System

Abstract

A system is provided.SOLUTION: A learning support system comprising: means for capturing an image by a user and extracting character information from the image; means for transmitting the extracted character information; means for analyzing the transmitted character information and determining whether a question is correct or incorrect; means for displaying a determination result to the user; means for automatically generating background information based on the correct / incorrect determination; and means for comparing a score of the user with a nationwide score.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, especially students preparing for exams and working adults taking qualification exams, are seeking effective learning support tools. Traditional learning tools are limited to simply checking whether test questions are correct or incorrect, and do not provide sufficient background or supplementary information to foster deeper understanding. Furthermore, there are limited ways for learners to compare their performance with the national level and understand their position. This makes it difficult for learners to maintain their motivation to study and effectively address their weaknesses. [Means for solving the problem]

[0005] In order to solve the above problems, the present invention provides the following means. First, it includes means for a user to take an image and extract text information from the image. Next, it includes means for transmitting the extracted text information and analyzing the information to determine whether the question is correct. Furthermore, it provides means for displaying the determination result to the user and automatically generating background information based on the determination. This background information can include individual supplemental information based on the user's learning history. Finally, it provides a learning support system that displays the generated background information to the user and includes means for comparing the user's performance with national performance. This system enables learners to improve their own learning effectiveness and effectively overcome their weaknesses.

[0006] "Taking pictures" refers to a user using the camera function of a smartphone or other electronic device to take a photo of a test question or document.

[0007] "Means for extracting character information" refers to technology (such as OCR technology) for recognizing characters from a captured image and generating text data.

[0008] "Transmitting means" refers to a function for transmitting extracted character information to an external server or other system via a communication network.

[0009] "Means for analyzing and determining whether the questions are correct" refers to algorithms and technologies for analyzing test questions and answers based on the transmitted text information and determining whether they are correct or incorrect.

[0010] "Means for displaying the judgment result to the user" refers to a display or application user interface for visually presenting the judged correct / incorrect result to the user.

[0011] "Means for automatically generating background information" refers to artificial intelligence and database search technology for generating additional relevant information based on the content of the question and its correct / incorrect judgment.

[0012] "Individual supplementary information based on learning history" refers to data and content that analyzes the content a user has studied in the past and their academic performance history, and provides each user with additional learning resources and information that are optimal for that user.

[0013] "Means for comparing performance with national performance" refers to technology or algorithms that aggregate an individual user's performance data and compare it with the performance data of other learners collected nationally.

[0014] "Learning support system" refers to a system in general that integrates and provides the above-mentioned means to comprehensively support users' learning activities. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11]FIG. 3 is a sequence diagram illustrating a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0016] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0018] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0019] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0020] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0021] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0022] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0023] [First embodiment]

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

[0025] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0026] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0027] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0028] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0029] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0030] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0032] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0033] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0034] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0035] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0036] The "learning support system" of this invention provides an integrated set of functions, including image recognition, natural language processing, performance analysis, and information provision, to fully support users' learning activities. This system is implemented as an application that users can easily use on devices such as smartphones and tablets.

[0037] 1. Photographing and recognizing test questions

[0038] Terminal

[0039] First, the user takes a photo of the test question using the smartphone camera. The device analyzes the captured image using its OCR (Optical Character Recognition) function and extracts the characters in the image as text information. This extracted text information is then sent to the server.

[0040] server

[0041] The server inputs the received text information into the generative AI model, analyzes the content of the question, and judges whether it is correct or not. The analyzed result of the judgment is then sent back to the device.

[0042] Terminal

[0043] The device displays the result of the accuracy judgment received from the server to the user. For example, if the user answers "1467" to the history question "What year did the Onin War start?", the result of the accuracy judgment is displayed as "correct answer."

[0044] 2. Providing background information

[0045] server

[0046] Based on the results of the correct / incorrect decision, the server automatically generates relevant background information, specifically by extracting historical background and additional explanatory information related to the problem from natural language generation (NLG) models and pre-built databases.

[0047] Terminal

[0048] The generated background information is sent to the terminal and displayed to the user. For example, detailed background information about the Onin War and the story of its outcome are displayed.

[0049] 3. Viewing learning history and supplementary information

[0050] Terminal

[0051] The user's learning history is stored in a local database, which records the correct / incorrect results for each question, the date and time of study, and so on.

[0052] server

[0053] The server analyzes the user's learning history data and generates supplementary information based on the user's learning patterns and interests. For example, if a user is not good at history, additional learning materials and video content that will help them improve their understanding will be recommended.

[0054] Terminal

[0055] The generated supplemental information is sent to the device and displayed to the user, and is displayed as a notification, allowing the user to further study based on it.

[0056] 4. Comparison of performance at the national level

[0057] server

[0058] The server compiles and analyzes the performance information of learners nationwide to generate statistical data, which makes it possible to compare each user's performance with the national average and deviation score.

[0059] Terminal

[0060] The user's academic performance data is sent to the server, and a comparison with the national average is sent back to the terminal, allowing the user to understand where their academic performance ranks nationally.

[0061] Specific examples

[0062] For example, suppose a user photographs a history test question, "What year did the Onin War start?" and answers "1467." In this case, the process would proceed as follows:

[0063] 1. The user takes a photo of the problem with their smartphone.

[0064] 2. The device uses its OCR function to extract "What year did the Onin War start?" and "1467" from the image and send it to the server.

[0065] 3. The server uses the generated AI model to determine whether the answer is correct, determines it is correct, and automatically generates background information about the progress and outcome of the Onin War.

[0066] 4. The device displays the correct / incorrect result and background information to the user.

[0067] 5. The server generates relevant supplementary information based on the user's learning history and sends it to the terminal.

[0068] 6. The terminal notifies and displays the supplementary information to the user.

[0069] 7. The server compares the user's grades with national data, calculates the standard deviation score, etc., and sends it to the terminal.

[0070] 8. The device displays the results of the comparison to the user.

[0071] In this way, the present invention is a system that supports the user's learning activities in many ways and enables more effective learning.

[0072] The processing flow will be explained below.

[0073] Step 1:

[0074] The user starts the smartphone app and takes a picture of the test paper. When the user presses the capture button, the smartphone's camera function is activated and an image of the test paper is captured.

[0075] Step 2:

[0076] The device analyzes the captured image using its OCR (Optical Character Recognition) function. The OCR engine extracts character data from the image data and converts it into text format.

[0077] Step 3:

[0078] The device sends the extracted text information to the server, using a communication network to generate an API request and send the data to a specified endpoint on the server.

[0079] Step 4:

[0080] The server inputs the received text information into a generative AI model, which analyzes the content of the problem and uses natural language processing (NLP) algorithms to distinguish between the question and the answer.

[0081] Step 5:

[0082] The server determines whether the question is correct, refers to an internal database or knowledge base, evaluates whether the analyzed answer is correct, formats the evaluation result, and generates a data packet to send back to the terminal.

[0083] Step 6:

[0084] The device displays the correctness judgment result received to the user. The result is reflected in the user interface (UI), and a message such as "correct" or "incorrect" is displayed.

[0085] Step 7:

[0086] The server generates background information based on the results of the accuracy assessment, automatically generating relevant historical context and additional explanations using a natural language generation (NLG) model, or searching and extracting relevant information from a database.

[0087] Step 8:

[0088] The server sends the generated background information to the device, which then converts it into JSON format and sends it to the device via the communication network.

[0089] Step 9:

[0090] The device displays the background information received to the user, providing detailed historical background and additional explanations in an easy-to-understand manner to help the user gain a deeper understanding.

[0091] Step 10:

[0092] The device stores the user's learning history in a local database, accumulating information such as the questions studied, answers, correct / incorrect results, and learning dates and times.

[0093] Step 11:

[0094] The server analyzes the user's learning history data, generates supplementary information based on the user's learning patterns and preferences, and recommends appropriate learning materials and video content.

[0095] Step 12:

[0096] The server generates supplementary information and sends it to the terminal. Information tailored to each user's individual learning needs is sent, improving learning efficiency.

[0097] Step 13:

[0098] The device notifies and displays the supplementary information it receives to the user. The supplementary information is displayed in the notification bar or a dedicated screen within the app, making it easy for the user to access.

[0099] Step 14:

[0100] The server aggregates and analyzes the performance data collected from other users nationwide, and uses statistical algorithms to calculate the national average and standard deviation.

[0101] Step 15:

[0102] The server generates information comparing the user's performance with the national average and sends it to the device, including the difference between the individual user's performance and the national average, as well as the deviation score.

[0103] Step 16:

[0104] The terminal displays the received performance comparison results to the user, visualizing the results in the form of graphs and charts to help the user clearly understand their position.

[0105] Through the above steps, the present invention is a system that supports users' learning in many ways and promotes efficient and deep understanding.

[0106] Example 1

[0107] 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."

[0108] Conventional learning support systems have limitations in their ability to accurately extract text information from images taken by users and determine whether questions are correct or incorrect. Furthermore, they lack the functionality to analyze learning history and provide users with appropriate supplementary information, or the ability to compare scores nationwide in real time. This makes it difficult for users to fully grasp their learning progress and level of understanding, making it difficult to study efficiently.

[0109] 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.

[0110] In this invention, the server includes means for a user to take an image and extract text information from the image, means for transmitting the extracted text information, means for analyzing the transmitted text information and judging whether the question is correct or incorrect, means for displaying the judgment result to the user, means for automatically generating background information based on the judgment of correctness, means for displaying the generated background information to the user, means for comparing the user's performance with national performance, means for accumulating and analyzing the user's learning history, means for generating supplementary information based on the analysis result and notifying the user, and means for aggregating and analyzing national performance data and comparing the user's performance with the national average and deviation score. This allows the user to grasp their own learning progress and level of understanding in real time and to study efficiently and effectively.

[0111] A "user" is a person who accesses the learning support system using a device such as a smartphone or tablet and engages in learning activities.

[0112] "Taking an image" refers to the act of a user using the camera on a smartphone or tablet to capture an image of a study question or document.

[0113] "Extracting text information" refers to the process of using OCR (Optical Character Recognition) technology to identify and extract text from a captured image as text data.

[0114] "Send" refers to the act of transferring data from a terminal to a server.

[0115] "Analysis" refers to the process in which the server analyzes the data it receives using technologies such as generative AI models to determine the content of the question and whether it is correct or incorrect.

[0116] "Determining correctness" means using a generative AI model to determine whether an answer to a question is correct or incorrect.

[0117] "Display" refers to the act of visually presenting analysis results, background information, performance information, etc. to the user on the terminal display.

[0118] "Background information" is information containing additional knowledge or explanation related to the content of the problem.

[0119] "Auto-generation" refers to the process of automatically generating the required information using pre-built databases and natural language generation (NLG) models.

[0120] "Study history" refers to a record of a user's past learning activities, answer results, study dates and times, etc.

[0121] "Analysis" refers to clarifying a user's learning patterns and tendencies based on accumulated learning history data.

[0122] "Supplemental information" refers to additional educational materials and video content provided to help users learn more efficiently.

[0123] "Notification" refers to the act of informing the user of generated information, supplementary materials, etc.

[0124] "National level results" refers to statistical data such as averages and standard deviations obtained by compiling and analyzing the results data of other users nationwide.

[0125] The "learning support system" of this invention provides an integrated set of functions, including image recognition, natural language processing, performance analysis, and information provision, to fully support users' learning activities. This system is implemented as an application that users can easily use on devices such as smartphones and tablets.

[0126] Details of the hardware and software you will use

[0127] Terminal

[0128] This system mainly uses mobile devices such as smartphones and tablets. The devices are equipped with a camera, OCR (Optical Character Recognition) functionality, and a display. The OCR functionality incorporates common OCR software and is used to identify characters from images.

[0129] System processing flow

[0130] 1. Photographing test questions and character recognition

[0131] First, the user takes a photo of the test question using the smartphone camera. The device analyzes the captured image using OCR and extracts the characters in the image as text information. This extracted text information is then sent to the server.

[0132] 2. Correct / incorrect judgment

[0133] The server inputs the received text information into the generative AI model, analyzes the content of the question, and judges whether it is correct or not. The analyzed result of the judgment is then sent back to the device.

[0134] Example: If a user answers "1467" to the history question "What year did the Onin War start?", the result of the accuracy test will be displayed as "correct answer."

[0135] 3. Automatic generation of background information

[0136] Based on the results of the correct / incorrect judgment, the server automatically generates relevant background information. Specifically, it extracts historical background and additional explanatory information related to the problem from natural language generation (NLG) models and pre-built databases. The generated background information is sent to the terminal and displayed to the user.

[0137] Example: The user gets the answer right and is shown detailed background information about the Onin War, as well as the story of its outcome.

[0138] 4. Accumulation and analysis of learning history

[0139] The device stores the user's learning history in a local database. The server periodically collects and analyzes the user's learning history data. For example, it can identify learning trends, such as the user's weakness in history questions. Based on the analysis, supplementary information to deepen understanding is generated and sent to the device.

[0140] Example: Additional educational materials and video content are recommended to help users who are weak in history to deepen their understanding.

[0141] 5. Comparison of performance at the national level

[0142] The server compiles and analyzes the academic performance information of learners nationwide. This makes it possible to compare each user's performance with the national average and deviation score. The user's performance data is sent to the server, and the results of the comparison with the national average are sent back to the terminal. This allows the user to understand where their learning performance ranks nationwide.

[0143] Example: Information such as the user's performance being in the top 20% of the national average is displayed on the device.

[0144] Prompt Sentence Examples

[0145] "Extract the text in the image below and determine whether the question is correct or incorrect. Then generate the answer and background information in the same image."

[0146] Through the above processing, the present invention is a system that supports the user's learning activities in many ways and realizes more effective learning.

[0147] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0148] Step 1: Photographing the test questions and recognizing the characters

[0149] Terminal

[0150] The user takes a photo of a test question using their smartphone camera. This image becomes the input. The device uses its OCR (Optical Character Recognition) function to analyze the character information from the captured image and extract text data. The extracted text data becomes the output.

[0151] Specific operation: The user launches the camera app on their smartphone and takes a picture of the test question. The app then performs OCR, generating the text data "What year did the Onin War start?" and "1467," and sends it to the server.

[0152] Step 2: Send text information

[0153] Terminal

[0154] The extracted text information is sent to the server. This text information is the input, and the transmission to the server is the output.

[0155] Specific operation: The device sends the text data "What year did the Onin War start?" and "1467" to the server.

[0156] Step 3: Correct / incorrect

[0157] server

[0158] The text information received by the server is input into the generative AI model. This text information becomes the input. The generative AI model analyzes the content of the question and determines whether it is correct or incorrect. The result of the correct or incorrect determination becomes the output.

[0159] Specific operation: The server inputs the received text data, "What year did the Onin War start?" and "1467," into the generative AI model, which determines that "1467" is the correct answer. The result is then sent back to the device.

[0160] Step 4: Displaying the results

[0161] Terminal

[0162] The terminal receives the result of the correctness judgment from the server and displays it to the user. The result of the correctness judgment is the input, and the display is the output.

[0163] Specific operation: The terminal displays the "correct" result received from the server on the user's screen.

[0164] Step 5: Automatically generate background information

[0165] server

[0166] Based on the accuracy assessment results, the server automatically generates background information. The accuracy assessment results are the input. The server generates related background information using a natural language generation (NLG) model or a pre-built database. This generated background information is the output.

[0167] Specific operation: The server obtains the "correct answer" result and automatically generates an "Outline of the Onin War and its historical background" using the NLG model and sends it to the terminal.

[0168] Step 6: Display background information

[0169] Terminal

[0170] The generated background information is received by the terminal and displayed to the user. The background information is the input and the display is the output.

[0171] Specific operation: The information received by the terminal, "Outline of the Onin War and its historical background," is displayed on the screen.

[0172] Step 7: Accumulating learning history

[0173] Terminal

[0174] The user's learning history is stored in a local database. The input is text information about the learning questions and correct / incorrect results. The output is to save this information in the local database.

[0175] Specific operation: The device stores the user's answers and study date and time in a local database.

[0176] Step 8: Analyze your learning history

[0177] server

[0178] The user's learning history data is collected periodically and analyzed. The accumulated learning history data is the input. The server analyzes this data to understand the user's learning patterns and tendencies. The results of this analysis are the output.

[0179] Specific operation: The server analyzes the user's learning history data and identifies a learning tendency such as "I feel uncomfortable with history questions."

[0180] Step 9: Generate and notify supplementary information

[0181] server

[0182] Supplementary information is generated based on the analysis results and notified to the user. The analysis results are the input. The server generates supplementary information and sends it to the user's terminal. This supplementary information is the output.

[0183] Specific operation: The server generates "educational materials and video content that will deepen the understanding of users who are not good at history" and sends them to the device. The device displays this to the user as a notification.

[0184] Step 10: Comparing performance at the national level

[0185] server

[0186] The performance information of learners nationwide is compiled and compared with the user's performance. National performance data is input. The server analyzes this data and compares the user's performance with the national average and deviation value. The results of this comparison are output.

[0187] Specific operation: The server analyzes the performance data collected from all over the country, calculates information such as whether the user's performance is in the top 20% of the national average, and sends it to the terminal. The terminal then displays this information to the user.

[0188] (Application example 1)

[0189] 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."

[0190] On conventional online shopping sites, users must research product details themselves when selecting a product, making it difficult to find related or recommended products. Furthermore, accurate product suggestions based on users' purchase history are not sufficiently provided. Therefore, there is a need for a system that allows users to efficiently select products and smoothly proceed with their purchasing process.

[0191] 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.

[0192] In this invention, the server includes means for allowing a user to take an image and extract text information from the image, means for transmitting the extracted text information, means for analyzing the transmitted text information and generating product information, means for displaying the generated product information to the user, means for automatically generating related products based on the generated product information, means for displaying the generated related product information to the user, means for generating individual recommended products based on the user's purchase history, means for displaying the generated recommended product information to the user, and means for comparing the generated recommended product information with purchase history data of other users. This enables a user to efficiently obtain detailed product information, related products, and recommended products simply by taking a photo of a product.

[0193] A "user" is a consumer who uses the system to obtain product information and assists in purchasing.

[0194] "Image" means visual data of a product or item photographed using a smartphone or other device.

[0195] "Text information" refers to text information contained in an image and is extracted using the OCR function.

[0196] "Extraction" is the process of extracting the necessary text information from an image.

[0197] "Sending" refers to transferring the extracted character information from the terminal to the server in the form of data.

[0198] "Analysis" refers to understanding product information and related information based on the text information sent and processing it using a generative AI model.

[0199] "Product information" refers to detailed data about a particular product, such as specifications, price, and ratings.

[0200] "Related products" are other products that may be of interest to the user that are suggested based on the analyzed product information.

[0201] "Display" refers to visually showing information sent from the server on the user's terminal screen.

[0202] "Purchase history" is data on products purchased in the past by a user, and is used to generate recommended products.

[0203] "Recommended products" are products that are likely to be purchased and are identified based on the user's purchase history and preferences.

[0204] "Purchase history data of other users" refers to data on the past purchase history of multiple users stored in the system.

[0205] "Comparison" is the process of evaluating and analyzing one piece of data (e.g., a user's purchase history) against other data (e.g., the purchase history of other users).

[0206] The "product proposal support system" for realizing the present invention can be implemented using the following hardware and software.

[0207] Hardware and software used

[0208] Smartphone: A device that takes a picture of a product and uses OCR to extract text information.

[0209] Server: A computer that performs analysis and data aggregation.

[0210] OCR (Optical Character Recognition) software: A tool that extracts textual information from images (e.g., Tesseract OCR).

[0211] Generative AI model: An AI model that analyzes product information and related products and makes suggestions (e.g., GPT-3).

[0212] Natural Language Processing (NLP) models: Models that understand and analyze extracted text information.

[0213] Database: A storage system (e.g., MySQL, PostgreSQL, etc.) that stores product information, user purchase history, and related products.

[0214] System Operation

[0215] 1. The user takes a photo of the product they plan to purchase using their smartphone.

[0216] 2. The device uses OCR software to extract text information from the image.

[0217] 3. The extracted text information is sent to the server.

[0218] 4. The server analyzes the received text information using a generative AI model to generate detailed product information and related product information.

[0219] 5. The server sends related product information and detailed information to the terminal.

[0220] 6. The terminal displays the received information to the user.

[0221] 7. The server retrieves the user's purchase history from the database and automatically generates recommended products using a generative AI model.

[0222] 8. The generated recommended product information is sent to the terminal and displayed to the user.

[0223] 9. The server compares your purchase history data with that of other users and performs a statistical evaluation of your purchasing patterns.

[0224] 10. The evaluation results are sent to the terminal and displayed to the user.

[0225] Specific examples

[0226] For example, if a user takes a picture of a new smartphone, a sample prompt might look like this:

[0227] "Take a photo of a smartphone (new model) and display detailed product information. Please suggest recommended products based on the user's purchase history."

[0228] A user takes a photo of their smartphone, and the OCR function extracts text information such as "new model smartphone." This text information is sent to a server, where a generative AI model generates detailed information about the "new model smartphone" and related products. This information is then sent to the device and displayed to the user. If the user has previously purchased electronic devices, related products and accessories are recommended based on their purchase history, and their purchasing trends are displayed in comparison with the purchasing behavior of other users. In this way, users can efficiently select products and smoothly proceed with their purchasing behavior.

[0229] This allows users to efficiently obtain detailed product information, related products, and recommended products simply by taking a photo of the product. Also, by comparing the purchase data of other users, users can understand their own purchasing trends and make better decisions.

[0230] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0231] Step 1:

[0232] The user takes a photo of the product they plan to purchase using their smartphone.

[0233] Input: Product image

[0234] Output: None

[0235] Specific action: A user takes a photo of a specific product using their smartphone camera.

[0236] Step 2:

[0237] The device uses OCR software to extract text information from the image.

[0238] Input: Product image

[0239] Output: Extracted text information

[0240] Specific operation: Image analysis is performed using on-device OCR (e.g., Tesseract OCR), and character information such as the product name and model number is extracted as text.

[0241] Step 3:

[0242] The extracted character information is sent to the server.

[0243] Input: Extracted text information

[0244] Output: Confirmation of sending text information to the server

[0245] What it does: The extracted text data is sent over a network to a server, often using the HTTP or HTTPS protocol.

[0246] Step 4:

[0247] The server uses a generative AI model based on the text information received to analyze and generate product information.

