system

Abstract

A system is provided.SOLUTION: A system comprising: means for inputting test conditions; means for collecting past test data and question collection; means for automatically generating a test based on the collected AI and the inputted conditions; means for outputting the generated test in a digital format; means for scanning an answer sheet of a student and uploading an image to a server; means for converting the uploaded image of the answer sheet into text data using AI and OCR; and means for automatically scoring the text data.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] Long working hours are a problem in the educational field. In particular, tasks such as creating and grading tests require a great deal of time and effort, placing a heavy burden on teachers. The objective of this invention is to streamline these test-related tasks, reduce the burden on teachers, and improve the quality of tests. [Means for solving the problem]

[0005] The present invention solves the above-mentioned problems by providing a system that includes a means for inputting test conditions, a means for collecting past test data and question bank data, a generation AI means for automatically generating tests based on the collected data and input conditions, a means for outputting the generated tests in digital form, a means for scanning students' answer sheets and uploading the images to a server, a means for converting the answers from the uploaded answer sheet images into text data using AI-OCR, a means for automatically scoring the converted text data, and a means for saving and analyzing each student's answer data and score data.

[0006] "Test conditions" refer to the various elements necessary to create a test, such as subject, grade level, scope, difficulty level, and target average score.

[0007] "Past test data" refers to data such as questions, answers, and scores from previous tests.

[0008] "Problem set data" refers to data relating to questions and answers contained in a problem set used as teaching material.

[0009] "Generative AI" is a program that uses artificial intelligence technology to select, edit, and create test questions.

[0010] "Digital format" means a format that can be read by an electronic device, specifically a PDF or other electronic file format.

[0011] An "answer sheet" is a paper or electronic medium on which a student writes answers to test questions.

[0012] "Scanning" is the process of obtaining a digital image of a paper answer sheet.

[0013] "AI-OCR" is an optical character recognition technology that uses artificial intelligence to convert the character information in an image into text data.

[0014] "Text data" is data that represents character information in digital form.

[0015] "Automatic scoring" is a process in which answers converted into text data are compared with correct answer data and scores are automatically calculated.

[0016] A "database" is a system for systematically storing and managing test data, answer results, score data, etc.

[0017] "Analysis" is the process of evaluating data stored in a database to derive specific trends or patterns.

[0018] "Individually optimized learning support" refers to providing appropriate learning materials and assignments according to each student's academic ability and level of understanding. [Brief explanation of the drawings]

[0019] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9]1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0020] 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.

[0021] First, the terms used in the following description will be explained.

[0022] 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).

[0023] 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.

[0024] 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.

[0025] 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.

[0026] 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."

[0027] [First embodiment]

[0028] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0029] 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.

[0030] 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).

[0031] 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.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0036] 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.

[0037] 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.

[0038] 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.

[0039] 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."

[0040] The present invention is a system that reduces long working hours in educational settings and improves the efficiency and quality of test-related work. Specific embodiments of the present invention will be described in detail below.

[0041] overview

[0042] This system involves a series of processes: teachers input test conditions, the system automatically creates tests using generative AI, and the system converts students' answers into text data using AI-OCR for automatic grading. Furthermore, the system provides optimal learning support for each student based on the data stored in the database.

[0043] Creating Tests

[0044] The user (teacher) inputs test conditions (subject, grade, scope, difficulty, target average score, etc.) from the terminal. The input conditions are sent from the terminal to the server.

[0045] The server collects past test data and question collection data from a database. The generation AI automatically creates tests based on the collected data and input conditions. The generation AI selects and edits questions based on the set target average score, taking into account the appropriate level of difficulty and question scope.

[0046] The server outputs the generated test in PDF or other digital format, and the output is sent to the teacher's device. If necessary, a QR code can also be generated for online distribution.

[0047] Test distribution and answer collection

[0048] The user (teacher) checks the generated test and distributes it to students, who then fill in the answers and submit the answer sheets.

[0049] The user (teacher) scans the students' answer sheets and uploads them to the server via their device. The scanned answer sheets are saved as image files on the server.

[0050] Answer analysis using AI-OCR

[0051] The server uses AI-OCR technology to analyze the uploaded image of the answer sheet. AI-OCR recognizes the text information in the image and converts it into text data. The converted text data is stored in a database.

[0052] Automatic scoring

[0053] The server compares students' answers, converted into text data, with the correct answers and scores them. The automatic scoring algorithm calculates a score, taking into account partial points and multiple answers. The results are sent to the teacher's device and stored in a database.

[0054] Data accumulation and individual optimization

[0055] The server stores each student's answer data and score data in a database. Based on the stored data, it analyzes each student's level of understanding and trends. It identifies the correct answer rate and incorrect answer patterns for specific questions and generates data for individually optimized learning support.

[0056] Based on the analysis results, the server generates customized learning materials and supplementary learning materials for each student and notifies the teacher's device, where the teacher can review and provide the materials to the student.

[0057] Specific examples

[0058] For example, let's say a teacher wants to create a test on linear equations for second-year junior high school mathematics students. The teacher enters the following criteria into their device: "second-year junior high school mathematics, linear equations, medium difficulty, target average score of 70 points." The server analyzes past test data and uses generative AI to select and edit questions that meet the criteria.

[0059] The generated test is output in PDF format and a download link is sent to the teacher's device. The teacher then distributes the test to the students and collects the answer sheets. The collected answer sheets are scanned and uploaded to the server as an image file.

[0060] The server uses AI-OCR to convert the image of the answer sheet into text data, and automatically grades the answers based on this data. The graded results are sent to the teacher's device and stored in a database. Finally, the server analyzes this data, generates customized teaching materials based on each student's level of understanding, and notifies the teacher.

[0061] The above is a specific embodiment of the present invention. This system improves the efficiency of test-related tasks, reduces the burden on teachers, and improves the academic ability of students.

[0062] The processing flow will be explained below.

[0063] Step 1:

[0064] The user enters the test conditions (subject, grade, scope, difficulty level, target average score, etc.) on the terminal.

[0065] Step 2:

[0066] The terminal transmits the input conditions to the server.

[0067] Step 3:

[0068] The server collects past test data and question collection data from a database.

[0069] Step 4:

[0070] The server automatically creates tests by selecting appropriate questions using a generation AI based on the collected data and input conditions. The generation AI adjusts the questions to meet the specified difficulty level and target average score.

[0071] Step 5:

[0072] The server outputs the generated test in PDF or other digital format and notifies the teacher's device with a download link.

[0073] Step 6:

[0074] The user checks the notification on their device, downloads the generated test, and distributes it to students.

[0075] Step 7:

[0076] The user collects the students' answers, scans the answer sheets on the device, and uploads them to the server as image files.

[0077] Step 8:

[0078] The server analyzes the uploaded image of the answer sheet using AI-OCR technology and converts the answers into text data.

[0079] Step 9:

[0080] The server compares the textual answers with the correct answers and calculates the score using an automated scoring algorithm, taking into account partial points and multiple correct answers.

[0081] Step 10:

[0082] The server sends the grading results to the teacher's terminal and stores them in a database.

[0083] Step 11:

[0084] The server analyzes each student's response data and score data and runs algorithms to identify their level of understanding and learning trends.

[0085] Step 12:

[0086] The server generates individually optimized learning support materials based on the analysis results and presents them on the teacher's device.

[0087] Step 13:

[0088] The user checks the learning support materials presented on the terminal and provides them to the students.

[0089] These are the specific processing steps in the system of the present invention. This process streamlines the process from test creation to grading, data accumulation, and individual learning support, and alleviates the problem of long working hours in educational settings.

[0090] Example 1

[0091] 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."

[0092] In educational settings, creating tests, collecting answers, grading, and individually optimizing learning support requires a great deal of time and effort. Providing testing quickly while maintaining test quality and fairness is a major challenge for many educators. Providing efficient and effective learning support based on each student's level of understanding is also a difficult problem. There is a need for a method to efficiently solve these issues, reduce the burden on educational settings, and improve the quality of learning.

[0093] 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.

[0094] In this invention, the server includes means for inputting test conditions, means for collecting past test data and question data, artificial intelligence means for automatically generating tests based on the collected data and input conditions, means for digitally outputting the generated tests, means for digitizing students' answer sheets and uploading the images to a computer, means for converting the answers from the uploaded answer sheet images into text data using optical character recognition technology, means for automatically scoring the converted text data, means for saving and analyzing answer data and score data for each student, means for generating individually optimized learning materials based on the analysis results, and means for notifying students of the generated learning materials. This makes it possible to improve the efficiency and quality of test-related work in educational settings.

[0095] "Means for inputting test conditions" refers to a function that allows educators to input test conditions such as subject, grade, scope, difficulty level, and target average score via a terminal.

[0096] "Means for collecting past test data and question data" is a function for collecting past test data and question set data from a database.

[0097] "Artificial intelligence means" means functionality that includes a generative AI model used to automatically generate tests based on collected data and input criteria.

[0098] "Means for outputting the generated test in digital format" refers to a function for outputting the generated test in PDF format or other digital format and notifying the teacher's terminal.

[0099] "Means for digitizing students' answer sheets and uploading the images to a computer" is a function that allows teachers to scan students' answer sheets with a scanner and upload the image data to a server.

[0100] "Means of converting answers into text data using optical character recognition technology" refers to a function that includes AI-OCR technology, which is used to analyze the image of the uploaded answer sheet and convert the answer content into text data.

[0101] The "means for automatically scoring answers converted into text data" is a function for automatically scoring students' answers converted into text data by comparing them with correct answer data.

[0102] "Means for saving and analyzing response data and score data for each student" refers to a function for saving student response data and score data in a database and analyzing this data.

[0103] The "means for generating individually optimized learning materials" is a function for generating customized learning materials according to each student's level of understanding based on the analyzed data.

[0104] The "means for notifying the generated learning materials" is a function for notifying the teacher's terminal of the generated customized learning materials.

[0105] The present invention is a system that improves the efficiency and quality of test-related tasks in educational settings. This system involves a series of processes: teachers input test conditions, automatically generate tests based on those conditions, analyze and score student answers using AI technology, and provide individually optimized learning materials. Specific embodiments of the present invention are described below.

[0106] Test condition input

[0107] The user (teacher) logs in to the terminal and opens the system's test creation screen. The user enters the following conditions into the form:

[0108] Subject (e.g. Mathematics)

[0109] Grade (e.g., second grade)

[0110] Range (e.g. linear equations)

[0111] Difficulty level (e.g. medium)

[0112] Target average score (e.g. 70 points)

[0113] Once you have completed entering the information, click the "Send" button and the input data will be sent from the terminal to the server in JSON format.

[0114] Automatic test generation

[0115] The server analyzes the received condition data and collects past test data and problem data from a database (e.g., MySQL, PostgreSQL, etc.). Based on the collected data, the server sends a prompt to a generative AI model (e.g., OpenAI GPT-4), inputting the following prompt to the generative AI model:

[0116] "Please create 10 test questions on linear equations for second-year junior high school mathematics students. The difficulty level should be medium, with a target average score of 70 points."

[0117] The generative AI model generates test questions based on this prompt. The generated test questions are converted to PDF format by the server and sent to the teacher's device as a download link. If necessary, a QR code for online distribution can also be generated.

[0118] Collecting and scanning answer sheets

[0119] The user (teacher) distributes the generated test to students and collects their answers. The collected answer sheets are scanned and the image files are uploaded to the server via the device. The device then sends the uploaded image files to the server.

[0120] Analysis of answers using AI-OCR

[0121] The server receives the uploaded image file and uses AI-OCR technology (e.g., Tesseract OCR) to recognize the text information in the image and convert it into text data. For example, if the image contains the answer "x = 2," AI-OCR converts it into the text data "x = 2." The converted text data is stored in a database.

[0122] Automatic scoring

[0123] The server retrieves the student's textual answers from the database and scores them by comparing them with the correct answer. The automated scoring algorithm calculates a score by taking into account partial credit and multiple answers. For example, if the correct answer for a question is "x = 2," a student who answers "x = 1" will receive partial credit (e.g., 0.5 out of 1).

[0124] Data accumulation and individual optimization

[0125] The server stores each student's answer data and score data in a database and analyzes it to analyze each student's level of understanding and learning trends. For example, it uses Python to analyze the data and identify the correct answer rate and incorrect answer patterns for specific questions. Based on these results, the server generates individually optimized learning materials and notifies the teacher's device. This customized material is provided to students as reinforcement for weak areas and additional practice questions.

[0126] Specific examples

[0127] For example, if a user inputs, "Please create 10 test questions on linear equations for second-year junior high school mathematics students. The difficulty level should be medium, and the target average score is 70 points," the server will send prompts to the generative AI model based on this, generating appropriate questions. The generated test is output in PDF format and sent to the teacher's device. The teacher then distributes the test to students and uploads the collected answer sheets to the server. The server then converts the uploaded answer sheets into text data using AI-OCR, and automatically grades them based on this. Finally, the server analyzes each student's data, generates individually optimized learning materials, and sends them to the teacher's device.

[0128] The above is a specific embodiment of the present invention. This system improves the efficiency of test-related tasks, reduces the burden on educational institutions, and improves students' academic abilities.

[0129] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0130] Program processing flow

[0131] Step 1: Enter conditions

[0132] The user (teacher) logs in to the device and opens the system's test creation screen. The user enters test conditions such as subject, grade, scope, difficulty, and target average score into the input form. The entered test conditions are sent from the device to the server in JSON format by clicking the "Send" button. (Input) Subject, grade, scope, difficulty, target average score. (Output) Test condition data in JSON format.

[0133] Step 2: Analyze and collect condition data

[0134] The server parses the received condition data using a JSON parser. Based on the parsed condition data, the server issues a query to a database (e.g., MySQL or PostgreSQL) to collect relevant past test data and problem data. (Input) Test condition data in JSON format. (Output) Set of past test data and problem data.

[0135] Step 3: Generate prompts and create tests

[0136] The server uses the collected data to generate prompts to be applied to a generative AI model (e.g., OpenAI GPT-4). Specifically, it converts the conditions entered by the user into a text prompt. Next, it inputs this prompt into the generative AI model to generate test questions that match the conditions. (Input) Past test data, problem data, and test conditions. (Output) Generated test questions.

[0137] Step 4: Digitally output the test

[0138] The server converts the generated test questions into PDF format or other digital format. The generated test is notified to the teacher's device as a download link. If necessary, a QR code for online distribution is also generated. (Input) Generated test questions. (Output) PDF or digital format test file, download link, QR code.

[0139] Step 5: Distribute the test and scan the answer sheets

[0140] The user (teacher) checks the generated test and distributes it to the students. After the students answer the test, the user scans the answer sheet with a scanner and uploads the image file to the server via their terminal. (Input) Student's answer sheet. (Output) Image file of the scanned answer sheet.

[0141] Step 6: Answer analysis using AI-OCR

[0142] The server receives the uploaded image file of the answer sheet and uses an AI-OCR engine (e.g., Tesseract OCR) to recognize the text information in the image. The resulting text data is converted into text data and stored in a database. (Input) Image file of the answer sheet. (Output) Answers converted into text data.

[0143] Step 7: Automated scoring

[0144] The server retrieves the answers converted to text data from the database and applies an automatic scoring algorithm. The algorithm compares the answers with the correct answer data and scores the answers taking into account partial points and multiple answers. The scoring results are saved back in the database and sent to the teacher's device. (Input) Answers converted to text data, correct answer data. (Output) Scoring results.

[0145] Step 8: Data accumulation and generation of individually optimized teaching materials

[0146] The server stores each student's answer data and score data in a database and performs data analysis based on this. It analyzes the student's level of understanding and learning tendencies and generates individually optimized learning materials based on this. The generated materials are notified to the teacher's device. (Input) Answer data, score data. (Output) Individually optimized learning materials, notification.

[0147] The above is the specific processing flow of the system of the present invention. This system improves the efficiency of test-related work in educational settings, reduces the burden on teachers, and improves the academic ability of students.

[0148] (Application example 1)

[0149] 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."

[0150] Test-related work in traditional educational settings is time-consuming and labor-intensive, resulting in long working hours for teachers. The process of creating and grading tests is particularly labor-intensive, necessitating greater efficiency. Furthermore, improving the efficiency of picking operations is a key issue in logistics centers, and a system that can efficiently perform tasks based on order data is needed. The purpose of this invention is to provide a system that solves these issues in both educational settings and logistics centers, improving the efficiency and quality of operations.

[0151] 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.

[0152] In this invention, the server includes means for inputting test conditions, means for collecting past test data and question book data, means for automatically generating tests based on the collected data and input conditions, means for digitally outputting the generated tests, means for scanning students' answer sheets and uploading the images to the server, means for converting the answers from the uploaded answer sheet images into text data using character recognition, means for automatically scoring the converted text data, means for saving and analyzing each student's answer data and score data, means for automatically generating an optimal picking list based on order data, means for providing work instructions in real time, and means for updating inventory data by reading item barcodes using a camera. This enables the efficiency and quality of test-related work in educational settings to be improved, as well as the efficiency of picking work in logistics centers.

[0153] "Means for inputting test conditions" refers to a device or software that has an interface that allows teachers to input conditions such as the test subject, grade level, scope, difficulty level, and target average score.

[0154] "Means for collecting past test data and question set data" refers to a device or software that has the function of searching and collecting previously used test data and question set data from a database.

[0155] A "generative model means for automatically generating tests based on collected data and input conditions" is a device or system that uses an artificial intelligence model to automatically generate appropriate tests based on collected data and input conditions.

[0156] A "means for digitally outputting the generated test" is a device or software that outputs the generated test in PDF or other digital format and provides it to teachers and students.

[0157] "Means for scanning students' answer sheets and uploading the images to the server" refers to a device or software for scanning answer sheets on which students have written their answers by hand and uploading the image data of the answer sheets to the server.

[0158] "Means for converting answers into text data using character recognition on the image of the uploaded answer sheet" refers to a device or software that has the function of analyzing the image data of the uploaded answer sheet and converting it into text data using character recognition technology.

[0159] The "means for automatically scoring answers converted into text data" refers to a device or software that uses character recognition technology to compare answers converted into text data with correct answer data and automatically score them.

[0160] "Means for storing and analyzing each student's response data and score data" refers to a device or software that stores each student's response data and score data in a database and analyzes the student's level of understanding and learning tendencies based on this data.

[0161] The "means for automatically generating an optimal picking list based on order data" refers to a device or software for automatically generating an efficient picking list based on order data from a logistics center.

[0162] The "means for providing work instructions in real time" refers to a device or software for providing picking work instructions to staff at the logistics center in real time.

[0163] The "means for reading item barcodes using a camera and updating inventory data" refers to a device or software in a logistics center that uses a camera to read item barcodes and automatically update inventory data.

[0164] An embodiment of the present invention will be described.

[0165] overview

[0166] This invention is a system that aims to improve the efficiency of work in educational institutions and logistics centers. In educational institutions, it aims to improve the efficiency and quality of test-related work, and in logistics centers, it aims to improve the efficiency of picking work.

[0167] Educational systems

[0168] The system includes the following means:

[0169] 1. How to enter test conditions:

[0170] A device with an interface that allows teachers to input conditions such as test subject, grade level, scope, difficulty level, and target average score.

[0171] 2. How to collect past test data and question bank data:

[0172] Software with the function of collecting past test data and question collection data from a database.

[0173] 3. Generative modeling means to automatically generate tests based on collected data and input conditions:

[0174] A system that automatically generates tests using a generative AI model based on collected data and input conditions.

[0175] 4. A means to digitally output the generated tests:

[0176] Software that outputs generated tests in PDF or other digital formats.

[0177] 5. How to scan students' answer sheets and upload the images to the server:

[0178] A device that scans answer sheets filled out by students and uploads the image data to a server.

[0179] 6. How to convert the answers into text data using character recognition on the uploaded image of the answer sheet:

[0180] Software that converts uploaded images into text data using AI-OCR technology.

[0181] 7. Methods for automatically scoring answers converted into text data:

[0182] Software equipped with an algorithm that automatically grades answers converted into text data.

[0183] 8. Means for storing and analyzing student response and score data:

[0184] A system that stores each student's response data and score data in a database and uses this data to analyze academic ability and patterns.

[0185] Hardware and software used

[0186] Hardware: scanners, teacher and student devices, servers

[0187] Software: Generative AI model, AI-OCR, database management system, PDF output tool

[0188] Specific examples

[0189] For example, if a teacher were to create a test on linear equations for second-year junior high school math students, they would use the system as follows:

[0190] Enter the conditions: "Junior high school mathematics, second year, linear equations, medium difficulty, target average score 70 points."

[0191] Logistics Center System

[0192] The system includes the following means:

[0193] 1. A method to automatically generate optimal picking lists based on order data:

[0194] A system that automatically generates efficient picking lists using generative AI models based on order data from logistics centers.

[0195] 2. Means of providing real-time work instructions:

[0196] A device that displays picking work instructions in real time to staff wearing smart glasses.

[0197] 3. Using a camera to read the item's barcode and update inventory data:

[0198] The software uses the camera in the smart glasses to read item barcodes and update inventory data in real time.

[0199] Hardware and software used

[0200] Hardware: smart glasses, servers, warehouse staff terminals

[0201] Software: Generative AI model, AI-OCR, database management system

[0202] Specific examples

[0203] For example, a logistics center might use the system as follows:

[0204] Order data: Item A, Location L3, Quantity 10

[0205] Product B, location L1, quantity 5

[0206] Product C, location L2, quantity 8

[0207] Based on this order data, an optimal picking list is generated and real-time instructions are provided to staff.

[0208] The above is a specific embodiment of the present invention. This system can improve the efficiency and quality of test-related work in educational settings and improve picking work in logistics centers.

[0209] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0210] Step 1:

[0211] The user inputs test conditions into the interface, including the subject, grade, scope, difficulty level, and target average score. The input condition data is then sent from the terminal to the server.

[0212] Input: Test subject, grade, scope, difficulty, target average score, and other conditions

[0213] Output: Input condition data

[0214] Step 2:

[0215] The server collects past test data and question collection data from a database, and the collected data is combined with the input condition data and passed to the generative AI model.

[0216] Input: Database

[0217] Output: Past test data and question collection data

[0218] Step 3:

[0219] The server uses a generative AI model to automatically generate tests based on the collected data and input conditions. The generated test questions are selected to be optimal based on the specified conditions.

[0220] Input: Past test data, question collection data, input condition data

[0221] Output: Generated tests

[0222] Step 4:

[0223] The server outputs the generated test in PDF or other digital format, and the digital file of the generated test is sent to the teacher's device and a download link is provided.

[0224] Input: Generated tests

[0225] Output: Test file in PDF or other digital format

[0226] Step 5:

[0227] The user distributes the generated test to students, who then fill in the answers on the answer sheets and submit them to the teacher, who then scans the answer sheets into their devices.

[0228] Input: Handwritten answer sheet

[0229] Output: Scanned image of answer sheet

[0230] Step 6:

[0231] The scanned image of the answer sheet is uploaded from the user's device to the server, where the uploaded image file is stored.

[0232] Input: Scanned image of answer sheet

[0233] Output: Image file saved on the server

[0234] Step 7:

[0235] The server uses AI-OCR technology to convert the uploaded image of the answer sheet into text data. AI-OCR recognizes the text information in the image and converts it into text format.

