system

Abstract

Provide a system. 【Solution means】 Means for receiving user input and analyzing the intention of the question based on the input, Means for referring to past test data and generating problem ideas based on the analysis results, Means for creating a specific problem statement from the generated problem ideas and checking logical consistency, Means for estimating the difficulty level of the problem statement, Means for analyzing the ideal answer and creating scoring criteria, Means for analyzing voice input and converting it into text data, Means for automatically generating problems using a generation AI model, Means for providing feedback to the user based on the evaluation results, A system including the above.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, and includes steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance that responds to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the conventional descriptive test question creation and scoring work, high-level expertise is required, and since it is often carried out by a limited number of people, there is a problem that the workload is extremely high. Also, it is necessary to ensure confidentiality in the process of question creation and scoring, and it has been considered difficult to operate the test efficiently and fairly. Furthermore, in order to more accurately identify excellent talents, it is required to maintain consistency in evaluation criteria.

Means for Solving the Problems

[0005] This invention provides a system that includes means for receiving and analyzing the intent of a question presented by a user, and means for generating question ideas and creating question texts based on the analysis results using past test data. Furthermore, by estimating the difficulty level of the question text and creating scoring criteria by analyzing ideal answers presented by the user, the system aims to standardize and streamline the test questions and scoring process. This makes it possible to conduct fair tests without relying on specialized knowledge, and to efficiently select outstanding personnel while ensuring the confidentiality and fairness of the test.

[0006] "User input" refers to the information and requests that users provide to the system, and this data is used by the system to analyze the intent behind the questions.

[0007] "Analysis means" refers to a method or apparatus for receiving user input and analyzing the information contained therein to identify the intent behind the question and related topics.

[0008] "Past exam data" refers to a collection of information about previously asked exam questions and their results, which is used to generate new question ideas.

[0009] A "problem idea" is a general concept or outline of the test questions tailored to the user's objectives, and will later be used to create the specific question text.

[0010] "Problem statement creation means" refers to a method or apparatus for creating a specific problem statement based on a generated problem idea.

[0011] "Logical consistency" means that the problem statement created has a coherent and logical structure without errors.

[0012] A "difficulty level estimation method" is a method or device for evaluating and quantifying the degree of challenge a problem presents to the solver, based on the content of the problem statement.

[0013] An "ideal answer" is a model answer as envisioned by the question setter, and is a specific answer used to create the scoring criteria.

[0014] "Scoring criteria" are indicators or standards used to evaluate answers, and are guidelines for scoring that are set based on ideal answers. [Brief explanation of the drawing]

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

Mode for Carrying Out the Invention

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

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

[0018] In the following embodiments, the labeled processor (hereinafter simply referred to as "processor") may be one arithmetic unit or a combination of a plurality of arithmetic units. Also, the processor may be one type of arithmetic unit or a combination of a plurality of types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

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

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

[0021] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0022] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0023] [First Embodiment]

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

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

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

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

[0028] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0029] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0030] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

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

[0032] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

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

[0034] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

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

[0036] The present invention relates to an AI-powered system for creating and grading written examination questions, and its embodiments are described below.

[0037] overview

[0038] This system consists of three main components: a server, a terminal, and a user. An AI model running on the server creates questions based on the user's intended question, and also creates scoring criteria and scores the test takers' answers.

[0039] Details of each component

[0040] server

[0041] Problem creation function

[0042] The server receives the user's submitted question intent and analyzes it using natural language processing (NLP) techniques to extract relevant topics and keywords. Based on this information, an AI model generates appropriate question ideas while referring to a database of past exams. It then generates specific question texts from the selected ideas and checks their logical consistency and readability. It also has a function to evaluate the characteristics of the questions and estimate their difficulty level.

[0043] Scoring support function

[0044] When the server receives an ideal answer from a user, it analyzes it to create scoring criteria. Based on these criteria, it scores the examinee's written answer. The AI ​​model evaluates the accuracy and relevance of the answer through syntactic and semantic analysis.

[0045] terminal

[0046] Interface function

[0047] The terminal provides an interface for users to access the system, input question intents, check questions and scoring criteria, and view scoring results. The UI is intuitive and designed for easy user operation.

[0048] User

[0049] Providing the intent behind the questions

[0050] Users provide the server with the abilities they wish to evaluate and the themes they wish to address via their device, and request the creation of problems based on this information.

[0051] Result confirmation and feedback

[0052] Users can view the questions, scoring criteria, and test results provided by the server through their terminals and provide feedback for improvement as needed.

[0053] Presentation of specific examples

[0054] For example, if a user wants to create an exam question on the theme of "assessing economic analytical skills related to sustainability," the user sends this intention to the server using their device. The server uses AI to identify relevant topics and generates questions such as, "Discuss the impact of sustainability policies on economic growth." The user then uploads their own example answer, and the server forms scoring criteria based on that answer and evaluates the answers of actual test takers.

[0055] This process enables efficient and fair testing without relying on specialized knowledge.

[0056] The following describes the processing flow.

[0057] Step 1:

[0058] Users use their devices to input the intent behind the questions and the objectives of the exam. By specifically describing the themes and abilities they wish to evaluate, the subsequent AI analysis will proceed more smoothly.

[0059] Step 2:

[0060] The terminal sends user input to the server. The transmitted data is stored on the server and prepared for use in subsequent analysis processes.

[0061] Step 3:

[0062] The server analyzes the received question intent using natural language processing (NLP) techniques. This extracts relevant keywords and topics, providing guidance for creating questions.

[0063] Step 4:

[0064] The server references a database of past exams based on the extracted keywords, and the AI ​​model generates appropriate problem ideas. This process verifies whether the ideas match the user's objectives.

[0065] Step 5:

[0066] The server generates a specific problem statement based on the selected problem idea. The generated problem statement is automatically checked for logical consistency and appropriate language.

[0067] Step 6:

[0068] The server estimates the difficulty level of the generated problem based on evaluation metrics. This numerically represents how challenging the problem is for the test-taker.

[0069] Step 7:

[0070] Users upload ideal answer examples to the server via their devices. These answers serve as the basic information necessary for creating the scoring criteria.

[0071] Step 8:

[0072] The server analyzes the user's ideal answer and automatically creates scoring criteria based on it. These criteria serve as a guideline for evaluation when test-takers answer the questions.

[0073] Step 9:

[0074] The device presents the completed problem statement and grading criteria to the user. At this stage, the user can review the content and provide corrections or feedback if necessary.

[0075] Step 10:

[0076] The server collects test-takers' answers during the exam and prepares them for analysis. This data is used for subsequent scoring.

[0077] Step 11:

[0078] The server evaluates the test-taker's answers based on scoring criteria and uses AI to determine the accuracy and relevance of each answer. The results are calculated as a score.

[0079] Step 12:

[0080] The terminal displays the scoring results to the user and provides information to obtain feedback useful for exam administration. The user evaluates the system's effectiveness through this information and uses it to create future exams.

[0081] (Example 1)

[0082] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0083] Traditional essay-style examination question creation and grading processes require significant time and expertise, making it difficult to ensure efficiency and fairness. In particular, consistently establishing standards to ensure the quality of exam questions and fairly evaluate examinees' responses is challenging.

[0084] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0085] In this invention, the server includes means for receiving user input and analyzing the intent based on the input, means for referring to past test information and generating problem ideas based on the analysis results, and means for creating specific problems from the generated ideas and checking their logical consistency. This makes it possible to improve the efficiency and fairness of the question creation and scoring process for written examinations.

[0086] "User input" refers to information and instructions provided by system users via their terminals, and serves as the starting point for creating exam questions and evaluating answers.

[0087] "Means for analyzing intent" refers to a system that uses natural language processing technology to analyze information provided by users and identify their purpose and specific needs.

[0088] "Past exam information" refers to historical questions and sample answers stored in existing exam databases, which are used as reference when creating new questions.

[0089] "Means of generating problem ideas" refers to the process of devising an outline of a problem, including related themes, based on the analyzed intent.

[0090] "Methods for creating specific problems" refer to the steps taken to construct a detailed and clear problem statement based on the generated ideas.

[0091] A "means of checking logical consistency" refers to a mechanism for evaluating whether a problem statement is logically consistent and free from contradictions.

[0092] An "ideal answer" is a model answer provided by a user, serving as a reference example for forming the scoring criteria.

[0093] "Means of creating evaluation criteria" refers to the process of establishing criteria for fairly evaluating test-takers' answers, based on ideal responses.

[0094] "Syntactic analysis" refers to the process of analyzing and understanding a test-taker's answers based on their grammatical structure.

[0095] "Semantic analysis" is the process of interpreting the content of a test-taker's answers and evaluating their intent and relevance.

[0096] "Means for evaluating responses" refers to a system that measures the quality of test-takers' responses based on syntactic and semantic analysis.

[0097] Specific embodiments of this invention will now be described.

[0098] This system consists of three main components: servers, terminals, and users.

[0099] server

[0100] The server operates on a cloud server with powerful computing capabilities. This server receives the intent of the question submitted by the user through their terminal and analyzes it using natural language processing (NLP) technology. The server has a generative AI model installed that accesses a database containing past exam information and extracts relevant topics based on the input intent. The AI ​​model then generates question ideas and creates specific question texts based on this. This process checks the logical consistency of the questions and selects the most suitable questions. Furthermore, it creates evaluation criteria based on the ideal answers received from the user and scores the answers obtained by the test taker. Mechanisms are in place to evaluate answers in detail through syntactic and semantic analysis.

[0101] terminal

[0102] The terminal provides the interface for users to access the system. This includes PCs and tablets used by users, which connect to the server via a web browser. The terminal is designed so that users can input the intent of the questions and ideal answers, and then check the questions, evaluation criteria, and scoring results generated by the server. The interface is also intuitive and designed with ease of operation in mind.

[0103] User

[0104] Users are the ones who use terminals to input intentions and themes in order to support the process of creating and grading exam questions. For example, if a user wants to create an exam question on the theme of "evaluating economic analytical skills related to sustainability," they use their terminal to send that intention to the server. Based on this, the server generates a question such as "Discuss the impact of sustainability policies on economic growth." The user then uploads their ideal answer to the server, and the server evaluates the test-taker's answer based on scoring criteria derived from that answer. This entire process ensures highly efficient and fair examination administration.

[0105] Example of a prompt

[0106] "Please create test questions to assess economic analytical skills related to sustainability."

[0107] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0108] Step 1:

[0109] Users input the intent behind the questions using their devices. Specifically, they use a web interface input form to describe in detail the abilities or themes they wish to evaluate. The data entered is primarily in text format and is transferred from the device to the server.

[0110] Step 2:

[0111] The server analyzes the question intent received from the user using natural language processing technology. This analysis extracts the intended evaluation points and topics. During the processing, the generative AI model analyzes keywords and main themes from the text and identifies relevant information. As a result, the analyzed data forms the basis for question generation.

[0112] Step 3:

[0113] The server generates question ideas based on analyzed topic information and references past exam data. It utilizes a generative AI model to suggest relevant questions and then creates specific question texts based on these suggestions. The output data, including a summary of the questions and specific question formats, is prepared for further evaluation.

[0114] Step 4:

[0115] The server checks the logical consistency and difficulty level of the generated problem statements. The AI ​​model analyzes whether the problems are consistent and of appropriate difficulty. Here, the quality of the problems is checked according to the set criteria, and the optimal content is selected. The checked problem statements are output and ready for user review.

[0116] Step 5:

[0117] Users review the question and its evaluation criteria presented by the server via their terminal. If necessary, users can request revisions to the content. A crucial step in the terminal review process is re-examining the question and approving the final question content.

[0118] Step 6:

[0119] Users input ideal answer examples from their terminals into the server. These example answers are required to be concise and appropriate, based on the intent of the question. The input data is then used to generate scoring criteria.

[0120] Step 7:

[0121] The server analyzes the ideal answers provided by users and creates scoring criteria based on them. It utilizes a generative AI model to identify key elements of the answers. Syntactic and semantic analysis are used to extract evaluation points for the answers and establish fair scoring criteria. These criteria are used as indicators when evaluating test-takers' answers.

[0122] Step 8:

[0123] The server receives the test taker's answers and evaluates them based on pre-defined scoring criteria. Using syntactic and semantic analysis, it determines how closely the test taker's answers resemble the ideal answers. A final evaluation result is generated and provided to the user as feedback. This feedback can be used to review the test results and improve future test designs.

[0124] (Application Example 1)

[0125] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0126] In educational settings, creating and grading essay-style examinations is a task that requires specialized knowledge and time, placing a significant burden on educators. Furthermore, automation is difficult, and many challenges remain in ensuring efficient and fair evaluation. This invention aims to solve these problems and improve the quality of education.

[0127] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0128] In this invention, the server includes means for receiving user input and analyzing the intent behind the questions, means for generating question ideas by referring to past test data, and means for automatically generating questions using a generation AI model. This enables efficient and fair creation and scoring of written test questions on an educational equipment platform.

[0129] "User input" refers to the information and instructions that users provide to the system.

[0130] "Exam intent" refers to the user's intentions regarding the purpose and theme of the exam questions they wish to create.

[0131] "Analysis" refers to the process of analyzing input information and extracting necessary data and patterns.

[0132] "Past exam data" refers to information about previously used exam questions and their results.

[0133] "Problem idea" refers to a hypothesis or draft of an exam question based on the examiner's intent.

[0134] A "generative AI model" refers to an algorithm or system that uses artificial intelligence to automatically create problems.

[0135] "Specific problem statement" refers to the written content of the questions used in the exam.

[0136] "Logical consistency" refers to a state in which the problem statement and the answer are logically consistent and free from contradictions.

[0137] "Difficulty level" refers to a measure that indicates how challenging an exam question is.

[0138] An "ideal answer" refers to a model and optimal response to an exam question.

[0139] "Scoring criteria" refers to the standards or criteria used to evaluate a test-taker's answers.

[0140] "Voice input" refers to a method of capturing audio as data through a device such as a microphone.

[0141] "Text data" refers to character information recorded in a format that can be processed by a computer.

[0142] An "educational equipment platform" refers to a hardware and software environment used for learning and educational purposes.

[0143] "Evaluation" refers to the process of reviewing the test takers' answers and performance and determining their value and quality.

[0144] "Feedback" refers to information and advice for improvement provided to users or test takers based on evaluation results.

[0145] The system for carrying out this invention operates on an educational equipment platform. Servers, terminals, and users are the main components.

[0146] On the server, a generative AI model is used to automatically generate questions based on the user's intended question. Specifically, the user communicates the exam theme to the server via voice or text input. For voice input, the server uses the Google® Speech-to-Text API to convert the voice data into text data. Next, it analyzes the intended question using NLP technology and refers to relevant past exam data. In this data analysis process, the AI ​​model infers and generates appropriate question ideas.

[0147] The terminal provides the user interface for accessing this system. Through an intuitive UI, users can review the generated problem statements and scoring criteria and suggest revisions as needed.