[0248] Input: Text information sent

[0249] Output: Generated product information

[0250] How it works: The server uses a generative AI model (e.g., GPT-3) to analyze the received text data and generate detailed product information (specifications, price, reviews, etc.). It may also collect information from external databases or APIs as needed.

[0251] Step 5:

[0252] The generated product information is transmitted to the terminal.

[0253] Input: Generated product information

[0254] Output: Confirmation of sending product information to the terminal

[0255] Specific operation: The product information generated by the server is sent to the terminal in a data format such as JSON.

[0256] Step 6:

[0257] The terminal displays the generated product information to the user.

[0258] Input: Generated product information

[0259] Output: Product details page shown to the user

[0260] What it does: Product information is displayed on the device screen, including the product name, price, specifications, reviews, etc.

[0261] Step 7:

[0262] The server retrieves the user's purchase history from the database and automatically generates recommended products using a generative AI model.

[0263] Input: User purchase history

[0264] Output: Generated product recommendations

[0265] Specific operation: The server retrieves purchase history from the database, analyzes the user's preferences and past purchase data using a generative AI model, and generates recommended products based on that.

[0266] Step 8:

[0267] The generated recommended product information is sent to the terminal and displayed to the user.

[0268] Input: Generated recommended product information

[0269] Output: Recommended products displayed to the user

[0270] Specific operation: The server sends recommended product information to the device, and the device displays the recommended products to the user, for example, recommended accessories and related products.

[0271] Step 9:

[0272] The server compares the purchase history data of other users and performs a statistical evaluation of the purchasing patterns.

[0273] Input: Purchase history data of other users

[0274] Output: Statistical evaluation results

[0275] What it does: The server compares your purchasing history data with that of other users and uses statistical analysis and machine learning models to assess how your purchasing patterns are different or similar to others.

[0276] Step 10:

[0277] The evaluation results are sent to the terminal and displayed to the user.

[0278] Input: Statistical evaluation results

[0279] Output: Screen showing the evaluation results

[0280] Specific operation: The server sends the statistical evaluation results to the terminal and displays them on the user's screen, allowing the user to understand how their purchasing behavior differs from that of other users.

[0281] 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.

[0282] The "learning support system" of this invention provides an integrated set of functions, including image recognition, natural language processing, performance analysis, and information provision, to fully support users' learning activities. Furthermore, this system combines an emotion engine that recognizes the user's emotions to provide optimal learning support according to the user's emotional state.

[0283] 1. Photographing and recognizing test questions

[0284] Terminal

[0285] First, the user takes a photo of the test question using the smartphone camera. The device analyzes the captured image using its OCR (Optical Character Recognition) function and extracts the characters in the image as text information. This extracted text information is then sent to the server.

[0286] server

[0287] The server inputs the received text information into the generative AI model, analyzes the content of the question, and judges whether it is correct or not. The analyzed result of the judgment is then sent back to the device.

[0288] Terminal

[0289] The device displays the result of the accuracy judgment received from the server to the user. For example, if the user answers "1467" to the history question "What year did the Onin War start?", the result of the accuracy judgment is displayed as "correct answer."

[0290] 2. Providing background information

[0291] server

[0292] Based on the results of the correct / incorrect decision, the server automatically generates relevant background information, specifically by extracting historical background and additional explanatory information related to the problem from natural language generation (NLG) models and pre-built databases.

[0293] Terminal

[0294] The generated background information is sent to the terminal and displayed to the user. For example, detailed background information about the Onin War and the story of its outcome are displayed.

[0295] 3. Viewing learning history and supplementary information

[0296] Terminal

[0297] The user's learning history is stored in a local database, which records the correct / incorrect results for each question, the date and time of study, and so on.

[0298] server

[0299] The server analyzes the user's learning history data and generates supplementary information based on the user's learning patterns and interests. For example, if a user is not good at history, additional learning materials and video content that will help them improve their understanding will be recommended.

[0300] Terminal

[0301] The generated supplemental information is sent to the device and displayed to the user, and is displayed as a notification, allowing the user to further study based on it.

[0302] 4. Comparison of performance at the national level

[0303] server

[0304] The server compiles and analyzes the performance information of learners nationwide to generate statistical data, which makes it possible to compare each user's performance with the national average and deviation score.

[0305] Terminal

[0306] The user's academic performance data is sent to the server, and a comparison with the national average is sent back to the terminal, allowing the user to understand where their academic performance ranks nationally.

[0307] 5. Incorporating an Emotional Engine

[0308] Terminal

[0309] The device uses a built-in emotion engine to recognize the user's emotions, analyzing facial expression data and voice tone obtained through the camera and microphone to identify the user's emotional state.

[0310] server

[0311] The server receives the emotion recognition results sent from the emotion engine and reflects them in the learning content and the provision of supplementary information. For example, if the user is feeling stressed, the server may provide content that helps them relax.

[0312] Terminal

[0313] The device displays adjusted learning content and supplementary information based on the emotion recognition results to the user, providing appropriate support in line with the user's emotions and improving the effectiveness of learning.

[0314] Specific examples

[0315] For example, suppose a user photographs a history test question, "What year did the Onin War start?" and answers "1467." In this case, the process would proceed as follows:

[0316] 1. The user takes a photo of the problem with their smartphone.

[0317] 2. The device uses its OCR function to extract "What year did the Onin War start?" and "1467" from the image and send it to the server.

[0318] 3. The server uses the generated AI model to determine whether the answer is correct, determines it is correct, and automatically generates background information about the progress and outcome of the Onin War.

[0319] 4. The device displays the correct / incorrect result and background information to the user.

[0320] 5. The server generates relevant supplementary information based on the user's learning history and sends it to the terminal.

[0321] 6. The terminal notifies and displays the supplementary information to the user.

[0322] 7. The server compares the user's grades with national data, calculates the standard deviation score, etc., and sends it to the terminal.

[0323] 8. The device displays the results of the comparison to the user.

[0324] 9. The device's emotion engine analyzes the user's emotions and sends the emotional state to the server.

[0325] 10. The server adjusts the learning content and supplementary information based on the emotion recognition results and sends them back to the device.

[0326] 11. The device displays the adjusted information to the user.

[0327] In this way, the present invention is a system that supports users' learning in multiple ways and promotes efficient and deep understanding. Furthermore, by taking into account the user's emotional state, it is possible to provide more personalized learning support.

[0328] The processing flow will be explained below.

[0329] Step 1:

[0330] The user starts the smartphone app and takes a picture of the test paper. When the user presses the capture button, the smartphone's camera function is activated and an image of the test paper is captured.

[0331] Step 2:

[0332] The device analyzes the captured image using its OCR (Optical Character Recognition) function. The OCR engine extracts character data from the image data and converts it into text format.

[0333] Step 3:

[0334] The device sends the extracted text information to the server, using a communication network to generate an API request and send the data to a specified endpoint on the server.

[0335] Step 4:

[0336] The server inputs the received text information into a generative AI model, which analyzes the content of the problem and uses natural language processing (NLP) algorithms to distinguish between the question and the answer.

[0337] Step 5:

[0338] The server determines whether the question is correct, refers to an internal database or knowledge base, evaluates whether the analyzed answer is correct, formats the evaluation result, and generates a data packet to send back to the terminal.

[0339] Step 6:

[0340] The device displays the correctness judgment result received to the user. The result is reflected in the user interface (UI), and a message such as "correct" or "incorrect" is displayed.

[0341] Step 7:

[0342] The server generates background information based on the results of the accuracy assessment, automatically generating relevant historical context and additional explanations using a natural language generation (NLG) model, or searching and extracting relevant information from a database.

[0343] Step 8:

[0344] The server sends the generated background information to the device, which then converts it into JSON format and sends it to the device via the communication network.

[0345] Step 9:

[0346] The device displays the background information received to the user, providing detailed historical background and additional explanations in an easy-to-understand manner to help the user gain a deeper understanding.

[0347] Step 10:

[0348] The device stores the user's learning history in a local database, accumulating information such as the questions studied, answers, correct / incorrect results, and learning dates and times.

[0349] Step 11:

[0350] The server analyzes the user's learning history data, generates supplementary information based on the user's learning patterns and preferences, and recommends appropriate learning materials and video content.

[0351] Step 12:

[0352] The server generates supplementary information and sends it to the terminal. Information tailored to each user's individual learning needs is sent, improving learning efficiency.

[0353] Step 13:

[0354] The device notifies and displays the supplementary information it receives to the user. The supplementary information is displayed in the notification bar or a dedicated screen within the app, making it easy for the user to access.

[0355] Step 14:

[0356] The server aggregates and analyzes the performance data collected from other users nationwide, and uses statistical algorithms to calculate the national average and standard deviation.

[0357] Step 15:

[0358] The server generates information comparing the user's performance with the national average and sends it to the device, including the difference between the individual user's performance and the national average, as well as the deviation score.

[0359] Step 16:

[0360] The terminal displays the received performance comparison results to the user, visualizing the results in the form of graphs and charts to help the user clearly understand their position.

[0361] Step 17:

[0362] The device captures the user's facial expressions and voice tone through a camera and microphone, and analyzes them with an emotion engine, which then recognizes the user's emotional state (e.g., joy, stress, concentration, etc.).

[0363] Step 18:

[0364] The server receives the emotion recognition results sent from the emotion engine and adjusts the learning content and supplementary information based on them. For example, if the user is feeling stressed, it provides relaxing content.

[0365] Step 19:

[0366] The device displays the adjusted learning content and supplementary information to the user, and reflects the adjustment results in the user interface, allowing the user to receive support that is in line with their emotions.

[0367] Through these steps, the present invention is a system that supports users' learning in multiple ways and promotes efficient and deep understanding. Furthermore, by taking into account the user's emotional state, it is possible to provide more personalized learning support.

[0368] Example 2

[0369] 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."

[0370] Conventional learning support systems are limited to determining whether questions submitted by users are correct or incorrect and providing background information, and lack the ability to flexibly respond to the user's learning situation and emotional state. Furthermore, because guidance and supplementary information are not provided efficiently or effectively to individual users, there are problems with not being able to improve learning efficiency or deepen understanding. Furthermore, when comparing scores at a national level, information based on the characteristics of individual users is often not provided, and sufficient support is often not provided. To solve these issues, a system with more advanced and flexible learning support functions is needed.

[0371] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0372] In this invention, the server includes a means for adjusting the learning content according to the user's emotions, a means for recognizing the user's emotions using image recognition and voice analysis, and a means for automatically generating background information based on a true / false judgment, thereby making it possible to provide individually optimized learning support based on the user's emotions and learning history.

[0373] A "user" is an individual who uses the system to carry out learning activities.

[0374] "Means for taking images" refers to the ability to take photos of test questions and study materials using a camera on a smartphone, tablet, etc.

[0375] "Means for extracting text information" refers to the function of extracting text data from images using OCR (optical character recognition) technology.

[0376] "Means for transmitting character information" refers to a function for transmitting extracted text data to a server via the Internet.

[0377] "Means for determining correctness" refers to the function of analyzing the text information sent and determining whether the answer to the question is correct.

[0378] "Means for displaying the judgment result to the user" refers to a function for displaying the result of the correctness judgment on the user's terminal.

[0379] "Means for automatically generating background information" refers to a function that automatically generates related supplementary explanations and detailed information based on a correct / incorrect judgment.

[0380] The "means for displaying background information to the user" refers to a function for displaying the generated background information on the user's terminal.

[0381] "Means for comparing performance with national performance" refers to a function that compares a user's performance data with that of other users nationwide and calculates rankings and standard deviations.

[0382] "Means for recognizing emotions" refers to a function that uses a camera or microphone to analyze and recognize the user's emotional state.

[0383] "Means for adjusting learning content" refers to a function that optimally adjusts the learning content and supplementary information provided based on the user's emotional state and learning history.

[0384] The "learning support system" of this invention provides multiple integrated functions to fully support users' learning activities. This system is realized by combining various functions, mainly image recognition, natural language processing, performance analysis, and information provision. Furthermore, by incorporating an emotion engine, it is possible to provide optimal learning support according to the user's emotional state.

[0385] 1. Photographing and recognizing test questions

[0386] User

[0387] Users take photos of test questions using their smartphone camera.

[0388] Terminal

[0389] The device analyzes the captured image using its OCR (Optical Character Recognition) function and extracts the character information in the image as text data. The specific software used is OCR technology (e.g., Google Cloud Vision or Tesseract). This extracted character information is sent from the device to a server.

[0390] server

[0391] The server inputs the received text information into a generative AI model, analyzes the content of the question, and determines whether it is correct or incorrect. For example, we use "OpenAI GPT-4" as the generative AI model. Examples of prompt sentences include:

[0392] Question: "What year did the Onin War start?"

[0393] Answer: "1467"

[0394] The generative AI model determines whether the prompt is correct or incorrect based on the prompt, and the server then sends the result back to the device.

[0395] Terminal

[0396] The terminal displays the result of the accuracy judgment received from the server to the user. For example, it displays the result in the format "What year did the Onin War start? 1467 -> Correct."

[0397] 2. Providing background information

[0398] server

[0399] Based on the results of the accuracy assessment, the server automatically generates relevant background information from a generative AI model or a pre-built database. Specific software examples include "natural language generation (NLG) models (e.g., OpenAI GPT-4 and Google BERT)."

[0400] Terminal

[0401] The generated background information is sent to the terminal and displayed to the user. For example, "Detailed background information about the Onin War" is displayed.

[0402] 3. Viewing learning history and supplementary information

[0403] Terminal

[0404] The device stores the user's learning history in a local database, including the correct / incorrect results for each question and the date and time of the learning.

[0405] server

[0406] The server analyzes the user's learning history data and generates supplemental information based on the user's learning patterns and preferences, using "machine learning algorithms and data analysis tools (e.g., Python, Pandas, Scikit-learn)."

[0407] Terminal

[0408] The generated supplementary information is sent to the terminal and notified and displayed to the user.

[0409] 4. Comparison of performance at the national level

[0410] server

[0411] The server will compile the academic performance information of students nationwide and generate statistical data. Specific technologies include "SQL databases and analytical tools (e.g., Tableau) used for data compilation and analysis."

[0412] Terminal

[0413] The user's academic performance data is sent to a server, and a comparison result with the national average is sent back to the terminal. The user can check this result and understand where their academic performance ranks nationwide.

[0414] 5. Incorporating an Emotional Engine

[0415] Terminal

[0416] The device uses a camera and microphone to recognize the user's emotions. Specific examples of software include emotion recognition software (e.g., Microsoft Azure Emotion API and IBM Watson Tone Analyzer).

[0417] server

[0418] The server receives the emotion recognition results sent from the emotion engine and adjusts the learning content and supplementary information based on the results. For example, if the user is feeling stressed, it provides relaxing content.

[0419] Terminal

[0420] The adjusted learning content and supplementary information are sent to the terminal and displayed to the user, allowing the user to receive the optimal learning environment and support.

[0421] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0422] Step 1:

[0423] User

[0424] The user takes a photo of the test question using the smartphone camera. The input is an image containing the test question, and the output is the captured image data. Specifically, the user opens the camera app, adjusts the frame so that the entire test paper fits within the frame, and presses the capture button.

[0425] Step 2:

[0426] Terminal

[0427] The device analyzes the captured image using its OCR (Optical Character Recognition) function and extracts the text information in the image as text data. The input is the captured image data, and the output is the extracted text data. Specifically, the device analyzes the image using OCR technology (for example, Google Cloud Vision or Tesseract) and extracts the question and answer in text format.

[0428] Step 3:

[0429] Terminal

[0430] The terminal sends the extracted text data to the server. The input is the extracted text data, and the output is the data sent to the server. Specifically, the terminal uses the HTTPS protocol to send the extracted text data to the specified endpoint of the server as a POST request.

[0431] Step 4:

[0432] server

[0433] The server inputs the received text data into the generative AI model, analyzes the content of the question, and determines whether it is correct or incorrect. The input is the received text data, and the output is the result of the correctness determination. Specifically, the server runs a Python script to generate a prompt for the generative AI model (e.g., GPT-4). An example of a prompt is as follows:

[0434] Question: "What year did the Onin War start?"

[0435] Answer: "1467"

[0436] The generative AI model determines whether the prompt is correct or not based on the prompt sentence and returns the result as "correct."

[0437] Step 5:

[0438] server

[0439] The server returns the result of the accuracy judgment to the terminal. The input is the accuracy judgment result, and the output is the response data to the terminal. Specifically, the server converts the accuracy judgment result into JSON format and returns it to the terminal as a response using the HTTPS protocol.

[0440] Step 6:

[0441] Terminal

[0442] The device displays the correctness judgment result received from the server to the user. The input is the correctness judgment result received from the server, and the output is the information displayed to the user. Specifically, the device analyzes the received JSON data and displays the result "correct" on the screen.

[0443] Step 7:

[0444] server

[0445] Based on the result of the accuracy judgment, the server automatically generates relevant background information from a generative AI model or a pre-built database. The input is the accuracy judgment result, and the output is the generated background information. Specifically, the server again sends a prompt to the generative AI model (e.g., GPT-4) to generate detailed historical background information.

[0446] Step 8:

[0447] server

[0448] The server sends the generated background information to the terminal. The input is the generated background information, and the output is the response data to the terminal. Specifically, the server returns the background information in JSON format to the terminal as a response.

[0449] Step 9:

[0450] Terminal

[0451] The terminal displays the background information received from the server to the user. The input is the background information received from the server, and the output is the information displayed to the user. Specifically, the terminal analyzes the received background information and displays it as "detailed background information about the Onin War."

[0452] Step 10:

[0453] Terminal

[0454] The device stores the user's learning history in a local database. The input is the learning result and the learning date and time, and the output is the data stored in the local database. Specifically, the device uses a local database such as SQLite and executes SQL queries to record the learning history.

[0455] Step 11:

[0456] server

[0457] The server analyzes the user's learning history data and generates related supplemental information. The input is the learning history data, and the output is the generated supplemental information. Specifically, the server analyzes the learning data using a machine learning algorithm and recommends additional learning materials and video content based on the user's interests and preferences.

[0458] Step 12:

[0459] server

[0460] The server sends the generated supplementary information to the terminal. The input is the generated supplementary information, and the output is the response data to the terminal. Specifically, the server returns the supplementary information in JSON format to the terminal as a response.

[0461] Step 13:

[0462] Terminal

[0463] The terminal displays the supplementary information received from the server to the user. The input is the supplementary information received from the server, and the output is the information displayed to the user. Specifically, the terminal displays the supplementary information as a notification, allowing the user to check the details.

[0464] Step 14:

[0465] server

[0466] The server aggregates the performance information of learners nationwide and generates statistical data. The input is the performance data of learners nationwide, and the output is the generated statistical data. Specifically, the server uses an SQL database to execute SQL queries that aggregate each user's performance and generate statistical data.

[0467] Step 15:

[0468] server

[0469] The server compares the user's performance data with the national average and sends the results to the device. The input is the user's performance data and the national average data, and the output is the comparison result. Specifically, the server returns the comparison result to the device as a response in JSON format.

[0470] Step 16:

[0471] Terminal

[0472] The terminal displays the comparison results with the national average to the user. The input is the comparison results received from the server, and the output is the information displayed to the user. Specifically, the terminal analyzes the received comparison results and displays the "comparison results with the national average" on the screen.

[0473] Step 17:

[0474] Terminal

[0475] The device uses its built-in emotion engine to analyze facial expression data and voice tone obtained from the camera and microphone to recognize the user's emotions. The input is facial expression data and voice tone, and the output is the emotion recognition result. Specifically, the device uses emotion recognition software (for example, Microsoft Azure Emotion API or IBM Watson Tone Analyzer) to perform the analysis.

[0476] Step 18:

[0477] Terminal

[0478] The device sends the emotion recognition results to the server. The input is the emotion recognition results, and the output is the data to be sent to the server. Specifically, the device sends the emotion recognition results in JSON format to the server as a POST request.

[0479] Step 19:

[0480] server

[0481] The server receives the emotion recognition results and adjusts the learning content and supplementary information based on them. The input is the emotion recognition results, and the output is the adjusted learning content and supplementary information. Specifically, the server analyzes the emotion data and adjusts the learning content to provide relaxing content if the user is feeling stressed.

[0482] Step 20:

[0483] server

[0484] The server sends the adjusted learning content and supplementary information to the terminal. The input is the adjusted learning content and supplementary information, and the output is the response data to the terminal. Specifically, the server returns the adjusted information in JSON format to the terminal as a response.

[0485] Step 21:

[0486] Terminal

[0487] The device displays the adjusted learning content and supplementary information to the user. The input is the adjusted information received from the server, and the output is the information displayed to the user. Specifically, the device analyzes the received information and displays it in a way that provides the user with the optimal learning environment.

[0488] (Application example 2)

[0489] 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."

[0490] Conventional learning support systems were able to judge whether test questions photographed by users were correct or incorrect and provide related background information, but they did not adjust the learning content based on the user's emotional state, and therefore lacked efficient learning support tailored to the emotions of each individual user. Furthermore, real-time learning support is required for use in physical stores, but a specific system to achieve this was lacking.

[0491] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for a user to take an image and extract text information from the image, means for analyzing the extracted text information based on a generative AI model and determining whether the question is correct, means for displaying the determination result to the user, means for automatically generating background information based on the correctness determination, means for displaying the generated background information to the user, means for comparing the user's performance with national performance, and means for recognizing the user's emotional state using an emotion recognition engine and adjusting the learning content. This enables efficient and personalized learning support according to the individual emotional state of each user, even in physical stores.

[0492] "User" refers to an individual who uses the system to carry out learning activities.

[0493] An "image" is a digital representation of visual information captured by a user.

[0494] "Text information" refers to text data extracted from an image.

[0495] "Generative AI models" refer to algorithms and models that use artificial intelligence technology to generate and analyze information.

[0496] "Correctness determination" refers to determining whether the answer given by the user is correct or incorrect.

[0497] "Background information" refers to additional explanation or information related to a question or problem.

[0498] "National level" refers to a scope that goes beyond a specific region or community and is effective nationwide.

[0499] "Grades" are evaluations and expressions of a user's learning progress and results expressed numerically or in letters.

[0500] An "emotion recognition engine" refers to hardware or software that analyzes and recognizes a user's emotional state.

