system

Abstract

Provide a system. A camera means of a terminal for acquiring input data; processing means at the terminal for extracting text from the image using optical character recognition techniques; a communication means for transmitting the extracted text data to a server; a receiving means for receiving text data by the server; an analysis means of a server that inputs text data into a natural language processing model for analysis; An evaluation means for calculating the probability that the sentence is an AI-generated sentence based on the analysis result; a result transmission means for transmitting the evaluation result to the terminal via a communication means; Display means for displaying evaluation results on the terminal A system including:

Description

[Technical Field]

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

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

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

[0004] In recent years, advances in AI technology have led to an increase in cheating, with students using AI to generate book reports and university assignments. This has made it difficult to properly evaluate students' academic abilities, resulting in a decline in the quality of education. To solve this problem, there is a need for technology that allows faculty and staff to quickly and accurately determine whether submitted writing has been generated by AI. Conventional methods require a great deal of time and expertise, making them impractical. [Means for solving the problem]

[0005] The present invention includes a camera means in the terminal that acquires input data, a processing means in the terminal that extracts text from the image using optical character recognition (OCR) technology, and a communication means that transmits the extracted text data to a server. The server further includes a receiving means that receives the text data, an analysis means in the server that inputs the text data into a natural language processing model for analysis, and an evaluation means that calculates the probability that the text is AI-generated based on the analysis results. Additionally, the system includes a result transmission means that transmits the evaluation results to the terminal via the communication means, and a display means on the terminal that displays the evaluation results. This configuration enables faculty and staff to quickly and accurately verify whether student submissions are AI-generated text.

[0006] "Input data" refers to image data acquired using the camera of the terminal.

[0007] The "camera means" refers to a camera device for taking an image including text information.

[0008] "Optical Character Recognition (OCR) technology" refers to technology that converts characters in an image into digital text.

[0009] "Processing means" refers to a processing device or software that applies OCR technology to acquired image data and extracts text data.

[0010] "Communication means" refers to the communication protocol and interface for transmitting the extracted text data from the terminal to the server.

[0011] A "server" is a computing device or set of computing devices that receives text data, analyzes it, and generates evaluation results.

[0012] The "receiving means" refers to a device or software that allows the server to receive text data sent from the terminal.

[0013] The "analysis means" refers to a processing device or software within the server that inputs received text data into a natural language processing model and analyzes it.

[0014] A "natural language processing model" is a machine learning model that analyzes text data and determines whether it is an AI-generated sentence.

[0015] "Evaluation means" refers to a processing device or software within the server that calculates the probability that a sentence is AI-generated based on the analysis results.

[0016] The "result transmission means" refers to a communication protocol and interface for the server to transmit the evaluation results to the terminal.

[0017] The "display means" refers to a display device and software for visually displaying to the user the evaluation results received by the terminal from the server.

[0018] "Formatting means" refers to a processing device or software for formatting and appropriately correcting text data extracted by OCR.

[0019] "Review generation means" refers to a processing device or software within the server for generating reviews in a user-understandable format.

[0020] A "secure protocol" is a communication standard and encryption method for ensuring secure communications. [Brief explanation of the drawings]

[0021] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0022] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0023] First, the terms used in the following description will be explained.

[0024] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0025] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0026] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0027] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0028] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0029] [First embodiment]

[0030] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0031] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0032] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0033] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0034] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0035] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0036] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0037] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0038] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0039] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0040] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0041] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0042] The present invention relates to a system that identifies whether book reports and university assignments submitted by students are AI-generated texts, which can be used by faculty and staff to quickly and accurately evaluate the submissions.

[0043] System Configuration

[0044] This system mainly consists of terminals, servers, and users. Terminals are mobile devices such as smartphones and tablets, and servers are computing resources deployed in a cloud environment. Users are primarily faculty and staff who use this system to evaluate student submissions.

[0045] Program processing

[0046] 1. The user captures input data using the smartphone camera.

[0047] The user launches the system's dedicated app and uses the camera function to take a photo of the book report paper.

[0048] The captured image is temporarily stored on the device.

[0049] 2. The device performs OCR processing to extract the text data.

[0050] The device performs OCR (optical character recognition) technology to extract text information from the captured image.

[0051] The extracted character information is stored in the terminal as text data.

[0052] 3. The device formats the text data and sends it to the server

[0053] The extracted text data is formatted on the device, and unnecessary line breaks, spaces, and misrecognized characters are corrected.

[0054] The formatted text data is sent to the server using a secure protocol (e.g., HTTPS).

[0055] 4. The server receives the text data and begins analyzing it.

[0056] The server receives the text data sent from the terminal.

[0057] The received text data is input into a natural language processing model and analysis begins.

[0058] 5. The server evaluates the analysis results and sends them to the device.

[0059] A natural language processing model analyzes the text data and calculates the probability that a sentence is AI-generated.

[0060] The server evaluates the analysis results and summarizes them in a format that is easy for the user to understand, such as "There is an 85% chance that this sentence was generated by an AI."

[0061] The evaluation results are then transmitted to the terminal again using a secure communication protocol.

[0062] 6. The device displays the evaluation results.

[0063] The terminal receives the analysis results sent from the server.

[0064] The received analysis results are displayed to the user, who can use the results to determine whether the student's submission is AI-generated.

[0065] Specific examples

[0066] For example, imagine a scenario in which a faculty member wants to evaluate a student's book report. The user launches a dedicated app and takes a photo of the report with their camera. The system extracts text from the image, formats it, and sends it to the server. Analysis is then performed on the server, and the result is determined to be "80% likely to be an AI-generated sentence." The user can review this result and, if necessary, conduct further research or provide feedback to the student.

[0067] In this way, the system provides faculty with advanced tools to effectively evaluate student work and helps prevent cheating.

[0068] The processing flow will be explained below.

[0069] Step 1:

[0070] The user activates the smartphone camera and takes a picture of the book report paper. The user then controls the camera through the app to capture the image with the appropriate framing.

[0071] Step 2:

[0072] The device saves the captured image and starts OCR processing. The device then analyzes the image and extracts the character information as text data.

[0073] Step 3:

[0074] The device then formats the extracted text data, specifically by removing unnecessary line breaks and spaces from the OCR results and correcting any misrecognized characters.

[0075] Step 4:

[0076] The device sends the formatted text data to the server. The device uses a secure communication protocol such as HTTPS to securely transmit the data to the server.

[0077] Step 5:

[0078] The server receives the text data sent from the device and prepares it for analysis. The server temporarily stores the received data for subsequent analysis.

[0079] Step 6:

[0080] The server inputs the text data into a natural language processing model and begins analysis, specifically to detect patterns and phrases characteristic of AI-generated text.

[0081] Step 7:

[0082] The server evaluates the analysis results and calculates the likelihood that the sentence is AI-generated. Based on the probability calculated by the AI ​​model, it generates an evaluation such as "high probability that the sentence is AI-generated" or "low probability."

[0083] Step 8:

[0084] The server formats the evaluation results and generates results in a format that is easy for users to understand. For example, it generates an evaluation result such as "There is an 80% chance that this sentence was generated by an AI."

[0085] Step 9:

[0086] The server transmits the evaluation results to the terminal, and again using a secure communication protocol, the server transmits the evaluation results to the terminal safely.

[0087] Step 10:

[0088] The terminal displays the evaluation results received from the server, and the terminal visually displays the evaluation results to the user, allowing the user to confirm the results.

[0089] Step 11:

[0090] Users can check the evaluation results and determine whether or not there has been any cheating. Based on the displayed evaluation results, users can conduct further detailed verification and provide feedback to students.

[0091] In this way, a system has been created in which users, devices, and servers work together to quickly and accurately evaluate whether student submissions are AI-generated texts.

[0092] Example 1

[0093] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0094] In recent years, advances in artificial intelligence technology have created a need for a fast and accurate way to assess whether student submissions are AI-generated. However, current systems rely heavily on manual review, which not only takes time and effort but also often lacks accuracy. Therefore, a method is needed for faculty and staff to efficiently assess submissions and prevent cheating.

[0095] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0096] In this invention, the server includes a photographing means of the terminal that acquires input data, an analysis means of the terminal that extracts characters from the image using optical character recognition technology, a communication means that transmits the extracted character data to the information processing device, a receiving means by the information processing device that receives the character data, an analysis means of the information processing device that inputs the character data into a natural language processing model and analyzes it, an evaluation means that calculates the probability that the sentence is an AI-generated sentence based on the analysis result, a result transmission means that transmits the evaluation result to the terminal via the communication means, and a display means on the terminal that displays the evaluation result. This enables faculty and staff to quickly and accurately evaluate whether a submitted work was generated by AI.

[0097] "Input data" is digital information that a user captures using a camera or other device.

[0098] A "terminal" is a device that is directly operated by a user, such as a smartphone, tablet, or PC.

[0099] "Photographing means" refers to the hardware and software, including the camera function, installed in the terminal.

[0100] "Optical character recognition technology" is a technology for reading and extracting text information from images.

[0101] "Analysis means" refers to the function of executing optical character recognition technology on the terminal and extracting character data from an image.

[0102] "Character data" refers to text-based data extracted using optical character recognition technology.

[0103] An "information processing device" is a device that performs data processing, including a server or cloud computing resource.

[0104] "Communication means" refers to an interface and protocol for transmitting and receiving data between a terminal and an information processing device.

[0105] The "receiving means" refers to a function by which the information processing device receives data transmitted from the terminal.

[0106] A "natural language processing model" is an algorithm and software that analyzes text data and determines whether it was generated by artificial intelligence.

[0107] The term "analysis means" refers to a function of an information processing device to input character data into a natural language processing model and analyze it.

[0108] "Evaluation means" refers to a function that calculates the probability that the input text was generated by AI based on the analysis results.

[0109] "Result transmission means" refers to a communication interface and protocol for transmitting evaluation results to a terminal.

[0110] "Display means" refers to an interface that allows the terminal to visually present the evaluation results to the user.

[0111] "Formatting means" refers to a function that formats extracted character data by correcting misrecognitions and deleting unnecessary line breaks and spaces.

[0112] The "evaluation generation means" refers to a function that summarizes and presents the analysis results in a format that is easy for the user to understand.

[0113] This invention relates to a system for identifying whether book reports and university assignments submitted by students have been generated by artificial intelligence. This system allows faculty and staff to quickly and accurately evaluate the submissions. The system mainly consists of a terminal, a server, and a user.

[0114] Hardware and software examples

[0115] Terminal

[0116] The terminal is a device that is directly operated by the user, such as a smartphone, tablet, or PC. The terminal is equipped with a camera, which is used to capture input data such as a book report. The terminal is equipped with optical character recognition (OCR) technology, such as Google® Tesseract OCR, to extract text data from images. The terminal also includes a communication means for transmitting the transmitted text data to a server. The HTTPS protocol is used for secure data communication.

[0117] server

[0118] The server is an information processing device that operates in a cloud environment and receives and analyzes text data sent from the terminal. The server has a receiving means and can receive data sent from the terminal. The received data is input into a natural language processing model (for example, "OpenAI's (registered trademark) GPT-4 (registered trademark)") for analysis. The analysis means inputs the text data into this natural language processing model and calculates the probability that the text was generated by AI. The evaluation means evaluates the analysis results and summarizes them in a format that is easy for users to understand. The evaluation results are then sent back to the terminal using the HTTPS protocol.

[0119] User

[0120] Users are primarily faculty and staff, who use the system to evaluate student submissions. Using a smartphone, tablet, or other device, users launch a dedicated app and take a picture of the submission. After taking the picture, text data extracted from the image is sent to a server, and the analysis results are displayed on the device. This allows users to quickly determine whether the submission was generated by artificial intelligence.

[0121] Examples of concrete examples and prompts

[0122] For example, imagine a scenario in which a faculty member wants to evaluate a student's book report. The user launches a dedicated app and takes a photo of the report with their camera. The system extracts text data from the image, formats it, and sends it to the server. Analysis is then performed on the server, and the result is that "there is an 80% chance that this sentence was generated by AI." The user can review this result and, if necessary, conduct further research or provide feedback to the student.

[0123] Examples of prompts include:

[0124] "Please rate whether the following sentences were generated by an AI:"

[0125] "Calculate the probability that the following text is AI-generated:"

[0126] This provides faculty with advanced tools to effectively assess student work and helps prevent cheating.

[0127] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0128] Step 1:

[0129] The user launches the system's dedicated app and takes a photo of the book report paper using the smartphone camera. The input data is an image of the paper, and the output is a digital image file that is temporarily stored inside the device. When the user taps the "take a photo" button, the camera starts and the image is saved.

[0130] Step 2:

[0131] The device analyzes the captured image using OCR (Optical Character Recognition) technology. The input data is the output image from step 1, and the output is the extracted character data. The OCR engine on the device analyzes the image and generates text data. Specifically, Google's Tesseract OCR library is used. The processed results are saved on the device as a text file.

[0132] Step 3:

[0133] The terminal formats the extracted text data. The input data is the output text from step 2, and the output is the formatted text data. Specifically, unnecessary line breaks and spaces are removed, and misrecognized characters are corrected. The formatted text data is converted to JSON format and is ready to be sent to the server.

[0134] Step 4:

[0135] The terminal sends the formatted text data to the server. The input data is the output text data of step 3, and the output is a data transmission success message to the server. The HTTPS protocol is used as the communication method, and the terminal sends the text data to the server in a secure state.

[0136] Step 5:

[0137] The server receives the text data sent from the terminal. The input data is the text data sent from the terminal, and the output is the received text data. The server's receiving means captures the data and stores it in a specific directory or database.

[0138] Step 6:

[0139] The server inputs the received text data into a natural language processing (NLP) model for analysis. The input data is the received text data from step 5, and the output is the analysis result. Specifically, a generative AI model such as OpenAI's GPT-4 is used to analyze whether the text data is AI-generated. In this process, a natural language processing algorithm is executed.

[0140] Step 7:

[0141] The server calculates the probability that the sentence is AI-generated based on the analysis results. The input data is the analysis result from step 6, and the output is an evaluation result including a probability value. Based on the characteristics of the analyzed text data and the prompt sentence, the server calculates an evaluation such as "There is an 85% chance that this sentence is AI-generated."

[0142] Step 8:

[0143] The server sends the evaluation result to the terminal. The input data is the evaluation result from step 7, and the output is a message to the terminal indicating that the transmission was successful. The evaluation result is formatted in a user-friendly format and sent to the terminal using the HTTPS protocol.

[0144] Step 9:

[0145] The device receives the evaluation results sent from the server and displays them to the user. The input data is the evaluation results sent from the server, and the output is the evaluation results displayed to the user. The device displays the results it receives on the screen, and the user is presented with a message saying, "There is an 85% chance that this sentence is AI-generated." Based on this information, the user can determine whether the submission was AI-generated.

[0146] This allows faculty and staff to quickly and accurately evaluate submissions through a series of processes, preventing cheating before it occurs.

[0147] (Application example 1)

[0148] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0149] With conventional systems, it was difficult to quickly and accurately determine whether student submissions or internal company documents were generated by AI, which caused problems for educational institutions and companies when verifying fraud and the authenticity of content. Furthermore, users could not obtain the results in real time, which led to a lack of efficiency. Furthermore, there were no concrete solutions using wearable devices.

[0150] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0151] In this invention, the server includes a camera means of the terminal that acquires input data, a processing means of the terminal that extracts text from images using optical character recognition (OCR) technology, a communication means that transmits the extracted text data to the server, a receiving means by the server that receives the text data, an analysis means of the server that inputs the text data into a natural language processing model for analysis, an evaluation means that calculates the probability that the text is an AI-generated sentence based on the analysis result, a result transmission means that transmits the evaluation result to the terminal via the communication means, a display means of the terminal that displays the evaluation result, a capture means that captures documents shared on the Internet or in internal documents and identifies whether they were generated by AI, and a wearable device that displays the evaluation result to the user in real time, thereby enabling the user to quickly and accurately obtain a judgment result on the AI-generated sentence.

[0152] The "camera means of the terminal for acquiring input data" refers to a means for capturing target documents or image data using a camera equipped on the terminal.

[0153] "Terminal processing means for extracting text from images using optical character recognition (OCR) techniques" means means for recognizing character information from a captured image and extracting it as text data.

[0154] The "communication means for transmitting extracted text data to a server" is a communication function for transmitting text data from a terminal to a server, for example, a means for using an Internet connection.

[0155] The "receiving means by which the server receives text data" refers to a function or device on the server side that receives text data sent from the terminal.

[0156] The "analysis means of the server that inputs text data into a natural language processing model for analysis" is a function of the server that uses natural language processing technology to analyze the acquired text data.

[0157] The "evaluation means for calculating the probability that a sentence is generated by AI based on the analysis results" is a function that calculates the probability that the sentence was generated by AI based on the analyzed text data.

[0158] The "result transmission means for transmitting the evaluation results to the terminal via the communication means" is a function for transmitting the analysis and evaluation results from the server to the terminal.

[0159] The "display means for the terminal to display the evaluation results" is a function for displaying the evaluation results on the terminal so that the user can visually confirm the analysis results.

[0160] "Capture means for capturing documents shared on the Internet or in internal documents and identifying whether they were generated by AI" refers to a means by which a user can capture documents on the Internet or in internal documents and determine whether their contents were generated by AI.

[0161] A "wearable device that displays evaluation results to a user in real time" is a pair of glasses or other device that can be worn by a user and that displays evaluation results in real time.

[0162] This invention relates to a system that can quickly and accurately identify whether book reports submitted by students, university assignments, internal corporate documents, and documents shared on the Internet have been generated by AI. This system is useful for preventing fraud and verifying the authenticity of content in security departments and educational institutions.

[0163] The basic system configuration is as follows:

[0164] 1. The device (smartphone or wearable device) is equipped with a camera that allows the user to capture the target document.

[0165] 2. The captured image is converted into text data using optical character recognition (OCR) technology such as Google Cloud Vision API or Tesseract OCR.

[0166] 3. The converted text data is formatted within the terminal, and unnecessary parts such as line breaks and spaces are removed.

[0167] 4. The formatted text data is sent to the server using a secure protocol (e.g., HTTPS).

[0168] 5. The server analyzes the text data using a natural language processing model (e.g., Hugging Face's Transformers library or GPT-3 (registered trademark)).

[0169] 6. The analysis results are evaluated by calculating the probability that the text data was generated by AI.

[0170] 7. The evaluation results are sent from the server to the device and displayed to the user in real time. Particularly when using a wearable device, users can check the evaluation results immediately.