[0236] Input: Uploaded image of answer sheet

[0237] Output: Answer converted to text data

[0238] Step 8:

[0239] The server automatically scores the answers converted into text data, and the automatic scoring algorithm compares them with the correct answer data and calculates the score for each answer.

[0240] Input: Answer converted to text data

[0241] Output:Scoring results

[0242] Step 9:

[0243] The server stores each student's answer data and score data in a database, which is then used for later analysis.

[0244] Input: Answer data, score data

[0245] Output: Data stored in the database

[0246] Step 10:

[0247] The server analyzes the accumulated data and determines each student's level of understanding and trends, and generates customized teaching materials and supplementary learning materials for each student.

[0248] Input: Answer data and score data stored in the database

[0249] Output: Customized teaching materials, supplementary teaching materials

[0250] Processing steps for logistics center systems

[0251] Step 1:

[0252] The server retrieves the order data and automatically generates the optimal picking list using an AI model. This prompt is also based on the input conditions.

[0253] Input: Order data

[0254] Output: Generated picking list

[0255] Step 2:

[0256] The server analyzes the generated picking list and provides work instructions in real time. Picking work instructions are displayed to staff wearing smart glasses.

[0257] Input: Generated picking list

[0258] Output: Work instructions

[0259] Step 3:

[0260] Following instructions, staff use the smart glasses' camera to scan the barcode of the item, and the data is sent to the server via the terminal.

[0261] Input: Item barcode

[0262] Output: Barcode data sent to the server

[0263] Step 4:

[0264] The server updates the inventory data based on the scanned barcode data, allowing you to grasp the inventory status in real time.

[0265] Input: Barcode data

[0266] Output: Updated inventory data

[0267] 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.

[0268] This invention is designed to reduce long working hours in educational settings and improve the efficiency and quality of test-related work. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the accuracy of individually optimized learning support can be improved.

[0269] overview

[0270] This system involves a series of processes: teachers input test conditions, generative AI automatically creates tests, and AI-OCR converts students' answers into text data for automatic scoring. Furthermore, it has a function that uses an emotion engine to recognize users' emotions and provides appropriate feedback and adjustments based on those emotions.

[0271] Creating Tests

[0272] The user (teacher) inputs test conditions (subject, grade, scope, difficulty, target average score, etc.) from the terminal. The input conditions are sent from the terminal to the server.

[0273] The server collects past test data and question collection data from a database. The generation AI automatically creates tests based on the collected data and input conditions. The generation AI selects and edits questions based on the set target average score, taking into account the appropriate level of difficulty and question scope.

[0274] The server outputs the generated test in PDF or other digital format, and the output is sent to the teacher's device. If necessary, a QR code can also be generated for online distribution.

[0275] Test distribution and answer collection

[0276] The user (teacher) checks the generated test and distributes it to students, who then fill in the answers and submit the answer sheets.

[0277] The user (teacher) scans the students' answer sheets and uploads them to the server via their device. The scanned answer sheets are saved as image files on the server.

[0278] Answer analysis using AI-OCR

[0279] The server uses AI-OCR technology to analyze the uploaded image of the answer sheet. AI-OCR recognizes the text information in the image and converts it into text data. The converted text data is stored in a database.

[0280] Automatic scoring

[0281] The server compares students' answers, converted into text data, with the correct answers and scores them. The automatic scoring algorithm calculates a score, taking into account partial points and multiple answers. The results are sent to the teacher's device and stored in a database.

[0282] Use of emotion engine

[0283] The server uses an emotion engine to recognize the emotions of users (teachers and students) in real time. The emotion data is used for test creation and feedback.

[0284] For example, if the emotion engine detects stress or impatience in a student, the generative AI will adjust the difficulty of the test accordingly to reduce stress. Similarly, if the emotion engine detects fatigue in a teacher, it will automatically provide candidate questions for test creation, reducing the teacher's burden.

[0285] Data accumulation and individual optimization

[0286] The server stores each student's response data, score data, and emotional data in a database. Based on the stored data, it analyzes each student's level of understanding and trends. It identifies the correct answer rate and incorrect answer patterns for specific questions and generates data to provide individually optimized learning support.

[0287] Based on the analysis results and emotion data, the server generates customized learning materials and supplementary learning materials for each student and notifies the teacher's device, where the teacher can review and provide the materials to the students.

[0288] Specific examples

[0289] For example, let's say a teacher wants to create a test on linear equations for second-year junior high school mathematics students. The teacher enters the following criteria into their device: "second-year junior high school mathematics, linear equations, medium difficulty, target average score of 70 points." The server analyzes past test data and uses generative AI to select and edit questions that meet the criteria.

[0290] The generated test is output in PDF format and a download link is sent to the teacher's device. The teacher then distributes the test to the students and collects their answer sheets. The collected answer sheets are scanned and uploaded to the server as an image file.

[0291] The server uses AI-OCR to convert the image of the answer sheet into text data, and automatically grades the answers based on this. The graded results are sent to the teacher's device and stored in a database. If the emotion engine indicates that the student is feeling stressed, the generative AI will automatically adjust the difficulty of the next test.

[0292] Finally, the server analyzes this data, generates customized learning materials based on each student's level of understanding and emotional data, and notifies the teacher, who can then provide guidance and feedback to the student.

[0293] The above is a specific embodiment of the present invention. This system improves the efficiency of test-related tasks, reduces the burden on teachers, and improves students' academic performance. Furthermore, by combining it with an emotion engine, it becomes possible to flexibly respond to the user's emotions.

[0294] The processing flow will be explained below.

[0295] Step 1:

[0296] The user enters the test conditions (subject, grade, scope, difficulty level, target average score, etc.) on the terminal.

[0297] Step 2:

[0298] The terminal transmits the input conditions to the server.

[0299] Step 3:

[0300] The server collects past test data and question collection data from a database.

[0301] Step 4:

[0302] The server automatically creates tests by selecting appropriate questions using a generation AI based on the collected data and input conditions. The generation AI adjusts the question structure based on the set target average score and difficulty level.

[0303] Step 5:

[0304] The server outputs the generated test in PDF or other digital format and sends a download link to the teacher's device.

[0305] Step 6:

[0306] The user checks the notification on their device, downloads the test, and distributes it to students.

[0307] Step 7:

[0308] The user collects the students' test answer sheets, scans them using the terminal, and uploads them to the server as image files.

[0309] Step 8:

[0310] The server analyzes the image of the answer sheet using AI-OCR technology and converts the answers into text data.

[0311] Step 9:

[0312] The server compares the student's textual answers with the correct answers and calculates a score using an automated scoring algorithm that takes into account partial credit and multiple correct answers.

[0313] Step 10:

[0314] The server sends the grading results to the teacher's terminal and stores them in a database.

[0315] Step 11:

[0316] The server uses an emotion engine to analyze emotion data collected from users' (teachers' and students') devices, for example, to detect student stress or a decline in motivation to learn.

[0317] Step 12:

[0318] The server will then adjust the difficulty of the next test based on the emotional data. For example, if a student is feeling highly stressed, the difficulty of the questions will be set lower.

[0319] Step 13:

[0320] The server generates customized feedback based on the emotional data and sends it to the teacher's device, providing specific advice such as, "Student A is under a lot of stress, so please review your teaching method next time."

[0321] Step 14:

[0322] The server stores each student's response data, score data, and emotional data in a database, and uses this data to analyze their level of understanding and learning trends.

[0323] Step 15:

[0324] Based on the analysis results and emotional data, the server generates individually optimized customized teaching materials and supplementary teaching materials and notifies the teacher's device.

[0325] Step 16:

[0326] The user checks the customized learning materials notified on the device and provides them to the students. The next lesson will be taught based on the feedback.

[0327] These are the specific processing steps of the present invention. This series of steps improves the efficiency of test-related tasks, reduces the burden on teachers, and improves students' academic ability. Furthermore, by combining it with an emotion engine, flexible responses based on the user's emotions become possible.

[0328] Example 2

[0329] 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."

[0330] Traditionally, creating and grading tests in the educational field has been extremely time-consuming and labor-intensive, placing a heavy burden on teachers. It has also been difficult to grasp each student's learning progress and emotional state in real time and provide optimal feedback to each individual student. These challenges have limited the improvement of educational quality and efficiency.

[0331] 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.

[0332] In this invention, the server includes a means for inputting test conditions, a means for collecting past test data and question collection data, a generation AI means for automatically generating tests based on the collected data and input conditions, a means for digitally outputting the generated tests, a means for scanning students' answer sheets and uploading the images to the server, a means for converting the answers from the uploaded answer sheet images into text data using AI-OCR, a means for automatically scoring the converted text data, a means for saving and analyzing answer data and score data for each student, an emotion engine means for recognizing user emotions in real time, a means for adjusting feedback and the difficulty of the next test based on the emotion data, and a means for generating teaching materials optimized for each student and notifying the teacher. This allows teachers to improve the efficiency and reduce the burden of test-related work, and further enables them to provide optimal learning support to each student.

[0333] "Means for inputting test conditions" refers to the means by which teachers input information such as subject, grade, scope, difficulty level, and target average score via a terminal.

[0334] The "means for collecting past test data and question set data" refers to the means by which the server collects appropriate past test data and question set data from the database.

[0335] "Generative AI means" means AI and related software for automatically generating tests based on collected data and input test conditions.

[0336] "Means for outputting the generated test in a digital format" refers to a means for exporting the test created by the generative AI in a digital file format such as PDF.

[0337] The "means for scanning students' answer sheets and uploading the images to the server" refers to a means for a teacher to scan students' answer sheets and send the image data to the server.

[0338] "Method of converting answers into text data using AI-OCR" refers to AI technology that recognizes character information from the image data of uploaded answer sheets and converts it into text data.

[0339] The "means for automatically scoring answers converted into text data" refers to a means for comparing the student's answers converted into text data with correct answer data and calculating the score.

[0340] "Means for saving and analyzing each student's response data and score data" refers to a means for saving each student's response content and scoring results in a database and analyzing their learning.

[0341] The "emotion engine means" refers to a technology and system for recognizing a user's emotions in real time.

[0342] The "means for adjusting the feedback and the difficulty of the next test based on emotional data" refers to a means for dynamically adjusting the feedback content and the difficulty of the next test based on the emotional data obtained by the emotion engine.

[0343] The "means for generating teaching materials optimized for each student and notifying the teacher" refers to a means for generating teaching materials customized for each student based on the analyzed data and emotional data and notifying the teacher of the same.

[0344] The present invention is a system that reduces the workload in educational settings, automates the process from test creation to grading and feedback, and improves the accuracy of learning support using an emotion engine. A specific embodiment of this system will be described.

[0345] System Overview

[0346] In this system, the user (teacher) inputs test conditions via a terminal, and the system automatically generates a test based on those conditions. The generated test is distributed to students, and the students' answer sheets are scanned and uploaded to a server. The server uses AI-OCR technology to convert the image of the answer sheet into text data, which is then automatically graded. Furthermore, an emotion engine is used to recognize the user's emotions, and feedback and the next test conditions are adjusted based on that. The system also includes a function to generate optimized teaching materials and notify the teacher.

[0347] Hardware and Software

[0348] Hardware: Teacher and student devices (PCs, tablets, etc.), servers, scanners

[0349] Software: Generative AI models (e.g., GPT-3), AI-OCR technology (e.g., Google Cloud Vision OCR), emotion engine

[0350] Usage and Examples

[0351] Test condition input

[0352] The user (teacher) enters conditions such as "subject," "grade," "scope," "difficulty level," and "target average score" into the input form on the terminal.

[0353] Examples:

[0354] If a teacher wants to "generate a test on linear equations for second-year junior high school mathematics students," they would enter "Subject: Mathematics," "Grade: Second-year junior high school," "Range: Linear equations," "Difficulty: Medium," and "Target average score: 70 points."

[0355] Test Data Collection and Generation

[0356] The server receives input conditions, collects past test data and question collection data from a database, and then sends prompts to the generative AI model to automatically generate tests.

[0357] Example prompt sentence:

[0358] "Generate a test on linear equations for second-year junior high school mathematics students. The conditions are as follows: difficulty level: medium, target average score: 70 points, test area: Chapter 5 of the textbook."

[0359] Test distribution and answer collection

[0360] The user (teacher) checks the generated test and distributes it to the students. After the students answer, the answer sheets are scanned and uploaded to the server via their terminal.

[0361] Answer analysis using AI-OCR

[0362] The server analyzes the image data of the uploaded answer sheet using AI-OCR technology and converts it into text data.

[0363] Automatic grading and feedback

[0364] The server compares the converted answer data with the correct answer data and automatically scores the answers. The results are sent to the teacher's device and stored in a database.

[0365] Use of emotion engine

[0366] The server uses an emotion engine to recognize the emotions of users (teachers and students) in real time. For example, if a student is feeling stressed, the system can adjust the difficulty of the next test, or if the system detects that a teacher is tired, it can automatically provide appropriate candidate questions.

[0367] Individually optimized learning support

[0368] The server analyzes the accumulated answer data and emotion data, and generates teaching materials optimized for each student based on their level of understanding. These generated teaching materials are then sent to the teacher's device.

[0369] This system allows teachers to streamline test-related tasks and reduce their workload, while also enabling them to provide optimal learning support to students. Furthermore, the emotion engine allows for flexible responses that take into account the user's emotions. As a result, this system not only improves the quality of education, but also contributes to the efficiency of classrooms.

[0370] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0371] Program processing steps

[0372] Step 1: Enter the test conditions

[0373] The user (teacher) uses the terminal to input conditions such as subject, grade, scope, difficulty level, and target average score. Specifically, the user fills in the input form on the GUI with "Subject: Mathematics," "Grade: 2nd year junior high school student," "Scope: Linear equations," "Difficulty level: Medium," and "Target average score: 70 points," and then presses the send button. At this time, these conditions are saved in the terminal's memory as input data.

[0374] Input: Subject, grade, range, difficulty level, target average score

[0375] Output: Test condition data sent to the server

[0376] Step 2: Sending test condition data

[0377] The device sends the entered test condition data to the server using an HTTP POST request. The specific data sent is in JSON format.

[0378] Input: Test condition data entered by the user

[0379] Output: Condition data in JSON format passed to the server

[0380] Step 3: Collect test data

[0381] The server receives the test condition data and accesses the database to collect past test data and problem set data. For example, it searches for problems that match the condition "Mathematics, Junior High School 2nd Grade, Linear Equations." It then uses a database query to extract the relevant data.

[0382] Input: Test condition data received by the server

[0383] Output: Past test data and question bank data retrieved from the database

[0384] Step 4: Automatically generate tests

[0385] The server sends prompts to the generative AI model based on the collected data and input conditions. The generative AI model (e.g., GPT-3) generates tests based on the prompts.

[0386] Specific prompt:

[0387] "Generate a test on linear equations for second-year junior high school mathematics students. The conditions are as follows: difficulty level: medium, target average score: 70 points, test area: Chapter 5 of the textbook."

[0388] Input: Past test data from the database, test conditions entered by the user, and prompts for the generative AI model

[0389] Output: Generated test data (questions, answer sheets, etc.)

[0390] Step 5: Generated Test Output

[0391] The server exports the generated test data in PDF or other digital file format, and the generated PDF file is saved in the server's storage.

[0392] Input: Test data generated by the generative AI model

[0393] Output: Test file in PDF format

[0394] Step 6: Notification of generated tests

[0395] The server generates a URL for the generated test file and notifies the user (teacher) of the URL. If necessary, it also generates and sends a QR code for online distribution.

[0396] Input: Generated test file in PDF format

[0397] Output: URL link and QR code

[0398] Step 7: Distribute the test

[0399] The user (teacher) reviews the generated test and distributes it to students by printing it on paper, or by email or via an online platform.

[0400] Input: URL link and QR code

[0401] Output: Tests distributed to students

[0402] Step 8: Collect answer sheets

[0403] The user (teacher) scans the answer sheets on which the students have completed the questions and imports them into the device. An image file of the answer sheet is generated through the scanner.

[0404] Input: Student's completed answer sheet

[0405] Output: Image file of the answer sheet captured on the device

[0406] Step 9: Upload your answer sheet

[0407] The device uploads the scanned image data of the answer sheet to the server using an HTTP POST request.

[0408] Input: Image file of answer sheet

[0409] Output: Image data of the answer sheet sent to the server

[0410] Step 10: Answer analysis using AI-OCR

[0411] The server receives the uploaded image data of the answer sheet and converts the character information into text data using AI-OCR technology. Specifically, the AI-OCR system recognizes the characters in the image and converts them into text format. For example, a mathematical formula such as "12x + 8 = 4y" is obtained as text data.

[0412] Input: Image data of the answer sheet

[0413] Output: Answers converted to text data

[0414] Step 11: Automated scoring

[0415] The server receives the answers converted into text data, compares them with the correct answer data, and automatically scores them. For example, if the answer is "12x + 8 = 4y," the server compares it with the correct answer data and calculates the score taking into account partial points and point allocation.

[0416] Input: Textual answer data, correct answer data

[0417] Output: Scoring results (score data)

[0418] Step 12: Notification of Scoring Results

[0419] The server notifies the user (teacher) of the grading results and simultaneously stores them in a database.

[0420] Input: Scoring results

[0421] Output: Marking results sent to the teacher's device

[0422] Step 13: Emotion Recognition

[0423] The server uses an emotion engine to recognize the emotions of users (teachers and students) in real time. For example, it analyzes the user's emotional state (stress, frustration, fatigue, etc.) from data acquired through a camera or microphone.

[0424] Input: Emotion data obtained from a camera or microphone

[0425] Output: Recognized emotion data

[0426] Step 14: Adjusting Feedback

[0427] Based on the recognized emotion data, the server sends prompts to the generative AI model to provide feedback and adjust the difficulty of the next test. For example, if the emotion engine detects stress, it will instruct the generative AI to lower the difficulty of the next test.

[0428] Input: Recognized emotion data

[0429] Output: Adjusted feedback and difficulty of next test

[0430] Step 15: Individually optimized learning support

[0431] The server analyzes each student's response data, score data, and emotion data to generate optimized teaching materials, which are then sent to the teacher's device and provided to the students.

[0432] Input: Answer data, score data, emotion data

[0433] Output: Optimized learning materials

[0434] These are the specific processing steps of this system. This system aims to reduce the burden on teachers and provide a more effective and efficient educational environment.

[0435] (Application example 2)

[0436] 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."

[0437] Physical stores are seeking training methods to efficiently improve their employees' customer service skills. Conventional training methods rely on subjective evaluations by trainers, making it difficult to provide objective feedback and resulting in inconsistencies in the quality and efficiency of training. Furthermore, there is a lack of means to recognize employees' emotions and stress levels and provide appropriate feedback. Given this background, a system is needed to support employee skill improvement and increase customer satisfaction.

[0438] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0439] In this invention, the server includes: means for inputting test conditions; means for collecting past test data and question collection data; AI generation means for automatically generating tests based on the collected data and input conditions; means for digitally outputting the generated tests; means for scanning images of answer sheets and training data and uploading them to the server; means for converting the uploaded images into text data using AI-OCR; means for automatically scoring the converted text data; an emotion recognition engine for recognizing user emotions in real time; means for providing appropriate feedback to users based on the recognized emotion data; means for saving and analyzing answer data, score data, and emotion data for each user; and means for generating a report that provides feedback on users' strengths and areas for improvement based on the response content and emotion data. This enables objective evaluation of employees' customer service skills and appropriate feedback based on the emotion data.

[0440] Definition of Terms

[0441] "Test conditions" refers to the information that a user enters to generate a specific test or training, specifically settings such as subject, target grade, question scope, difficulty level, and target average score.

[0442] "Past test data and question bank data" refers to data collected from previously administered tests or existing question banks and used for test creation and training.

[0443] "Generative AI means" refers to artificial intelligence technology that automatically generates appropriate test questions or training tasks based on input conditions.

[0444] "Digital output means" means a means for providing the generated test or training assignment in PDF or other digital file format.

[0445] "Means for scanning images of answer sheets and training data and uploading them to a server" refers to a means for capturing answer sheets and training records completed by students or employees as digital images and transmitting them to a server via a network.

[0446] "Means for converting scanned image data into text data using AI-OCR" refers to a means for converting scanned image data into text data using optical character recognition technology.

[0447] An "automatic scoring means" is a means that has an algorithm that calculates scores by comparing the answers converted into text data with the correct answer data.

[0448] An "emotion recognition engine" refers to technology that identifies emotions in real time from a user's facial expressions, voice, text data, etc.

[0449] The "means for providing feedback" is a means for providing appropriate comments and advice to the user based on the recognized emotion data and analysis results.

[0450] The "means for storing and analyzing response data, score data, and emotional data for each user" refers to a means for storing performance data and emotional data for each user and analyzing them.

[0451] The "means for generating a feedback report" is a means for organizing the user's strengths and areas for improvement based on the response content and emotional data, and providing them in the form of a report.

[0452] MODE FOR CARRYING OUT THE INVENTION

[0453] The present invention relates to an emotion recognition training system for helping store employees improve their customer service skills. This system allows employees to receive real-time feedback using a smartphone or smart glasses while undergoing customer service training. Detailed embodiments of the present invention are described below.

[0454] Hardware and software used

[0455] Hardware

[0456] Smartphone

[0457] Smart Glasses

[0458] server

[0459] software

[0460] AI-OCR engine (e.g. Tesseract OCR)

[0461] Emotion recognition engine (e.g. Microsoft Azure Emotion API)

[0462] Generative AI methods (e.g., text generation AI)

[0463] Database Management Systems

[0464] System configuration

[0465] User terminal

[0466] Users (employees) use smartphones or smart glasses to conduct customer service training. At the start of a training session, users input training conditions (e.g., customer service scenario, target skills, training time, etc.). This allows users to focus on the training scenario.

[0467] server

[0468] The server manages the training session and processes the data in the following manner.

[0469] 1. Enter and collect test conditions

[0470] Collect training conditions entered by the user.

[0471] Past training data and question collection data are collected from the database.

[0472] 2. Training and generation using generative AI models

[0473] Training content is automatically generated based on collected data and conditions.

[0474] The generated training is output in digital format and delivered to the user's smart device.

[0475] 3. Scan and upload your answers and training data

[0476] The data collected by the user during training (audio, video, text, etc.) is scanned and uploaded to the server.

[0477] 4. Text data conversion using AI-OCR

[0478] The uploaded data is converted into text data using an AI-OCR engine.

[0479] 5. Use of automatic scoring and emotion recognition engine

[0480] The system automatically scores answers converted into text data and uses an emotion recognition engine to recognize the emotions of users and virtual customers in real time.

[0481] 6. Providing Feedback

[0482] Based on the recognized emotion data, appropriate feedback is provided to the user.

[0483] The report shows users' strengths and areas for improvement, providing useful advice for the next training session.

[0484] Specific examples

[0485] For example, consider the case where employee A receives customer service training on explaining a new product. Employee A wears smart glasses and conducts the training while interacting with a virtual customer. When the training conditions are entered as "new product description, target skill: product knowledge, training time: 10 minutes," the server analyzes past training data and generates a scenario. During the training, employee A's responses and facial expressions are recorded and uploaded to the server. AI-OCR converts the responses into text, and an emotion recognition engine analyzes the emotions of employee A and the virtual customer. After the training is completed, the server generates a feedback report, assessing that "employee A was able to explain the product's features in detail, but lacked confidence in answering customer questions." This allows employee A to know specific areas for improvement for the next training session.