[0148] Users aim to improve the quality of education through this system. In particular, voice input allows for the provision of question intent under various conditions, enabling immediate question creation and grading in different situations. This flexibility helps to meet the diverse needs of educational settings and reduces the burden on educators.

[0149] For example, if a user wants to create an exam question on "environmental issues," the server will automatically generate a question such as "Discuss the relationship between environmental issues and economic policy." OpenAI's GPT series is used for analysis, enabling smooth question generation. An example of a prompt message is: "Generate an exam question on 'environmental issues' using OpenAI GPT-3. Please grade the answer considering semantics and syntax."

[0150] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0151] Step 1:

[0152] Users input the exam question's theme and intent via their device. Input can be in voice or text format; voice input is converted to text using the Google Speech-to-Text API. This process involves analyzing the voice data and converting it to text data, with the output being text data including the intent behind the question.

[0153] Step 2:

[0154] The server analyzes the input question intent using natural language processing (NLP) techniques. Specifically, it uses an NLP library (e.g., spaCy) to syntactically analyze the text data and extract important keywords and themes. This process outputs structured data of the analyzed question intent.

[0155] Step 3:

[0156] The server generates problem ideas based on extracted keywords and themes, utilizing a generative AI model (e.g., OpenAI GPT-3). Here, the AI ​​model combines relevant information to create a draft problem that matches the intended question. This process outputs a preliminary draft of the generated problem.

[0157] Step 4:

[0158] Based on the generated problem ideas, the server creates a specific problem statement. The automatically generated problem statement is then checked for logical consistency and difficulty. At this stage, the text data of the problem statement is output, along with evaluation values ​​for its consistency and difficulty.

[0159] Step 5:

[0160] The server analyzes the ideal answers provided by the user and creates scoring criteria. At this stage, NLP techniques are again used to analyze the syntax and meaning of the ideal answers. This process outputs structured data to be used as scoring criteria.

[0161] Step 6:

[0162] Users review the completed question text and scoring criteria on their devices and suggest revisions if necessary. Based on user feedback, data is then corrected, and the final revised question text and scoring criteria are output.

[0163] Step 7:

[0164] In educational settings, test takers submit their answers, and the server collects these answers. The collected answers are evaluated based on previously established scoring criteria. This results in an output score for each answer.

[0165] Step 8:

[0166] Based on the evaluation results, the server provides feedback to the user via the terminal. In this step, text data showing the results and areas for improvement is generated and output as feedback in a format that the user can review.

[0167] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0168] This invention aims to achieve more sophisticated exam management by combining an emotion engine with an AI-powered system for creating and grading written exam questions. This emotion engine is characterized by its ability to analyze the user's emotional state and adjust each process based on that information.

[0169] overview

[0170] The system consists of a server, terminals, and an emotion engine. The emotion engine analyzes the user's emotional state and uses the resulting data to adjust each process accordingly.

[0171] Details of each component

[0172] server

[0173] Problem creation support

[0174] The server processes the question intent received from the user, along with data including the sentiment analysis results from the sentiment engine, and extracts relevant topics. This process ensures that the question intent is appropriately interpreted based on the user's emotions, helping the AI ​​model accurately generate question ideas.

[0175] Grading support and feedback

[0176] By incorporating emotional information from the emotion engine into the evaluation of test responses, it is possible to optimize scoring criteria and feedback content. This allows for more flexible and personalized feedback.

[0177] terminal

[0178] Collection of interface and sentiment data

[0179] The terminal provides an interface for the user to access the system and collects data to obtain the user's emotional state. This data is analyzed by an emotion engine and sent to the server.

[0180] Emotional Engine

[0181] Emotional analysis and process adjustment

[0182] The emotion engine analyzes the user's emotions based on data transmitted from the device and reflects the results in each process. Specifically, it is used to adjust the difficulty level when creating questions and to adaptively adjust the scoring criteria.

[0183] Presentation of specific examples

[0184] For example, if a user creates a test to "assess critical thinking skills regarding environmental issues," and the emotion engine analyzes that the user is experiencing stress, the server will take that information into account and present questions with slightly adjusted difficulty levels. In this way, tests that take into account the user's emotions and mental state are constructed. After the test, when grading the test-takers' answers, the feedback is also adjusted based on the data obtained from the emotion analysis, helping test-takers accept the results in a way that is more easily understood and accepted.

[0185] This format makes it possible to provide users and test takers with a more personalized and effective testing experience.

[0186] The following describes the processing flow.

[0187] Step 1:

[0188] Users use a terminal to input the intent behind the questions regarding the themes and abilities they wish to evaluate. The input requires a detailed description of the test's purpose and target areas.

[0189] Step 2:

[0190] The device collects emotional data simultaneously with user input. This emotional data includes information estimated from factors such as keyboard typing speed and eye movements.

[0191] Step 3:

[0192] The device transmits question intent and emotional data to the server. The transmission is done in real time and is designed to respond to changes in emotions.

[0193] Step 4:

[0194] The server analyzes the received question intent using natural language processing (NLP). During this process, it references sentiment data provided by the sentiment engine to supplement the interpretation with one that reflects the user's mental state.

[0195] Step 5:

[0196] Based on the information obtained from the analysis, the server generates multiple problem ideas using an AI model. It then utilizes emotional data to select the idea that best suits the user's emotional state from among the generated ideas.

[0197] Step 6:

[0198] The server creates specific problem statements based on the selected problem ideas. During this process, the difficulty level is adjusted considering the emotion engine data, ensuring the content is adapted to the user's state.

[0199] Step 7:

[0200] The device presents the user with the completed problem statement and scoring criteria that reflect sentiment data. The user can review the content and provide feedback as needed.

[0201] Step 8:

[0202] When the test is administered, the device collects not only the test-taker's answers but also emotional data from the test. This data, along with the answers, is sent to the server.

[0203] Step 9:

[0204] The server uses the collected responses and sentiment data to perform scoring. The sentiment engine assists in evaluating the responses and calculates an appropriate score based on the scoring criteria.

[0205] Step 10:

[0206] The device presents test takers and users with emotionally-based feedback along with their scoring results. This feedback includes advice that reflects emotional elements to aid the test taker's understanding.

[0207] (Example 2)

[0208] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0209] Traditional exam creation and scoring systems have the drawback of not considering the emotional state of users and test-takers, making it difficult to provide a personalized exam experience. Furthermore, because the questions and scoring are standardized, the diverse learning styles and emotions of test-takers do not significantly influence the results.

[0210] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0211] In this invention, the server includes means for collecting and analyzing user emotional data, means for adjusting the difficulty level and scoring criteria of questions based on the emotional analysis results, and means for adjusting the content of feedback based on the emotional analysis results. This enables the creation of personalized tests and flexible scoring and feedback that are tailored to the emotional state of the user and test taker.

[0212] "User input" refers to the information and instructions that users provide to the system, including the intent behind the questions necessary for creating the exam.

[0213] "Exam intent" refers to the purpose and criteria that the user expresses when creating exam questions.

[0214] "Past data" refers to information and statistics collected previously, which are used in the generation and evaluation of exam questions.

[0215] An "idea" refers to a concept or theme for giving concrete form to the exam questions.

[0216] "Checking sentence consistency" is the process of verifying whether the generated problem sentences are grammatically and logically correct.

[0217] "Difficulty level" is an indicator that shows the complexity and ease of solving exam questions.

[0218] "Criteria for analyzing answers" are standard guidelines used to evaluate the answers of test-takers.

[0219] "Emotional data" refers to information that indicates a user's emotional state, and is collected and used by the system.

[0220] "Sentiment analysis" is the process of evaluating collected emotional data to understand the user's emotional state.

[0221] "Adjusting the feedback content" refers to the process of optimizing responses and comments to test-takers' answers based on the results of sentiment analysis.

[0222] This invention provides a test creation and scoring support system that utilizes AI technology, including user emotion analysis. This system consists of a server, terminals, and emotion analysis devices, which function through mutual cooperation.

[0223] The terminal provides the user with an interface for creating exams. Through this interface, the user inputs the intent behind the questions, and simultaneously, data for sentiment analysis is collected. For example, keyboard input speed, mouse movements, or direct user feedback can be used. The data obtained in this way is essential for capturing the user's real-time emotional state.

[0224] The server receives data transmitted from the terminal and analyzes the user's input regarding the intent behind the questions and emotional data. Emotional analysis is performed by an emotional analysis device, utilizing natural language processing and machine learning techniques. This analysis determines the user's stress level and agitation state, which is used to adjust the difficulty of the test. The server then uses a generative AI model to extract relevant topics that align with the intent behind the questions and creates questions based on those topics.

[0225] For example, if a user creates a test to "evaluate critical thinking skills regarding environmental issues," the system will determine the user's stress level through sentiment analysis and adjust the difficulty level appropriately. This allows for a flexible test format tailored to the user's state. Furthermore, when generating test questions, the AI ​​model applies appropriate prompts based on the legal provision that "the user's emotions should be taken into consideration using a generative AI model."

[0226] Thus, the present invention enables the creation and scoring of exams that reflect the individual emotional states of users and test takers, thereby providing a personalized educational environment.

[0227] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0228] Step 1:

[0229] The terminal receives input from the user and obtains the user's intent for the question. Simultaneously, it collects emotional data through sensors and interfaces. Specifically, it records the themes and keywords entered by the user, as well as the emotional state inferred from keyboard typing speed and mouse movements. This input data is collected as basic information necessary for analysis on the server.

[0230] Step 2:

[0231] The terminal transmits the collected question intent and emotional data to the server. After receiving this data, the server passes it to an emotional analysis device for analysis. Using natural language processing technology and machine learning algorithms, the device evaluates the user's emotional state and determines whether the user is experiencing stress. The results of this analysis are then used in the subsequent process of adjusting the difficulty level of the test questions.

[0232] Step 3:

[0233] The server takes the sentiment analysis results into account and uses a generative AI model to extract relevant topics that match the intent of the question. Specifically, the AI ​​model uses prompt sentences to generate specific topics based on the given theme. These prompt sentences are such as, "The AI ​​will generate questions based on the user's sentiment about the selected topic."

[0234] Step 4:

[0235] The server generates test questions that take sentiment data into account, based on the extracted topics. If the analysis indicates that the user is experiencing stress, the difficulty level of the questions is appropriately adjusted before generation. In this process, the question text is created based on the topics generated by the AI, and a consistency check is performed.

[0236] Step 5:

[0237] The server verifies the difficulty level and logical consistency of the generated exam questions, and then provides the user with the exam questions and predicted scoring criteria. It also provides flexible editing tools through an interface, allowing users to modify the questions and scoring criteria as needed.

[0238] Step 6:

[0239] After completing the exam, the user submits their answers to the server. The server evaluates the received answers according to scoring criteria and adjusts the feedback based on the results of sentiment analysis. In particular, if the server determines that the test-taker is experiencing excessive stress, the feedback is optimized to emphasize positive aspects.

[0240] (Application Example 2)

[0241] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0242] Traditional exam question creation and scoring systems have a problem in that they do not take into account the emotional state of test-takers, and in particular, they do not adequately consider the impact of stress and anxiety on test-takers' performance. As a result, there is a problem in conducting flexible and individualized exams that respond to the diverse needs and emotions of learners.

[0243] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0244] In this invention, the server includes means for receiving user input and analyzing the intent behind the questions based on the input; means for referring to past test data and generating question ideas based on the analysis results; means for creating specific question sentences from the generated question ideas and checking their logical consistency; and means for analyzing the emotional state of the user using a device for recognizing emotions and adjusting the difficulty level and content of the questions based on that analysis. This makes it possible to provide a flexible and individualized test that takes into account the emotional state of the test taker.

[0245] "Means for receiving user input and analyzing the intent behind the questions based on that input" refers to a process of receiving input data from users, analyzing that data, and interpreting what purpose and direction questions should be created in.

[0246] "A means of generating problem ideas based on analysis results by referring to past test data" refers to a process of creating new test question ideas based on the analyzed intent of previously conducted tests.

[0247] "A means of creating a specific problem statement from generated problem ideas and checking its logical consistency" refers to the process of developing a problem designed at the idea level into a detailed statement and verifying whether its content is logically consistent.

[0248] "A means of analyzing a user's emotional state using a device for recognizing their emotions and adjusting the difficulty level and content of the questions based on that analysis" refers to a process that utilizes emotion recognition technology to analyze the user's psychological state and adaptively adjusts the content and difficulty level of the exam questions based on that information.

[0249] "Providing a flexible and individualized examination that takes into account the emotional state of the test-taker" means customizing the test questions and evaluation methods according to the emotions and psychological state of each individual test-taker, thereby providing an appropriate examination experience.

[0250] The system implementing this invention mainly consists of a server, a terminal, and an emotion analysis device. The server receives input from the user and analyzes the intent behind the question based on the input information. To support this analysis, it refers to past test data and extracts relevant information to improve the accuracy of the analysis. Based on the analysis results, the server uses a generative AI model to generate new question ideas and outputs them as specific question statements.

[0251] Simultaneously, the device acts as the interface with the user and collects data on the user's emotional state. This utilizes hardware that supports emotion recognition technology, such as cameras and microphones. Specifically, the Emotion Recognition library is used for emotion recognition. The collected emotional data is sent to a server and processed by an emotion analysis device.

[0252] The emotion analysis device analyzes this data to determine the user's emotional state. This result is used as input to adaptively adjust the difficulty level and content of the questions. For example, if it detects that the user is nervous, measures are taken to reduce the user's psychological burden, such as lowering the difficulty level of the questions or simplifying the introduction.

[0253] Even after the exam ends, the server collects the test-taker's answers and provides feedback based on sentiment analysis results. This feedback helps users understand themselves and provides a more personalized learning experience by including advice for future learning.

[0254] As a concrete example, in an exam conducted at a certain educational institution, the system can utilize the examinee's smartphone as a terminal to analyze their emotions in real time and enable flexible exam management tailored to the situation. An example of a prompt message could be, "Develop a system that sets the optimal difficulty level of exam questions based on the user's emotions."

[0255] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0256] Step 1:

[0257] The device receives input data from the user. This input includes data related to the intent of the test questions and the user's emotional state. The device uses sensors such as a camera and microphone to collect the user's facial expressions and voice tone, and generates data to recognize their emotional state.

[0258] Step 2:

[0259] The terminal sends the collected user input data and emotional state data to the server. The transmitted data serves as foundational data for analyzing the user's intent behind the questions and determining how emotional information influences them.

[0260] Step 3:

[0261] The server analyzes past exam data based on the received question intent and emotional state data. During the analysis, data calculations are performed to identify how the question intent relates to the user's emotions. Using these analysis results, a generative AI model is utilized to generate question ideas.

[0262] Step 4:

[0263] The server creates specific problem statements from the generated problem ideas and checks for logical consistency. This is the problem statement generation process, which aims to create problems of appropriate difficulty and content that reflect emotional information and align with the intent of the question.

[0264] Step 5:

[0265] The server finalizes the question and its scoring criteria, and sends them back to the terminal. The terminal then presents this to the user and receives user feedback as needed. This feedback is valuable input for the system to make further adjustments based on the sentiment analysis results.