[0171] Examples:

[0172] Example 1: Corporate security department use

[0173] During a meeting, a company security officer captures a printed internal document with the smart glasses' camera. The captured image is converted into text data using the smart glasses' built-in OCR system, and the data is sent to a server via secure communication. The server analyzes the text data and displays the assessment result in real time on the glasses' display: "There is an 85% chance that this document was generated by AI."

[0174] Example 2: Faculty and staff at an educational institution reviewing student submissions

[0175] Faculty and staff use their smartphones to take photos of students' book reports. The captured images are then processed by OCR on the smartphone and converted into text data. This text data is then sent to a server and analyzed using a natural language processing model. The analysis results, such as "There is a 70% chance that this review was generated by AI," are displayed on the smartphone screen.

[0176] Prompt Sentence Examples

[0177] For internal company documents: "Please determine whether the following document was generated by AI:\n\n'Regarding customer data management methods, we would like to propose a new system based on the content of our recent meeting.'"

[0178] For educational student submissions: "Please determine if this student's book report was generated by AI: 'Reading this book helped me to gain a deeper understanding of different cultural backgrounds.'"

[0179] In this way, the system of the present invention enables faculty and security personnel to efficiently and accurately identify AI-generated sentences and quickly provide necessary feedback and countermeasures.

[0180] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0181] Step 1:

[0182] The user captures a document using the device's camera or smart glasses. The input is image data of the document. Specifically, the user launches the device's camera app and takes a picture of the target document. This image data is temporarily stored inside the device.

[0183] Step 2:

[0184] The device performs OCR processing on the captured image data to extract text data. The input is image data and the output is text data. Specifically, the device uses the Google Cloud Vision API and Tesseract OCR engine to recognize and extract character information from the image. This extraction process obtains the document contents as text data.

[0185] Step 3:

[0186] The terminal formats the extracted text data and removes unnecessary characters such as line breaks and spaces. The input is text data obtained by OCR, and the output is formatted text data. Specifically, using regular expressions and string manipulation libraries, unnecessary characters and line breaks are removed from the text, and the formatted text data is generated.

[0187] Step 4:

[0188] The formatted text data is sent from the terminal to the server. The input is the formatted text data, and the output is the state after transmission to the server has been completed. Specifically, the text data is sent to the server using a secure communication protocol (e.g., HTTPS). This procedure allows the text data to be received in a format that can be analyzed on the server side.

[0189] Step 5:

[0190] The server inputs the received text data into a natural language processing model to begin analysis. The input is formatted text data, and the output is the analysis result. Specifically, the text data is analyzed using natural language processing models such as Hugging Face's Transformers library and GPT-3. This analysis calculates the probability that the text was generated by AI.

[0191] Step 6:

[0192] The server calculates the probability that the text is AI-generated based on the analysis results and performs an evaluation. The input is the analyzed text data, and the output is the evaluation result. Specifically, based on the output of the analysis model, an evaluation such as "There is an 85% chance that this document is AI-generated text" is generated.

[0193] Step 7:

[0194] The server sends the evaluation results to the terminal. The input is the evaluation results, and the output is the completion of transmission to the terminal. Specifically, the evaluation results are sent to the terminal again using a secure communication protocol such as HTTPS.

[0195] Step 8:

[0196] The device displays the evaluation results and notifies the user. The input is the evaluation results received from the server, and the output is what is displayed to the user. Specifically, a message such as "There is an 85% chance that this document is AI-generated" is displayed in real time using the device's display or the display function of the smart glasses. This allows the user to quickly check the evaluation results.

[0197] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0198] This invention combines a system that identifies whether book reports and university assignments submitted by students are generated by AI with a function that recognizes user emotions. This system allows faculty and staff to quickly and accurately evaluate submissions, and by understanding the user's emotional state, it provides more flexible and effective feedback.

[0199] System Configuration

[0200] This system mainly consists of terminals, servers, users, and an emotion engine. Terminals are mobile devices such as smartphones and tablets, and the server is a computing resource deployed in a cloud environment. Users are primarily faculty and staff who use this system to evaluate students' submissions. The emotion engine has the ability to analyze users' facial expressions and voices and identify their emotional state.

[0201] Program processing

[0202] 1. The user captures input data using the smartphone camera.

[0203] The user launches the system's dedicated app and uses the camera function to take a photo of the book report paper.

[0204] The captured image is temporarily stored on the device.

[0205] 2. The device performs OCR processing to extract the text data.

[0206] The device performs OCR (optical character recognition) technology to extract text information from the captured image.

[0207] The extracted character information is stored in the terminal as text data.

[0208] 3. The device formats the text data and sends it to the server

[0209] The extracted text data is formatted on the device, and unnecessary line breaks, spaces, and misrecognized characters are corrected.

[0210] The formatted text data is sent to the server using a secure protocol (e.g., HTTPS).

[0211] 4. The server receives the text data and begins analyzing it.

[0212] The server receives the text data sent from the terminal.

[0213] The received text data is input into a natural language processing model and analysis begins.

[0214] 5. The server evaluates the analysis results and sends them to the device.

[0215] A natural language processing model analyzes the text data and calculates the probability that a sentence is AI-generated.

[0216] The server evaluates the analysis results and summarizes them in a format that is easy for the user to understand, such as "There is an 80% chance that this sentence was generated by an AI."

[0217] The evaluation results are then transmitted to the terminal again using a secure communication protocol.

[0218] 6. Emotion engine analyzes user emotions

[0219] The emotion engine installed on the device captures facial expressions and voice data from the user through a facial recognition camera and microphone.

[0220] Based on the acquired data, the user's current emotional state is classified and stored as emotional data in the device.

[0221] 7. The device sends the emotion data to the server

[0222] The device sends emotion data to the server, which is also sent using a secure protocol.

[0223] 8. The server reflects the emotion data in the analysis results

[0224] The server reflects the received emotion data in the analysis results. For example, if the user shows strong emotion (surprise or confusion), it adjusts the reliability of the analysis results.

[0225] As a result, the evaluation results can be corrected based on the user's emotional state.

[0226] 9. The device displays the final evaluation result and emotional feedback.

[0227] The device displays the final evaluation results and emotional feedback to the user, who can check not only the probability of the AI-generated sentences but also their own emotional data.

[0228] For example, it might say, "There is an 80% chance that this sentence was generated by AI. User's emotional state: Confused 45%."

[0229] Specific examples

[0230] For example, imagine a scenario in which a faculty member tries to evaluate a student's book review. The user launches a dedicated app and takes a photo of the review with their camera. The system extracts text from the image, formats it, and sends it to the server. Analysis is then performed on the server, resulting in an evaluation such as "There is an 80% chance that this sentence was generated by AI." When the user checks the evaluation results, the emotion engine analyzes the user's emotions and provides feedback such as "Confused 45%." This allows the user to objectively confirm the reliability of the evaluation results.

[0231] The system not only provides faculty with advanced tools to effectively grade student submissions, helping to prevent cheating, but also allows for more flexible grading through feedback based on sentiment data.

[0232] The processing flow will be explained below.

[0233] Step 1:

[0234] The user activates the smartphone camera and takes a picture of the book report paper. The user then controls the camera through a dedicated app to capture the image with the appropriate framing.

[0235] Step 2:

[0236] The device saves the captured image and starts OCR processing. The device then analyzes the image and extracts the character information as text data.

[0237] Step 3:

[0238] The device then formats the extracted text data, specifically by removing unnecessary line breaks and spaces from the OCR results and correcting any misrecognized characters.

[0239] Step 4:

[0240] The device sends the formatted text data to the server. The device uses a secure communication protocol such as HTTPS to securely transmit the data to the server.

[0241] Step 5:

[0242] The server receives the text data sent from the device and prepares it for analysis. The server temporarily stores the received data for subsequent analysis.

[0243] Step 6:

[0244] The server inputs the text data into a natural language processing model and begins analysis, specifically to detect patterns and phrases characteristic of AI-generated text.

[0245] Step 7:

[0246] The server evaluates the analysis results and calculates the likelihood that the sentence is AI-generated. Based on the probability calculated by the AI ​​model, it generates an evaluation such as "high probability that the sentence is AI-generated" or "low probability."

[0247] Step 8:

[0248] The server formats the evaluation results and generates results in a format that is easy for users to understand. For example, it generates an evaluation result such as "There is an 80% chance that this sentence was generated by an AI."

[0249] Step 9:

[0250] The server transmits the evaluation results to the terminal, and again using a secure communication protocol, the server transmits the evaluation results to the terminal safely.

[0251] Step 10:

[0252] The terminal displays the evaluation results received from the server, and the terminal visually displays the evaluation results to the user, allowing the user to confirm the results.

[0253] Step 11:

[0254] The emotion engine acquires the user's emotion data. The device uses a camera and microphone to collect the user's facial expression and voice data, which the emotion engine analyzes to identify the user's emotional state.

[0255] Step 12:

[0256] The device transmits the acquired emotion data to the server, which also uses a secure protocol.

[0257] Step 13:

[0258] The server receives the emotion data and performs corrections based on the analysis. The server takes into account the user's emotional state identified by the emotion engine as a factor that influences the evaluation results.

[0259] Step 14:

[0260] The server generates the final evaluation result and emotional feedback and sends it to the terminal. The evaluation result takes into account the user's emotional data and is sent to the terminal as the final evaluation result.

[0261] Step 15:

[0262] The device will then display the final evaluation results and emotional feedback to the user, such as "There is an 80% chance that this sentence was generated by AI. User's emotional state: Confused 45%."

[0263] Step 16:

[0264] The user checks the final evaluation results and determines whether there was any cheating. Based on the displayed evaluation results and emotional feedback, the user can conduct further investigations and provide feedback to the students.

[0265] Through these steps, users can not only identify AI-generated sentences but also check the reliability of the evaluation based on their own emotional state, which can effectively prevent cheating in educational settings.

[0266] Example 2

[0267] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0268] Conventional student submission evaluation systems have difficulty identifying whether book reports or university assignments are generated by AI. Furthermore, they do not provide feedback that takes into account the evaluator's emotional state. As a result, it is difficult to provide reliable evaluations and effective feedback. This invention solves these problems, prevents cheating in submissions, and enables flexible feedback based on the evaluator's emotions.

[0269] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for inputting text data into a natural language processing model for analysis, an evaluation means for calculating the probability that the text is an AI-generated sentence based on the analysis result, and an analysis reflection means for reflecting emotion data in the analysis result. This makes it possible to identify with high accuracy whether the submitted work was generated by an AI and provide feedback appropriate to the evaluator's emotions.

[0270] The "camera means of the terminal that acquires input data" refers to a function that allows a user to take a photo of a submission such as a book report or a university assignment using the camera of a smartphone or tablet.

[0271] "Processing means of a terminal that extracts text from images using optical character recognition technology" refers to technology for extracting text information from captured images and the functionality within the terminal that executes this technology.

[0272] The "communication means for transmitting extracted text data to a server" refers to a communication protocol and function for transmitting text data from a terminal to a server.

[0273] The "receiving means for the server to receive text data" is a function for the server to receive text data sent from the terminal.

[0274] The "server's analysis means for inputting text data into a natural language processing model for analysis" refers to the server's function for inputting received text data into a natural language processing algorithm for analysis.

[0275] The "evaluation means for calculating the probability that a sentence is generated by AI based on the analysis results" is a server function that calculates the probability that a sentence was generated by AI based on the analysis results of a natural language processing model.

[0276] The "result transmission means for transmitting the evaluation results to the terminal via the communication means" refers to a communication protocol and function for transmitting the evaluation results obtained on the server side to the terminal.

[0277] The "emotion data collection means for the terminal to acquire the user's facial expressions and voice" is a function that acquires the user's facial expressions and voice using the camera and microphone installed in the terminal and uses them as emotion data.

[0278] "Emotion data transmission means for transmitting emotion data from a terminal to a server" refers to a communication protocol and function for transmitting emotion data collected by a terminal to a server.

[0279] The "analysis reflection means for the server to reflect emotion data in the analysis results" is a function for reflecting emotion data received by the server in the analysis results and adjusting the reliability of the evaluation.

[0280] The "display means by which the terminal displays the evaluation results and emotion feedback" is a function of the terminal for visually displaying the evaluation results and emotion data received from the server to the user.

[0281] The present invention provides a system that identifies whether a book report or university assignment submitted by a student was generated by AI and further reflects the user's emotional state in the evaluation. The system includes a terminal, a server, and an emotion engine. A specific embodiment of the system is described below.

[0282] Device hardware and software

[0283] The terminal is primarily a mobile device such as a smartphone or tablet. This terminal includes a camera means, a processing means, a communication means, an emotional data collection means, and a display means. The camera means has a function that allows the user to take a photo of a book report or assignment. The processing means includes optical character recognition (OCR) software, specifically "Tesseract OCR." The communication means has a function to send and receive data using the HTTPS protocol. The emotional data collection means includes a face recognition camera and microphone, and has a function to capture facial expressions and voice. The display means visually displays the evaluation results and emotional feedback to the user.

[0284] Server Hardware and Software

[0285] The server is located in a cloud environment and has powerful computing resources. The server includes a receiving means, an analyzing means, an evaluating means, an analysis reflecting means, and a result transmitting means. The receiving means receives data transmitted from the terminal and stores it. The analyzing means analyzes the text data using a natural language processing (NLP) model. Specifically, a generative AI model such as "GPT-4" is used. The evaluating means calculates the probability that the sentence is AI-generated based on the analysis results. Furthermore, the analysis reflecting means reflects the user's emotional data in the analysis results and adjusts the reliability of the evaluation. The result transmitting means transmits the evaluation results to the terminal.

[0286] System Operation

[0287] The system works as follows: The user uses the device's camera to take an image of a book report or assignment. The device's OCR software extracts text information from the image and formats this data. The formatted text data is sent to the server using a secure communications protocol. The server receives the text data and inputs it into a natural language processing model for analysis. Based on the analysis results, it calculates the probability that the text is AI-generated and sends this evaluation result to the device.

[0288] Furthermore, the device's emotional data collection means uses the user's facial recognition camera and microphone to capture facial expressions and voice data, generating emotional data. This emotional data is sent to the server, which then reflects it in the analysis results. Finally, the device displays the evaluation results and emotional feedback to the user. For example, it may display, "There is an 80% chance that this sentence is an AI-generated sentence. User's emotional state: Confused 45%."

[0289] Specific examples

[0290] For example, consider the case where a faculty member wants to evaluate a student's book report. The user launches a dedicated app and takes a photo of the report with their camera. The system extracts text from the image, formats it, and sends it to the server. Analysis is performed on the server side, and an evaluation is given, such as "There is an 80% chance that this sentence was generated by AI." Furthermore, when the user checks the evaluation results, the emotion engine analyzes the user's emotions and provides feedback such as "Confused 45%." This allows the user to objectively confirm the reliability of the evaluation results.

[0291] Prompt Sentence Examples

[0292] Here is an example prompt:

[0293] "The following text is a book report written by a student. Please rate whether this text was generated by AI. Also, please analyze the user's emotional state at the time of rating."

[0294] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0295] Step 1:

[0296] The user launches the app and uses the smartphone camera to take a photo of a book report or university assignment. The input is the image captured by the camera, and the output is the image stored in the device's temporary memory. Specifically, the user taps the "take a photo" button in the app, focuses the camera on the object, and takes a photo.

[0297] Step 2:

[0298] The device uses OCR technology to extract text information from a captured image. The input is image data from temporary memory, and the output is the extracted text data. Specifically, the device's OCR engine (e.g., "Tesseract OCR") analyzes the image and converts the text information into text format.

[0299] Step 3:

[0300] The terminal formats the extracted text data. The input is the text data obtained by OCR processing, and the output is the formatted text data. Specifically, a program is executed to correct unnecessary line breaks, spaces, and misrecognized characters.

[0301] Step 4:

[0302] Sends formatted text data to the server. The input is the formatted text data, and the output is the data sent to the server. Specifically, the text data is encrypted and sent using a secure communication protocol (e.g., HTTPS).

[0303] Step 5:

[0304] The server receives the text data and inputs it into a natural language processing model for analysis. The input is the received text data, and the output is the analyzed result. Specifically, the server receives the HTTPS request, saves it in a database, and then inputs the data into an NLP engine (e.g., "GPT-4") to perform analysis.

[0305] Step 6:

[0306] The server calculates the probability that the sentence is AI-generated based on the analysis results. The input is the analysis result of the NLP engine, and the output is a probability value. Specifically, the generative AI model evaluates the analysis result and calculates the probability in the form of "There is an 80% chance that this sentence is AI-generated."

[0307] Step 7:

[0308] The server sends the evaluation results to the terminal. The input is the evaluation results including the probability values, and the output is the data sent to the terminal. Specifically, the evaluation results are sent to the terminal in JSON format using a secure communication protocol.

[0309] Step 8:

[0310] The device's emotion data collection means acquires the user's facial expressions and voice. The input is the user's face and voice, and the output is the acquired emotion data. Specifically, the device's camera and microphone operate simultaneously to capture the facial expressions and voice.

[0311] Step 9:

[0312] The device sends emotion data to the server. The input is the acquired emotion data, and the output is the data sent to the server. Specifically, the emotion data is formatted in JSON format and sent using a secure protocol.

[0313] Step 10:

[0314] The server reflects the emotional data in the analysis results. The input is the received emotional data and the analysis results, and the output is the final evaluation result that reflects the emotions. Specifically, the server analyzes the emotional data and adjusts the reliability of the evaluation results.

[0315] Step 11:

[0316] The device displays the final evaluation result and emotional feedback. The input is the final evaluation result received from the server, and the output is the evaluation result and emotional feedback displayed to the user. Specifically, the device's UI displays "There is an 80% chance that this sentence is an AI-generated sentence. User's emotional state: Confused 45%."

[0317] (Application example 2)

[0318] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0319] Passengers in autonomous vehicles are likely to be in various emotional states, and safer and more comfortable driving is required. However, current autonomous driving systems have difficulty recognizing passengers' emotional states in real time and adjusting vehicle operating parameters accordingly. This can lead to inability to respond appropriately to passengers' anxiety or confusion, which could result in a decrease in safety and comfort.

[0320] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0321] In this invention, the server includes a camera means of the terminal that acquires input data, a processing means of the terminal that extracts text from images using optical character recognition (OCR) technology, an analysis means of the server that inputs the text data into a natural language processing model for analysis, an emotion analysis means of the terminal or server that analyzes the user's emotion using the camera and voice input means, and an operation adjustment means that adjusts the vehicle's operation parameters based on the analysis results of the emotion analysis means. This makes it possible to recognize the emotional state of passengers in real time and appropriately adjust the vehicle's operation parameters based on the emotional state.