[0486] Example prompts to input to the generative AI model

[0487] Analysis of customer service training logs:

[0488] 1. Convert the response content into text using AI-OCR

[0489] 2. Identify emotional data with an emotion recognition engine

[0490] 3. Feedback on strengths and areas for improvement

[0491] As a result, the present invention can efficiently and objectively support the improvement of customer service skills of employees in physical stores, thereby increasing customer satisfaction.

[0492] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0493] Program processing steps

[0494] Step 1:

[0495] The user starts a training session using a smartphone or smart glasses. They input the training conditions (e.g., customer service scenario, target skills, training time, etc.). This sets specific training objectives, and the information is sent to the server. Input data: Training conditions. Output data: Set training conditions.

[0496] Step 2:

[0497] The server receives the training condition data sent by the user and collects past training data and question set data from the database. The collected data is input into the generative AI model. The server generates an appropriate training scenario. Input data: training conditions, past data. Output data: generated training scenario.

[0498] Step 3:

[0499] The server outputs the generated training scenario in a digital format (e.g. PDF, text file) and delivers it to the user's smart device. The user then performs training according to this training scenario. Input data: The generated training scenario. Output data: The digital file of the training scenario.

[0500] Step 4:

[0501] Users conduct training and record video and audio data and input text during the training process. This data is uploaded to the server from their smart device as answer sheets and training data. Input data: video data, audio data, text data. Output data: training data uploaded to the server.

[0502] Step 5:

[0503] The server receives the uploaded training data and converts it into text data using the AI-OCR engine. The converted text data is saved for analysis. Input data: image data, audio data. Output data: text data.

[0504] Step 6:

[0505] The server automatically scores the text data and uses an emotion recognition engine to recognize the emotions of the user and virtual customers in real time. The recognized emotion data is saved for analysis. Input data: text data, emotion data. Output data: scoring results, emotion data.

[0506] Step 7:

[0507] The server generates a report based on the text data and emotion data to provide appropriate feedback to the user. The report includes the user's strengths and areas for improvement and serves as a guide for the next training session. Input data: text data, emotion data. Output data: feedback report.

[0508] Step 8:

[0509] A feedback report is sent to the user's device, and the user can check the contents. The feedback can be used for the next training session. Input data: Feedback report. Output data: Feedback received by the user.

[0510] Through these steps, users can receive objective evaluations and feedback based on their emotions, allowing them to efficiently improve their customer service skills.

[0511] 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.

[0512] 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.

[0513] 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.

[0514] [Second embodiment]

[0515] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0516] 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.

[0517] 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).

[0518] 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.

[0519] 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.

[0520] 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).

[0521] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0522] 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.

[0523] 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.

[0524] 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.

[0525] 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.

[0526] 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."

[0527] The present invention is a system that reduces long working hours in educational settings and improves the efficiency and quality of test-related work. Specific embodiments of the present invention will be described in detail below.

[0528] overview

[0529] This system involves a series of processes: teachers input test conditions, the system automatically creates tests using generative AI, and the system converts students' answers into text data using AI-OCR for automatic grading. Furthermore, the system provides optimal learning support for each student based on the data stored in the database.

[0530] Creating Tests

[0531] The user (teacher) inputs test conditions (subject, grade, scope, difficulty, target average score, etc.) from the terminal. The input conditions are sent from the terminal to the server.

[0532] The server collects past test data and question collection data from a database. The generation AI automatically creates tests based on the collected data and input conditions. The generation AI selects and edits questions based on the set target average score, taking into account the appropriate level of difficulty and question scope.

[0533] The server outputs the generated test in PDF or other digital format, and the output is sent to the teacher's device. If necessary, a QR code can also be generated for online distribution.

[0534] Test distribution and answer collection

[0535] The user (teacher) checks the generated test and distributes it to students, who then fill in the answers and submit the answer sheets.

[0536] The user (teacher) scans the students' answer sheets and uploads them to the server via their device. The scanned answer sheets are saved as image files on the server.

[0537] Answer analysis using AI-OCR

[0538] The server uses AI-OCR technology to analyze the uploaded image of the answer sheet. AI-OCR recognizes the text information in the image and converts it into text data. The converted text data is stored in a database.

[0539] Automatic scoring

[0540] The server compares students' answers, converted into text data, with the correct answers and scores them. The automatic scoring algorithm calculates a score, taking into account partial points and multiple answers. The results are sent to the teacher's device and stored in a database.

[0541] Data accumulation and individual optimization

[0542] The server stores each student's answer data and score data in a database. Based on the stored data, it analyzes each student's level of understanding and trends. It identifies the correct answer rate and incorrect answer patterns for specific questions and generates data for individually optimized learning support.

[0543] Based on the analysis results, the server generates customized learning materials and supplementary learning materials for each student and notifies the teacher's device, where the teacher can review and provide the materials to the student.

[0544] Specific examples

[0545] For example, let's say a teacher wants to create a test on linear equations for second-year junior high school mathematics students. The teacher enters the following criteria into their device: "second-year junior high school mathematics, linear equations, medium difficulty, target average score of 70 points." The server analyzes past test data and uses generative AI to select and edit questions that meet the criteria.

[0546] The generated test is output in PDF format and a download link is sent to the teacher's device. The teacher then distributes the test to the students and collects the answer sheets. The collected answer sheets are scanned and uploaded to the server as an image file.

[0547] The server uses AI-OCR to convert the image of the answer sheet into text data, and automatically grades the answers based on this data. The graded results are sent to the teacher's device and stored in a database. Finally, the server analyzes this data, generates customized teaching materials based on each student's level of understanding, and notifies the teacher.

[0548] The above is a specific embodiment of the present invention. This system improves the efficiency of test-related tasks, reduces the burden on teachers, and improves the academic ability of students.

[0549] The processing flow will be explained below.

[0550] Step 1:

[0551] The user enters the test conditions (subject, grade, scope, difficulty level, target average score, etc.) on the terminal.

[0552] Step 2:

[0553] The terminal transmits the input conditions to the server.

[0554] Step 3:

[0555] The server collects past test data and question collection data from a database.

[0556] Step 4:

[0557] The server automatically creates tests by selecting appropriate questions using a generation AI based on the collected data and input conditions. The generation AI adjusts the questions to meet the specified difficulty level and target average score.

[0558] Step 5:

[0559] The server outputs the generated test in PDF or other digital format and notifies the teacher's device with a download link.

[0560] Step 6:

[0561] The user checks the notification on their device, downloads the generated test, and distributes it to students.

[0562] Step 7:

[0563] The user collects the students' answers, scans the answer sheets on the device, and uploads them to the server as image files.

[0564] Step 8:

[0565] The server analyzes the uploaded image of the answer sheet using AI-OCR technology and converts the answers into text data.

[0566] Step 9:

[0567] The server compares the textual answers with the correct answers and calculates the score using an automated scoring algorithm, taking into account partial points and multiple correct answers.

[0568] Step 10:

[0569] The server sends the grading results to the teacher's terminal and stores them in a database.

[0570] Step 11:

[0571] The server analyzes each student's response data and score data and runs algorithms to identify their level of understanding and learning trends.

[0572] Step 12:

[0573] The server generates individually optimized learning support materials based on the analysis results and presents them on the teacher's device.

[0574] Step 13:

[0575] The user checks the learning support materials presented on the terminal and provides them to the students.

[0576] These are the specific processing steps in the system of the present invention. This process streamlines the process from test creation to grading, data accumulation, and individual learning support, and alleviates the problem of long working hours in educational settings.

[0577] Example 1

[0578] 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."

[0579] In educational settings, creating tests, collecting answers, grading, and individually optimizing learning support requires a great deal of time and effort. Providing testing quickly while maintaining test quality and fairness is a major challenge for many educators. Providing efficient and effective learning support based on each student's level of understanding is also a difficult problem. There is a need for a method to efficiently solve these issues, reduce the burden on educational settings, and improve the quality of learning.

[0580] 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.

[0581] In this invention, the server includes means for inputting test conditions, means for collecting past test data and question data, artificial intelligence means for automatically generating tests based on the collected data and input conditions, means for digitally outputting the generated tests, means for digitizing students' answer sheets and uploading the images to a computer, means for converting the answers from the uploaded answer sheet images into text data using optical character recognition technology, means for automatically scoring the converted text data, means for saving and analyzing answer data and score data for each student, means for generating individually optimized learning materials based on the analysis results, and means for notifying students of the generated learning materials. This makes it possible to improve the efficiency and quality of test-related work in educational settings.

[0582] "Means for inputting test conditions" refers to a function that allows educators to input test conditions such as subject, grade, scope, difficulty level, and target average score via a terminal.

[0583] "Means for collecting past test data and question data" is a function for collecting past test data and question set data from a database.

[0584] "Artificial intelligence means" means functionality that includes a generative AI model used to automatically generate tests based on collected data and input criteria.

[0585] "Means for outputting the generated test in digital format" refers to a function for outputting the generated test in PDF format or other digital format and notifying the teacher's terminal.

[0586] "Means for digitizing students' answer sheets and uploading the images to a computer" is a function that allows teachers to scan students' answer sheets with a scanner and upload the image data to a server.

[0587] "Means of converting answers into text data using optical character recognition technology" refers to a function that includes AI-OCR technology, which is used to analyze the image of the uploaded answer sheet and convert the answer content into text data.

[0588] The "means for automatically scoring answers converted into text data" is a function for automatically scoring students' answers converted into text data by comparing them with correct answer data.

[0589] "Means for saving and analyzing response data and score data for each student" refers to a function for saving student response data and score data in a database and analyzing this data.

[0590] The "means for generating individually optimized learning materials" is a function for generating customized learning materials according to each student's level of understanding based on the analyzed data.

[0591] The "means for notifying the generated learning materials" is a function for notifying the teacher's terminal of the generated customized learning materials.

[0592] The present invention is a system that improves the efficiency and quality of test-related tasks in educational settings. This system involves a series of processes: teachers input test conditions, automatically generate tests based on those conditions, analyze and score student answers using AI technology, and provide individually optimized learning materials. Specific embodiments of the present invention are described below.

[0593] Test condition input

[0594] The user (teacher) logs in to the terminal and opens the system's test creation screen. The user enters the following conditions into the form:

[0595] Subject (e.g. Mathematics)

[0596] Grade (e.g., second grade)

[0597] Range (e.g. linear equations)

[0598] Difficulty level (e.g. medium)

[0599] Target average score (e.g. 70 points)

[0600] Once you have completed entering the information, click the "Send" button and the input data will be sent from the terminal to the server in JSON format.

[0601] Automatic test generation

[0602] The server analyzes the received condition data and collects past test data and problem data from a database (e.g., MySQL, PostgreSQL, etc.). Based on the collected data, the server sends a prompt to a generative AI model (e.g., OpenAI GPT-4), inputting the following prompt to the generative AI model:

[0603] "Please create 10 test questions on linear equations for second-year junior high school mathematics students. The difficulty level should be medium, with a target average score of 70 points."

[0604] The generative AI model generates test questions based on this prompt. The generated test questions are converted to PDF format by the server and sent to the teacher's device as a download link. If necessary, a QR code for online distribution can also be generated.

[0605] Collecting and scanning answer sheets

[0606] The user (teacher) distributes the generated test to students and collects their answers. The collected answer sheets are scanned and the image files are uploaded to the server via the device. The device then sends the uploaded image files to the server.

[0607] Analysis of answers using AI-OCR

[0608] The server receives the uploaded image file and uses AI-OCR technology (e.g., Tesseract OCR) to recognize the text information in the image and convert it into text data. For example, if the image contains the answer "x = 2," AI-OCR converts it into the text data "x = 2." The converted text data is stored in a database.

[0609] Automatic scoring

[0610] The server retrieves the student's textual answers from the database and scores them by comparing them with the correct answer. The automated scoring algorithm calculates a score by taking into account partial credit and multiple answers. For example, if the correct answer for a question is "x = 2," a student who answers "x = 1" will receive partial credit (e.g., 0.5 out of 1).

[0611] Data accumulation and individual optimization

[0612] The server stores each student's answer data and score data in a database and analyzes it to analyze each student's level of understanding and learning trends. For example, it uses Python to analyze the data and identify the correct answer rate and incorrect answer patterns for specific questions. Based on these results, the server generates individually optimized learning materials and notifies the teacher's device. This customized material is provided to students as reinforcement for weak areas and additional practice questions.

[0613] Specific examples

[0614] For example, if a user inputs, "Please create 10 test questions on linear equations for second-year junior high school mathematics students. The difficulty level should be medium, and the target average score is 70 points," the server will send prompts to the generative AI model based on this, generating appropriate questions. The generated test is output in PDF format and sent to the teacher's device. The teacher then distributes the test to students and uploads the collected answer sheets to the server. The server then converts the uploaded answer sheets into text data using AI-OCR, and automatically grades them based on this. Finally, the server analyzes each student's data, generates individually optimized learning materials, and sends them to the teacher's device.

[0615] The above is a specific embodiment of the present invention. This system improves the efficiency of test-related tasks, reduces the burden on educational institutions, and improves students' academic abilities.

[0616] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0617] Program processing flow

[0618] Step 1: Enter conditions

[0619] The user (teacher) logs in to the device and opens the system's test creation screen. The user enters test conditions such as subject, grade, scope, difficulty, and target average score into the input form. The entered test conditions are sent from the device to the server in JSON format by clicking the "Send" button. (Input) Subject, grade, scope, difficulty, target average score. (Output) Test condition data in JSON format.

[0620] Step 2: Analyze and collect condition data

[0621] The server parses the received condition data using a JSON parser. Based on the parsed condition data, the server issues a query to a database (e.g., MySQL or PostgreSQL) to collect relevant past test data and problem data. (Input) Test condition data in JSON format. (Output) Set of past test data and problem data.

[0622] Step 3: Generate prompts and create tests

[0623] The server uses the collected data to generate prompts to be applied to a generative AI model (e.g., OpenAI GPT-4). Specifically, it converts the conditions entered by the user into a text prompt. Next, it inputs this prompt into the generative AI model to generate test questions that match the conditions. (Input) Past test data, problem data, and test conditions. (Output) Generated test questions.

[0624] Step 4: Digitally output the test

[0625] The server converts the generated test questions into PDF format or other digital format. The generated test is notified to the teacher's device as a download link. If necessary, a QR code for online distribution is also generated. (Input) Generated test questions. (Output) PDF or digital format test file, download link, QR code.

[0626] Step 5: Distribute the test and scan the answer sheets

[0627] The user (teacher) checks the generated test and distributes it to the students. After the students answer the test, the user scans the answer sheet with a scanner and uploads the image file to the server via their terminal. (Input) Student's answer sheet. (Output) Image file of the scanned answer sheet.

[0628] Step 6: Answer analysis using AI-OCR

[0629] The server receives the uploaded image file of the answer sheet and uses an AI-OCR engine (e.g., Tesseract OCR) to recognize the text information in the image. The resulting text data is converted into text data and stored in a database. (Input) Image file of the answer sheet. (Output) Answers converted into text data.

[0630] Step 7: Automated scoring

[0631] The server retrieves the answers converted to text data from the database and applies an automatic scoring algorithm. The algorithm compares the answers with the correct answer data and scores the answers taking into account partial points and multiple answers. The scoring results are saved back in the database and sent to the teacher's device. (Input) Answers converted to text data, correct answer data. (Output) Scoring results.

[0632] Step 8: Data accumulation and generation of individually optimized teaching materials

[0633] The server stores each student's answer data and score data in a database and performs data analysis based on this. It analyzes the student's level of understanding and learning tendencies and generates individually optimized learning materials based on this. The generated materials are notified to the teacher's device. (Input) Answer data, score data. (Output) Individually optimized learning materials, notification.

[0634] The above is the specific processing flow of the system of the present invention. This system improves the efficiency of test-related work in educational settings, reduces the burden on teachers, and improves the academic ability of students.

[0635] (Application example 1)

[0636] 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."

[0637] Test-related work in traditional educational settings is time-consuming and labor-intensive, resulting in long working hours for teachers. The process of creating and grading tests is particularly labor-intensive, necessitating greater efficiency. Furthermore, improving the efficiency of picking operations is a key issue in logistics centers, and a system that can efficiently perform tasks based on order data is needed. The purpose of this invention is to provide a system that solves these issues in both educational settings and logistics centers, improving the efficiency and quality of operations.

[0638] 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.

[0639] In this invention, the server includes means for inputting test conditions, means for collecting past test data and question book data, means for automatically generating tests based on the collected data and input conditions, means for digitally outputting the generated tests, means for scanning students' answer sheets and uploading the images to the server, means for converting the answers from the uploaded answer sheet images into text data using character recognition, means for automatically scoring the converted text data, means for saving and analyzing each student's answer data and score data, means for automatically generating an optimal picking list based on order data, means for providing work instructions in real time, and means for updating inventory data by reading item barcodes using a camera. This enables the efficiency and quality of test-related work in educational settings to be improved, as well as the efficiency of picking work in logistics centers.

[0640] "Means for inputting test conditions" refers to a device or software that has an interface that allows teachers to input conditions such as the test subject, grade level, scope, difficulty level, and target average score.

[0641] "Means for collecting past test data and question set data" refers to a device or software that has the function of searching and collecting previously used test data and question set data from a database.

[0642] A "generative model means for automatically generating tests based on collected data and input conditions" is a device or system that uses an artificial intelligence model to automatically generate appropriate tests based on collected data and input conditions.

[0643] A "means for digitally outputting the generated test" is a device or software that outputs the generated test in PDF or other digital format and provides it to teachers and students.

[0644] "Means for scanning students' answer sheets and uploading the images to the server" refers to a device or software for scanning answer sheets on which students have written their answers by hand and uploading the image data of the answer sheets to the server.

[0645] "Means for converting answers into text data using character recognition on the image of the uploaded answer sheet" refers to a device or software that has the function of analyzing the image data of the uploaded answer sheet and converting it into text data using character recognition technology.

[0646] The "means for automatically scoring answers converted into text data" refers to a device or software that uses character recognition technology to compare answers converted into text data with correct answer data and automatically score them.

[0647] "Means for storing and analyzing each student's response data and score data" refers to a device or software that stores each student's response data and score data in a database and analyzes the student's level of understanding and learning tendencies based on this data.

[0648] The "means for automatically generating an optimal picking list based on order data" refers to a device or software for automatically generating an efficient picking list based on order data from a logistics center.

[0649] The "means for providing work instructions in real time" refers to a device or software for providing picking work instructions to staff at the logistics center in real time.

[0650] The "means for reading item barcodes using a camera and updating inventory data" refers to a device or software in a logistics center that uses a camera to read item barcodes and automatically update inventory data.

[0651] An embodiment of the present invention will be described.

[0652] overview

[0653] This invention is a system that aims to improve the efficiency of work in educational institutions and logistics centers. In educational institutions, it aims to improve the efficiency and quality of test-related work, and in logistics centers, it aims to improve the efficiency of picking work.

[0654] Educational systems

[0655] The system includes the following means:

[0656] 1. How to enter test conditions:

[0657] A device with an interface that allows teachers to input conditions such as test subject, grade level, scope, difficulty level, and target average score.

[0658] 2. How to collect past test data and question bank data:

[0659] Software with the function of collecting past test data and question collection data from a database.

[0660] 3. Generative modeling means to automatically generate tests based on collected data and input conditions:

[0661] A system that automatically generates tests using a generative AI model based on collected data and input conditions.

[0662] 4. A means to digitally output the generated tests:

[0663] Software that outputs generated tests in PDF or other digital formats.

[0664] 5. How to scan students' answer sheets and upload the images to the server:

[0665] A device that scans answer sheets filled out by students and uploads the image data to a server.

[0666] 6. How to convert the answers into text data using character recognition on the uploaded image of the answer sheet:

[0667] Software that converts uploaded images into text data using AI-OCR technology.

[0668] 7. Methods for automatically scoring answers converted into text data:

[0669] Software equipped with an algorithm that automatically grades answers converted into text data.

[0670] 8. Means for storing and analyzing student response and score data:

[0671] A system that stores each student's response data and score data in a database and uses this data to analyze academic ability and patterns.

[0672] Hardware and software used

[0673] Hardware: scanners, teacher and student devices, servers

[0674] Software: Generative AI model, AI-OCR, database management system, PDF output tool

[0675] Specific examples

[0676] For example, if a teacher were to create a test on linear equations for second-year junior high school math students, they would use the system as follows:

[0677] Enter the conditions: "Junior high school mathematics, second year, linear equations, medium difficulty, target average score 70 points."

[0678] Logistics Center System

[0679] The system includes the following means:

[0680] 1. A method to automatically generate optimal picking lists based on order data:

[0681] A system that automatically generates efficient picking lists using generative AI models based on order data from logistics centers.

[0682] 2. Means of providing real-time work instructions:

[0683] A device that displays picking work instructions in real time to staff wearing smart glasses.

[0684] 3. Using a camera to read the item's barcode and update inventory data:

[0685] The software uses the camera in the smart glasses to read item barcodes and update inventory data in real time.

[0686] Hardware and software used

[0687] Hardware: smart glasses, servers, warehouse staff terminals

[0688] Software: Generative AI model, AI-OCR, database management system

[0689] Specific examples

[0690] For example, a logistics center might use the system as follows:

[0691] Order data: Item A, Location L3, Quantity 10

[0692] Product B, location L1, quantity 5

[0693] Product C, location L2, quantity 8

[0694] Based on this order data, an optimal picking list is generated and real-time instructions are provided to staff.

[0695] The above is a specific embodiment of the present invention. This system can improve the efficiency and quality of test-related work in educational settings and improve picking work in logistics centers.

[0696] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0697] Step 1:

[0698] The user inputs test conditions into the interface, including the subject, grade, scope, difficulty level, and target average score. The input condition data is then sent from the terminal to the server.

[0699] Input: Test subject, grade, scope, difficulty, target average score, and other conditions

[0700] Output: Input condition data

[0701] Step 2:

[0702] The server collects past test data and question collection data from a database, and the collected data is combined with the input condition data and passed to the generative AI model.

[0703] Input: Database

[0704] Output: Past test data and question collection data

[0705] Step 3:

[0706] The server uses a generative AI model to automatically generate tests based on the collected data and input conditions. The generated test questions are selected to be optimal based on the specified conditions.

[0707] Input: Past test data, question collection data, input condition data

[0708] Output: Generated tests

[0709] Step 4:

[0710] The server outputs the generated test in PDF or other digital format, and the digital file of the generated test is sent to the teacher's device and a download link is provided.

[0711] Input: Generated tests

[0712] Output: Test file in PDF or other digital format

[0713] Step 5:

[0714] The user distributes the generated test to students, who then fill in the answers on the answer sheets and submit them to the teacher, who then scans the answer sheets into their devices.

[0715] Input: Handwritten answer sheet

[0716] Output: Scanned image of answer sheet

[0717] Step 6:

[0718] The scanned image of the answer sheet is uploaded from the user's device to the server, where the uploaded image file is stored.

[0719] Input: Scanned image of answer sheet

[0720] Output: Image file saved on the server

[0721] Step 7:

[0722] The server uses AI-OCR technology to convert the uploaded image of the answer sheet into text data. AI-OCR recognizes the text information in the image and converts it into text format.