[0266] Step 6:

[0267] Once a user submits their test answers, the device sends those answers to the server. The server evaluates the answers based on scoring criteria and generates feedback that takes into account the user's emotional state.

[0268] Step 7:

[0269] The server provides emotionally sensitive feedback back to the device. The device then presents this feedback to the user, helping them deepen their understanding as they move towards the next learning step. This allows the user to experience a personalized test that is tailored to their emotions and learning progress.

[0270] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0271] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0272] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0273] [Second Embodiment]

[0274] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0275] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0276] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0277] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0278] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0279] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0280] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0281] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0282] The specific processing program 56 is an example of the "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 operating as the specific processing unit 290 according to the specific processing program 56 executed by the processor 28 on the RAM 30.

[0283] 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 specific processing unit 290.

[0284] In the smart glasses 214, reception / output processing is performed by the processor 46. The storage 50 stores a reception / output program 60. 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 processing is realized by operating as the control unit 46A according to the reception / output program 60 executed by the processor 46 on the RAM 48.

[0285] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 is referred to as a "server", and the smart glasses 214 are referred to as a "terminal".

[0286] The present invention is based on a question creation and scoring support system for descriptive tests using AI, and its embodiments will be described below.

[0287] Overview

[0288] This system is composed of three main components: a server, a terminal, and a user. The AI model operating on the server creates questions based on the question intention provided by the user, and further performs the creation of scoring criteria and the scoring of the examinee's answers.

[0289] Details of Each Component

[0290] server

[0291] Problem creation function

[0292] The server receives the user's submitted question intent and analyzes it using natural language processing (NLP) techniques to extract relevant topics and keywords. Based on this information, an AI model generates appropriate question ideas while referring to a database of past exams. It then generates specific question texts from the selected ideas and checks their logical consistency and readability. It also has a function to evaluate the characteristics of the questions and estimate their difficulty level.

[0293] Scoring support function

[0294] When the server receives an ideal answer from a user, it analyzes it to create scoring criteria. Based on these criteria, it scores the examinee's written answer. The AI ​​model evaluates the accuracy and relevance of the answer through syntactic and semantic analysis.

[0295] terminal

[0296] Interface function

[0297] The terminal provides an interface for users to access the system, input question intents, check questions and scoring criteria, and view scoring results. The UI is intuitive and designed for easy user operation.

[0298] User

[0299] Providing the intent behind the questions

[0300] Users provide the server with the abilities they wish to evaluate and the themes they wish to address via their device, and request the creation of problems based on this information.

[0301] Result confirmation and feedback

[0302] The user checks the questions, scoring criteria, and test results provided by the server through the terminal, and provides feedback for improvement if necessary.

[0303] Presentation of specific examples

[0304] For example, when the user wants to create a test question with the theme of "evaluating the economic analysis ability related to sustainability", the user uses the terminal to send this intention to the server. The server uses AI to identify relevant topics and generates questions such as "Discuss the impact of sustainability policies on economic growth". Then, the user uploads their example answer, and the server forms the scoring criteria based on this answer and evaluates the actual test takers' answers.

[0305] Through such a process, efficient and fair test operation can be achieved without relying on specialized knowledge.

[0306] The following describes the processing flow.

[0307] Step 1:

[0308] The user uses the terminal to input the intention of the question and the purpose of the test. At this time, by specifically describing the theme and ability to be evaluated, the subsequent AI analysis can proceed smoothly.

[0309] Step 2:

[0310] The terminal sends the user's input to the server. The transmitted data is stored on the server, and preparations are made for use in subsequent analysis processes.

[0311] Step 3:

[0312] The server analyzes the received question intention using natural language processing (NLP) technology. As a result, relevant keywords and topics are extracted to obtain guidelines for question creation.

[0313] Step 4:

[0314] The server references a database of past exams based on the extracted keywords, and the AI ​​model generates appropriate problem ideas. This process verifies whether the ideas match the user's objectives.

[0315] Step 5:

[0316] The server generates a specific problem statement based on the selected problem idea. The generated problem statement is automatically checked for logical consistency and appropriate language.

[0317] Step 6:

[0318] The server estimates the difficulty level of the generated problem based on evaluation metrics. This numerically represents how challenging the problem is for the test-taker.

[0319] Step 7:

[0320] Users upload ideal answer examples to the server via their devices. These answers serve as the basic information necessary for creating the scoring criteria.

[0321] Step 8:

[0322] The server analyzes the user's ideal answer and automatically creates scoring criteria based on it. These criteria serve as a guideline for evaluation when test-takers answer the questions.

[0323] Step 9:

[0324] The device presents the completed problem statement and grading criteria to the user. At this stage, the user can review the content and provide corrections or feedback if necessary.

[0325] Step 10:

[0326] The server collects test-takers' answers during the exam and prepares them for analysis. This data is used for subsequent scoring.

[0327] Step 11:

[0328] The server evaluates the test-taker's answers based on scoring criteria and uses AI to determine the accuracy and relevance of each answer. The results are calculated as a score.

[0329] Step 12:

[0330] The terminal displays the scoring results to the user and provides information to obtain feedback useful for exam administration. The user evaluates the system's effectiveness through this information and uses it to create future exams.

[0331] (Example 1)

[0332] Next, we will describe Example 1. 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."

[0333] Traditional essay-style examination question creation and grading processes require significant time and expertise, making it difficult to ensure efficiency and fairness. In particular, consistently establishing standards to ensure the quality of exam questions and fairly evaluate examinees' responses is challenging.

[0334] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0335] In this invention, the server includes means for receiving user input and analyzing the intent based on the input, means for referring to past test information and generating problem ideas based on the analysis results, and means for creating specific problems from the generated ideas and checking their logical consistency. This makes it possible to improve the efficiency and fairness of the question creation and scoring process for written examinations.

[0336] "User input" refers to information and instructions provided by system users via their terminals, and serves as the starting point for creating exam questions and evaluating answers.

[0337] "Means for analyzing intent" refers to a system that uses natural language processing technology to analyze information provided by users and identify their purpose and specific needs.

[0338] "Past exam information" refers to historical questions and sample answers stored in existing exam databases, which are used as reference when creating new questions.

[0339] "Means of generating problem ideas" refers to the process of devising an outline of a problem, including related themes, based on the analyzed intent.

[0340] "Methods for creating specific problems" refer to the steps taken to construct a detailed and clear problem statement based on the generated ideas.

[0341] A "means of checking logical consistency" refers to a mechanism for evaluating whether a problem statement is logically consistent and free from contradictions.

[0342] An "ideal answer" is a model answer provided by a user, serving as a reference example for forming the scoring criteria.

[0343] "Means of creating evaluation criteria" refers to the process of establishing criteria for fairly evaluating test-takers' answers, based on ideal responses.

[0344] "Syntactic analysis" refers to the process of analyzing and understanding a test-taker's answers based on their grammatical structure.

[0345] "Semantic analysis" is the process of interpreting the content of a test-taker's answers and evaluating their intent and relevance.

[0346] "Means for evaluating responses" refers to a system that measures the quality of test-takers' responses based on syntactic and semantic analysis.

[0347] Specific embodiments of this invention will now be described.

[0348] This system consists of three main components: servers, terminals, and users.

[0349] server

[0350] The server operates on a cloud server with powerful computing capabilities. This server receives the intent of the question submitted by the user through their terminal and analyzes it using natural language processing (NLP) technology. The server has a generative AI model installed that accesses a database containing past exam information and extracts relevant topics based on the input intent. The AI ​​model then generates question ideas and creates specific question texts based on this. This process checks the logical consistency of the questions and selects the most suitable questions. Furthermore, it creates evaluation criteria based on the ideal answers received from the user and scores the answers obtained by the test taker. Mechanisms are in place to evaluate answers in detail through syntactic and semantic analysis.

[0351] terminal

[0352] The terminal provides the interface for users to access the system. This includes PCs and tablets used by users, which connect to the server via a web browser. The terminal is designed so that users can input the intent of the questions and ideal answers, and then check the questions, evaluation criteria, and scoring results generated by the server. The interface is also intuitive and designed with ease of operation in mind.

[0353] User

[0354] Users are the ones who use terminals to input intentions and themes in order to support the process of creating and grading exam questions. For example, if a user wants to create an exam question on the theme of "evaluating economic analytical skills related to sustainability," they use their terminal to send that intention to the server. Based on this, the server generates a question such as "Discuss the impact of sustainability policies on economic growth." The user then uploads their ideal answer to the server, and the server evaluates the test-taker's answer based on scoring criteria derived from that answer. This entire process ensures highly efficient and fair examination administration.

[0355] Example of a prompt

[0356] "Please create test questions to assess economic analytical skills related to sustainability."

[0357] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0358] Step 1:

[0359] Users input the intent behind the questions using their devices. Specifically, they use a web interface input form to describe in detail the abilities or themes they wish to evaluate. The data entered is primarily in text format and is transferred from the device to the server.

[0360] Step 2:

[0361] The server analyzes the question intent received from the user using natural language processing technology. This analysis extracts the intended evaluation points and topics. During the processing, the generative AI model analyzes keywords and main themes from the text and identifies relevant information. As a result, the analyzed data forms the basis for question generation.

[0362] Step 3:

[0363] The server generates question ideas based on analyzed topic information and references past exam data. It utilizes a generative AI model to suggest relevant questions and then creates specific question texts based on these suggestions. The output data, including a summary of the questions and specific question formats, is prepared for further evaluation.

[0364] Step 4:

[0365] The server checks the logical consistency and difficulty level of the generated problem statements. The AI ​​model analyzes whether the problems are consistent and of appropriate difficulty. Here, the quality of the problems is checked according to the set criteria, and the optimal content is selected. The checked problem statements are output and ready for user review.

[0366] Step 5:

[0367] Users review the question and its evaluation criteria presented by the server via their terminal. If necessary, users can request revisions to the content. A crucial step in the terminal review process is re-examining the question and approving the final question content.

[0368] Step 6:

[0369] Users input ideal answer examples from their terminals into the server. These example answers are required to be concise and appropriate, based on the intent of the question. The input data is then used to generate scoring criteria.

[0370] Step 7:

[0371] The server analyzes the ideal answers provided by users and creates scoring criteria based on them. It utilizes a generative AI model to identify key elements of the answers. Syntactic and semantic analysis are used to extract evaluation points for the answers and establish fair scoring criteria. These criteria are used as indicators when evaluating test-takers' answers.

[0372] Step 8:

[0373] The server receives the test taker's answers and evaluates them based on pre-defined scoring criteria. Using syntactic and semantic analysis, it determines how closely the test taker's answers resemble the ideal answers. A final evaluation result is generated and provided to the user as feedback. This feedback can be used to review the test results and improve future test designs.

[0374] (Application Example 1)

[0375] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0376] In educational settings, creating and grading essay-style examinations is a task that requires specialized knowledge and time, placing a significant burden on educators. Furthermore, automation is difficult, and many challenges remain in ensuring efficient and fair evaluation. This invention aims to solve these problems and improve the quality of education.

[0377] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0378] In this invention, the server includes means for receiving user input and analyzing the intent behind the questions, means for generating question ideas by referring to past test data, and means for automatically generating questions using a generation AI model. This enables efficient and fair creation and scoring of written test questions on an educational equipment platform.

[0379] "User input" refers to the information and instructions that users provide to the system.

[0380] "Exam intent" refers to the user's intentions regarding the purpose and theme of the exam questions they wish to create.

[0381] "Analysis" refers to the process of analyzing input information and extracting necessary data and patterns.

[0382] "Past exam data" refers to information about previously used exam questions and their results.

[0383] "Problem idea" refers to a hypothesis or draft of an exam question based on the examiner's intent.

[0384] A "generative AI model" refers to an algorithm or system that uses artificial intelligence to automatically create problems.

[0385] "Specific problem statement" refers to the written content of the questions used in the exam.

[0386] "Logical consistency" refers to a state in which the problem statement and the answer are logically consistent and free from contradictions.

[0387] "Difficulty level" refers to a measure that indicates how challenging an exam question is.

[0388] An "ideal answer" refers to a model and optimal response to an exam question.

[0389] "Scoring criteria" refers to the standards or criteria used to evaluate a test-taker's answers.

[0390] "Voice input" refers to a method of capturing audio as data through a device such as a microphone.

[0391] "Text data" refers to character information recorded in a format that can be processed by a computer.

[0392] An "educational equipment platform" refers to a hardware and software environment used for learning and educational purposes.

[0393] "Evaluation" refers to the process of reviewing the test takers' answers and performance and determining their value and quality.

[0394] "Feedback" refers to information and advice for improvement provided to users or test takers based on evaluation results.

[0395] The system for carrying out this invention operates on an educational equipment platform. Servers, terminals, and users are the main components.

[0396] On the server, a generative AI model is used to automatically generate questions based on the user's intended question. Specifically, the user communicates the exam theme to the server via voice or text input. For voice input, the server uses the Google Speech-to-Text API to convert the voice data into text data. Next, NLP technology is used to analyze the intended question and refer to relevant past exam data. In this data analysis process, the AI ​​model infers and generates appropriate question ideas.

[0397] The terminal provides the user interface for accessing this system. Through an intuitive UI, users can review the generated problem statements and scoring criteria and suggest revisions as needed.

[0398] Users aim to improve the quality of education through this system. In particular, voice input allows for the provision of question intent under various conditions, enabling immediate question creation and grading in different situations. This flexibility helps to meet the diverse needs of educational settings and reduces the burden on educators.

[0399] For example, if a user wants to create an exam question on "environmental issues," the server will automatically generate a question such as "Discuss the relationship between environmental issues and economic policy." OpenAI's GPT series is used for analysis, enabling smooth question generation. An example of a prompt message is: "Generate an exam question on 'environmental issues' using OpenAI GPT-3. Please grade the answer considering semantics and syntax."

[0400] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0401] Step 1:

[0402] Users input the exam question's theme and intent via their device. Input can be in voice or text format; voice input is converted to text using the Google Speech-to-Text API. This process involves analyzing the voice data and converting it to text data, with the output being text data including the intent behind the question.

[0403] Step 2:

[0404] The server analyzes the input question intent using natural language processing (NLP) techniques. Specifically, it uses an NLP library (e.g., spaCy) to syntactically analyze the text data and extract important keywords and themes. This process outputs structured data of the analyzed question intent.

[0405] Step 3:

[0406] The server generates problem ideas based on extracted keywords and themes, utilizing a generative AI model (e.g., OpenAI GPT-3). Here, the AI ​​model combines relevant information to create a draft problem that matches the intended question. This process outputs a preliminary draft of the generated problem.

[0407] Step 4:

[0408] Based on the generated problem ideas, the server creates a specific problem statement. The automatically generated problem statement is then checked for logical consistency and difficulty. At this stage, the text data of the problem statement is output, along with evaluation values ​​for its consistency and difficulty.

[0409] Step 5:

[0410] The server analyzes the ideal answers provided by the user and creates scoring criteria. At this stage, NLP techniques are again used to analyze the syntax and meaning of the ideal answers. This process outputs structured data to be used as scoring criteria.

