple choice, essay, etc.), so that the generated questions are in a format that is appropriate for the user's learning content.

[1189] Storage and provision of exam questions

[1190] The server stores the generated test questions in a database and makes them available for users to access. User devices can access and display these question sets through a web app or a dedicated app.

[1191] Emotion Recognition and Feedback

[1192] As users answer questions, the emotion engine uses camera footage and audio data to recognize their emotions. Based on the recognized emotions, the server adjusts the difficulty of the questions and provides timely feedback and support messages.

[1193] Answers and feedback

[1194] When a user answers a question, the server automatically scores the answer and provides immediate feedback, taking into account emotional data recognized by the emotion engine, according to the user's level of understanding and state.

[1195] Generate print data

[1196] The system also provides users with the ability to download and print the generated problem sets in PDF format if they wish, allowing them to study offline.

[1197] Specific examples

[1198] The steps for using a high school mathematics textbook PDF as learning content are as follows:

[1199] 1. User operation: The user uploads a PDF file of a mathematics textbook. The user allows the emotion engine to be used.

[1200] 2. Server processing: Receive the PDF file, save it in storage, extract the text, and analyze it using the AI ​​model.

[1201] 3. Question generation: Based on the analyzed content, fill-in-the-blank and multiple-choice questions are generated.

[1202] 4. Submit and Answer: The user answers the generated questions in the web app.

[1203] 5. Emotion recognition and feedback: The emotion engine analyzes the user's facial expressions and voice and adjusts feedback and problem difficulty in real time.

[1204] 6. Answer result: Once the answer is completed, the server immediately scores it and provides feedback that takes into account emotional data.

[1205] 7. Printable data: Upon user request, the problem set will be converted into PDF format and a download link will be provided.

[1206] As described above, the present invention efficiently generates test questions from a user's learning content and combines them with an emotion engine to provide a personalized learning experience that responds to the user's emotions.

[1207] The processing flow will be explained below.

[1208] Step 1:

[1209] The user accesses the learning content upload screen on their device (PC, smartphone, etc.). The user selects and uploads the content they want to study (textbook PDF, website URL, audio file, video file, etc.). They also set up permission to use the emotion engine.

[1210] Step 2:

[1211] The server receives the content file sent from the user's device. The received file is safely processed and saved in the designated storage within the system. The emotion engine usage settings are also saved at the same time.

[1212] Step 3:

[1213] The server determines the format of the received file (PDF, URL, audio, video, etc.) and prepares to select an analysis algorithm depending on the file format.

[1214] Step 4:

[1215] The server applies a parsing algorithm based on the file format.

[1216] For PDF: The server uses a PDF text extraction library (e.g., PyMuPDF) to extract text from the PDF.

[1217] For a URL: The server uses a web scraping tool (e.g., BeautifulSoup, Scrapy) to extract the text of the web page.

[1218] For audio files: The server uses a speech recognition API (e.g., Google Cloud Speech-to-Text) to convert the audio to text.

[1219] For video files: The server extracts the audio portion of the video and converts it to text using a speech recognition API.

[1220] Step 5:

[1221] The AI ​​model on the server analyzes the extracted text, using natural language processing techniques to identify important information in the content (keywords, themes, key points, etc.), and builds analytical data for generating questions.

[1222] Step 6:

[1223] The server automatically generates test questions based on the analyzed data. The generated questions are in the following formats:

[1224] Fill-in-the-blank questions: Questions that leave specific keywords or phrases blank and require users to fill them in.

[1225] Multiple choice questions: Questions that provide multiple options and require the user to select the correct answer.

[1226] Essay questions: Questions that require users to write freely.

[1227] Step 7:

[1228] The server stores the generated exam questions in a database, where they are prepared for user access.

[1229] Step 8:

[1230] Users access the test questions generated on their own devices and answer them through a web or app interface. As they begin answering questions, the emotion engine uses camera footage and audio data to recognize the user's emotions in real time.

[1231] Step 9:

[1232] The server receives and analyzes the user's emotional data, and based on that data, adjusts the difficulty of the test questions provided and provides feedback and support messages at appropriate times.

[1233] Step 10:

[1234] Once the user has completed the questions, the server automatically scores the answers and provides immediate feedback, taking into account the emotional data recognized by the emotion engine, providing detailed feedback tailored to the user's level of understanding and state.

[1235] Step 11:

[1236] If the user wishes, the server will make the generated test question set available for download in PDF format. When the user clicks the "Generate Printable PDF" button, the server will convert the question set into a high-quality PDF format and provide a download link.

[1237] Step 12:

[1238] Users can download the PDF file from the download link and print it out, allowing them to continue their studies offline.

[1239] As a concrete example, let's consider the case where a user uses a high school mathematics textbook PDF as learning content:

[1240] 1. User operation: The user uploads a PDF file of a mathematics textbook. The user allows the emotion engine to be used.

[1241] 2. Server processing: The PDF file is received and stored in storage. The text is extracted and analyzed using the AI ​​model.

[1242] 3. Question generation: Based on the analyzed content, fill-in-the-blank and multiple-choice questions are generated.

[1243] 4. Providing and answering: The user answers the generated questions in the web app. The emotion engine analyzes the user's facial expressions and voice.

[1244] 5. Emotion recognition and feedback: The emotion engine recognizes the user's emotions in real time, and the server provides appropriate feedback and problem adjustments.

[1245] 6. Answer result: Once the answer is completed, the server immediately scores it and provides feedback that takes into account emotional data.

[1246] 7. Printable data: Upon user request, the problem set will be converted into PDF format and a download link will be provided.

[1247] As described above, the present invention efficiently generates test questions from learning content and combines them with an emotion engine to provide a personalized learning experience that responds to the user's emotions.

[1248] Example 2

[1249] 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."

[1250] While conventional learning systems can generate test questions from users' learning content, they have the problem of being unable to provide personalized feedback or adjust questions based on the user's emotions. Furthermore, because they do not take into account the user's level of understanding or emotional state, efficient and effective learning is difficult.

[1251] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving study content from a user, means for analyzing the received content, means for automatically generating test questions based on the analyzed content, means for providing the generated test questions to the user, means for recognizing the user's emotions, and means for providing feedback and adjusting the questions in real time based on the recognized emotions. This enables a personalized learning experience that takes into account the user's emotional state.

[1252] "Content to be studied" refers to information provided by a user for study purposes, and includes formats such as text data, web page data, audio data, and video data.

[1253] "Means for receiving from users" refers to interfaces and functions that allow users to upload content to be studied to the system.

[1254] "Means for analyzing the content" refers to technology that mechanically analyzes the received learning content and identifies important keywords, themes, key points, etc.

[1255] "Means for automatically generating test questions" refers to a function that automatically creates fill-in-the-blank questions, multiple-choice questions, essay questions, etc. based on the content of the analyzed content.

[1256] "Means for providing test questions to users" refers to an interface or function that allows users to access and answer the generated test questions.

[1257] "Means for recognizing user emotions" refers to technology that automatically determines emotions from a user's facial expressions and tone of voice.

[1258] "Means for adjusting feedback and questions in real time" refers to a mechanism that instantly and adaptively changes the learning content, difficulty of questions, and feedback based on the recognized user emotions.

[1259] MODE FOR CARRYING OUT THE INVENTION

[1260] This invention is a system that generates test questions from learning content provided by a user and provides feedback according to the user's emotions using an emotion engine. Specific embodiments of this system are described below.

[1261] System Configuration

[1262] The system mainly consists of the following components:

[1263] 1. User Device

[1264] 2. Server

[1265] 3. Emotion Engine

[1266] User terminal

[1267] A user device refers to the device on which a user uploads learning content and views and answers generated test questions. User devices include PCs, smartphones, tablets, etc. These devices are equipped with cameras and microphones, which the emotion engine uses to analyze the user's emotions.

[1268] server

[1269] The server receives learning content sent from the user's device, analyzes the content, generates test questions, and provides them. This server incorporates the following systems:

[1270] AI models (e.g., generative AI models)

[1271] Database (e.g. PostgreSQL)

[1272] Text Analysis Algorithms

[1273] Speech recognition API (e.g., Google Cloud's speech recognition API)

[1274] The server automatically determines the format of the uploaded learning content (PDF, URL, audio, video, etc.) and processes it appropriately.

[1275] Emotion Engine

[1276] The emotion engine is a technology that recognizes emotions from the user's facial expressions and tone of voice. Specifically, it collects camera footage and audio data while the user is answering test questions and analyzes it using technologies such as Microsoft's Emotion API.