[0723] Input: Uploaded image of answer sheet

[0724] Output: Answer converted to text data

[0725] Step 8:

[0726] The server automatically scores the answers converted into text data, and the automatic scoring algorithm compares them with the correct answer data and calculates the score for each answer.

[0727] Input: Answer converted to text data

[0728] Output:Scoring results

[0729] Step 9:

[0730] The server stores each student's answer data and score data in a database, which is then used for later analysis.

[0731] Input: Answer data, score data

[0732] Output: Data stored in the database

[0733] Step 10:

[0734] The server analyzes the accumulated data and determines each student's level of understanding and trends, and generates customized teaching materials and supplementary learning materials for each student.

[0735] Input: Answer data and score data stored in the database

[0736] Output: Customized teaching materials, supplementary teaching materials

[0737] Processing steps for logistics center systems

[0738] Step 1:

[0739] The server retrieves the order data and automatically generates the optimal picking list using an AI model. This prompt is also based on the input conditions.

[0740] Input: Order data

[0741] Output: Generated picking list

[0742] Step 2:

[0743] The server analyzes the generated picking list and provides work instructions in real time. Picking work instructions are displayed to staff wearing smart glasses.

[0744] Input: Generated picking list

[0745] Output: Work instructions

[0746] Step 3:

[0747] Following instructions, staff use the smart glasses' camera to scan the barcode of the item, and the data is sent to the server via the terminal.

[0748] Input: Item barcode

[0749] Output: Barcode data sent to the server

[0750] Step 4:

[0751] The server updates the inventory data based on the scanned barcode data, allowing you to grasp the inventory status in real time.

[0752] Input: Barcode data

[0753] Output: Updated inventory data

[0754] 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.

[0755] This invention is designed to reduce long working hours in educational settings and improve the efficiency and quality of test-related work. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the accuracy of individually optimized learning support can be improved.

[0756] overview

[0757] This system involves a series of processes: teachers input test conditions, generative AI automatically creates tests, and AI-OCR converts students' answers into text data for automatic scoring. Furthermore, it has a function that uses an emotion engine to recognize users' emotions and provides appropriate feedback and adjustments based on those emotions.

[0758] Creating Tests

[0759] The user (teacher) inputs test conditions (subject, grade, scope, difficulty, target average score, etc.) from the terminal. The input conditions are sent from the terminal to the server.

[0760] The server collects past test data and question collection data from a database. The generation AI automatically creates tests based on the collected data and input conditions. The generation AI selects and edits questions based on the set target average score, taking into account the appropriate level of difficulty and question scope.

[0761] The server outputs the generated test in PDF or other digital format, and the output is sent to the teacher's device. If necessary, a QR code can also be generated for online distribution.

[0762] Test distribution and answer collection

[0763] The user (teacher) checks the generated test and distributes it to students, who then fill in the answers and submit the answer sheets.

[0764] The user (teacher) scans the students' answer sheets and uploads them to the server via their device. The scanned answer sheets are saved as image files on the server.

[0765] Answer analysis using AI-OCR

[0766] The server uses AI-OCR technology to analyze the uploaded image of the answer sheet. AI-OCR recognizes the text information in the image and converts it into text data. The converted text data is stored in a database.

[0767] Automatic scoring

[0768] The server compares students' answers, converted into text data, with the correct answers and scores them. The automatic scoring algorithm calculates a score, taking into account partial points and multiple answers. The results are sent to the teacher's device and stored in a database.

[0769] Use of emotion engine

[0770] The server uses an emotion engine to recognize the emotions of users (teachers and students) in real time. The emotion data is used for test creation and feedback.

[0771] For example, if the emotion engine detects stress or impatience in a student, the generative AI will adjust the difficulty of the test accordingly to reduce stress. Similarly, if the emotion engine detects fatigue in a teacher, it will automatically provide candidate questions for test creation, reducing the teacher's burden.

[0772] Data accumulation and individual optimization

[0773] The server stores each student's response data, score data, and emotional data in a database. Based on the stored data, it analyzes each student's level of understanding and trends. It identifies the correct answer rate and incorrect answer patterns for specific questions and generates data to provide individually optimized learning support.

[0774] Based on the analysis results and emotion data, the server generates customized learning materials and supplementary learning materials for each student and notifies the teacher's device, where the teacher can review and provide the materials to the students.

[0775] Specific examples

[0776] For example, let's say a teacher wants to create a test on linear equations for second-year junior high school mathematics students. The teacher enters the following criteria into their device: "second-year junior high school mathematics, linear equations, medium difficulty, target average score of 70 points." The server analyzes past test data and uses generative AI to select and edit questions that meet the criteria.

[0777] The generated test is output in PDF format and a download link is sent to the teacher's device. The teacher then distributes the test to the students and collects their answer sheets. The collected answer sheets are scanned and uploaded to the server as an image file.

[0778] The server uses AI-OCR to convert the image of the answer sheet into text data, and automatically grades the answers based on this. The graded results are sent to the teacher's device and stored in a database. If the emotion engine indicates that the student is feeling stressed, the generative AI will automatically adjust the difficulty of the next test.

[0779] Finally, the server analyzes this data, generates customized learning materials based on each student's level of understanding and emotional data, and notifies the teacher, who can then provide guidance and feedback to the student.

[0780] The above is a specific embodiment of the present invention. This system improves the efficiency of test-related tasks, reduces the burden on teachers, and improves students' academic performance. Furthermore, by combining it with an emotion engine, it becomes possible to flexibly respond to the user's emotions.

[0781] The processing flow will be explained below.

[0782] Step 1:

[0783] The user enters the test conditions (subject, grade, scope, difficulty level, target average score, etc.) on the terminal.

[0784] Step 2:

[0785] The terminal transmits the input conditions to the server.

[0786] Step 3:

[0787] The server collects past test data and question collection data from a database.

[0788] Step 4:

[0789] The server automatically creates tests by selecting appropriate questions using a generation AI based on the collected data and input conditions. The generation AI adjusts the question structure based on the set target average score and difficulty level.

[0790] Step 5:

[0791] The server outputs the generated test in PDF or other digital format and sends a download link to the teacher's device.

[0792] Step 6:

[0793] The user checks the notification on their device, downloads the test, and distributes it to students.

[0794] Step 7:

[0795] The user collects the students' test answer sheets, scans them using the terminal, and uploads them to the server as image files.

[0796] Step 8:

[0797] The server analyzes the image of the answer sheet using AI-OCR technology and converts the answers into text data.

[0798] Step 9:

[0799] The server compares the student's textual answers with the correct answers and calculates a score using an automated scoring algorithm that takes into account partial credit and multiple correct answers.

[0800] Step 10:

[0801] The server sends the grading results to the teacher's terminal and stores them in a database.

[0802] Step 11:

[0803] The server uses an emotion engine to analyze emotion data collected from users' (teachers' and students') devices, for example, to detect student stress or a decline in motivation to learn.

[0804] Step 12:

[0805] The server will then adjust the difficulty of the next test based on the emotional data. For example, if a student is feeling highly stressed, the difficulty of the questions will be set lower.

[0806] Step 13:

[0807] The server generates customized feedback based on the emotional data and sends it to the teacher's device, providing specific advice such as, "Student A is under a lot of stress, so please review your teaching method next time."

[0808] Step 14:

[0809] The server stores each student's response data, score data, and emotional data in a database, and uses this data to analyze their level of understanding and learning trends.

[0810] Step 15:

[0811] Based on the analysis results and emotional data, the server generates individually optimized customized teaching materials and supplementary teaching materials and notifies the teacher's device.

[0812] Step 16:

[0813] The user checks the customized learning materials notified on the device and provides them to the students. The next lesson will be taught based on the feedback.

[0814] These are the specific processing steps of the present invention. This series of steps improves the efficiency of test-related tasks, reduces the burden on teachers, and improves students' academic ability. Furthermore, by combining it with an emotion engine, flexible responses based on the user's emotions become possible.

[0815] Example 2

[0816] 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."

[0817] Traditionally, creating and grading tests in the educational field has been extremely time-consuming and labor-intensive, placing a heavy burden on teachers. It has also been difficult to grasp each student's learning progress and emotional state in real time and provide optimal feedback to each individual student. These challenges have limited the improvement of educational quality and efficiency.

[0818] 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.

[0819] In this invention, the server includes a means for inputting test conditions, a means for collecting past test data and question collection data, a generation AI means for automatically generating tests based on the collected data and input conditions, a means for digitally outputting the generated tests, a means for scanning students' answer sheets and uploading the images to the server, a means for converting the answers from the uploaded answer sheet images into text data using AI-OCR, a means for automatically scoring the converted text data, a means for saving and analyzing answer data and score data for each student, an emotion engine means for recognizing user emotions in real time, a means for adjusting feedback and the difficulty of the next test based on the emotion data, and a means for generating teaching materials optimized for each student and notifying the teacher. This allows teachers to improve the efficiency and reduce the burden of test-related work, and further enables them to provide optimal learning support to each student.

[0820] "Means for inputting test conditions" refers to the means by which teachers input information such as subject, grade, scope, difficulty level, and target average score via a terminal.

[0821] The "means for collecting past test data and question set data" refers to the means by which the server collects appropriate past test data and question set data from the database.

[0822] "Generative AI means" means AI and related software for automatically generating tests based on collected data and input test conditions.

[0823] "Means for outputting the generated test in a digital format" refers to a means for exporting the test created by the generative AI in a digital file format such as PDF.

[0824] The "means for scanning students' answer sheets and uploading the images to the server" refers to a means for a teacher to scan students' answer sheets and send the image data to the server.

[0825] "Method of converting answers into text data using AI-OCR" refers to AI technology that recognizes character information from the image data of uploaded answer sheets and converts it into text data.

[0826] The "means for automatically scoring answers converted into text data" refers to a means for comparing the student's answers converted into text data with correct answer data and calculating the score.

[0827] "Means for saving and analyzing each student's response data and score data" refers to a means for saving each student's response content and scoring results in a database and analyzing their learning.

[0828] The "emotion engine means" refers to a technology and system for recognizing a user's emotions in real time.

[0829] The "means for adjusting the feedback and the difficulty of the next test based on emotional data" refers to a means for dynamically adjusting the feedback content and the difficulty of the next test based on the emotional data obtained by the emotion engine.

[0830] The "means for generating teaching materials optimized for each student and notifying the teacher" refers to a means for generating teaching materials customized for each student based on the analyzed data and emotional data and notifying the teacher of the same.

[0831] The present invention is a system that reduces the workload in educational settings, automates the process from test creation to grading and feedback, and improves the accuracy of learning support using an emotion engine. A specific embodiment of this system will be described.

[0832] System Overview

[0833] In this system, the user (teacher) inputs test conditions via a terminal, and the system automatically generates a test based on those conditions. The generated test is distributed to students, and the students' answer sheets are scanned and uploaded to a server. The server uses AI-OCR technology to convert the image of the answer sheet into text data, which is then automatically graded. Furthermore, an emotion engine is used to recognize the user's emotions, and feedback and the next test conditions are adjusted based on that. The system also includes a function to generate optimized teaching materials and notify the teacher.

[0834] Hardware and Software

[0835] Hardware: Teacher and student devices (PCs, tablets, etc.), servers, scanners

[0836] Software: Generative AI models (e.g., GPT-3), AI-OCR technology (e.g., Google Cloud Vision OCR), emotion engine

[0837] Usage and Examples

[0838] Test condition input

[0839] The user (teacher) enters conditions such as "subject," "grade," "scope," "difficulty level," and "target average score" into the input form on the terminal.

[0840] Examples:

[0841] If a teacher wants to "generate a test on linear equations for second-year junior high school mathematics students," they would enter "Subject: Mathematics," "Grade: Second-year junior high school," "Range: Linear equations," "Difficulty: Medium," and "Target average score: 70 points."

[0842] Test Data Collection and Generation

[0843] The server receives input conditions, collects past test data and question collection data from a database, and then sends prompts to the generative AI model to automatically generate tests.

[0844] Example prompt sentence:

[0845] "Generate a test on linear equations for second-year junior high school mathematics students. The conditions are as follows: difficulty level: medium, target average score: 70 points, test area: Chapter 5 of the textbook."

[0846] Test distribution and answer collection

[0847] The user (teacher) checks the generated test and distributes it to the students. After the students answer, the answer sheets are scanned and uploaded to the server via their terminal.

[0848] Answer analysis using AI-OCR

[0849] The server analyzes the image data of the uploaded answer sheet using AI-OCR technology and converts it into text data.

[0850] Automatic grading and feedback

[0851] The server compares the converted answer data with the correct answer data and automatically scores the answers. The results are sent to the teacher's device and stored in a database.

[0852] Use of emotion engine

[0853] The server uses an emotion engine to recognize the emotions of users (teachers and students) in real time. For example, if a student is feeling stressed, the system can adjust the difficulty of the next test, or if the system detects that a teacher is tired, it can automatically provide appropriate candidate questions.

[0854] Individually optimized learning support

[0855] The server analyzes the accumulated answer data and emotion data, and generates teaching materials optimized for each student based on their level of understanding. These generated teaching materials are then sent to the teacher's device.

[0856] This system allows teachers to streamline test-related tasks and reduce their workload, while also enabling them to provide optimal learning support to students. Furthermore, the emotion engine allows for flexible responses that take into account the user's emotions. As a result, this system not only improves the quality of education, but also contributes to the efficiency of classrooms.

[0857] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0858] Program processing steps

[0859] Step 1: Enter the test conditions

[0860] The user (teacher) uses the terminal to input conditions such as subject, grade, scope, difficulty level, and target average score. Specifically, the user fills in the input form on the GUI with "Subject: Mathematics," "Grade: 2nd year junior high school student," "Scope: Linear equations," "Difficulty level: Medium," and "Target average score: 70 points," and then presses the send button. At this time, these conditions are saved in the terminal's memory as input data.

[0861] Input: Subject, grade, range, difficulty level, target average score

[0862] Output: Test condition data sent to the server

[0863] Step 2: Sending test condition data

[0864] The device sends the entered test condition data to the server using an HTTP POST request. The specific data sent is in JSON format.

[0865] Input: Test condition data entered by the user

[0866] Output: Condition data in JSON format passed to the server

[0867] Step 3: Collect test data

[0868] The server receives the test condition data and accesses the database to collect past test data and problem set data. For example, it searches for problems that match the condition "Mathematics, Junior High School 2nd Grade, Linear Equations." It then uses a database query to extract the relevant data.

[0869] Input: Test condition data received by the server

[0870] Output: Past test data and question bank data retrieved from the database

[0871] Step 4: Automatically generate tests

[0872] The server sends prompts to the generative AI model based on the collected data and input conditions. The generative AI model (e.g., GPT-3) generates tests based on the prompts.

[0873] Specific prompt:

[0874] "Generate a test on linear equations for second-year junior high school mathematics students. The conditions are as follows: difficulty level: medium, target average score: 70 points, test area: Chapter 5 of the textbook."

[0875] Input: Past test data from the database, test conditions entered by the user, and prompts for the generative AI model

[0876] Output: Generated test data (questions, answer sheets, etc.)

[0877] Step 5: Generated Test Output

[0878] The server exports the generated test data in PDF or other digital file format, and the generated PDF file is saved in the server's storage.

[0879] Input: Test data generated by the generative AI model

[0880] Output: Test file in PDF format

[0881] Step 6: Notification of generated tests

[0882] The server generates a URL for the generated test file and notifies the user (teacher) of the URL. If necessary, it also generates and sends a QR code for online distribution.

[0883] Input: Generated test file in PDF format

[0884] Output: URL link and QR code

[0885] Step 7: Distribute the test

[0886] The user (teacher) reviews the generated test and distributes it to students by printing it on paper, or by email or via an online platform.

[0887] Input: URL link and QR code

[0888] Output: Tests distributed to students

[0889] Step 8: Collect answer sheets

[0890] The user (teacher) scans the answer sheets on which the students have completed the questions and imports them into the device. An image file of the answer sheet is generated through the scanner.

[0891] Input: Student's completed answer sheet

[0892] Output: Image file of the answer sheet captured on the device

[0893] Step 9: Upload your answer sheet

[0894] The device uploads the scanned image data of the answer sheet to the server using an HTTP POST request.

[0895] Input: Image file of answer sheet

[0896] Output: Image data of the answer sheet sent to the server

[0897] Step 10: Answer analysis using AI-OCR

[0898] The server receives the uploaded image data of the answer sheet and converts the character information into text data using AI-OCR technology. Specifically, the AI-OCR system recognizes the characters in the image and converts them into text format. For example, a mathematical formula such as "12x + 8 = 4y" is obtained as text data.

[0899] Input: Image data of the answer sheet

[0900] Output: Answers converted to text data

[0901] Step 11: Automated scoring

[0902] The server receives the answers converted into text data, compares them with the correct answer data, and automatically scores them. For example, if the answer is "12x + 8 = 4y," the server compares it with the correct answer data and calculates the score taking into account partial points and point allocation.

[0903] Input: Textual answer data, correct answer data

[0904] Output: Scoring results (score data)

[0905] Step 12: Notification of Scoring Results

[0906] The server notifies the user (teacher) of the grading results and simultaneously stores them in a database.

[0907] Input: Scoring results

[0908] Output: Marking results sent to the teacher's device

[0909] Step 13: Emotion Recognition

[0910] The server uses an emotion engine to recognize the emotions of users (teachers and students) in real time. For example, it analyzes the user's emotional state (stress, frustration, fatigue, etc.) from data acquired through a camera or microphone.

[0911] Input: Emotion data obtained from a camera or microphone

[0912] Output: Recognized emotion data

[0913] Step 14: Adjusting Feedback

[0914] Based on the recognized emotion data, the server sends prompts to the generative AI model to provide feedback and adjust the difficulty of the next test. For example, if the emotion engine detects stress, it will instruct the generative AI to lower the difficulty of the next test.

[0915] Input: Recognized emotion data

[0916] Output: Adjusted feedback and difficulty of next test

[0917] Step 15: Individually optimized learning support

[0918] The server analyzes each student's response data, score data, and emotion data to generate optimized teaching materials, which are then sent to the teacher's device and provided to the students.

[0919] Input: Answer data, score data, emotion data

[0920] Output: Optimized learning materials

[0921] These are the specific processing steps of this system. This system aims to reduce the burden on teachers and provide a more effective and efficient educational environment.

[0922] (Application example 2)

[0923] 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."

[0924] Physical stores are seeking training methods to efficiently improve their employees' customer service skills. Conventional training methods rely on subjective evaluations by trainers, making it difficult to provide objective feedback and resulting in inconsistencies in the quality and efficiency of training. Furthermore, there is a lack of means to recognize employees' emotions and stress levels and provide appropriate feedback. Given this background, a system is needed to support employee skill improvement and increase customer satisfaction.

[0925] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0926] In this invention, the server includes: means for inputting test conditions; means for collecting past test data and question collection data; AI generation means for automatically generating tests based on the collected data and input conditions; means for digitally outputting the generated tests; means for scanning images of answer sheets and training data and uploading them to the server; means for converting the uploaded images into text data using AI-OCR; means for automatically scoring the converted text data; an emotion recognition engine for recognizing user emotions in real time; means for providing appropriate feedback to users based on the recognized emotion data; means for saving and analyzing answer data, score data, and emotion data for each user; and means for generating a report that provides feedback on users' strengths and areas for improvement based on the response content and emotion data. This enables objective evaluation of employees' customer service skills and appropriate feedback based on the emotion data.

[0927] Definition of Terms

[0928] "Test conditions" refers to the information that a user enters to generate a specific test or training, specifically settings such as subject, target grade, question scope, difficulty level, and target average score.

[0929] "Past test data and question bank data" refers to data collected from previously administered tests or existing question banks and used for test creation and training.

[0930] "Generative AI means" refers to artificial intelligence technology that automatically generates appropriate test questions or training tasks based on input conditions.

[0931] "Digital output means" means a means for providing the generated test or training assignment in PDF or other digital file format.

[0932] "Means for scanning images of answer sheets and training data and uploading them to a server" refers to a means for capturing answer sheets and training records completed by students or employees as digital images and transmitting them to a server via a network.

[0933] "Means for converting scanned image data into text data using AI-OCR" refers to a means for converting scanned image data into text data using optical character recognition technology.

[0934] An "automatic scoring means" is a means that has an algorithm that calculates scores by comparing the answers converted into text data with the correct answer data.

[0935] An "emotion recognition engine" refers to technology that identifies emotions in real time from a user's facial expressions, voice, text data, etc.

[0936] The "means for providing feedback" is a means for providing appropriate comments and advice to the user based on the recognized emotion data and analysis results.

[0937] The "means for storing and analyzing response data, score data, and emotional data for each user" refers to a means for storing performance data and emotional data for each user and analyzing them.

[0938] The "means for generating a feedback report" is a means for organizing the user's strengths and areas for improvement based on the response content and emotional data, and providing them in the form of a report.

[0939] MODE FOR CARRYING OUT THE INVENTION

[0940] The present invention relates to an emotion recognition training system for helping store employees improve their customer service skills. This system allows employees to receive real-time feedback using a smartphone or smart glasses while undergoing customer service training. Detailed embodiments of the present invention are described below.

[0941] Hardware and software used

[0942] Hardware

[0943] Smartphone

[0944] Smart Glasses

[0945] server

[0946] software

[0947] AI-OCR engine (e.g. Tesseract OCR)

[0948] Emotion recognition engine (e.g. Microsoft Azure Emotion API)

[0949] Generative AI methods (e.g., text generation AI)

[0950] Database Management Systems

[0951] System configuration

[0952] User terminal

[0953] Users (employees) use smartphones or smart glasses to conduct customer service training. At the start of a training session, users input training conditions (e.g., customer service scenario, target skills, training time, etc.). This allows users to focus on the training scenario.

[0954] server

[0955] The server manages the training session and processes the data in the following manner.

[0956] 1. Enter and collect test conditions

[0957] Collect training conditions entered by the user.

[0958] Past training data and question collection data are collected from the database.

[0959] 2. Training and generation using generative AI models

[0960] Training content is automatically generated based on collected data and conditions.

[0961] The generated training is output in digital format and delivered to the user's smart device.

[0962] 3. Scan and upload your answers and training data

[0963] The data collected by the user during training (audio, video, text, etc.) is scanned and uploaded to the server.

[0964] 4. Text data conversion using AI-OCR

[0965] The uploaded data is converted into text data using an AI-OCR engine.

[0966] 5. Use of automatic scoring and emotion recognition engine

[0967] The system automatically scores answers converted into text data and uses an emotion recognition engine to recognize the emotions of users and virtual customers in real time.

[0968] 6. Providing Feedback

[0969] Based on the recognized emotion data, appropriate feedback is provided to the user.

[0970] The report shows users' strengths and areas for improvement, providing useful advice for the next training session.

[0971] Specific examples

[0972] For example, consider the case where employee A receives customer service training on explaining a new product. Employee A wears smart glasses and conducts the training while interacting with a virtual customer. When the training conditions are entered as "new product description, target skill: product knowledge, training time: 10 minutes," the server analyzes past training data and generates a scenario. During the training, employee A's responses and facial expressions are recorded and uploaded to the server. AI-OCR converts the responses into text, and an emotion recognition engine analyzes the emotions of employee A and the virtual customer. After the training is completed, the server generates a feedback report, assessing that "employee A was able to explain the product's features in detail, but lacked confidence in answering customer questions." This allows employee A to know specific areas for improvement for the next training session.