[0501] "Learning content" refers to the materials, questions, information, and content generally provided for users to study.

[0502] "Adjustment" refers to changing content or settings based on specific conditions or circumstances.

[0503] This invention is a learning support system that integrates image recognition, natural language processing, performance analysis, and information provision functions to support users' learning activities. It is intended for use in brick-and-mortar stores, but it also functions effectively in other environments.

[0504] System Configuration

[0505] The system consists of multiple hardware and software components, including a user terminal, a server, and a learning assistant robot.

[0506] User terminal

[0507] User devices are primarily smartphones and tablets with the following features:

[0508] 1. Image capture: The user takes a picture of the study question using the smartphone camera.

[0509] 2. OCR (Optical Character Recognition): Uses OCR technology (e.g., Tesseract OCR) to extract text information from captured images.

[0510] server

[0511] The server has the following features:

[0512] 1. Character information analysis and accuracy determination: The character information extracted by OCR is analyzed based on a generative AI model (e.g., the GPT-3.5-turbo model) to determine whether the question is correct.

[0513] 2. Generating background information: Based on the results of the correct / incorrect judgment, background information related to the problem is automatically generated using a natural language generation (NLG) model.

[0514] 3. Performance analysis and comparison: Aggregate the performance of all users and compare it at a national level.

[0515] 4. Emotion Recognition: Use an emotion recognition engine to analyze the user's emotional state and adjust the learning content.

[0516] Learning Assistant Robot

[0517] The physical store will be equipped with a learning assistant robot, which will have the following functions:

[0518] 1. User support: Providing real-time learning support in-store and responding to user questions and problems.

[0519] 2. Facial Recognition and Emotion Analysis: Uses a camera and microphone to analyze a user's facial expressions and tone of voice to recognize their emotional state.

[0520] 3. Information provision: The generated background information and performance information are displayed to the user.

[0521] Specific examples

[0522] For example, consider a scenario where a user is in a store and asks a history question to a dedicated learning assistant robot, "What year did the Onin War start?", while simultaneously taking a photo of the question with their smartphone. In this scenario, the process would proceed as follows:

[0523] 1. The user takes a photo of the problem with their smartphone.

[0524] 2. The device uses its OCR function to extract text information from the image, such as "What year did the Onin War start?", and sends it to the server.

[0525] 3. The server uses a generative AI model to determine whether the answer is correct or not, and generates background information on the progress and outcome of the Onin War.

[0526] 4. The device displays the correct / incorrect result and background information to the user.

[0527] 5. The server generates relevant supplementary information based on the user's learning history and sends it to the terminal.

[0528] 6. The terminal notifies and displays the supplementary information to the user.

[0529] 7. The server compares the user's grades with national data, calculates the standard deviation score, etc., and sends it to the terminal.

[0530] 8. The device displays the results of the comparison to the user.

[0531] 9. The device's emotion recognition engine analyzes the user's emotions and sends the emotional state to the server.

[0532] 10. The server adjusts the learning content and supplementary information based on the emotion recognition results and sends them back to the device.

[0533] 11. The device displays the adjusted information to the user.

[0534] Examples of prompts:

[0535] "Provide detailed background information for the following question: In what year did the Onin War begin?"

[0536] As a result, this system enables efficient and personalized learning support based on each user's emotional state, even in physical stores.

[0537] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0538] Step 1:

[0539] The user takes a photo of the problem with their smartphone.

[0540] Input: Image of the training question

[0541] Output: Captured image data

[0542] Specific operation: The user takes a photo of the study question using the smartphone camera. The image data is saved on the device.

[0543] Step 2:

[0544] The device uses the OCR function to extract text information from the image and send it to the server.

[0545] Input: Captured image data

[0546] Output: Extracted text

[0547] Specific operation: Extract text information from the captured image using pytesseract, convert the text information into text format, and then send this text data to the server.

[0548] Step 3:

[0549] The server uses a generative AI model to determine accuracy and generates background information on the progress and results of the Onin War.

[0550] Input: Extracted text

[0551] Output: Correct / incorrect result and background information

[0552] Specific operation: Based on the received text information, the server uses a generative AI model (e.g., the GPT-3.5-turbo model) to analyze whether the question is correct. After determining whether the question is correct, the server uses the model or a pre-built database to generate background information related to the question.

[0553] Step 4:

[0554] The terminal displays the result of the correct / incorrect judgment and background information to the user.

[0555] Input: Correct / incorrect result and background information

[0556] Output: The result of the test and background information displayed on the user's screen

[0557] Specific operation: The result of the correct answer and background information sent from the server are received and displayed on the user's device. For example, "The answer to the question is correct" is displayed along with "the progress and outcome of the Onin War."

[0558] Step 5:

[0559] The server generates related supplementary information based on the user's learning history and sends it to the terminal.

[0560] Input: User learning history data

[0561] Output: Supplementary information

[0562] Specific operation: The server analyzes the user's learning history data and generates supplementary information to deepen understanding of the relevant topic. The generated supplementary information is sent to the terminal.

[0563] Step 6:

[0564] The terminal notifies and displays the supplementary information to the user.

[0565] Input: Additional information

[0566] Output: Supplementary information notified and displayed to the user

[0567] Specific operation: The supplementary information received from the server is notified to the user and displayed on the device screen. Specifically, it provides links such as "Video content related to the Onin War."

[0568] Step 7:

[0569] The server compares the user's grades with national data, calculates the standard deviation score, etc., and sends it to the terminal.

[0570] Input: User performance data, nationwide learner data

[0571] Output: Standard deviation and comparison results

[0572] Specific operation: The server compares the user's performance data with the data of learners nationwide, calculates the deviation score and ranking, and sends the calculation results to the terminal.

[0573] Step 8:

[0574] The terminal displays the results of the comparison of the results to the user.

[0575] Input: Standard deviation or comparison result

[0576] Output: The result of the comparison is displayed on the user's screen.

[0577] Specific operation: The results of the grade comparison sent from the server are notified to the user and displayed on the device screen. Specifically, a message such as "Your grades are above the national average" is displayed.

[0578] Step 9:

[0579] The device's emotion recognition engine analyzes the user's emotions and transmits the emotional state to the server.

[0580] Input: facial expression data, voice data

[0581] Output: Emotion recognition result

[0582] Specific operation: The device uses a built-in camera and microphone to collect and analyze the user's facial expressions and voice tone. The analysis results are sent to the server.

[0583] Step 10:

[0584] The server adjusts the learning content and supplementary information based on the emotion recognition results and sends them back to the device.

[0585] Input: Emotion recognition results

[0586] Output: Adjusted learning results and supplementary information

[0587] Specific operation: The server adjusts the learning content and supplementary information based on the received emotion recognition results. For example, if the user is feeling stressed, it will provide relaxing content. The adjusted information is then sent back to the device.

[0588] Step 11:

[0589] The terminal displays the adjusted information to the user.

[0590] Input: Adjusted learning content and supplementary information

[0591] Output: Adjusted information displayed on the user screen

[0592] Specific operation: The adjusted learning content and supplementary information sent from the server are received and displayed on the user's device. Specifically, a "relaxing video about the Onin War" is displayed.

[0593] 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.

[0594] 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.

[0595] 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.

[0596] [Second embodiment]

[0597] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0598] 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.

[0599] 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).

[0600] 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.

[0601] 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.

[0602] 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).

[0603] 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. 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.

[0604] 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.

[0605] 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.

[0606] 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.

[0607] 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.

[0608] 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."

[0609] The "learning support system" of this invention provides an integrated set of functions, including image recognition, natural language processing, performance analysis, and information provision, to fully support users' learning activities. This system is implemented as an application that users can easily use on devices such as smartphones and tablets.

[0610] 1. Photographing and recognizing test questions

[0611] Terminal

[0612] First, the user takes a photo of the test question using the smartphone camera. The device analyzes the captured image using its OCR (Optical Character Recognition) function and extracts the characters in the image as text information. This extracted text information is then sent to the server.

[0613] server

[0614] The server inputs the received text information into the generative AI model, analyzes the content of the question, and judges whether it is correct or not. The analyzed result of the judgment is then sent back to the device.

[0615] Terminal

[0616] The device displays the result of the accuracy judgment received from the server to the user. For example, if the user answers "1467" to the history question "What year did the Onin War start?", the result of the accuracy judgment is displayed as "correct answer."

[0617] 2. Providing background information

[0618] server

[0619] Based on the results of the correct / incorrect decision, the server automatically generates relevant background information, specifically by extracting historical background and additional explanatory information related to the problem from natural language generation (NLG) models and pre-built databases.

[0620] Terminal

[0621] The generated background information is sent to the terminal and displayed to the user. For example, detailed background information about the Onin War and the story of its outcome are displayed.

[0622] 3. Viewing learning history and supplementary information

[0623] Terminal

[0624] The user's learning history is stored in a local database, which records the correct / incorrect results for each question, the date and time of study, and so on.

[0625] server

[0626] The server analyzes the user's learning history data and generates supplementary information based on the user's learning patterns and interests. For example, if a user is not good at history, additional learning materials and video content that will help them improve their understanding will be recommended.

[0627] Terminal

[0628] The generated supplemental information is sent to the device and displayed to the user, and is displayed as a notification, allowing the user to further study based on it.

[0629] 4. Comparison of performance at the national level

[0630] server

[0631] The server compiles and analyzes the performance information of learners nationwide to generate statistical data, which makes it possible to compare each user's performance with the national average and deviation score.

[0632] Terminal

[0633] The user's academic performance data is sent to the server, and a comparison with the national average is sent back to the terminal, allowing the user to understand where their academic performance ranks nationally.

[0634] Specific examples

[0635] For example, suppose a user photographs a history test question, "What year did the Onin War start?" and answers "1467." In this case, the process would proceed as follows:

[0636] 1. The user takes a photo of the problem with their smartphone.

[0637] 2. The device uses its OCR function to extract "What year did the Onin War start?" and "1467" from the image and send it to the server.

[0638] 3. The server uses the generated AI model to determine whether the answer is correct, determines it is correct, and automatically generates background information about the progress and outcome of the Onin War.

[0639] 4. The device displays the correct / incorrect result and background information to the user.

[0640] 5. The server generates relevant supplementary information based on the user's learning history and sends it to the terminal.

[0641] 6. The terminal notifies and displays the supplementary information to the user.

[0642] 7. The server compares the user's grades with national data, calculates the standard deviation score, etc., and sends it to the terminal.

[0643] 8. The device displays the results of the comparison to the user.

[0644] In this way, the present invention is a system that supports the user's learning activities in many ways and enables more effective learning.

[0645] The processing flow will be explained below.

[0646] Step 1:

[0647] The user starts the smartphone app and takes a picture of the test paper. When the user presses the capture button, the smartphone's camera function is activated and an image of the test paper is captured.

[0648] Step 2:

[0649] The device analyzes the captured image using its OCR (Optical Character Recognition) function. The OCR engine extracts character data from the image data and converts it into text format.

[0650] Step 3:

[0651] The device sends the extracted text information to the server, using a communication network to generate an API request and send the data to a specified endpoint on the server.

[0652] Step 4:

[0653] The server inputs the received text information into a generative AI model, which analyzes the content of the problem and uses natural language processing (NLP) algorithms to distinguish between the question and the answer.

[0654] Step 5:

[0655] The server determines whether the question is correct, refers to an internal database or knowledge base, evaluates whether the analyzed answer is correct, formats the evaluation result, and generates a data packet to send back to the terminal.

[0656] Step 6:

[0657] The device displays the correctness judgment result received to the user. The result is reflected in the user interface (UI), and a message such as "correct" or "incorrect" is displayed.

[0658] Step 7:

[0659] The server generates background information based on the results of the accuracy assessment, automatically generating relevant historical context and additional explanations using a natural language generation (NLG) model, or searching and extracting relevant information from a database.

[0660] Step 8:

[0661] The server sends the generated background information to the device, which then converts it into JSON format and sends it to the device via the communication network.

[0662] Step 9:

[0663] The device displays the background information received to the user, providing detailed historical background and additional explanations in an easy-to-understand manner to help the user gain a deeper understanding.

[0664] Step 10:

[0665] The device stores the user's learning history in a local database, accumulating information such as the questions studied, answers, correct / incorrect results, and learning dates and times.

[0666] Step 11:

[0667] The server analyzes the user's learning history data, generates supplementary information based on the user's learning patterns and preferences, and recommends appropriate learning materials and video content.

[0668] Step 12:

[0669] The server generates supplementary information and sends it to the terminal. Information tailored to each user's individual learning needs is sent, improving learning efficiency.

[0670] Step 13:

[0671] The device notifies and displays the supplementary information it receives to the user. The supplementary information is displayed in the notification bar or a dedicated screen within the app, making it easy for the user to access.

[0672] Step 14:

[0673] The server aggregates and analyzes the performance data collected from other users nationwide, and uses statistical algorithms to calculate the national average and standard deviation.

[0674] Step 15:

[0675] The server generates information comparing the user's performance with the national average and sends it to the device, including the difference between the individual user's performance and the national average, as well as the deviation score.

[0676] Step 16:

[0677] The terminal displays the received performance comparison results to the user, visualizing the results in the form of graphs and charts to help the user clearly understand their position.

[0678] Through the above steps, the present invention is a system that supports users' learning in many ways and promotes efficient and deep understanding.

[0679] Example 1

[0680] 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."

[0681] Conventional learning support systems have limitations in their ability to accurately extract text information from images taken by users and determine whether questions are correct or incorrect. Furthermore, they lack the functionality to analyze learning history and provide users with appropriate supplementary information, or the ability to compare scores nationwide in real time. This makes it difficult for users to fully grasp their learning progress and level of understanding, making it difficult to study efficiently.

[0682] 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.

[0683] In this invention, the server includes means for a user to take an image and extract text information from the image, means for transmitting the extracted text information, means for analyzing the transmitted text information and judging whether the question is correct or incorrect, means for displaying the judgment result to the user, means for automatically generating background information based on the judgment of correctness, means for displaying the generated background information to the user, means for comparing the user's performance with national performance, means for accumulating and analyzing the user's learning history, means for generating supplementary information based on the analysis result and notifying the user, and means for aggregating and analyzing national performance data and comparing the user's performance with the national average and deviation score. This allows the user to grasp their own learning progress and level of understanding in real time and to study efficiently and effectively.

[0684] A "user" is a person who accesses the learning support system using a device such as a smartphone or tablet and engages in learning activities.

[0685] "Taking an image" refers to the act of a user using the camera on a smartphone or tablet to capture an image of a study question or document.

[0686] "Extracting text information" refers to the process of using OCR (Optical Character Recognition) technology to identify and extract text from a captured image as text data.

[0687] "Send" refers to the act of transferring data from a terminal to a server.

[0688] "Analysis" refers to the process in which the server analyzes the data it receives using technologies such as generative AI models to determine the content of the question and whether it is correct or incorrect.

[0689] "Determining correctness" means using a generative AI model to determine whether an answer to a question is correct or incorrect.

[0690] "Display" refers to the act of visually presenting analysis results, background information, performance information, etc. to the user on the terminal display.

[0691] "Background information" is information containing additional knowledge or explanation related to the content of the problem.

[0692] "Auto-generation" refers to the process of automatically generating the required information using pre-built databases and natural language generation (NLG) models.

[0693] "Study history" refers to a record of a user's past learning activities, answer results, study dates and times, etc.

[0694] "Analysis" refers to clarifying a user's learning patterns and tendencies based on accumulated learning history data.

[0695] "Supplemental information" refers to additional educational materials and video content provided to help users learn more efficiently.

[0696] "Notification" refers to the act of informing the user of generated information, supplementary materials, etc.

[0697] "National level results" refers to statistical data such as averages and standard deviations obtained by compiling and analyzing the results data of other users nationwide.

[0698] The "learning support system" of this invention provides an integrated set of functions, including image recognition, natural language processing, performance analysis, and information provision, to fully support users' learning activities. This system is implemented as an application that users can easily use on devices such as smartphones and tablets.

[0699] Details of the hardware and software you will use

[0700] Terminal

[0701] This system mainly uses mobile devices such as smartphones and tablets. The devices are equipped with a camera, OCR (Optical Character Recognition) functionality, and a display. The OCR functionality incorporates common OCR software and is used to identify characters from images.

[0702] System processing flow

[0703] 1. Photographing test questions and character recognition

[0704] First, the user takes a photo of the test question using the smartphone camera. The device analyzes the captured image using OCR and extracts the characters in the image as text information. This extracted text information is then sent to the server.

[0705] 2. Correct / incorrect judgment

[0706] The server inputs the received text information into the generative AI model, analyzes the content of the question, and judges whether it is correct or not. The analyzed result of the judgment is then sent back to the device.

[0707] Example: If a user answers "1467" to the history question "What year did the Onin War start?", the result of the accuracy test will be displayed as "correct answer."

[0708] 3. Automatic generation of background information

[0709] Based on the results of the correct / incorrect judgment, the server automatically generates relevant background information. Specifically, it extracts historical background and additional explanatory information related to the problem from natural language generation (NLG) models and pre-built databases. The generated background information is sent to the terminal and displayed to the user.

[0710] Example: The user gets the answer right and is shown detailed background information about the Onin War, as well as the story of its outcome.

[0711] 4. Accumulation and analysis of learning history

[0712] The device stores the user's learning history in a local database. The server periodically collects and analyzes the user's learning history data. For example, it can identify learning trends, such as the user's weakness in history questions. Based on the analysis, supplementary information to deepen understanding is generated and sent to the device.

[0713] Example: Additional educational materials and video content are recommended to help users who are weak in history to deepen their understanding.

[0714] 5. Comparison of performance at the national level

[0715] The server compiles and analyzes the academic performance information of learners nationwide. This makes it possible to compare each user's performance with the national average and deviation score. The user's performance data is sent to the server, and the results of the comparison with the national average are sent back to the terminal. This allows the user to understand where their learning performance ranks nationwide.

[0716] Example: Information such as the user's performance being in the top 20% of the national average is displayed on the device.

[0717] Prompt Sentence Examples

[0718] "Extract the text in the image below and determine whether the question is correct or incorrect. Then generate the answer and background information in the same image."

[0719] Through the above processing, the present invention is a system that supports the user's learning activities in many ways and realizes more effective learning.

[0720] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0721] Step 1: Photographing the test questions and recognizing the characters

[0722] Terminal

[0723] The user takes a photo of a test question using their smartphone camera. This image becomes the input. The device uses its OCR (Optical Character Recognition) function to analyze the character information from the captured image and extract text data. The extracted text data becomes the output.

[0724] Specific operation: The user launches the camera app on their smartphone and takes a picture of the test question. The app then performs OCR, generating the text data "What year did the Onin War start?" and "1467," and sends it to the server.

[0725] Step 2: Send text information

[0726] Terminal

[0727] The extracted text information is sent to the server. This text information is the input, and the transmission to the server is the output.

[0728] Specific operation: The device sends the text data "What year did the Onin War start?" and "1467" to the server.

[0729] Step 3: Correct / incorrect

[0730] server

[0731] The text information received by the server is input into the generative AI model. This text information becomes the input. The generative AI model analyzes the content of the question and determines whether it is correct or incorrect. The result of the correct or incorrect determination becomes the output.

[0732] Specific operation: The server inputs the received text data, "What year did the Onin War start?" and "1467," into the generative AI model, which determines that "1467" is the correct answer. The result is then sent back to the device.

[0733] Step 4: Displaying the results

[0734] Terminal

[0735] The terminal receives the result of the correctness judgment from the server and displays it to the user. The result of the correctness judgment is the input, and the display is the output.

[0736] Specific operation: The terminal displays the "correct" result received from the server on the user's screen.

[0737] Step 5: Automatically generate background information

[0738] server

[0739] Based on the accuracy assessment results, the server automatically generates background information. The accuracy assessment results are the input. The server generates related background information using a natural language generation (NLG) model or a pre-built database. This generated background information is the output.

[0740] Specific operation: The server obtains the "correct answer" result and automatically generates an "Outline of the Onin War and its historical background" using the NLG model and sends it to the terminal.

[0741] Step 6: Display background information

[0742] Terminal

[0743] The generated background information is received by the terminal and displayed to the user. The background information is the input and the display is the output.

[0744] Specific operation: The information received by the terminal, "Outline of the Onin War and its historical background," is displayed on the screen.

[0745] Step 7: Accumulating learning history

[0746] Terminal

[0747] The user's learning history is stored in a local database. The input is text information about the learning questions and correct / incorrect results. The output is to save this information in the local database.

[0748] Specific operation: The device stores the user's answers and study date and time in a local database.

[0749] Step 8: Analyze your learning history

[0750] server

[0751] The user's learning history data is collected periodically and analyzed. The accumulated learning history data is the input. The server analyzes this data to understand the user's learning patterns and tendencies. The results of this analysis are the output.

[0752] Specific operation: The server analyzes the user's learning history data and identifies a learning tendency such as "I feel uncomfortable with history questions."

[0753] Step 9: Generate and notify supplementary information

[0754] server

[0755] Supplementary information is generated based on the analysis results and notified to the user. The analysis results are the input. The server generates supplementary information and sends it to the user's terminal. This supplementary information is the output.

[0756] Specific operation: The server generates "educational materials and video content that will deepen the understanding of users who are not good at history" and sends them to the device. The device displays this to the user as a notification.

[0757] Step 10: Comparing performance at the national level

[0758] server

[0759] The performance information of learners nationwide is compiled and compared with the user's performance. National performance data is input. The server analyzes this data and compares the user's performance with the national average and deviation value. The results of this comparison are output.

[0760] Specific operation: The server analyzes the performance data collected from all over the country, calculates information such as whether the user's performance is in the top 20% of the national average, and sends it to the terminal. The terminal then displays this information to the user.

[0761] (Application example 1)

[0762] 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."

[0763] On conventional online shopping sites, users must research product details themselves when selecting a product, making it difficult to find related or recommended products. Furthermore, accurate product suggestions based on users' purchase history are not sufficiently provided. Therefore, there is a need for a system that allows users to efficiently select products and smoothly proceed with their purchasing process.

[0764] 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.

[0765] In this invention, the server includes means for allowing a user to take an image and extract text information from the image, means for transmitting the extracted text information, means for analyzing the transmitted text information and generating product information, means for displaying the generated product information to the user, means for automatically generating related products based on the generated product information, means for displaying the generated related product information to the user, means for generating individual recommended products based on the user's purchase history, means for displaying the generated recommended product information to the user, and means for comparing the generated recommended product information with purchase history data of other users. This enables a user to efficiently obtain detailed product information, related products, and recommended products simply by taking a photo of a product.