[0322] "Input data" refers to data such as images and sounds that are acquired by the system and are the subject of analysis.

[0323] "Camera means of the terminal" refers to a means for acquiring image or video data using a camera mounted on the terminal.

[0324] "Optical character recognition (OCR) technology" is a technology that extracts character information from image data.

[0325] "Terminal processing means" refers to a means that has the function of analyzing and processing data within the terminal.

[0326] "Communication means" refers to a means for transmitting and receiving data between a terminal and a server.

[0327] A "server" is a central computer system that performs data analysis and processing.

[0328] The "receiving means" is a means by which the server receives data transmitted from the terminal.

[0329] A "natural language processing model" is an algorithm or technology for analyzing text data and understanding and generating its content.

[0330] The "analysis means" is a means for analyzing the data received by the server.

[0331] The "evaluation means" is a means for calculating the probability that a text is an AI-generated sentence based on the analysis results.

[0332] The "result transmission means" is a means by which the server transmits the analysis results to the terminal.

[0333] The "display means" is a means for visually presenting the evaluation results received by the terminal to the user.

[0334] The "emotion analysis means" is a means for analyzing the user's emotional state using a camera and a voice input means.

[0335] The "operation adjustment means" is a means for adjusting the vehicle operation parameters based on the analysis results of the emotion analysis means.

[0336] The present invention provides a system for recognizing a user's emotional state in real time and adjusting vehicle operating parameters accordingly. The system includes the following means:

[0337] Hardware and software used

[0338] The system requires a device equipped with a camera and microphone, and a cloud server for analyzing the data. The device includes a camera for capturing video data and a voice input for capturing audio data. The software includes optical character recognition (OCR) technology, natural language processing models, and DeepFace and SpeechRecognition libraries for emotion recognition.

[0339] Data processing and calculation

[0340] The server converts the image data sent from the device into text data using OCR technology. This allows it to analyze the content of the text data and calculate the probability that the text was generated by AI. Meanwhile, the device or server uses a camera and voice input means to capture the user's facial expressions and voice in real time, and analyzes the user's emotional state based on this data.

[0341] Emotion analysis uses the DeepFace library to analyze facial expression data and identify the user's dominant emotion (e.g., anger, sadness, surprise, etc.). For voice data, the SpeechRecognition library is used to first convert the voice to text, and then the text is analyzed for emotion. The results of the emotion analysis are sent from the device to a server, which uses this data to generate instructions for adjusting the vehicle's operating parameters (speed, safety distance, etc.).

[0342] Specific examples

[0343] For example, if a passenger expresses anxiety while using the autonomous taxi application, the system will detect the passenger's emotional state and reduce the vehicle's speed and increase the safety distance in real time. The in-car environment (music, lighting, etc.) will also be adjusted according to the passenger's emotions, allowing passengers to enjoy a safer and more comfortable riding experience.

[0344] Prompt Sentence Examples

[0345] A possible prompt would be:

[0346] Generate a Python program to recognize passenger emotions in real time and adjust the driving parameters of an autonomous vehicle based on the emotions. Use cameras and microphones installed in the vehicle to perform facial and voice emotion analysis using DeepFace and SpeechRecognition libraries. Obtain emotion data in real time and reduce the vehicle's speed and increase the safety distance if the passenger is anxious.

[0347] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0348] Step 1:

[0349] The terminal uses a camera and microphone to capture passenger image and voice data. These data are captured in real time and used for subsequent processing. The input is image and voice data, and the output is a real-time capture of these data.

[0350] Step 2:

[0351] The DeepFace library is used to analyze the image data captured by the device to identify the passenger's emotional state. It analyzes facial patterns and expressions to identify the dominant emotion (e.g., joy, anger, sadness, etc.). The input is the image data, and the output is the identified emotional state.

[0352] Step 3:

[0353] The voice data acquired by the device is converted into text using the SpeechRecognition library, and then the text content is analyzed to infer the emotional state. The emotional nuances of the spoken content are evaluated. The input is voice data, and the output is text data containing the emotional state.

[0354] Step 4:

[0355] The device integrates the emotional states identified in steps 2 and 3 to generate an overall emotional assessment. The device outputs the overall emotional assessment from multiple emotional data. The input is emotional data from images and text, and the output is an overall emotional assessment.

[0356] Step 5:

[0357] The device sends a comprehensive emotional evaluation to the server, which then generates instructions to adjust the vehicle's operating parameters based on the evaluation results. Specifically, it adjusts the speed or changes the safety distance. The input is comprehensive emotional evaluation data, and the output is instructions to adjust the vehicle's operating parameters.

[0358] Step 6:

[0359] The server generates an instruction to adjust the operating parameters and sends it to the terminal, which then adjusts the vehicle's operating parameters in real time based on the instruction received. Specific actions include reducing the vehicle's speed, maintaining distance, etc. The input is the instruction to adjust the operating parameters, and the output is the adjusted vehicle's operating parameters.

[0360] Step 7:

[0361] The user (passenger) checks the displayed evaluation results and the vehicle's operating status. The terminal displays the analyzed emotional state and the operating parameters based on it to the user. The input is the analyzed emotional state and operating parameters, and the output is visual feedback to the user.

[0362] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0363] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0364] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0365] [Second embodiment]

[0366] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0367] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0368] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0369] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0370] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0371] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0372] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0373] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0374] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0375] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0376] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0377] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0378] The present invention relates to a system that identifies whether book reports and university assignments submitted by students are AI-generated texts, which can be used by faculty and staff to quickly and accurately evaluate the submissions.

[0379] System Configuration

[0380] This system mainly consists of terminals, servers, and users. Terminals are mobile devices such as smartphones and tablets, and servers are computing resources deployed in a cloud environment. Users are primarily faculty and staff who use this system to evaluate student submissions.

[0381] Program processing

[0382] 1. The user captures input data using the smartphone camera.

[0383] The user launches the system's dedicated app and uses the camera function to take a photo of the book report paper.

[0384] The captured image is temporarily stored on the device.

[0385] 2. The device performs OCR processing to extract the text data.

[0386] The device performs OCR (optical character recognition) technology to extract text information from the captured image.

[0387] The extracted character information is stored in the terminal as text data.

[0388] 3. The device formats the text data and sends it to the server

[0389] The extracted text data is formatted on the device, and unnecessary line breaks, spaces, and misrecognized characters are corrected.

[0390] The formatted text data is sent to the server using a secure protocol (e.g., HTTPS).

[0391] 4. The server receives the text data and begins analyzing it.

[0392] The server receives the text data sent from the terminal.

[0393] The received text data is input into a natural language processing model and analysis begins.

[0394] 5. The server evaluates the analysis results and sends them to the device.

[0395] A natural language processing model analyzes the text data and calculates the probability that a sentence is AI-generated.

[0396] The server evaluates the analysis results and summarizes them in a format that is easy for the user to understand, such as "There is an 85% chance that this sentence was generated by an AI."

[0397] The evaluation results are then transmitted to the terminal again using a secure communication protocol.

[0398] 6. The device displays the evaluation results.

[0399] The terminal receives the analysis results sent from the server.

[0400] The received analysis results are displayed to the user, who can use the results to determine whether the student's submission is AI-generated.

[0401] Specific examples

[0402] For example, imagine a scenario in which a faculty member wants to evaluate a student's book report. The user launches a dedicated app and takes a photo of the report with their camera. The system extracts text from the image, formats it, and sends it to the server. Analysis is then performed on the server, and the result is determined to be "80% likely to be an AI-generated sentence." The user can review this result and, if necessary, conduct further research or provide feedback to the student.

[0403] In this way, the system provides faculty with advanced tools to effectively evaluate student work and helps prevent cheating.

[0404] The processing flow will be explained below.

[0405] Step 1:

[0406] The user activates the smartphone camera and takes a picture of the book report paper. The user then controls the camera through the app to capture the image with the appropriate framing.

[0407] Step 2:

[0408] The device saves the captured image and starts OCR processing. The device then analyzes the image and extracts the character information as text data.

[0409] Step 3:

[0410] The device then formats the extracted text data, specifically by removing unnecessary line breaks and spaces from the OCR results and correcting any misrecognized characters.

[0411] Step 4:

[0412] The device sends the formatted text data to the server. The device uses a secure communication protocol such as HTTPS to securely transmit the data to the server.

[0413] Step 5:

[0414] The server receives the text data sent from the device and prepares it for analysis. The server temporarily stores the received data for subsequent analysis.

[0415] Step 6:

[0416] The server inputs the text data into a natural language processing model and begins analysis, specifically to detect patterns and phrases characteristic of AI-generated text.

[0417] Step 7:

[0418] The server evaluates the analysis results and calculates the likelihood that the sentence is AI-generated. Based on the probability calculated by the AI ​​model, it generates an evaluation such as "high probability that the sentence is AI-generated" or "low probability."

[0419] Step 8:

[0420] The server formats the evaluation results and generates results in a format that is easy for users to understand. For example, it generates an evaluation result such as "There is an 80% chance that this sentence was generated by an AI."

[0421] Step 9:

[0422] The server transmits the evaluation results to the terminal, and again using a secure communication protocol, the server transmits the evaluation results to the terminal safely.

[0423] Step 10:

[0424] The terminal displays the evaluation results received from the server, and the terminal visually displays the evaluation results to the user, allowing the user to confirm the results.

[0425] Step 11:

[0426] Users can check the evaluation results and determine whether or not there has been any cheating. Based on the displayed evaluation results, users can conduct further detailed verification and provide feedback to students.

[0427] In this way, a system has been created in which users, devices, and servers work together to quickly and accurately evaluate whether student submissions are AI-generated texts.

[0428] Example 1

[0429] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0430] In recent years, advances in artificial intelligence technology have created a need for a fast and accurate way to assess whether student submissions are AI-generated. However, current systems rely heavily on manual review, which not only takes time and effort but also often lacks accuracy. Therefore, a method is needed for faculty and staff to efficiently assess submissions and prevent cheating.

[0431] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0432] In this invention, the server includes a photographing means of the terminal that acquires input data, an analysis means of the terminal that extracts characters from the image using optical character recognition technology, a communication means that transmits the extracted character data to the information processing device, a receiving means by the information processing device that receives the character data, an analysis means of the information processing device that inputs the character data into a natural language processing model and analyzes it, an evaluation means that calculates the probability that the sentence is an AI-generated sentence based on the analysis result, a result transmission means that transmits the evaluation result to the terminal via the communication means, and a display means on the terminal that displays the evaluation result. This enables faculty and staff to quickly and accurately evaluate whether a submitted work was generated by AI.

[0433] "Input data" is digital information that a user captures using a camera or other device.

[0434] A "terminal" is a device that is directly operated by a user, such as a smartphone, tablet, or PC.

[0435] "Photographing means" refers to the hardware and software, including the camera function, installed in the terminal.

[0436] "Optical character recognition technology" is a technology for reading and extracting text information from images.

[0437] "Analysis means" refers to the function of executing optical character recognition technology on the terminal and extracting character data from an image.

[0438] "Character data" refers to text-based data extracted using optical character recognition technology.

[0439] An "information processing device" is a device that performs data processing, including a server or cloud computing resource.

[0440] "Communication means" refers to an interface and protocol for transmitting and receiving data between a terminal and an information processing device.

[0441] The "receiving means" refers to a function by which the information processing device receives data transmitted from the terminal.

[0442] A "natural language processing model" is an algorithm and software that analyzes text data and determines whether it was generated by artificial intelligence.

[0443] The term "analysis means" refers to a function of an information processing device to input character data into a natural language processing model and analyze it.

[0444] "Evaluation means" refers to a function that calculates the probability that the input text was generated by AI based on the analysis results.

[0445] "Result transmission means" refers to a communication interface and protocol for transmitting evaluation results to a terminal.

[0446] "Display means" refers to an interface that allows the terminal to visually present the evaluation results to the user.

[0447] "Formatting means" refers to a function that formats extracted character data by correcting misrecognitions and deleting unnecessary line breaks and spaces.

[0448] The "evaluation generation means" refers to a function that summarizes and presents the analysis results in a format that is easy for the user to understand.

[0449] This invention relates to a system for identifying whether book reports and university assignments submitted by students have been generated by artificial intelligence. This system allows faculty and staff to quickly and accurately evaluate the submissions. The system mainly consists of a terminal, a server, and a user.

[0450] Hardware and software examples

[0451] Terminal

[0452] The terminal is a device that is directly operated by the user, such as a smartphone, tablet, or PC. The terminal is equipped with a camera, which is used to capture input data such as a book report. The terminal is equipped with optical character recognition (OCR) technology, such as Google Tesseract OCR, to extract text data from images. In addition, the terminal includes a communication means for sending the submitted text data to a server. The HTTPS protocol is used for secure data communication.

[0453] server

[0454] The server is an information processing device that operates in a cloud environment and receives and analyzes text data sent from the terminal. The server has a receiving means and can receive data sent from the terminal. The received data is input into a natural language processing model (for example, "OpenAI's GPT-4") for analysis. The analysis means inputs the text data into this natural language processing model and calculates the probability that the text was generated by AI. The evaluation means evaluates the analysis results and summarizes them in a format that is easy for the user to understand. The evaluation results are then sent back to the terminal using the HTTPS protocol.

[0455] User

[0456] Users are primarily faculty and staff, who use the system to evaluate student submissions. Using a smartphone, tablet, or other device, users launch a dedicated app and take a picture of the submission. After taking the picture, text data extracted from the image is sent to a server, and the analysis results are displayed on the device. This allows users to quickly determine whether the submission was generated by artificial intelligence.

[0457] Examples of concrete examples and prompts

[0458] For example, imagine a scenario in which a faculty member wants to evaluate a student's book report. The user launches a dedicated app and takes a photo of the report with their camera. The system extracts text data from the image, formats it, and sends it to the server. Analysis is then performed on the server, and the result is that "there is an 80% chance that this sentence was generated by AI." The user can review this result and, if necessary, conduct further research or provide feedback to the student.

[0459] Examples of prompts include:

[0460] "Please rate whether the following sentences were generated by an AI:"

[0461] "Calculate the probability that the following text is AI-generated:"

[0462] This provides faculty with advanced tools to effectively assess student work and helps prevent cheating.

[0463] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0464] Step 1:

[0465] The user launches the system's dedicated app and takes a photo of the book report paper using the smartphone camera. The input data is an image of the paper, and the output is a digital image file that is temporarily stored inside the device. When the user taps the "take a photo" button, the camera starts and the image is saved.

[0466] Step 2:

[0467] The device analyzes the captured image using OCR (Optical Character Recognition) technology. The input data is the output image from step 1, and the output is the extracted character data. The OCR engine on the device analyzes the image and generates text data. Specifically, Google's Tesseract OCR library is used. The processed results are saved on the device as a text file.

[0468] Step 3:

[0469] The terminal formats the extracted text data. The input data is the output text from step 2, and the output is the formatted text data. Specifically, unnecessary line breaks and spaces are removed, and misrecognized characters are corrected. The formatted text data is converted to JSON format and is ready to be sent to the server.

[0470] Step 4:

[0471] The terminal sends the formatted text data to the server. The input data is the output text data of step 3, and the output is a data transmission success message to the server. The HTTPS protocol is used as the communication method, and the terminal sends the text data to the server in a secure state.

[0472] Step 5:

[0473] The server receives the text data sent from the terminal. The input data is the text data sent from the terminal, and the output is the received text data. The server's receiving means captures the data and stores it in a specific directory or database.

[0474] Step 6:

[0475] The server inputs the received text data into a natural language processing (NLP) model for analysis. The input data is the received text data from step 5, and the output is the analysis result. Specifically, a generative AI model such as OpenAI's GPT-4 is used to analyze whether the text data is AI-generated. In this process, a natural language processing algorithm is executed.

[0476] Step 7:

[0477] The server calculates the probability that the sentence is AI-generated based on the analysis results. The input data is the analysis result from step 6, and the output is an evaluation result including a probability value. Based on the characteristics of the analyzed text data and the prompt sentence, the server calculates an evaluation such as "There is an 85% chance that this sentence is AI-generated."

[0478] Step 8:

[0479] The server sends the evaluation result to the terminal. The input data is the evaluation result from step 7, and the output is a message to the terminal indicating that the transmission was successful. The evaluation result is formatted in a user-friendly format and sent to the terminal using the HTTPS protocol.

[0480] Step 9:

[0481] The device receives the evaluation results sent from the server and displays them to the user. The input data is the evaluation results sent from the server, and the output is the evaluation results displayed to the user. The device displays the results it receives on the screen, and the user is presented with a message saying, "There is an 85% chance that this sentence is AI-generated." Based on this information, the user can determine whether the submission was AI-generated.

[0482] This allows faculty and staff to quickly and accurately evaluate submissions through a series of processes, preventing cheating before it occurs.

[0483] (Application example 1)

[0484] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0485] With conventional systems, it was difficult to quickly and accurately determine whether student submissions or internal company documents were generated by AI, which caused problems for educational institutions and companies when verifying fraud and the authenticity of content. Furthermore, users could not obtain the results in real time, which led to a lack of efficiency. Furthermore, there were no concrete solutions using wearable devices.

[0486] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0487] In this invention, the server includes a camera means of the terminal that acquires input data, a processing means of the terminal that extracts text from images using optical character recognition (OCR) technology, a communication means that transmits the extracted text data to the server, a receiving means by the server that receives the text data, an analysis means of the server that inputs the text data into a natural language processing model for analysis, an evaluation means that calculates the probability that the text is an AI-generated sentence based on the analysis result, a result transmission means that transmits the evaluation result to the terminal via the communication means, a display means of the terminal that displays the evaluation result, a capture means that captures documents shared on the Internet or in internal documents and identifies whether they were generated by AI, and a wearable device that displays the evaluation result to the user in real time, thereby enabling the user to quickly and accurately obtain a judgment result on the AI-generated sentence.

[0488] The "camera means of the terminal for acquiring input data" refers to a means for capturing target documents or image data using a camera equipped on the terminal.

[0489] "Terminal processing means for extracting text from images using optical character recognition (OCR) techniques" means means for recognizing character information from a captured image and extracting it as text data.

[0490] The "communication means for transmitting extracted text data to a server" is a communication function for transmitting text data from a terminal to a server, for example, a means for using an Internet connection.

[0491] The "receiving means by which the server receives text data" refers to a function or device on the server side that receives text data sent from the terminal.

[0492] The "analysis means of the server that inputs text data into a natural language processing model for analysis" is a function of the server that uses natural language processing technology to analyze the acquired text data.

[0493] The "evaluation means for calculating the probability that a sentence is generated by AI based on the analysis results" is a function that calculates the probability that the sentence was generated by AI based on the analyzed text data.