[0411] Step 6:

[0412] Users review the completed question text and scoring criteria on their devices and suggest revisions if necessary. Based on user feedback, data is then corrected, and the final revised question text and scoring criteria are output.

[0413] Step 7:

[0414] In educational settings, test takers submit their answers, and the server collects these answers. The collected answers are evaluated based on previously established scoring criteria. This results in an output score for each answer.

[0415] Step 8:

[0416] Based on the evaluation results, the server provides feedback to the user via the terminal. In this step, text data showing the results and areas for improvement is generated and output as feedback in a format that the user can review.

[0417] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0418] This invention aims to achieve more sophisticated exam management by combining an emotion engine with an AI-powered system for creating and grading written exam questions. This emotion engine is characterized by its ability to analyze the user's emotional state and adjust each process based on that information.

[0419] overview

[0420] The system consists of a server, terminals, and an emotion engine. The emotion engine analyzes the user's emotional state and uses the resulting data to adjust each process accordingly.

[0421] Details of each component

[0422] server

[0423] Problem creation support

[0424] The server processes the question intent received from the user, along with data including the sentiment analysis results from the sentiment engine, and extracts relevant topics. This process ensures that the question intent is appropriately interpreted based on the user's emotions, helping the AI ​​model accurately generate question ideas.

[0425] Grading support and feedback

[0426] By incorporating emotional information from the emotion engine into the evaluation of test responses, it is possible to optimize scoring criteria and feedback content. This allows for more flexible and personalized feedback.

[0427] terminal

[0428] Collection of interface and sentiment data

[0429] The terminal provides an interface for the user to access the system and collects data to obtain the user's emotional state. This data is analyzed by an emotion engine and sent to the server.

[0430] Emotional Engine

[0431] Emotional analysis and process adjustment

[0432] The emotion engine analyzes the user's emotions based on data transmitted from the device and reflects the results in each process. Specifically, it is used to adjust the difficulty level when creating questions and to adaptively adjust the scoring criteria.

[0433] Presentation of specific examples

[0434] For example, if a user creates a test to "assess critical thinking skills regarding environmental issues," and the emotion engine analyzes that the user is experiencing stress, the server will take that information into account and present questions with slightly adjusted difficulty levels. In this way, tests that take into account the user's emotions and mental state are constructed. After the test, when grading the test-takers' answers, the feedback is also adjusted based on the data obtained from the emotion analysis, helping test-takers accept the results in a way that is more easily understood and accepted.

[0435] This format makes it possible to provide users and test takers with a more personalized and effective testing experience.

[0436] The following describes the processing flow.

[0437] Step 1:

[0438] Users use a terminal to input the intent behind the questions regarding the themes and abilities they wish to evaluate. The input requires a detailed description of the test's purpose and target areas.

[0439] Step 2:

[0440] The device collects emotional data simultaneously with user input. This emotional data includes information estimated from factors such as keyboard typing speed and eye movements.

[0441] Step 3:

[0442] The device transmits question intent and emotional data to the server. The transmission is done in real time and is designed to respond to changes in emotions.

[0443] Step 4:

[0444] The server analyzes the received question intent using natural language processing (NLP). During this process, it references sentiment data provided by the sentiment engine to supplement the interpretation with one that reflects the user's mental state.

[0445] Step 5:

[0446] Based on the information obtained from the analysis, the server generates multiple problem ideas using an AI model. It then utilizes emotional data to select the idea that best suits the user's emotional state from among the generated ideas.

[0447] Step 6:

[0448] The server creates specific problem statements based on the selected problem ideas. During this process, the difficulty level is adjusted considering the emotion engine data, ensuring the content is adapted to the user's state.

[0449] Step 7:

[0450] The device presents the user with the completed problem statement and scoring criteria that reflect sentiment data. The user can review the content and provide feedback as needed.

[0451] Step 8:

[0452] When the test is administered, the device collects not only the test-taker's answers but also emotional data from the test. This data, along with the answers, is sent to the server.

[0453] Step 9:

[0454] The server uses the collected responses and sentiment data to perform scoring. The sentiment engine assists in evaluating the responses and calculates an appropriate score based on the scoring criteria.

[0455] Step 10:

[0456] The device presents test takers and users with emotionally-based feedback along with their scoring results. This feedback includes advice that reflects emotional elements to aid the test taker's understanding.

[0457] (Example 2)

[0458] Next, we will describe Example 2. 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".

[0459] Traditional exam creation and scoring systems have the drawback of not considering the emotional state of users and test-takers, making it difficult to provide a personalized exam experience. Furthermore, because the questions and scoring are standardized, the diverse learning styles and emotions of test-takers do not significantly influence the results.

[0460] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0461] In this invention, the server includes means for collecting and analyzing user emotional data, means for adjusting the difficulty level and scoring criteria of questions based on the emotional analysis results, and means for adjusting the content of feedback based on the emotional analysis results. This enables the creation of personalized tests and flexible scoring and feedback that are tailored to the emotional state of the user and test taker.

[0462] "User input" refers to the information and instructions that users provide to the system, including the intent behind the questions necessary for creating the exam.

[0463] "Exam intent" refers to the purpose and criteria that the user expresses when creating exam questions.

[0464] "Past data" refers to information and statistics collected previously, which are used in the generation and evaluation of exam questions.

[0465] An "idea" refers to a concept or theme for giving concrete form to the exam questions.

[0466] "Checking sentence consistency" is the process of verifying whether the generated problem sentences are grammatically and logically correct.

[0467] "Difficulty level" is an indicator that shows the complexity and ease of solving exam questions.

[0468] "Criteria for analyzing answers" are standard guidelines used to evaluate the answers of test-takers.

[0469] "Emotional data" refers to information that indicates a user's emotional state, and is collected and used by the system.

[0470] "Sentiment analysis" is the process of evaluating collected emotional data to understand the user's emotional state.

[0471] "Adjusting the feedback content" refers to the process of optimizing responses and comments to test-takers' answers based on the results of sentiment analysis.

[0472] This invention provides a test creation and scoring support system that utilizes AI technology, including user emotion analysis. This system consists of a server, terminals, and emotion analysis devices, which function through mutual cooperation.

[0473] The terminal provides the user with an interface for creating exams. Through this interface, the user inputs the intent behind the questions, and simultaneously, data for sentiment analysis is collected. For example, keyboard input speed, mouse movements, or direct user feedback can be used. The data obtained in this way is essential for capturing the user's real-time emotional state.

[0474] The server receives data transmitted from the terminal and analyzes the user's input regarding the intent behind the questions and emotional data. Emotional analysis is performed by an emotional analysis device, utilizing natural language processing and machine learning techniques. This analysis determines the user's stress level and agitation state, which is used to adjust the difficulty of the test. The server then uses a generative AI model to extract relevant topics that align with the intent behind the questions and creates questions based on those topics.

[0475] For example, if a user creates a test to "evaluate critical thinking skills regarding environmental issues," the system will determine the user's stress level through sentiment analysis and adjust the difficulty level appropriately. This allows for a flexible test format tailored to the user's state. Furthermore, when generating test questions, the AI ​​model applies appropriate prompts based on the legal provision that "the user's emotions should be taken into consideration using a generative AI model."

[0476] Thus, the present invention enables the creation and scoring of exams that reflect the individual emotional states of users and test takers, thereby providing a personalized educational environment.

[0477] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0478] Step 1:

[0479] The terminal receives input from the user and obtains the user's intent for the question. Simultaneously, it collects emotional data through sensors and interfaces. Specifically, it records the themes and keywords entered by the user, as well as the emotional state inferred from keyboard typing speed and mouse movements. This input data is collected as basic information necessary for analysis on the server.

[0480] Step 2:

[0481] The terminal transmits the collected question intent and emotional data to the server. After receiving this data, the server passes it to an emotional analysis device for analysis. Using natural language processing technology and machine learning algorithms, the device evaluates the user's emotional state and determines whether the user is experiencing stress. The results of this analysis are then used in the subsequent process of adjusting the difficulty level of the test questions.

[0482] Step 3:

[0483] The server takes the sentiment analysis results into account and uses a generative AI model to extract relevant topics that match the intent of the question. Specifically, the AI ​​model uses prompt sentences to generate specific topics based on the given theme. These prompt sentences are such as, "The AI ​​will generate questions based on the user's sentiment about the selected topic."

[0484] Step 4:

[0485] The server generates test questions that take sentiment data into account, based on the extracted topics. If the analysis indicates that the user is experiencing stress, the difficulty level of the questions is appropriately adjusted before generation. In this process, the question text is created based on the topics generated by the AI, and a consistency check is performed.

[0486] Step 5:

[0487] The server verifies the difficulty level and logical consistency of the generated exam questions, and then provides the user with the exam questions and predicted scoring criteria. It also provides flexible editing tools through an interface, allowing users to modify the questions and scoring criteria as needed.

[0488] Step 6:

[0489] After completing the exam, the user submits their answers to the server. The server evaluates the received answers according to scoring criteria and adjusts the feedback based on the results of sentiment analysis. In particular, if the server determines that the test-taker is experiencing excessive stress, the feedback is optimized to emphasize positive aspects.

[0490] (Application Example 2)

[0491] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0492] Traditional exam question creation and scoring systems have a problem in that they do not take into account the emotional state of test-takers, and in particular, they do not adequately consider the impact of stress and anxiety on test-takers' performance. As a result, there is a problem in conducting flexible and individualized exams that respond to the diverse needs and emotions of learners.

[0493] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0494] In this invention, the server includes means for receiving user input and analyzing the intent behind the questions based on the input; means for referring to past test data and generating question ideas based on the analysis results; means for creating specific question sentences from the generated question ideas and checking their logical consistency; and means for analyzing the emotional state of the user using a device for recognizing emotions and adjusting the difficulty level and content of the questions based on that analysis. This makes it possible to provide a flexible and individualized test that takes into account the emotional state of the test taker.

[0495] "Means for receiving user input and analyzing the intent behind the questions based on that input" refers to a process of receiving input data from users, analyzing that data, and interpreting what purpose and direction questions should be created in.

[0496] "A means of generating problem ideas based on analysis results by referring to past test data" refers to a process of creating new test question ideas based on the analyzed intent of previously conducted tests.

[0497] "A means of creating a specific problem statement from generated problem ideas and checking its logical consistency" refers to the process of developing a problem designed at the idea level into a detailed statement and verifying whether its content is logically consistent.

[0498] "A means of analyzing a user's emotional state using a device for recognizing their emotions and adjusting the difficulty level and content of the questions based on that analysis" refers to a process that utilizes emotion recognition technology to analyze the user's psychological state and adaptively adjusts the content and difficulty level of the exam questions based on that information.

[0499] "Providing a flexible and individualized examination that takes into account the emotional state of the test-taker" means customizing the test questions and evaluation methods according to the emotions and psychological state of each individual test-taker, thereby providing an appropriate examination experience.

[0500] The system implementing this invention mainly consists of a server, a terminal, and an emotion analysis device. The server receives input from the user and analyzes the intent behind the question based on the input information. To support this analysis, it refers to past test data and extracts relevant information to improve the accuracy of the analysis. Based on the analysis results, the server uses a generative AI model to generate new question ideas and outputs them as specific question statements.

[0501] Simultaneously, the device acts as the interface with the user and collects data on the user's emotional state. This utilizes hardware that supports emotion recognition technology, such as cameras and microphones. Specifically, the Emotion Recognition library is used for emotion recognition. The collected emotional data is sent to a server and processed by an emotion analysis device.

[0502] The emotion analysis device analyzes this data to determine the user's emotional state. This result is used as input to adaptively adjust the difficulty level and content of the questions. For example, if it detects that the user is nervous, measures are taken to reduce the user's psychological burden, such as lowering the difficulty level of the questions or simplifying the introduction.

[0503] Even after the exam ends, the server collects the test-taker's answers and provides feedback based on sentiment analysis results. This feedback helps users understand themselves and provides a more personalized learning experience by including advice for future learning.

[0504] As a concrete example, in an exam conducted at a certain educational institution, the system can utilize the examinee's smartphone as a terminal to analyze their emotions in real time and enable flexible exam management tailored to the situation. An example of a prompt message could be, "Develop a system that sets the optimal difficulty level of exam questions based on the user's emotions."

[0505] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0506] Step 1:

[0507] The device receives input data from the user. This input includes data related to the intent of the test questions and the user's emotional state. The device uses sensors such as a camera and microphone to collect the user's facial expressions and voice tone, and generates data to recognize their emotional state.

[0508] Step 2:

[0509] The terminal sends the collected user input data and emotional state data to the server. The transmitted data serves as foundational data for analyzing the user's intent behind the questions and determining how emotional information influences them.

[0510] Step 3:

[0511] The server analyzes past exam data based on the received question intent and emotional state data. During the analysis, data calculations are performed to identify how the question intent relates to the user's emotions. Using these analysis results, a generative AI model is utilized to generate question ideas.

[0512] Step 4:

[0513] The server creates specific problem statements from the generated problem ideas and checks for logical consistency. This is the problem statement generation process, which aims to create problems of appropriate difficulty and content that reflect emotional information and align with the intent of the question.

[0514] Step 5:

[0515] The server finalizes the question and its scoring criteria, and sends them back to the terminal. The terminal then presents this to the user and receives user feedback as needed. This feedback is valuable input for the system to make further adjustments based on the sentiment analysis results.

[0516] Step 6:

[0517] Once a user submits their test answers, the device sends those answers to the server. The server evaluates the answers based on scoring criteria and generates feedback that takes into account the user's emotional state.

[0518] Step 7:

[0519] The server provides emotionally sensitive feedback back to the device. The device then presents this feedback to the user, helping them deepen their understanding as they move towards the next learning step. This allows the user to experience a personalized test that is tailored to their emotions and learning progress.

[0520] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0521] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0522] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0523] [Third Embodiment]

[0524] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0525] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0526] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0527] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0528] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0529] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0530] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0531] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0532] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0533] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0534] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0535] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0536] The present invention relates to an AI-powered system for creating and grading written examination questions, and its embodiments are described below.

[0537] overview

[0538] This system consists of three main components: a server, a terminal, and a user. An AI model running on the server creates questions based on the user's intended question, and also creates scoring criteria and scores the test takers' answers.

[0539] Details of each component

[0540] server

[0541] Problem creation function

[0542] The server receives the user's submitted question intent and analyzes it using natural language processing (NLP) techniques to extract relevant topics and keywords. Based on this information, an AI model generates appropriate question ideas while referring to a database of past exams. It then generates specific question texts from the selected ideas and checks their logical consistency and readability. It also has a function to evaluate the characteristics of the questions and estimate their difficulty level.

[0543] Scoring support function

[0544] When the server receives an ideal answer from a user, it analyzes it to create scoring criteria. Based on these criteria, it scores the examinee's written answer. The AI ​​model evaluates the accuracy and relevance of the answer through syntactic and semantic analysis.

[0545] terminal

[0546] Interface function

[0547] The terminal provides an interface for users to access the system, input question intents, check questions and scoring criteria, and view scoring results. The UI is intuitive and designed for easy user operation.