[0766] A "user" is a consumer who uses the system to obtain product information and assists in purchasing.

[0767] "Image" means visual data of a product or item photographed using a smartphone or other device.

[0768] "Text information" refers to text information contained in an image and is extracted using the OCR function.

[0769] "Extraction" is the process of extracting the necessary text information from an image.

[0770] "Sending" refers to transferring the extracted character information from the terminal to the server in the form of data.

[0771] "Analysis" refers to understanding product information and related information based on the text information sent and processing it using a generative AI model.

[0772] "Product information" refers to detailed data about a particular product, such as specifications, price, and ratings.

[0773] "Related products" are other products that may be of interest to the user that are suggested based on the analyzed product information.

[0774] "Display" refers to visually showing information sent from the server on the user's terminal screen.

[0775] "Purchase history" is data on products purchased in the past by a user, and is used to generate recommended products.

[0776] "Recommended products" are products that are likely to be purchased and are identified based on the user's purchase history and preferences.

[0777] "Purchase history data of other users" refers to data on the past purchase history of multiple users stored in the system.

[0778] "Comparison" is the process of evaluating and analyzing one piece of data (e.g., a user's purchase history) against other data (e.g., the purchase history of other users).

[0779] The "product proposal support system" for realizing the present invention can be implemented using the following hardware and software.

[0780] Hardware and software used

[0781] Smartphone: A device that takes a picture of a product and uses OCR to extract text information.

[0782] Server: A computer that performs analysis and data aggregation.

[0783] OCR (Optical Character Recognition) software: A tool that extracts textual information from images (e.g., Tesseract OCR).

[0784] Generative AI model: An AI model that analyzes product information and related products and makes suggestions (e.g., GPT-3).

[0785] Natural Language Processing (NLP) models: Models that understand and analyze extracted text information.

[0786] Database: A storage system (e.g., MySQL, PostgreSQL, etc.) that stores product information, user purchase history, and related products.

[0787] System Operation

[0788] 1. The user takes a photo of the product they plan to purchase using their smartphone.

[0789] 2. The device uses OCR software to extract text information from the image.

[0790] 3. The extracted text information is sent to the server.

[0791] 4. The server analyzes the received text information using a generative AI model to generate detailed product information and related product information.

[0792] 5. The server sends related product information and detailed information to the terminal.

[0793] 6. The terminal displays the received information to the user.

[0794] 7. The server retrieves the user's purchase history from the database and automatically generates recommended products using a generative AI model.

[0795] 8. The generated recommended product information is sent to the terminal and displayed to the user.

[0796] 9. The server compares your purchase history data with that of other users and performs a statistical evaluation of your purchasing patterns.

[0797] 10. The evaluation results are sent to the terminal and displayed to the user.

[0798] Specific examples

[0799] For example, if a user takes a picture of a new smartphone, a sample prompt might look like this:

[0800] "Take a photo of a smartphone (new model) and display detailed product information. Please suggest recommended products based on the user's purchase history."

[0801] A user takes a photo of their smartphone, and the OCR function extracts text information such as "new model smartphone." This text information is sent to a server, where a generative AI model generates detailed information about the "new model smartphone" and related products. This information is then sent to the device and displayed to the user. If the user has previously purchased electronic devices, related products and accessories are recommended based on their purchase history, and their purchasing trends are displayed in comparison with the purchasing behavior of other users. In this way, users can efficiently select products and smoothly proceed with their purchasing behavior.

[0802] This allows users to efficiently obtain detailed product information, related products, and recommended products simply by taking a photo of the product. Also, by comparing the purchase data of other users, users can understand their own purchasing trends and make better decisions.

[0803] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0804] Step 1:

[0805] The user takes a photo of the product they plan to purchase using their smartphone.

[0806] Input: Product image

[0807] Output: None

[0808] Specific action: A user takes a photo of a specific product using their smartphone camera.

[0809] Step 2:

[0810] The device uses OCR software to extract text information from the image.

[0811] Input: Product image

[0812] Output: Extracted text information

[0813] Specific operation: Image analysis is performed using on-device OCR (e.g., Tesseract OCR), and character information such as the product name and model number is extracted as text.

[0814] Step 3:

[0815] The extracted character information is sent to the server.

[0816] Input: Extracted text information

[0817] Output: Confirmation of sending text information to the server

[0818] What it does: The extracted text data is sent over a network to a server, often using the HTTP or HTTPS protocol.

[0819] Step 4:

[0820] The server uses a generative AI model based on the text information received to analyze and generate product information.

[0821] Input: Text information sent

[0822] Output: Generated product information

[0823] How it works: The server uses a generative AI model (e.g., GPT-3) to analyze the received text data and generate detailed product information (specifications, price, reviews, etc.). It may also collect information from external databases or APIs as needed.

[0824] Step 5:

[0825] The generated product information is transmitted to the terminal.

[0826] Input: Generated product information

[0827] Output: Confirmation of sending product information to the terminal

[0828] Specific operation: The product information generated by the server is sent to the terminal in a data format such as JSON.

[0829] Step 6:

[0830] The terminal displays the generated product information to the user.

[0831] Input: Generated product information

[0832] Output: Product details page shown to the user

[0833] What it does: Product information is displayed on the device screen, including the product name, price, specifications, reviews, etc.

[0834] Step 7:

[0835] The server retrieves the user's purchase history from the database and automatically generates recommended products using a generative AI model.

[0836] Input: User purchase history

[0837] Output: Generated product recommendations

[0838] Specific operation: The server retrieves purchase history from the database, analyzes the user's preferences and past purchase data using a generative AI model, and generates recommended products based on that.

[0839] Step 8:

[0840] The generated recommended product information is sent to the terminal and displayed to the user.

[0841] Input: Generated recommended product information

[0842] Output: Recommended products displayed to the user

[0843] Specific operation: The server sends recommended product information to the device, and the device displays the recommended products to the user, for example, recommended accessories and related products.

[0844] Step 9:

[0845] The server compares the purchase history data of other users and performs a statistical evaluation of the purchasing patterns.

[0846] Input: Purchase history data of other users

[0847] Output: Statistical evaluation results

[0848] What it does: The server compares your purchasing history data with that of other users and uses statistical analysis and machine learning models to assess how your purchasing patterns are different or similar to others.

[0849] Step 10:

[0850] The evaluation results are sent to the terminal and displayed to the user.

[0851] Input: Statistical evaluation results

[0852] Output: Screen showing the evaluation results

[0853] Specific operation: The server sends the statistical evaluation results to the terminal and displays them on the user's screen, allowing the user to understand how their purchasing behavior differs from that of other users.

[0854] 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.

[0855] The "learning support system" of this invention provides an integrated set of functions, including image recognition, natural language processing, performance analysis, and information provision, to fully support users' learning activities. Furthermore, this system combines an emotion engine that recognizes the user's emotions to provide optimal learning support according to the user's emotional state.

[0856] 1. Photographing and recognizing test questions

[0857] Terminal

[0858] First, the user takes a photo of the test question using the smartphone camera. The device analyzes the captured image using its OCR (Optical Character Recognition) function and extracts the characters in the image as text information. This extracted text information is then sent to the server.

[0859] server

[0860] The server inputs the received text information into the generative AI model, analyzes the content of the question, and judges whether it is correct or not. The analyzed result of the judgment is then sent back to the device.

[0861] Terminal

[0862] The device displays the result of the accuracy judgment received from the server to the user. For example, if the user answers "1467" to the history question "What year did the Onin War start?", the result of the accuracy judgment is displayed as "correct answer."

[0863] 2. Providing background information

[0864] server

[0865] Based on the results of the correct / incorrect decision, the server automatically generates relevant background information, specifically by extracting historical background and additional explanatory information related to the problem from natural language generation (NLG) models and pre-built databases.

[0866] Terminal

[0867] The generated background information is sent to the terminal and displayed to the user. For example, detailed background information about the Onin War and the story of its outcome are displayed.

[0868] 3. Viewing learning history and supplementary information

[0869] Terminal

[0870] The user's learning history is stored in a local database, which records the correct / incorrect results for each question, the date and time of study, and so on.

[0871] server

[0872] The server analyzes the user's learning history data and generates supplementary information based on the user's learning patterns and interests. For example, if a user is not good at history, additional learning materials and video content that will help them improve their understanding will be recommended.

[0873] Terminal

[0874] The generated supplemental information is sent to the device and displayed to the user, and is displayed as a notification, allowing the user to further study based on it.

[0875] 4. Comparison of performance at the national level

[0876] server

[0877] The server compiles and analyzes the performance information of learners nationwide to generate statistical data, which makes it possible to compare each user's performance with the national average and deviation score.

[0878] Terminal

[0879] The user's academic performance data is sent to the server, and a comparison with the national average is sent back to the terminal, allowing the user to understand where their academic performance ranks nationally.

[0880] 5. Incorporating an Emotional Engine

[0881] Terminal

[0882] The device uses a built-in emotion engine to recognize the user's emotions, analyzing facial expression data and voice tone obtained through the camera and microphone to identify the user's emotional state.

[0883] server

[0884] The server receives the emotion recognition results sent from the emotion engine and reflects them in the learning content and the provision of supplementary information. For example, if the user is feeling stressed, the server may provide content that helps them relax.

[0885] Terminal

[0886] The device displays adjusted learning content and supplementary information based on the emotion recognition results to the user, providing appropriate support in line with the user's emotions and improving the effectiveness of learning.

[0887] Specific examples

[0888] For example, suppose a user photographs a history test question, "What year did the Onin War start?" and answers "1467." In this case, the process would proceed as follows:

[0889] 1. The user takes a photo of the problem with their smartphone.

[0890] 2. The device uses its OCR function to extract "What year did the Onin War start?" and "1467" from the image and send it to the server.

[0891] 3. The server uses the generated AI model to determine whether the answer is correct, determines it is correct, and automatically generates background information about the progress and outcome of the Onin War.

[0892] 4. The device displays the correct / incorrect result and background information to the user.

[0893] 5. The server generates relevant supplementary information based on the user's learning history and sends it to the terminal.

[0894] 6. The terminal notifies and displays the supplementary information to the user.

[0895] 7. The server compares the user's grades with national data, calculates the standard deviation score, etc., and sends it to the terminal.

[0896] 8. The device displays the results of the comparison to the user.

[0897] 9. The device's emotion engine analyzes the user's emotions and sends the emotional state to the server.

[0898] 10. The server adjusts the learning content and supplementary information based on the emotion recognition results and sends them back to the device.

[0899] 11. The device displays the adjusted information to the user.

[0900] In this way, the present invention is a system that supports users' learning in multiple ways and promotes efficient and deep understanding. Furthermore, by taking into account the user's emotional state, it is possible to provide more personalized learning support.

[0901] The processing flow will be explained below.

[0902] Step 1:

[0903] The user starts the smartphone app and takes a picture of the test paper. When the user presses the capture button, the smartphone's camera function is activated and an image of the test paper is captured.

[0904] Step 2:

[0905] The device analyzes the captured image using its OCR (Optical Character Recognition) function. The OCR engine extracts character data from the image data and converts it into text format.

[0906] Step 3:

[0907] The device sends the extracted text information to the server, using a communication network to generate an API request and send the data to a specified endpoint on the server.

[0908] Step 4:

[0909] The server inputs the received text information into a generative AI model, which analyzes the content of the problem and uses natural language processing (NLP) algorithms to distinguish between the question and the answer.

[0910] Step 5:

[0911] The server determines whether the question is correct, refers to an internal database or knowledge base, evaluates whether the analyzed answer is correct, formats the evaluation result, and generates a data packet to send back to the terminal.

[0912] Step 6:

[0913] The device displays the correctness judgment result received to the user. The result is reflected in the user interface (UI), and a message such as "correct" or "incorrect" is displayed.

[0914] Step 7:

[0915] The server generates background information based on the results of the accuracy assessment, automatically generating relevant historical context and additional explanations using a natural language generation (NLG) model, or searching and extracting relevant information from a database.

[0916] Step 8:

[0917] The server sends the generated background information to the device, which then converts it into JSON format and sends it to the device via the communication network.

[0918] Step 9:

[0919] The device displays the background information received to the user, providing detailed historical background and additional explanations in an easy-to-understand manner to help the user gain a deeper understanding.

[0920] Step 10:

[0921] The device stores the user's learning history in a local database, accumulating information such as the questions studied, answers, correct / incorrect results, and learning dates and times.

[0922] Step 11:

[0923] The server analyzes the user's learning history data, generates supplementary information based on the user's learning patterns and preferences, and recommends appropriate learning materials and video content.

[0924] Step 12:

[0925] The server generates supplementary information and sends it to the terminal. Information tailored to each user's individual learning needs is sent, improving learning efficiency.

[0926] Step 13:

[0927] The device notifies and displays the supplementary information it receives to the user. The supplementary information is displayed in the notification bar or a dedicated screen within the app, making it easy for the user to access.

[0928] Step 14:

[0929] The server aggregates and analyzes the performance data collected from other users nationwide, and uses statistical algorithms to calculate the national average and standard deviation.

[0930] Step 15:

[0931] The server generates information comparing the user's performance with the national average and sends it to the device, including the difference between the individual user's performance and the national average, as well as the deviation score.

[0932] Step 16:

[0933] The terminal displays the received performance comparison results to the user, visualizing the results in the form of graphs and charts to help the user clearly understand their position.

[0934] Step 17:

[0935] The device captures the user's facial expressions and voice tone through a camera and microphone, and analyzes them with an emotion engine, which then recognizes the user's emotional state (e.g., joy, stress, concentration, etc.).

[0936] Step 18:

[0937] The server receives the emotion recognition results sent from the emotion engine and adjusts the learning content and supplementary information based on them. For example, if the user is feeling stressed, it provides relaxing content.

[0938] Step 19:

[0939] The device displays the adjusted learning content and supplementary information to the user, and reflects the adjustment results in the user interface, allowing the user to receive support that is in line with their emotions.

[0940] Through these steps, the present invention is a system that supports users' learning in multiple ways and promotes efficient and deep understanding. Furthermore, by taking into account the user's emotional state, it is possible to provide more personalized learning support.

[0941] Example 2

[0942] 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."

[0943] Conventional learning support systems are limited to determining whether questions submitted by users are correct or incorrect and providing background information, and lack the ability to flexibly respond to the user's learning situation and emotional state. Furthermore, because guidance and supplementary information are not provided efficiently or effectively to individual users, there are problems with not being able to improve learning efficiency or deepen understanding. Furthermore, when comparing scores at a national level, information based on the characteristics of individual users is often not provided, and sufficient support is often not provided. To solve these issues, a system with more advanced and flexible learning support functions is needed.

[0944] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0945] In this invention, the server includes a means for adjusting the learning content according to the user's emotions, a means for recognizing the user's emotions using image recognition and voice analysis, and a means for automatically generating background information based on a true / false judgment, thereby making it possible to provide individually optimized learning support based on the user's emotions and learning history.

[0946] A "user" is an individual who uses the system to carry out learning activities.

[0947] "Means for taking images" refers to the ability to take photos of test questions and study materials using a camera on a smartphone, tablet, etc.

[0948] "Means for extracting text information" refers to the function of extracting text data from images using OCR (optical character recognition) technology.

[0949] "Means for transmitting character information" refers to a function for transmitting extracted text data to a server via the Internet.

[0950] "Means for determining correctness" refers to the function of analyzing the text information sent and determining whether the answer to the question is correct.

[0951] "Means for displaying the judgment result to the user" refers to a function for displaying the result of the correctness judgment on the user's terminal.

[0952] "Means for automatically generating background information" refers to a function that automatically generates related supplementary explanations and detailed information based on a correct / incorrect judgment.

[0953] The "means for displaying background information to the user" refers to a function for displaying the generated background information on the user's terminal.

[0954] "Means for comparing performance with national performance" refers to a function that compares a user's performance data with that of other users nationwide and calculates rankings and standard deviations.

[0955] "Means for recognizing emotions" refers to a function that uses a camera or microphone to analyze and recognize the user's emotional state.

[0956] "Means for adjusting learning content" refers to a function that optimally adjusts the learning content and supplementary information provided based on the user's emotional state and learning history.

[0957] The "learning support system" of this invention provides multiple integrated functions to fully support users' learning activities. This system is realized by combining various functions, mainly image recognition, natural language processing, performance analysis, and information provision. Furthermore, by incorporating an emotion engine, it is possible to provide optimal learning support according to the user's emotional state.

[0958] 1. Photographing and recognizing test questions

[0959] User

[0960] Users take photos of test questions using their smartphone camera.

[0961] Terminal

[0962] The device analyzes the captured image using its OCR (Optical Character Recognition) function and extracts the character information in the image as text data. The specific software used is OCR technology (e.g., Google Cloud Vision or Tesseract). This extracted character information is sent from the device to a server.

[0963] server

[0964] The server inputs the received text information into a generative AI model, analyzes the content of the question, and determines whether it is correct or incorrect. For example, we use "OpenAI GPT-4" as the generative AI model. Examples of prompt sentences include:

[0965] Question: "What year did the Onin War start?"

[0966] Answer: "1467"

[0967] The generative AI model determines whether the prompt is correct or incorrect based on the prompt, and the server then sends the result back to the device.

[0968] Terminal

[0969] The terminal displays the result of the accuracy judgment received from the server to the user. For example, it displays the result in the format "What year did the Onin War start? 1467 -> Correct."

[0970] 2. Providing background information

[0971] server

[0972] Based on the results of the accuracy assessment, the server automatically generates relevant background information from a generative AI model or a pre-built database. Specific software examples include "natural language generation (NLG) models (e.g., OpenAI GPT-4 and Google BERT)."

[0973] Terminal

[0974] The generated background information is sent to the terminal and displayed to the user. For example, "Detailed background information about the Onin War" is displayed.

[0975] 3. Viewing learning history and supplementary information

[0976] Terminal

[0977] The device stores the user's learning history in a local database, including the correct / incorrect results for each question and the date and time of the learning.

[0978] server

[0979] The server analyzes the user's learning history data and generates supplemental information based on the user's learning patterns and preferences, using "machine learning algorithms and data analysis tools (e.g., Python, Pandas, Scikit-learn)."

[0980] Terminal

[0981] The generated supplementary information is sent to the terminal and notified and displayed to the user.

[0982] 4. Comparison of performance at the national level

[0983] server

[0984] The server will compile the academic performance information of students nationwide and generate statistical data. Specific technologies include "SQL databases and analytical tools (e.g., Tableau) used for data compilation and analysis."

[0985] Terminal

[0986] The user's academic performance data is sent to a server, and a comparison result with the national average is sent back to the terminal. The user can check this result and understand where their academic performance ranks nationwide.

[0987] 5. Incorporating an Emotional Engine

[0988] Terminal

[0989] The device uses a camera and microphone to recognize the user's emotions. Specific examples of software include emotion recognition software (e.g., Microsoft Azure Emotion API and IBM Watson Tone Analyzer).

[0990] server

[0991] The server receives the emotion recognition results sent from the emotion engine and adjusts the learning content and supplementary information based on the results. For example, if the user is feeling stressed, it provides relaxing content.

[0992] Terminal

[0993] The adjusted learning content and supplementary information are sent to the terminal and displayed to the user, allowing the user to receive the optimal learning environment and support.

[0994] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0995] Step 1:

[0996] User

[0997] The user takes a photo of the test question using the smartphone camera. The input is an image containing the test question, and the output is the captured image data. Specifically, the user opens the camera app, adjusts the frame so that the entire test paper fits within the frame, and presses the capture button.

[0998] Step 2:

[0999] Terminal

[1000] The device analyzes the captured image using its OCR (Optical Character Recognition) function and extracts the text information in the image as text data. The input is the captured image data, and the output is the extracted text data. Specifically, the device analyzes the image using OCR technology (for example, Google Cloud Vision or Tesseract) and extracts the question and answer in text format.

[1001] Step 3:

[1002] Terminal

[1003] The terminal sends the extracted text data to the server. The input is the extracted text data, and the output is the data sent to the server. Specifically, the terminal uses the HTTPS protocol to send the extracted text data to the specified endpoint of the server as a POST request.

[1004] Step 4:

[1005] server

[1006] The server inputs the received text data into the generative AI model, analyzes the content of the question, and determines whether it is correct or incorrect. The input is the received text data, and the output is the result of the correctness determination. Specifically, the server runs a Python script to generate a prompt for the generative AI model (e.g., GPT-4). An example of a prompt is as follows:

[1007] Question: "What year did the Onin War start?"

[1008] Answer: "1467"

[1009] The generative AI model determines whether the prompt is correct or not based on the prompt sentence and returns the result as "correct."

[1010] Step 5:

[1011] server

[1012] The server returns the result of the accuracy judgment to the terminal. The input is the accuracy judgment result, and the output is the response data to the terminal. Specifically, the server converts the accuracy judgment result into JSON format and returns it to the terminal as a response using the HTTPS protocol.

[1013] Step 6:

[1014] Terminal

[1015] The device displays the correctness judgment result received from the server to the user. The input is the correctness judgment result received from the server, and the output is the information displayed to the user. Specifically, the device analyzes the received JSON data and displays the result "correct" on the screen.

[1016] Step 7:

[1017] server

[1018] Based on the result of the accuracy judgment, the server automatically generates relevant background information from a generative AI model or a pre-built database. The input is the accuracy judgment result, and the output is the generated background information. Specifically, the server again sends a prompt to the generative AI model (e.g., GPT-4) to generate detailed historical background information.

[1019] Step 8:

[1020] server

[1021] The server sends the generated background information to the terminal. The input is the generated background information, and the output is the response data to the terminal. Specifically, the server returns the background information in JSON format to the terminal as a response.

[1022] Step 9:

[1023] Terminal

[1024] The terminal displays the background information received from the server to the user. The input is the background information received from the server, and the output is the information displayed to the user. Specifically, the terminal analyzes the received background information and displays it as "detailed background information about the Onin War."

[1025] Step 10:

[1026] Terminal

[1027] The device stores the user's learning history in a local database. The input is the learning result and the learning date and time, and the output is the data stored in the local database. Specifically, the device uses a local database such as SQLite and executes SQL queries to record the learning history.