[0494] The "result transmission means for transmitting the evaluation results to the terminal via the communication means" is a function for transmitting the analysis and evaluation results from the server to the terminal.

[0495] The "display means for the terminal to display the evaluation results" is a function for displaying the evaluation results on the terminal so that the user can visually confirm the analysis results.

[0496] "Capture means for capturing documents shared on the Internet or in internal documents and identifying whether they were generated by AI" refers to a means by which a user can capture documents on the Internet or in internal documents and determine whether their contents were generated by AI.

[0497] A "wearable device that displays evaluation results to a user in real time" is a pair of glasses or other device that can be worn by a user and that displays evaluation results in real time.

[0498] This invention relates to a system that can quickly and accurately identify whether book reports submitted by students, university assignments, internal corporate documents, and documents shared on the Internet have been generated by AI. This system is useful for preventing fraud and verifying the authenticity of content in security departments and educational institutions.

[0499] The basic system configuration is as follows:

[0500] 1. The device (smartphone or wearable device) is equipped with a camera that allows the user to capture the target document.

[0501] 2. The captured image is converted into text data using optical character recognition (OCR) technology such as Google Cloud Vision API or Tesseract OCR.

[0502] 3. The converted text data is formatted within the terminal, and unnecessary parts such as line breaks and spaces are removed.

[0503] 4. The formatted text data is sent to the server using a secure protocol (e.g., HTTPS).

[0504] 5. The server analyzes the text data using a natural language processing model (e.g., Hugging Face's Transformers library or GPT-3).

[0505] 6. The analysis results are evaluated by calculating the probability that the text data was generated by AI.

[0506] 7. The evaluation results are sent from the server to the device and displayed to the user in real time. Particularly when using a wearable device, users can check the evaluation results immediately.

[0507] Examples:

[0508] Example 1: Corporate security department use

[0509] During a meeting, a company security officer captures a printed internal document with the smart glasses' camera. The captured image is converted into text data using the smart glasses' built-in OCR system, and the data is sent to a server via secure communication. The server analyzes the text data and displays the assessment result in real time on the glasses' display: "There is an 85% chance that this document was generated by AI."

[0510] Example 2: Faculty and staff at an educational institution reviewing student submissions

[0511] Faculty and staff use their smartphones to take photos of students' book reports. The captured images are then processed by OCR on the smartphone and converted into text data. This text data is then sent to a server and analyzed using a natural language processing model. The analysis results, such as "There is a 70% chance that this review was generated by AI," are displayed on the smartphone screen.

[0512] Prompt Sentence Examples

[0513] For internal company documents: "Please determine whether the following document was generated by AI:\n\n'Regarding customer data management methods, we would like to propose a new system based on the content of our recent meeting.'"

[0514] For educational student submissions: "Please determine if this student's book report was generated by AI: 'Reading this book helped me to gain a deeper understanding of different cultural backgrounds.'"

[0515] In this way, the system of the present invention enables faculty and security personnel to efficiently and accurately identify AI-generated sentences and quickly provide necessary feedback and countermeasures.

[0516] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0517] Step 1:

[0518] The user captures a document using the device's camera or smart glasses. The input is image data of the document. Specifically, the user launches the device's camera app and takes a picture of the target document. This image data is temporarily stored inside the device.

[0519] Step 2:

[0520] The device performs OCR processing on the captured image data to extract text data. The input is image data and the output is text data. Specifically, the device uses the Google Cloud Vision API and Tesseract OCR engine to recognize and extract character information from the image. This extraction process obtains the document contents as text data.

[0521] Step 3:

[0522] The terminal formats the extracted text data and removes unnecessary characters such as line breaks and spaces. The input is text data obtained by OCR, and the output is formatted text data. Specifically, using regular expressions and string manipulation libraries, unnecessary characters and line breaks are removed from the text, and the formatted text data is generated.

[0523] Step 4:

[0524] The formatted text data is sent from the terminal to the server. The input is the formatted text data, and the output is the state after transmission to the server has been completed. Specifically, the text data is sent to the server using a secure communication protocol (e.g., HTTPS). This procedure allows the text data to be received in a format that can be analyzed on the server side.

[0525] Step 5:

[0526] The server inputs the received text data into a natural language processing model to begin analysis. The input is formatted text data, and the output is the analysis result. Specifically, the text data is analyzed using natural language processing models such as Hugging Face's Transformers library and GPT-3. This analysis calculates the probability that the text was generated by AI.

[0527] Step 6:

[0528] The server calculates the probability that the text is AI-generated based on the analysis results and performs an evaluation. The input is the analyzed text data, and the output is the evaluation result. Specifically, based on the output of the analysis model, an evaluation such as "There is an 85% chance that this document is AI-generated text" is generated.

[0529] Step 7:

[0530] The server sends the evaluation results to the terminal. The input is the evaluation results, and the output is the completion of transmission to the terminal. Specifically, the evaluation results are sent to the terminal again using a secure communication protocol such as HTTPS.

[0531] Step 8:

[0532] The device displays the evaluation results and notifies the user. The input is the evaluation results received from the server, and the output is what is displayed to the user. Specifically, a message such as "There is an 85% chance that this document is AI-generated" is displayed in real time using the device's display or the display function of the smart glasses. This allows the user to quickly check the evaluation results.

[0533] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0534] This invention combines a system that identifies whether book reports and university assignments submitted by students are generated by AI with a function that recognizes user emotions. This system allows faculty and staff to quickly and accurately evaluate submissions, and by understanding the user's emotional state, it provides more flexible and effective feedback.

[0535] System Configuration

[0536] This system mainly consists of terminals, servers, users, and an emotion engine. Terminals are mobile devices such as smartphones and tablets, and the server is a computing resource deployed in a cloud environment. Users are primarily faculty and staff who use this system to evaluate students' submissions. The emotion engine has the ability to analyze users' facial expressions and voices and identify their emotional state.

[0537] Program processing

[0538] 1. The user captures input data using the smartphone camera.

[0539] The user launches the system's dedicated app and uses the camera function to take a photo of the book report paper.

[0540] The captured image is temporarily stored on the device.

[0541] 2. The device performs OCR processing to extract the text data.

[0542] The device performs OCR (optical character recognition) technology to extract text information from the captured image.

[0543] The extracted character information is stored in the terminal as text data.

[0544] 3. The device formats the text data and sends it to the server

[0545] The extracted text data is formatted on the device, and unnecessary line breaks, spaces, and misrecognized characters are corrected.

[0546] The formatted text data is sent to the server using a secure protocol (e.g., HTTPS).

[0547] 4. The server receives the text data and begins analyzing it.

[0548] The server receives the text data sent from the terminal.

[0549] The received text data is input into a natural language processing model and analysis begins.

[0550] 5. The server evaluates the analysis results and sends them to the device.

[0551] A natural language processing model analyzes the text data and calculates the probability that a sentence is AI-generated.

[0552] The server evaluates the analysis results and summarizes them in a format that is easy for the user to understand, such as "There is an 80% chance that this sentence was generated by an AI."

[0553] The evaluation results are then transmitted to the terminal again using a secure communication protocol.

[0554] 6. Emotion engine analyzes user emotions

[0555] The emotion engine installed on the device captures facial expressions and voice data from the user through a facial recognition camera and microphone.

[0556] Based on the acquired data, the user's current emotional state is classified and stored as emotional data in the device.

[0557] 7. The device sends the emotion data to the server

[0558] The device sends emotion data to the server, which is also sent using a secure protocol.

[0559] 8. The server reflects the emotion data in the analysis results

[0560] The server reflects the received emotion data in the analysis results. For example, if the user shows strong emotion (surprise or confusion), it adjusts the reliability of the analysis results.

[0561] As a result, the evaluation results can be corrected based on the user's emotional state.

[0562] 9. The device displays the final evaluation result and emotional feedback.

[0563] The device displays the final evaluation results and emotional feedback to the user, who can check not only the probability of the AI-generated sentences but also their own emotional data.

[0564] For example, it might say, "There is an 80% chance that this sentence was generated by AI. User's emotional state: Confused 45%."

[0565] Specific examples

[0566] For example, imagine a scenario in which a faculty member tries to evaluate a student's book review. The user launches a dedicated app and takes a photo of the review with their camera. The system extracts text from the image, formats it, and sends it to the server. Analysis is then performed on the server, resulting in an evaluation such as "There is an 80% chance that this sentence was generated by AI." When the user checks the evaluation results, the emotion engine analyzes the user's emotions and provides feedback such as "Confused 45%." This allows the user to objectively confirm the reliability of the evaluation results.

[0567] The system not only provides faculty with advanced tools to effectively grade student submissions, helping to prevent cheating, but also allows for more flexible grading through feedback based on sentiment data.

[0568] The processing flow will be explained below.

[0569] Step 1:

[0570] The user activates the smartphone camera and takes a picture of the book report paper. The user then controls the camera through a dedicated app to capture the image with the appropriate framing.

[0571] Step 2:

[0572] The device saves the captured image and starts OCR processing. The device then analyzes the image and extracts the character information as text data.

[0573] Step 3:

[0574] The device then formats the extracted text data, specifically by removing unnecessary line breaks and spaces from the OCR results and correcting any misrecognized characters.

[0575] Step 4:

[0576] The device sends the formatted text data to the server. The device uses a secure communication protocol such as HTTPS to securely transmit the data to the server.

[0577] Step 5:

[0578] The server receives the text data sent from the device and prepares it for analysis. The server temporarily stores the received data for subsequent analysis.

[0579] Step 6:

[0580] The server inputs the text data into a natural language processing model and begins analysis, specifically to detect patterns and phrases characteristic of AI-generated text.

[0581] Step 7:

[0582] The server evaluates the analysis results and calculates the likelihood that the sentence is AI-generated. Based on the probability calculated by the AI ​​model, it generates an evaluation such as "high probability that the sentence is AI-generated" or "low probability."

[0583] Step 8:

[0584] The server formats the evaluation results and generates results in a format that is easy for users to understand. For example, it generates an evaluation result such as "There is an 80% chance that this sentence was generated by an AI."

[0585] Step 9:

[0586] The server transmits the evaluation results to the terminal, and again using a secure communication protocol, the server transmits the evaluation results to the terminal safely.

[0587] Step 10:

[0588] The terminal displays the evaluation results received from the server, and the terminal visually displays the evaluation results to the user, allowing the user to confirm the results.

[0589] Step 11:

[0590] The emotion engine acquires the user's emotion data. The device uses a camera and microphone to collect the user's facial expression and voice data, which the emotion engine analyzes to identify the user's emotional state.

[0591] Step 12:

[0592] The device transmits the acquired emotion data to the server, which also uses a secure protocol.

[0593] Step 13:

[0594] The server receives the emotion data and performs corrections based on the analysis. The server takes into account the user's emotional state identified by the emotion engine as a factor that influences the evaluation results.

[0595] Step 14:

[0596] The server generates the final evaluation result and emotional feedback and sends it to the terminal. The evaluation result takes into account the user's emotional data and is sent to the terminal as the final evaluation result.

[0597] Step 15:

[0598] The device will then display the final evaluation results and emotional feedback to the user, such as "There is an 80% chance that this sentence was generated by AI. User's emotional state: Confused 45%."

[0599] Step 16:

[0600] The user checks the final evaluation results and determines whether there was any cheating. Based on the displayed evaluation results and emotional feedback, the user can conduct further investigations and provide feedback to the students.

[0601] Through these steps, users can not only identify AI-generated sentences but also check the reliability of the evaluation based on their own emotional state, which can effectively prevent cheating in educational settings.

[0602] Example 2

[0603] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0604] Conventional student submission evaluation systems have difficulty identifying whether book reports or university assignments are generated by AI. Furthermore, they do not provide feedback that takes into account the evaluator's emotional state. As a result, it is difficult to provide reliable evaluations and effective feedback. This invention solves these problems, prevents cheating in submissions, and enables flexible feedback based on the evaluator's emotions.

[0605] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for inputting text data into a natural language processing model for analysis, an evaluation means for calculating the probability that the text is an AI-generated sentence based on the analysis result, and an analysis reflection means for reflecting emotion data in the analysis result. This makes it possible to identify with high accuracy whether the submitted work was generated by an AI and provide feedback appropriate to the evaluator's emotions.

[0606] The "camera means of the terminal that acquires input data" refers to a function that allows a user to take a photo of a submission such as a book report or a university assignment using the camera of a smartphone or tablet.

[0607] "Processing means of a terminal that extracts text from images using optical character recognition technology" refers to technology for extracting text information from captured images and the functionality within the terminal that executes this technology.

[0608] The "communication means for transmitting extracted text data to a server" refers to a communication protocol and function for transmitting text data from a terminal to a server.

[0609] The "receiving means for the server to receive text data" is a function for the server to receive text data sent from the terminal.

[0610] The "server's analysis means for inputting text data into a natural language processing model for analysis" refers to the server's function for inputting received text data into a natural language processing algorithm for analysis.

[0611] The "evaluation means for calculating the probability that a sentence is generated by AI based on the analysis results" is a server function that calculates the probability that a sentence was generated by AI based on the analysis results of a natural language processing model.

[0612] The "result transmission means for transmitting the evaluation results to the terminal via the communication means" refers to a communication protocol and function for transmitting the evaluation results obtained on the server side to the terminal.

[0613] The "emotion data collection means for the terminal to acquire the user's facial expressions and voice" is a function that acquires the user's facial expressions and voice using the camera and microphone installed in the terminal and uses them as emotion data.

[0614] "Emotion data transmission means for transmitting emotion data from a terminal to a server" refers to a communication protocol and function for transmitting emotion data collected by a terminal to a server.

[0615] The "analysis reflection means for the server to reflect emotion data in the analysis results" is a function for reflecting emotion data received by the server in the analysis results and adjusting the reliability of the evaluation.

[0616] The "display means by which the terminal displays the evaluation results and emotion feedback" is a function of the terminal for visually displaying the evaluation results and emotion data received from the server to the user.

[0617] The present invention provides a system that identifies whether a book report or university assignment submitted by a student was generated by AI and further reflects the user's emotional state in the evaluation. The system includes a terminal, a server, and an emotion engine. A specific embodiment of the system is described below.

[0618] Device hardware and software

[0619] The terminal is primarily a mobile device such as a smartphone or tablet. This terminal includes a camera means, a processing means, a communication means, an emotional data collection means, and a display means. The camera means has a function that allows the user to take a photo of a book report or assignment. The processing means includes optical character recognition (OCR) software, specifically "Tesseract OCR." The communication means has a function to send and receive data using the HTTPS protocol. The emotional data collection means includes a face recognition camera and microphone, and has a function to capture facial expressions and voice. The display means visually displays the evaluation results and emotional feedback to the user.

[0620] Server Hardware and Software

[0621] The server is located in a cloud environment and has powerful computing resources. The server includes a receiving means, an analyzing means, an evaluating means, an analysis reflecting means, and a result transmitting means. The receiving means receives data transmitted from the terminal and stores it. The analyzing means analyzes the text data using a natural language processing (NLP) model. Specifically, a generative AI model such as "GPT-4" is used. The evaluating means calculates the probability that the sentence is AI-generated based on the analysis results. Furthermore, the analysis reflecting means reflects the user's emotional data in the analysis results and adjusts the reliability of the evaluation. The result transmitting means transmits the evaluation results to the terminal.

[0622] System Operation

[0623] The system works as follows: The user uses the device's camera to take an image of a book report or assignment. The device's OCR software extracts text information from the image and formats this data. The formatted text data is sent to the server using a secure communications protocol. The server receives the text data and inputs it into a natural language processing model for analysis. Based on the analysis results, it calculates the probability that the text is AI-generated and sends this evaluation result to the device.

[0624] Furthermore, the device's emotional data collection means uses the user's facial recognition camera and microphone to capture facial expressions and voice data, generating emotional data. This emotional data is sent to the server, which then reflects it in the analysis results. Finally, the device displays the evaluation results and emotional feedback to the user. For example, it may display, "There is an 80% chance that this sentence is an AI-generated sentence. User's emotional state: Confused 45%."

[0625] Specific examples

[0626] For example, consider the case where a faculty member wants to evaluate a student's book report. The user launches a dedicated app and takes a photo of the report with their camera. The system extracts text from the image, formats it, and sends it to the server. Analysis is performed on the server side, and an evaluation is given, such as "There is an 80% chance that this sentence was generated by AI." Furthermore, when the user checks the evaluation results, the emotion engine analyzes the user's emotions and provides feedback such as "Confused 45%." This allows the user to objectively confirm the reliability of the evaluation results.

[0627] Prompt Sentence Examples

[0628] Here is an example prompt:

[0629] "The following text is a book report written by a student. Please rate whether this text was generated by AI. Also, please analyze the user's emotional state at the time of rating."

[0630] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0631] Step 1:

[0632] The user launches the app and uses the smartphone camera to take a photo of a book report or university assignment. The input is the image captured by the camera, and the output is the image stored in the device's temporary memory. Specifically, the user taps the "take a photo" button in the app, focuses the camera on the object, and takes a photo.

[0633] Step 2:

[0634] The device uses OCR technology to extract text information from a captured image. The input is image data from temporary memory, and the output is the extracted text data. Specifically, the device's OCR engine (e.g., "Tesseract OCR") analyzes the image and converts the text information into text format.

[0635] Step 3:

[0636] The terminal formats the extracted text data. The input is the text data obtained by OCR processing, and the output is the formatted text data. Specifically, a program is executed to correct unnecessary line breaks, spaces, and misrecognized characters.

[0637] Step 4:

[0638] Sends formatted text data to the server. The input is the formatted text data, and the output is the data sent to the server. Specifically, the text data is encrypted and sent using a secure communication protocol (e.g., HTTPS).

[0639] Step 5:

[0640] The server receives the text data and inputs it into a natural language processing model for analysis. The input is the received text data, and the output is the analyzed result. Specifically, the server receives the HTTPS request, saves it in a database, and then inputs the data into an NLP engine (e.g., "GPT-4") to perform analysis.

[0641] Step 6:

[0642] The server calculates the probability that the sentence is AI-generated based on the analysis results. The input is the analysis result of the NLP engine, and the output is a probability value. Specifically, the generative AI model evaluates the analysis result and calculates the probability in the form of "There is an 80% chance that this sentence is AI-generated."

[0643] Step 7:

[0644] The server sends the evaluation results to the terminal. The input is the evaluation results including the probability values, and the output is the data sent to the terminal. Specifically, the evaluation results are sent to the terminal in JSON format using a secure communication protocol.