[0973] Example prompts to input to the generative AI model

[0974] Analysis of customer service training logs:

[0975] 1. Convert the response content into text using AI-OCR

[0976] 2. Identify emotional data with an emotion recognition engine

[0977] 3. Feedback on strengths and areas for improvement

[0978] As a result, the present invention can efficiently and objectively support the improvement of customer service skills of employees in physical stores, thereby increasing customer satisfaction.

[0979] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0980] Program processing steps

[0981] Step 1:

[0982] The user starts a training session using a smartphone or smart glasses. They input the training conditions (e.g., customer service scenario, target skills, training time, etc.). This sets specific training objectives, and the information is sent to the server. Input data: Training conditions. Output data: Set training conditions.

[0983] Step 2:

[0984] The server receives the training condition data sent by the user and collects past training data and question set data from the database. The collected data is input into the generative AI model. The server generates an appropriate training scenario. Input data: training conditions, past data. Output data: generated training scenario.

[0985] Step 3:

[0986] The server outputs the generated training scenario in a digital format (e.g. PDF, text file) and delivers it to the user's smart device. The user then performs training according to this training scenario. Input data: The generated training scenario. Output data: The digital file of the training scenario.

[0987] Step 4:

[0988] Users conduct training and record video and audio data and input text during the training process. This data is uploaded to the server from their smart device as answer sheets and training data. Input data: video data, audio data, text data. Output data: training data uploaded to the server.

[0989] Step 5:

[0990] The server receives the uploaded training data and converts it into text data using the AI-OCR engine. The converted text data is saved for analysis. Input data: image data, audio data. Output data: text data.

[0991] Step 6:

[0992] The server automatically scores the text data and uses an emotion recognition engine to recognize the emotions of the user and virtual customers in real time. The recognized emotion data is saved for analysis. Input data: text data, emotion data. Output data: scoring results, emotion data.

[0993] Step 7:

[0994] The server generates a report based on the text data and emotion data to provide appropriate feedback to the user. The report includes the user's strengths and areas for improvement and serves as a guide for the next training session. Input data: text data, emotion data. Output data: feedback report.

[0995] Step 8:

[0996] A feedback report is sent to the user's device, and the user can check the contents. The feedback can be used for the next training session. Input data: Feedback report. Output data: Feedback received by the user.

[0997] Through these steps, users can receive objective evaluations and feedback based on their emotions, allowing them to efficiently improve their customer service skills.

[0998] 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.

[0999] 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.

[1000] 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.

[1001] [Third embodiment]

[1002] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1003] 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.

[1004] 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).

[1005] 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.

[1006] 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.

[1007] 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).

[1008] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1009] 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.

[1010] 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.

[1011] 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.

[1012] 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.

[1013] 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."

[1014] The present invention is a system that reduces long working hours in educational settings and improves the efficiency and quality of test-related work. Specific embodiments of the present invention will be described in detail below.

[1015] overview

[1016] This system involves a series of processes: teachers input test conditions, the system automatically creates tests using generative AI, and the system converts students' answers into text data using AI-OCR for automatic grading. Furthermore, the system provides optimal learning support for each student based on the data stored in the database.

[1017] Creating Tests

[1018] The user (teacher) inputs test conditions (subject, grade, scope, difficulty, target average score, etc.) from the terminal. The input conditions are sent from the terminal to the server.

[1019] The server collects past test data and question collection data from a database. The generation AI automatically creates tests based on the collected data and input conditions. The generation AI selects and edits questions based on the set target average score, taking into account the appropriate level of difficulty and question scope.

[1020] The server outputs the generated test in PDF or other digital format, and the output is sent to the teacher's device. If necessary, a QR code can also be generated for online distribution.

[1021] Test distribution and answer collection

[1022] The user (teacher) checks the generated test and distributes it to students, who then fill in the answers and submit the answer sheets.

[1023] The user (teacher) scans the students' answer sheets and uploads them to the server via their device. The scanned answer sheets are saved as image files on the server.

[1024] Answer analysis using AI-OCR

[1025] The server uses AI-OCR technology to analyze the uploaded image of the answer sheet. AI-OCR recognizes the text information in the image and converts it into text data. The converted text data is stored in a database.

[1026] Automatic scoring

[1027] The server compares students' answers, converted into text data, with the correct answers and scores them. The automatic scoring algorithm calculates a score, taking into account partial points and multiple answers. The results are sent to the teacher's device and stored in a database.

[1028] Data accumulation and individual optimization

[1029] The server stores each student's answer data and score data in a database. Based on the stored data, it analyzes each student's level of understanding and trends. It identifies the correct answer rate and incorrect answer patterns for specific questions and generates data for individually optimized learning support.

[1030] Based on the analysis results, the server generates customized learning materials and supplementary learning materials for each student and notifies the teacher's device, where the teacher can review and provide the materials to the student.

[1031] Specific examples

[1032] For example, let's say a teacher wants to create a test on linear equations for second-year junior high school mathematics students. The teacher enters the following criteria into their device: "second-year junior high school mathematics, linear equations, medium difficulty, target average score of 70 points." The server analyzes past test data and uses generative AI to select and edit questions that meet the criteria.

[1033] The generated test is output in PDF format and a download link is sent to the teacher's device. The teacher then distributes the test to the students and collects the answer sheets. The collected answer sheets are scanned and uploaded to the server as an image file.

[1034] The server uses AI-OCR to convert the image of the answer sheet into text data, and automatically grades the answers based on this data. The graded results are sent to the teacher's device and stored in a database. Finally, the server analyzes this data, generates customized teaching materials based on each student's level of understanding, and notifies the teacher.

[1035] The above is a specific embodiment of the present invention. This system improves the efficiency of test-related tasks, reduces the burden on teachers, and improves the academic ability of students.

[1036] The processing flow will be explained below.

[1037] Step 1:

[1038] The user enters the test conditions (subject, grade, scope, difficulty level, target average score, etc.) on the terminal.

[1039] Step 2:

[1040] The terminal transmits the input conditions to the server.

[1041] Step 3:

[1042] The server collects past test data and question collection data from a database.

[1043] Step 4:

[1044] The server automatically creates tests by selecting appropriate questions using a generation AI based on the collected data and input conditions. The generation AI adjusts the questions to meet the specified difficulty level and target average score.

[1045] Step 5:

[1046] The server outputs the generated test in PDF or other digital format and notifies the teacher's device with a download link.

[1047] Step 6:

[1048] The user checks the notification on their device, downloads the generated test, and distributes it to students.

[1049] Step 7:

[1050] The user collects the students' answers, scans the answer sheets on the device, and uploads them to the server as image files.

[1051] Step 8:

[1052] The server analyzes the uploaded image of the answer sheet using AI-OCR technology and converts the answers into text data.

[1053] Step 9:

[1054] The server compares the textual answers with the correct answers and calculates the score using an automated scoring algorithm, taking into account partial points and multiple correct answers.

[1055] Step 10:

[1056] The server sends the grading results to the teacher's terminal and stores them in a database.

[1057] Step 11:

[1058] The server analyzes each student's response data and score data and runs algorithms to identify their level of understanding and learning trends.

[1059] Step 12:

[1060] The server generates individually optimized learning support materials based on the analysis results and presents them on the teacher's device.

[1061] Step 13:

[1062] The user checks the learning support materials presented on the terminal and provides them to the students.

[1063] These are the specific processing steps in the system of the present invention. This process streamlines the process from test creation to grading, data accumulation, and individual learning support, and alleviates the problem of long working hours in educational settings.

[1064] Example 1

[1065] 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."

[1066] In educational settings, creating tests, collecting answers, grading, and individually optimizing learning support requires a great deal of time and effort. Providing testing quickly while maintaining test quality and fairness is a major challenge for many educators. Providing efficient and effective learning support based on each student's level of understanding is also a difficult problem. There is a need for a method to efficiently solve these issues, reduce the burden on educational settings, and improve the quality of learning.

[1067] 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.

[1068] In this invention, the server includes means for inputting test conditions, means for collecting past test data and question data, artificial intelligence means for automatically generating tests based on the collected data and input conditions, means for digitally outputting the generated tests, means for digitizing students' answer sheets and uploading the images to a computer, means for converting the answers from the uploaded answer sheet images into text data using optical character recognition technology, means for automatically scoring the converted text data, means for saving and analyzing answer data and score data for each student, means for generating individually optimized learning materials based on the analysis results, and means for notifying students of the generated learning materials. This makes it possible to improve the efficiency and quality of test-related work in educational settings.

[1069] "Means for inputting test conditions" refers to a function that allows educators to input test conditions such as subject, grade, scope, difficulty level, and target average score via a terminal.

[1070] "Means for collecting past test data and question data" is a function for collecting past test data and question set data from a database.

[1071] "Artificial intelligence means" means functionality that includes a generative AI model used to automatically generate tests based on collected data and input criteria.

[1072] "Means for outputting the generated test in digital format" refers to a function for outputting the generated test in PDF format or other digital format and notifying the teacher's terminal.

[1073] "Means for digitizing students' answer sheets and uploading the images to a computer" is a function that allows teachers to scan students' answer sheets with a scanner and upload the image data to a server.

[1074] "Means of converting answers into text data using optical character recognition technology" refers to a function that includes AI-OCR technology, which is used to analyze the image of the uploaded answer sheet and convert the answer content into text data.

[1075] The "means for automatically scoring answers converted into text data" is a function for automatically scoring students' answers converted into text data by comparing them with correct answer data.

[1076] "Means for saving and analyzing response data and score data for each student" refers to a function for saving student response data and score data in a database and analyzing this data.

[1077] The "means for generating individually optimized learning materials" is a function for generating customized learning materials according to each student's level of understanding based on the analyzed data.

[1078] The "means for notifying the generated learning materials" is a function for notifying the teacher's terminal of the generated customized learning materials.

[1079] The present invention is a system that improves the efficiency and quality of test-related tasks in educational settings. This system involves a series of processes: teachers input test conditions, automatically generate tests based on those conditions, analyze and score student answers using AI technology, and provide individually optimized learning materials. Specific embodiments of the present invention are described below.

[1080] Test condition input

[1081] The user (teacher) logs in to the terminal and opens the system's test creation screen. The user enters the following conditions into the form:

[1082] Subject (e.g. Mathematics)

[1083] Grade (e.g., second grade)

[1084] Range (e.g. linear equations)

[1085] Difficulty level (e.g. medium)

[1086] Target average score (e.g. 70 points)

[1087] Once you have completed entering the information, click the "Send" button and the input data will be sent from the terminal to the server in JSON format.

[1088] Automatic test generation

[1089] The server analyzes the received condition data and collects past test data and problem data from a database (e.g., MySQL, PostgreSQL, etc.). Based on the collected data, the server sends a prompt to a generative AI model (e.g., OpenAI GPT-4), inputting the following prompt to the generative AI model:

[1090] "Please create 10 test questions on linear equations for second-year junior high school mathematics students. The difficulty level should be medium, with a target average score of 70 points."

[1091] The generative AI model generates test questions based on this prompt. The generated test questions are converted to PDF format by the server and sent to the teacher's device as a download link. If necessary, a QR code for online distribution can also be generated.

[1092] Collecting and scanning answer sheets

[1093] The user (teacher) distributes the generated test to students and collects their answers. The collected answer sheets are scanned and the image files are uploaded to the server via the device. The device then sends the uploaded image files to the server.

[1094] Analysis of answers using AI-OCR

[1095] The server receives the uploaded image file and uses AI-OCR technology (e.g., Tesseract OCR) to recognize the text information in the image and convert it into text data. For example, if the image contains the answer "x = 2," AI-OCR converts it into the text data "x = 2." The converted text data is stored in a database.

[1096] Automatic scoring

[1097] The server retrieves the student's textual answers from the database and scores them by comparing them with the correct answer. The automated scoring algorithm calculates a score by taking into account partial credit and multiple answers. For example, if the correct answer for a question is "x = 2," a student who answers "x = 1" will receive partial credit (e.g., 0.5 out of 1).

[1098] Data accumulation and individual optimization

[1099] The server stores each student's answer data and score data in a database and analyzes it to analyze each student's level of understanding and learning trends. For example, it uses Python to analyze the data and identify the correct answer rate and incorrect answer patterns for specific questions. Based on these results, the server generates individually optimized learning materials and notifies the teacher's device. This customized material is provided to students as reinforcement for weak areas and additional practice questions.

[1100] Specific examples

[1101] For example, if a user inputs, "Please create 10 test questions on linear equations for second-year junior high school mathematics students. The difficulty level should be medium, and the target average score is 70 points," the server will send prompts to the generative AI model based on this, generating appropriate questions. The generated test is output in PDF format and sent to the teacher's device. The teacher then distributes the test to students and uploads the collected answer sheets to the server. The server then converts the uploaded answer sheets into text data using AI-OCR, and automatically grades them based on this. Finally, the server analyzes each student's data, generates individually optimized learning materials, and sends them to the teacher's device.

[1102] The above is a specific embodiment of the present invention. This system improves the efficiency of test-related tasks, reduces the burden on educational institutions, and improves students' academic abilities.

[1103] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1104] Program processing flow

[1105] Step 1: Enter conditions

[1106] The user (teacher) logs in to the device and opens the system's test creation screen. The user enters test conditions such as subject, grade, scope, difficulty, and target average score into the input form. The entered test conditions are sent from the device to the server in JSON format by clicking the "Send" button. (Input) Subject, grade, scope, difficulty, target average score. (Output) Test condition data in JSON format.

[1107] Step 2: Analyze and collect condition data

[1108] The server parses the received condition data using a JSON parser. Based on the parsed condition data, the server issues a query to a database (e.g., MySQL or PostgreSQL) to collect relevant past test data and problem data. (Input) Test condition data in JSON format. (Output) Set of past test data and problem data.

[1109] Step 3: Generate prompts and create tests

[1110] The server uses the collected data to generate prompts to be applied to a generative AI model (e.g., OpenAI GPT-4). Specifically, it converts the conditions entered by the user into a text prompt. Next, it inputs this prompt into the generative AI model to generate test questions that match the conditions. (Input) Past test data, problem data, and test conditions. (Output) Generated test questions.

[1111] Step 4: Digitally output the test

[1112] The server converts the generated test questions into PDF format or other digital format. The generated test is notified to the teacher's device as a download link. If necessary, a QR code for online distribution is also generated. (Input) Generated test questions. (Output) PDF or digital format test file, download link, QR code.

[1113] Step 5: Distribute the test and scan the answer sheets

[1114] The user (teacher) checks the generated test and distributes it to the students. After the students answer the test, the user scans the answer sheet with a scanner and uploads the image file to the server via their terminal. (Input) Student's answer sheet. (Output) Image file of the scanned answer sheet.

[1115] Step 6: Answer analysis using AI-OCR

[1116] The server receives the uploaded image file of the answer sheet and uses an AI-OCR engine (e.g., Tesseract OCR) to recognize the text information in the image. The resulting text data is converted into text data and stored in a database. (Input) Image file of the answer sheet. (Output) Answers converted into text data.

[1117] Step 7: Automated scoring

[1118] The server retrieves the answers converted to text data from the database and applies an automatic scoring algorithm. The algorithm compares the answers with the correct answer data and scores the answers taking into account partial points and multiple answers. The scoring results are saved back in the database and sent to the teacher's device. (Input) Answers converted to text data, correct answer data. (Output) Scoring results.

[1119] Step 8: Data accumulation and generation of individually optimized teaching materials

[1120] The server stores each student's answer data and score data in a database and performs data analysis based on this. It analyzes the student's level of understanding and learning tendencies and generates individually optimized learning materials based on this. The generated materials are notified to the teacher's device. (Input) Answer data, score data. (Output) Individually optimized learning materials, notification.

[1121] The above is the specific processing flow of the system of the present invention. This system improves the efficiency of test-related work in educational settings, reduces the burden on teachers, and improves the academic ability of students.

[1122] (Application example 1)

[1123] 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."

[1124] Test-related work in traditional educational settings is time-consuming and labor-intensive, resulting in long working hours for teachers. The process of creating and grading tests is particularly labor-intensive, necessitating greater efficiency. Furthermore, improving the efficiency of picking operations is a key issue in logistics centers, and a system that can efficiently perform tasks based on order data is needed. The purpose of this invention is to provide a system that solves these issues in both educational settings and logistics centers, improving the efficiency and quality of operations.

[1125] 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.

[1126] In this invention, the server includes means for inputting test conditions, means for collecting past test data and question book data, means for automatically generating tests based on the collected data and input conditions, means for digitally outputting the generated tests, means for scanning students' answer sheets and uploading the images to the server, means for converting the answers from the uploaded answer sheet images into text data using character recognition, means for automatically scoring the converted text data, means for saving and analyzing each student's answer data and score data, means for automatically generating an optimal picking list based on order data, means for providing work instructions in real time, and means for updating inventory data by reading item barcodes using a camera. This enables the efficiency and quality of test-related work in educational settings to be improved, as well as the efficiency of picking work in logistics centers.

[1127] "Means for inputting test conditions" refers to a device or software that has an interface that allows teachers to input conditions such as the test subject, grade level, scope, difficulty level, and target average score.

[1128] "Means for collecting past test data and question set data" refers to a device or software that has the function of searching and collecting previously used test data and question set data from a database.

[1129] A "generative model means for automatically generating tests based on collected data and input conditions" is a device or system that uses an artificial intelligence model to automatically generate appropriate tests based on collected data and input conditions.

[1130] A "means for digitally outputting the generated test" is a device or software that outputs the generated test in PDF or other digital format and provides it to teachers and students.

[1131] "Means for scanning students' answer sheets and uploading the images to the server" refers to a device or software for scanning answer sheets on which students have written their answers by hand and uploading the image data of the answer sheets to the server.

[1132] "Means for converting answers into text data using character recognition on the image of the uploaded answer sheet" refers to a device or software that has the function of analyzing the image data of the uploaded answer sheet and converting it into text data using character recognition technology.

[1133] The "means for automatically scoring answers converted into text data" refers to a device or software that uses character recognition technology to compare answers converted into text data with correct answer data and automatically score them.

[1134] "Means for storing and analyzing each student's response data and score data" refers to a device or software that stores each student's response data and score data in a database and analyzes the student's level of understanding and learning tendencies based on this data.

[1135] The "means for automatically generating an optimal picking list based on order data" refers to a device or software for automatically generating an efficient picking list based on order data from a logistics center.

[1136] The "means for providing work instructions in real time" refers to a device or software for providing picking work instructions to staff at the logistics center in real time.

[1137] The "means for reading item barcodes using a camera and updating inventory data" refers to a device or software in a logistics center that uses a camera to read item barcodes and automatically update inventory data.

[1138] An embodiment of the present invention will be described.

[1139] overview

[1140] This invention is a system that aims to improve the efficiency of work in educational institutions and logistics centers. In educational institutions, it aims to improve the efficiency and quality of test-related work, and in logistics centers, it aims to improve the efficiency of picking work.

[1141] Educational systems

[1142] The system includes the following means:

[1143] 1. How to enter test conditions:

[1144] A device with an interface that allows teachers to input conditions such as test subject, grade level, scope, difficulty level, and target average score.

[1145] 2. How to collect past test data and question bank data:

[1146] Software with the function of collecting past test data and question collection data from a database.

[1147] 3. Generative modeling means to automatically generate tests based on collected data and input conditions:

[1148] A system that automatically generates tests using a generative AI model based on collected data and input conditions.

[1149] 4. A means to digitally output the generated tests:

[1150] Software that outputs generated tests in PDF or other digital formats.

[1151] 5. How to scan students' answer sheets and upload the images to the server:

[1152] A device that scans answer sheets filled out by students and uploads the image data to a server.

[1153] 6. How to convert the answers into text data using character recognition on the uploaded image of the answer sheet:

[1154] Software that converts uploaded images into text data using AI-OCR technology.

[1155] 7. Methods for automatically scoring answers converted into text data:

[1156] Software equipped with an algorithm that automatically grades answers converted into text data.

[1157] 8. Means for storing and analyzing student response and score data:

[1158] A system that stores each student's response data and score data in a database and uses this data to analyze academic ability and patterns.

[1159] Hardware and software used

[1160] Hardware: scanners, teacher and student devices, servers

[1161] Software: Generative AI model, AI-OCR, database management system, PDF output tool

[1162] Specific examples

[1163] For example, if a teacher were to create a test on linear equations for second-year junior high school math students, they would use the system as follows:

[1164] Enter the conditions: "Junior high school mathematics, second year, linear equations, medium difficulty, target average score 70 points."

[1165] Logistics Center System

[1166] The system includes the following means:

[1167] 1. A method to automatically generate optimal picking lists based on order data:

[1168] A system that automatically generates efficient picking lists using generative AI models based on order data from logistics centers.

[1169] 2. Means of providing real-time work instructions:

[1170] A device that displays picking work instructions in real time to staff wearing smart glasses.

[1171] 3. Using a camera to read the item's barcode and update inventory data:

[1172] The software uses the camera in the smart glasses to read item barcodes and update inventory data in real time.

[1173] Hardware and software used

[1174] Hardware: smart glasses, servers, warehouse staff terminals

[1175] Software: Generative AI model, AI-OCR, database management system

[1176] Specific examples

[1177] For example, a logistics center might use the system as follows:

[1178] Order data: Item A, Location L3, Quantity 10

[1179] Product B, location L1, quantity 5

[1180] Product C, location L2, quantity 8

[1181] Based on this order data, an optimal picking list is generated and real-time instructions are provided to staff.

[1182] The above is a specific embodiment of the present invention. This system can improve the efficiency and quality of test-related work in educational settings and improve picking work in logistics centers.

[1183] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1184] Step 1:

[1185] The user inputs test conditions into the interface, including the subject, grade, scope, difficulty level, and target average score. The input condition data is then sent from the terminal to the server.

[1186] Input: Test subject, grade, scope, difficulty, target average score, and other conditions

[1187] Output: Input condition data

[1188] Step 2:

[1189] The server collects past test data and question collection data from a database, and the collected data is combined with the input condition data and passed to the generative AI model.

[1190] Input: Database

[1191] Output: Past test data and question collection data

[1192] Step 3:

[1193] The server uses a generative AI model to automatically generate tests based on the collected data and input conditions. The generated test questions are selected to be optimal based on the specified conditions.

[1194] Input: Past test data, question collection data, input condition data

[1195] Output: Generated tests

[1196] Step 4:

[1197] The server outputs the generated test in PDF or other digital format, and the digital file of the generated test is sent to the teacher's device and a download link is provided.

[1198] Input: Generated tests

[1199] Output: Test file in PDF or other digital format

[1200] Step 5:

[1201] The user distributes the generated test to students, who then fill in the answers on the answer sheets and submit them to the teacher, who then scans the answer sheets into their devices.

[1202] Input: Handwritten answer sheet

[1203] Output: Scanned image of answer sheet

[1204] Step 6:

[1205] The scanned image of the answer sheet is uploaded from the user's device to the server, where the uploaded image file is stored.