[0548] User

[0549] Providing the intent behind the questions

[0550] Users provide the server with the abilities they wish to evaluate and the themes they wish to address via their device, and request the creation of problems based on this information.

[0551] Result confirmation and feedback

[0552] Users can view the questions, scoring criteria, and test results provided by the server through their terminals and provide feedback for improvement as needed.

[0553] Presentation of specific examples

[0554] For example, if a user wants to create an exam question on the theme of "assessing economic analytical skills related to sustainability," the user sends this intention to the server using their device. The server uses AI to identify relevant topics and generates questions such as, "Discuss the impact of sustainability policies on economic growth." The user then uploads their own example answer, and the server forms scoring criteria based on that answer and evaluates the answers of actual test takers.

[0555] This process enables efficient and fair testing without relying on specialized knowledge.

[0556] The following describes the processing flow.

[0557] Step 1:

[0558] Users use their devices to input the intent behind the questions and the objectives of the exam. By specifically describing the themes and abilities they wish to evaluate, the subsequent AI analysis will proceed more smoothly.

[0559] Step 2:

[0560] The terminal sends user input to the server. The transmitted data is stored on the server and prepared for use in subsequent analysis processes.

[0561] Step 3:

[0562] The server analyzes the received question intent using natural language processing (NLP) techniques. This extracts relevant keywords and topics, providing guidance for creating questions.

[0563] Step 4:

[0564] The server references a database of past exams based on the extracted keywords, and the AI ​​model generates appropriate problem ideas. This process verifies whether the ideas match the user's objectives.

[0565] Step 5:

[0566] The server generates a specific problem statement based on the selected problem idea. The generated problem statement is automatically checked for logical consistency and appropriate language.

[0567] Step 6:

[0568] The server estimates the difficulty level of the generated problem based on evaluation metrics. This numerically represents how challenging the problem is for the test-taker.

[0569] Step 7:

[0570] Users upload ideal answer examples to the server via their devices. These answers serve as the basic information necessary for creating the scoring criteria.

[0571] Step 8:

[0572] The server analyzes the user's ideal answer and automatically creates scoring criteria based on it. These criteria serve as a guideline for evaluation when test-takers answer the questions.

[0573] Step 9:

[0574] The device presents the completed problem statement and grading criteria to the user. At this stage, the user can review the content and provide corrections or feedback if necessary.

[0575] Step 10:

[0576] The server collects test-takers' answers during the exam and prepares them for analysis. This data is used for subsequent scoring.

[0577] Step 11:

[0578] The server evaluates the test-taker's answers based on scoring criteria and uses AI to determine the accuracy and relevance of each answer. The results are calculated as a score.

[0579] Step 12:

[0580] The terminal displays the scoring results to the user and provides information to obtain feedback useful for exam administration. The user evaluates the system's effectiveness through this information and uses it to create future exams.

[0581] (Example 1)

[0582] Next, we will describe Example 1. 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."

[0583] Traditional essay-style examination question creation and grading processes require significant time and expertise, making it difficult to ensure efficiency and fairness. In particular, consistently establishing standards to ensure the quality of exam questions and fairly evaluate examinees' responses is challenging.

[0584] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0585] In this invention, the server includes means for receiving user input and analyzing the intent based on the input, means for referring to past test information and generating problem ideas based on the analysis results, and means for creating specific problems from the generated ideas and checking their logical consistency. This makes it possible to improve the efficiency and fairness of the question creation and scoring process for written examinations.

[0586] "User input" refers to information and instructions provided by system users via their terminals, and serves as the starting point for creating exam questions and evaluating answers.

[0587] "Means for analyzing intent" refers to a system that uses natural language processing technology to analyze information provided by users and identify their purpose and specific needs.

[0588] "Past exam information" refers to historical questions and sample answers stored in existing exam databases, which are used as reference when creating new questions.

[0589] "Means of generating problem ideas" refers to the process of devising an outline of a problem, including related themes, based on the analyzed intent.

[0590] "Methods for creating specific problems" refer to the steps taken to construct a detailed and clear problem statement based on the generated ideas.

[0591] A "means of checking logical consistency" refers to a mechanism for evaluating whether a problem statement is logically consistent and free from contradictions.

[0592] An "ideal answer" is a model answer provided by a user, serving as a reference example for forming the scoring criteria.

[0593] "Means of creating evaluation criteria" refers to the process of establishing criteria for fairly evaluating test-takers' answers, based on ideal responses.

[0594] "Syntactic analysis" refers to the process of analyzing and understanding a test-taker's answers based on their grammatical structure.

[0595] "Semantic analysis" is the process of interpreting the content of a test-taker's answers and evaluating their intent and relevance.

[0596] "Means for evaluating responses" refers to a system that measures the quality of test-takers' responses based on syntactic and semantic analysis.

[0597] Specific embodiments of this invention will now be described.

[0598] This system consists of three main components: servers, terminals, and users.

[0599] server

[0600] The server operates on a cloud server with powerful computing capabilities. This server receives the intent of the question submitted by the user through their terminal and analyzes it using natural language processing (NLP) technology. The server has a generative AI model installed that accesses a database containing past exam information and extracts relevant topics based on the input intent. The AI ​​model then generates question ideas and creates specific question texts based on this. This process checks the logical consistency of the questions and selects the most suitable questions. Furthermore, it creates evaluation criteria based on the ideal answers received from the user and scores the answers obtained by the test taker. Mechanisms are in place to evaluate answers in detail through syntactic and semantic analysis.

[0601] terminal

[0602] The terminal provides the interface for users to access the system. This includes PCs and tablets used by users, which connect to the server via a web browser. The terminal is designed so that users can input the intent of the questions and ideal answers, and then check the questions, evaluation criteria, and scoring results generated by the server. The interface is also intuitive and designed with ease of operation in mind.

[0603] User

[0604] Users are the ones who use terminals to input intentions and themes in order to support the process of creating and grading exam questions. For example, if a user wants to create an exam question on the theme of "evaluating economic analytical skills related to sustainability," they use their terminal to send that intention to the server. Based on this, the server generates a question such as "Discuss the impact of sustainability policies on economic growth." The user then uploads their ideal answer to the server, and the server evaluates the test-taker's answer based on scoring criteria derived from that answer. This entire process ensures highly efficient and fair examination administration.

[0605] Example of a prompt

[0606] "Please create test questions to assess economic analytical skills related to sustainability."

[0607] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0608] Step 1:

[0609] Users input the intent behind the questions using their devices. Specifically, they use a web interface input form to describe in detail the abilities or themes they wish to evaluate. The data entered is primarily in text format and is transferred from the device to the server.

[0610] Step 2:

[0611] The server analyzes the question intent received from the user using natural language processing technology. This analysis extracts the intended evaluation points and topics. During the processing, the generative AI model analyzes keywords and main themes from the text and identifies relevant information. As a result, the analyzed data forms the basis for question generation.

[0612] Step 3:

[0613] The server generates question ideas based on analyzed topic information and references past exam data. It utilizes a generative AI model to suggest relevant questions and then creates specific question texts based on these suggestions. The output data, including a summary of the questions and specific question formats, is prepared for further evaluation.

[0614] Step 4:

[0615] The server checks the logical consistency and difficulty level of the generated problem statements. The AI ​​model analyzes whether the problems are consistent and of appropriate difficulty. Here, the quality of the problems is checked according to the set criteria, and the optimal content is selected. The checked problem statements are output and ready for user review.

[0616] Step 5:

[0617] Users review the question and its evaluation criteria presented by the server via their terminal. If necessary, users can request revisions to the content. A crucial step in the terminal review process is re-examining the question and approving the final question content.

[0618] Step 6:

[0619] Users input ideal answer examples from their terminals into the server. These example answers are required to be concise and appropriate, based on the intent of the question. The input data is then used to generate scoring criteria.

[0620] Step 7:

[0621] The server analyzes the ideal answers provided by users and creates scoring criteria based on them. It utilizes a generative AI model to identify key elements of the answers. Syntactic and semantic analysis are used to extract evaluation points for the answers and establish fair scoring criteria. These criteria are used as indicators when evaluating test-takers' answers.

[0622] Step 8:

[0623] The server receives the test taker's answers and evaluates them based on pre-defined scoring criteria. Using syntactic and semantic analysis, it determines how closely the test taker's answers resemble the ideal answers. A final evaluation result is generated and provided to the user as feedback. This feedback can be used to review the test results and improve future test designs.

[0624] (Application Example 1)

[0625] Next, we will explain Application Example 1. In the following explanation, 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."

[0626] In educational settings, creating and grading essay-style examinations is a task that requires specialized knowledge and time, placing a significant burden on educators. Furthermore, automation is difficult, and many challenges remain in ensuring efficient and fair evaluation. This invention aims to solve these problems and improve the quality of education.

[0627] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0628] In this invention, the server includes means for receiving user input and analyzing the intent behind the questions, means for generating question ideas by referring to past test data, and means for automatically generating questions using a generation AI model. This enables efficient and fair creation and scoring of written test questions on an educational equipment platform.

[0629] "User input" refers to the information and instructions that users provide to the system.

[0630] "Exam intent" refers to the user's intentions regarding the purpose and theme of the exam questions they wish to create.

[0631] "Analysis" refers to the process of analyzing input information and extracting necessary data and patterns.

[0632] "Past exam data" refers to information about previously used exam questions and their results.

[0633] "Problem idea" refers to a hypothesis or draft of an exam question based on the examiner's intent.

[0634] A "generative AI model" refers to an algorithm or system that uses artificial intelligence to automatically create problems.

[0635] "Specific problem statement" refers to the written content of the questions used in the exam.

[0636] "Logical consistency" refers to a state in which the problem statement and the answer are logically consistent and free from contradictions.

[0637] "Difficulty level" refers to a measure that indicates how challenging an exam question is.

[0638] An "ideal answer" refers to a model and optimal response to an exam question.

[0639] "Scoring criteria" refers to the standards or criteria used to evaluate a test-taker's answers.

[0640] "Voice input" refers to a method of capturing audio as data through a device such as a microphone.

[0641] "Text data" refers to character information recorded in a format that can be processed by a computer.

[0642] An "educational equipment platform" refers to a hardware and software environment used for learning and educational purposes.

[0643] "Evaluation" refers to the process of reviewing the test takers' answers and performance and determining their value and quality.

[0644] "Feedback" refers to information and advice for improvement provided to users or test takers based on evaluation results.

[0645] The system for carrying out this invention operates on an educational equipment platform. Servers, terminals, and users are the main components.

[0646] On the server, a generative AI model is used to automatically generate questions based on the user's intended question. Specifically, the user communicates the exam theme to the server via voice or text input. For voice input, the server uses the Google Speech-to-Text API to convert the voice data into text data. Next, NLP technology is used to analyze the intended question and refer to relevant past exam data. In this data analysis process, the AI ​​model infers and generates appropriate question ideas.

[0647] The terminal provides the user interface for accessing this system. Through an intuitive UI, users can review the generated problem statements and scoring criteria and suggest revisions as needed.

[0648] Users aim to improve the quality of education through this system. In particular, voice input allows for the provision of question intent under various conditions, enabling immediate question creation and grading in different situations. This flexibility helps to meet the diverse needs of educational settings and reduces the burden on educators.

[0649] For example, if a user wants to create an exam question on "environmental issues," the server will automatically generate a question such as "Discuss the relationship between environmental issues and economic policy." OpenAI's GPT series is used for analysis, enabling smooth question generation. An example of a prompt message is: "Generate an exam question on 'environmental issues' using OpenAI GPT-3. Please grade the answer considering semantics and syntax."

[0650] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0651] Step 1:

[0652] Users input the exam question's theme and intent via their device. Input can be in voice or text format; voice input is converted to text using the Google Speech-to-Text API. This process involves analyzing the voice data and converting it to text data, with the output being text data including the intent behind the question.

[0653] Step 2:

[0654] The server analyzes the input question intent using natural language processing (NLP) techniques. Specifically, it uses an NLP library (e.g., spaCy) to syntactically analyze the text data and extract important keywords and themes. This process outputs structured data of the analyzed question intent.

[0655] Step 3:

[0656] The server generates problem ideas based on extracted keywords and themes, utilizing a generative AI model (e.g., OpenAI GPT-3). Here, the AI ​​model combines relevant information to create a draft problem that matches the intended question. This process outputs a preliminary draft of the generated problem.

[0657] Step 4:

[0658] Based on the generated problem ideas, the server creates a specific problem statement. The automatically generated problem statement is then checked for logical consistency and difficulty. At this stage, the text data of the problem statement is output, along with evaluation values ​​for its consistency and difficulty.

[0659] Step 5:

[0660] The server analyzes the ideal answers provided by the user and creates scoring criteria. At this stage, NLP techniques are again used to analyze the syntax and meaning of the ideal answers. This process outputs structured data to be used as scoring criteria.

[0661] Step 6:

[0662] Users review the completed question text and scoring criteria on their devices and suggest revisions if necessary. Based on user feedback, data is then corrected, and the final revised question text and scoring criteria are output.

[0663] Step 7:

[0664] In educational settings, test takers submit their answers, and the server collects these answers. The collected answers are evaluated based on previously established scoring criteria. This results in an output score for each answer.

[0665] Step 8:

[0666] Based on the evaluation results, the server provides feedback to the user via the terminal. In this step, text data showing the results and areas for improvement is generated and output as feedback in a format that the user can review.

[0667] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0668] This invention aims to achieve more sophisticated exam management by combining an emotion engine with an AI-powered system for creating and grading written exam questions. This emotion engine is characterized by its ability to analyze the user's emotional state and adjust each process based on that information.

[0669] overview

[0670] The system consists of a server, terminals, and an emotion engine. The emotion engine analyzes the user's emotional state and uses the resulting data to adjust each process accordingly.

[0671] Details of each component

[0672] server

[0673] Problem creation support

[0674] The server processes the question intent received from the user, along with data including the sentiment analysis results from the sentiment engine, and extracts relevant topics. This process ensures that the question intent is appropriately interpreted based on the user's emotions, helping the AI ​​model accurately generate question ideas.

[0675] Grading support and feedback

[0676] By incorporating emotional information from the emotion engine into the evaluation of test responses, it is possible to optimize scoring criteria and feedback content. This allows for more flexible and personalized feedback.

[0677] terminal

[0678] Collection of interface and sentiment data

[0679] The terminal provides an interface for the user to access the system and collects data to obtain the user's emotional state. This data is analyzed by an emotion engine and sent to the server.

[0680] Emotional Engine

[0681] Emotional analysis and process adjustment

[0682] The emotion engine analyzes the user's emotions based on data transmitted from the device and reflects the results in each process. Specifically, it is used to adjust the difficulty level when creating questions and to adaptively adjust the scoring criteria.

[0683] Presentation of specific examples

[0684] For example, if a user creates a test to "assess critical thinking skills regarding environmental issues," and the emotion engine analyzes that the user is experiencing stress, the server will take that information into account and present questions with slightly adjusted difficulty levels. In this way, tests that take into account the user's emotions and mental state are constructed. After the test, when grading the test-takers' answers, the feedback is also adjusted based on the data obtained from the emotion analysis, helping test-takers accept the results in a way that is more easily understood and accepted.