[0645] Step 8:

[0646] The device's emotion data collection means acquires the user's facial expressions and voice. The input is the user's face and voice, and the output is the acquired emotion data. Specifically, the device's camera and microphone operate simultaneously to capture the facial expressions and voice.

[0647] Step 9:

[0648] The device sends emotion data to the server. The input is the acquired emotion data, and the output is the data sent to the server. Specifically, the emotion data is formatted in JSON format and sent using a secure protocol.

[0649] Step 10:

[0650] The server reflects the emotional data in the analysis results. The input is the received emotional data and the analysis results, and the output is the final evaluation result that reflects the emotions. Specifically, the server analyzes the emotional data and adjusts the reliability of the evaluation results.

[0651] Step 11:

[0652] The device displays the final evaluation result and emotional feedback. The input is the final evaluation result received from the server, and the output is the evaluation result and emotional feedback displayed to the user. Specifically, the device's UI displays "There is an 80% chance that this sentence is an AI-generated sentence. User's emotional state: Confused 45%."

[0653] (Application example 2)

[0654] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0655] Passengers in autonomous vehicles are likely to be in various emotional states, and safer and more comfortable driving is required. However, current autonomous driving systems have difficulty recognizing passengers' emotional states in real time and adjusting vehicle operating parameters accordingly. This can lead to inability to respond appropriately to passengers' anxiety or confusion, which could result in a decrease in safety and comfort.

[0656] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0657] In this invention, the server includes a camera means of the terminal that acquires input data, a processing means of the terminal that extracts text from images using optical character recognition (OCR) technology, an analysis means of the server that inputs the text data into a natural language processing model for analysis, an emotion analysis means of the terminal or server that analyzes the user's emotion using the camera and voice input means, and an operation adjustment means that adjusts the vehicle's operation parameters based on the analysis results of the emotion analysis means. This makes it possible to recognize the emotional state of passengers in real time and appropriately adjust the vehicle's operation parameters based on the emotional state.

[0658] "Input data" refers to data such as images and sounds that are acquired by the system and are the subject of analysis.

[0659] "Camera means of the terminal" refers to a means for acquiring image or video data using a camera mounted on the terminal.

[0660] "Optical character recognition (OCR) technology" is a technology that extracts character information from image data.

[0661] "Terminal processing means" refers to a means that has the function of analyzing and processing data within the terminal.

[0662] "Communication means" refers to a means for transmitting and receiving data between a terminal and a server.

[0663] A "server" is a central computer system that performs data analysis and processing.

[0664] The "receiving means" is a means by which the server receives data transmitted from the terminal.

[0665] A "natural language processing model" is an algorithm or technology for analyzing text data and understanding and generating its content.

[0666] The "analysis means" is a means for analyzing the data received by the server.

[0667] The "evaluation means" is a means for calculating the probability that a text is an AI-generated sentence based on the analysis results.

[0668] The "result transmission means" is a means by which the server transmits the analysis results to the terminal.

[0669] The "display means" is a means for visually presenting the evaluation results received by the terminal to the user.

[0670] The "emotion analysis means" is a means for analyzing the user's emotional state using a camera and a voice input means.

[0671] The "operation adjustment means" is a means for adjusting the vehicle operation parameters based on the analysis results of the emotion analysis means.

[0672] The present invention provides a system for recognizing a user's emotional state in real time and adjusting vehicle operating parameters accordingly. The system includes the following means:

[0673] Hardware and software used

[0674] The system requires a device equipped with a camera and microphone, and a cloud server for analyzing the data. The device includes a camera for capturing video data and a voice input for capturing audio data. The software includes optical character recognition (OCR) technology, natural language processing models, and DeepFace and SpeechRecognition libraries for emotion recognition.

[0675] Data processing and calculation

[0676] The server converts the image data sent from the device into text data using OCR technology. This allows it to analyze the content of the text data and calculate the probability that the text was generated by AI. Meanwhile, the device or server uses a camera and voice input means to capture the user's facial expressions and voice in real time, and analyzes the user's emotional state based on this data.

[0677] Emotion analysis uses the DeepFace library to analyze facial expression data and identify the user's dominant emotion (e.g., anger, sadness, surprise, etc.). For voice data, the SpeechRecognition library is used to first convert the voice to text, and then the text is analyzed for emotion. The results of the emotion analysis are sent from the device to a server, which uses this data to generate instructions for adjusting the vehicle's operating parameters (speed, safety distance, etc.).

[0678] Specific examples

[0679] For example, if a passenger expresses anxiety while using the autonomous taxi application, the system will detect the passenger's emotional state and reduce the vehicle's speed and increase the safety distance in real time. The in-car environment (music, lighting, etc.) will also be adjusted according to the passenger's emotions, allowing passengers to enjoy a safer and more comfortable riding experience.

[0680] Prompt Sentence Examples

[0681] A possible prompt would be:

[0682] Generate a Python program to recognize passenger emotions in real time and adjust the driving parameters of an autonomous vehicle based on the emotions. Use cameras and microphones installed in the vehicle to perform facial and voice emotion analysis using DeepFace and SpeechRecognition libraries. Obtain emotion data in real time and reduce the vehicle's speed and increase the safety distance if the passenger is anxious.

[0683] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0684] Step 1:

[0685] The terminal uses a camera and microphone to capture passenger image and voice data. These data are captured in real time and used for subsequent processing. The input is image and voice data, and the output is a real-time capture of these data.

[0686] Step 2:

[0687] The DeepFace library is used to analyze the image data captured by the device to identify the passenger's emotional state. It analyzes facial patterns and expressions to identify the dominant emotion (e.g., joy, anger, sadness, etc.). The input is the image data, and the output is the identified emotional state.

[0688] Step 3:

[0689] The voice data acquired by the device is converted into text using the SpeechRecognition library, and then the text content is analyzed to infer the emotional state. The emotional nuances of the spoken content are evaluated. The input is voice data, and the output is text data containing the emotional state.

[0690] Step 4:

[0691] The device integrates the emotional states identified in steps 2 and 3 to generate an overall emotional assessment. The device outputs the overall emotional assessment from multiple emotional data. The input is emotional data from images and text, and the output is an overall emotional assessment.

[0692] Step 5:

[0693] The device sends a comprehensive emotional evaluation to the server, which then generates instructions to adjust the vehicle's operating parameters based on the evaluation results. Specifically, it adjusts the speed or changes the safety distance. The input is comprehensive emotional evaluation data, and the output is instructions to adjust the vehicle's operating parameters.

[0694] Step 6:

[0695] The server generates an instruction to adjust the operating parameters and sends it to the terminal, which then adjusts the vehicle's operating parameters in real time based on the instruction received. Specific actions include reducing the vehicle's speed, maintaining distance, etc. The input is the instruction to adjust the operating parameters, and the output is the adjusted vehicle's operating parameters.

[0696] Step 7:

[0697] The user (passenger) checks the displayed evaluation results and the vehicle's operating status. The terminal displays the analyzed emotional state and the operating parameters based on it to the user. The input is the analyzed emotional state and operating parameters, and the output is visual feedback to the user.

[0698] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0699] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0700] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0701] [Third embodiment]

[0702] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0703] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0704] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0705] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0706] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0707] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0708] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0709] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0710] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0711] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0712] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0713] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0714] The present invention relates to a system that identifies whether book reports and university assignments submitted by students are AI-generated texts, which can be used by faculty and staff to quickly and accurately evaluate the submissions.

[0715] System Configuration

[0716] This system mainly consists of terminals, servers, and users. Terminals are mobile devices such as smartphones and tablets, and servers are computing resources deployed in a cloud environment. Users are primarily faculty and staff who use this system to evaluate student submissions.

[0717] Program processing

[0718] 1. The user captures input data using the smartphone camera.

[0719] The user launches the system's dedicated app and uses the camera function to take a photo of the book report paper.

[0720] The captured image is temporarily stored on the device.

[0721] 2. The device performs OCR processing to extract the text data.

[0722] The device performs OCR (optical character recognition) technology to extract text information from the captured image.

[0723] The extracted character information is stored in the terminal as text data.

[0724] 3. The device formats the text data and sends it to the server

[0725] The extracted text data is formatted on the device, and unnecessary line breaks, spaces, and misrecognized characters are corrected.

[0726] The formatted text data is sent to the server using a secure protocol (e.g., HTTPS).

[0727] 4. The server receives the text data and begins analyzing it.

[0728] The server receives the text data sent from the terminal.

[0729] The received text data is input into a natural language processing model and analysis begins.

[0730] 5. The server evaluates the analysis results and sends them to the device.

[0731] A natural language processing model analyzes the text data and calculates the probability that a sentence is AI-generated.

[0732] The server evaluates the analysis results and summarizes them in a format that is easy for the user to understand, such as "There is an 85% chance that this sentence was generated by an AI."

[0733] The evaluation results are then transmitted to the terminal again using a secure communication protocol.

[0734] 6. The device displays the evaluation results.

[0735] The terminal receives the analysis results sent from the server.

[0736] The received analysis results are displayed to the user, who can use the results to determine whether the student's submission is AI-generated.

[0737] Specific examples

[0738] For example, imagine a scenario in which a faculty member wants to evaluate a student's book report. The user launches a dedicated app and takes a photo of the report with their camera. The system extracts text from the image, formats it, and sends it to the server. Analysis is then performed on the server, and the result is determined to be "80% likely to be an AI-generated sentence." The user can review this result and, if necessary, conduct further research or provide feedback to the student.

[0739] In this way, the system provides faculty with advanced tools to effectively evaluate student work and helps prevent cheating.

[0740] The processing flow will be explained below.

[0741] Step 1:

[0742] The user activates the smartphone camera and takes a picture of the book report paper. The user then controls the camera through the app to capture the image with the appropriate framing.

[0743] Step 2:

[0744] The device saves the captured image and starts OCR processing. The device then analyzes the image and extracts the character information as text data.

[0745] Step 3:

[0746] The device then formats the extracted text data, specifically by removing unnecessary line breaks and spaces from the OCR results and correcting any misrecognized characters.

[0747] Step 4:

[0748] The device sends the formatted text data to the server. The device uses a secure communication protocol such as HTTPS to securely transmit the data to the server.

[0749] Step 5:

[0750] The server receives the text data sent from the device and prepares it for analysis. The server temporarily stores the received data for subsequent analysis.

[0751] Step 6:

[0752] The server inputs the text data into a natural language processing model and begins analysis, specifically to detect patterns and phrases characteristic of AI-generated text.

[0753] Step 7:

[0754] The server evaluates the analysis results and calculates the likelihood that the sentence is AI-generated. Based on the probability calculated by the AI ​​model, it generates an evaluation such as "high probability that the sentence is AI-generated" or "low probability."

[0755] Step 8:

[0756] The server formats the evaluation results and generates results in a format that is easy for users to understand. For example, it generates an evaluation result such as "There is an 80% chance that this sentence was generated by an AI."

[0757] Step 9:

[0758] The server transmits the evaluation results to the terminal, and again using a secure communication protocol, the server transmits the evaluation results to the terminal safely.

[0759] Step 10:

[0760] The terminal displays the evaluation results received from the server, and the terminal visually displays the evaluation results to the user, allowing the user to confirm the results.

[0761] Step 11:

[0762] Users can check the evaluation results and determine whether or not there has been any cheating. Based on the displayed evaluation results, users can conduct further detailed verification and provide feedback to students.

[0763] In this way, a system has been created in which users, devices, and servers work together to quickly and accurately evaluate whether student submissions are AI-generated texts.

[0764] Example 1

[0765] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0766] In recent years, advances in artificial intelligence technology have created a need for a fast and accurate way to assess whether student submissions are AI-generated. However, current systems rely heavily on manual review, which not only takes time and effort but also often lacks accuracy. Therefore, a method is needed for faculty and staff to efficiently assess submissions and prevent cheating.

[0767] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0768] In this invention, the server includes a photographing means of the terminal that acquires input data, an analysis means of the terminal that extracts characters from the image using optical character recognition technology, a communication means that transmits the extracted character data to the information processing device, a receiving means by the information processing device that receives the character data, an analysis means of the information processing device that inputs the character data into a natural language processing model and analyzes it, an evaluation means that calculates the probability that the sentence is an AI-generated sentence based on the analysis result, a result transmission means that transmits the evaluation result to the terminal via the communication means, and a display means on the terminal that displays the evaluation result. This enables faculty and staff to quickly and accurately evaluate whether a submitted work was generated by AI.

[0769] "Input data" is digital information that a user captures using a camera or other device.

[0770] A "terminal" is a device that is directly operated by a user, such as a smartphone, tablet, or PC.

[0771] "Photographing means" refers to the hardware and software, including the camera function, installed in the terminal.

[0772] "Optical character recognition technology" is a technology for reading and extracting text information from images.

[0773] "Analysis means" refers to the function of executing optical character recognition technology on the terminal and extracting character data from an image.

[0774] "Character data" refers to text-based data extracted using optical character recognition technology.

[0775] An "information processing device" is a device that performs data processing, including a server or cloud computing resource.

[0776] "Communication means" refers to an interface and protocol for transmitting and receiving data between a terminal and an information processing device.

[0777] The "receiving means" refers to a function by which the information processing device receives data transmitted from the terminal.

[0778] A "natural language processing model" is an algorithm and software that analyzes text data and determines whether it was generated by artificial intelligence.

[0779] The term "analysis means" refers to a function of an information processing device to input character data into a natural language processing model and analyze it.

[0780] "Evaluation means" refers to a function that calculates the probability that the input text was generated by AI based on the analysis results.

[0781] "Result transmission means" refers to a communication interface and protocol for transmitting evaluation results to a terminal.

[0782] "Display means" refers to an interface that allows the terminal to visually present the evaluation results to the user.

[0783] "Formatting means" refers to a function that formats extracted character data by correcting misrecognitions and deleting unnecessary line breaks and spaces.

[0784] The "evaluation generation means" refers to a function that summarizes and presents the analysis results in a format that is easy for the user to understand.

[0785] This invention relates to a system for identifying whether book reports and university assignments submitted by students have been generated by artificial intelligence. This system allows faculty and staff to quickly and accurately evaluate the submissions. The system mainly consists of a terminal, a server, and a user.

[0786] Hardware and software examples

[0787] Terminal

[0788] The terminal is a device that is directly operated by the user, such as a smartphone, tablet, or PC. The terminal is equipped with a camera, which is used to capture input data such as a book report. The terminal is equipped with optical character recognition (OCR) technology, such as Google Tesseract OCR, to extract text data from images. In addition, the terminal includes a communication means for sending the submitted text data to a server. The HTTPS protocol is used for secure data communication.

[0789] server

[0790] The server is an information processing device that operates in a cloud environment and receives and analyzes text data sent from the terminal. The server has a receiving means and can receive data sent from the terminal. The received data is input into a natural language processing model (for example, "OpenAI's GPT-4") for analysis. The analysis means inputs the text data into this natural language processing model and calculates the probability that the text was generated by AI. The evaluation means evaluates the analysis results and summarizes them in a format that is easy for the user to understand. The evaluation results are then sent back to the terminal using the HTTPS protocol.

[0791] User

[0792] Users are primarily faculty and staff, who use the system to evaluate student submissions. Using a smartphone, tablet, or other device, users launch a dedicated app and take a picture of the submission. After taking the picture, text data extracted from the image is sent to a server, and the analysis results are displayed on the device. This allows users to quickly determine whether the submission was generated by artificial intelligence.

[0793] Examples of concrete examples and prompts

[0794] For example, imagine a scenario in which a faculty member wants to evaluate a student's book report. The user launches a dedicated app and takes a photo of the report with their camera. The system extracts text data from the image, formats it, and sends it to the server. Analysis is then performed on the server, and the result is that "there is an 80% chance that this sentence was generated by AI." The user can review this result and, if necessary, conduct further research or provide feedback to the student.

[0795] Examples of prompts include:

[0796] "Please rate whether the following sentences were generated by an AI:"

[0797] "Calculate the probability that the following text is AI-generated:"

[0798] This provides faculty with advanced tools to effectively assess student work and helps prevent cheating.

[0799] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0800] Step 1:

[0801] The user launches the system's dedicated app and takes a photo of the book report paper using the smartphone camera. The input data is an image of the paper, and the output is a digital image file that is temporarily stored inside the device. When the user taps the "take a photo" button, the camera starts and the image is saved.

[0802] Step 2:

[0803] The device analyzes the captured image using OCR (Optical Character Recognition) technology. The input data is the output image from step 1, and the output is the extracted character data. The OCR engine on the device analyzes the image and generates text data. Specifically, Google's Tesseract OCR library is used. The processed results are saved on the device as a text file.

[0804] Step 3:

[0805] The terminal formats the extracted text data. The input data is the output text from step 2, and the output is the formatted text data. Specifically, unnecessary line breaks and spaces are removed, and misrecognized characters are corrected. The formatted text data is converted to JSON format and is ready to be sent to the server.

[0806] Step 4:

[0807] The terminal sends the formatted text data to the server. The input data is the output text data of step 3, and the output is a data transmission success message to the server. The HTTPS protocol is used as the communication method, and the terminal sends the text data to the server in a secure state.

[0808] Step 5:

[0809] The server receives the text data sent from the terminal. The input data is the text data sent from the terminal, and the output is the received text data. The server's receiving means captures the data and stores it in a specific directory or database.

[0810] Step 6:

[0811] The server inputs the received text data into a natural language processing (NLP) model for analysis. The input data is the received text data from step 5, and the output is the analysis result. Specifically, a generative AI model such as OpenAI's GPT-4 is used to analyze whether the text data is AI-generated. In this process, a natural language processing algorithm is executed.

[0812] Step 7:

[0813] The server calculates the probability that the sentence is AI-generated based on the analysis results. The input data is the analysis result from step 6, and the output is an evaluation result including a probability value. Based on the characteristics of the analyzed text data and the prompt sentence, the server calculates an evaluation such as "There is an 85% chance that this sentence is AI-generated."

[0814] Step 8:

[0815] The server sends the evaluation result to the terminal. The input data is the evaluation result from step 7, and the output is a message to the terminal indicating that the transmission was successful. The evaluation result is formatted in a user-friendly format and sent to the terminal using the HTTPS protocol.

[0816] Step 9:

[0817] The device receives the evaluation results sent from the server and displays them to the user. The input data is the evaluation results sent from the server, and the output is the evaluation results displayed to the user. The device displays the results it receives on the screen, and the user is presented with a message saying, "There is an 85% chance that this sentence is AI-generated." Based on this information, the user can determine whether the submission was AI-generated.

[0818] This allows faculty and staff to quickly and accurately evaluate submissions through a series of processes, preventing cheating before it occurs.