[1277] Specific examples

[1278] The steps for using a high school mathematics textbook PDF as learning content are as follows:

[1279] 1. User operation: A user uploads a PDF file of a mathematics textbook. When uploading, the user selects the option to allow the use of the emotion engine.

[1280] 2. Server processing: The server receives the PDF file, stores it in the storage system, extracts the text using a PDF text extraction algorithm, and analyzes it using an AI model.

[1281] 3. Question generation: Based on the analyzed content, fill-in-the-blank questions, multiple-choice questions, etc. are generated.

[1282] 4. Submit and answer: Users answer questions generated through a web app or dedicated app.

[1283] 5. Emotion recognition and feedback: The emotion engine analyzes the user's facial expressions and voice and provides real-time feedback and adjusts the difficulty of the questions.

[1284] 6. Answer result: After the answer is given, the server automatically scores it and provides feedback that takes into account emotional data.

[1285] 7. Print Data Generation: If the user wishes, convert the problem set into PDF format and provide a download link.

[1286] Prompt Sentence Examples

[1287] "I have uploaded a PDF of a high school mathematics textbook. Please extract important keywords and themes from this content and generate fill-in-the-blank and multiple-choice questions based on them. Also, please adjust the difficulty of the questions by analyzing the user's emotions when answering them."

[1288] The above is a specific embodiment of this system, which efficiently generates test questions from the user's learning content and combines it with an emotion engine to provide a personalized learning experience that responds to the user's emotions.

[1289] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1290] Step 1:

[1291] The user device uploads learning content. The user clicks the "Upload Content" button on the web app, and a file selection dialog appears. The user selects the file, checks the option to allow the use of the emotion engine, and clicks the upload button.

[1292] Input: User's learning content file, emotion engine permission settings

[1293] Output: Learning content is sent from the user device to the server.

[1294] Step 2:

[1295] The server receives the learning content and stores it in storage. The server receives a file upload request from the user device and stores the file in temporary storage. The server then moves the stored file to a specific location in the storage system (e.g., Amazon S3).

[1296] Input: Learning content sent from the user's device

[1297] Output: Learning content files saved to storage

[1298] Step 3:

[1299] The server analyzes the file format and selects the appropriate analysis algorithm. The server analyzes the metadata of the uploaded file and determines the file format (PDF, URL, audio, video, etc.).

[1300] Input: Learning content files saved in storage

[1301] Output: Analysis algorithms depending on the file format

[1302] Step 4:

[1303] The server analyzes the content and runs the analysis algorithm it has selected. For PDF files, it uses the PDF text extraction algorithm, and for audio files, it uses a speech recognition API (e.g., Google Cloud's speech recognition API).

[1304] Input: Parsing algorithm depending on the file format

[1305] Output: Extracted text data

[1306] Step 5:

[1307] The AI ​​model on the server analyzes the extracted text data, using natural language processing techniques to identify important keywords, themes, and key points. This analysis is performed using tools such as Amazon Comprehend.

[1308] Input: Extracted text data

[1309] Output: Analyzed keywords, themes, and key points data

[1310] Step 6:

[1311] The server generates test questions based on the analysis results. The server calls up a generative AI model and automatically creates fill-in-the-blank, multiple-choice, and essay questions based on the analyzed keywords and themes.

[1312] Input: Analyzed keywords, themes, and key points data

[1313] Output: Generated test question data

[1314] Step 7:

[1315] The server saves the generated test questions in a database and prepares them for serving to users.The server saves the generated questions in a database (e.g., PostgreSQL) and generates a URL for the question set via an API so that it can be accessed from a web app or dedicated app.

[1316] Input: Generated test question data

[1317] Output: Question data stored in the database, URL for accessing it

[1318] Step 8:

[1319] The user's device displays the test questions, and the user answers them. The user accesses the provided URL and the questions are displayed on the web app. The user answers the questions in the browser.

[1320] Input: The URL the user accessed in the web app

[1321] Output: User's answers to the questions

[1322] Step 9:

[1323] The emotion engine recognizes the user's emotions while they are answering. While the user is entering their answer, the emotion engine periodically collects camera footage and microphone audio, and calls the Emotion API to obtain the analysis results.

[1324] Input: User's camera video and audio data

[1325] Output: Parsed emotion data

[1326] Step 10:

[1327] The server provides real-time feedback and adjusts the difficulty level. The server recalculates the difficulty of the questions based on the emotional data sent from the emotion engine, and displays feedback messages or easy questions as needed.

[1328] Input: Parsed emotion data and user response data

[1329] Output: Adjusted difficulty of the problem, feedback message

[1330] Step 11:

[1331] The server scores the user's answers and provides feedback. When the user submits their answer, the server automatically scores it and provides immediate feedback that takes into account emotional data.

[1332] Input: User's answer data

[1333] Output: Marking results and feedback

[1334] Step 12:

[1335] The server generates the printable data and provides it to the user. If the user wishes, the server converts the generated problem set into PDF format and provides a download link.

[1336] Input: User request

[1337] Output: PDF problem set and download link

[1338] The above are the specific processing steps of this system.

[1339] (Application example 2)

[1340] 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."

[1341] Conventional learning systems have struggled to provide a personalized learning experience that responds to the user's emotions. This has resulted in a lack of appropriate feedback and support based on emotions such as stress and excitement felt during learning, resulting in reduced learning efficiency. Furthermore, conventional learning systems often lacked sufficient support for factory employees to efficiently master specific skills and operations.

[1342] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1343] In this invention, the server

[1344] means for receiving content to be studied from a user;

[1345] means for analyzing the content of the received content;

[1346] means for automatically generating test questions based on the analyzed content;

[1347] means for providing the generated test questions to a user;

[1348] A sentiment analysis means;

[1349] a means for adjusting the difficulty and format of test questions based on user sentiment;

[1350] Includes.

[1351] This allows for a personalized learning experience by understanding employees' emotions in real time while they are studying and providing appropriate feedback. Furthermore, the generated test questions are dynamically adjusted according to the user's emotions, maximizing the effectiveness of learning. Furthermore, using emotion analysis technology can help factory employees efficiently master operations and skills, improving overall production efficiency.

[1352] "Content to be studied" refers to information or learning materials that a user wishes to study, and includes text data, web page data, audio data, video data, and the like.

[1353] "Means for analysis" refers to the technologies and algorithms used to understand the content of received content and extract important parts, including natural language processing and speech recognition.

[1354] "Means for automatically generating test questions" refers to a system that generates questions in a format appropriate for the user's learning content based on analyzed content, and uses an AI model or a generative AI model.

[1355] "Means for providing" refers to devices or applications for providing the generated test questions in a form that users can access, including smartphone apps and web apps.

[1356] "Emotion analysis means" refers to technology or devices for analyzing emotions from a user's facial expressions and voice, and includes image analysis and voice analysis using a camera or microphone.

[1357] "Feedback" refers to instructions or comments provided based on a user's answers and feelings, which can help improve the effectiveness of learning.

[1358] "Means for adjusting difficulty and format" refers to a system that changes the difficulty and format of test questions according to the user's emotions and learning situation, and is dynamically adjusted in real time.

[1359] A "personalized learning experience" aims to maximize learning effectiveness by providing learning methods and content optimized for each user's individual learning situation and emotions.

[1360] "Real-time" means reacting and responding immediately to changes in the user's learning and emotions.

[1361] "Downloadable in PDF format" means that the generated exam questions are saved as electronic files and are available for users to download via the Internet.

[1362] The system necessary to implement this invention mainly comprises the following elements: a server, a user terminal, an emotion analysis means, and an AI model.

[1363] server

[1364] The server plays a central role in analyzing learning content received from users and automatically generating test questions. Specifically, the server receives content uploaded by users, such as PDFs, web pages, audio, and video, and extracts text using appropriate analysis algorithms. The extracted text is analyzed using natural language processing (NLP) techniques to identify important keywords and themes. Test questions are then automatically generated based on the analysis results using an AI model. These test questions are then made available to users for access, viewing, and answering via a web app or dedicated app.

[1365] User terminal

[1366] User devices include PCs, smartphones, tablets, etc., and are used by users to upload learning content and answer generated test questions. User devices are equipped with cameras and microphones, which are used as emotion analysis tools. This makes it possible to analyze emotions in real time from the user's facial expressions and tone of voice, and send the results to the server.