[1206] Input: Scanned image of answer sheet

[1207] Output: Image file saved on the server

[1208] Step 7:

[1209] The server uses AI-OCR technology to convert the uploaded image of the answer sheet into text data. AI-OCR recognizes the text information in the image and converts it into text format.

[1210] Input: Uploaded image of answer sheet

[1211] Output: Answer converted to text data

[1212] Step 8:

[1213] The server automatically scores the answers converted into text data, and the automatic scoring algorithm compares them with the correct answer data and calculates the score for each answer.

[1214] Input: Answer converted to text data

[1215] Output:Scoring results

[1216] Step 9:

[1217] The server stores each student's answer data and score data in a database, which is then used for later analysis.

[1218] Input: Answer data, score data

[1219] Output: Data stored in the database

[1220] Step 10:

[1221] The server analyzes the accumulated data and determines each student's level of understanding and trends, and generates customized teaching materials and supplementary learning materials for each student.

[1222] Input: Answer data and score data stored in the database

[1223] Output: Customized teaching materials, supplementary teaching materials

[1224] Processing steps for logistics center systems

[1225] Step 1:

[1226] The server retrieves the order data and automatically generates the optimal picking list using an AI model. This prompt is also based on the input conditions.

[1227] Input: Order data

[1228] Output: Generated picking list

[1229] Step 2:

[1230] The server analyzes the generated picking list and provides work instructions in real time. Picking work instructions are displayed to staff wearing smart glasses.

[1231] Input: Generated picking list

[1232] Output: Work instructions

[1233] Step 3:

[1234] Following instructions, staff use the smart glasses' camera to scan the barcode of the item, and the data is sent to the server via the terminal.

[1235] Input: Item barcode

[1236] Output: Barcode data sent to the server

[1237] Step 4:

[1238] The server updates the inventory data based on the scanned barcode data, allowing you to grasp the inventory status in real time.

[1239] Input: Barcode data

[1240] Output: Updated inventory data

[1241] 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.

[1242] This invention is designed to reduce long working hours in educational settings and improve the efficiency and quality of test-related work. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the accuracy of individually optimized learning support can be improved.

[1243] overview

[1244] This system involves a series of processes: teachers input test conditions, generative AI automatically creates tests, and AI-OCR converts students' answers into text data for automatic scoring. Furthermore, it has a function that uses an emotion engine to recognize users' emotions and provides appropriate feedback and adjustments based on those emotions.

[1245] Creating Tests

[1246] The user (teacher) inputs test conditions (subject, grade, scope, difficulty, target average score, etc.) from the terminal. The input conditions are sent from the terminal to the server.

[1247] The server collects past test data and question collection data from a database. The generation AI automatically creates tests based on the collected data and input conditions. The generation AI selects and edits questions based on the set target average score, taking into account the appropriate level of difficulty and question scope.

[1248] The server outputs the generated test in PDF or other digital format, and the output is sent to the teacher's device. If necessary, a QR code can also be generated for online distribution.

[1249] Test distribution and answer collection

[1250] The user (teacher) checks the generated test and distributes it to students, who then fill in the answers and submit the answer sheets.

[1251] The user (teacher) scans the students' answer sheets and uploads them to the server via their device. The scanned answer sheets are saved as image files on the server.

[1252] Answer analysis using AI-OCR

[1253] The server uses AI-OCR technology to analyze the uploaded image of the answer sheet. AI-OCR recognizes the text information in the image and converts it into text data. The converted text data is stored in a database.

[1254] Automatic scoring

[1255] The server compares students' answers, converted into text data, with the correct answers and scores them. The automatic scoring algorithm calculates a score, taking into account partial points and multiple answers. The results are sent to the teacher's device and stored in a database.

[1256] Use of emotion engine

[1257] The server uses an emotion engine to recognize the emotions of users (teachers and students) in real time. The emotion data is used for test creation and feedback.

[1258] For example, if the emotion engine detects stress or impatience in a student, the generative AI will adjust the difficulty of the test accordingly to reduce stress. Similarly, if the emotion engine detects fatigue in a teacher, it will automatically provide candidate questions for test creation, reducing the teacher's burden.

[1259] Data accumulation and individual optimization

[1260] The server stores each student's response data, score data, and emotional data in a database. Based on the stored data, it analyzes each student's level of understanding and trends. It identifies the correct answer rate and incorrect answer patterns for specific questions and generates data to provide individually optimized learning support.

[1261] Based on the analysis results and emotion data, the server generates customized learning materials and supplementary learning materials for each student and notifies the teacher's device, where the teacher can review and provide the materials to the students.

[1262] Specific examples

[1263] For example, let's say a teacher wants to create a test on linear equations for second-year junior high school mathematics students. The teacher enters the following criteria into their device: "second-year junior high school mathematics, linear equations, medium difficulty, target average score of 70 points." The server analyzes past test data and uses generative AI to select and edit questions that meet the criteria.

[1264] The generated test is output in PDF format and a download link is sent to the teacher's device. The teacher then distributes the test to the students and collects their answer sheets. The collected answer sheets are scanned and uploaded to the server as an image file.

[1265] The server uses AI-OCR to convert the image of the answer sheet into text data, and automatically grades the answers based on this. The graded results are sent to the teacher's device and stored in a database. If the emotion engine indicates that the student is feeling stressed, the generative AI will automatically adjust the difficulty of the next test.

[1266] Finally, the server analyzes this data, generates customized learning materials based on each student's level of understanding and emotional data, and notifies the teacher, who can then provide guidance and feedback to the student.

[1267] The above is a specific embodiment of the present invention. This system improves the efficiency of test-related tasks, reduces the burden on teachers, and improves students' academic performance. Furthermore, by combining it with an emotion engine, it becomes possible to flexibly respond to the user's emotions.

[1268] The processing flow will be explained below.

[1269] Step 1:

[1270] The user enters the test conditions (subject, grade, scope, difficulty level, target average score, etc.) on the terminal.

[1271] Step 2:

[1272] The terminal transmits the input conditions to the server.

[1273] Step 3:

[1274] The server collects past test data and question collection data from a database.

[1275] Step 4:

[1276] The server automatically creates tests by selecting appropriate questions using a generation AI based on the collected data and input conditions. The generation AI adjusts the question structure based on the set target average score and difficulty level.

[1277] Step 5:

[1278] The server outputs the generated test in PDF or other digital format and sends a download link to the teacher's device.

[1279] Step 6:

[1280] The user checks the notification on their device, downloads the test, and distributes it to students.

[1281] Step 7:

[1282] The user collects the students' test answer sheets, scans them using the terminal, and uploads them to the server as image files.

[1283] Step 8:

[1284] The server analyzes the image of the answer sheet using AI-OCR technology and converts the answers into text data.

[1285] Step 9:

[1286] The server compares the student's textual answers with the correct answers and calculates a score using an automated scoring algorithm that takes into account partial credit and multiple correct answers.

[1287] Step 10:

[1288] The server sends the grading results to the teacher's terminal and stores them in a database.

[1289] Step 11:

[1290] The server uses an emotion engine to analyze emotion data collected from users' (teachers' and students') devices, for example, to detect student stress or a decline in motivation to learn.

[1291] Step 12:

[1292] The server will then adjust the difficulty of the next test based on the emotional data. For example, if a student is feeling highly stressed, the difficulty of the questions will be set lower.

[1293] Step 13:

[1294] The server generates customized feedback based on the emotional data and sends it to the teacher's device, providing specific advice such as, "Student A is under a lot of stress, so please review your teaching method next time."

[1295] Step 14:

[1296] The server stores each student's response data, score data, and emotional data in a database, and uses this data to analyze their level of understanding and learning trends.

[1297] Step 15:

[1298] Based on the analysis results and emotional data, the server generates individually optimized customized teaching materials and supplementary teaching materials and notifies the teacher's device.

[1299] Step 16:

[1300] The user checks the customized learning materials notified on the device and provides them to the students. The next lesson will be taught based on the feedback.

[1301] These are the specific processing steps of the present invention. This series of steps improves the efficiency of test-related tasks, reduces the burden on teachers, and improves students' academic ability. Furthermore, by combining it with an emotion engine, flexible responses based on the user's emotions become possible.

[1302] Example 2

[1303] 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."

[1304] Traditionally, creating and grading tests in the educational field has been extremely time-consuming and labor-intensive, placing a heavy burden on teachers. It has also been difficult to grasp each student's learning progress and emotional state in real time and provide optimal feedback to each individual student. These challenges have limited the improvement of educational quality and efficiency.

[1305] 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.

[1306] In this invention, the server includes a means for inputting test conditions, a means for collecting past test data and question collection data, a generation AI means for automatically generating tests based on the collected data and input conditions, a means for digitally outputting the generated tests, a means for scanning students' answer sheets and uploading the images to the server, a means for converting the answers from the uploaded answer sheet images into text data using AI-OCR, a means for automatically scoring the converted text data, a means for saving and analyzing answer data and score data for each student, an emotion engine means for recognizing user emotions in real time, a means for adjusting feedback and the difficulty of the next test based on the emotion data, and a means for generating teaching materials optimized for each student and notifying the teacher. This allows teachers to improve the efficiency and reduce the burden of test-related work, and further enables them to provide optimal learning support to each student.

[1307] "Means for inputting test conditions" refers to the means by which teachers input information such as subject, grade, scope, difficulty level, and target average score via a terminal.

[1308] The "means for collecting past test data and question set data" refers to the means by which the server collects appropriate past test data and question set data from the database.

[1309] "Generative AI means" means AI and related software for automatically generating tests based on collected data and input test conditions.

[1310] "Means for outputting the generated test in a digital format" refers to a means for exporting the test created by the generative AI in a digital file format such as PDF.

[1311] The "means for scanning students' answer sheets and uploading the images to the server" refers to a means for a teacher to scan students' answer sheets and send the image data to the server.

[1312] "Method of converting answers into text data using AI-OCR" refers to AI technology that recognizes character information from the image data of uploaded answer sheets and converts it into text data.

[1313] The "means for automatically scoring answers converted into text data" refers to a means for comparing the student's answers converted into text data with correct answer data and calculating the score.

[1314] "Means for saving and analyzing each student's response data and score data" refers to a means for saving each student's response content and scoring results in a database and analyzing their learning.

[1315] The "emotion engine means" refers to a technology and system for recognizing a user's emotions in real time.

[1316] The "means for adjusting the feedback and the difficulty of the next test based on emotional data" refers to a means for dynamically adjusting the feedback content and the difficulty of the next test based on the emotional data obtained by the emotion engine.

[1317] The "means for generating teaching materials optimized for each student and notifying the teacher" refers to a means for generating teaching materials customized for each student based on the analyzed data and emotional data and notifying the teacher of the same.

[1318] The present invention is a system that reduces the workload in educational settings, automates the process from test creation to grading and feedback, and improves the accuracy of learning support using an emotion engine. A specific embodiment of this system will be described.

[1319] System Overview

[1320] In this system, the user (teacher) inputs test conditions via a terminal, and the system automatically generates a test based on those conditions. The generated test is distributed to students, and the students' answer sheets are scanned and uploaded to a server. The server uses AI-OCR technology to convert the image of the answer sheet into text data, which is then automatically graded. Furthermore, an emotion engine is used to recognize the user's emotions, and feedback and the next test conditions are adjusted based on that. The system also includes a function to generate optimized teaching materials and notify the teacher.

[1321] Hardware and Software

[1322] Hardware: Teacher and student devices (PCs, tablets, etc.), servers, scanners

[1323] Software: Generative AI models (e.g., GPT-3), AI-OCR technology (e.g., Google Cloud Vision OCR), emotion engine

[1324] Usage and Examples

[1325] Test condition input

[1326] The user (teacher) enters conditions such as "subject," "grade," "scope," "difficulty level," and "target average score" into the input form on the terminal.

[1327] Examples:

[1328] If a teacher wants to "generate a test on linear equations for second-year junior high school mathematics students," they would enter "Subject: Mathematics," "Grade: Second-year junior high school," "Range: Linear equations," "Difficulty: Medium," and "Target average score: 70 points."

[1329] Test Data Collection and Generation

[1330] The server receives input conditions, collects past test data and question collection data from a database, and then sends prompts to the generative AI model to automatically generate tests.

[1331] Example prompt sentence:

[1332] "Generate a test on linear equations for second-year junior high school mathematics students. The conditions are as follows: difficulty level: medium, target average score: 70 points, test area: Chapter 5 of the textbook."

[1333] Test distribution and answer collection

[1334] The user (teacher) checks the generated test and distributes it to the students. After the students answer, the answer sheets are scanned and uploaded to the server via their terminal.

[1335] Answer analysis using AI-OCR

[1336] The server analyzes the image data of the uploaded answer sheet using AI-OCR technology and converts it into text data.

[1337] Automatic grading and feedback

[1338] The server compares the converted answer data with the correct answer data and automatically scores the answers. The results are sent to the teacher's device and stored in a database.

[1339] Use of emotion engine

[1340] The server uses an emotion engine to recognize the emotions of users (teachers and students) in real time. For example, if a student is feeling stressed, the system can adjust the difficulty of the next test, or if the system detects that a teacher is tired, it can automatically provide appropriate candidate questions.

[1341] Individually optimized learning support

[1342] The server analyzes the accumulated answer data and emotion data, and generates teaching materials optimized for each student based on their level of understanding. These generated teaching materials are then sent to the teacher's device.

[1343] This system allows teachers to streamline test-related tasks and reduce their workload, while also enabling them to provide optimal learning support to students. Furthermore, the emotion engine allows for flexible responses that take into account the user's emotions. As a result, this system not only improves the quality of education, but also contributes to the efficiency of classrooms.

[1344] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1345] Program processing steps

[1346] Step 1: Enter the test conditions

[1347] The user (teacher) uses the terminal to input conditions such as subject, grade, scope, difficulty level, and target average score. Specifically, the user fills in the input form on the GUI with "Subject: Mathematics," "Grade: 2nd year junior high school student," "Scope: Linear equations," "Difficulty level: Medium," and "Target average score: 70 points," and then presses the send button. At this time, these conditions are saved in the terminal's memory as input data.

[1348] Input: Subject, grade, range, difficulty level, target average score

[1349] Output: Test condition data sent to the server

[1350] Step 2: Sending test condition data

[1351] The device sends the entered test condition data to the server using an HTTP POST request. The specific data sent is in JSON format.

[1352] Input: Test condition data entered by the user

[1353] Output: Condition data in JSON format passed to the server

[1354] Step 3: Collect test data

[1355] The server receives the test condition data and accesses the database to collect past test data and problem set data. For example, it searches for problems that match the condition "Mathematics, Junior High School 2nd Grade, Linear Equations." It then uses a database query to extract the relevant data.

[1356] Input: Test condition data received by the server

[1357] Output: Past test data and question bank data retrieved from the database

[1358] Step 4: Automatically generate tests

[1359] The server sends prompts to the generative AI model based on the collected data and input conditions. The generative AI model (e.g., GPT-3) generates tests based on the prompts.

[1360] Specific prompt:

[1361] "Generate a test on linear equations for second-year junior high school mathematics students. The conditions are as follows: difficulty level: medium, target average score: 70 points, test area: Chapter 5 of the textbook."

[1362] Input: Past test data from the database, test conditions entered by the user, and prompts for the generative AI model

[1363] Output: Generated test data (questions, answer sheets, etc.)

[1364] Step 5: Generated Test Output

[1365] The server exports the generated test data in PDF or other digital file format, and the generated PDF file is saved in the server's storage.

[1366] Input: Test data generated by the generative AI model

[1367] Output: Test file in PDF format

[1368] Step 6: Notification of generated tests

[1369] The server generates a URL for the generated test file and notifies the user (teacher) of the URL. If necessary, it also generates and sends a QR code for online distribution.

[1370] Input: Generated test file in PDF format

[1371] Output: URL link and QR code

[1372] Step 7: Distribute the test

[1373] The user (teacher) reviews the generated test and distributes it to students by printing it on paper, or by email or via an online platform.

[1374] Input: URL link and QR code

[1375] Output: Tests distributed to students

[1376] Step 8: Collect answer sheets

[1377] The user (teacher) scans the answer sheets on which the students have completed the questions and imports them into the device. An image file of the answer sheet is generated through the scanner.

[1378] Input: Student's completed answer sheet

[1379] Output: Image file of the answer sheet captured on the device

[1380] Step 9: Upload your answer sheet

[1381] The device uploads the scanned image data of the answer sheet to the server using an HTTP POST request.

[1382] Input: Image file of answer sheet

[1383] Output: Image data of the answer sheet sent to the server

[1384] Step 10: Answer analysis using AI-OCR

[1385] The server receives the uploaded image data of the answer sheet and converts the character information into text data using AI-OCR technology. Specifically, the AI-OCR system recognizes the characters in the image and converts them into text format. For example, a mathematical formula such as "12x + 8 = 4y" is obtained as text data.

[1386] Input: Image data of the answer sheet

[1387] Output: Answers converted to text data

[1388] Step 11: Automated scoring

[1389] The server receives the answers converted into text data, compares them with the correct answer data, and automatically scores them. For example, if the answer is "12x + 8 = 4y," the server compares it with the correct answer data and calculates the score taking into account partial points and point allocation.

[1390] Input: Textual answer data, correct answer data

[1391] Output: Scoring results (score data)

[1392] Step 12: Notification of Scoring Results

[1393] The server notifies the user (teacher) of the grading results and simultaneously stores them in a database.

[1394] Input: Scoring results

[1395] Output: Marking results sent to the teacher's device

[1396] Step 13: Emotion Recognition

[1397] The server uses an emotion engine to recognize the emotions of users (teachers and students) in real time. For example, it analyzes the user's emotional state (stress, frustration, fatigue, etc.) from data acquired through a camera or microphone.

[1398] Input: Emotion data obtained from a camera or microphone

[1399] Output: Recognized emotion data

[1400] Step 14: Adjusting Feedback

[1401] Based on the recognized emotion data, the server sends prompts to the generative AI model to provide feedback and adjust the difficulty of the next test. For example, if the emotion engine detects stress, it will instruct the generative AI to lower the difficulty of the next test.

[1402] Input: Recognized emotion data

[1403] Output: Adjusted feedback and difficulty of next test

[1404] Step 15: Individually optimized learning support

[1405] The server analyzes each student's response data, score data, and emotion data to generate optimized teaching materials, which are then sent to the teacher's device and provided to the students.

[1406] Input: Answer data, score data, emotion data

[1407] Output: Optimized learning materials

[1408] These are the specific processing steps of this system. This system aims to reduce the burden on teachers and provide a more effective and efficient educational environment.

[1409] (Application example 2)

[1410] 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."

[1411] Physical stores are seeking training methods to efficiently improve their employees' customer service skills. Conventional training methods rely on subjective evaluations by trainers, making it difficult to provide objective feedback and resulting in inconsistencies in the quality and efficiency of training. Furthermore, there is a lack of means to recognize employees' emotions and stress levels and provide appropriate feedback. Given this background, a system is needed to support employee skill improvement and increase customer satisfaction.

[1412] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1413] In this invention, the server includes: means for inputting test conditions; means for collecting past test data and question collection data; AI generation means for automatically generating tests based on the collected data and input conditions; means for digitally outputting the generated tests; means for scanning images of answer sheets and training data and uploading them to the server; means for converting the uploaded images into text data using AI-OCR; means for automatically scoring the converted text data; an emotion recognition engine for recognizing user emotions in real time; means for providing appropriate feedback to users based on the recognized emotion data; means for saving and analyzing answer data, score data, and emotion data for each user; and means for generating a report that provides feedback on users' strengths and areas for improvement based on the response content and emotion data. This enables objective evaluation of employees' customer service skills and appropriate feedback based on the emotion data.

[1414] Definition of Terms

[1415] "Test conditions" refers to the information that a user enters to generate a specific test or training, specifically settings such as subject, target grade, question scope, difficulty level, and target average score.

[1416] "Past test data and question bank data" refers to data collected from previously administered tests or existing question banks and used for test creation and training.

[1417] "Generative AI means" refers to artificial intelligence technology that automatically generates appropriate test questions or training tasks based on input conditions.

[1418] "Digital output means" means a means for providing the generated test or training assignment in PDF or other digital file format.

[1419] "Means for scanning images of answer sheets and training data and uploading them to a server" refers to a means for capturing answer sheets and training records completed by students or employees as digital images and transmitting them to a server via a network.

[1420] "Means for converting scanned image data into text data using AI-OCR" refers to a means for converting scanned image data into text data using optical character recognition technology.

[1421] An "automatic scoring means" is a means that has an algorithm that calculates scores by comparing the answers converted into text data with the correct answer data.

[1422] An "emotion recognition engine" refers to technology that identifies emotions in real time from a user's facial expressions, voice, text data, etc.

[1423] The "means for providing feedback" is a means for providing appropriate comments and advice to the user based on the recognized emotion data and analysis results.

[1424] The "means for storing and analyzing response data, score data, and emotional data for each user" refers to a means for storing performance data and emotional data for each user and analyzing them.

[1425] The "means for generating a feedback report" is a means for organizing the user's strengths and areas for improvement based on the response content and emotional data, and providing them in the form of a report.

[1426] MODE FOR CARRYING OUT THE INVENTION

[1427] The present invention relates to an emotion recognition training system for helping store employees improve their customer service skills. This system allows employees to receive real-time feedback using a smartphone or smart glasses while undergoing customer service training. Detailed embodiments of the present invention are described below.

[1428] Hardware and software used

[1429] Hardware

[1430] Smartphone

[1431] Smart Glasses

[1432] server

[1433] software

[1434] AI-OCR engine (e.g. Tesseract OCR)

[1435] Emotion recognition engine (e.g. Microsoft Azure Emotion API)

[1436] Generative AI methods (e.g., text generation AI)

[1437] Database Management Systems

[1438] System configuration

[1439] User terminal

[1440] Users (employees) use smartphones or smart glasses to conduct customer service training. At the start of a training session, users input training conditions (e.g., customer service scenario, target skills, training time, etc.). This allows users to focus on the training scenario.

[1441] server

[1442] The server manages the training session and processes the data in the following manner.

[1443] 1. Enter and collect test conditions

[1444] Collect training conditions entered by the user.

[1445] Past training data and question collection data are collected from the database.

[1446] 2. Training and generation using generative AI models

[1447] Training content is automatically generated based on collected data and conditions.

[1448] The generated training is output in digital format and delivered to the user's smart device.

[1449] 3. Scan and upload your answers and training data

[1450] The data collected by the user during training (audio, video, text, etc.) is scanned and uploaded to the server.

[1451] 4. Text data conversion using AI-OCR

[1452] The uploaded data is converted into text data using an AI-OCR engine.

[1453] 5. Use of automatic scoring and emotion recognition engine

[1454] The system automatically scores answers converted into text data and uses an emotion recognition engine to recognize the emotions of users and virtual customers in real time.

[1455] 6. Providing Feedback

[1456] Based on the recognized emotion data, appropriate feedback is provided to the user.

[1457] The report shows users' strengths and areas for improvement, providing useful advice for the next training session.

[1458] Specific examples

[1459] For example, consider the case where employee A receives customer service training on explaining a new product. Employee A wears smart glasses and conducts the training while interacting with a virtual customer. When the training conditions are entered as "new product description, target skill: product knowledge, training time: 10 minutes," the server analyzes past training data and generates a scenario. During the training, employee A's responses and facial expressions are recorded and uploaded to the server. AI-OCR converts the responses into text, and an emotion recognition engine analyzes the emotions of employee A and the virtual customer. After the training is completed, the server generates a feedback report, assessing that "employee A was able to explain the product's features in detail, but lacked confidence in answering customer questions." This allows employee A to know specific areas for improvement for the next training session.