[1028] Step 11:

[1029] server

[1030] The server analyzes the user's learning history data and generates related supplemental information. The input is the learning history data, and the output is the generated supplemental information. Specifically, the server analyzes the learning data using a machine learning algorithm and recommends additional learning materials and video content based on the user's interests and preferences.

[1031] Step 12:

[1032] server

[1033] The server sends the generated supplementary information to the terminal. The input is the generated supplementary information, and the output is the response data to the terminal. Specifically, the server returns the supplementary information in JSON format to the terminal as a response.

[1034] Step 13:

[1035] Terminal

[1036] The terminal displays the supplementary information received from the server to the user. The input is the supplementary information received from the server, and the output is the information displayed to the user. Specifically, the terminal displays the supplementary information as a notification, allowing the user to check the details.

[1037] Step 14:

[1038] server

[1039] The server aggregates the performance information of learners nationwide and generates statistical data. The input is the performance data of learners nationwide, and the output is the generated statistical data. Specifically, the server uses an SQL database to execute SQL queries that aggregate each user's performance and generate statistical data.

[1040] Step 15:

[1041] server

[1042] The server compares the user's performance data with the national average and sends the results to the device. The input is the user's performance data and the national average data, and the output is the comparison result. Specifically, the server returns the comparison result to the device as a response in JSON format.

[1043] Step 16:

[1044] Terminal

[1045] The terminal displays the comparison results with the national average to the user. The input is the comparison results received from the server, and the output is the information displayed to the user. Specifically, the terminal analyzes the received comparison results and displays the "comparison results with the national average" on the screen.

[1046] Step 17:

[1047] Terminal

[1048] The device uses its built-in emotion engine to analyze facial expression data and voice tone obtained from the camera and microphone to recognize the user's emotions. The input is facial expression data and voice tone, and the output is the emotion recognition result. Specifically, the device uses emotion recognition software (for example, Microsoft Azure Emotion API or IBM Watson Tone Analyzer) to perform the analysis.

[1049] Step 18:

[1050] Terminal

[1051] The device sends the emotion recognition results to the server. The input is the emotion recognition results, and the output is the data to be sent to the server. Specifically, the device sends the emotion recognition results in JSON format to the server as a POST request.

[1052] Step 19:

[1053] server

[1054] The server receives the emotion recognition results and adjusts the learning content and supplementary information based on them. The input is the emotion recognition results, and the output is the adjusted learning content and supplementary information. Specifically, the server analyzes the emotion data and adjusts the learning content to provide relaxing content if the user is feeling stressed.

[1055] Step 20:

[1056] server

[1057] The server sends the adjusted learning content and supplementary information to the terminal. The input is the adjusted learning content and supplementary information, and the output is the response data to the terminal. Specifically, the server returns the adjusted information in JSON format to the terminal as a response.

[1058] Step 21:

[1059] Terminal

[1060] The device displays the adjusted learning content and supplementary information to the user. The input is the adjusted information received from the server, and the output is the information displayed to the user. Specifically, the device analyzes the received information and displays it in a way that provides the user with the optimal learning environment.

[1061] (Application example 2)

[1062] 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."

[1063] Conventional learning support systems were able to judge whether test questions photographed by users were correct or incorrect and provide related background information, but they did not adjust the learning content based on the user's emotional state, and therefore lacked efficient learning support tailored to the emotions of each individual user. Furthermore, real-time learning support is required for use in physical stores, but a specific system to achieve this was lacking.

[1064] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for a user to take an image and extract text information from the image, means for analyzing the extracted text information based on a generative AI model and determining whether the question is correct, means for displaying the determination result to the user, means for automatically generating background information based on the correctness determination, means for displaying the generated background information to the user, means for comparing the user's performance with national performance, and means for recognizing the user's emotional state using an emotion recognition engine and adjusting the learning content. This enables efficient and personalized learning support according to the individual emotional state of each user, even in physical stores.

[1065] "User" refers to an individual who uses the system to carry out learning activities.

[1066] An "image" is a digital representation of visual information captured by a user.

[1067] "Text information" refers to text data extracted from an image.

[1068] "Generative AI models" refer to algorithms and models that use artificial intelligence technology to generate and analyze information.

[1069] "Correctness determination" refers to determining whether the answer given by the user is correct or incorrect.

[1070] "Background information" refers to additional explanation or information related to a question or problem.

[1071] "National level" refers to a scope that goes beyond a specific region or community and is effective nationwide.

[1072] "Grades" are evaluations and expressions of a user's learning progress and results expressed numerically or in letters.

[1073] An "emotion recognition engine" refers to hardware or software that analyzes and recognizes a user's emotional state.

[1074] "Learning content" refers to the materials, questions, information, and content generally provided for users to study.

[1075] "Adjustment" refers to changing content or settings based on specific conditions or circumstances.

[1076] This invention is a learning support system that integrates image recognition, natural language processing, performance analysis, and information provision functions to support users' learning activities. It is intended for use in brick-and-mortar stores, but it also functions effectively in other environments.

[1077] System Configuration

[1078] The system consists of multiple hardware and software components, including a user terminal, a server, and a learning assistant robot.

[1079] User terminal

[1080] User devices are primarily smartphones and tablets with the following features:

[1081] 1. Image capture: The user takes a picture of the study question using the smartphone camera.

[1082] 2. OCR (Optical Character Recognition): Uses OCR technology (e.g., Tesseract OCR) to extract text information from captured images.

[1083] server

[1084] The server has the following features:

[1085] 1. Character information analysis and accuracy determination: The character information extracted by OCR is analyzed based on a generative AI model (e.g., the GPT-3.5-turbo model) to determine whether the question is correct.

[1086] 2. Generating background information: Based on the results of the correct / incorrect judgment, background information related to the problem is automatically generated using a natural language generation (NLG) model.

[1087] 3. Performance analysis and comparison: Aggregate the performance of all users and compare it at a national level.

[1088] 4. Emotion Recognition: Use an emotion recognition engine to analyze the user's emotional state and adjust the learning content.

[1089] Learning Assistant Robot

[1090] The physical store will be equipped with a learning assistant robot, which will have the following functions:

[1091] 1. User support: Providing real-time learning support in-store and responding to user questions and problems.

[1092] 2. Facial Recognition and Emotion Analysis: Uses a camera and microphone to analyze a user's facial expressions and tone of voice to recognize their emotional state.

[1093] 3. Information provision: The generated background information and performance information are displayed to the user.

[1094] Specific examples

[1095] For example, consider a scenario where a user is in a store and asks a history question to a dedicated learning assistant robot, "What year did the Onin War start?", while simultaneously taking a photo of the question with their smartphone. In this scenario, the process would proceed as follows:

[1096] 1. The user takes a photo of the problem with their smartphone.

[1097] 2. The device uses its OCR function to extract text information from the image, such as "What year did the Onin War start?", and sends it to the server.

[1098] 3. The server uses a generative AI model to determine whether the answer is correct or not, and generates background information on the progress and outcome of the Onin War.

[1099] 4. The device displays the correct / incorrect result and background information to the user.

[1100] 5. The server generates relevant supplementary information based on the user's learning history and sends it to the terminal.

[1101] 6. The terminal notifies and displays the supplementary information to the user.

[1102] 7. The server compares the user's grades with national data, calculates the standard deviation score, etc., and sends it to the terminal.

[1103] 8. The device displays the results of the comparison to the user.

[1104] 9. The device's emotion recognition engine analyzes the user's emotions and sends the emotional state to the server.

[1105] 10. The server adjusts the learning content and supplementary information based on the emotion recognition results and sends them back to the device.

[1106] 11. The device displays the adjusted information to the user.

[1107] Examples of prompts:

[1108] "Provide detailed background information for the following question: In what year did the Onin War begin?"

[1109] As a result, this system enables efficient and personalized learning support based on each user's emotional state, even in physical stores.

[1110] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1111] Step 1:

[1112] The user takes a photo of the problem with their smartphone.

[1113] Input: Image of the training question

[1114] Output: Captured image data

[1115] Specific operation: The user takes a photo of the study question using the smartphone camera. The image data is saved on the device.

[1116] Step 2:

[1117] The device uses the OCR function to extract text information from the image and send it to the server.

[1118] Input: Captured image data

[1119] Output: Extracted text

[1120] Specific operation: Extract text information from the captured image using pytesseract, convert the text information into text format, and then send this text data to the server.

[1121] Step 3:

[1122] The server uses a generative AI model to determine accuracy and generates background information on the progress and results of the Onin War.

[1123] Input: Extracted text

[1124] Output: Correct / incorrect result and background information

[1125] Specific operation: Based on the received text information, the server uses a generative AI model (e.g., the GPT-3.5-turbo model) to analyze whether the question is correct. After determining whether the question is correct, the server uses the model or a pre-built database to generate background information related to the question.

[1126] Step 4:

[1127] The terminal displays the result of the correct / incorrect judgment and background information to the user.

[1128] Input: Correct / incorrect result and background information

[1129] Output: The result of the test and background information displayed on the user's screen

[1130] Specific operation: The result of the correct answer and background information sent from the server are received and displayed on the user's device. For example, "The answer to the question is correct" is displayed along with "the progress and outcome of the Onin War."

[1131] Step 5:

[1132] The server generates related supplementary information based on the user's learning history and sends it to the terminal.

[1133] Input: User learning history data

[1134] Output: Supplementary information

[1135] Specific operation: The server analyzes the user's learning history data and generates supplementary information to deepen understanding of the relevant topic. The generated supplementary information is sent to the terminal.

[1136] Step 6:

[1137] The terminal notifies and displays the supplementary information to the user.

[1138] Input: Additional information

[1139] Output: Supplementary information notified and displayed to the user

[1140] Specific operation: The supplementary information received from the server is notified to the user and displayed on the device screen. Specifically, it provides links such as "Video content related to the Onin War."

[1141] Step 7:

[1142] The server compares the user's grades with national data, calculates the standard deviation score, etc., and sends it to the terminal.

[1143] Input: User performance data, nationwide learner data

[1144] Output: Standard deviation and comparison results

[1145] Specific operation: The server compares the user's performance data with the data of learners nationwide, calculates the deviation score and ranking, and sends the calculation results to the terminal.

[1146] Step 8:

[1147] The terminal displays the results of the comparison of the results to the user.

[1148] Input: Standard deviation or comparison result

[1149] Output: The result of the comparison is displayed on the user's screen.

[1150] Specific operation: The results of the grade comparison sent from the server are notified to the user and displayed on the device screen. Specifically, a message such as "Your grades are above the national average" is displayed.

[1151] Step 9:

[1152] The device's emotion recognition engine analyzes the user's emotions and transmits the emotional state to the server.

[1153] Input: facial expression data, voice data

[1154] Output: Emotion recognition result

[1155] Specific operation: The device uses a built-in camera and microphone to collect and analyze the user's facial expressions and voice tone. The analysis results are sent to the server.

[1156] Step 10:

[1157] The server adjusts the learning content and supplementary information based on the emotion recognition results and sends them back to the device.

[1158] Input: Emotion recognition results

[1159] Output: Adjusted learning results and supplementary information

[1160] Specific operation: The server adjusts the learning content and supplementary information based on the received emotion recognition results. For example, if the user is feeling stressed, it will provide relaxing content. The adjusted information is then sent back to the device.

[1161] Step 11:

[1162] The terminal displays the adjusted information to the user.

[1163] Input: Adjusted learning content and supplementary information

[1164] Output: Adjusted information displayed on the user screen

[1165] Specific operation: The adjusted learning content and supplementary information sent from the server are received and displayed on the user's device. Specifically, a "relaxing video about the Onin War" is displayed.

[1166] 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.

[1167] 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.

[1168] 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.

[1169] [Third embodiment]

[1170] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1171] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[1172] 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).

[1173] 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.

[1174] 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.

[1175] 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).

[1176] 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. 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.

[1177] 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.

[1178] 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.

[1179] 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.

[1180] 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.

[1181] 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."

[1182] The "learning support system" of this invention provides an integrated set of functions, including image recognition, natural language processing, performance analysis, and information provision, to fully support users' learning activities. This system is implemented as an application that users can easily use on devices such as smartphones and tablets.

[1183] 1. Photographing and recognizing test questions

[1184] Terminal

[1185] First, the user takes a photo of the test question using the smartphone camera. The device analyzes the captured image using its OCR (Optical Character Recognition) function and extracts the characters in the image as text information. This extracted text information is then sent to the server.

[1186] server

[1187] The server inputs the received text information into the generative AI model, analyzes the content of the question, and judges whether it is correct or not. The analyzed result of the judgment is then sent back to the device.

[1188] Terminal

[1189] The device displays the result of the accuracy judgment received from the server to the user. For example, if the user answers "1467" to the history question "What year did the Onin War start?", the result of the accuracy judgment is displayed as "correct answer."

[1190] 2. Providing background information

[1191] server

[1192] Based on the results of the correct / incorrect decision, the server automatically generates relevant background information, specifically by extracting historical background and additional explanatory information related to the problem from natural language generation (NLG) models and pre-built databases.

[1193] Terminal

[1194] The generated background information is sent to the terminal and displayed to the user. For example, detailed background information about the Onin War and the story of its outcome are displayed.

[1195] 3. Viewing learning history and supplementary information

[1196] Terminal

[1197] The user's learning history is stored in a local database, which records the correct / incorrect results for each question, the date and time of study, and so on.

[1198] server

[1199] The server analyzes the user's learning history data and generates supplementary information based on the user's learning patterns and interests. For example, if a user is not good at history, additional learning materials and video content that will help them improve their understanding will be recommended.

[1200] Terminal

[1201] The generated supplemental information is sent to the device and displayed to the user, and is displayed as a notification, allowing the user to further study based on it.

[1202] 4. Comparison of performance at the national level

[1203] server

[1204] The server compiles and analyzes the performance information of learners nationwide to generate statistical data, which makes it possible to compare each user's performance with the national average and deviation score.

[1205] Terminal

[1206] The user's academic performance data is sent to the server, and a comparison with the national average is sent back to the terminal, allowing the user to understand where their academic performance ranks nationally.

[1207] Specific examples

[1208] For example, suppose a user photographs a history test question, "What year did the Onin War start?" and answers "1467." In this case, the process would proceed as follows:

[1209] 1. The user takes a photo of the problem with their smartphone.

[1210] 2. The device uses its OCR function to extract "What year did the Onin War start?" and "1467" from the image and send it to the server.

[1211] 3. The server uses the generated AI model to determine whether the answer is correct, determines it is correct, and automatically generates background information about the progress and outcome of the Onin War.

[1212] 4. The device displays the correct / incorrect result and background information to the user.

[1213] 5. The server generates relevant supplementary information based on the user's learning history and sends it to the terminal.

[1214] 6. The terminal notifies and displays the supplementary information to the user.

[1215] 7. The server compares the user's grades with national data, calculates the standard deviation score, etc., and sends it to the terminal.

[1216] 8. The device displays the results of the comparison to the user.

[1217] In this way, the present invention is a system that supports the user's learning activities in many ways and enables more effective learning.

[1218] The processing flow will be explained below.

[1219] Step 1:

[1220] The user starts the smartphone app and takes a picture of the test paper. When the user presses the capture button, the smartphone's camera function is activated and an image of the test paper is captured.

[1221] Step 2:

[1222] The device analyzes the captured image using its OCR (Optical Character Recognition) function. The OCR engine extracts character data from the image data and converts it into text format.

[1223] Step 3:

[1224] The device sends the extracted text information to the server, using a communication network to generate an API request and send the data to a specified endpoint on the server.

[1225] Step 4:

[1226] The server inputs the received text information into a generative AI model, which analyzes the content of the problem and uses natural language processing (NLP) algorithms to distinguish between the question and the answer.

[1227] Step 5:

[1228] The server determines whether the question is correct, refers to an internal database or knowledge base, evaluates whether the analyzed answer is correct, formats the evaluation result, and generates a data packet to send back to the terminal.

[1229] Step 6:

[1230] The device displays the correctness judgment result received to the user. The result is reflected in the user interface (UI), and a message such as "correct" or "incorrect" is displayed.

[1231] Step 7:

[1232] The server generates background information based on the results of the accuracy assessment, automatically generating relevant historical context and additional explanations using a natural language generation (NLG) model, or searching and extracting relevant information from a database.

[1233] Step 8:

[1234] The server sends the generated background information to the device, which then converts it into JSON format and sends it to the device via the communication network.

[1235] Step 9:

[1236] The device displays the background information received to the user, providing detailed historical background and additional explanations in an easy-to-understand manner to help the user gain a deeper understanding.

[1237] Step 10:

[1238] The device stores the user's learning history in a local database, accumulating information such as the questions studied, answers, correct / incorrect results, and learning dates and times.

[1239] Step 11:

[1240] The server analyzes the user's learning history data, generates supplementary information based on the user's learning patterns and preferences, and recommends appropriate learning materials and video content.

[1241] Step 12:

[1242] The server generates supplementary information and sends it to the terminal. Information tailored to each user's individual learning needs is sent, improving learning efficiency.

[1243] Step 13:

[1244] The device notifies and displays the supplementary information it receives to the user. The supplementary information is displayed in the notification bar or a dedicated screen within the app, making it easy for the user to access.

[1245] Step 14:

[1246] The server aggregates and analyzes the performance data collected from other users nationwide, and uses statistical algorithms to calculate the national average and standard deviation.

[1247] Step 15:

[1248] The server generates information comparing the user's performance with the national average and sends it to the device, including the difference between the individual user's performance and the national average, as well as the deviation score.

[1249] Step 16:

[1250] The terminal displays the received performance comparison results to the user, visualizing the results in the form of graphs and charts to help the user clearly understand their position.

[1251] Through the above steps, the present invention is a system that supports users' learning in many ways and promotes efficient and deep understanding.

[1252] Example 1

[1253] 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."

[1254] Conventional learning support systems have limitations in their ability to accurately extract text information from images taken by users and determine whether questions are correct or incorrect. Furthermore, they lack the functionality to analyze learning history and provide users with appropriate supplementary information, or the ability to compare scores nationwide in real time. This makes it difficult for users to fully grasp their learning progress and level of understanding, making it difficult to study efficiently.

[1255] 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.

[1256] In this invention, the server includes means for a user to take an image and extract text information from the image, means for transmitting the extracted text information, means for analyzing the transmitted text information and judging whether the question is correct or incorrect, means for displaying the judgment result to the user, means for automatically generating background information based on the judgment of correctness, means for displaying the generated background information to the user, means for comparing the user's performance with national performance, means for accumulating and analyzing the user's learning history, means for generating supplementary information based on the analysis result and notifying the user, and means for aggregating and analyzing national performance data and comparing the user's performance with the national average and deviation score. This allows the user to grasp their own learning progress and level of understanding in real time and to study efficiently and effectively.

[1257] A "user" is a person who accesses the learning support system using a device such as a smartphone or tablet and engages in learning activities.

[1258] "Taking an image" refers to the act of a user using the camera on a smartphone or tablet to capture an image of a study question or document.

[1259] "Extracting text information" refers to the process of using OCR (Optical Character Recognition) technology to identify and extract text from a captured image as text data.

[1260] "Send" refers to the act of transferring data from a terminal to a server.

[1261] "Analysis" refers to the process in which the server analyzes the data it receives using technologies such as generative AI models to determine the content of the question and whether it is correct or incorrect.

[1262] "Determining correctness" means using a generative AI model to determine whether an answer to a question is correct or incorrect.

[1263] "Display" refers to the act of visually presenting analysis results, background information, performance information, etc. to the user on the terminal display.

[1264] "Background information" is information containing additional knowledge or explanation related to the content of the problem.

[1265] "Auto-generation" refers to the process of automatically generating the required information using pre-built databases and natural language generation (NLG) models.

[1266] "Study history" refers to a record of a user's past learning activities, answer results, study dates and times, etc.

[1267] "Analysis" refers to clarifying a user's learning patterns and tendencies based on accumulated learning history data.

[1268] "Supplemental information" refers to additional educational materials and video content provided to help users learn more efficiently.

[1269] "Notification" refers to the act of informing the user of generated information, supplementary materials, etc.

[1270] "National level results" refers to statistical data such as averages and standard deviations obtained by compiling and analyzing the results data of other users nationwide.

[1271] The "learning support system" of this invention provides an integrated set of functions, including image recognition, natural language processing, performance analysis, and information provision, to fully support users' learning activities. This system is implemented as an application that users can easily use on devices such as smartphones and tablets.

[1272] Details of the hardware and software you will use

[1273] Terminal

[1274] This system mainly uses mobile devices such as smartphones and tablets. The devices are equipped with a camera, OCR (Optical Character Recognition) functionality, and a display. The OCR functionality incorporates common OCR software and is used to identify characters from images.

[1275] System processing flow

[1276] 1. Photographing test questions and character recognition

[1277] First, the user takes a photo of the test question using the smartphone camera. The device analyzes the captured image using OCR and extracts the characters in the image as text information. This extracted text information is then sent to the server.

[1278] 2. Correct / incorrect judgment

[1279] The server inputs the received text information into the generative AI model, analyzes the content of the question, and judges whether it is correct or not. The analyzed result of the judgment is then sent back to the device.

[1280] Example: If a user answers "1467" to the history question "What year did the Onin War start?", the result of the accuracy test will be displayed as "correct answer."

[1281] 3. Automatic generation of background information

[1282] Based on the results of the correct / incorrect judgment, the server automatically generates relevant background information. Specifically, it extracts historical background and additional explanatory information related to the problem from natural language generation (NLG) models and pre-built databases. The generated background information is sent to the terminal and displayed to the user.

[1283] Example: The user gets the answer right and is shown detailed background information about the Onin War, as well as the story of its outcome.

[1284] 4. Accumulation and analysis of learning history

[1285] The device stores the user's learning history in a local database. The server periodically collects and analyzes the user's learning history data. For example, it can identify learning trends, such as the user's weakness in history questions. Based on the analysis, supplementary information to deepen understanding is generated and sent to the device.

[1286] Example: Additional educational materials and video content are recommended to help users who are weak in history to deepen their understanding.

[1287] 5. Comparison of performance at the national level

[1288] The server compiles and analyzes the academic performance information of learners nationwide. This makes it possible to compare each user's performance with the national average and deviation score. The user's performance data is sent to the server, and the results of the comparison with the national average are sent back to the terminal. This allows the user to understand where their learning performance ranks nationwide.