[1367] Emotion analysis means

[1368] Emotion analysis is a technology that uses a camera and microphone to analyze the user's facial expressions and voice. Specifically, it collects camera footage and microphone audio from the user's device and analyzes the user's emotions in real time using the EmotionRecognizer library. This emotional data is sent to a server and used to dynamically adjust the difficulty and format of test questions according to the user's emotions.

[1369] Feedback and Adjustments

[1370] The generated test questions are provided to the user, and as the user answers them, the server monitors the user's state through emotional analysis. For example, if the user is feeling "frustrated," the server will lower the difficulty of the questions based on the emotional data, reducing the user's stress and improving learning efficiency. Similarly, if the user is determined to be "confident," the server will provide questions of increased difficulty. This ensures that the user always has the optimal learning experience.

[1371] Specific examples

[1372] For example, suppose a factory employee uploads "Robot Operation Manual.pdf" to the system as learning content. The system extracts text from the PDF file and analyzes important operating procedures and troubleshooting methods. Based on this analysis, test questions are generated to deepen the employee's understanding and are provided through the smart glasses. When the employee answers the test questions, the smart glasses' camera and microphone are used to analyze their emotions. If the system detects "frustration," it automatically adjusts the difficulty of the questions and provides appropriate feedback.

[1373] Prompt Sentence Examples

[1374] An example of a prompt sentence to input to the generative AI model is:

[1375] "You have an employee learning how to operate a robot. If the analysis reveals that his state is 'frustrated,' create questions that are easy to understand. Conversely, if the analysis reveals that he is 'confident,' create questions that are difficult to understand."

[1376] Instructions include:

[1377] In this way, the present invention is a system that maximizes learning effectiveness by providing a personalized learning experience according to the user's emotions.

[1378] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1379] Step 1:

[1380] Users upload learning content.

[1381] The user uploads the content they want to learn (e.g., "Robot Operation Manual.pdf") from their device to the system. The device receives the file and sends it to the server. The input is the learning content file, and the output is the content saved on the server.

[1382] Step 2:

[1383] The server stores the received content and prepares it for analysis.

[1384] The server saves the uploaded file in storage and automatically detects the file format. For example, if it is a PDF file, the server uses a PDF text extraction algorithm to extract the text. The input is the received content and the output is the data ready to be parsed.

[1385] Step 3:

[1386] The server analyzes the learning content.

[1387] The server uses natural language processing (NLP) techniques to analyze the text and identify important keywords and themes. The specific software used here is an NLP library. The input is prepared text data, and the output is the analyzed keywords and themes.

[1388] Step 4:

[1389] The server generates test questions based on the analysis results.

[1390] The server uses a generative AI model to automatically generate test questions based on the analyzed keywords and themes. The generated questions can be fill-in-the-blank, multiple choice, or essay questions. The input is the analysis results, and the output is the generated test questions.

[1391] Step 5:

[1392] The server provides the generated test questions to the user terminal.

[1393] The server stores the generated test questions in a database and makes them accessible to users. Users can view the questions through a web app or a dedicated app. The input is the generated test question data, and the output is the questions displayed on the user's device.

[1394] Step 6:

[1395] The user answers the test questions.

[1396] The user answers the provided test questions and sends the answers from the terminal to the server. The input is the user's answer data, and the output is the answer data saved on the server.

[1397] Step 7:

[1398] The emotion analysis means analyzes the emotion of the user.

[1399] The system uses the camera and microphone on the user's device to collect facial expressions and voice while the user is answering test questions. It then uses an emotion analysis library (e.g., EmotionRecognizer) to analyze the user's emotions in real time. The input is camera video and audio data, and the output is analyzed emotion data.

[1400] Step 8:

[1401] The server adjusts the difficulty and format of the test questions based on the user's feelings.

[1402] The server adjusts the difficulty of the test questions based on the analyzed emotional data. For example, if the user is judged to be "frustrated," the server sets the difficulty of the questions low, and if the user is judged to be "confident," the server increases the difficulty. The input is emotional data, and the output is the adjusted test questions.

[1403] Step 9:

[1404] The server provides feedback to the user.

[1405] The system provides users with adjusted test questions and appropriate learning feedback in real time, allowing them to receive feedback that is tailored to their learning situation. The input is the adjusted test questions and feedback data, and the output is the feedback displayed on the user's device.

[1406] 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.

[1407] 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.

[1408] 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.

[1409] [Fourth embodiment]

[1410] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1411] 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.

[1412] 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).

[1413] 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.

[1414] 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.

[1415] 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).

[1416] 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.

[1417] 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.

[1418] 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.

[1419] 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.

[1420] 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.

[1421] 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.

[1422] 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."

[1423] The present invention provides a system for efficiently generating and providing test questions from a learner's learning content. Specific embodiments of this system will be described below.

[1424] System Configuration

[1425] The system mainly consists of the following components:

[1426] User terminal

[1427] server

[1428] User terminal

[1429] A user terminal is a device on which a user uploads learning content, views generated test questions, and answers them. Examples of such devices include personal computers (PCs), smartphones, and tablets.

[1430] server

[1431] The server plays a central role in receiving learning content sent from user devices, analyzing it, generating test questions, and providing them to users. This server is equipped with powerful AI models, databases, text analysis, speech recognition, and other systems.

[1432] User operation (input)

[1433] Users upload the content they want to study (e.g., textbooks, websites, audio files, video files, etc.) from their terminals to the system. At this time, users can select the content and send it to the server in a specific format (e.g., PDF, URL, MP3, MP4, etc.).

[1434] Receiving and storing content

[1435] The server receives the learning content sent from the user's device and saves it in the designated storage. Within this storage, the format of the received file (e.g. PDF, audio, video, etc.) is automatically identified and the appropriate processing is performed.

[1436] Preparing and running content analysis

[1437] The server selects a different parsing algorithm for each file format received: for example, for a PDF file, the server uses a PDF text extraction algorithm to extract text, and for an audio file, it uses a speech recognition API to convert the audio to text.

[1438] Content Analysis

[1439] The AI ​​model on the server analyzes the extracted text to identify important keywords, themes, and key points. This analysis process uses natural language processing techniques to understand sentence structure and meaning.

[1440] Test question generation

[1441] Based on the analysis results, the server generates test questions in various formats (e.g., fill-in-the-blank, multiple-choice, essay, etc.). For example, questions using specific formulas and theorems are generated from mathematics textbooks. The content of the options and fill-in-the-blanks is also automatically generated by the AI ​​model.

[1442] Storage and provision of exam questions

[1443] The generated test questions are stored in a database on the server and made accessible to users. User devices can access and display these question sets through a web app or dedicated app.

[1444] Answers and feedback

[1445] The user answers the generated questions on their device. Once the answer is complete, the server automatically scores the answer and provides immediate feedback to the user. The user can then check the accuracy rate and explanations to assess their own understanding.

[1446] Generate print data

[1447] The system also provides a function for users to download and print the generated problem set in PDF format. When a user requests a printable PDF, the server generates a PDF with high-quality layout and provides it to the user.

[1448] Specific examples

[1449] For example, the following shows the process when using a high school mathematics textbook PDF as learning content:

[1450] 1. User operation: The user uploads a PDF file of a mathematics textbook from the user's terminal.

[1451] 2. Server processing: The server receives the PDF file and saves it in storage.

[1452] 3. Content analysis: Using a PDF text extraction algorithm, the text of the textbook is extracted, and the AI ​​model analyzes the content.

[1453] 4. Question Generation: Generate fill-in-the-blank and multiple-choice questions based on important mathematical formulas and theorems.

[1454] 5. Submit and Answer: The user answers the generated questions in the web app, and the server instantly scores and provides feedback.

[1455] 6. Printable data: When requested by the user, the server converts the problem set into PDF format and provides a download link.

[1456] In this way, this system is a powerful tool for efficiently generating test questions from the learning content desired by the user and for improving learning effectiveness.

[1457] The processing flow will be explained below.

[1458] Step 1:

[1459] The user accesses the learning content upload screen on their device (PC, smartphone, etc.). The user selects and uploads the content they want to study (e.g., textbook PDF, website URL, audio file, video file, etc.).

[1460] Step 2:

[1461] The server receives the content file sent from the user terminal, processes the received file securely, and saves it in the designated storage within the system.

[1462] Step 3:

[1463] The server determines the format of the received file (PDF, URL, audio, video, etc.) and prepares to select an analysis algorithm depending on the file format.