[1460] Example prompts to input to the generative AI model

[1461] Analysis of customer service training logs:

[1462] 1. Convert the response content into text using AI-OCR

[1463] 2. Identify emotional data with an emotion recognition engine

[1464] 3. Feedback on strengths and areas for improvement

[1465] As a result, the present invention can efficiently and objectively support the improvement of customer service skills of employees in physical stores, thereby increasing customer satisfaction.

[1466] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1467] Program processing steps

[1468] Step 1:

[1469] The user starts a training session using a smartphone or smart glasses. They input the training conditions (e.g., customer service scenario, target skills, training time, etc.). This sets specific training objectives, and the information is sent to the server. Input data: Training conditions. Output data: Set training conditions.

[1470] Step 2:

[1471] The server receives the training condition data sent by the user and collects past training data and question set data from the database. The collected data is input into the generative AI model. The server generates an appropriate training scenario. Input data: training conditions, past data. Output data: generated training scenario.

[1472] Step 3:

[1473] The server outputs the generated training scenario in a digital format (e.g. PDF, text file) and delivers it to the user's smart device. The user then performs training according to this training scenario. Input data: The generated training scenario. Output data: The digital file of the training scenario.

[1474] Step 4:

[1475] Users conduct training and record video and audio data and input text during the training process. This data is uploaded to the server from their smart device as answer sheets and training data. Input data: video data, audio data, text data. Output data: training data uploaded to the server.

[1476] Step 5:

[1477] The server receives the uploaded training data and converts it into text data using the AI-OCR engine. The converted text data is saved for analysis. Input data: image data, audio data. Output data: text data.

[1478] Step 6:

[1479] The server automatically scores the text data and uses an emotion recognition engine to recognize the emotions of the user and virtual customers in real time. The recognized emotion data is saved for analysis. Input data: text data, emotion data. Output data: scoring results, emotion data.

[1480] Step 7:

[1481] The server generates a report based on the text data and emotion data to provide appropriate feedback to the user. The report includes the user's strengths and areas for improvement and serves as a guide for the next training session. Input data: text data, emotion data. Output data: feedback report.

[1482] Step 8:

[1483] A feedback report is sent to the user's device, and the user can check the contents. The feedback can be used for the next training session. Input data: Feedback report. Output data: Feedback received by the user.

[1484] Through these steps, users can receive objective evaluations and feedback based on their emotions, allowing them to efficiently improve their customer service skills.

[1485] 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.

[1486] 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.

[1487] 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.

[1488] [Fourth embodiment]

[1489] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1490] 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.

[1491] 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).

[1492] 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.

[1493] 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.

[1494] 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).

[1495] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1496] 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.

[1497] 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.

[1498] 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.

[1499] 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.

[1500] 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.

[1501] 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."

[1502] The present invention is a system that reduces long working hours in educational settings and improves the efficiency and quality of test-related work. Specific embodiments of the present invention will be described in detail below.

[1503] overview

[1504] This system involves a series of processes: teachers input test conditions, the system automatically creates tests using generative AI, and the system converts students' answers into text data using AI-OCR for automatic grading. Furthermore, the system provides optimal learning support for each student based on the data stored in the database.

[1505] Creating Tests

[1506] The user (teacher) inputs test conditions (subject, grade, scope, difficulty, target average score, etc.) from the terminal. The input conditions are sent from the terminal to the server.

[1507] The server collects past test data and question collection data from a database. The generation AI automatically creates tests based on the collected data and input conditions. The generation AI selects and edits questions based on the set target average score, taking into account the appropriate level of difficulty and question scope.

[1508] The server outputs the generated test in PDF or other digital format, and the output is sent to the teacher's device. If necessary, a QR code can also be generated for online distribution.

[1509] Test distribution and answer collection

[1510] The user (teacher) checks the generated test and distributes it to students, who then fill in the answers and submit the answer sheets.

[1511] The user (teacher) scans the students' answer sheets and uploads them to the server via their device. The scanned answer sheets are saved as image files on the server.

[1512] Answer analysis using AI-OCR

[1513] The server uses AI-OCR technology to analyze the uploaded image of the answer sheet. AI-OCR recognizes the text information in the image and converts it into text data. The converted text data is stored in a database.

[1514] Automatic scoring

[1515] The server compares students' answers, converted into text data, with the correct answers and scores them. The automatic scoring algorithm calculates a score, taking into account partial points and multiple answers. The results are sent to the teacher's device and stored in a database.

[1516] Data accumulation and individual optimization

[1517] The server stores each student's answer data and score data in a database. Based on the stored data, it analyzes each student's level of understanding and trends. It identifies the correct answer rate and incorrect answer patterns for specific questions and generates data for individually optimized learning support.

[1518] Based on the analysis results, the server generates customized learning materials and supplementary learning materials for each student and notifies the teacher's device, where the teacher can review and provide the materials to the student.

[1519] Specific examples

[1520] For example, let's say a teacher wants to create a test on linear equations for second-year junior high school mathematics students. The teacher enters the following criteria into their device: "second-year junior high school mathematics, linear equations, medium difficulty, target average score of 70 points." The server analyzes past test data and uses generative AI to select and edit questions that meet the criteria.

[1521] The generated test is output in PDF format and a download link is sent to the teacher's device. The teacher then distributes the test to the students and collects the answer sheets. The collected answer sheets are scanned and uploaded to the server as an image file.

[1522] The server uses AI-OCR to convert the image of the answer sheet into text data, and automatically grades the answers based on this data. The graded results are sent to the teacher's device and stored in a database. Finally, the server analyzes this data, generates customized teaching materials based on each student's level of understanding, and notifies the teacher.

[1523] The above is a specific embodiment of the present invention. This system improves the efficiency of test-related tasks, reduces the burden on teachers, and improves the academic ability of students.

[1524] The processing flow will be explained below.

[1525] Step 1:

[1526] The user enters the test conditions (subject, grade, scope, difficulty level, target average score, etc.) on the terminal.

[1527] Step 2:

[1528] The terminal transmits the input conditions to the server.

[1529] Step 3:

[1530] The server collects past test data and question collection data from a database.

[1531] Step 4:

[1532] The server automatically creates tests by selecting appropriate questions using a generation AI based on the collected data and input conditions. The generation AI adjusts the questions to meet the specified difficulty level and target average score.

[1533] Step 5:

[1534] The server outputs the generated test in PDF or other digital format and notifies the teacher's device with a download link.

[1535] Step 6:

[1536] The user checks the notification on their device, downloads the generated test, and distributes it to students.

[1537] Step 7:

[1538] The user collects the students' answers, scans the answer sheets on the device, and uploads them to the server as image files.

[1539] Step 8:

[1540] The server analyzes the uploaded image of the answer sheet using AI-OCR technology and converts the answers into text data.

[1541] Step 9:

[1542] The server compares the textual answers with the correct answers and calculates the score using an automated scoring algorithm, taking into account partial points and multiple correct answers.

[1543] Step 10:

[1544] The server sends the grading results to the teacher's terminal and stores them in a database.

[1545] Step 11:

[1546] The server analyzes each student's response data and score data and runs algorithms to identify their level of understanding and learning trends.

[1547] Step 12:

[1548] The server generates individually optimized learning support materials based on the analysis results and presents them on the teacher's device.

[1549] Step 13:

[1550] The user checks the learning support materials presented on the terminal and provides them to the students.

[1551] These are the specific processing steps in the system of the present invention. This process streamlines the process from test creation to grading, data accumulation, and individual learning support, and alleviates the problem of long working hours in educational settings.

[1552] Example 1

[1553] 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."

[1554] In educational settings, creating tests, collecting answers, grading, and individually optimizing learning support requires a great deal of time and effort. Providing testing quickly while maintaining test quality and fairness is a major challenge for many educators. Providing efficient and effective learning support based on each student's level of understanding is also a difficult problem. There is a need for a method to efficiently solve these issues, reduce the burden on educational settings, and improve the quality of learning.

[1555] 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.

[1556] In this invention, the server includes means for inputting test conditions, means for collecting past test data and question data, artificial intelligence means for automatically generating tests based on the collected data and input conditions, means for digitally outputting the generated tests, means for digitizing students' answer sheets and uploading the images to a computer, means for converting the answers from the uploaded answer sheet images into text data using optical character recognition technology, means for automatically scoring the converted text data, means for saving and analyzing answer data and score data for each student, means for generating individually optimized learning materials based on the analysis results, and means for notifying students of the generated learning materials. This makes it possible to improve the efficiency and quality of test-related work in educational settings.

[1557] "Means for inputting test conditions" refers to a function that allows educators to input test conditions such as subject, grade, scope, difficulty level, and target average score via a terminal.

[1558] "Means for collecting past test data and question data" is a function for collecting past test data and question set data from a database.

[1559] "Artificial intelligence means" means functionality that includes a generative AI model used to automatically generate tests based on collected data and input criteria.

[1560] "Means for outputting the generated test in digital format" refers to a function for outputting the generated test in PDF format or other digital format and notifying the teacher's terminal.

[1561] "Means for digitizing students' answer sheets and uploading the images to a computer" is a function that allows teachers to scan students' answer sheets with a scanner and upload the image data to a server.

[1562] "Means of converting answers into text data using optical character recognition technology" refers to a function that includes AI-OCR technology, which is used to analyze the image of the uploaded answer sheet and convert the answer content into text data.

[1563] The "means for automatically scoring answers converted into text data" is a function for automatically scoring students' answers converted into text data by comparing them with correct answer data.

[1564] "Means for saving and analyzing response data and score data for each student" refers to a function for saving student response data and score data in a database and analyzing this data.

[1565] The "means for generating individually optimized learning materials" is a function for generating customized learning materials according to each student's level of understanding based on the analyzed data.

[1566] The "means for notifying the generated learning materials" is a function for notifying the teacher's terminal of the generated customized learning materials.

[1567] The present invention is a system that improves the efficiency and quality of test-related tasks in educational settings. This system involves a series of processes: teachers input test conditions, automatically generate tests based on those conditions, analyze and score student answers using AI technology, and provide individually optimized learning materials. Specific embodiments of the present invention are described below.

[1568] Test condition input

[1569] The user (teacher) logs in to the terminal and opens the system's test creation screen. The user enters the following conditions into the form:

[1570] Subject (e.g. Mathematics)

[1571] Grade (e.g., second grade)

[1572] Range (e.g. linear equations)

[1573] Difficulty level (e.g. medium)

[1574] Target average score (e.g. 70 points)

[1575] Once you have completed entering the information, click the "Send" button and the input data will be sent from the terminal to the server in JSON format.

[1576] Automatic test generation

[1577] The server analyzes the received condition data and collects past test data and problem data from a database (e.g., MySQL, PostgreSQL, etc.). Based on the collected data, the server sends a prompt to a generative AI model (e.g., OpenAI GPT-4), inputting the following prompt to the generative AI model:

[1578] "Please create 10 test questions on linear equations for second-year junior high school mathematics students. The difficulty level should be medium, with a target average score of 70 points."

[1579] The generative AI model generates test questions based on this prompt. The generated test questions are converted to PDF format by the server and sent to the teacher's device as a download link. If necessary, a QR code for online distribution can also be generated.

[1580] Collecting and scanning answer sheets

[1581] The user (teacher) distributes the generated test to students and collects their answers. The collected answer sheets are scanned and the image files are uploaded to the server via the device. The device then sends the uploaded image files to the server.

[1582] Analysis of answers using AI-OCR

[1583] The server receives the uploaded image file and uses AI-OCR technology (e.g., Tesseract OCR) to recognize the text information in the image and convert it into text data. For example, if the image contains the answer "x = 2," AI-OCR converts it into the text data "x = 2." The converted text data is stored in a database.

[1584] Automatic scoring

[1585] The server retrieves the student's textual answers from the database and scores them by comparing them with the correct answer. The automated scoring algorithm calculates a score by taking into account partial credit and multiple answers. For example, if the correct answer for a question is "x = 2," a student who answers "x = 1" will receive partial credit (e.g., 0.5 out of 1).

[1586] Data accumulation and individual optimization

[1587] The server stores each student's answer data and score data in a database and analyzes it to analyze each student's level of understanding and learning trends. For example, it uses Python to analyze the data and identify the correct answer rate and incorrect answer patterns for specific questions. Based on these results, the server generates individually optimized learning materials and notifies the teacher's device. This customized material is provided to students as reinforcement for weak areas and additional practice questions.

[1588] Specific examples

[1589] For example, if a user inputs, "Please create 10 test questions on linear equations for second-year junior high school mathematics students. The difficulty level should be medium, and the target average score is 70 points," the server will send prompts to the generative AI model based on this, generating appropriate questions. The generated test is output in PDF format and sent to the teacher's device. The teacher then distributes the test to students and uploads the collected answer sheets to the server. The server then converts the uploaded answer sheets into text data using AI-OCR, and automatically grades them based on this. Finally, the server analyzes each student's data, generates individually optimized learning materials, and sends them to the teacher's device.

[1590] The above is a specific embodiment of the present invention. This system improves the efficiency of test-related tasks, reduces the burden on educational institutions, and improves students' academic abilities.

[1591] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1592] Program processing flow

[1593] Step 1: Enter conditions

[1594] The user (teacher) logs in to the device and opens the system's test creation screen. The user enters test conditions such as subject, grade, scope, difficulty, and target average score into the input form. The entered test conditions are sent from the device to the server in JSON format by clicking the "Send" button. (Input) Subject, grade, scope, difficulty, target average score. (Output) Test condition data in JSON format.

[1595] Step 2: Analyze and collect condition data

[1596] The server parses the received condition data using a JSON parser. Based on the parsed condition data, the server issues a query to a database (e.g., MySQL or PostgreSQL) to collect relevant past test data and problem data. (Input) Test condition data in JSON format. (Output) Set of past test data and problem data.

[1597] Step 3: Generate prompts and create tests

[1598] The server uses the collected data to generate prompts to be applied to a generative AI model (e.g., OpenAI GPT-4). Specifically, it converts the conditions entered by the user into a text prompt. Next, it inputs this prompt into the generative AI model to generate test questions that match the conditions. (Input) Past test data, problem data, and test conditions. (Output) Generated test questions.

[1599] Step 4: Digitally output the test

[1600] The server converts the generated test questions into PDF format or other digital format. The generated test is notified to the teacher's device as a download link. If necessary, a QR code for online distribution is also generated. (Input) Generated test questions. (Output) PDF or digital format test file, download link, QR code.

[1601] Step 5: Distribute the test and scan the answer sheets

[1602] The user (teacher) checks the generated test and distributes it to the students. After the students answer the test, the user scans the answer sheet with a scanner and uploads the image file to the server via their terminal. (Input) Student's answer sheet. (Output) Image file of the scanned answer sheet.

[1603] Step 6: Answer analysis using AI-OCR

[1604] The server receives the uploaded image file of the answer sheet and uses an AI-OCR engine (e.g., Tesseract OCR) to recognize the text information in the image. The resulting text data is converted into text data and stored in a database. (Input) Image file of the answer sheet. (Output) Answers converted into text data.

[1605] Step 7: Automated scoring

[1606] The server retrieves the answers converted to text data from the database and applies an automatic scoring algorithm. The algorithm compares the answers with the correct answer data and scores the answers taking into account partial points and multiple answers. The scoring results are saved back in the database and sent to the teacher's device. (Input) Answers converted to text data, correct answer data. (Output) Scoring results.

[1607] Step 8: Data accumulation and generation of individually optimized teaching materials

[1608] The server stores each student's answer data and score data in a database and performs data analysis based on this. It analyzes the student's level of understanding and learning tendencies and generates individually optimized learning materials based on this. The generated materials are notified to the teacher's device. (Input) Answer data, score data. (Output) Individually optimized learning materials, notification.

[1609] The above is the specific processing flow of the system of the present invention. This system improves the efficiency of test-related work in educational settings, reduces the burden on teachers, and improves the academic ability of students.

[1610] (Application example 1)

[1611] 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."

[1612] Test-related work in traditional educational settings is time-consuming and labor-intensive, resulting in long working hours for teachers. The process of creating and grading tests is particularly labor-intensive, necessitating greater efficiency. Furthermore, improving the efficiency of picking operations is a key issue in logistics centers, and a system that can efficiently perform tasks based on order data is needed. The purpose of this invention is to provide a system that solves these issues in both educational settings and logistics centers, improving the efficiency and quality of operations.

[1613] 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.

[1614] In this invention, the server includes means for inputting test conditions, means for collecting past test data and question book data, means for automatically generating tests based on the collected data and input conditions, means for digitally outputting the generated tests, means for scanning students' answer sheets and uploading the images to the server, means for converting the answers from the uploaded answer sheet images into text data using character recognition, means for automatically scoring the converted text data, means for saving and analyzing each student's answer data and score data, means for automatically generating an optimal picking list based on order data, means for providing work instructions in real time, and means for updating inventory data by reading item barcodes using a camera. This enables the efficiency and quality of test-related work in educational settings to be improved, as well as the efficiency of picking work in logistics centers.

[1615] "Means for inputting test conditions" refers to a device or software that has an interface that allows teachers to input conditions such as the test subject, grade level, scope, difficulty level, and target average score.

[1616] "Means for collecting past test data and question set data" refers to a device or software that has the function of searching and collecting previously used test data and question set data from a database.

[1617] A "generative model means for automatically generating tests based on collected data and input conditions" is a device or system that uses an artificial intelligence model to automatically generate appropriate tests based on collected data and input conditions.

[1618] A "means for digitally outputting the generated test" is a device or software that outputs the generated test in PDF or other digital format and provides it to teachers and students.

[1619] "Means for scanning students' answer sheets and uploading the images to the server" refers to a device or software for scanning answer sheets on which students have written their answers by hand and uploading the image data of the answer sheets to the server.

[1620] "Means for converting answers into text data using character recognition on the image of the uploaded answer sheet" refers to a device or software that has the function of analyzing the image data of the uploaded answer sheet and converting it into text data using character recognition technology.

[1621] The "means for automatically scoring answers converted into text data" refers to a device or software that uses character recognition technology to compare answers converted into text data with correct answer data and automatically score them.

[1622] "Means for storing and analyzing each student's response data and score data" refers to a device or software that stores each student's response data and score data in a database and analyzes the student's level of understanding and learning tendencies based on this data.

[1623] The "means for automatically generating an optimal picking list based on order data" refers to a device or software for automatically generating an efficient picking list based on order data from a logistics center.

[1624] The "means for providing work instructions in real time" refers to a device or software for providing picking work instructions to staff at the logistics center in real time.

[1625] The "means for reading item barcodes using a camera and updating inventory data" refers to a device or software in a logistics center that uses a camera to read item barcodes and automatically update inventory data.

[1626] An embodiment of the present invention will be described.

[1627] overview

[1628] This invention is a system that aims to improve the efficiency of work in educational institutions and logistics centers. In educational institutions, it aims to improve the efficiency and quality of test-related work, and in logistics centers, it aims to improve the efficiency of picking work.

[1629] Educational systems

[1630] The system includes the following means:

[1631] 1. How to enter test conditions:

[1632] A device with an interface that allows teachers to input conditions such as test subject, grade level, scope, difficulty level, and target average score.

[1633] 2. How to collect past test data and question bank data:

[1634] Software with the function of collecting past test data and question collection data from a database.

[1635] 3. Generative modeling means to automatically generate tests based on collected data and input conditions:

[1636] A system that automatically generates tests using a generative AI model based on collected data and input conditions.

[1637] 4. A means to digitally output the generated tests:

[1638] Software that outputs generated tests in PDF or other digital formats.

[1639] 5. How to scan students' answer sheets and upload the images to the server:

[1640] A device that scans answer sheets filled out by students and uploads the image data to a server.

[1641] 6. How to convert the answers into text data using character recognition on the uploaded image of the answer sheet:

[1642] Software that converts uploaded images into text data using AI-OCR technology.

[1643] 7. Methods for automatically scoring answers converted into text data:

[1644] Software equipped with an algorithm that automatically grades answers converted into text data.

[1645] 8. Means for storing and analyzing student response and score data:

[1646] A system that stores each student's response data and score data in a database and uses this data to analyze academic ability and patterns.

[1647] Hardware and software used

[1648] Hardware: scanners, teacher and student devices, servers

[1649] Software: Generative AI model, AI-OCR, database management system, PDF output tool

[1650] Specific examples

[1651] For example, if a teacher were to create a test on linear equations for second-year junior high school math students, they would use the system as follows:

[1652] Enter the conditions: "Junior high school mathematics, second year, linear equations, medium difficulty, target average score 70 points."

[1653] Logistics Center System

[1654] The system includes the following means:

[1655] 1. A method to automatically generate optimal picking lists based on order data:

[1656] A system that automatically generates efficient picking lists using generative AI models based on order data from logistics centers.

[1657] 2. Means of providing real-time work instructions:

[1658] A device that displays picking work instructions in real time to staff wearing smart glasses.

[1659] 3. Using a camera to read the item's barcode and update inventory data:

[1660] The software uses the camera in the smart glasses to read item barcodes and update inventory data in real time.

[1661] Hardware and software used

[1662] Hardware: smart glasses, servers, warehouse staff terminals

[1663] Software: Generative AI model, AI-OCR, database management system

[1664] Specific examples

[1665] For example, a logistics center might use the system as follows:

[1666] Order data: Item A, Location L3, Quantity 10

[1667] Product B, location L1, quantity 5

[1668] Product C, location L2, quantity 8

[1669] Based on this order data, an optimal picking list is generated and real-time instructions are provided to staff.

[1670] The above is a specific embodiment of the present invention. This system can improve the efficiency and quality of test-related work in educational settings and improve picking work in logistics centers.

[1671] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1672] Step 1:

[1673] The user inputs test conditions into the interface, including the subject, grade, scope, difficulty level, and target average score. The input condition data is then sent from the terminal to the server.

[1674] Input: Test subject, grade, scope, difficulty, target average score, and other conditions

[1675] Output: Input condition data

[1676] Step 2:

[1677] The server collects past test data and question collection data from a database, and the collected data is combined with the input condition data and passed to the generative AI model.

[1678] Input: Database

[1679] Output: Past test data and question collection data

[1680] Step 3:

[1681] The server uses a generative AI model to automatically generate tests based on the collected data and input conditions. The generated test questions are selected to be optimal based on the specified conditions.

[1682] Input: Past test data, question collection data, input condition data

[1683] Output: Generated tests

[1684] Step 4:

[1685] The server outputs the generated test in PDF or other digital format, and the digital file of the generated test is sent to the teacher's device and a download link is provided.

[1686] Input: Generated tests

[1687] Output: Test file in PDF or other digital format

[1688] Step 5:

[1689] The user distributes the generated test to students, who then fill in the answers on the answer sheets and submit them to the teacher, who then scans the answer sheets into their devices.

[1690] Input: Handwritten answer sheet

[1691] Output: Scanned image of answer sheet

[1692] Step 6:

[1693] The scanned image of the answer sheet is uploaded from the user's device to the server, where the uploaded image file is stored.