[1289] Example: Information such as the user's performance being in the top 20% of the national average is displayed on the device.

[1290] Prompt Sentence Examples

[1291] "Extract the text in the image below and determine whether the question is correct or incorrect. Then generate the answer and background information in the same image."

[1292] Through the above processing, the present invention is a system that supports the user's learning activities in many ways and realizes more effective learning.

[1293] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1294] Step 1: Photographing the test questions and recognizing the characters

[1295] Terminal

[1296] The user takes a photo of a test question using their smartphone camera. This image becomes the input. The device uses its OCR (Optical Character Recognition) function to analyze the character information from the captured image and extract text data. The extracted text data becomes the output.

[1297] Specific operation: The user launches the camera app on their smartphone and takes a picture of the test question. The app then performs OCR, generating the text data "What year did the Onin War start?" and "1467," and sends it to the server.

[1298] Step 2: Send text information

[1299] Terminal

[1300] The extracted text information is sent to the server. This text information is the input, and the transmission to the server is the output.

[1301] Specific operation: The device sends the text data "What year did the Onin War start?" and "1467" to the server.

[1302] Step 3: Correct / incorrect

[1303] server

[1304] The text information received by the server is input into the generative AI model. This text information becomes the input. The generative AI model analyzes the content of the question and determines whether it is correct or incorrect. The result of the correct or incorrect determination becomes the output.

[1305] Specific operation: The server inputs the received text data, "What year did the Onin War start?" and "1467," into the generative AI model, which determines that "1467" is the correct answer. The result is then sent back to the device.

[1306] Step 4: Displaying the results

[1307] Terminal

[1308] The terminal receives the result of the correctness judgment from the server and displays it to the user. The result of the correctness judgment is the input, and the display is the output.

[1309] Specific operation: The terminal displays the "correct" result received from the server on the user's screen.

[1310] Step 5: Automatically generate background information

[1311] server

[1312] Based on the accuracy assessment results, the server automatically generates background information. The accuracy assessment results are the input. The server generates related background information using a natural language generation (NLG) model or a pre-built database. This generated background information is the output.

[1313] Specific operation: The server obtains the "correct answer" result and automatically generates an "Outline of the Onin War and its historical background" using the NLG model and sends it to the terminal.

[1314] Step 6: Display background information

[1315] Terminal

[1316] The generated background information is received by the terminal and displayed to the user. The background information is the input and the display is the output.

[1317] Specific operation: The information received by the terminal, "Outline of the Onin War and its historical background," is displayed on the screen.

[1318] Step 7: Accumulating learning history

[1319] Terminal

[1320] The user's learning history is stored in a local database. The input is text information about the learning questions and correct / incorrect results. The output is to save this information in the local database.

[1321] Specific operation: The device stores the user's answers and study date and time in a local database.

[1322] Step 8: Analyze your learning history

[1323] server

[1324] The user's learning history data is collected periodically and analyzed. The accumulated learning history data is the input. The server analyzes this data to understand the user's learning patterns and tendencies. The results of this analysis are the output.

[1325] Specific operation: The server analyzes the user's learning history data and identifies a learning tendency such as "I feel uncomfortable with history questions."

[1326] Step 9: Generate and notify supplementary information

[1327] server

[1328] Supplementary information is generated based on the analysis results and notified to the user. The analysis results are the input. The server generates supplementary information and sends it to the user's terminal. This supplementary information is the output.

[1329] Specific operation: The server generates "educational materials and video content that will deepen the understanding of users who are not good at history" and sends them to the device. The device displays this to the user as a notification.

[1330] Step 10: Comparing performance at the national level

[1331] server

[1332] The performance information of learners nationwide is compiled and compared with the user's performance. National performance data is input. The server analyzes this data and compares the user's performance with the national average and deviation value. The results of this comparison are output.

[1333] Specific operation: The server analyzes the performance data collected from all over the country, calculates information such as whether the user's performance is in the top 20% of the national average, and sends it to the terminal. The terminal then displays this information to the user.

[1334] (Application example 1)

[1335] 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."

[1336] On conventional online shopping sites, users must research product details themselves when selecting a product, making it difficult to find related or recommended products. Furthermore, accurate product suggestions based on users' purchase history are not sufficiently provided. Therefore, there is a need for a system that allows users to efficiently select products and smoothly proceed with their purchasing process.

[1337] 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.

[1338] In this invention, the server includes means for allowing a user to take an image and extract text information from the image, means for transmitting the extracted text information, means for analyzing the transmitted text information and generating product information, means for displaying the generated product information to the user, means for automatically generating related products based on the generated product information, means for displaying the generated related product information to the user, means for generating individual recommended products based on the user's purchase history, means for displaying the generated recommended product information to the user, and means for comparing the generated recommended product information with purchase history data of other users. This enables a user to efficiently obtain detailed product information, related products, and recommended products simply by taking a photo of a product.

[1339] A "user" is a consumer who uses the system to obtain product information and assists in purchasing.

[1340] "Image" means visual data of a product or item photographed using a smartphone or other device.

[1341] "Text information" refers to text information contained in an image and is extracted using the OCR function.

[1342] "Extraction" is the process of extracting the necessary text information from an image.

[1343] "Sending" refers to transferring the extracted character information from the terminal to the server in the form of data.

[1344] "Analysis" refers to understanding product information and related information based on the text information sent and processing it using a generative AI model.

[1345] "Product information" refers to detailed data about a particular product, such as specifications, price, and ratings.

[1346] "Related products" are other products that may be of interest to the user that are suggested based on the analyzed product information.

[1347] "Display" refers to visually showing information sent from the server on the user's terminal screen.

[1348] "Purchase history" is data on products purchased in the past by a user, and is used to generate recommended products.

[1349] "Recommended products" are products that are likely to be purchased and are identified based on the user's purchase history and preferences.

[1350] "Purchase history data of other users" refers to data on the past purchase history of multiple users stored in the system.

[1351] "Comparison" is the process of evaluating and analyzing one piece of data (e.g., a user's purchase history) against other data (e.g., the purchase history of other users).

[1352] The "product proposal support system" for realizing the present invention can be implemented using the following hardware and software.

[1353] Hardware and software used

[1354] Smartphone: A device that takes a picture of a product and uses OCR to extract text information.

[1355] Server: A computer that performs analysis and data aggregation.

[1356] OCR (Optical Character Recognition) software: A tool that extracts textual information from images (e.g., Tesseract OCR).

[1357] Generative AI model: An AI model that analyzes product information and related products and makes suggestions (e.g., GPT-3).

[1358] Natural Language Processing (NLP) models: Models that understand and analyze extracted text information.

[1359] Database: A storage system (e.g., MySQL, PostgreSQL, etc.) that stores product information, user purchase history, and related products.

[1360] System Operation

[1361] 1. The user takes a photo of the product they plan to purchase using their smartphone.

[1362] 2. The device uses OCR software to extract text information from the image.

[1363] 3. The extracted text information is sent to the server.

[1364] 4. The server analyzes the received text information using a generative AI model to generate detailed product information and related product information.

[1365] 5. The server sends related product information and detailed information to the terminal.

[1366] 6. The terminal displays the received information to the user.

[1367] 7. The server retrieves the user's purchase history from the database and automatically generates recommended products using a generative AI model.

[1368] 8. The generated recommended product information is sent to the terminal and displayed to the user.

[1369] 9. The server compares your purchase history data with that of other users and performs a statistical evaluation of your purchasing patterns.

[1370] 10. The evaluation results are sent to the terminal and displayed to the user.

[1371] Specific examples

[1372] For example, if a user takes a picture of a new smartphone, a sample prompt might look like this:

[1373] "Take a photo of a smartphone (new model) and display detailed product information. Please suggest recommended products based on the user's purchase history."

[1374] A user takes a photo of their smartphone, and the OCR function extracts text information such as "new model smartphone." This text information is sent to a server, where a generative AI model generates detailed information about the "new model smartphone" and related products. This information is then sent to the device and displayed to the user. If the user has previously purchased electronic devices, related products and accessories are recommended based on their purchase history, and their purchasing trends are displayed in comparison with the purchasing behavior of other users. In this way, users can efficiently select products and smoothly proceed with their purchasing behavior.

[1375] This allows users to efficiently obtain detailed product information, related products, and recommended products simply by taking a photo of the product. Also, by comparing the purchase data of other users, users can understand their own purchasing trends and make better decisions.

[1376] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1377] Step 1:

[1378] The user takes a photo of the product they plan to purchase using their smartphone.

[1379] Input: Product image

[1380] Output: None

[1381] Specific action: A user takes a photo of a specific product using their smartphone camera.

[1382] Step 2:

[1383] The device uses OCR software to extract text information from the image.

[1384] Input: Product image

[1385] Output: Extracted text information

[1386] Specific operation: Image analysis is performed using on-device OCR (e.g., Tesseract OCR), and character information such as the product name and model number is extracted as text.

[1387] Step 3:

[1388] The extracted character information is sent to the server.

[1389] Input: Extracted text information

[1390] Output: Confirmation of sending text information to the server

[1391] What it does: The extracted text data is sent over a network to a server, often using the HTTP or HTTPS protocol.

[1392] Step 4:

[1393] The server uses a generative AI model based on the text information received to analyze and generate product information.

[1394] Input: Text information sent

[1395] Output: Generated product information

[1396] How it works: The server uses a generative AI model (e.g., GPT-3) to analyze the received text data and generate detailed product information (specifications, price, reviews, etc.). It may also collect information from external databases or APIs as needed.

[1397] Step 5:

[1398] The generated product information is transmitted to the terminal.

[1399] Input: Generated product information

[1400] Output: Confirmation of sending product information to the terminal

[1401] Specific operation: The product information generated by the server is sent to the terminal in a data format such as JSON.

[1402] Step 6:

[1403] The terminal displays the generated product information to the user.

[1404] Input: Generated product information

[1405] Output: Product details page shown to the user

[1406] What it does: Product information is displayed on the device screen, including the product name, price, specifications, reviews, etc.

[1407] Step 7:

[1408] The server retrieves the user's purchase history from the database and automatically generates recommended products using a generative AI model.

[1409] Input: User purchase history

[1410] Output: Generated product recommendations

[1411] Specific operation: The server retrieves purchase history from the database, analyzes the user's preferences and past purchase data using a generative AI model, and generates recommended products based on that.

[1412] Step 8:

[1413] The generated recommended product information is sent to the terminal and displayed to the user.

[1414] Input: Generated recommended product information

[1415] Output: Recommended products displayed to the user

[1416] Specific operation: The server sends recommended product information to the device, and the device displays the recommended products to the user, for example, recommended accessories and related products.

[1417] Step 9:

[1418] The server compares the purchase history data of other users and performs a statistical evaluation of the purchasing patterns.

[1419] Input: Purchase history data of other users

[1420] Output: Statistical evaluation results

[1421] What it does: The server compares your purchasing history data with that of other users and uses statistical analysis and machine learning models to assess how your purchasing patterns are different or similar to others.

[1422] Step 10:

[1423] The evaluation results are sent to the terminal and displayed to the user.

[1424] Input: Statistical evaluation results

[1425] Output: Screen showing the evaluation results

[1426] Specific operation: The server sends the statistical evaluation results to the terminal and displays them on the user's screen, allowing the user to understand how their purchasing behavior differs from that of other users.

[1427] 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.

[1428] The "learning support system" of this invention provides an integrated set of functions, including image recognition, natural language processing, performance analysis, and information provision, to fully support users' learning activities. Furthermore, this system combines an emotion engine that recognizes the user's emotions to provide optimal learning support according to the user's emotional state.

[1429] 1. Photographing and recognizing test questions

[1430] Terminal

[1431] First, the user takes a photo of the test question using the smartphone camera. The device analyzes the captured image using its OCR (Optical Character Recognition) function and extracts the characters in the image as text information. This extracted text information is then sent to the server.

[1432] server

[1433] The server inputs the received text information into the generative AI model, analyzes the content of the question, and judges whether it is correct or not. The analyzed result of the judgment is then sent back to the device.

[1434] Terminal

[1435] The device displays the result of the accuracy judgment received from the server to the user. For example, if the user answers "1467" to the history question "What year did the Onin War start?", the result of the accuracy judgment is displayed as "correct answer."

[1436] 2. Providing background information

[1437] server

[1438] Based on the results of the correct / incorrect decision, the server automatically generates relevant background information, specifically by extracting historical background and additional explanatory information related to the problem from natural language generation (NLG) models and pre-built databases.

[1439] Terminal

[1440] The generated background information is sent to the terminal and displayed to the user. For example, detailed background information about the Onin War and the story of its outcome are displayed.

[1441] 3. Viewing learning history and supplementary information

[1442] Terminal

[1443] The user's learning history is stored in a local database, which records the correct / incorrect results for each question, the date and time of study, and so on.

[1444] server

[1445] The server analyzes the user's learning history data and generates supplementary information based on the user's learning patterns and interests. For example, if a user is not good at history, additional learning materials and video content that will help them improve their understanding will be recommended.

[1446] Terminal

[1447] The generated supplemental information is sent to the device and displayed to the user, and is displayed as a notification, allowing the user to further study based on it.

[1448] 4. Comparison of performance at the national level

[1449] server

[1450] The server compiles and analyzes the performance information of learners nationwide to generate statistical data, which makes it possible to compare each user's performance with the national average and deviation score.

[1451] Terminal

[1452] The user's academic performance data is sent to the server, and a comparison with the national average is sent back to the terminal, allowing the user to understand where their academic performance ranks nationally.

[1453] 5. Incorporating an Emotional Engine

[1454] Terminal

[1455] The device uses a built-in emotion engine to recognize the user's emotions, analyzing facial expression data and voice tone obtained through the camera and microphone to identify the user's emotional state.

[1456] server

[1457] The server receives the emotion recognition results sent from the emotion engine and reflects them in the learning content and the provision of supplementary information. For example, if the user is feeling stressed, the server may provide content that helps them relax.

[1458] Terminal

[1459] The device displays adjusted learning content and supplementary information based on the emotion recognition results to the user, providing appropriate support in line with the user's emotions and improving the effectiveness of learning.

[1460] Specific examples

[1461] For example, suppose a user photographs a history test question, "What year did the Onin War start?" and answers "1467." In this case, the process would proceed as follows:

[1462] 1. The user takes a photo of the problem with their smartphone.

[1463] 2. The device uses its OCR function to extract "What year did the Onin War start?" and "1467" from the image and send it to the server.

[1464] 3. The server uses the generated AI model to determine whether the answer is correct, determines it is correct, and automatically generates background information about the progress and outcome of the Onin War.

[1465] 4. The device displays the correct / incorrect result and background information to the user.

[1466] 5. The server generates relevant supplementary information based on the user's learning history and sends it to the terminal.

[1467] 6. The terminal notifies and displays the supplementary information to the user.

[1468] 7. The server compares the user's grades with national data, calculates the standard deviation score, etc., and sends it to the terminal.

[1469] 8. The device displays the results of the comparison to the user.

[1470] 9. The device's emotion engine analyzes the user's emotions and sends the emotional state to the server.

[1471] 10. The server adjusts the learning content and supplementary information based on the emotion recognition results and sends them back to the device.

[1472] 11. The device displays the adjusted information to the user.

[1473] In this way, the present invention is a system that supports users' learning in multiple ways and promotes efficient and deep understanding. Furthermore, by taking into account the user's emotional state, it is possible to provide more personalized learning support.

[1474] The processing flow will be explained below.

[1475] Step 1:

[1476] The user starts the smartphone app and takes a picture of the test paper. When the user presses the capture button, the smartphone's camera function is activated and an image of the test paper is captured.

[1477] Step 2:

[1478] The device analyzes the captured image using its OCR (Optical Character Recognition) function. The OCR engine extracts character data from the image data and converts it into text format.

[1479] Step 3:

[1480] The device sends the extracted text information to the server, using a communication network to generate an API request and send the data to a specified endpoint on the server.

[1481] Step 4:

[1482] The server inputs the received text information into a generative AI model, which analyzes the content of the problem and uses natural language processing (NLP) algorithms to distinguish between the question and the answer.

[1483] Step 5:

[1484] The server determines whether the question is correct, refers to an internal database or knowledge base, evaluates whether the analyzed answer is correct, formats the evaluation result, and generates a data packet to send back to the terminal.

[1485] Step 6:

[1486] The device displays the correctness judgment result received to the user. The result is reflected in the user interface (UI), and a message such as "correct" or "incorrect" is displayed.

[1487] Step 7:

[1488] The server generates background information based on the results of the accuracy assessment, automatically generating relevant historical context and additional explanations using a natural language generation (NLG) model, or searching and extracting relevant information from a database.

[1489] Step 8:

[1490] The server sends the generated background information to the device, which then converts it into JSON format and sends it to the device via the communication network.

[1491] Step 9:

[1492] The device displays the background information received to the user, providing detailed historical background and additional explanations in an easy-to-understand manner to help the user gain a deeper understanding.

[1493] Step 10:

[1494] The device stores the user's learning history in a local database, accumulating information such as the questions studied, answers, correct / incorrect results, and learning dates and times.

[1495] Step 11:

[1496] The server analyzes the user's learning history data, generates supplementary information based on the user's learning patterns and preferences, and recommends appropriate learning materials and video content.

[1497] Step 12:

[1498] The server generates supplementary information and sends it to the terminal. Information tailored to each user's individual learning needs is sent, improving learning efficiency.

[1499] Step 13:

[1500] The device notifies and displays the supplementary information it receives to the user. The supplementary information is displayed in the notification bar or a dedicated screen within the app, making it easy for the user to access.

[1501] Step 14:

[1502] The server aggregates and analyzes the performance data collected from other users nationwide, and uses statistical algorithms to calculate the national average and standard deviation.

[1503] Step 15:

[1504] The server generates information comparing the user's performance with the national average and sends it to the device, including the difference between the individual user's performance and the national average, as well as the deviation score.

[1505] Step 16:

[1506] The terminal displays the received performance comparison results to the user, visualizing the results in the form of graphs and charts to help the user clearly understand their position.

[1507] Step 17:

[1508] The device captures the user's facial expressions and voice tone through a camera and microphone, and analyzes them with an emotion engine, which then recognizes the user's emotional state (e.g., joy, stress, concentration, etc.).

[1509] Step 18:

[1510] The server receives the emotion recognition results sent from the emotion engine and adjusts the learning content and supplementary information based on them. For example, if the user is feeling stressed, it provides relaxing content.

[1511] Step 19:

[1512] The device displays the adjusted learning content and supplementary information to the user, and reflects the adjustment results in the user interface, allowing the user to receive support that is in line with their emotions.

[1513] Through these steps, the present invention is a system that supports users' learning in multiple ways and promotes efficient and deep understanding. Furthermore, by taking into account the user's emotional state, it is possible to provide more personalized learning support.

[1514] Example 2

[1515] 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."

[1516] Conventional learning support systems are limited to determining whether questions submitted by users are correct or incorrect and providing background information, and lack the ability to flexibly respond to the user's learning situation and emotional state. Furthermore, because guidance and supplementary information are not provided efficiently or effectively to individual users, there are problems with not being able to improve learning efficiency or deepen understanding. Furthermore, when comparing scores at a national level, information based on the characteristics of individual users is often not provided, and sufficient support is often not provided. To solve these issues, a system with more advanced and flexible learning support functions is needed.

[1517] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1518] In this invention, the server includes a means for adjusting the learning content according to the user's emotions, a means for recognizing the user's emotions using image recognition and voice analysis, and a means for automatically generating background information based on a true / false judgment, thereby making it possible to provide individually optimized learning support based on the user's emotions and learning history.

[1519] A "user" is an individual who uses the system to carry out learning activities.

[1520] "Means for taking images" refers to the ability to take photos of test questions and study materials using a camera on a smartphone, tablet, etc.

[1521] "Means for extracting text information" refers to the function of extracting text data from images using OCR (optical character recognition) technology.

[1522] "Means for transmitting character information" refers to a function for transmitting extracted text data to a server via the Internet.

[1523] "Means for determining correctness" refers to the function of analyzing the text information sent and determining whether the answer to the question is correct.

[1524] "Means for displaying the judgment result to the user" refers to a function for displaying the result of the correctness judgment on the user's terminal.

[1525] "Means for automatically generating background information" refers to a function that automatically generates related supplementary explanations and detailed information based on a correct / incorrect judgment.

[1526] The "means for displaying background information to the user" refers to a function for displaying the generated background information on the user's terminal.

[1527] "Means for comparing performance with national performance" refers to a function that compares a user's performance data with that of other users nationwide and calculates rankings and standard deviations.

[1528] "Means for recognizing emotions" refers to a function that uses a camera or microphone to analyze and recognize the user's emotional state.

[1529] "Means for adjusting learning content" refers to a function that optimally adjusts the learning content and supplementary information provided based on the user's emotional state and learning history.

[1530] The "learning support system" of this invention provides multiple integrated functions to fully support users' learning activities. This system is realized by combining various functions, mainly image recognition, natural language processing, performance analysis, and information provision. Furthermore, by incorporating an emotion engine, it is possible to provide optimal learning support according to the user's emotional state.

[1531] 1. Photographing and recognizing test questions

[1532] User

[1533] Users take photos of test questions using their smartphone camera.

[1534] Terminal

[1535] The device analyzes the captured image using its OCR (Optical Character Recognition) function and extracts the character information in the image as text data. The specific software used is OCR technology (e.g., Google Cloud Vision or Tesseract). This extracted character information is sent from the device to a server.

[1536] server

[1537] The server inputs the received text information into a generative AI model, analyzes the content of the question, and determines whether it is correct or incorrect. For example, we use "OpenAI GPT-4" as the generative AI model. Examples of prompt sentences include:

[1538] Question: "What year did the Onin War start?"

[1539] Answer: "1467"

[1540] The generative AI model determines whether the prompt is correct or incorrect based on the prompt, and the server then sends the result back to the device.

[1541] Terminal

[1542] The terminal displays the result of the accuracy judgment received from the server to the user. For example, it displays the result in the format "What year did the Onin War start? 1467 -> Correct."

[1543] 2. Providing background information

[1544] server

[1545] Based on the results of the accuracy assessment, the server automatically generates relevant background information from a generative AI model or a pre-built database. Specific software examples include "natural language generation (NLG) models (e.g., OpenAI GPT-4 and Google BERT)."