[0819] (Application example 1)

[0820] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0821] With conventional systems, it was difficult to quickly and accurately determine whether student submissions or internal company documents were generated by AI, which caused problems for educational institutions and companies when verifying fraud and the authenticity of content. Furthermore, users could not obtain the results in real time, which led to a lack of efficiency. Furthermore, there were no concrete solutions using wearable devices.

[0822] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0823] In this invention, the server includes a camera means of the terminal that acquires input data, a processing means of the terminal that extracts text from images using optical character recognition (OCR) technology, a communication means that transmits the extracted text data to the server, a receiving means by the server that receives the text data, an analysis means of the server that inputs the text data into a natural language processing model for analysis, an evaluation means that calculates the probability that the text is an AI-generated sentence based on the analysis result, a result transmission means that transmits the evaluation result to the terminal via the communication means, a display means of the terminal that displays the evaluation result, a capture means that captures documents shared on the Internet or in internal documents and identifies whether they were generated by AI, and a wearable device that displays the evaluation result to the user in real time, thereby enabling the user to quickly and accurately obtain a judgment result on the AI-generated sentence.

[0824] The "camera means of the terminal for acquiring input data" refers to a means for capturing target documents or image data using a camera equipped on the terminal.

[0825] "Terminal processing means for extracting text from images using optical character recognition (OCR) techniques" means means for recognizing character information from a captured image and extracting it as text data.

[0826] The "communication means for transmitting extracted text data to a server" is a communication function for transmitting text data from a terminal to a server, for example, a means for using an Internet connection.

[0827] The "receiving means by which the server receives text data" refers to a function or device on the server side that receives text data sent from the terminal.

[0828] The "analysis means of the server that inputs text data into a natural language processing model for analysis" is a function of the server that uses natural language processing technology to analyze the acquired text data.

[0829] The "evaluation means for calculating the probability that a sentence is generated by AI based on the analysis results" is a function that calculates the probability that the sentence was generated by AI based on the analyzed text data.

[0830] The "result transmission means for transmitting the evaluation results to the terminal via the communication means" is a function for transmitting the analysis and evaluation results from the server to the terminal.

[0831] The "display means for the terminal to display the evaluation results" is a function for displaying the evaluation results on the terminal so that the user can visually confirm the analysis results.

[0832] "Capture means for capturing documents shared on the Internet or in internal documents and identifying whether they were generated by AI" refers to a means by which a user can capture documents on the Internet or in internal documents and determine whether their contents were generated by AI.

[0833] A "wearable device that displays evaluation results to a user in real time" is a pair of glasses or other device that can be worn by a user and that displays evaluation results in real time.

[0834] This invention relates to a system that can quickly and accurately identify whether book reports submitted by students, university assignments, internal corporate documents, and documents shared on the Internet have been generated by AI. This system is useful for preventing fraud and verifying the authenticity of content in security departments and educational institutions.

[0835] The basic system configuration is as follows:

[0836] 1. The device (smartphone or wearable device) is equipped with a camera that allows the user to capture the target document.

[0837] 2. The captured image is converted into text data using optical character recognition (OCR) technology such as Google Cloud Vision API or Tesseract OCR.

[0838] 3. The converted text data is formatted within the terminal, and unnecessary parts such as line breaks and spaces are removed.

[0839] 4. The formatted text data is sent to the server using a secure protocol (e.g., HTTPS).

[0840] 5. The server analyzes the text data using a natural language processing model (e.g., Hugging Face's Transformers library or GPT-3).

[0841] 6. The analysis results are evaluated by calculating the probability that the text data was generated by AI.

[0842] 7. The evaluation results are sent from the server to the device and displayed to the user in real time. Particularly when using a wearable device, users can check the evaluation results immediately.

[0843] Examples:

[0844] Example 1: Corporate security department use

[0845] During a meeting, a company security officer captures a printed internal document with the smart glasses' camera. The captured image is converted into text data using the smart glasses' built-in OCR system, and the data is sent to a server via secure communication. The server analyzes the text data and displays the assessment result in real time on the glasses' display: "There is an 85% chance that this document was generated by AI."

[0846] Example 2: Faculty and staff at an educational institution reviewing student submissions

[0847] Faculty and staff use their smartphones to take photos of students' book reports. The captured images are then processed by OCR on the smartphone and converted into text data. This text data is then sent to a server and analyzed using a natural language processing model. The analysis results, such as "There is a 70% chance that this review was generated by AI," are displayed on the smartphone screen.

[0848] Prompt Sentence Examples

[0849] For internal company documents: "Please determine whether the following document was generated by AI:\n\n'Regarding customer data management methods, we would like to propose a new system based on the content of our recent meeting.'"

[0850] For educational student submissions: "Please determine if this student's book report was generated by AI: 'Reading this book helped me to gain a deeper understanding of different cultural backgrounds.'"

[0851] In this way, the system of the present invention enables faculty and security personnel to efficiently and accurately identify AI-generated sentences and quickly provide necessary feedback and countermeasures.

[0852] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0853] Step 1:

[0854] The user captures a document using the device's camera or smart glasses. The input is image data of the document. Specifically, the user launches the device's camera app and takes a picture of the target document. This image data is temporarily stored inside the device.

[0855] Step 2:

[0856] The device performs OCR processing on the captured image data to extract text data. The input is image data and the output is text data. Specifically, the device uses the Google Cloud Vision API and Tesseract OCR engine to recognize and extract character information from the image. This extraction process obtains the document contents as text data.

[0857] Step 3:

[0858] The terminal formats the extracted text data and removes unnecessary characters such as line breaks and spaces. The input is text data obtained by OCR, and the output is formatted text data. Specifically, using regular expressions and string manipulation libraries, unnecessary characters and line breaks are removed from the text, and the formatted text data is generated.

[0859] Step 4:

[0860] The formatted text data is sent from the terminal to the server. The input is the formatted text data, and the output is the state after transmission to the server has been completed. Specifically, the text data is sent to the server using a secure communication protocol (e.g., HTTPS). This procedure allows the text data to be received in a format that can be analyzed on the server side.

[0861] Step 5:

[0862] The server inputs the received text data into a natural language processing model to begin analysis. The input is formatted text data, and the output is the analysis result. Specifically, the text data is analyzed using natural language processing models such as Hugging Face's Transformers library and GPT-3. This analysis calculates the probability that the text was generated by AI.

[0863] Step 6:

[0864] The server calculates the probability that the text is AI-generated based on the analysis results and performs an evaluation. The input is the analyzed text data, and the output is the evaluation result. Specifically, based on the output of the analysis model, an evaluation such as "There is an 85% chance that this document is AI-generated text" is generated.

[0865] Step 7:

[0866] The server sends the evaluation results to the terminal. The input is the evaluation results, and the output is the completion of transmission to the terminal. Specifically, the evaluation results are sent to the terminal again using a secure communication protocol such as HTTPS.

[0867] Step 8:

[0868] The device displays the evaluation results and notifies the user. The input is the evaluation results received from the server, and the output is what is displayed to the user. Specifically, a message such as "There is an 85% chance that this document is AI-generated" is displayed in real time using the device's display or the display function of the smart glasses. This allows the user to quickly check the evaluation results.

[0869] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0870] This invention combines a system that identifies whether book reports and university assignments submitted by students are generated by AI with a function that recognizes user emotions. This system allows faculty and staff to quickly and accurately evaluate submissions, and by understanding the user's emotional state, it provides more flexible and effective feedback.

[0871] System Configuration

[0872] This system mainly consists of terminals, servers, users, and an emotion engine. Terminals are mobile devices such as smartphones and tablets, and the server is a computing resource deployed in a cloud environment. Users are primarily faculty and staff who use this system to evaluate students' submissions. The emotion engine has the ability to analyze users' facial expressions and voices and identify their emotional state.

[0873] Program processing

[0874] 1. The user captures input data using the smartphone camera.

[0875] The user launches the system's dedicated app and uses the camera function to take a photo of the book report paper.

[0876] The captured image is temporarily stored on the device.

[0877] 2. The device performs OCR processing to extract the text data.

[0878] The device performs OCR (optical character recognition) technology to extract text information from the captured image.

[0879] The extracted character information is stored in the terminal as text data.

[0880] 3. The device formats the text data and sends it to the server

[0881] The extracted text data is formatted on the device, and unnecessary line breaks, spaces, and misrecognized characters are corrected.

[0882] The formatted text data is sent to the server using a secure protocol (e.g., HTTPS).

[0883] 4. The server receives the text data and begins analyzing it.

[0884] The server receives the text data sent from the terminal.

[0885] The received text data is input into a natural language processing model and analysis begins.

[0886] 5. The server evaluates the analysis results and sends them to the device.

[0887] A natural language processing model analyzes the text data and calculates the probability that a sentence is AI-generated.

[0888] The server evaluates the analysis results and summarizes them in a format that is easy for the user to understand, such as "There is an 80% chance that this sentence was generated by an AI."

[0889] The evaluation results are then transmitted to the terminal again using a secure communication protocol.

[0890] 6. Emotion engine analyzes user emotions

[0891] The emotion engine installed on the device captures facial expressions and voice data from the user through a facial recognition camera and microphone.

[0892] Based on the acquired data, the user's current emotional state is classified and stored as emotional data in the device.

[0893] 7. The device sends the emotion data to the server

[0894] The device sends emotion data to the server, which is also sent using a secure protocol.

[0895] 8. The server reflects the emotion data in the analysis results

[0896] The server reflects the received emotion data in the analysis results. For example, if the user shows strong emotion (surprise or confusion), it adjusts the reliability of the analysis results.

[0897] As a result, the evaluation results can be corrected based on the user's emotional state.

[0898] 9. The device displays the final evaluation result and emotional feedback.

[0899] The device displays the final evaluation results and emotional feedback to the user, who can check not only the probability of the AI-generated sentences but also their own emotional data.

[0900] For example, it might say, "There is an 80% chance that this sentence was generated by AI. User's emotional state: Confused 45%."

[0901] Specific examples

[0902] For example, imagine a scenario in which a faculty member tries to evaluate a student's book review. The user launches a dedicated app and takes a photo of the review with their camera. The system extracts text from the image, formats it, and sends it to the server. Analysis is then performed on the server, resulting in an evaluation such as "There is an 80% chance that this sentence was generated by AI." When the user checks the evaluation results, the emotion engine analyzes the user's emotions and provides feedback such as "Confused 45%." This allows the user to objectively confirm the reliability of the evaluation results.

[0903] The system not only provides faculty with advanced tools to effectively grade student submissions, helping to prevent cheating, but also allows for more flexible grading through feedback based on sentiment data.

[0904] The processing flow will be explained below.

[0905] Step 1:

[0906] The user activates the smartphone camera and takes a picture of the book report paper. The user then controls the camera through a dedicated app to capture the image with the appropriate framing.

[0907] Step 2:

[0908] The device saves the captured image and starts OCR processing. The device then analyzes the image and extracts the character information as text data.

[0909] Step 3:

[0910] The device then formats the extracted text data, specifically by removing unnecessary line breaks and spaces from the OCR results and correcting any misrecognized characters.

[0911] Step 4:

[0912] The device sends the formatted text data to the server. The device uses a secure communication protocol such as HTTPS to securely transmit the data to the server.

[0913] Step 5:

[0914] The server receives the text data sent from the device and prepares it for analysis. The server temporarily stores the received data for subsequent analysis.

[0915] Step 6:

[0916] The server inputs the text data into a natural language processing model and begins analysis, specifically to detect patterns and phrases characteristic of AI-generated text.

[0917] Step 7:

[0918] The server evaluates the analysis results and calculates the likelihood that the sentence is AI-generated. Based on the probability calculated by the AI ​​model, it generates an evaluation such as "high probability that the sentence is AI-generated" or "low probability."

[0919] Step 8:

[0920] The server formats the evaluation results and generates results in a format that is easy for users to understand. For example, it generates an evaluation result such as "There is an 80% chance that this sentence was generated by an AI."

[0921] Step 9:

[0922] The server transmits the evaluation results to the terminal, and again using a secure communication protocol, the server transmits the evaluation results to the terminal safely.

[0923] Step 10:

[0924] The terminal displays the evaluation results received from the server, and the terminal visually displays the evaluation results to the user, allowing the user to confirm the results.

[0925] Step 11:

[0926] The emotion engine acquires the user's emotion data. The device uses a camera and microphone to collect the user's facial expression and voice data, which the emotion engine analyzes to identify the user's emotional state.

[0927] Step 12:

[0928] The device transmits the acquired emotion data to the server, which also uses a secure protocol.

[0929] Step 13:

[0930] The server receives the emotion data and performs corrections based on the analysis. The server takes into account the user's emotional state identified by the emotion engine as a factor that influences the evaluation results.

[0931] Step 14:

[0932] The server generates the final evaluation result and emotional feedback and sends it to the terminal. The evaluation result takes into account the user's emotional data and is sent to the terminal as the final evaluation result.

[0933] Step 15:

[0934] The device will then display the final evaluation results and emotional feedback to the user, such as "There is an 80% chance that this sentence was generated by AI. User's emotional state: Confused 45%."

[0935] Step 16:

[0936] The user checks the final evaluation results and determines whether there was any cheating. Based on the displayed evaluation results and emotional feedback, the user can conduct further investigations and provide feedback to the students.

[0937] Through these steps, users can not only identify AI-generated sentences but also check the reliability of the evaluation based on their own emotional state, which can effectively prevent cheating in educational settings.

[0938] Example 2

[0939] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0940] Conventional student submission evaluation systems have difficulty identifying whether book reports or university assignments are generated by AI. Furthermore, they do not provide feedback that takes into account the evaluator's emotional state. As a result, it is difficult to provide reliable evaluations and effective feedback. This invention solves these problems, prevents cheating in submissions, and enables flexible feedback based on the evaluator's emotions.

[0941] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for inputting text data into a natural language processing model for analysis, an evaluation means for calculating the probability that the text is an AI-generated sentence based on the analysis result, and an analysis reflection means for reflecting emotion data in the analysis result. This makes it possible to identify with high accuracy whether the submitted work was generated by an AI and provide feedback appropriate to the evaluator's emotions.

[0942] The "camera means of the terminal that acquires input data" refers to a function that allows a user to take a photo of a submission such as a book report or a university assignment using the camera of a smartphone or tablet.

[0943] "Processing means of a terminal that extracts text from images using optical character recognition technology" refers to technology for extracting text information from captured images and the functionality within the terminal that executes this technology.

[0944] The "communication means for transmitting extracted text data to a server" refers to a communication protocol and function for transmitting text data from a terminal to a server.

[0945] The "receiving means for the server to receive text data" is a function for the server to receive text data sent from the terminal.

[0946] The "server's analysis means for inputting text data into a natural language processing model for analysis" refers to the server's function for inputting received text data into a natural language processing algorithm for analysis.

[0947] The "evaluation means for calculating the probability that a sentence is generated by AI based on the analysis results" is a server function that calculates the probability that a sentence was generated by AI based on the analysis results of a natural language processing model.

[0948] The "result transmission means for transmitting the evaluation results to the terminal via the communication means" refers to a communication protocol and function for transmitting the evaluation results obtained on the server side to the terminal.

[0949] The "emotion data collection means for the terminal to acquire the user's facial expressions and voice" is a function that acquires the user's facial expressions and voice using the camera and microphone installed in the terminal and uses them as emotion data.

[0950] "Emotion data transmission means for transmitting emotion data from a terminal to a server" refers to a communication protocol and function for transmitting emotion data collected by a terminal to a server.

[0951] The "analysis reflection means for the server to reflect emotion data in the analysis results" is a function for reflecting emotion data received by the server in the analysis results and adjusting the reliability of the evaluation.

[0952] The "display means by which the terminal displays the evaluation results and emotion feedback" is a function of the terminal for visually displaying the evaluation results and emotion data received from the server to the user.

[0953] The present invention provides a system that identifies whether a book report or university assignment submitted by a student was generated by AI and further reflects the user's emotional state in the evaluation. The system includes a terminal, a server, and an emotion engine. A specific embodiment of the system is described below.

[0954] Device hardware and software

[0955] The terminal is primarily a mobile device such as a smartphone or tablet. This terminal includes a camera means, a processing means, a communication means, an emotional data collection means, and a display means. The camera means has a function that allows the user to take a photo of a book report or assignment. The processing means includes optical character recognition (OCR) software, specifically "Tesseract OCR." The communication means has a function to send and receive data using the HTTPS protocol. The emotional data collection means includes a face recognition camera and microphone, and has a function to capture facial expressions and voice. The display means visually displays the evaluation results and emotional feedback to the user.

[0956] Server Hardware and Software

[0957] The server is located in a cloud environment and has powerful computing resources. The server includes a receiving means, an analyzing means, an evaluating means, an analysis reflecting means, and a result transmitting means. The receiving means receives data transmitted from the terminal and stores it. The analyzing means analyzes the text data using a natural language processing (NLP) model. Specifically, a generative AI model such as "GPT-4" is used. The evaluating means calculates the probability that the sentence is AI-generated based on the analysis results. Furthermore, the analysis reflecting means reflects the user's emotional data in the analysis results and adjusts the reliability of the evaluation. The result transmitting means transmits the evaluation results to the terminal.

[0958] System Operation

[0959] The system works as follows: The user uses the device's camera to take an image of a book report or assignment. The device's OCR software extracts text information from the image and formats this data. The formatted text data is sent to the server using a secure communications protocol. The server receives the text data and inputs it into a natural language processing model for analysis. Based on the analysis results, it calculates the probability that the text is AI-generated and sends this evaluation result to the device.

[0960] Furthermore, the device's emotional data collection means uses the user's facial recognition camera and microphone to capture facial expressions and voice data, generating emotional data. This emotional data is sent to the server, which then reflects it in the analysis results. Finally, the device displays the evaluation results and emotional feedback to the user. For example, it may display, "There is an 80% chance that this sentence is an AI-generated sentence. User's emotional state: Confused 45%."

[0961] Specific examples

[0962] For example, consider the case where a faculty member wants to evaluate a student's book report. The user launches a dedicated app and takes a photo of the report with their camera. The system extracts text from the image, formats it, and sends it to the server. Analysis is performed on the server side, and an evaluation is given, such as "There is an 80% chance that this sentence was generated by AI." Furthermore, when the user checks the evaluation results, the emotion engine analyzes the user's emotions and provides feedback such as "Confused 45%." This allows the user to objectively confirm the reliability of the evaluation results.

[0963] Prompt Sentence Examples

[0964] Here is an example prompt:

[0965] "The following text is a book report written by a student. Please rate whether this text was generated by AI. Also, please analyze the user's emotional state at the time of rating."