[1464] Step 4:

[1465] The server applies a parsing algorithm based on the file format.

[1466] For PDF: The server uses a PDF text extraction library (e.g., PyMuPDF) to extract text from the PDF.

[1467] For a URL: The server uses a web scraping tool (e.g., BeautifulSoup, Scrapy) to extract the text of the web page.

[1468] For audio files: The server uses a speech recognition API (e.g., Google Cloud Speech-to-Text) to convert the audio to text.

[1469] For video files: The server extracts the audio portion of the video and converts it to text using a speech recognition API.

[1470] Step 5:

[1471] The AI ​​model on the server analyzes the extracted text, using natural language processing techniques to identify important information in the content (keywords, themes, key points, etc.), and builds analytical data for generating questions.

[1472] Step 6:

[1473] The server automatically generates test questions based on the analyzed data. The generated questions are in the following formats:

[1474] Fill-in-the-blank questions: Questions that leave specific keywords or phrases blank and require users to fill them in.

[1475] Multiple choice questions: Questions that provide multiple options and require the user to select the correct answer.

[1476] Essay questions: Questions that require users to write freely.

[1477] Step 7:

[1478] The server stores the generated exam questions in a database, where they are prepared for user access.

[1479] Step 8:

[1480] Users access the generated exam questions on their device, answer them through a web or app interface, and submit their answers once they have completed the questions.

[1481] Step 9:

[1482] The server receives the user's answers and automatically scores them. The server immediately returns the results and feedback to the user, who can then check the answers and evaluate their own understanding.

[1483] Step 10:

[1484] If the user wishes, the server will make the generated test question set available for download in PDF format. When the user clicks the "Generate Printable PDF" button, the server will convert the question set into a high-quality PDF format and provide a download link.

[1485] Step 11:

[1486] Users can download the PDF file from the download link and print it out, allowing them to continue their studies offline.

[1487] The above are the specific processing steps for the system to automatically generate and provide test questions from learning content.

[1488] Example 1

[1489] 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."

[1490] Conventional learning systems require users to manually generate test questions from learning content, which is time-consuming and labor-intensive, and limits the quality and variety of questions generated. Furthermore, when content formats differ, it is difficult to select an appropriate analysis method for each, making efficient question generation impossible. The present invention aims to solve these problems and provide an environment in which users can study efficiently and effectively.

[1491] 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.

[1492] In this invention, the server includes means for receiving study content from a user, means for automatically determining the format of the received content and selecting an appropriate analysis algorithm, means for analyzing the content based on the selected analysis algorithm, means for automatically generating test questions based on the analyzed content, means for providing the generated test questions to the user, means for receiving the user's answers to the test questions and performing automatic grading and feedback, and means for providing the generated test questions in a printable PDF format. This allows users to efficiently generate and answer test questions from content in a variety of formats and have the results evaluated.

[1493] "Content to be studied" refers to information formats such as text data, web page data, audio data, and video data provided by users for study purposes.

[1494] "User" refers to a person who uses the learning system to upload learning content and answer generated exam questions.

[1495] "Means for receiving" refers to the function by which the system receives and records learning content sent by a user.

[1496] "Means for automatically identifying the format and selecting the appropriate analysis algorithm" refers to technology for identifying the format of received content and automatically selecting the analysis method that is optimal for that format.

[1497] "Means of analysis" refers to the function of analyzing the content of learning content using selected algorithms and extracting important information.

[1498] "Means for automatically generating test questions" refers to a function that automatically creates test questions such as fill-in-the-blank questions, multiple-choice questions, and essay questions based on the analyzed content.

[1499] "Means for providing to the user" refers to a function for presenting the generated test questions in a form that can be used by the user.

[1500] "Means for receiving answers and automatically scoring and providing feedback" refers to the function that allows the system to receive the answers provided by the user, determine whether they are correct, and immediately return evaluation results and advice.

[1501] "Means for providing in printable PDF format" refers to the ability to generate generated exam questions as high-quality PDF documents that users can download and print.

[1502] This invention is a system that allows learners to efficiently generate and provide test questions from any learning content. This system is mainly composed of a user terminal and a server.

[1503] User terminal

[1504] A user terminal is a device on which a user uploads learning content, views generated test questions, and answers them. This includes personal computers (PCs), smartphones, tablets, etc. Users access the system using a browser or dedicated application on these devices.

[1505] server

[1506] The server plays a central role in receiving learning content sent from user devices, analyzing it, and generating and providing test questions. The server incorporates powerful AI models, databases, text analysis, speech recognition, and other systems. Specifically, the following software and web services are used:

[1507] PDF Text Extraction Algorithm

[1508] Speech Recognition API

[1509] Natural language processing technology

[1510] Uploading learning content

[1511] Users upload the content they want to study (e.g., textbooks, websites, audio files, video files, etc.) from their terminals to the system. At this time, users can send this content to the server in a specific format (e.g., PDF, URL, MP3, MP4, etc.).

[1512] Receiving and storing content

[1513] The server receives the learning content sent from the user's device and stores it in storage. Within this storage, the format of the received file (e.g., PDF, audio, video, etc.) is automatically identified and the appropriate processing is performed.

[1514] Preparing and running content analysis

[1515] The server selects a different parsing algorithm for each file format received: for example, for a PDF file, the server uses a PDF text extraction algorithm to extract text, and for an audio file, it uses a speech recognition API to convert the audio to text.

[1516] Content Analysis

[1517] The AI ​​model on the server analyzes the extracted text to identify important keywords, themes, and key points. This analysis process uses natural language processing techniques to understand sentence structure and meaning.

[1518] Test question generation

[1519] Based on the analysis results, the server generates test questions in various formats (e.g., fill-in-the-blank, multiple-choice, essay, etc.). For example, questions using specific formulas and theorems are generated from mathematics textbooks. The content of the options and fill-in-the-blanks is also automatically generated by the AI ​​model.

[1520] Storage and provision of exam questions

[1521] The generated test questions are stored in a database on the server and made accessible to users. User devices can access and display these question sets through a web app or dedicated app.

[1522] Answers and feedback

[1523] The user answers the generated questions on their device. Once the answer is complete, the server automatically scores the answer and provides immediate feedback to the user. The user can then check the accuracy rate and explanations to assess their level of understanding.

[1524] Generate print data

[1525] The system also provides a function for users to download and print the generated problem set in PDF format. When a user requests a printable PDF, the server generates a PDF with high-quality layout and provides it to the user.

[1526] Specific examples

[1527] For example, the following shows the process when using a high school mathematics textbook PDF as learning content:

[1528] 1. User operation: The user uploads a PDF file of a mathematics textbook from the user's terminal.

[1529] 2. Server processing: The server receives the PDF file and saves it in storage.

[1530] 3. Content analysis: Using a PDF text extraction algorithm, the text of the textbook is extracted, and the AI ​​model analyzes the content.

[1531] 4. Question Generation: Generate fill-in-the-blank and multiple-choice questions based on important mathematical formulas and theorems. Prompt: "Generate exam questions based on the contents of this mathematics textbook."

[1532] 5. Submit and Answer: The user answers the generated questions in the web app, and the server instantly scores and provides feedback.

[1533] 6. Printable data: When requested by the user, the server converts the problem set into PDF format and provides a download link.

[1534] This system is a powerful tool for efficiently generating test questions from the learning content desired by the user, thereby enhancing learning effectiveness.

[1535] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1536] Step 1:

[1537] The user selects and uploads learning content from the user device. The learning content file (e.g., PDF, URL, MP3, MP4, etc.) is provided as input. The user device then sends the selected file to the server. Specifically, the user clicks the "Upload" button in a browser or dedicated application and selects the file.

[1538] Step 2:

[1539] The server receives learning content sent from the user's device. The file sent by the user is provided as input. After receiving the file, the server saves it in storage and automatically determines the file format (e.g., PDF, URL, MP3, MP4). Specifically, it analyzes the received file name and format, and sorts and saves it in the specified folder.

[1540] Step 3:

[1541] The server selects the appropriate analysis algorithm based on the format of the saved file. The saved file format is provided as input. The server selects the PDF text extraction algorithm for PDF files and the speech recognition API for audio files. Specifically, it uses conditional branching to determine the file format and calls the corresponding analysis algorithm.

[1542] Step 4:

[1543] The server analyzes the content using the selected algorithm. The stored content and the selected analysis algorithm are provided as input. In the case of a PDF file, the server extracts text using a PDF text extraction algorithm, and in the case of an audio file, it converts the audio to text using a speech recognition API. Specifically, the analysis engine extracts information from the file and generates text data.