[0685] This format makes it possible to provide users and test takers with a more personalized and effective testing experience.

[0686] The following describes the processing flow.

[0687] Step 1:

[0688] Users use a terminal to input the intent behind the questions regarding the themes and abilities they wish to evaluate. The input requires a detailed description of the test's purpose and target areas.

[0689] Step 2:

[0690] The device collects emotional data simultaneously with user input. This emotional data includes information estimated from factors such as keyboard typing speed and eye movements.

[0691] Step 3:

[0692] The device transmits question intent and emotional data to the server. The transmission is done in real time and is designed to respond to changes in emotions.

[0693] Step 4:

[0694] The server analyzes the received question intent using natural language processing (NLP). During this process, it references sentiment data provided by the sentiment engine to supplement the interpretation with one that reflects the user's mental state.

[0695] Step 5:

[0696] Based on the information obtained from the analysis, the server generates multiple problem ideas using an AI model. It then utilizes emotional data to select the idea that best suits the user's emotional state from among the generated ideas.

[0697] Step 6:

[0698] The server creates specific problem statements based on the selected problem ideas. During this process, the difficulty level is adjusted considering the emotion engine data, ensuring the content is adapted to the user's state.

[0699] Step 7:

[0700] The device presents the user with the completed problem statement and scoring criteria that reflect sentiment data. The user can review the content and provide feedback as needed.

[0701] Step 8:

[0702] When the test is administered, the device collects not only the test-taker's answers but also emotional data from the test. This data, along with the answers, is sent to the server.

[0703] Step 9:

[0704] The server uses the collected responses and sentiment data to perform scoring. The sentiment engine assists in evaluating the responses and calculates an appropriate score based on the scoring criteria.

[0705] Step 10:

[0706] The device presents test takers and users with emotionally-based feedback along with their scoring results. This feedback includes advice that reflects emotional elements to aid the test taker's understanding.

[0707] (Example 2)

[0708] Next, we will describe Example 2. 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."

[0709] Traditional exam creation and scoring systems have the drawback of not considering the emotional state of users and test-takers, making it difficult to provide a personalized exam experience. Furthermore, because the questions and scoring are standardized, the diverse learning styles and emotions of test-takers do not significantly influence the results.

[0710] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0711] In this invention, the server includes means for collecting and analyzing user emotional data, means for adjusting the difficulty level and scoring criteria of questions based on the emotional analysis results, and means for adjusting the content of feedback based on the emotional analysis results. This enables the creation of personalized tests and flexible scoring and feedback that are tailored to the emotional state of the user and test taker.

[0712] "User input" refers to the information and instructions that users provide to the system, including the intent behind the questions necessary for creating the exam.

[0713] "Exam intent" refers to the purpose and criteria that the user expresses when creating exam questions.

[0714] "Past data" refers to information and statistics collected previously, which are used in the generation and evaluation of exam questions.

[0715] An "idea" refers to a concept or theme for giving concrete form to the exam questions.

[0716] "Checking sentence consistency" is the process of verifying whether the generated problem sentences are grammatically and logically correct.

[0717] "Difficulty level" is an indicator that shows the complexity and ease of solving exam questions.

[0718] "Criteria for analyzing answers" are standard guidelines used to evaluate the answers of test-takers.

[0719] "Emotional data" refers to information that indicates a user's emotional state, and is collected and used by the system.

[0720] "Sentiment analysis" is the process of evaluating collected emotional data to understand the user's emotional state.

[0721] "Adjusting the feedback content" refers to the process of optimizing responses and comments to test-takers' answers based on the results of sentiment analysis.

[0722] This invention provides a test creation and scoring support system that utilizes AI technology, including user emotion analysis. This system consists of a server, terminals, and emotion analysis devices, which function through mutual cooperation.

[0723] The terminal provides the user with an interface for creating exams. Through this interface, the user inputs the intent behind the questions, and simultaneously, data for sentiment analysis is collected. For example, keyboard input speed, mouse movements, or direct user feedback can be used. The data obtained in this way is essential for capturing the user's real-time emotional state.

[0724] The server receives data transmitted from the terminal and analyzes the user's input regarding the intent behind the questions and emotional data. Emotional analysis is performed by an emotional analysis device, utilizing natural language processing and machine learning techniques. This analysis determines the user's stress level and agitation state, which is used to adjust the difficulty of the test. The server then uses a generative AI model to extract relevant topics that align with the intent behind the questions and creates questions based on those topics.

[0725] For example, if a user creates a test to "evaluate critical thinking skills regarding environmental issues," the system will determine the user's stress level through sentiment analysis and adjust the difficulty level appropriately. This allows for a flexible test format tailored to the user's state. Furthermore, when generating test questions, the AI ​​model applies appropriate prompts based on the legal provision that "the user's emotions should be taken into consideration using a generative AI model."

[0726] Thus, the present invention enables the creation and scoring of exams that reflect the individual emotional states of users and test takers, thereby providing a personalized educational environment.

[0727] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0728] Step 1:

[0729] The terminal receives input from the user and obtains the user's intent for the question. Simultaneously, it collects emotional data through sensors and interfaces. Specifically, it records the themes and keywords entered by the user, as well as the emotional state inferred from keyboard typing speed and mouse movements. This input data is collected as basic information necessary for analysis on the server.

[0730] Step 2:

[0731] The terminal transmits the collected question intent and emotional data to the server. After receiving this data, the server passes it to an emotional analysis device for analysis. Using natural language processing technology and machine learning algorithms, the device evaluates the user's emotional state and determines whether the user is experiencing stress. The results of this analysis are then used in the subsequent process of adjusting the difficulty level of the test questions.

[0732] Step 3:

[0733] The server takes the sentiment analysis results into account and uses a generative AI model to extract relevant topics that match the intent of the question. Specifically, the AI ​​model uses prompt sentences to generate specific topics based on the given theme. These prompt sentences are such as, "The AI ​​will generate questions based on the user's sentiment about the selected topic."

[0734] Step 4:

[0735] The server generates test questions that take sentiment data into account, based on the extracted topics. If the analysis indicates that the user is experiencing stress, the difficulty level of the questions is appropriately adjusted before generation. In this process, the question text is created based on the topics generated by the AI, and a consistency check is performed.

[0736] Step 5:

[0737] The server verifies the difficulty level and logical consistency of the generated exam questions, and then provides the user with the exam questions and predicted scoring criteria. It also provides flexible editing tools through an interface, allowing users to modify the questions and scoring criteria as needed.

[0738] Step 6:

[0739] After completing the exam, the user submits their answers to the server. The server evaluates the received answers according to scoring criteria and adjusts the feedback based on the results of sentiment analysis. In particular, if the server determines that the test-taker is experiencing excessive stress, the feedback is optimized to emphasize positive aspects.

[0740] (Application Example 2)

[0741] Next, we will explain application example 2. In the following explanation, 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."

[0742] Traditional exam question creation and scoring systems have a problem in that they do not take into account the emotional state of test-takers, and in particular, they do not adequately consider the impact of stress and anxiety on test-takers' performance. As a result, there is a problem in conducting flexible and individualized exams that respond to the diverse needs and emotions of learners.

[0743] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0744] In this invention, the server includes means for receiving user input and analyzing the intent behind the questions based on the input; means for referring to past test data and generating question ideas based on the analysis results; means for creating specific question sentences from the generated question ideas and checking their logical consistency; and means for analyzing the emotional state of the user using a device for recognizing emotions and adjusting the difficulty level and content of the questions based on that analysis. This makes it possible to provide a flexible and individualized test that takes into account the emotional state of the test taker.

[0745] "Means for receiving user input and analyzing the intent behind the questions based on that input" refers to a process of receiving input data from users, analyzing that data, and interpreting what purpose and direction questions should be created in.

[0746] "A means of generating problem ideas based on analysis results by referring to past test data" refers to a process of creating new test question ideas based on the analyzed intent of previously conducted tests.

[0747] "A means of creating a specific problem statement from generated problem ideas and checking its logical consistency" refers to the process of developing a problem designed at the idea level into a detailed statement and verifying whether its content is logically consistent.

[0748] "A means of analyzing a user's emotional state using a device for recognizing their emotions and adjusting the difficulty level and content of the questions based on that analysis" refers to a process that utilizes emotion recognition technology to analyze the user's psychological state and adaptively adjusts the content and difficulty level of the exam questions based on that information.

[0749] "Providing a flexible and individualized examination that takes into account the emotional state of the test-taker" means customizing the test questions and evaluation methods according to the emotions and psychological state of each individual test-taker, thereby providing an appropriate examination experience.

[0750] The system implementing this invention mainly consists of a server, a terminal, and an emotion analysis device. The server receives input from the user and analyzes the intent behind the question based on the input information. To support this analysis, it refers to past test data and extracts relevant information to improve the accuracy of the analysis. Based on the analysis results, the server uses a generative AI model to generate new question ideas and outputs them as specific question statements.

[0751] Simultaneously, the device acts as the interface with the user and collects data on the user's emotional state. This utilizes hardware that supports emotion recognition technology, such as cameras and microphones. Specifically, the Emotion Recognition library is used for emotion recognition. The collected emotional data is sent to a server and processed by an emotion analysis device.

[0752] The emotion analysis device analyzes this data to determine the user's emotional state. This result is used as input to adaptively adjust the difficulty level and content of the questions. For example, if it detects that the user is nervous, measures are taken to reduce the user's psychological burden, such as lowering the difficulty level of the questions or simplifying the introduction.

[0753] Even after the exam ends, the server collects the test-taker's answers and provides feedback based on sentiment analysis results. This feedback helps users understand themselves and provides a more personalized learning experience by including advice for future learning.

[0754] As a concrete example, in an exam conducted at a certain educational institution, the system can utilize the examinee's smartphone as a terminal to analyze their emotions in real time and enable flexible exam management tailored to the situation. An example of a prompt message could be, "Develop a system that sets the optimal difficulty level of exam questions based on the user's emotions."

[0755] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0756] Step 1:

[0757] The device receives input data from the user. This input includes data related to the intent of the test questions and the user's emotional state. The device uses sensors such as a camera and microphone to collect the user's facial expressions and voice tone, and generates data to recognize their emotional state.

[0758] Step 2:

[0759] The terminal sends the collected user input data and emotional state data to the server. The transmitted data serves as foundational data for analyzing the user's intent behind the questions and determining how emotional information influences them.

[0760] Step 3:

[0761] The server analyzes past exam data based on the received question intent and emotional state data. During the analysis, data calculations are performed to identify how the question intent relates to the user's emotions. Using these analysis results, a generative AI model is utilized to generate question ideas.

[0762] Step 4:

[0763] The server creates specific problem statements from the generated problem ideas and checks for logical consistency. This is the problem statement generation process, which aims to create problems of appropriate difficulty and content that reflect emotional information and align with the intent of the question.

[0764] Step 5:

[0765] The server finalizes the question and its scoring criteria, and sends them back to the terminal. The terminal then presents this to the user and receives user feedback as needed. This feedback is valuable input for the system to make further adjustments based on the sentiment analysis results.

[0766] Step 6:

[0767] Once a user submits their test answers, the device sends those answers to the server. The server evaluates the answers based on scoring criteria and generates feedback that takes into account the user's emotional state.

[0768] Step 7:

[0769] The server provides emotionally sensitive feedback back to the device. The device then presents this feedback to the user, helping them deepen their understanding as they move towards the next learning step. This allows the user to experience a personalized test that is tailored to their emotions and learning progress.

[0770] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0771] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0772] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0773] [Fourth Embodiment]

[0774] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0775] As shown in Figure 7, the 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.

[0776] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0777] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0778] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0779] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0780] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0781] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0782] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0783] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0784] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0785] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0786] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0787] The present invention relates to an AI-powered system for creating and grading written examination questions, and its embodiments are described below.

[0788] overview

[0789] This system consists of three main components: a server, a terminal, and a user. An AI model running on the server creates questions based on the user's intended question, and also creates scoring criteria and scores the test takers' answers.

[0790] Details of each component

[0791] server

[0792] Problem creation function

[0793] The server receives the user's submitted question intent and analyzes it using natural language processing (NLP) techniques to extract relevant topics and keywords. Based on this information, an AI model generates appropriate question ideas while referring to a database of past exams. It then generates specific question texts from the selected ideas and checks their logical consistency and readability. It also has a function to evaluate the characteristics of the questions and estimate their difficulty level.

[0794] Scoring support function

[0795] When the server receives an ideal answer from a user, it analyzes it to create scoring criteria. Based on these criteria, it scores the examinee's written answer. The AI ​​model evaluates the accuracy and relevance of the answer through syntactic and semantic analysis.

[0796] terminal

[0797] Interface function

[0798] The terminal provides an interface for users to access the system, input question intents, check questions and scoring criteria, and view scoring results. The UI is intuitive and designed for easy user operation.

[0799] User

[0800] Providing the intent behind the questions

[0801] Users provide the server with the abilities they wish to evaluate and the themes they wish to address via their device, and request the creation of problems based on this information.

[0802] Result confirmation and feedback

[0803] Users can view the questions, scoring criteria, and test results provided by the server through their terminals and provide feedback for improvement as needed.

[0804] Presentation of specific examples

[0805] For example, if a user wants to create an exam question on the theme of "assessing economic analytical skills related to sustainability," the user sends this intention to the server using their device. The server uses AI to identify relevant topics and generates questions such as, "Discuss the impact of sustainability policies on economic growth." The user then uploads their own example answer, and the server forms scoring criteria based on that answer and evaluates the answers of actual test takers.

[0806] This process enables efficient and fair testing without relying on specialized knowledge.

[0807] The following describes the processing flow.

[0808] Step 1:

[0809] Users use their devices to input the intent behind the questions and the objectives of the exam. By specifically describing the themes and abilities they wish to evaluate, the subsequent AI analysis will proceed more smoothly.

[0810] Step 2:

[0811] The terminal sends user input to the server. The transmitted data is stored on the server and prepared for use in subsequent analysis processes.

[0812] Step 3:

[0813] The server analyzes the received question intent using natural language processing (NLP) techniques. This extracts relevant keywords and topics, providing guidance for creating questions.

[0814] Step 4:

[0815] The server references a database of past exams based on the extracted keywords, and the AI ​​model generates appropriate problem ideas. This process verifies whether the ideas match the user's objectives.

[0816] Step 5:

[0817] The server generates a specific problem statement based on the selected problem idea. The generated problem statement is automatically checked for logical consistency and appropriate language.

[0818] Step 6:

[0819] The server estimates the difficulty level of the generated problem based on evaluation metrics. This numerically represents how challenging the problem is for the test-taker.

[0820] Step 7:

[0821] Users upload ideal answer examples to the server via their devices. These answers serve as the basic information necessary for creating the scoring criteria.