[0966] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0967] Step 1:

[0968] The user launches the app and uses the smartphone camera to take a photo of a book report or university assignment. The input is the image captured by the camera, and the output is the image stored in the device's temporary memory. Specifically, the user taps the "take a photo" button in the app, focuses the camera on the object, and takes a photo.

[0969] Step 2:

[0970] The device uses OCR technology to extract text information from a captured image. The input is image data from temporary memory, and the output is the extracted text data. Specifically, the device's OCR engine (e.g., "Tesseract OCR") analyzes the image and converts the text information into text format.

[0971] Step 3:

[0972] The terminal formats the extracted text data. The input is the text data obtained by OCR processing, and the output is the formatted text data. Specifically, a program is executed to correct unnecessary line breaks, spaces, and misrecognized characters.

[0973] Step 4:

[0974] Sends formatted text data to the server. The input is the formatted text data, and the output is the data sent to the server. Specifically, the text data is encrypted and sent using a secure communication protocol (e.g., HTTPS).

[0975] Step 5:

[0976] The server receives the text data and inputs it into a natural language processing model for analysis. The input is the received text data, and the output is the analyzed result. Specifically, the server receives the HTTPS request, saves it in a database, and then inputs the data into an NLP engine (e.g., "GPT-4") to perform analysis.

[0977] Step 6:

[0978] The server calculates the probability that the sentence is AI-generated based on the analysis results. The input is the analysis result of the NLP engine, and the output is a probability value. Specifically, the generative AI model evaluates the analysis result and calculates the probability in the form of "There is an 80% chance that this sentence is AI-generated."

[0979] Step 7:

[0980] The server sends the evaluation results to the terminal. The input is the evaluation results including the probability values, and the output is the data sent to the terminal. Specifically, the evaluation results are sent to the terminal in JSON format using a secure communication protocol.

[0981] Step 8:

[0982] The device's emotion data collection means acquires the user's facial expressions and voice. The input is the user's face and voice, and the output is the acquired emotion data. Specifically, the device's camera and microphone operate simultaneously to capture the facial expressions and voice.

[0983] Step 9:

[0984] The device sends emotion data to the server. The input is the acquired emotion data, and the output is the data sent to the server. Specifically, the emotion data is formatted in JSON format and sent using a secure protocol.

[0985] Step 10:

[0986] The server reflects the emotional data in the analysis results. The input is the received emotional data and the analysis results, and the output is the final evaluation result that reflects the emotions. Specifically, the server analyzes the emotional data and adjusts the reliability of the evaluation results.

[0987] Step 11:

[0988] The device displays the final evaluation result and emotional feedback. The input is the final evaluation result received from the server, and the output is the evaluation result and emotional feedback displayed to the user. Specifically, the device's UI displays "There is an 80% chance that this sentence is an AI-generated sentence. User's emotional state: Confused 45%."

[0989] (Application example 2)

[0990] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0991] Passengers in autonomous vehicles are likely to be in various emotional states, and safer and more comfortable driving is required. However, current autonomous driving systems have difficulty recognizing passengers' emotional states in real time and adjusting vehicle operating parameters accordingly. This can lead to inability to respond appropriately to passengers' anxiety or confusion, which could result in a decrease in safety and comfort.

[0992] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0993] In this invention, the server includes a camera means of the terminal that acquires input data, a processing means of the terminal that extracts text from images using optical character recognition (OCR) technology, an analysis means of the server that inputs the text data into a natural language processing model for analysis, an emotion analysis means of the terminal or server that analyzes the user's emotion using the camera and voice input means, and an operation adjustment means that adjusts the vehicle's operation parameters based on the analysis results of the emotion analysis means. This makes it possible to recognize the emotional state of passengers in real time and appropriately adjust the vehicle's operation parameters based on the emotional state.

[0994] "Input data" refers to data such as images and sounds that are acquired by the system and are the subject of analysis.

[0995] "Camera means of the terminal" refers to a means for acquiring image or video data using a camera mounted on the terminal.

[0996] "Optical character recognition (OCR) technology" is a technology that extracts character information from image data.

[0997] "Terminal processing means" refers to a means that has the function of analyzing and processing data within the terminal.

[0998] "Communication means" refers to a means for transmitting and receiving data between a terminal and a server.

[0999] A "server" is a central computer system that performs data analysis and processing.

[1000] The "receiving means" is a means by which the server receives data transmitted from the terminal.

[1001] A "natural language processing model" is an algorithm or technology for analyzing text data and understanding and generating its content.

[1002] The "analysis means" is a means for analyzing the data received by the server.

[1003] The "evaluation means" is a means for calculating the probability that a text is an AI-generated sentence based on the analysis results.

[1004] The "result transmission means" is a means by which the server transmits the analysis results to the terminal.

[1005] The "display means" is a means for visually presenting the evaluation results received by the terminal to the user.

[1006] The "emotion analysis means" is a means for analyzing the user's emotional state using a camera and a voice input means.

[1007] The "operation adjustment means" is a means for adjusting the vehicle operation parameters based on the analysis results of the emotion analysis means.

[1008] The present invention provides a system for recognizing a user's emotional state in real time and adjusting vehicle operating parameters accordingly. The system includes the following means:

[1009] Hardware and software used

[1010] The system requires a device equipped with a camera and microphone, and a cloud server for analyzing the data. The device includes a camera for capturing video data and a voice input for capturing audio data. The software includes optical character recognition (OCR) technology, natural language processing models, and DeepFace and SpeechRecognition libraries for emotion recognition.

[1011] Data processing and calculation

[1012] The server converts the image data sent from the device into text data using OCR technology. This allows it to analyze the content of the text data and calculate the probability that the text was generated by AI. Meanwhile, the device or server uses a camera and voice input means to capture the user's facial expressions and voice in real time, and analyzes the user's emotional state based on this data.

[1013] Emotion analysis uses the DeepFace library to analyze facial expression data and identify the user's dominant emotion (e.g., anger, sadness, surprise, etc.). For voice data, the SpeechRecognition library is used to first convert the voice to text, and then the text is analyzed for emotion. The results of the emotion analysis are sent from the device to a server, which uses this data to generate instructions for adjusting the vehicle's operating parameters (speed, safety distance, etc.).

[1014] Specific examples

[1015] For example, if a passenger expresses anxiety while using the autonomous taxi application, the system will detect the passenger's emotional state and reduce the vehicle's speed and increase the safety distance in real time. The in-car environment (music, lighting, etc.) will also be adjusted according to the passenger's emotions, allowing passengers to enjoy a safer and more comfortable riding experience.

[1016] Prompt Sentence Examples

[1017] A possible prompt would be:

[1018] Generate a Python program to recognize passenger emotions in real time and adjust the driving parameters of an autonomous vehicle based on the emotions. Use cameras and microphones installed in the vehicle to perform facial and voice emotion analysis using DeepFace and SpeechRecognition libraries. Obtain emotion data in real time and reduce the vehicle's speed and increase the safety distance if the passenger is anxious.

[1019] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1020] Step 1:

[1021] The terminal uses a camera and microphone to capture passenger image and voice data. These data are captured in real time and used for subsequent processing. The input is image and voice data, and the output is a real-time capture of these data.

[1022] Step 2:

[1023] The DeepFace library is used to analyze the image data captured by the device to identify the passenger's emotional state. It analyzes facial patterns and expressions to identify the dominant emotion (e.g., joy, anger, sadness, etc.). The input is the image data, and the output is the identified emotional state.

[1024] Step 3:

[1025] The voice data acquired by the device is converted into text using the SpeechRecognition library, and then the text content is analyzed to infer the emotional state. The emotional nuances of the spoken content are evaluated. The input is voice data, and the output is text data containing the emotional state.

[1026] Step 4:

[1027] The device integrates the emotional states identified in steps 2 and 3 to generate an overall emotional assessment. The device outputs the overall emotional assessment from multiple emotional data. The input is emotional data from images and text, and the output is an overall emotional assessment.

[1028] Step 5:

[1029] The device sends a comprehensive emotional evaluation to the server, which then generates instructions to adjust the vehicle's operating parameters based on the evaluation results. Specifically, it adjusts the speed or changes the safety distance. The input is comprehensive emotional evaluation data, and the output is instructions to adjust the vehicle's operating parameters.

[1030] Step 6:

[1031] The server generates an instruction to adjust the operating parameters and sends it to the terminal, which then adjusts the vehicle's operating parameters in real time based on the instruction received. Specific actions include reducing the vehicle's speed, maintaining distance, etc. The input is the instruction to adjust the operating parameters, and the output is the adjusted vehicle's operating parameters.

[1032] Step 7:

[1033] The user (passenger) checks the displayed evaluation results and the vehicle's operating status. The terminal displays the analyzed emotional state and the operating parameters based on it to the user. The input is the analyzed emotional state and operating parameters, and the output is visual feedback to the user.

[1034] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1035] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1036] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1037] [Fourth embodiment]

[1038] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1039] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1040] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1041] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1042] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1043] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1044] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1045] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1046] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1047] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1048] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1049] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1050] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1051] The present invention relates to a system that identifies whether book reports and university assignments submitted by students are AI-generated texts, which can be used by faculty and staff to quickly and accurately evaluate the submissions.

[1052] System Configuration

[1053] This system mainly consists of terminals, servers, and users. Terminals are mobile devices such as smartphones and tablets, and servers are computing resources deployed in a cloud environment. Users are primarily faculty and staff who use this system to evaluate student submissions.

[1054] Program processing

[1055] 1. The user captures input data using the smartphone camera.

[1056] The user launches the system's dedicated app and uses the camera function to take a photo of the book report paper.

[1057] The captured image is temporarily stored on the device.

[1058] 2. The device performs OCR processing to extract the text data.

[1059] The device performs OCR (optical character recognition) technology to extract text information from the captured image.

[1060] The extracted character information is stored in the terminal as text data.

[1061] 3. The device formats the text data and sends it to the server

[1062] The extracted text data is formatted on the device, and unnecessary line breaks, spaces, and misrecognized characters are corrected.

[1063] The formatted text data is sent to the server using a secure protocol (e.g., HTTPS).

[1064] 4. The server receives the text data and begins analyzing it.

[1065] The server receives the text data sent from the terminal.

[1066] The received text data is input into a natural language processing model and analysis begins.

[1067] 5. The server evaluates the analysis results and sends them to the device.

[1068] A natural language processing model analyzes the text data and calculates the probability that a sentence is AI-generated.

[1069] The server evaluates the analysis results and summarizes them in a format that is easy for the user to understand, such as "There is an 85% chance that this sentence was generated by an AI."

[1070] The evaluation results are then transmitted to the terminal again using a secure communication protocol.

[1071] 6. The device displays the evaluation results.

[1072] The terminal receives the analysis results sent from the server.

[1073] The received analysis results are displayed to the user, who can use the results to determine whether the student's submission is AI-generated.

[1074] Specific examples

[1075] For example, imagine a scenario in which a faculty member wants to evaluate a student's book report. The user launches a dedicated app and takes a photo of the report with their camera. The system extracts text from the image, formats it, and sends it to the server. Analysis is then performed on the server, and the result is determined to be "80% likely to be an AI-generated sentence." The user can review this result and, if necessary, conduct further research or provide feedback to the student.

[1076] In this way, the system provides faculty with advanced tools to effectively evaluate student work and helps prevent cheating.

[1077] The processing flow will be explained below.

[1078] Step 1:

[1079] The user activates the smartphone camera and takes a picture of the book report paper. The user then controls the camera through the app to capture the image with the appropriate framing.

[1080] Step 2:

[1081] The device saves the captured image and starts OCR processing. The device then analyzes the image and extracts the character information as text data.

[1082] Step 3:

[1083] The device then formats the extracted text data, specifically by removing unnecessary line breaks and spaces from the OCR results and correcting any misrecognized characters.

[1084] Step 4:

[1085] The device sends the formatted text data to the server. The device uses a secure communication protocol such as HTTPS to securely transmit the data to the server.

[1086] Step 5:

[1087] The server receives the text data sent from the device and prepares it for analysis. The server temporarily stores the received data for subsequent analysis.

[1088] Step 6:

[1089] The server inputs the text data into a natural language processing model and begins analysis, specifically to detect patterns and phrases characteristic of AI-generated text.

[1090] Step 7:

[1091] The server evaluates the analysis results and calculates the likelihood that the sentence is AI-generated. Based on the probability calculated by the AI ​​model, it generates an evaluation such as "high probability that the sentence is AI-generated" or "low probability."

[1092] Step 8:

[1093] The server formats the evaluation results and generates results in a format that is easy for users to understand. For example, it generates an evaluation result such as "There is an 80% chance that this sentence was generated by an AI."

[1094] Step 9:

[1095] The server transmits the evaluation results to the terminal, and again using a secure communication protocol, the server transmits the evaluation results to the terminal safely.

[1096] Step 10:

[1097] The terminal displays the evaluation results received from the server, and the terminal visually displays the evaluation results to the user, allowing the user to confirm the results.

[1098] Step 11:

[1099] Users can check the evaluation results and determine whether or not there has been any cheating. Based on the displayed evaluation results, users can conduct further detailed verification and provide feedback to students.

[1100] In this way, a system has been created in which users, devices, and servers work together to quickly and accurately evaluate whether student submissions are AI-generated texts.

[1101] Example 1

[1102] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1103] In recent years, advances in artificial intelligence technology have created a need for a fast and accurate way to assess whether student submissions are AI-generated. However, current systems rely heavily on manual review, which not only takes time and effort but also often lacks accuracy. Therefore, a method is needed for faculty and staff to efficiently assess submissions and prevent cheating.

[1104] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1105] In this invention, the server includes a photographing means of the terminal that acquires input data, an analysis means of the terminal that extracts characters from the image using optical character recognition technology, a communication means that transmits the extracted character data to the information processing device, a receiving means by the information processing device that receives the character data, an analysis means of the information processing device that inputs the character data into a natural language processing model and analyzes it, an evaluation means that calculates the probability that the sentence is an AI-generated sentence based on the analysis result, a result transmission means that transmits the evaluation result to the terminal via the communication means, and a display means on the terminal that displays the evaluation result. This enables faculty and staff to quickly and accurately evaluate whether a submitted work was generated by AI.

[1106] "Input data" is digital information that a user captures using a camera or other device.

[1107] A "terminal" is a device that is directly operated by a user, such as a smartphone, tablet, or PC.

[1108] "Photographing means" refers to the hardware and software, including the camera function, installed in the terminal.

[1109] "Optical character recognition technology" is a technology for reading and extracting text information from images.

[1110] "Analysis means" refers to the function of executing optical character recognition technology on the terminal and extracting character data from an image.

[1111] "Character data" refers to text-based data extracted using optical character recognition technology.

[1112] An "information processing device" is a device that performs data processing, including a server or cloud computing resource.

[1113] "Communication means" refers to an interface and protocol for transmitting and receiving data between a terminal and an information processing device.

[1114] The "receiving means" refers to a function by which the information processing device receives data transmitted from the terminal.

[1115] A "natural language processing model" is an algorithm and software that analyzes text data and determines whether it was generated by artificial intelligence.

[1116] The term "analysis means" refers to a function of an information processing device to input character data into a natural language processing model and analyze it.

[1117] "Evaluation means" refers to a function that calculates the probability that the input text was generated by AI based on the analysis results.

[1118] "Result transmission means" refers to a communication interface and protocol for transmitting evaluation results to a terminal.

[1119] "Display means" refers to an interface that allows the terminal to visually present the evaluation results to the user.

[1120] "Formatting means" refers to a function that formats extracted character data by correcting misrecognitions and deleting unnecessary line breaks and spaces.

[1121] The "evaluation generation means" refers to a function that summarizes and presents the analysis results in a format that is easy for the user to understand.

[1122] This invention relates to a system for identifying whether book reports and university assignments submitted by students have been generated by artificial intelligence. This system allows faculty and staff to quickly and accurately evaluate the submissions. The system mainly consists of a terminal, a server, and a user.

[1123] Hardware and software examples

[1124] Terminal

[1125] The terminal is a device that is directly operated by the user, such as a smartphone, tablet, or PC. The terminal is equipped with a camera, which is used to capture input data such as a book report. The terminal is equipped with optical character recognition (OCR) technology, such as Google Tesseract OCR, to extract text data from images. In addition, the terminal includes a communication means for sending the submitted text data to a server. The HTTPS protocol is used for secure data communication.

[1126] server

[1127] The server is an information processing device that operates in a cloud environment and receives and analyzes text data sent from the terminal. The server has a receiving means and can receive data sent from the terminal. The received data is input into a natural language processing model (for example, "OpenAI's GPT-4") for analysis. The analysis means inputs the text data into this natural language processing model and calculates the probability that the text was generated by AI. The evaluation means evaluates the analysis results and summarizes them in a format that is easy for the user to understand. The evaluation results are then sent back to the terminal using the HTTPS protocol.

[1128] User

[1129] Users are primarily faculty and staff, who use the system to evaluate student submissions. Using a smartphone, tablet, or other device, users launch a dedicated app and take a picture of the submission. After taking the picture, text data extracted from the image is sent to a server, and the analysis results are displayed on the device. This allows users to quickly determine whether the submission was generated by artificial intelligence.

[1130] Examples of concrete examples and prompts

[1131] For example, imagine a scenario in which a faculty member wants to evaluate a student's book report. The user launches a dedicated app and takes a photo of the report with their camera. The system extracts text data from the image, formats it, and sends it to the server. Analysis is then performed on the server, and the result is that "there is an 80% chance that this sentence was generated by AI." The user can review this result and, if necessary, conduct further research or provide feedback to the student.

[1132] Examples of prompts include:

[1133] "Please rate whether the following sentences were generated by an AI:"

[1134] "Calculate the probability that the following text is AI-generated:"

[1135] This provides faculty with advanced tools to effectively assess student work and helps prevent cheating.

[1136] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1137] Step 1:

[1138] The user launches the system's dedicated app and takes a photo of the book report paper using the smartphone camera. The input data is an image of the paper, and the output is a digital image file that is temporarily stored inside the device. When the user taps the "take a photo" button, the camera starts and the image is saved.

[1139] Step 2:

[1140] The device analyzes the captured image using OCR (Optical Character Recognition) technology. The input data is the output image from step 1, and the output is the extracted character data. The OCR engine on the device analyzes the image and generates text data. Specifically, Google's Tesseract OCR library is used. The processed results are saved on the device as a text file.

[1141] Step 3:

[1142] The terminal formats the extracted text data. The input data is the output text from step 2, and the output is the formatted text data. Specifically, unnecessary line breaks and spaces are removed, and misrecognized characters are corrected. The formatted text data is converted to JSON format and is ready to be sent to the server.