[1546] Terminal

[1547] The generated background information is sent to the terminal and displayed to the user. For example, "Detailed background information about the Onin War" is displayed.

[1548] 3. Viewing learning history and supplementary information

[1549] Terminal

[1550] The device stores the user's learning history in a local database, including the correct / incorrect results for each question and the date and time of the learning.

[1551] server

[1552] The server analyzes the user's learning history data and generates supplemental information based on the user's learning patterns and preferences, using "machine learning algorithms and data analysis tools (e.g., Python, Pandas, Scikit-learn)."

[1553] Terminal

[1554] The generated supplementary information is sent to the terminal and notified and displayed to the user.

[1555] 4. Comparison of performance at the national level

[1556] server

[1557] The server will compile the academic performance information of students nationwide and generate statistical data. Specific technologies include "SQL databases and analytical tools (e.g., Tableau) used for data compilation and analysis."

[1558] Terminal

[1559] The user's academic performance data is sent to a server, and a comparison result with the national average is sent back to the terminal. The user can check this result and understand where their academic performance ranks nationwide.

[1560] 5. Incorporating an Emotional Engine

[1561] Terminal

[1562] The device uses a camera and microphone to recognize the user's emotions. Specific examples of software include emotion recognition software (e.g., Microsoft Azure Emotion API and IBM Watson Tone Analyzer).

[1563] server

[1564] The server receives the emotion recognition results sent from the emotion engine and adjusts the learning content and supplementary information based on the results. For example, if the user is feeling stressed, it provides relaxing content.

[1565] Terminal

[1566] The adjusted learning content and supplementary information are sent to the terminal and displayed to the user, allowing the user to receive the optimal learning environment and support.

[1567] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1568] Step 1:

[1569] User

[1570] The user takes a photo of the test question using the smartphone camera. The input is an image containing the test question, and the output is the captured image data. Specifically, the user opens the camera app, adjusts the frame so that the entire test paper fits within the frame, and presses the capture button.

[1571] Step 2:

[1572] Terminal

[1573] The device analyzes the captured image using its OCR (Optical Character Recognition) function and extracts the text information in the image as text data. The input is the captured image data, and the output is the extracted text data. Specifically, the device analyzes the image using OCR technology (for example, Google Cloud Vision or Tesseract) and extracts the question and answer in text format.

[1574] Step 3:

[1575] Terminal

[1576] The terminal sends the extracted text data to the server. The input is the extracted text data, and the output is the data sent to the server. Specifically, the terminal uses the HTTPS protocol to send the extracted text data to the specified endpoint of the server as a POST request.

[1577] Step 4:

[1578] server

[1579] The server inputs the received text data into the generative AI model, analyzes the content of the question, and determines whether it is correct or incorrect. The input is the received text data, and the output is the result of the correctness determination. Specifically, the server runs a Python script to generate a prompt for the generative AI model (e.g., GPT-4). An example of a prompt is as follows:

[1580] Question: "What year did the Onin War start?"

[1581] Answer: "1467"

[1582] The generative AI model determines whether the prompt is correct or not based on the prompt sentence and returns the result as "correct."

[1583] Step 5:

[1584] server

[1585] The server returns the result of the accuracy judgment to the terminal. The input is the accuracy judgment result, and the output is the response data to the terminal. Specifically, the server converts the accuracy judgment result into JSON format and returns it to the terminal as a response using the HTTPS protocol.

[1586] Step 6:

[1587] Terminal

[1588] The device displays the correctness judgment result received from the server to the user. The input is the correctness judgment result received from the server, and the output is the information displayed to the user. Specifically, the device analyzes the received JSON data and displays the result "correct" on the screen.

[1589] Step 7:

[1590] server

[1591] Based on the result of the accuracy judgment, the server automatically generates relevant background information from a generative AI model or a pre-built database. The input is the accuracy judgment result, and the output is the generated background information. Specifically, the server again sends a prompt to the generative AI model (e.g., GPT-4) to generate detailed historical background information.

[1592] Step 8:

[1593] server

[1594] The server sends the generated background information to the terminal. The input is the generated background information, and the output is the response data to the terminal. Specifically, the server returns the background information in JSON format to the terminal as a response.

[1595] Step 9:

[1596] Terminal

[1597] The terminal displays the background information received from the server to the user. The input is the background information received from the server, and the output is the information displayed to the user. Specifically, the terminal analyzes the received background information and displays it as "detailed background information about the Onin War."

[1598] Step 10:

[1599] Terminal

[1600] The device stores the user's learning history in a local database. The input is the learning result and the learning date and time, and the output is the data stored in the local database. Specifically, the device uses a local database such as SQLite and executes SQL queries to record the learning history.

[1601] Step 11:

[1602] server

[1603] The server analyzes the user's learning history data and generates related supplemental information. The input is the learning history data, and the output is the generated supplemental information. Specifically, the server analyzes the learning data using a machine learning algorithm and recommends additional learning materials and video content based on the user's interests and preferences.

[1604] Step 12:

[1605] server

[1606] The server sends the generated supplementary information to the terminal. The input is the generated supplementary information, and the output is the response data to the terminal. Specifically, the server returns the supplementary information in JSON format to the terminal as a response.

[1607] Step 13:

[1608] Terminal

[1609] The terminal displays the supplementary information received from the server to the user. The input is the supplementary information received from the server, and the output is the information displayed to the user. Specifically, the terminal displays the supplementary information as a notification, allowing the user to check the details.

[1610] Step 14:

[1611] server

[1612] The server aggregates the performance information of learners nationwide and generates statistical data. The input is the performance data of learners nationwide, and the output is the generated statistical data. Specifically, the server uses an SQL database to execute SQL queries that aggregate each user's performance and generate statistical data.

[1613] Step 15:

[1614] server

[1615] The server compares the user's performance data with the national average and sends the results to the device. The input is the user's performance data and the national average data, and the output is the comparison result. Specifically, the server returns the comparison result to the device as a response in JSON format.

[1616] Step 16:

[1617] Terminal

[1618] The terminal displays the comparison results with the national average to the user. The input is the comparison results received from the server, and the output is the information displayed to the user. Specifically, the terminal analyzes the received comparison results and displays the "comparison results with the national average" on the screen.

[1619] Step 17:

[1620] Terminal

[1621] The device uses its built-in emotion engine to analyze facial expression data and voice tone obtained from the camera and microphone to recognize the user's emotions. The input is facial expression data and voice tone, and the output is the emotion recognition result. Specifically, the device uses emotion recognition software (for example, Microsoft Azure Emotion API or IBM Watson Tone Analyzer) to perform the analysis.

[1622] Step 18:

[1623] Terminal

[1624] The device sends the emotion recognition results to the server. The input is the emotion recognition results, and the output is the data to be sent to the server. Specifically, the device sends the emotion recognition results in JSON format to the server as a POST request.

[1625] Step 19:

[1626] server

[1627] The server receives the emotion recognition results and adjusts the learning content and supplementary information based on them. The input is the emotion recognition results, and the output is the adjusted learning content and supplementary information. Specifically, the server analyzes the emotion data and adjusts the learning content to provide relaxing content if the user is feeling stressed.

[1628] Step 20:

[1629] server

[1630] The server sends the adjusted learning content and supplementary information to the terminal. The input is the adjusted learning content and supplementary information, and the output is the response data to the terminal. Specifically, the server returns the adjusted information in JSON format to the terminal as a response.

[1631] Step 21:

[1632] Terminal

[1633] The device displays the adjusted learning content and supplementary information to the user. The input is the adjusted information received from the server, and the output is the information displayed to the user. Specifically, the device analyzes the received information and displays it in a way that provides the user with the optimal learning environment.

[1634] (Application example 2)

[1635] 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."

[1636] Conventional learning support systems were able to judge whether test questions photographed by users were correct or incorrect and provide related background information, but they did not adjust the learning content based on the user's emotional state, and therefore lacked efficient learning support tailored to the emotions of each individual user. Furthermore, real-time learning support is required for use in physical stores, but a specific system to achieve this was lacking.

[1637] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for a user to take an image and extract text information from the image, means for analyzing the extracted text information based on a generative AI model and determining whether the question is correct, means for displaying the determination result to the user, means for automatically generating background information based on the correctness determination, means for displaying the generated background information to the user, means for comparing the user's performance with national performance, and means for recognizing the user's emotional state using an emotion recognition engine and adjusting the learning content. This enables efficient and personalized learning support according to the individual emotional state of each user, even in physical stores.

[1638] "User" refers to an individual who uses the system to carry out learning activities.

[1639] An "image" is a digital representation of visual information captured by a user.

[1640] "Text information" refers to text data extracted from an image.

[1641] "Generative AI models" refer to algorithms and models that use artificial intelligence technology to generate and analyze information.

[1642] "Correctness determination" refers to determining whether the answer given by the user is correct or incorrect.

[1643] "Background information" refers to additional explanation or information related to a question or problem.

[1644] "National level" refers to a scope that goes beyond a specific region or community and is effective nationwide.

[1645] "Grades" are evaluations and expressions of a user's learning progress and results expressed numerically or in letters.

[1646] An "emotion recognition engine" refers to hardware or software that analyzes and recognizes a user's emotional state.

[1647] "Learning content" refers to the materials, questions, information, and content generally provided for users to study.

[1648] "Adjustment" refers to changing content or settings based on specific conditions or circumstances.

[1649] This invention is a learning support system that integrates image recognition, natural language processing, performance analysis, and information provision functions to support users' learning activities. It is intended for use in brick-and-mortar stores, but it also functions effectively in other environments.

[1650] System Configuration

[1651] The system consists of multiple hardware and software components, including a user terminal, a server, and a learning assistant robot.

[1652] User terminal

[1653] User devices are primarily smartphones and tablets with the following features:

[1654] 1. Image capture: The user takes a picture of the study question using the smartphone camera.

[1655] 2. OCR (Optical Character Recognition): Uses OCR technology (e.g., Tesseract OCR) to extract text information from captured images.

[1656] server

[1657] The server has the following features:

[1658] 1. Character information analysis and accuracy determination: The character information extracted by OCR is analyzed based on a generative AI model (e.g., the GPT-3.5-turbo model) to determine whether the question is correct.

[1659] 2. Generating background information: Based on the results of the correct / incorrect judgment, background information related to the problem is automatically generated using a natural language generation (NLG) model.

[1660] 3. Performance analysis and comparison: Aggregate the performance of all users and compare it at a national level.

[1661] 4. Emotion Recognition: Use an emotion recognition engine to analyze the user's emotional state and adjust the learning content.

[1662] Learning Assistant Robot

[1663] The physical store will be equipped with a learning assistant robot, which will have the following functions:

[1664] 1. User support: Providing real-time learning support in-store and responding to user questions and problems.

[1665] 2. Facial Recognition and Emotion Analysis: Uses a camera and microphone to analyze a user's facial expressions and tone of voice to recognize their emotional state.

[1666] 3. Information provision: The generated background information and performance information are displayed to the user.

[1667] Specific examples

[1668] For example, consider a scenario where a user is in a store and asks a history question to a dedicated learning assistant robot, "What year did the Onin War start?", while simultaneously taking a photo of the question with their smartphone. In this scenario, the process would proceed as follows:

[1669] 1. The user takes a photo of the problem with their smartphone.

[1670] 2. The device uses its OCR function to extract text information from the image, such as "What year did the Onin War start?", and sends it to the server.

[1671] 3. The server uses a generative AI model to determine whether the answer is correct or not, and generates background information on the progress and outcome of the Onin War.

[1672] 4. The device displays the correct / incorrect result and background information to the user.

[1673] 5. The server generates relevant supplementary information based on the user's learning history and sends it to the terminal.

[1674] 6. The terminal notifies and displays the supplementary information to the user.

[1675] 7. The server compares the user's grades with national data, calculates the standard deviation score, etc., and sends it to the terminal.

[1676] 8. The device displays the results of the comparison to the user.

[1677] 9. The device's emotion recognition engine analyzes the user's emotions and sends the emotional state to the server.

[1678] 10. The server adjusts the learning content and supplementary information based on the emotion recognition results and sends them back to the device.

[1679] 11. The device displays the adjusted information to the user.

[1680] Examples of prompts:

[1681] "Provide detailed background information for the following question: In what year did the Onin War begin?"

[1682] As a result, this system enables efficient and personalized learning support based on each user's emotional state, even in physical stores.

[1683] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1684] Step 1:

[1685] The user takes a photo of the problem with their smartphone.

[1686] Input: Image of the training question

[1687] Output: Captured image data

[1688] Specific operation: The user takes a photo of the study question using the smartphone camera. The image data is saved on the device.

[1689] Step 2:

[1690] The device uses the OCR function to extract text information from the image and send it to the server.

[1691] Input: Captured image data

[1692] Output: Extracted text

[1693] Specific operation: Extract text information from the captured image using pytesseract, convert the text information into text format, and then send this text data to the server.

[1694] Step 3:

[1695] The server uses a generative AI model to determine accuracy and generates background information on the progress and results of the Onin War.

[1696] Input: Extracted text

[1697] Output: Correct / incorrect result and background information

[1698] Specific operation: Based on the received text information, the server uses a generative AI model (e.g., the GPT-3.5-turbo model) to analyze whether the question is correct. After determining whether the question is correct, the server uses the model or a pre-built database to generate background information related to the question.

[1699] Step 4:

[1700] The terminal displays the result of the correct / incorrect judgment and background information to the user.

[1701] Input: Correct / incorrect result and background information

[1702] Output: The result of the test and background information displayed on the user's screen

[1703] Specific operation: The result of the correct answer and background information sent from the server are received and displayed on the user's device. For example, "The answer to the question is correct" is displayed along with "the progress and outcome of the Onin War."

[1704] Step 5:

[1705] The server generates related supplementary information based on the user's learning history and sends it to the terminal.

[1706] Input: User learning history data

[1707] Output: Supplementary information

[1708] Specific operation: The server analyzes the user's learning history data and generates supplementary information to deepen understanding of the relevant topic. The generated supplementary information is sent to the terminal.

[1709] Step 6:

[1710] The terminal notifies and displays the supplementary information to the user.

[1711] Input: Additional information

[1712] Output: Supplementary information notified and displayed to the user

[1713] Specific operation: The supplementary information received from the server is notified to the user and displayed on the device screen. Specifically, it provides links such as "Video content related to the Onin War."

[1714] Step 7:

[1715] The server compares the user's grades with national data, calculates the standard deviation score, etc., and sends it to the terminal.

[1716] Input: User performance data, nationwide learner data

[1717] Output: Standard deviation and comparison results

[1718] Specific operation: The server compares the user's performance data with the data of learners nationwide, calculates the deviation score and ranking, and sends the calculation results to the terminal.

[1719] Step 8:

[1720] The terminal displays the results of the comparison of the results to the user.

[1721] Input: Standard deviation or comparison result

[1722] Output: The result of the comparison is displayed on the user's screen.

[1723] Specific operation: The results of the grade comparison sent from the server are notified to the user and displayed on the device screen. Specifically, a message such as "Your grades are above the national average" is displayed.

[1724] Step 9:

[1725] The device's emotion recognition engine analyzes the user's emotions and transmits the emotional state to the server.

[1726] Input: facial expression data, voice data

[1727] Output: Emotion recognition result

[1728] Specific operation: The device uses a built-in camera and microphone to collect and analyze the user's facial expressions and voice tone. The analysis results are sent to the server.

[1729] Step 10:

[1730] The server adjusts the learning content and supplementary information based on the emotion recognition results and sends them back to the device.

[1731] Input: Emotion recognition results

[1732] Output: Adjusted learning results and supplementary information

[1733] Specific operation: The server adjusts the learning content and supplementary information based on the received emotion recognition results. For example, if the user is feeling stressed, it will provide relaxing content. The adjusted information is then sent back to the device.

[1734] Step 11:

[1735] The terminal displays the adjusted information to the user.

[1736] Input: Adjusted learning content and supplementary information

[1737] Output: Adjusted information displayed on the user screen

[1738] Specific operation: The adjusted learning content and supplementary information sent from the server are received and displayed on the user's device. Specifically, a "relaxing video about the Onin War" is displayed.

[1739] 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.

[1740] 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.

[1741] 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.

[1742] [Fourth embodiment]

[1743] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1744] 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.

[1745] 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).

[1746] 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.

[1747] 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.

[1748] 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).

[1749] 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. 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.

[1750] 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.

[1751] 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.

[1752] 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.

[1753] 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.

[1754] 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.

[1755] 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."

[1756] The "learning support system" of this invention provides an integrated set of functions, including image recognition, natural language processing, performance analysis, and information provision, to fully support users' learning activities. This system is implemented as an application that users can easily use on devices such as smartphones and tablets.

[1757] 1. Photographing and recognizing test questions

[1758] Terminal

[1759] First, the user takes a photo of the test question using the smartphone camera. The device analyzes the captured image using its OCR (Optical Character Recognition) function and extracts the characters in the image as text information. This extracted text information is then sent to the server.

[1760] server

[1761] The server inputs the received text information into the generative AI model, analyzes the content of the question, and judges whether it is correct or not. The analyzed result of the judgment is then sent back to the device.

[1762] Terminal

[1763] The device displays the result of the accuracy judgment received from the server to the user. For example, if the user answers "1467" to the history question "What year did the Onin War start?", the result of the accuracy judgment is displayed as "correct answer."

[1764] 2. Providing background information

[1765] server

[1766] Based on the results of the correct / incorrect decision, the server automatically generates relevant background information, specifically by extracting historical background and additional explanatory information related to the problem from natural language generation (NLG) models and pre-built databases.

[1767] Terminal

[1768] The generated background information is sent to the terminal and displayed to the user. For example, detailed background information about the Onin War and the story of its outcome are displayed.

[1769] 3. Viewing learning history and supplementary information

[1770] Terminal

[1771] The user's learning history is stored in a local database, which records the correct / incorrect results for each question, the date and time of study, and so on.

[1772] server

[1773] The server analyzes the user's learning history data and generates supplementary information based on the user's learning patterns and interests. For example, if a user is not good at history, additional learning materials and video content that will help them improve their understanding will be recommended.

[1774] Terminal

[1775] The generated supplemental information is sent to the device and displayed to the user, and is displayed as a notification, allowing the user to further study based on it.

[1776] 4. Comparison of performance at the national level

[1777] server

[1778] The server compiles and analyzes the performance information of learners nationwide to generate statistical data, which makes it possible to compare each user's performance with the national average and deviation score.

[1779] Terminal

[1780] The user's academic performance data is sent to the server, and a comparison with the national average is sent back to the terminal, allowing the user to understand where their academic performance ranks nationally.

[1781] Specific examples

[1782] For example, suppose a user photographs a history test question, "What year did the Onin War start?" and answers "1467." In this case, the process would proceed as follows:

[1783] 1. The user takes a photo of the problem with their smartphone.

[1784] 2. The device uses its OCR function to extract "What year did the Onin War start?" and "1467" from the image and send it to the server.

[1785] 3. The server uses the generated AI model to determine whether the answer is correct, determines it is correct, and automatically generates background information about the progress and outcome of the Onin War.

[1786] 4. The device displays the correct / incorrect result and background information to the user.

[1787] 5. The server generates relevant supplementary information based on the user's learning history and sends it to the terminal.

[1788] 6. The terminal notifies and displays the supplementary information to the user.

[1789] 7. The server compares the user's grades with national data, calculates the standard deviation score, etc., and sends it to the terminal.

[1790] 8. The device displays the results of the comparison to the user.

[1791] In this way, the present invention is a system that supports the user's learning activities in many ways and enables more effective learning.

[1792] The processing flow will be explained below.

[1793] Step 1:

[1794] The user starts the smartphone app and takes a picture of the test paper. When the user presses the capture button, the smartphone's camera function is activated and an image of the test paper is captured.

[1795] Step 2:

[1796] The device analyzes the captured image using its OCR (Optical Character Recognition) function. The OCR engine extracts character data from the image data and converts it into text format.

[1797] Step 3:

[1798] The device sends the extracted text information to the server, using a communication network to generate an API request and send the data to a specified endpoint on the server.

[1799] Step 4:

[1800] The server inputs the received text information into a generative AI model, which analyzes the content of the problem and uses natural language processing (NLP) algorithms to distinguish between the question and the answer.

[1801] Step 5:

[1802] The server determines whether the question is correct, refers to an internal database or knowledge base, evaluates whether the analyzed answer is correct, formats the evaluation result, and generates a data packet to send back to the terminal.

[1803] Step 6:

[1804] The device displays the correctness judgment result received to the user. The result is reflected in the user interface (UI), and a message such as "correct" or "incorrect" is displayed.

[1805] Step 7:

[1806] The server generates background information based on the results of the accuracy assessment, automatically generating relevant historical context and additional explanations using a natural language generation (NLG) model, or searching and extracting relevant information from a database.

[1807] Step 8:

[1808] The server sends the generated background information to the device, which then converts it into JSON format and sends it to the device via the communication network.

[1809] Step 9:

[1810] The device displays the background information received to the user, providing detailed historical background and additional explanations in an easy-to-understand manner to help the user gain a deeper understanding.

[1811] Step 10:

[1812] The device stores the user's learning history in a local database, accumulating information such as the questions studied, answers, correct / incorrect results, and learning dates and times.

[1813] Step 11:

[1814] The server analyzes the user's learning history data, generates supplementary information based on the user's learning patterns and preferences, and recommends appropriate learning materials and video content.

[1815] Step 12:

[1816] The server generates supplementary information and sends it to the terminal. Information tailored to each user's individual learning needs is sent, improving learning efficiency.

[1817] Step 13:

[1818] The device notifies and displays the supplementary information it receives to the user. The supplementary information is displayed in the notification bar or a dedicated screen within the app, making it easy for the user to access.

[1819] Step 14:

[1820] The server aggregates and analyzes the performance data collected from other users nationwide, and uses statistical algorithms to calculate the national average and standard deviation.

[1821] Step 15:

[1822] The server generates information comparing the user's performance with the national average and sends it to the device, including the difference between the individual user's performance and the national average, as well as the deviation score.

[1823] Step 16:

[1824] The terminal displays the received performance comparison results to the user, visualizing the results in the form of graphs and charts to help the user clearly understand their position.

[1825] Through the above steps, the present invention is a system that supports users' learning in many ways and promotes efficient and deep understanding.