[0822] Step 8:

[0823] The server analyzes the user's ideal answer and automatically creates scoring criteria based on it. These criteria serve as a guideline for evaluation when test-takers answer the questions.

[0824] Step 9:

[0825] The device presents the completed problem statement and grading criteria to the user. At this stage, the user can review the content and provide corrections or feedback if necessary.

[0826] Step 10:

[0827] The server collects test-takers' answers during the exam and prepares them for analysis. This data is used for subsequent scoring.

[0828] Step 11:

[0829] The server evaluates the test-taker's answers based on scoring criteria and uses AI to determine the accuracy and relevance of each answer. The results are calculated as a score.

[0830] Step 12:

[0831] The terminal displays the scoring results to the user and provides information to obtain feedback useful for exam administration. The user evaluates the system's effectiveness through this information and uses it to create future exams.

[0832] (Example 1)

[0833] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0834] Traditional essay-style examination question creation and grading processes require significant time and expertise, making it difficult to ensure efficiency and fairness. In particular, consistently establishing standards to ensure the quality of exam questions and fairly evaluate examinees' responses is challenging.

[0835] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0836] In this invention, the server includes means for receiving user input and analyzing the intent based on the input, means for referring to past test information and generating problem ideas based on the analysis results, and means for creating specific problems from the generated ideas and checking their logical consistency. This makes it possible to improve the efficiency and fairness of the question creation and scoring process for written examinations.

[0837] "User input" refers to information and instructions provided by system users via their terminals, and serves as the starting point for creating exam questions and evaluating answers.

[0838] "Means for analyzing intent" refers to a system that uses natural language processing technology to analyze information provided by users and identify their purpose and specific needs.

[0839] "Past exam information" refers to historical questions and sample answers stored in existing exam databases, which are used as reference when creating new questions.

[0840] "Means of generating problem ideas" refers to the process of devising an outline of a problem, including related themes, based on the analyzed intent.

[0841] "Methods for creating specific problems" refer to the steps taken to construct a detailed and clear problem statement based on the generated ideas.

[0842] A "means of checking logical consistency" refers to a mechanism for evaluating whether a problem statement is logically consistent and free from contradictions.

[0843] An "ideal answer" is a model answer provided by a user, serving as a reference example for forming the scoring criteria.

[0844] "Means of creating evaluation criteria" refers to the process of establishing criteria for fairly evaluating test-takers' answers, based on ideal responses.

[0845] "Syntactic analysis" refers to the process of analyzing and understanding a test-taker's answers based on their grammatical structure.

[0846] "Semantic analysis" is the process of interpreting the content of a test-taker's answers and evaluating their intent and relevance.

[0847] "Means for evaluating responses" refers to a system that measures the quality of test-takers' responses based on syntactic and semantic analysis.

[0848] Specific embodiments of this invention will now be described.

[0849] This system consists of three main components: servers, terminals, and users.

[0850] server

[0851] The server operates on a cloud server with powerful computing capabilities. This server receives the intent of the question submitted by the user through their terminal and analyzes it using natural language processing (NLP) technology. The server has a generative AI model installed that accesses a database containing past exam information and extracts relevant topics based on the input intent. The AI ​​model then generates question ideas and creates specific question texts based on this. This process checks the logical consistency of the questions and selects the most suitable questions. Furthermore, it creates evaluation criteria based on the ideal answers received from the user and scores the answers obtained by the test taker. Mechanisms are in place to evaluate answers in detail through syntactic and semantic analysis.

[0852] terminal

[0853] The terminal provides the interface for users to access the system. This includes PCs and tablets used by users, which connect to the server via a web browser. The terminal is designed so that users can input the intent of the questions and ideal answers, and then check the questions, evaluation criteria, and scoring results generated by the server. The interface is also intuitive and designed with ease of operation in mind.

[0854] User

[0855] Users are the ones who use terminals to input intentions and themes in order to support the process of creating and grading exam questions. For example, if a user wants to create an exam question on the theme of "evaluating economic analytical skills related to sustainability," they use their terminal to send that intention to the server. Based on this, the server generates a question such as "Discuss the impact of sustainability policies on economic growth." The user then uploads their ideal answer to the server, and the server evaluates the test-taker's answer based on scoring criteria derived from that answer. This entire process ensures highly efficient and fair examination administration.

[0856] Example of a prompt

[0857] "Please create test questions to assess economic analytical skills related to sustainability."

[0858] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0859] Step 1:

[0860] Users input the intent behind the questions using their devices. Specifically, they use a web interface input form to describe in detail the abilities or themes they wish to evaluate. The data entered is primarily in text format and is transferred from the device to the server.

[0861] Step 2:

[0862] The server analyzes the question intent received from the user using natural language processing technology. This analysis extracts the intended evaluation points and topics. During the processing, the generative AI model analyzes keywords and main themes from the text and identifies relevant information. As a result, the analyzed data forms the basis for question generation.

[0863] Step 3:

[0864] The server generates question ideas based on analyzed topic information and references past exam data. It utilizes a generative AI model to suggest relevant questions and then creates specific question texts based on these suggestions. The output data, including a summary of the questions and specific question formats, is prepared for further evaluation.

[0865] Step 4:

[0866] The server checks the logical consistency and difficulty level of the generated problem statements. The AI ​​model analyzes whether the problems are consistent and of appropriate difficulty. Here, the quality of the problems is checked according to the set criteria, and the optimal content is selected. The checked problem statements are output and ready for user review.

[0867] Step 5:

[0868] Users review the question and its evaluation criteria presented by the server via their terminal. If necessary, users can request revisions to the content. A crucial step in the terminal review process is re-examining the question and approving the final question content.

[0869] Step 6:

[0870] Users input ideal answer examples from their terminals into the server. These example answers are required to be concise and appropriate, based on the intent of the question. The input data is then used to generate scoring criteria.

[0871] Step 7:

[0872] The server analyzes the ideal answers provided by users and creates scoring criteria based on them. It utilizes a generative AI model to identify key elements of the answers. Syntactic and semantic analysis are used to extract evaluation points for the answers and establish fair scoring criteria. These criteria are used as indicators when evaluating test-takers' answers.

[0873] Step 8:

[0874] The server receives the test taker's answers and evaluates them based on pre-defined scoring criteria. Using syntactic and semantic analysis, it determines how closely the test taker's answers resemble the ideal answers. A final evaluation result is generated and provided to the user as feedback. This feedback can be used to review the test results and improve future test designs.

[0875] (Application Example 1)

[0876] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0877] In educational settings, creating and grading essay-style examinations is a task that requires specialized knowledge and time, placing a significant burden on educators. Furthermore, automation is difficult, and many challenges remain in ensuring efficient and fair evaluation. This invention aims to solve these problems and improve the quality of education.

[0878] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0879] In this invention, the server includes means for receiving user input and analyzing the intent behind the questions, means for generating question ideas by referring to past test data, and means for automatically generating questions using a generation AI model. This enables efficient and fair creation and scoring of written test questions on an educational equipment platform.

[0880] "User input" refers to the information and instructions that users provide to the system.

[0881] "Exam intent" refers to the user's intentions regarding the purpose and theme of the exam questions they wish to create.

[0882] "Analysis" refers to the process of analyzing input information and extracting necessary data and patterns.

[0883] "Past exam data" refers to information about previously used exam questions and their results.

[0884] "Problem idea" refers to a hypothesis or draft of an exam question based on the examiner's intent.

[0885] A "generative AI model" refers to an algorithm or system that uses artificial intelligence to automatically create problems.

[0886] "Specific problem statement" refers to the written content of the questions used in the exam.

[0887] "Logical consistency" refers to a state in which the problem statement and the answer are logically consistent and free from contradictions.

[0888] "Difficulty level" refers to a measure that indicates how challenging an exam question is.

[0889] An "ideal answer" refers to a model and optimal response to an exam question.

[0890] "Scoring criteria" refers to the standards or criteria used to evaluate a test-taker's answers.

[0891] "Voice input" refers to a method of capturing audio as data through a device such as a microphone.

[0892] "Text data" refers to character information recorded in a format that can be processed by a computer.

[0893] An "educational equipment platform" refers to a hardware and software environment used for learning and educational purposes.

[0894] "Evaluation" refers to the process of reviewing the test takers' answers and performance and determining their value and quality.

[0895] "Feedback" refers to information and advice for improvement provided to users or test takers based on evaluation results.

[0896] The system for carrying out this invention operates on an educational equipment platform. Servers, terminals, and users are the main components.

[0897] On the server, a generative AI model is used to automatically generate questions based on the user's intended question. Specifically, the user communicates the exam theme to the server via voice or text input. For voice input, the server uses the Google Speech-to-Text API to convert the voice data into text data. Next, NLP technology is used to analyze the intended question and refer to relevant past exam data. In this data analysis process, the AI ​​model infers and generates appropriate question ideas.

[0898] The terminal provides the user interface for accessing this system. Through an intuitive UI, users can review the generated problem statements and scoring criteria and suggest revisions as needed.

[0899] Users aim to improve the quality of education through this system. In particular, voice input allows for the provision of question intent under various conditions, enabling immediate question creation and grading in different situations. This flexibility helps to meet the diverse needs of educational settings and reduces the burden on educators.

[0900] For example, if a user wants to create an exam question on "environmental issues," the server will automatically generate a question such as "Discuss the relationship between environmental issues and economic policy." OpenAI's GPT series is used for analysis, enabling smooth question generation. An example of a prompt message is: "Generate an exam question on 'environmental issues' using OpenAI GPT-3. Please grade the answer considering semantics and syntax."

[0901] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0902] Step 1:

[0903] Users input the exam question's theme and intent via their device. Input can be in voice or text format; voice input is converted to text using the Google Speech-to-Text API. This process involves analyzing the voice data and converting it to text data, with the output being text data including the intent behind the question.

[0904] Step 2:

[0905] The server analyzes the input question intent using natural language processing (NLP) techniques. Specifically, it uses an NLP library (e.g., spaCy) to syntactically analyze the text data and extract important keywords and themes. This process outputs structured data of the analyzed question intent.

[0906] Step 3:

[0907] The server generates problem ideas based on extracted keywords and themes, utilizing a generative AI model (e.g., OpenAI GPT-3). Here, the AI ​​model combines relevant information to create a draft problem that matches the intended question. This process outputs a preliminary draft of the generated problem.

[0908] Step 4:

[0909] Based on the generated problem ideas, the server creates a specific problem statement. The automatically generated problem statement is then checked for logical consistency and difficulty. At this stage, the text data of the problem statement is output, along with evaluation values ​​for its consistency and difficulty.

[0910] Step 5:

[0911] The server analyzes the ideal answers provided by the user and creates scoring criteria. At this stage, NLP techniques are again used to analyze the syntax and meaning of the ideal answers. This process outputs structured data to be used as scoring criteria.

[0912] Step 6:

[0913] Users review the completed question text and scoring criteria on their devices and suggest revisions if necessary. Based on user feedback, data is then corrected, and the final revised question text and scoring criteria are output.

[0914] Step 7:

[0915] In educational settings, test takers submit their answers, and the server collects these answers. The collected answers are evaluated based on previously established scoring criteria. This results in an output score for each answer.

[0916] Step 8:

[0917] Based on the evaluation results, the server provides feedback to the user via the terminal. In this step, text data showing the results and areas for improvement is generated and output as feedback in a format that the user can review.

[0918] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0919] This invention aims to achieve more sophisticated exam management by combining an emotion engine with an AI-powered system for creating and grading written exam questions. This emotion engine is characterized by its ability to analyze the user's emotional state and adjust each process based on that information.

[0920] overview

[0921] The system consists of a server, terminals, and an emotion engine. The emotion engine analyzes the user's emotional state and uses the resulting data to adjust each process accordingly.

[0922] Details of each component

[0923] server

[0924] Problem creation support

[0925] The server processes the question intent received from the user, along with data including the sentiment analysis results from the sentiment engine, and extracts relevant topics. This process ensures that the question intent is appropriately interpreted based on the user's emotions, helping the AI ​​model accurately generate question ideas.

[0926] Grading support and feedback

[0927] By incorporating emotional information from the emotion engine into the evaluation of test responses, it is possible to optimize scoring criteria and feedback content. This allows for more flexible and personalized feedback.

[0928] terminal

[0929] Collection of interface and sentiment data

[0930] The terminal provides an interface for the user to access the system and collects data to obtain the user's emotional state. This data is analyzed by an emotion engine and sent to the server.

[0931] Emotional Engine

[0932] Emotional analysis and process adjustment

[0933] The emotion engine analyzes the user's emotions based on data transmitted from the device and reflects the results in each process. Specifically, it is used to adjust the difficulty level when creating questions and to adaptively adjust the scoring criteria.

[0934] Presentation of specific examples

[0935] For example, if a user creates a test to "assess critical thinking skills regarding environmental issues," and the emotion engine analyzes that the user is experiencing stress, the server will take that information into account and present questions with slightly adjusted difficulty levels. In this way, tests that take into account the user's emotions and mental state are constructed. After the test, when grading the test-takers' answers, the feedback is also adjusted based on the data obtained from the emotion analysis, helping test-takers accept the results in a way that is more easily understood and accepted.

[0936] This format makes it possible to provide users and test takers with a more personalized and effective testing experience.

[0937] The following describes the processing flow.

[0938] Step 1:

[0939] Users use a terminal to input the intent behind the questions regarding the themes and abilities they wish to evaluate. The input requires a detailed description of the test's purpose and target areas.

[0940] Step 2:

[0941] The device collects emotional data simultaneously with user input. This emotional data includes information estimated from factors such as keyboard typing speed and eye movements.

[0942] Step 3:

[0943] The device transmits question intent and emotional data to the server. The transmission is done in real time and is designed to respond to changes in emotions.

[0944] Step 4:

[0945] The server analyzes the received question intent using natural language processing (NLP). During this process, it references sentiment data provided by the sentiment engine to supplement the interpretation with one that reflects the user's mental state.

[0946] Step 5:

[0947] Based on the information obtained from the analysis, the server generates multiple problem ideas using an AI model. It then utilizes emotional data to select the idea that best suits the user's emotional state from among the generated ideas.

[0948] Step 6:

[0949] The server creates specific problem statements based on the selected problem ideas. During this process, the difficulty level is adjusted considering the emotion engine data, ensuring the content is adapted to the user's state.

[0950] Step 7:

[0951] The device presents the user with the completed problem statement and scoring criteria that reflect sentiment data. The user can review the content and provide feedback as needed.

[0952] Step 8:

[0953] When the test is administered, the device collects not only the test-taker's answers but also emotional data from the test. This data, along with the answers, is sent to the server.

[0954] Step 9:

[0955] The server uses the collected responses and sentiment data to perform scoring. The sentiment engine assists in evaluating the responses and calculates an appropriate score based on the scoring criteria.

[0956] Step 10:

[0957] The device presents test takers and users with emotionally-based feedback along with their scoring results. This feedback includes advice that reflects emotional elements to aid the test taker's understanding.