[1544] Step 5:

[1545] The AI ​​model on the server analyzes the extracted text and identifies important keywords and themes. The extracted text is provided as input. The server uses natural language processing techniques to extract keywords and identify key points. Specifically, the text analysis engine tokenizes the text and extracts themes and key information.

[1546] Step 6:

[1547] The server automatically generates test questions based on the analysis results. The key points and keywords of the analyzed text are provided as input. The server uses the generative AI model to create a prompt sentence and generates test questions based on that. Specifically, the server inputs the prompt sentence into the generative AI model and obtains the generated question sentence.

[1548] Step 7:

[1549] The server saves the generated test questions in a database so that they can be provided to users. The generated test questions are provided as input. The server inserts the generated question data into the database and provides a question set in response to a user request. Specifically, the server stores the question data in the database and provides it to the user's device via a WebAPI.

[1550] Step 8:

[1551] The user answers the generated questions using the user terminal. The test questions provided as input are displayed on the user terminal. After answering, the user sends the answers to the server. Specifically, the user answers the questions on the application and clicks the "Submit" button.

[1552] Step 9:

[1553] The server receives the answers submitted by the user, automatically scores them, and returns the feedback to the user. The user's answer data is provided as input. The server uses a scoring algorithm to determine whether the answer is correct and generates a result. Specifically, the server inputs the answer data into the scoring algorithm, generates a score result, and displays it to the user in real time.

[1554] Step 10:

[1555] When a user requests a printable PDF, the server generates a PDF with a high-quality layout and provides it to the user. The user's request is provided as input. The server converts the problem set into PDF format using a PDF generation tool and generates a download link. Specifically, the server starts the PDF generation process in response to the request and displays a link to the generated PDF file in the web application.

[1556] In this way, through a series of processing steps, the system provides the user with the functionality to efficiently generate test questions from their own learning content, answer them, and evaluate the results.

[1557] (Application example 1)

[1558] 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."

[1559] Conventional learning systems have had problems such as the difficulty for learners to efficiently generate test questions from any learning content, and the inability to properly analyze the format of uploaded content and automatically generate test questions, resulting in a significant lack of user convenience. Another issue is the difficulty in effectively providing the generated test questions to users and providing immediate feedback.

[1560] 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.

[1561] In this invention, the server includes means for receiving study content from a user, means for analyzing the received content, means for automatically generating test questions based on the analyzed content, means for providing the generated test questions to the user, means for extracting text according to the format of the received content, means for generating prompt sentences from the extracted text, means for using the prompt sentences to generate test questions using a generative AI model, and means for providing the generated test questions to the user via the web and accepting answers. This allows learners to efficiently generate test questions from any study content, automates analysis and text extraction, and provides the generated test questions instantly for rapid feedback.

[1562] "Content to be studied" refers to information such as text data, audio data, and video data that is uploaded by a user for study purposes.

[1563] "Means for receiving from users" refers to the interface and processing functions that allow users to upload learning content to the system.

[1564] "Content analysis means" refers to analytical techniques used to understand uploaded content and identify important keywords and themes.

[1565] "Means for automatically generating test questions" refers to algorithms or generative AI models that create appropriate test questions based on the analyzed content.

[1566] The "means for providing test questions to the user" refers to a function for displaying or transmitting the generated test questions to the user's device.

[1567] The "means for extracting text according to the format of received content" refers to a technology for extracting text information in a manner appropriate for the format of received content, such as text, audio, or video.

[1568] The "means for generating prompt sentences" refers to a technology that creates input sentences (prompt sentences) for the generative AI model to generate test questions based on the extracted text.

[1569] A "generative AI model" is an artificial intelligence model for generating test questions based on a given prompt.

[1570] "Means for providing and accepting answers via the web" refers to the interface and processing functions that allow users to access test questions and submit answers via a web browser or dedicated application.

[1571] To implement this invention, a system including the following elements is required: a user terminal, a server, a powerful AI model, text analysis, speech recognition, etc.

[1572] System Configuration

[1573] The system mainly consists of the following components:

[1574] User terminal: A device on which a user uploads learning content, views generated test questions, and answers them. Examples include smartphones, tablets, and personal computers (PCs).

[1575] Server: Plays a central role in receiving learning content sent from user devices, analyzing it, generating and providing test questions. This server is equipped with powerful AI models, databases, text analysis, speech recognition, and other systems.

[1576] User operations

[1577] Users can upload any learning content (e.g., textbooks, websites, audio files, video files, etc.) from their devices. At this time, users can select these contents and send them to the server in a specific format (e.g., PDF, URL, MP3, MP4, etc.).

[1578] Server Processing

[1579] 1. Receiving and storing content

[1580] The server receives the learning content sent from the user's device and saves it in the designated storage. Within this storage, the format of the received file is automatically identified and the appropriate processing is performed.

[1581] 2. Preparing and running content analysis

[1582] The server selects different parsing algorithms for each file format received: for example, for PDF files, the server performs PDF text extraction; for audio files, it uses speech recognition technology to convert speech to text.

[1583] 3. Content Analysis

[1584] The AI ​​model on the server analyzes the extracted text to identify important keywords and themes, using natural language processing techniques to understand sentence structure and meaning.

[1585] 4. Prompt generation

[1586] Based on the extracted text, the AI ​​model creates an input sentence (prompt sentence) for generating test questions, for example, in the format "Please create a test question from the following content: [extracted text]."

[1587] 5. Test Question Generation

[1588] Using the prompt, a generative AI model generates test questions, which can range from fill-in-the-blank questions, multiple-choice questions, and essay questions.

[1589] 6. Provision of test questions and answers

[1590] The generated test questions are stored in a database on the server and made accessible to users, who can then answer them via a web app or a dedicated app.

[1591] Specific examples

[1592] For example, if you use a high school mathematics textbook PDF as learning content, the process is as follows:

[1593] 1. User operation: The user uploads a PDF file of a mathematics textbook from the user terminal to the server.

[1594] 2. Server processing:

[1595] Receive PDF files and save them to storage.

[1596] Using PDF text extraction technology, textbook content is extracted.

[1597] The extracted text is analyzed by an AI model.

[1598] 3. Prompt generation: Based on the extracted text, a prompt is generated: "Please create test questions from the following content: [Textbook content]."

[1599] 4. Test question generation: Using a generative AI model, test questions are generated based on the analysis results.

[1600] 5. Submit and answer: The user answers the generated test questions in the web app, and the server immediately scores and provides feedback.

[1601] Example prompt sentence:

[1602] "Create an exam question from the following content: A description of the growth of morning glories. Morning glories are summer flowers and require watering and sunlight."

[1603] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1604] Step 1:

[1605] Users upload learning content

[1606] Users upload the content they want to study (e.g., PDF files, audio files, video files, etc.) from their own devices to the system. The input here is the study content selected by the user, and the output is the transmission of this content to the server.

[1607] Step 2:

[1608] The server receives and stores the learning content.

[1609] The server receives the learning content sent from the user's device and saves it in the specified storage. The input here is the content file sent by the user, and the output is the content file saved in the storage. At this time, the content format (e.g., PDF, MP3, MP4, etc.) is also automatically determined.

[1610] Step 3:

[1611] The server determines the content type and selects the appropriate parsing algorithm.

[1612] The server selects an appropriate parsing algorithm depending on the format of the received content, for example, a text extraction algorithm for a PDF file, or a speech recognition algorithm for an audio file. The input here is the saved content file and its format information, and the output is the selected parsing algorithm.

[1613] Step 4:

[1614] The server parses the content and extracts the text

[1615] The server analyzes the content using the selected analysis algorithm and extracts the required text data, for example extracting text from a PDF file or converting audio to text from an audio file. The input here is the content file and the selected analysis algorithm, and the output is the extracted text data.

[1616] Step 5:

[1617] The server generates a prompt

[1618] The server creates a prompt based on the extracted text data for the generative AI model to generate questions. An example of a prompt is: "Please create a test question from the following content: [extracted text]". Here, the input is the extracted text data, and the output is the generated prompt.

[1619] Step 6:

[1620] The server generates test questions using the generative AI model

[1621] The server generates test questions by inputting the generative AI model using the generated prompt sentences, where the inputs are the prompt sentences and the generative AI model, and the output is the generated test questions.

[1622] Step 7:

[1623] The server provides the test questions to the user.