[1694] Input: Scanned image of answer sheet

[1695] Output: Image file saved on the server

[1696] Step 7:

[1697] The server uses AI-OCR technology to convert the uploaded image of the answer sheet into text data. AI-OCR recognizes the text information in the image and converts it into text format.

[1698] Input: Uploaded image of answer sheet

[1699] Output: Answer converted to text data

[1700] Step 8:

[1701] The server automatically scores the answers converted into text data, and the automatic scoring algorithm compares them with the correct answer data and calculates the score for each answer.

[1702] Input: Answer converted to text data

[1703] Output:Scoring results

[1704] Step 9:

[1705] The server stores each student's answer data and score data in a database, which is then used for later analysis.

[1706] Input: Answer data, score data

[1707] Output: Data stored in the database

[1708] Step 10:

[1709] The server analyzes the accumulated data and determines each student's level of understanding and trends, and generates customized teaching materials and supplementary learning materials for each student.

[1710] Input: Answer data and score data stored in the database

[1711] Output: Customized teaching materials, supplementary teaching materials

[1712] Processing steps for logistics center systems

[1713] Step 1:

[1714] The server retrieves the order data and automatically generates the optimal picking list using an AI model. This prompt is also based on the input conditions.

[1715] Input: Order data

[1716] Output: Generated picking list

[1717] Step 2:

[1718] The server analyzes the generated picking list and provides work instructions in real time. Picking work instructions are displayed to staff wearing smart glasses.

[1719] Input: Generated picking list

[1720] Output: Work instructions

[1721] Step 3:

[1722] Following instructions, staff use the smart glasses' camera to scan the barcode of the item, and the data is sent to the server via the terminal.

[1723] Input: Item barcode

[1724] Output: Barcode data sent to the server

[1725] Step 4:

[1726] The server updates the inventory data based on the scanned barcode data, allowing you to grasp the inventory status in real time.

[1727] Input: Barcode data

[1728] Output: Updated inventory data

[1729] 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.

[1730] This invention is designed to reduce long working hours in educational settings and improve the efficiency and quality of test-related work. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the accuracy of individually optimized learning support can be improved.

[1731] overview

[1732] This system involves a series of processes: teachers input test conditions, generative AI automatically creates tests, and AI-OCR converts students' answers into text data for automatic scoring. Furthermore, it has a function that uses an emotion engine to recognize users' emotions and provides appropriate feedback and adjustments based on those emotions.

[1733] Creating Tests

[1734] The user (teacher) inputs test conditions (subject, grade, scope, difficulty, target average score, etc.) from the terminal. The input conditions are sent from the terminal to the server.

[1735] The server collects past test data and question collection data from a database. The generation AI automatically creates tests based on the collected data and input conditions. The generation AI selects and edits questions based on the set target average score, taking into account the appropriate level of difficulty and question scope.

[1736] The server outputs the generated test in PDF or other digital format, and the output is sent to the teacher's device. If necessary, a QR code can also be generated for online distribution.

[1737] Test distribution and answer collection

[1738] The user (teacher) checks the generated test and distributes it to students, who then fill in the answers and submit the answer sheets.

[1739] The user (teacher) scans the students' answer sheets and uploads them to the server via their device. The scanned answer sheets are saved as image files on the server.

[1740] Answer analysis using AI-OCR

[1741] The server uses AI-OCR technology to analyze the uploaded image of the answer sheet. AI-OCR recognizes the text information in the image and converts it into text data. The converted text data is stored in a database.

[1742] Automatic scoring

[1743] The server compares students' answers, converted into text data, with the correct answers and scores them. The automatic scoring algorithm calculates a score, taking into account partial points and multiple answers. The results are sent to the teacher's device and stored in a database.

[1744] Use of emotion engine

[1745] The server uses an emotion engine to recognize the emotions of users (teachers and students) in real time. The emotion data is used for test creation and feedback.

[1746] For example, if the emotion engine detects stress or impatience in a student, the generative AI will adjust the difficulty of the test accordingly to reduce stress. Similarly, if the emotion engine detects fatigue in a teacher, it will automatically provide candidate questions for test creation, reducing the teacher's burden.

[1747] Data accumulation and individual optimization

[1748] The server stores each student's response data, score data, and emotional data in a database. Based on the stored data, it analyzes each student's level of understanding and trends. It identifies the correct answer rate and incorrect answer patterns for specific questions and generates data to provide individually optimized learning support.

[1749] Based on the analysis results and emotion data, the server generates customized learning materials and supplementary learning materials for each student and sends them to the teacher's device. The teacher can then review these materials and provide them to the students.

[1750] Specific examples

[1751] For example, let's say a teacher wants to create a test on linear equations for second-year junior high school mathematics students. The teacher enters the following criteria into their device: "second-year junior high school mathematics, linear equations, medium difficulty, target average score of 70 points." The server analyzes past test data and uses generative AI to select and edit questions that meet the criteria.

[1752] The generated test is output in PDF format and a download link is sent to the teacher's device. The teacher then distributes the test to the students and collects their answer sheets. The collected answer sheets are scanned and uploaded to the server as an image file.

[1753] The server uses AI-OCR to convert the image of the answer sheet into text data, and automatically grades the answers based on this. The graded results are sent to the teacher's device and stored in a database. If the emotion engine indicates that the student is feeling stressed, the generative AI will automatically adjust the difficulty of the next test.

[1754] Finally, the server analyzes this data, generates customized learning materials based on each student's level of understanding and emotional data, and notifies the teacher, who can then provide guidance and feedback to the student.

[1755] The above is a specific embodiment of the present invention. This system improves the efficiency of test-related tasks, reduces the burden on teachers, and improves students' academic performance. Furthermore, by combining it with an emotion engine, it becomes possible to flexibly respond to the user's emotions.

[1756] The processing flow will be explained below.

[1757] Step 1:

[1758] The user enters the test conditions (subject, grade, scope, difficulty level, target average score, etc.) on the terminal.

[1759] Step 2:

[1760] The terminal transmits the input conditions to the server.

[1761] Step 3:

[1762] The server collects past test data and question collection data from a database.

[1763] Step 4:

[1764] The server automatically creates tests by selecting appropriate questions using a generation AI based on the collected data and input conditions. The generation AI adjusts the question structure based on the set target average score and difficulty level.

[1765] Step 5:

[1766] The server outputs the generated test in PDF or other digital format and sends a download link to the teacher's device.

[1767] Step 6:

[1768] The user checks the notification on their device, downloads the test, and distributes it to students.

[1769] Step 7:

[1770] The user collects the students' test answer sheets, scans them using the terminal, and uploads them to the server as image files.

[1771] Step 8:

[1772] The server analyzes the image of the answer sheet using AI-OCR technology and converts the answers into text data.

[1773] Step 9:

[1774] The server compares the student's textual answers with the correct answers and calculates a score using an automated scoring algorithm that takes into account partial credit and multiple correct answers.

[1775] Step 10:

[1776] The server sends the grading results to the teacher's terminal and stores them in a database.

[1777] Step 11:

[1778] The server uses an emotion engine to analyze emotion data collected from users' (teachers' and students') devices, for example, to detect student stress or a decline in motivation to learn.

[1779] Step 12:

[1780] The server will then adjust the difficulty of the next test based on the emotional data. For example, if a student is feeling highly stressed, the difficulty of the questions will be set lower.

[1781] Step 13:

[1782] The server generates customized feedback based on the emotional data and sends it to the teacher's device, providing specific advice such as, "Student A is under a lot of stress, so please review your teaching method next time."

[1783] Step 14:

[1784] The server stores each student's response data, score data, and emotional data in a database, and uses this data to analyze their level of understanding and learning trends.

[1785] Step 15:

[1786] Based on the analysis results and emotional data, the server generates individually optimized customized teaching materials and supplementary teaching materials and notifies the teacher's device.

[1787] Step 16:

[1788] The user checks the customized learning materials notified on the device and provides them to the students. The next lesson will be taught based on the feedback.

[1789] These are the specific processing steps of the present invention. This series of steps improves the efficiency of test-related tasks, reduces the burden on teachers, and improves students' academic ability. Furthermore, by combining it with an emotion engine, flexible responses based on the user's emotions become possible.

[1790] Example 2

[1791] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1792] Traditionally, creating and grading tests in the educational field has been extremely time-consuming and labor-intensive, placing a heavy burden on teachers. It has also been difficult to grasp each student's learning progress and emotional state in real time and provide optimal feedback to each individual student. These challenges have limited the improvement of educational quality and efficiency.

[1793] 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.

[1794] In this invention, the server includes a means for inputting test conditions, a means for collecting past test data and question collection data, a generation AI means for automatically generating tests based on the collected data and input conditions, a means for digitally outputting the generated tests, a means for scanning students' answer sheets and uploading the images to the server, a means for converting the answers from the uploaded answer sheet images into text data using AI-OCR, a means for automatically scoring the converted text data, a means for saving and analyzing answer data and score data for each student, an emotion engine means for recognizing user emotions in real time, a means for adjusting feedback and the difficulty of the next test based on the emotion data, and a means for generating teaching materials optimized for each student and notifying the teacher. This allows teachers to improve the efficiency and reduce the burden of test-related work, and further enables them to provide optimal learning support to each student.

[1795] "Means for inputting test conditions" refers to the means by which teachers input information such as subject, grade, scope, difficulty level, and target average score via a terminal.

[1796] The "means for collecting past test data and question set data" refers to the means by which the server collects appropriate past test data and question set data from the database.

[1797] "Generative AI means" means AI and related software for automatically generating tests based on collected data and input test conditions.

[1798] "Means for outputting the generated test in a digital format" refers to a means for exporting the test created by the generative AI in a digital file format such as PDF.

[1799] The "means for scanning students' answer sheets and uploading the images to the server" refers to a means for a teacher to scan students' answer sheets and send the image data to the server.

[1800] "Method of converting answers into text data using AI-OCR" refers to AI technology that recognizes character information from the image data of uploaded answer sheets and converts it into text data.

[1801] The "means for automatically scoring answers converted into text data" refers to a means for comparing the student's answers converted into text data with correct answer data and calculating the score.

[1802] "Means for saving and analyzing each student's response data and score data" refers to a means for saving each student's response content and scoring results in a database and analyzing their learning.

[1803] The "emotion engine means" refers to a technology and system for recognizing a user's emotions in real time.

[1804] The "means for adjusting the feedback and the difficulty of the next test based on emotional data" refers to a means for dynamically adjusting the feedback content and the difficulty of the next test based on the emotional data obtained by the emotion engine.

[1805] The "means for generating teaching materials optimized for each student and notifying the teacher" refers to a means for generating teaching materials customized for each student based on the analyzed data and emotional data and notifying the teacher of the same.

[1806] The present invention is a system that reduces the workload in educational settings, automates the process from test creation to grading and feedback, and improves the accuracy of learning support using an emotion engine. A specific embodiment of this system will be described.

[1807] System Overview

[1808] In this system, the user (teacher) inputs test conditions via a terminal, and the system automatically generates a test based on those conditions. The generated test is distributed to students, and the students' answer sheets are scanned and uploaded to a server. The server uses AI-OCR technology to convert the image of the answer sheet into text data, which is then automatically graded. Furthermore, an emotion engine is used to recognize the user's emotions, and feedback and the next test conditions are adjusted based on that. The system also includes a function to generate optimized teaching materials and notify the teacher.

[1809] Hardware and Software

[1810] Hardware: Teacher and student devices (PCs, tablets, etc.), servers, scanners

[1811] Software: Generative AI models (e.g., GPT-3), AI-OCR technology (e.g., Google Cloud Vision OCR), emotion engine

[1812] Usage and Examples

[1813] Test condition input

[1814] The user (teacher) enters conditions such as "subject," "grade," "scope," "difficulty level," and "target average score" into the input form on the terminal.

[1815] Examples:

[1816] If a teacher wants to "generate a test on linear equations for second-year junior high school mathematics students," they would enter "Subject: Mathematics," "Grade: Second-year junior high school," "Range: Linear equations," "Difficulty: Medium," and "Target average score: 70 points."

[1817] Test Data Collection and Generation

[1818] The server receives input conditions, collects past test data and question collection data from a database, and then sends prompts to the generative AI model to automatically generate tests.

[1819] Example prompt sentence:

[1820] "Generate a test on linear equations for second-year junior high school mathematics students. The conditions are as follows: difficulty level: medium, target average score: 70 points, test area: Chapter 5 of the textbook."

[1821] Test distribution and answer collection

[1822] The user (teacher) checks the generated test and distributes it to the students. After the students answer, the answer sheets are scanned and uploaded to the server via their terminal.

[1823] Answer analysis using AI-OCR

[1824] The server analyzes the image data of the uploaded answer sheet using AI-OCR technology and converts it into text data.

[1825] Automatic grading and feedback

[1826] The server compares the converted answer data with the correct answer data and automatically scores the answers. The results are sent to the teacher's device and stored in a database.

[1827] Use of emotion engine

[1828] The server uses an emotion engine to recognize the emotions of users (teachers and students) in real time. For example, if a student is feeling stressed, the system can adjust the difficulty of the next test, or if the system detects that a teacher is tired, it can automatically provide appropriate candidate questions.

[1829] Individually optimized learning support

[1830] The server analyzes the accumulated answer data and emotion data, and generates teaching materials optimized for each student based on their level of understanding. These generated teaching materials are then sent to the teacher's device.

[1831] This system allows teachers to streamline test-related tasks and reduce their workload, while also enabling them to provide optimal learning support to students. Furthermore, the emotion engine allows for flexible responses that take into account the user's emotions. As a result, this system not only improves the quality of education, but also contributes to the efficiency of classrooms.

[1832] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1833] Program processing steps

[1834] Step 1: Enter the test conditions

[1835] The user (teacher) uses the terminal to input conditions such as subject, grade, scope, difficulty level, and target average score. Specifically, the user fills in the input form on the GUI with "Subject: Mathematics," "Grade: 2nd year junior high school student," "Scope: Linear equations," "Difficulty level: Medium," and "Target average score: 70 points," and then presses the send button. At this time, these conditions are saved in the terminal's memory as input data.

[1836] Input: Subject, grade, range, difficulty level, target average score

[1837] Output: Test condition data sent to the server

[1838] Step 2: Sending test condition data

[1839] The device sends the entered test condition data to the server using an HTTP POST request. The specific data sent is in JSON format.

[1840] Input: Test condition data entered by the user

[1841] Output: Condition data in JSON format passed to the server

[1842] Step 3: Collect test data

[1843] The server receives the test condition data and accesses the database to collect past test data and problem set data. For example, it searches for problems that match the condition "Mathematics, Junior High School 2nd Grade, Linear Equations." It then uses a database query to extract the relevant data.

[1844] Input: Test condition data received by the server

[1845] Output: Past test data and question bank data retrieved from the database

[1846] Step 4: Automatically generate tests

[1847] The server sends prompts to the generative AI model based on the collected data and input conditions. The generative AI model (e.g., GPT-3) generates tests based on the prompts.

[1848] Specific prompt:

[1849] "Generate a test on linear equations for second-year junior high school mathematics students. The conditions are as follows: difficulty level: medium, target average score: 70 points, test area: Chapter 5 of the textbook."

[1850] Input: Past test data from the database, test conditions entered by the user, and prompts for the generative AI model

[1851] Output: Generated test data (questions, answer sheets, etc.)

[1852] Step 5: Generated Test Output

[1853] The server exports the generated test data in PDF or other digital file format, and the generated PDF file is saved in the server's storage.

[1854] Input: Test data generated by the generative AI model

[1855] Output: Test file in PDF format

[1856] Step 6: Notification of generated tests

[1857] The server generates a URL for the generated test file and notifies the user (teacher) of the URL. If necessary, it also generates and sends a QR code for online distribution.

[1858] Input: Generated test file in PDF format

[1859] Output: URL link and QR code

[1860] Step 7: Distribute the test

[1861] The user (teacher) reviews the generated test and distributes it to students by printing it on paper, or by email or via an online platform.

[1862] Input: URL link and QR code

[1863] Output: Tests distributed to students

[1864] Step 8: Collect answer sheets

[1865] The user (teacher) scans the answer sheets on which the students have completed the questions and imports them into the device. An image file of the answer sheet is generated through the scanner.

[1866] Input: Student's completed answer sheet

[1867] Output: Image file of the answer sheet captured on the device

[1868] Step 9: Upload your answer sheet

[1869] The device uploads the scanned image data of the answer sheet to the server using an HTTP POST request.

[1870] Input: Image file of answer sheet

[1871] Output: Image data of the answer sheet sent to the server

[1872] Step 10: Answer analysis using AI-OCR

[1873] The server receives the uploaded image data of the answer sheet and converts the character information into text data using AI-OCR technology. Specifically, the AI-OCR system recognizes the characters in the image and converts them into text format. For example, a mathematical formula such as "12x + 8 = 4y" is obtained as text data.

[1874] Input: Image data of the answer sheet

[1875] Output: Answers converted to text data

[1876] Step 11: Automated scoring

[1877] The server receives the answers converted into text data, compares them with the correct answer data, and automatically scores them. For example, if the answer is "12x + 8 = 4y," the server compares it with the correct answer data and calculates the score taking into account partial points and point allocation.

[1878] Input: Textual answer data, correct answer data

[1879] Output: Scoring results (score data)

[1880] Step 12: Notification of Scoring Results

[1881] The server notifies the user (teacher) of the grading results and simultaneously stores them in a database.

[1882] Input: Scoring results

[1883] Output: Marking results sent to the teacher's device

[1884] Step 13: Emotion Recognition

[1885] The server uses an emotion engine to recognize the emotions of users (teachers and students) in real time. For example, it analyzes the user's emotional state (stress, frustration, fatigue, etc.) from data acquired through a camera or microphone.

[1886] Input: Emotion data obtained from a camera or microphone

[1887] Output: Recognized emotion data

[1888] Step 14: Adjusting Feedback

[1889] Based on the recognized emotion data, the server sends prompts to the generative AI model to provide feedback and adjust the difficulty of the next test. For example, if the emotion engine detects stress, it will instruct the generative AI to lower the difficulty of the next test.

[1890] Input: Recognized emotion data

[1891] Output: Adjusted feedback and difficulty of next test

[1892] Step 15: Individually optimized learning support

[1893] The server analyzes each student's response data, score data, and emotion data to generate optimized teaching materials, which are then sent to the teacher's device and provided to the students.

[1894] Input: Answer data, score data, emotion data

[1895] Output: Optimized learning materials

[1896] These are the specific processing steps of this system. This system aims to reduce the burden on teachers and provide a more effective and efficient educational environment.

[1897] (Application example 2)

[1898] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1899] Physical stores are seeking training methods to efficiently improve their employees' customer service skills. Conventional training methods rely on subjective evaluations by trainers, making it difficult to provide objective feedback and resulting in inconsistencies in the quality and efficiency of training. Furthermore, there is a lack of means to recognize employees' emotions and stress levels and provide appropriate feedback. Given this background, a system is needed to support employee skill improvement and increase customer satisfaction.

[1900] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1901] In this invention, the server includes: means for inputting test conditions; means for collecting past test data and question collection data; AI generation means for automatically generating tests based on the collected data and input conditions; means for digitally outputting the generated tests; means for scanning images of answer sheets and training data and uploading them to the server; means for converting the uploaded images into text data using AI-OCR; means for automatically scoring the converted text data; an emotion recognition engine for recognizing user emotions in real time; means for providing appropriate feedback to users based on the recognized emotion data; means for saving and analyzing answer data, score data, and emotion data for each user; and means for generating a report that provides feedback on users' strengths and areas for improvement based on the response content and emotion data. This enables objective evaluation of employees' customer service skills and appropriate feedback based on the emotion data.

[1902] Definition of Terms

[1903] "Test conditions" refers to the information that a user enters to generate a specific test or training, specifically settings such as subject, target grade, question scope, difficulty level, and target average score.

[1904] "Past test data and question bank data" refers to data collected from previously administered tests or existing question banks and used for test creation and training.

[1905] "Generative AI means" refers to artificial intelligence technology that automatically generates appropriate test questions or training tasks based on input conditions.

[1906] "Digital output means" means a means for providing the generated test or training assignment in PDF or other digital file format.

[1907] "Means for scanning images of answer sheets and training data and uploading them to a server" refers to a means for capturing answer sheets and training records completed by students or employees as digital images and transmitting them to a server via a network.

[1908] "Means for converting scanned image data into text data using AI-OCR" refers to a means for converting scanned image data into text data using optical character recognition technology.

[1909] An "automatic scoring means" is a means that has an algorithm that calculates scores by comparing the answers converted into text data with the correct answer data.

[1910] An "emotion recognition engine" refers to technology that identifies emotions in real time from a user's facial expressions, voice, text data, etc.

[1911] The "means for providing feedback" is a means for providing appropriate comments and advice to the user based on the recognized emotion data and analysis results.

[1912] The "means for storing and analyzing response data, score data, and emotional data for each user" refers to a means for storing performance data and emotional data for each user and analyzing them.

[1913] The "means for generating a feedback report" is a means for organizing the user's strengths and areas for improvement based on the response content and emotional data, and providing them in the form of a report.

[1914] MODE FOR CARRYING OUT THE INVENTION

[1915] The present invention relates to an emotion recognition training system for helping store employees improve their customer service skills. This system allows employees to receive real-time feedback using a smartphone or smart glasses while undergoing customer service training. Detailed embodiments of the present invention are described below.

[1916] Hardware and software used

[1917] Hardware

[1918] Smartphone

[1919] Smart Glasses

[1920] server

[1921] software

[1922] AI-OCR engine (e.g. Tesseract OCR)

[1923] Emotion recognition engine (e.g. Microsoft Azure Emotion API)

[1924] Generative AI methods (e.g., text generation AI)

[1925] Database Management Systems

[1926] System configuration

[1927] User terminal

[1928] Users (employees) use smartphones or smart glasses to conduct customer service training. At the start of a training session, users input training conditions (e.g., customer service scenario, target skills, training time, etc.). This allows users to focus on the training scenario.

[1929] server

[1930] The server manages the training session and processes the data in the following manner.

[1931] 1. Enter and collect test conditions

[1932] Collect training conditions entered by the user.

[1933] Past training data and question collection data are collected from the database.

[1934] 2. Training and generation using generative AI models

[1935] Training content is automatically generated based on collected data and conditions.

[1936] The generated training is output in digital format and delivered to the user's smart device.

[1937] 3. Scan and upload your answers and training data

[1938] The data collected by the user during training (audio, video, text, etc.) is scanned and uploaded to the server.

[1939] 4. Text data conversion using AI-OCR

[1940] The uploaded data is converted into text data using an AI-OCR en...

Claims

1. a means for inputting test conditions; A means of collecting past test data and question collection data, A generative AI means for automatically generating tests based on collected data and input conditions; means for outputting the generated tests in digital form; A means to scan students' answer sheets and upload the images to a server; A method of converting the answers into text data using AI-OCR from the uploaded image of the answer sheet, A means for automatically scoring answers converted into text data; A system that includes a means for storing and analyzing response data and score data for each student.

2. The system of claim 1 includes an algorithm that takes into account partial points and multiple answers when converting an image of an answer sheet into text data using AI-OCR.

3. 2. The system according to claim 1, further comprising means for providing optimal learning support to each student based on the analyzed data.

Images