[0958] (Example 2)

[0959] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0960] Traditional exam creation and scoring systems have the drawback of not considering the emotional state of users and test-takers, making it difficult to provide a personalized exam experience. Furthermore, because the questions and scoring are standardized, the diverse learning styles and emotions of test-takers do not significantly influence the results.

[0961] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0962] In this invention, the server includes means for collecting and analyzing user emotional data, means for adjusting the difficulty level and scoring criteria of questions based on the emotional analysis results, and means for adjusting the content of feedback based on the emotional analysis results. This enables the creation of personalized tests and flexible scoring and feedback that are tailored to the emotional state of the user and test taker.

[0963] "User input" refers to the information and instructions that users provide to the system, including the intent behind the questions necessary for creating the exam.

[0964] "Exam intent" refers to the purpose and criteria that the user expresses when creating exam questions.

[0965] "Past data" refers to information and statistics collected previously, which are used in the generation and evaluation of exam questions.

[0966] An "idea" refers to a concept or theme for giving concrete form to the exam questions.

[0967] "Checking sentence consistency" is the process of verifying whether the generated problem sentences are grammatically and logically correct.

[0968] "Difficulty level" is an indicator that shows the complexity and ease of solving exam questions.

[0969] "Criteria for analyzing answers" are standard guidelines used to evaluate the answers of test-takers.

[0970] "Emotional data" refers to information that indicates a user's emotional state, and is collected and used by the system.

[0971] "Sentiment analysis" is the process of evaluating collected emotional data to understand the user's emotional state.

[0972] "Adjusting the feedback content" refers to the process of optimizing responses and comments to test-takers' answers based on the results of sentiment analysis.

[0973] This invention provides a test creation and scoring support system that utilizes AI technology, including user emotion analysis. This system consists of a server, terminals, and emotion analysis devices, which function through mutual cooperation.

[0974] The terminal provides the user with an interface for creating exams. Through this interface, the user inputs the intent behind the questions, and simultaneously, data for sentiment analysis is collected. For example, keyboard input speed, mouse movements, or direct user feedback can be used. The data obtained in this way is essential for capturing the user's real-time emotional state.

[0975] The server receives data transmitted from the terminal and analyzes the user's input regarding the intent behind the questions and emotional data. Emotional analysis is performed by an emotional analysis device, utilizing natural language processing and machine learning techniques. This analysis determines the user's stress level and agitation state, which is used to adjust the difficulty of the test. The server then uses a generative AI model to extract relevant topics that align with the intent behind the questions and creates questions based on those topics.

[0976] For example, if a user creates a test to "evaluate critical thinking skills regarding environmental issues," the system will determine the user's stress level through sentiment analysis and adjust the difficulty level appropriately. This allows for a flexible test format tailored to the user's state. Furthermore, when generating test questions, the AI ​​model applies appropriate prompts based on the legal provision that "the user's emotions should be taken into consideration using a generative AI model."

[0977] Thus, the present invention enables the creation and scoring of exams that reflect the individual emotional states of users and test takers, thereby providing a personalized educational environment.

[0978] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0979] Step 1:

[0980] The terminal receives input from the user and obtains the user's intent for the question. Simultaneously, it collects emotional data through sensors and interfaces. Specifically, it records the themes and keywords entered by the user, as well as the emotional state inferred from keyboard typing speed and mouse movements. This input data is collected as basic information necessary for analysis on the server.

[0981] Step 2:

[0982] The terminal transmits the collected question intent and emotional data to the server. After receiving this data, the server passes it to an emotional analysis device for analysis. Using natural language processing technology and machine learning algorithms, the device evaluates the user's emotional state and determines whether the user is experiencing stress. The results of this analysis are then used in the subsequent process of adjusting the difficulty level of the test questions.

[0983] Step 3:

[0984] The server takes the sentiment analysis results into account and uses a generative AI model to extract relevant topics that match the intent of the question. Specifically, the AI ​​model uses prompt sentences to generate specific topics based on the given theme. These prompt sentences are such as, "The AI ​​will generate questions based on the user's sentiment about the selected topic."

[0985] Step 4:

[0986] The server generates test questions that take sentiment data into account, based on the extracted topics. If the analysis indicates that the user is experiencing stress, the difficulty level of the questions is appropriately adjusted before generation. In this process, the question text is created based on the topics generated by the AI, and a consistency check is performed.

[0987] Step 5:

[0988] The server verifies the difficulty level and logical consistency of the generated exam questions, and then provides the user with the exam questions and predicted scoring criteria. It also provides flexible editing tools through an interface, allowing users to modify the questions and scoring criteria as needed.

[0989] Step 6:

[0990] After completing the exam, the user submits their answers to the server. The server evaluates the received answers according to scoring criteria and adjusts the feedback based on the results of sentiment analysis. In particular, if the server determines that the test-taker is experiencing excessive stress, the feedback is optimized to emphasize positive aspects.

[0991] (Application Example 2)

[0992] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0993] Traditional exam question creation and scoring systems have a problem in that they do not take into account the emotional state of test-takers, and in particular, they do not adequately consider the impact of stress and anxiety on test-takers' performance. As a result, there is a problem in conducting flexible and individualized exams that respond to the diverse needs and emotions of learners.

[0994] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0995] In this invention, the server includes means for receiving user input and analyzing the intent behind the questions based on the input; means for referring to past test data and generating question ideas based on the analysis results; means for creating specific question sentences from the generated question ideas and checking their logical consistency; and means for analyzing the emotional state of the user using a device for recognizing emotions and adjusting the difficulty level and content of the questions based on that analysis. This makes it possible to provide a flexible and individualized test that takes into account the emotional state of the test taker.

[0996] "Means for receiving user input and analyzing the intent behind the questions based on that input" refers to a process of receiving input data from users, analyzing that data, and interpreting what purpose and direction questions should be created in.

[0997] "A means of generating problem ideas based on analysis results by referring to past test data" refers to a process of creating new test question ideas based on the analyzed intent of previously conducted tests.

[0998] "A means of creating a specific problem statement from generated problem ideas and checking its logical consistency" refers to the process of developing a problem designed at the idea level into a detailed statement and verifying whether its content is logically consistent.

[0999] "A means of analyzing a user's emotional state using a device for recognizing their emotions and adjusting the difficulty level and content of the questions based on that analysis" refers to a process that utilizes emotion recognition technology to analyze the user's psychological state and adaptively adjusts the content and difficulty level of the exam questions based on that information.

[1000] "Providing a flexible and individualized examination that takes into account the emotional state of the test-taker" means customizing the test questions and evaluation methods according to the emotions and psychological state of each individual test-taker, thereby providing an appropriate examination experience.

[1001] The system implementing this invention mainly consists of a server, a terminal, and an emotion analysis device. The server receives input from the user and analyzes the intent behind the question based on the input information. To support this analysis, it refers to past test data and extracts relevant information to improve the accuracy of the analysis. Based on the analysis results, the server uses a generative AI model to generate new question ideas and outputs them as specific question statements.

[1002] Simultaneously, the device acts as the interface with the user and collects data on the user's emotional state. This utilizes hardware that supports emotion recognition technology, such as cameras and microphones. Specifically, the Emotion Recognition library is used for emotion recognition. The collected emotional data is sent to a server and processed by an emotion analysis device.

[1003] The emotion analysis device analyzes this data to determine the user's emotional state. This result is used as input to adaptively adjust the difficulty level and content of the questions. For example, if it detects that the user is nervous, measures are taken to reduce the user's psychological burden, such as lowering the difficulty level of the questions or simplifying the introduction.

[1004] Even after the exam ends, the server collects the test-taker's answers and provides feedback based on sentiment analysis results. This feedback helps users understand themselves and provides a more personalized learning experience by including advice for future learning.

[1005] As a concrete example, in an exam conducted at a certain educational institution, the system can utilize the examinee's smartphone as a terminal to analyze their emotions in real time and enable flexible exam management tailored to the situation. An example of a prompt message could be, "Develop a system that sets the optimal difficulty level of exam questions based on the user's emotions."

[1006] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1007] Step 1:

[1008] The device receives input data from the user. This input includes data related to the intent of the test questions and the user's emotional state. The device uses sensors such as a camera and microphone to collect the user's facial expressions and voice tone, and generates data to recognize their emotional state.

[1009] Step 2:

[1010] The terminal sends the collected user input data and emotional state data to the server. The transmitted data serves as foundational data for analyzing the user's intent behind the questions and determining how emotional information influences them.

[1011] Step 3:

[1012] The server analyzes past exam data based on the received question intent and emotional state data. During the analysis, data calculations are performed to identify how the question intent relates to the user's emotions. Using these analysis results, a generative AI model is utilized to generate question ideas.

[1013] Step 4:

[1014] The server creates specific problem statements from the generated problem ideas and checks for logical consistency. This is the problem statement generation process, which aims to create problems of appropriate difficulty and content that reflect emotional information and align with the intent of the question.

[1015] Step 5:

[1016] The server finalizes the question and its scoring criteria, and sends them back to the terminal. The terminal then presents this to the user and receives user feedback as needed. This feedback is valuable input for the system to make further adjustments based on the sentiment analysis results.

[1017] Step 6:

[1018] Once a user submits their test answers, the device sends those answers to the server. The server evaluates the answers based on scoring criteria and generates feedback that takes into account the user's emotional state.

[1019] Step 7:

[1020] The server provides emotionally sensitive feedback back to the device. The device then presents this feedback to the user, helping them deepen their understanding as they move towards the next learning step. This allows the user to experience a personalized test that is tailored to their emotions and learning progress.

[1021] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1022] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1023] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1024] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1025] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1026] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1027] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1028] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1029] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1030] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1031] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1032] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1033] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1034] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1035] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1036] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1037] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1038] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1039] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1040] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1041] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[1042] The following is further disclosed regarding the embodiments described above.

[1043] (Claim 1)

[1044] A means for receiving user input and analyzing the intent behind the question based on that input,

[1045] A means of generating problem ideas based on analysis results by referring to past test data,

[1046] A method for creating specific problem statements from generated problem ideas and checking their logical consistency,

[1047] A method for estimating the difficulty level of the problem statement,

[1048] A method for analyzing ideal answers and creating scoring criteria,

[1049] A system that includes this.

[1050] (Claim 2)

[1051] The system according to claim 1, further comprising means for presenting the completed problem statement and scoring criteria to the user, and means for reflecting modifications as necessary.

[1052] (Claim 3)

[1053] The system according to claim 1, comprising means for collecting test answers, evaluating the answers based on scoring criteria, and providing feedback to the user based on the evaluation results.

[1054] "Example 1"

[1055] (Claim 1)

[1056] A means for receiving user input and analyzing the intent based on that input,

[1057] A means of generating problem ideas based on analysis results by referring to past test information,

[1058] A means of creating specific problems from generated ideas and checking their logical consistency,

[1059] A means of estimating the difficulty level of a problem,

[1060] A method for analyzing ideal answers and creating evaluation criteria,

[1061] A means for evaluating examinees' answers based on syntactic and semantic analysis,

[1062] A system that includes this.

[1063] (Claim 2)

[1064] The system according to claim 1, comprising means for presenting the user with the completed problem and evaluation criteria, and for reflecting modifications as necessary.

[1065] (Claim 3)

[1066] The system according to claim 1, comprising means for collecting test responses, evaluating the responses based on evaluation criteria, and providing feedback to the user based on the evaluation results.

[1067] "Application Example 1"

[1068] (Claim 1)

[1069] A means for receiving user input and analyzing the intent behind the question based on that input,

[1070] A means of generating problem ideas based on analysis results by referring to past test data,

[1071] A method for creating specific problem statements from generated problem ideas and checking their logical consistency,

[1072] A method for estimating the difficulty level of the problem statement,

[1073] A method for analyzing ideal answers and creating scoring criteria,

[1074] A means of analyzing voice input and converting it into text data,

[1075] A method for automatically generating problems using a generative AI model,

[1076] A means of providing feedback to users based on evaluation results,

[1077] A system that includes this.

[1078] (Claim 2)

[1079] The system according to claim 1, further comprising means for presenting the completed problem statement and scoring criteria to the user, means for reflecting modifications as necessary, and means for installation on an educational equipment platform.

[1080] (Claim 3)

[1081] The system according to claim 1, comprising means for collecting test answers and evaluating the answers based on scoring criteria, and means for converting students' speech into text using speech recognition technology and using that text for evaluation.

[1082] "Example 2 of combining an emotion engine"

[1083] (Claim 1)

[1084] A means for receiving user input and analyzing the intent behind the question based on that input,

[1085] A means of generating ideas based on past data and analysis results,

[1086] A means of creating concrete sentences from the generated ideas and checking their consistency,

[1087] A method for estimating the difficulty level of a sentence,

[1088] A means of analyzing the answers and creating criteria,

[1089] A means of collecting and analyzing user sentiment data,

[1090] A means of adjusting the difficulty level and scoring criteria of a problem based on the results of sentiment analysis,

[1091] A system that includes this.

[1092] (Claim 2)

[1093] The system according to claim 1, comprising means for presenting the user with a completed sentence and criteria, and for reflecting modifications as necessary.

[1094] (Claim 3)

[1095] The system according to claim 1, comprising means for collecting responses, evaluating them based on criteria, providing feedback based on the evaluation results, and adjusting the content of the feedback based on the sentiment analysis results.

[1096] "Application example 2 when combining with an emotional engine"

[1097] (Claim 1)

[1098] A means for receiving user input and analyzing the intent behind the question based on that input,

[1099] A means of generating problem ideas based on analysis results by referring to past test data,

[1100] A method for creating specific problem statements from generated problem ideas and checking their logical consistency,

[1101] A method for estimating the difficulty level of the problem statement,

[1102] A method for analyzing ideal answers and creating scoring criteria,

[1103] A means of analyzing the emotional state using a device to recognize the user's emotions and adjusting the difficulty level and content of the questions based on that analysis,

[1104] A system that includes this.

[1105] (Claim 2)

[1106] The system according to claim 1, further comprising means for presenting the user with the completed problem statement and scoring criteria, means for reflecting revisions as necessary, and means for providing feedback according to the user's emotional state.

[1107] (Claim 3)

[1108] The system according to claim 1, comprising means for collecting test answers, evaluating the answers based on scoring criteria, providing feedback to the user based on the evaluation results, and adjusting the content of the feedback according to the user's emotional state. [Explanation of symbols]

[1109] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for receiving user input and analyzing the intent behind the question based on that input, A means of generating problem ideas based on analysis results by referring to past test data, A method for creating specific problem statements from generated problem ideas and checking their logical consistency, A method for estimating the difficulty level of the problem statement, A method for analyzing ideal answers and creating scoring criteria, A means of analyzing voice input and converting it into text data, A method for automatically generating problems using a generative AI model, A means of providing feedback to users based on evaluation results, A system that includes this.

2. The system according to claim 1, further comprising means for presenting the completed problem statement and scoring criteria to the user, means for reflecting modifications as necessary, and means for installation on an educational equipment platform.

3. The system according to claim 1, comprising means for collecting test answers and evaluating the answers based on scoring criteria, and means for converting students' speech into text using speech recognition technology and using that text for evaluation.

Images