[1624] The server stores the generated test questions in a database and makes them accessible to users. Users can access and answer these questions through a web application or a dedicated application. The input here is the generated test questions, and the output is the test questions provided via the web.

[1625] Step 8:

[1626] The user answers the questions and the server provides feedback

[1627] The user answers the provided test questions and sends the answers to the server, which instantly marks the answers and provides feedback to the user. The input here is the user's answer, and the output is the marking results and feedback.

[1628] 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.

[1629] This invention is a system that generates test questions from a user's study content and provides personalized questions according to the user's emotions by combining it with an emotion engine. Specific embodiments of this system are described below.

[1630] System Configuration

[1631] The system mainly consists of the following components:

[1632] User terminal

[1633] server

[1634] Emotion Engine

[1635] User terminal

[1636] A user device refers to the device on which a user uploads learning content, views generated test questions, and answers them. This includes PCs, smartphones, tablets, etc. The device is also equipped with a camera and microphone, and the emotion engine uses these devices to analyze the user's emotions.

[1637] server

[1638] The server plays a central role in receiving learning content sent from user devices, analyzing it, generating test questions, and providing them to users. This server is equipped with systems such as AI models, databases, text analysis, and speech recognition.

[1639] Emotion Engine

[1640] The emotion engine is a technology that recognizes emotions from the user's facial expressions and tone of voice, allowing it to analyze the user's emotions while answering test questions and provide appropriate feedback and adjust the questions in real time.

[1641] User operation (input)

[1642] Users upload the content they want to study (textbooks, websites, audio files, video files, etc.) from their devices. When uploading content, users also set up their devices to allow the use of the emotion engine.

[1643] Receiving and storing content

[1644] The server receives the content file sent from the user's device and saves it in the specified storage. The format of the received file (PDF, URL, audio, video, etc.) is automatically determined and the appropriate processing is performed.

[1645] Preparing and running content analysis

[1646] The server selects a different parsing algorithm for each file format received: for example, for PDF files, it uses a PDF text extraction algorithm to extract text, and for audio files, it uses a speech recognition API to convert speech to text.

[1647] Content Analysis

[1648] An AI model on the server analyzes the extracted text to identify important keywords, themes, and key points. This analysis process uses natural language processing techniques to understand sentence structure and meaning.

[1649] Test question generation

[1650] Based on the analysis results, the server generates test questions in various formats (fill-in-the-blank, multiple choice, essay, etc.), so that the generated questions are in a format that is appropriate for the user's learning content.

[1651] Storage and provision of exam questions

[1652] The server stores the generated test questions in a database and makes them available for users to access. User devices can access and display these question sets through a web app or a dedicated app.

[1653] Emotion Recognition and Feedback

[1654] As users answer questions, the emotion engine uses camera footage and audio data to recognize their emotions. Based on the recognized emotions, the server adjusts the difficulty of the questions and provides timely feedback and support messages.

[1655] Answers and feedback

[1656] When a user answers a question, the server automatically scores the answer and provides immediate feedback, taking into account emotional data recognized by the emotion engine, according to the user's level of understanding and state.

[1657] Generate print data

[1658] The system also provides users with the ability to download and print the generated problem sets in PDF format if they wish, allowing them to study offline.

[1659] Specific examples

[1660] The steps for using a high school mathematics textbook PDF as learning content are as follows:

[1661] 1. User operation: The user uploads a PDF file of a mathematics textbook. The user allows the emotion engine to be used.

[1662] 2. Server processing: Receive the PDF file, save it in storage, extract the text, and analyze it using the AI ​​model.

[1663] 3. Question generation: Based on the analyzed content, fill-in-the-blank and multiple-choice questions are generated.

[1664] 4. Submit and Answer: The user answers the generated questions in the web app.

[1665] 5. Emotion recognition and feedback: The emotion engine analyzes the user's facial expressions and voice and adjusts feedback and problem difficulty in real time.

[1666] 6. Answer result: Once the answer is completed, the server immediately scores it and provides feedback that takes into account emotional data.

[1667] 7. Printable data: Upon user request, the problem set will be converted into PDF format and a download link will be provided.

[1668] As described above, the present invention efficiently generates test questions from a user's learning content and combines them with an emotion engine to provide a personalized learning experience that responds to the user's emotions.

[1669] The processing flow will be explained below.

[1670] Step 1:

[1671] The user accesses the learning content upload screen on their device (PC, smartphone, etc.). The user selects and uploads the content they want to study (textbook PDF, website URL, audio file, video file, etc.). They also set up permission to use the emotion engine.

[1672] Step 2:

[1673] The server receives the content file sent from the user's device. The received file is safely processed and saved in the designated storage within the system. The emotion engine usage settings are also saved at the same time.

[1674] Step 3:

[1675] The server determines the format of the received file (PDF, URL, audio, video, etc.) and prepares to select an analysis algorithm depending on the file format.

[1676] Step 4:

[1677] The server applies a parsing algorithm based on the file format.

[1678] For PDF: The server uses a PDF text extraction library (e.g., PyMuPDF) to extract text from the PDF.

[1679] For a URL: The server uses a web scraping tool (e.g., BeautifulSoup, Scrapy) to extract the text of the web page.

[1680] For audio files: The server uses a speech recognition API (e.g., Google Cloud Speech-to-Text) to convert the audio to text.

[1681] For video files: The server extracts the audio portion of the video and converts it to text using a speech recognition API.

[1682] Step 5:

[1683] The AI ​​model on the server analyzes the extracted text, using natural language processing techniques to identify important information in the content (keywords, themes, key points, etc.), and builds analytical data for generating questions.

[1684] Step 6:

[1685] The server automatically generates test questions based on the analyzed data. The generated questions are in the following formats:

[1686] Fill-in-the-blank questions: Questions that leave specific keywords or phrases blank and require users to fill them in.

[1687] Multiple choice questions: Questions that provide multiple options and require the user to select the correct answer.

[1688] Essay questions: Questions that require users to write freely.

[1689] Step 7:

[1690] The server stores the generated exam questions in a database, where they are prepared for user access.

[1691] Step 8:

[1692] Users access the test questions generated on their own devices and answer them through a web or app interface. As they begin answering questions, the emotion engine uses camera footage and audio data to recognize the user's emotions in real time.

[1693] Step 9:

[1694] The server receives and analyzes the user's emotional data, and based on that data, adjusts the difficulty of the test questions provided and provides feedback and support messages at appropriate times.

[1695] Step 10:

[1696] Once the user has completed the questions, the server automatically scores the answers and provides immediate feedback, taking into account the emotional data recognized by the emotion engine, providing detailed feedback tailored to the user's level of understanding and state.

[1697] Step 11:

[1698] If the user wishes, the server will make the generated test question set available for download in PDF format. When the user clicks the "Generate Printable PDF" button, the server will convert the question set into a high-quality PDF format and provide a download link.

[1699] Step 12:

[1700] Users can download the PDF file from the download link and print it out, allowing them to continue their studies offline.

[1701] As a concrete example, let's consider the case where a user uses a high school mathematics textbook PDF as learning content:

[1702] 1. User operation: The user uploads a PDF file of a mathematics textbook. The user allows the emotion engine to be used.

[1703] 2. Server processing: The PDF file is received and stored in storage. The text is extracted and analyzed using the AI ​​model.

[1704] 3. Question generation: Based on the analyzed content, fill-in-the-blank and multiple-choice questions are generated.

[1705] 4. Providing and answering: The user answers the generated questions in the web app. The emotion engine analyzes the user's facial expressions and voice.

[1706] 5. Emotion recognition and feedback: The emotion engine recognizes the user's emotions in real time, and the server provides appropriate feedback and problem adjustments.

[1707] 6. Answer result: Once the answer is completed, the server immediately scores it and provides feedback that takes into account emotional data.

[1708] 7. Printable data: Upon user request, the problem set will be converted into PDF format and a download link will be provided.

[1709] As described above, the present invention efficiently generates test questions from learning content and combines them with an emotion engine to provide a personalized learning experience that responds to the user's emotions.

[1710] Example 2

[1711] 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."

[1712] While conventional learning systems can generate test questions from users' learning content, they have the problem of being unable to provide personalized feedback or adjust questions based on the user's emotions. Furthermore, because they do not take into account the user's level of understanding or emotional state, efficient and effective learning is difficult.