[1826] Example 1

[1827] 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."

[1828] Conventional learning support systems have limitations in their ability to accurately extract text information from images taken by users and determine whether questions are correct or incorrect. Furthermore, they lack the functionality to analyze learning history and provide users with appropriate supplementary information, or the ability to compare scores nationwide in real time. This makes it difficult for users to fully grasp their learning progress and level of understanding, making it difficult to study efficiently.

[1829] 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.

[1830] In this invention, the server includes means for a user to take an image and extract text information from the image, means for transmitting the extracted text information, means for analyzing the transmitted text information and judging whether the question is correct or incorrect, means for displaying the judgment result to the user, means for automatically generating background information based on the judgment of correctness, means for displaying the generated background information to the user, means for comparing the user's performance with national performance, means for accumulating and analyzing the user's learning history, means for generating supplementary information based on the analysis result and notifying the user, and means for aggregating and analyzing national performance data and comparing the user's performance with the national average and deviation score. This allows the user to grasp their own learning progress and level of understanding in real time and to study efficiently and effectively.

[1831] A "user" is a person who accesses the learning support system using a device such as a smartphone or tablet and engages in learning activities.

[1832] "Taking an image" refers to the act of a user using the camera on a smartphone or tablet to capture an image of a study question or document.

[1833] "Extracting text information" refers to the process of using OCR (Optical Character Recognition) technology to identify and extract text from a captured image as text data.

[1834] "Send" refers to the act of transferring data from a terminal to a server.

[1835] "Analysis" refers to the process in which the server analyzes the data it receives using technologies such as generative AI models to determine the content of the question and whether it is correct or incorrect.

[1836] "Determining correctness" means using a generative AI model to determine whether an answer to a question is correct or incorrect.

[1837] "Display" refers to the act of visually presenting analysis results, background information, performance information, etc. to the user on the terminal display.

[1838] "Background information" is information containing additional knowledge or explanation related to the content of the problem.

[1839] "Auto-generation" refers to the process of automatically generating the required information using pre-built databases and natural language generation (NLG) models.

[1840] "Study history" refers to a record of a user's past learning activities, answer results, study dates and times, etc.

[1841] "Analysis" refers to clarifying a user's learning patterns and tendencies based on accumulated learning history data.

[1842] "Supplemental information" refers to additional educational materials and video content provided to help users learn more efficiently.

[1843] "Notification" refers to the act of informing the user of generated information, supplementary materials, etc.

[1844] "National level results" refers to statistical data such as averages and standard deviations obtained by compiling and analyzing the results data of other users nationwide.

[1845] The "learning support system" of this invention provides an integrated set of functions, including image recognition, natural language processing, performance analysis, and information provision, to fully support users' learning activities. This system is implemented as an application that users can easily use on devices such as smartphones and tablets.

[1846] Details of the hardware and software you will use

[1847] Terminal

[1848] This system mainly uses mobile devices such as smartphones and tablets. The devices are equipped with a camera, OCR (Optical Character Recognition) functionality, and a display. The OCR functionality incorporates common OCR software and is used to identify characters from images.

[1849] System processing flow

[1850] 1. Photographing test questions and character recognition

[1851] First, the user takes a photo of the test question using the smartphone camera. The device analyzes the captured image using OCR and extracts the characters in the image as text information. This extracted text information is then sent to the server.

[1852] 2. Correct / incorrect judgment

[1853] The server inputs the received text information into the generative AI model, analyzes the content of the question, and judges whether it is correct or not. The analyzed result of the judgment is then sent back to the device.

[1854] Example: If a user answers "1467" to the history question "What year did the Onin War start?", the result of the accuracy test will be displayed as "correct answer."

[1855] 3. Automatic generation of background information

[1856] Based on the results of the correct / incorrect judgment, the server automatically generates relevant background information. Specifically, it extracts historical background and additional explanatory information related to the problem from natural language generation (NLG) models and pre-built databases. The generated background information is sent to the terminal and displayed to the user.

[1857] Example: The user gets the answer right and is shown detailed background information about the Onin War, as well as the story of its outcome.

[1858] 4. Accumulation and analysis of learning history

[1859] The device stores the user's learning history in a local database. The server periodically collects and analyzes the user's learning history data. For example, it can identify learning trends, such as the user's weakness in history questions. Based on the analysis, supplementary information to deepen understanding is generated and sent to the device.

[1860] Example: Additional educational materials and video content are recommended to help users who are weak in history to deepen their understanding.

[1861] 5. Comparison of performance at the national level

[1862] The server compiles and analyzes the academic performance information of learners nationwide. This makes it possible to compare each user's performance with the national average and deviation score. The user's performance data is sent to the server, and the results of the comparison with the national average are sent back to the terminal. This allows the user to understand where their learning performance ranks nationwide.

[1863] Example: Information such as the user's performance being in the top 20% of the national average is displayed on the device.

[1864] Prompt Sentence Examples

[1865] "Extract the text in the image below and determine whether the question is correct or incorrect. Then generate the answer and background information in the same image."

[1866] Through the above processing, the present invention is a system that supports the user's learning activities in many ways and realizes more effective learning.

[1867] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1868] Step 1: Photographing the test questions and recognizing the characters

[1869] Terminal

[1870] The user takes a photo of a test question using their smartphone camera. This image becomes the input. The device uses its OCR (Optical Character Recognition) function to analyze the character information from the captured image and extract text data. The extracted text data becomes the output.

[1871] Specific operation: The user launches the camera app on their smartphone and takes a picture of the test question. The app then performs OCR, generating the text data "What year did the Onin War start?" and "1467," and sends it to the server.

[1872] Step 2: Send text information

[1873] Terminal

[1874] The extracted text information is sent to the server. This text information is the input, and the transmission to the server is the output.

[1875] Specific operation: The device sends the text data "What year did the Onin War start?" and "1467" to the server.

[1876] Step 3: Correct / incorrect

[1877] server

[1878] The text information received by the server is input into the generative AI model. This text information becomes the input. The generative AI model analyzes the content of the question and determines whether it is correct or incorrect. The result of the correct or incorrect determination becomes the output.

[1879] Specific operation: The server inputs the received text data, "What year did the Onin War start?" and "1467," into the generative AI model, which determines that "1467" is the correct answer. The result is then sent back to the device.

[1880] Step 4: Displaying the results

[1881] Terminal

[1882] The terminal receives the result of the correctness judgment from the server and displays it to the user. The result of the correctness judgment is the input, and the display is the output.

[1883] Specific operation: The terminal displays the "correct" result received from the server on the user's screen.

[1884] Step 5: Automatically generate background information

[1885] server

[1886] Based on the accuracy assessment results, the server automatically generates background information. The accuracy assessment results are the input. The server generates related background information using a natural language generation (NLG) model or a pre-built database. This generated background information is the output.

[1887] Specific operation: The server obtains the "correct answer" result and automatically generates an "Outline of the Onin War and its historical background" using the NLG model and sends it to the terminal.

[1888] Step 6: Display background information

[1889] Terminal

[1890] The generated background information is received by the terminal and displayed to the user. The background information is the input and the display is the output.

[1891] Specific operation: The information received by the terminal, "Outline of the Onin War and its historical background," is displayed on the screen.

[1892] Step 7: Accumulating learning history

[1893] Terminal

[1894] The user's learning history is stored in a local database. The input is text information about the learning questions and correct / incorrect results. The output is to save this information in the local database.

[1895] Specific operation: The device stores the user's answers and study date and time in a local database.

[1896] Step 8: Analyze your learning history

[1897] server

[1898] The user's learning history data is collected periodically and analyzed. The accumulated learning history data is the input. The server analyzes this data to understand the user's learning patterns and tendencies. The results of this analysis are the output.

[1899] Specific operation: The server analyzes the user's learning history data and identifies a learning tendency such as "I feel uncomfortable with history questions."

[1900] Step 9: Generate and notify supplementary information

[1901] server

[1902] Supplementary information is generated based on the analysis results and notified to the user. The analysis results are the input. The server generates supplementary information and sends it to the user's terminal. This supplementary information is the output.

[1903] Specific operation: The server generates "educational materials and video content that will deepen the understanding of users who are not good at history" and sends them to the device. The device displays this to the user as a notification.

[1904] Step 10: Comparing performance at the national level

[1905] server

[1906] The performance information of learners nationwide is compiled and compared with the user's performance. National performance data is input. The server analyzes this data and compares the user's performance with the national average and deviation value. The results of this comparison are output.

[1907] Specific operation: The server analyzes the performance data collected from all over the country, calculates information such as whether the user's performance is in the top 20% of the national average, and sends it to the terminal. The terminal then displays this information to the user.

[1908] (Application example 1)

[1909] 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."

[1910] On conventional online shopping sites, users must research product details themselves when selecting a product, making it difficult to find related or recommended products. Furthermore, accurate product suggestions based on users' purchase history are not sufficiently provided. Therefore, there is a need for a system that allows users to efficiently select products and smoothly proceed with their purchasing process.

[1911] 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.

[1912] In this invention, the server includes means for allowing a user to take an image and extract text information from the image, means for transmitting the extracted text information, means for analyzing the transmitted text information and generating product information, means for displaying the generated product information to the user, means for automatically generating related products based on the generated product information, means for displaying the generated related product information to the user, means for generating individual recommended products based on the user's purchase history, means for displaying the generated recommended product information to the user, and means for comparing the generated recommended product information with purchase history data of other users. This enables a user to efficiently obtain detailed product information, related products, and recommended products simply by taking a photo of a product.

[1913] A "user" is a consumer who uses the system to obtain product information and assists in purchasing.

[1914] "Image" means visual data of a product or item photographed using a smartphone or other device.

[1915] "Text information" refers to text information contained in an image and is extracted using the OCR function.

[1916] "Extraction" is the process of extracting the necessary text information from an image.

[1917] "Sending" refers to transferring the extracted character information from the terminal to the server in the form of data.

[1918] "Analysis" refers to understanding product information and related information based on the text information sent and processing it using a generative AI model.

[1919] "Product information" refers to detailed data about a particular product, such as specifications, price, and ratings.

[1920] "Related products" are other products that may be of interest to the user that are suggested based on the analyzed product information.

[1921] "Display" refers to visually showing information sent from the server on the user's terminal screen.

[1922] "Purchase history" is data on products purchased in the past by a user, and is used to generate recommended products.

[1923] "Recommended products" are products that are likely to be purchased and are identified based on the user's purchase history and preferences.

[1924] "Purchase history data of other users" refers to data on the past purchase history of multiple users stored in the system.

[1925] "Comparison" is the process of evaluating and analyzing one piece of data (e.g., a user's purchase history) against other data (e.g., the purchase history of other users).

[1926] The "product proposal support system" for realizing the present invention can be implemented using the following hardware and software.

[1927] Hardware and software used

[1928] Smartphone: A device that takes a picture of a product and uses OCR to extract text information.

[1929] Server: A computer that performs analysis and data aggregation.

[1930] OCR (Optical Character Recognition) software: A tool that extracts textual information from images (e.g., Tesseract OCR).

[1931] Generative AI model: An AI model that analyzes product information and related products and makes suggestions (e.g., GPT-3).

[1932] Natural Language Processing (NLP) models: Models that understand and analyze extracted text information.

[1933] Database: A storage system (e.g., MySQL, PostgreSQL, etc.) that stores product information, user purchase history, and related products.

[1934] System Operation

[1935] 1. The user takes a photo of the product they plan to purchase using their smartphone.

[1936] 2. The device uses OCR software to extract text information from the image.

[1937] 3. The extracted text information is sent to the server.

[1938] 4. The server analyzes the received text information using a generative AI model to generate detailed product information and related product information.

[1939] 5. The server sends related product information and detailed information to the terminal.

[1940] 6. The terminal displays the received information to the user.

[1941] 7. The server retrieves the user's purchase history from the database and automatically generates recommended products using a generative AI model.

[1942] 8. The generated recommended product information is sent to the terminal and displayed to the user.

[1943] 9. The server compares your purchase history data with that of other users and performs a statistical evaluation of your purchasing patterns.

[1944] 10. The evaluation results are sent to the terminal and displayed to the user.

[1945] Specific examples

[1946] For example, if a user takes a picture of a new smartphone, a sample prompt might look like this:

[1947] "Take a photo of a smartphone (new model) and display detailed product information. Please suggest recommended products based on the user's purchase history."

[1948] A user takes a photo of their smartphone, and the OCR function extracts text information such as "new model smartphone." This text information is sent to a server, where a generative AI model generates detailed information about the "new model smartphone" and related products. This information is then sent to the device and displayed to the user. If the user has previously purchased electronic devices, related products and accessories are recommended based on their purchase history, and their purchasing trends are displayed in comparison with the purchasing behavior of other users. In this way, users can efficiently select products and smoothly proceed with their purchasing behavior.

[1949] This allows users to efficiently obtain detailed product information, related products, and recommended products simply by taking a photo of the product. Also, by comparing the purchase data of other users, users can understand their own purchasing trends and make better decisions.

[1950] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1951] Step 1:

[1952] The user takes a photo of the product they plan to purchase using their smartphone.

[1953] Input: Product image

[1954] Output: None

[1955] Specific action: A user takes a photo of a specific product using their smartphone camera.

[1956] Step 2:

[1957] The device uses OCR software to extract text information from the image.

[1958] Input: Product image

[1959] Output: Extracted text information

[1960] Specific operation: Image analysis is performed using on-device OCR (e.g., Tesseract OCR), and character information such as the product name and model number is extracted as text.

[1961] Step 3:

[1962] The extracted character information is sent to the server.

[1963] Input: Extracted text information

[1964] Output: Confirmation of sending text information to the server

[1965] What it does: The extracted text data is sent over a network to a server, often using the HTTP or HTTPS protocol.

[1966] Step 4:

[1967] The server uses a generative AI model based on the text information received to analyze and generate product information.

[1968] Input: Text information sent

[1969] Output: Generated product information

[1970] How it works: The server uses a generative AI model (e.g., GPT-3) to analyze the received text data and generate detailed product information (specifications, price, reviews, etc.). It may also collect information from external databases or APIs as needed.

[1971] Step 5:

[1972] The generated product information is transmitted to the terminal.

[1973] Input: Generated product information

[1974] Output: Confirmation of sending product information to the terminal

[1975] Specific operation: The product information generated by the server is sent to the terminal in a data format such as JSON.

[1976] Step 6:

[1977] The terminal displays the generated product information to the user.

[1978] Input: Generated product information

[1979] Output: Product details page shown to the user

[1980] What it does: Product information is displayed on the device screen, including the product name, price, specifications, reviews, etc.

[1981] Step 7:

[1982] The server retrieves the user's purchase history from the database and automatically generates recommended products using a generative AI model.

[1983] Input: User purchase history

[1984] Output: Generated product recommendations

[1985] Specific operation: The server retrieves purchase history from the database, analyzes the user's preferences and past purchase data using a generative AI model, and generates recommended products based on that.

[1986] Step 8:

[1987] The generated recommended product information is sent to the terminal and displayed to the user.

[1988] Input: Generated recommended product information

[1989] Output: Recommended products displayed to the user

[1990] Specific operation: The server sends recommended product information to the device, and the device displays the recommended products to the user, for example, recommended accessories and related products.

[1991] Step 9:

[1992] The server compares the purchase history data of other users and performs a statistical evaluation of the purchasing patterns.

[1993] Input: Purchase history data of other users

[1994] Output: Statistical evaluation results

[1995] What it does: The server compares your purchasing history data with that of other users and uses statistical analysis and machine learning models to assess how your purchasing patterns are different or similar to others.

[1996] Step 10:

[1997] The evaluation results are sent to the terminal and displayed to the user.

[1998] Input: Statistical evaluation results

[1999] Output: Screen showing the evaluation results

[2000] Specific operation: The server sends the statistical evaluation results to the terminal and displays them on the user's screen, allowing the user to understand how their purchasing behavior differs from that of other users.

[2001] 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.

[2002] The "learning support system" of this invention provides an integrated set of functions, including image recognition, natural language processing, performance analysis, and information provision, to fully support users' learning activities. Furthermore, this system combines an emotion engine that recognizes the user's emotions to provide optimal learning support according to the user's emotional state.

[2003] 1. Photographing and recognizing test questions

[2004] Terminal

[2005] First, the user takes a photo of the test question using the smartphone camera. The device analyzes the captured image using its OCR (Optical Character Recognition) function and extracts the characters in the image as text information. This extracted text information is then sent to the server.

[2006] server

[2007] The server inputs the received text information into the generative AI model, analyzes the content of the question, and judges whether it is correct or not. The analyzed result of the judgment is then sent back to the device.

[2008] Terminal

[2009] The device displays the result of the accuracy judgment received from the server to the user. For example, if the user answers "1467" to the history question "What year did the Onin War start?", the result of the accuracy judgment is displayed as "correct answer."

[2010] 2. Providing background information

[2011] server

[2012] Based on the results of the correct / incorrect decision, the server automatically generates relevant background information, specifically by extracting historical background and additional explanatory information related to the problem from natural language generation (NLG) models and pre-built databases.

[2013] Terminal

[2014] The generated background information is sent to the terminal and displayed to the user. For example, detailed background information about the Onin War and the story of its outcome are displayed.

[2015] 3. Viewing learning history and supplementary information

[2016] Terminal

[2017] The user's learning history is stored in a local database, which records the correct / incorrect results for each question, the date and time of study, and so on.

[2018] server

[2019] The server analyzes the user's learning history data and generates supplementary information based on the user's learning patterns and interests. For example, if a user is not good at history, additional learning materials and video content that will help them improve their understanding will be recommended.

[2020] Terminal

[2021] The generated supplemental information is sent to the device and displayed to the user, and is displayed as a notification, allowing the user to further study based on it.

[2022] 4. Comparison of performance at the national level

[2023] server

[2024] The server compiles and analyzes the performance information of learners nationwide to generate statistical data, which makes it possible to compare each user's performance with the national average and deviation score.

[2025] Terminal

[2026] The user's academic performance data is sent to the server, and a comparison with the national average is sent back to the terminal, allowing the user to understand where their academic performance ranks nationally.

[2027] 5. Incorporating an Emotional Engine

[2028] Terminal

[2029] The device uses a built-in emotion engine to recognize the user's emotions, analyzing facial expression data and voice tone obtained through the camera and microphone to identify the user's emotional state.

[2030] server

[2031] The server receives the emotion recognition results sent from the emotion engine and reflects them in the learning content and the provision of supplementary information. For example, if the user is feeling stressed, the server may provide content that helps them relax.

[2032] Terminal

[2033] The device displays adjusted learning content and supplementary information based on the emotion recognition results to the user, providing appropriate support in line with the user's emotions and improving the effectiveness of learning.

[2034] Specific examples

[2035] For example, suppose a user photographs a history test question, "What year did the Onin War start?" and answers "1467." In this case, the process would proceed as follows:

[2036] 1. The user takes a photo of the problem with their smartphone.

[2037] 2. The device uses its OCR function to extract "What year did the Onin War start?" and "1467" from the image and send it to the server.

[2038] 3. The server uses the generated AI model to determine whether the answer is correct, determines it is correct, and automatically generates background information about the progress and outcome of the Onin War.

[2039] 4. The device displays the correct / incorrect result and background information to the user.

[2040] 5. The server generates relevant supplementary information based on the user's learning history and sends it to the terminal.

[2041] 6. The terminal notifies and displays the supplementary information to the user.

[2042] 7. The server compares the user's grades with national data, calculates the standard deviation score, etc., and sends it to the terminal.

[2043] 8. The device displays the results of the comparison to the user.

[2044] 9. The device's emotion engine analyzes the user's emotions and sends the emotional state to the server.

[2045] 10. The server adjusts the learning content and supplementary information based on the emotion recognition results and sends them back to the device.

[2046] 11. The device displays the adjusted information to the user.

[2047] In this way, the present invention is a system that supports users' learning in multiple ways and promotes efficient and deep understanding. Furthermore, by taking into account the user's emotional state, it is possible to provide more personalized learning support.

[2048] The processing flow will be explained below.

[2049] Step 1:

[2050] The user starts the smartphone app and takes a picture of the test paper. When the user presses the capture button, the smartphone's camera function is activated and an image of the test paper is captured.

[2051] Step 2:

[2052] The device analyzes the captured image using its OCR (Optical Character Recognition) function. The OCR engine extracts character data from the image data and converts it into text format.

[2053] Step 3:

[2054] The device sends the extracted text information to the server, using a communication network to generate an API request and send the data to a specified endpoint on the server.

[2055] Step 4:

[2056] The server inputs the received text information into a generative AI model, which analyzes the content of the problem and uses natural language processing (NLP) algorithms to distinguish between the question and the answer.

[2057] Step 5:

[2058] The server determines whether the question is correct, refers to an internal database or knowledge base, evaluates whether the analyzed answer is correct, formats the evaluation result, and generates a data packet to send back to the terminal.

[2059] Step 6:

[2060] The device displays the correctness judgment result received to the user. The result is reflected in the user interface (UI), and a message such as "correct" or "incorrect" is displayed.

[2061] Step 7:

[2062] The server generates background information based on the results of the accuracy assessment, automatically generating relevant historical context and additional explanations using a natural language generation (NLG) model, or searching and extracting relevant information from a database.

[2063] Step 8:

[2064] The server sends the generated background information to the device, which then converts it into JSON format and sends it to the device via the communication network.

[2065] Step 9:

[2066] The device displays the background information received to the user, providing detailed historical background and additional explanations in an easy-to-understand manner to help the user gain a deeper understanding.

[2067] Step 10:

[2068] The device stores the user's learning history in a local database, accumulating information such as the questions studied, answers, correct / incorrect results, and learning dates and times.

[2069] Step 11:

[2070] The server analyzes the user's learning history data, generates suppl...

Claims

1. The user takes a picture, means for extracting textual information from the image; means for transmitting the extracted character information; A means for analyzing the transmitted character information and determining whether the question is correct or incorrect; means for displaying the determination result to a user; means for automatically generating background information based on the correct / incorrect determination; means for displaying the generated background information to a user; a means for comparing the user's performance with national performance; A learning support system including:

2. 2. The learning support system according to claim 1, wherein the generated background information includes individual supplemental information based on the user's learning history.

3. 2. The learning support system according to claim 1, further comprising means for comparing the results of the user with those of other users nationwide and displaying the results to the user.

Images