[1713] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving study content from a user, means for analyzing the received content, means for automatically generating test questions based on the analyzed content, means for providing the generated test questions to the user, means for recognizing the user's emotions, and means for providing feedback and adjusting the questions in real time based on the recognized emotions. This enables a personalized learning experience that takes into account the user's emotional state.

[1714] "Content to be studied" refers to information provided by a user for study purposes, and includes formats such as text data, web page data, audio data, and video data.

[1715] "Means for receiving from users" refers to interfaces and functions that allow users to upload content to be studied to the system.

[1716] "Means for analyzing the content" refers to technology that mechanically analyzes the received learning content and identifies important keywords, themes, key points, etc.

[1717] "Means for automatically generating test questions" refers to a function that automatically creates fill-in-the-blank questions, multiple-choice questions, essay questions, etc. based on the content of the analyzed content.

[1718] "Means for providing test questions to users" refers to an interface or function that allows users to access and answer the generated test questions.

[1719] "Means for recognizing user emotions" refers to technology that automatically determines emotions from a user's facial expressions and tone of voice.

[1720] "Means for adjusting feedback and questions in real time" refers to a mechanism that instantly and adaptively changes the learning content, difficulty of questions, and feedback based on the recognized user emotions.

[1721] MODE FOR CARRYING OUT THE INVENTION

[1722] This invention is a system that generates test questions from learning content provided by a user and provides feedback according to the user's emotions using an emotion engine. Specific embodiments of this system are described below.

[1723] System Configuration

[1724] The system mainly consists of the following components:

[1725] 1. User Device

[1726] 2. Server

[1727] 3. Emotion Engine

[1728] User terminal

[1729] A user device refers to the device on which a user uploads learning content and views and answers generated test questions. User devices include PCs, smartphones, tablets, etc. These devices are equipped with cameras and microphones, which the emotion engine uses to analyze the user's emotions.

[1730] server

[1731] The server receives learning content sent from the user's device, analyzes the content, generates test questions, and provides them. This server incorporates the following systems:

[1732] AI models (e.g., generative AI models)

[1733] Database (e.g. PostgreSQL)

[1734] Text Analysis Algorithms

[1735] Speech recognition API (e.g., Google Cloud's speech recognition API)

[1736] The server automatically determines the format of the uploaded learning content (PDF, URL, audio, video, etc.) and processes it appropriately.

[1737] Emotion Engine

[1738] The emotion engine is a technology that recognizes emotions from the user's facial expressions and tone of voice. Specifically, it collects camera footage and audio data while the user is answering test questions and analyzes it using technologies such as Microsoft's Emotion API.

[1739] Specific examples

[1740] The steps for using a high school mathematics textbook PDF as learning content are as follows:

[1741] 1. User operation: A user uploads a PDF file of a mathematics textbook. When uploading, the user selects the option to allow the use of the emotion engine.

[1742] 2. Server processing: The server receives the PDF file, stores it in the storage system, extracts the text using a PDF text extraction algorithm, and analyzes it using an AI model.

[1743] 3. Question generation: Based on the analyzed content, fill-in-the-blank questions, multiple-choice questions, etc. are generated.

[1744] 4. Submit and answer: Users answer questions generated through a web app or dedicated app.

[1745] 5. Emotion recognition and feedback: The emotion engine analyzes the user's facial expressions and voice and provides real-time feedback and adjusts the difficulty of the questions.

[1746] 6. Answer result: After the answer is given, the server automatically scores it and provides feedback that takes into account emotional data.

[1747] 7. Print Data Generation: If the user wishes, convert the problem set into PDF format and provide a download link.

[1748] Prompt Sentence Examples

[1749] "I have uploaded a PDF of a high school mathematics textbook. Please extract important keywords and themes from this content and generate fill-in-the-blank and multiple-choice questions based on them. Also, please adjust the difficulty of the questions by analyzing the user's emotions when answering them."

[1750] The above is a specific embodiment of this system, which efficiently generates test questions from the user's learning content and combines it with an emotion engine to provide a personalized learning experience that responds to the user's emotions.

[1751] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1752] Step 1:

[1753] The user device uploads learning content. The user clicks the "Upload Content" button on the web app, and a file selection dialog appears. The user selects the file, checks the option to allow the use of the emotion engine, and clicks the upload button.

[1754] Input: User's learning content file, emotion engine permission settings

[1755] Output: Learning content is sent from the user device to the server.

[1756] Step 2:

[1757] The server receives the learning content and stores it in storage. The server receives a file upload request from the user device and stores the file in temporary storage. The server then moves the stored file to a specific location in the storage system (e.g., Amazon S3).

[1758] Input: Learning content sent from the user's device

[1759] Output: Learning content files saved to storage

[1760] Step 3:

[1761] The server analyzes the file format and selects the appropriate analysis algorithm. The server analyzes the metadata of the uploaded file and determines the file format (PDF, URL, audio, video, etc.).

[1762] Input: Learning content files saved in storage

[1763] Output: Analysis algorithms depending on the file format

[1764] Step 4:

[1765] The server analyzes the content and runs the analysis algorithm it has selected. For PDF files, it uses the PDF text extraction algorithm, and for audio files, it uses a speech recognition API (e.g., Google Cloud's speech recognition API).

[1766] Input: Parsing algorithm depending on the file format

[1767] Output: Extracted text data

[1768] Step 5:

[1769] The AI ​​model on the server analyzes the extracted text data, using natural language processing techniques to identify important keywords, themes, and key points. This analysis is performed using tools such as Amazon Comprehend.

[1770] Input: Extracted text data

[1771] Output: Analyzed keywords, themes, and key points data

[1772] Step 6:

[1773] The server generates test questions based on the analysis results. The server calls up a generative AI model and automatically creates fill-in-the-blank, multiple-choice, and essay questions based on the analyzed keywords and themes.

[1774] Input: Analyzed keywords, themes, and key points data

[1775] Output: Generated test question data

[1776] Step 7:

[1777] The server saves the generated test questions in a database and prepares them for serving to users.The server saves the generated questions in a database (e.g., PostgreSQL) and generates a URL for the question set via an API so that it can be accessed from a web app or dedicated app.

[1778] Input: Generated test question data

[1779] Output: Question data stored in the database, URL for accessing it

[1780] Step 8:

[1781] The user's device displays the test questions, and the user answers them. The user accesses the provided URL and the questions are displayed on the web app. The user answers the questions in the browser.

[1782] Input: The URL the user accessed in the web app

[1783] Output: User's answers to the questions

[1784] Step 9:

[1785] The emotion engine recognizes the user's emotions while they are answering. While the user is entering their answer, the emotion engine periodically collects camera footage and microphone audio, and calls the Emotion API to obtain the analysis results.

[1786] Input: User's camera video and audio data

[1787] Output: Parsed emotion data

[1788] Step 10:

[1789] The server provides real-time feedback and adjusts the difficulty level. The server recalculates the difficulty of the questions based on the emotional data sent from the emotion engine, and displays feedback messages or easy questions as needed.

[1790] Input: Parsed emotion data and user response data

[1791] Output: Adjusted difficulty of the problem, feedback message

[1792] Step 11:

[1793] The server scores the user's answers and provides feedback. When the user submits their answer, the server automatically scores it and provides immediate feedback that takes into account emotional data.

[1794] Input: User's answer data

[1795] Output: Marking results and feedback

[1796] Step 12:

[1797] The server generates the printable data and provides it to the user. If the user wishes, the server converts the generated problem set into PDF format and provides a download link.

[1798] Input: User request

[1799] Output: PDF problem set and download link

[1800] The above are the specific processing steps of this system.

[1801] (Application example 2)

[1802] 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."

[1803] Conventional learning systems have struggled to provide a personalized learning experience that responds to the user's emotions. This has resulted in a lack of appropriate feedback and support based on emotions such as stress and excitement felt during learning, resulting in reduced learning efficiency. Furthermore, conventional learning systems often lacked sufficient support for factory employees to efficiently master specific skills and operations.

[1804] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1805] In this invention, the server

[1806] means for receiving content to be studied from a user;

[1807] means for analyzing the content of the received content;

[1808] means for automatically generating test questions based on the analyzed content;

[1809] means for providing the generated test questions to a user;

[1810] A sentiment analysis means;

[1811] a means for adjusting the difficulty and format of test questions based on user sentiment;

[1812] Includes.

[1813] This allows for a personalized learning experience by understanding employees' emotions in real time while they are studying and providing appropriate feedback. Furthermore, the generated test questions are dynamically adjusted according to the user's emotions, maximizing the effectiveness of learning. Furthermore, using emotion analysis technology can help factory employees efficiently master operations and skills, improving overall production efficiency.

[1814] "Content to be studied" refers to information or learning materials that a user wishes to study, and includes text data, web page data, audio data, video data, and the like.

[1815] "Means for analysis" refers to the technologies and algorithms used to understand the content of received content and extract important parts, including natural language processing and speech recognition.

[1816] "Means for automatically generating test questions" refers to a system that generates questions in a format appropriate for the user's learning content based on analyzed content, and uses an AI model or a generative AI model.

[1817] "Means for providing" refers to devices or applications for providing the generated test questions in a form that users can access, including smartphone apps and web apps.

[1818] "Emotion analysis means" refers to technology or devices for analyzing emotions from a user's facial expressions and voice, and includes image analysis and voice analysis using a camera or microphone.

[1819] "Feedback" refers to instructions or comments provided based on a user's answers and feelings, which can help improve the effectiveness of learning.

[1820] "Means for adjusting difficulty and format" refers to a system that changes the difficulty and format of test questions according to the user's emotions and learning situation, and is dynamically adjusted in real time.

[1821] A "personalized learning experience" aims to maximize learning effectiveness by providing learning methods and content optimized for each user's individual learning situation and emotions.

[1822] "Real-time" means reacting and responding immediately to changes in the user's learning and emotions.

[1823] "Downloadable in PDF format" means that the generated exam questions are saved as electronic files and are available for users to download via the Internet.

[1824] The system necessary to implement this invention mainly comprises the following elements: a server, a user terminal, an emotion analysis means, and an AI model.

[1825] server

[1826] The server plays a central role in analyzing learning content received from users and automatically generating test questions. Specifically, the server receives content uploaded by users, such as PDFs, web pages, audio, and video, and extracts text using appropriate analysis algorithms. The extracted text is analyzed using natural language processing (NLP) techniques to identify important keywords and themes. Test questions are then automatically generated based on the analysis results using an AI model. These test questions are then made available to users for access, viewing, and answering via a web app or dedicated app.

[1827] User terminal

[1828] User devices include PCs, smartphones, tablets, etc., and are used by users to upload learning content and answer generated test questions. User devices are equipped with cameras and microphones, which are used as emotion analysis tools. This makes it possible to analyze emotions in real time from the user's facial expressions and tone of voice, and send the results to the server.

[1829] Emotion analysis means

[1830] Emotion analysis is a technology that uses a camera and microphone to analyze the user's facial expressions and voice. Specifically, it collects camera footage and microphone audio from the user's device and analyzes the user's emotions in real time using the EmotionRecognizer library. This emotional data is sent to a server and used to dynamically adjust the difficulty and format of test questions according to the user's emotions.

[1831] Feedback and Adjustments

[1832] The generated test questions are provided to the user, and as the user answers them, the server monitors the user's state through emotional analysis. For example, if the user is feeling "frustrated," the server will lower the difficulty of the questions based on the emotional data, reducing the user's stress and improving learning efficiency. Similarly, if the user is determined to be "confident," the server will provide questions of increased difficulty. This ensures that the user always has the optimal learning experience.

[1833] Specific examples

[1834] For example, suppose a factory employee uploads "Robot Operation Manual.pdf" to the system as learning content. The system extracts text from the PDF file and analyzes important operating procedures and troubleshooting methods. Based on this analysis, test questions are generated to deepen the employee's understanding and are provided through the smart glasses. When the employee answers the test questions, the smart glasses' camera and microphone are used to analyze their emotions. If the system detects "frustration," it automatically adjusts the difficulty of the questions and provides appropriate feedback.

[1835] Prompt Sentence Examples

[1836] An example of a prompt sentence to input to the generative AI model is:

[1837] "You have an employee learning how to operate a robot. If the analysis reveals that his state is 'frustrated,' create questions that are easy to understand. Conversely, if the analysis reveals that he is 'confident,' create questions that are difficult to understand."

[1838] Instructions include:

[1839] In this way, the present invention is a system that maximizes learning effectiveness by providing a personalized learning experience according to the user's emotions.

[1840] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1841] Step 1:

[1842] Users upload learning content.

[1843] The user uploads the content they want to learn (e.g., "Robot Operation Manual.pdf") from their device to the system. The device receives the file and sends it to the server. The input is the learning content file, and the output is the content saved on the server.

[1844] Step 2:

[1845] The server stores the received content and prepares it for analysis.

[1846] The server saves the uploaded file in storage and automatically detects the file format. For example, if it is a PDF file, the server uses a PDF text extraction algorithm to extract the text. The input is the received content and the output is the data ready to be parsed.

[1847] Step 3:

[1848] The server analyzes the learning content.

[1849] The server uses natural language processing (NLP) techniques to analyze the text and identify important keywords and themes. The specific software used here is an NLP library. The input is prepared text data, and the output is the analyzed keywords and themes.

[1850] Step 4:

[1851] The server generates test questions based on the analysis results.

[1852] The server uses a generative AI model to automatically generate test questions based on the analyzed keywords and themes. The generated questions can be fill-in-the-blank, multiple choice, or essay questions. The input is the analysis results, and the output is the generated test questions.

[1853] Step 5:

[1854] The server provides the generated test questions to the user terminal.

[1855] The server stores the generated test questions in a database and makes them accessible to users. Users can view the questions through a web app or a dedicated app. The input is the generated test question data, and the output is the questions displayed on the user's device.

[1856] Step 6:

[1857] The user answers the test questions.

[1858] The user answers the provided test questions and sends the answers from the terminal to the server. The input is the user's answer data, and the output is the answer data saved on the server.

[1859] Step 7:

[1860] The emotion analysis means analyzes the emotion of the user.

[1861] The system uses the camera and microphone on the user's device to collect facial expressions and voice while the user is answering test questions. It then uses an emotion analysis library (e.g., EmotionRecognizer) to analyze the user's emotions in real time. The input is camera video and audio data, and the output is analyzed emotion data.

[1862] Step 8:

[1863] The server adjusts the difficulty and format of the test questions based on the user's feelings.

[1864] The server adjusts the difficulty of the test questions based on the analyzed emotional data. For example, if the user is judged to be "frustrated," the server sets the difficulty of the questions low, and if the user is judged to be "confident," the server increases the difficulty. The input is emotional data, and the output is the adjusted test questions.

[1865] Step 9:

[1866] The server provides feedback to the user.

[1867] The system provides users with adjusted test questions and appropriate learning feedback in real time, allowing them to receive feedback that is tailored to their learning situation. The input is the adjusted test questions and feedback data, and the output is the feedback displayed on the user's device.

[1868] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1869] 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.

[1870] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1871] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1872] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1873] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1874] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1875] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1876] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1877] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1878] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1879] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1880] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1881] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1882] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1883] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1884] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1885] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1886] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1887] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1888] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1889] The following is further disclosed regarding the above embodiment.

[1890] (Claim 1)

[1891] means for receiving content to be studied from a user;

[1892] means for analyzing the content of the received content;

[1893] means for automatically generating test questions based on the analyzed content;

[1894] means for providing the generated test questions to a user;

[1895] A system including:

[1896] (Claim 2)

[1897] 2. The system according to claim 1, wherein the test questions generated based on the analyzed content are in the form of fill-in-the-blank questions, multiple-choice questions, or essay questions.

[1898] (Claim 3)

[1899] 2. The system according to claim 1, wherein the received study content is at least one of text data, web page data, audio data, and video data.

[1900] (Claim 4)

[1901] 2. The system according to claim 1, further comprising a means for displaying the generated test questions on a user terminal and for allowing the user to answer and check the answers.

[1902] (Claim 5)

[1903] 10. The system of claim 1, further c...

Claims

1. means for receiving content to be studied from a user; means for analyzing the content of the received content; means for automatically generating test questions based on the analyzed content; means for providing the generated test questions to a user; A system including:

2. 2. The system according to claim 1, wherein the test questions generated based on the analyzed content are in the form of fill-in-the-blank questions, multiple-choice questions, or essay questions.

3. 2. The system according to claim 1, wherein the received learning content is at least one of text data, web page data, audio data, and video data.

4. 2. The system according to claim 1, further comprising means for displaying the generated test questions on a user terminal and for allowing the user to answer and check the answers.

5. 10. The system of claim 1, further comprising means for allowing a user to download and print the generated test questions in PDF format.

Images