[1143] Step 4:

[1144] The terminal sends the formatted text data to the server. The input data is the output text data of step 3, and the output is a data transmission success message to the server. The HTTPS protocol is used as the communication method, and the terminal sends the text data to the server in a secure state.

[1145] Step 5:

[1146] The server receives the text data sent from the terminal. The input data is the text data sent from the terminal, and the output is the received text data. The server's receiving means captures the data and stores it in a specific directory or database.

[1147] Step 6:

[1148] The server inputs the received text data into a natural language processing (NLP) model for analysis. The input data is the received text data from step 5, and the output is the analysis result. Specifically, a generative AI model such as OpenAI's GPT-4 is used to analyze whether the text data is AI-generated. In this process, a natural language processing algorithm is executed.

[1149] Step 7:

[1150] The server calculates the probability that the sentence is AI-generated based on the analysis results. The input data is the analysis result from step 6, and the output is an evaluation result including a probability value. Based on the characteristics of the analyzed text data and the prompt sentence, the server calculates an evaluation such as "There is an 85% chance that this sentence is AI-generated."

[1151] Step 8:

[1152] The server sends the evaluation result to the terminal. The input data is the evaluation result from step 7, and the output is a message to the terminal indicating that the transmission was successful. The evaluation result is formatted in a user-friendly format and sent to the terminal using the HTTPS protocol.

[1153] Step 9:

[1154] The device receives the evaluation results sent from the server and displays them to the user. The input data is the evaluation results sent from the server, and the output is the evaluation results displayed to the user. The device displays the results it receives on the screen, and the user is presented with a message saying, "There is an 85% chance that this sentence is AI-generated." Based on this information, the user can determine whether the submission was AI-generated.

[1155] This allows faculty and staff to quickly and accurately evaluate submissions through a series of processes, preventing cheating before it occurs.

[1156] (Application example 1)

[1157] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1158] With conventional systems, it was difficult to quickly and accurately determine whether student submissions or internal company documents were generated by AI, which caused problems for educational institutions and companies when verifying fraud and the authenticity of content. Furthermore, users could not obtain the results in real time, which led to a lack of efficiency. Furthermore, there were no concrete solutions using wearable devices.

[1159] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1160] In this invention, the server includes a camera means of the terminal that acquires input data, a processing means of the terminal that extracts text from images using optical character recognition (OCR) technology, a communication means that transmits the extracted text data to the server, a receiving means by the server that receives the text data, an analysis means of the server that inputs the text data into a natural language processing model for analysis, an evaluation means that calculates the probability that the text is an AI-generated sentence based on the analysis result, a result transmission means that transmits the evaluation result to the terminal via the communication means, a display means of the terminal that displays the evaluation result, a capture means that captures documents shared on the Internet or in internal documents and identifies whether they were generated by AI, and a wearable device that displays the evaluation result to the user in real time, thereby enabling the user to quickly and accurately obtain a judgment result on the AI-generated sentence.

[1161] The "camera means of the terminal for acquiring input data" refers to a means for capturing target documents or image data using a camera equipped on the terminal.

[1162] "Terminal processing means for extracting text from images using optical character recognition (OCR) techniques" means means for recognizing character information from a captured image and extracting it as text data.

[1163] The "communication means for transmitting extracted text data to a server" is a communication function for transmitting text data from a terminal to a server, for example, a means for using an Internet connection.

[1164] The "receiving means by which the server receives text data" refers to a function or device on the server side that receives text data sent from the terminal.

[1165] The "analysis means of the server that inputs text data into a natural language processing model for analysis" is a function of the server that uses natural language processing technology to analyze the acquired text data.

[1166] The "evaluation means for calculating the probability that a sentence is generated by AI based on the analysis results" is a function that calculates the probability that the sentence was generated by AI based on the analyzed text data.

[1167] The "result transmission means for transmitting the evaluation results to the terminal via the communication means" is a function for transmitting the analysis and evaluation results from the server to the terminal.

[1168] The "display means for the terminal to display the evaluation results" is a function for displaying the evaluation results on the terminal so that the user can visually confirm the analysis results.

[1169] "Capture means for capturing documents shared on the Internet or in internal documents and identifying whether they were generated by AI" refers to a means by which a user can capture documents on the Internet or in internal documents and determine whether their contents were generated by AI.

[1170] A "wearable device that displays evaluation results to a user in real time" is a pair of glasses or other device that can be worn by a user and that displays evaluation results in real time.

[1171] This invention relates to a system that can quickly and accurately identify whether book reports submitted by students, university assignments, internal corporate documents, and documents shared on the Internet have been generated by AI. This system is useful for preventing fraud and verifying the authenticity of content in security departments and educational institutions.

[1172] The basic system configuration is as follows:

[1173] 1. The device (smartphone or wearable device) is equipped with a camera that allows the user to capture the target document.

[1174] 2. The captured image is converted into text data using optical character recognition (OCR) technology such as Google Cloud Vision API or Tesseract OCR.

[1175] 3. The converted text data is formatted within the terminal, and unnecessary parts such as line breaks and spaces are removed.

[1176] 4. The formatted text data is sent to the server using a secure protocol (e.g., HTTPS).

[1177] 5. The server analyzes the text data using a natural language processing model (e.g., Hugging Face's Transformers library or GPT-3).

[1178] 6. The analysis results are evaluated by calculating the probability that the text data was generated by AI.

[1179] 7. The evaluation results are sent from the server to the device and displayed to the user in real time. Particularly when using a wearable device, users can check the evaluation results immediately.

[1180] Examples:

[1181] Example 1: Corporate security department use

[1182] During a meeting, a company security officer captures a printed internal document with the smart glasses' camera. The captured image is converted into text data using the smart glasses' built-in OCR system, and the data is sent to a server via secure communication. The server analyzes the text data and displays the assessment result in real time on the glasses' display: "There is an 85% chance that this document was generated by AI."

[1183] Example 2: Faculty and staff at an educational institution reviewing student submissions

[1184] Faculty and staff use their smartphones to take photos of students' book reports. The captured images are then processed by OCR on the smartphone and converted into text data. This text data is then sent to a server and analyzed using a natural language processing model. The analysis results, such as "There is a 70% chance that this review was generated by AI," are displayed on the smartphone screen.

[1185] Prompt Sentence Examples

[1186] For internal company documents: "Please determine whether the following document was generated by AI:\n\n'Regarding customer data management methods, we would like to propose a new system based on the content of our recent meeting.'"

[1187] For educational student submissions: "Please determine if this student's book report was generated by AI: 'Reading this book helped me to gain a deeper understanding of different cultural backgrounds.'"

[1188] In this way, the system of the present invention enables faculty and security personnel to efficiently and accurately identify AI-generated sentences and quickly provide necessary feedback and countermeasures.

[1189] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1190] Step 1:

[1191] The user captures a document using the device's camera or smart glasses. The input is image data of the document. Specifically, the user launches the device's camera app and takes a picture of the target document. This image data is temporarily stored inside the device.

[1192] Step 2:

[1193] The device performs OCR processing on the captured image data to extract text data. The input is image data and the output is text data. Specifically, the device uses the Google Cloud Vision API and Tesseract OCR engine to recognize and extract character information from the image. This extraction process obtains the document contents as text data.

[1194] Step 3:

[1195] The terminal formats the extracted text data and removes unnecessary characters such as line breaks and spaces. The input is text data obtained by OCR, and the output is formatted text data. Specifically, using regular expressions and string manipulation libraries, unnecessary characters and line breaks are removed from the text, and the formatted text data is generated.

[1196] Step 4:

[1197] The formatted text data is sent from the terminal to the server. The input is the formatted text data, and the output is the state after transmission to the server has been completed. Specifically, the text data is sent to the server using a secure communication protocol (e.g., HTTPS). This procedure allows the text data to be received in a format that can be analyzed on the server side.

[1198] Step 5:

[1199] The server inputs the received text data into a natural language processing model to begin analysis. The input is formatted text data, and the output is the analysis result. Specifically, the text data is analyzed using natural language processing models such as Hugging Face's Transformers library and GPT-3. This analysis calculates the probability that the text was generated by AI.

[1200] Step 6:

[1201] The server calculates the probability that the text is AI-generated based on the analysis results and performs an evaluation. The input is the analyzed text data, and the output is the evaluation result. Specifically, based on the output of the analysis model, an evaluation such as "There is an 85% chance that this document is AI-generated text" is generated.

[1202] Step 7:

[1203] The server sends the evaluation results to the terminal. The input is the evaluation results, and the output is the completion of transmission to the terminal. Specifically, the evaluation results are sent to the terminal again using a secure communication protocol such as HTTPS.

[1204] Step 8:

[1205] The device displays the evaluation results and notifies the user. The input is the evaluation results received from the server, and the output is what is displayed to the user. Specifically, a message such as "There is an 85% chance that this document is AI-generated" is displayed in real time using the device's display or the display function of the smart glasses. This allows the user to quickly check the evaluation results.

[1206] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1207] This invention combines a system that identifies whether book reports and university assignments submitted by students are generated by AI with a function that recognizes user emotions. This system allows faculty and staff to quickly and accurately evaluate submissions, and by understanding the user's emotional state, it provides more flexible and effective feedback.

[1208] System Configuration

[1209] This system mainly consists of terminals, servers, users, and an emotion engine. Terminals are mobile devices such as smartphones and tablets, and the server is a computing resource deployed in a cloud environment. Users are primarily faculty and staff who use this system to evaluate students' submissions. The emotion engine has the ability to analyze users' facial expressions and voices and identify their emotional state.

[1210] Program processing

[1211] 1. The user captures input data using the smartphone camera.

[1212] The user launches the system's dedicated app and uses the camera function to take a photo of the book report paper.

[1213] The captured image is temporarily stored on the device.

[1214] 2. The device performs OCR processing to extract the text data.

[1215] The device performs OCR (optical character recognition) technology to extract text information from the captured image.

[1216] The extracted character information is stored in the terminal as text data.

[1217] 3. The device formats the text data and sends it to the server

[1218] The extracted text data is formatted on the device, and unnecessary line breaks, spaces, and misrecognized characters are corrected.

[1219] The formatted text data is sent to the server using a secure protocol (e.g., HTTPS).

[1220] 4. The server receives the text data and begins analyzing it.

[1221] The server receives the text data sent from the terminal.

[1222] The received text data is input into a natural language processing model and analysis begins.

[1223] 5. The server evaluates the analysis results and sends them to the device.

[1224] A natural language processing model analyzes the text data and calculates the probability that a sentence is AI-generated.

[1225] The server evaluates the analysis results and summarizes them in a format that is easy for the user to understand, such as "There is an 80% chance that this sentence was generated by an AI."

[1226] The evaluation results are then transmitted to the terminal again using a secure communication protocol.

[1227] 6. Emotion engine analyzes user emotions

[1228] The emotion engine installed on the device captures facial expressions and voice data from the user through a facial recognition camera and microphone.

[1229] Based on the acquired data, the user's current emotional state is classified and stored as emotional data in the device.

[1230] 7. The device sends the emotion data to the server

[1231] The device sends emotion data to the server, which is also sent using a secure protocol.

[1232] 8. The server reflects the emotion data in the analysis results

[1233] The server reflects the received emotion data in the analysis results. For example, if the user shows strong emotion (surprise or confusion), it adjusts the reliability of the analysis results.

[1234] As a result, the evaluation results can be corrected based on the user's emotional state.

[1235] 9. The device displays the final evaluation result and emotional feedback.

[1236] The device displays the final evaluation results and emotional feedback to the user, who can check not only the probability of the AI-generated sentences but also their own emotional data.

[1237] For example, it might say, "There is an 80% chance that this sentence was generated by AI. User's emotional state: Confused 45%."

[1238] Specific examples

[1239] For example, imagine a scenario in which a faculty member tries to evaluate a student's book review. The user launches a dedicated app and takes a photo of the review with their camera. The system extracts text from the image, formats it, and sends it to the server. Analysis is then performed on the server, resulting in an evaluation such as "There is an 80% chance that this sentence was generated by AI." When the user checks the evaluation results, the emotion engine analyzes the user's emotions and provides feedback such as "Confused 45%." This allows the user to objectively confirm the reliability of the evaluation results.

[1240] The system not only provides faculty with advanced tools to effectively grade student submissions, helping to prevent cheating, but also allows for more flexible grading through feedback based on sentiment data.

[1241] The processing flow will be explained below.

[1242] Step 1:

[1243] The user activates the smartphone camera and takes a picture of the book report paper. The user then controls the camera through a dedicated app to capture the image with the appropriate framing.

[1244] Step 2:

[1245] The device saves the captured image and starts OCR processing. The device then analyzes the image and extracts the character information as text data.

[1246] Step 3:

[1247] The device then formats the extracted text data, specifically by removing unnecessary line breaks and spaces from the OCR results and correcting any misrecognized characters.

[1248] Step 4:

[1249] The device sends the formatted text data to the server. The device uses a secure communication protocol such as HTTPS to securely transmit the data to the server.

[1250] Step 5:

[1251] The server receives the text data sent from the device and prepares it for analysis. The server temporarily stores the received data for subsequent analysis.

[1252] Step 6:

[1253] The server inputs the text data into a natural language processing model and begins analysis, specifically to detect patterns and phrases characteristic of AI-generated text.

[1254] Step 7:

[1255] The server evaluates the analysis results and calculates the likelihood that the sentence is AI-generated. Based on the probability calculated by the AI ​​model, it generates an evaluation such as "high probability that the sentence is AI-generated" or "low probability."

[1256] Step 8:

[1257] The server formats the evaluation results and generates results in a format that is easy for users to understand. For example, it generates an evaluation result such as "There is an 80% chance that this sentence was generated by an AI."

[1258] Step 9:

[1259] The server transmits the evaluation results to the terminal, and again using a secure communication protocol, the server transmits the evaluation results to the terminal safely.

[1260] Step 10:

[1261] The terminal displays the evaluation results received from the server, and the terminal visually displays the evaluation results to the user, allowing the user to confirm the results.

[1262] Step 11:

[1263] The emotion engine acquires the user's emotion data. The device uses a camera and microphone to collect the user's facial expression and voice data, which the emotion engine analyzes to identify the user's emotional state.

[1264] Step 12:

[1265] The device transmits the acquired emotion data to the server, which also uses a secure protocol.

[1266] Step 13:

[1267] The server receives the emotion data and performs corrections based on the analysis. The server takes into account the user's emotional state identified by the emotion engine as a factor that influences the evaluation results.

[1268] Step 14:

[1269] The server generates the final evaluation result and emotional feedback and sends it to the terminal. The evaluation result takes into account the user's emotional data and is sent to the terminal as the final evaluation result.

[1270] Step 15:

[1271] The device will then display the final evaluation results and emotional feedback to the user, such as "There is an 80% chance that this sentence was generated by AI. User's emotional state: Confused 45%."

[1272] Step 16:

[1273] The user checks the final evaluation results and determines whether there was any cheating. Based on the displayed evaluation results and emotional feedback, the user can conduct further investigations and provide feedback to the students.

[1274] Through these steps, users can not only identify AI-generated sentences but also check the reliability of the evaluation based on their own emotional state, which can effectively prevent cheating in educational settings.

[1275] Example 2

[1276] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1277] Conventional student submission evaluation systems have difficulty identifying whether book reports or university assignments are generated by AI. Furthermore, they do not provide feedback that takes into account the evaluator's emotional state. As a result, it is difficult to provide reliable evaluations and effective feedback. This invention solves these problems, prevents cheating in submissions, and enables flexible feedback based on the evaluator's emotions.

[1278] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for inputting text data into a natural language processing model for analysis, an evaluation means for calculating the probability that the text is an AI-generated sentence based on the analysis result, and an analysis reflection means for reflecting emotion data in the analysis result. This makes it possible to identify with high accuracy whether the submitted work was generated by an AI and provide feedback appropriate to the evaluator's emotions.

[1279] The "camera means of the terminal that acquires input data" refers to a function that allows a user to take a photo of a submission such as a book report or a university assignment using the camera of a smartphone or tablet.

[1280] "Processing means of a terminal that extracts text from images using optical character recognition technology" refers to technology for extracting text information from captured images and the functionality within the terminal that executes this technology.

[1281] The "communication means for transmitting extracted text data to a server" refers to a communication protocol and function for transmitting text data from a terminal to a server.

[1282] The "receiving means for the server to receive text data" is a function for the server to receive text data sent from the terminal.

[1283] The "server's analysis means for inputting text data into a natural language processing model for analysis" refers to the server's function for inputting received text data into a natural language processing algorithm for analysis.

[1284] The "evaluation means for calculating the probability that a sentence is generated by AI based on the analysis results" is a server function that calculates the probability that a sentence was generated by AI based on the analysis results of a natural language processing model.

[1285] The "result transmission means for transmitting the evaluation results to the terminal via the communication means" refers to a communication protocol and function for transmitting the evaluation results obtained on the server side to the terminal.

[1286] The "emotion data collection means for the terminal to acquire the user's facial expressions and voice" is a function that acquires the user's facial expressions and voice using the camera and microphone installed in the terminal and uses them as emotion data.

[1287] "Emotion data transmission means for transmitting emotion data from a terminal to a server" refers to a communication protocol and function for transmitting emotion data collected by a terminal to a server.

[1288] The "analysis reflection means for the server to reflect emotion data in the analysis results" is a function for reflecting emotion data received by the server in the analysis results and adjusting the reliability of the evaluation.

[1289] The "display means by which the terminal displays the evaluation results and emotion feedback" is a function of the terminal for visually displaying the evaluation results and emotion data received from the server to the user.

[1290] The present invention provides a system that identifies whether a book report or university assignment submitted by a student was generated by AI and further reflects the user's emotional state in the evaluation. The system includes a terminal, a server, and an emotion engine. A specific embodiment of the system is described below.

[1291] Device hardware and software

[1292] The terminal is primarily a mobile device such as a smartphone or tablet. This terminal includes a camera means, a processing means, a communication means, an emotional data collection means, and a display means. The camera means has a function that allows the user to take a photo of a book report or assignment. The processing means includes optical character recognition (OCR) software, specifically "Tesseract OCR." The communication means has a function to send and receive data using the HTTPS protocol. The emotional data collection means includes a face recognition camera and microphone, and has a function to capture facial expressions and voice. The display means visually displays the evaluation results and emotional feedback to the user.

[1293] Server Hardware and Software

[1294] The server is located in a cloud environment and has powerful computing resources. The server includes a receiving means, an analyzing means, an evaluating means, an analysis reflecting means, and a result transmitting means. The receiving means receives data transmitted from the terminal and stores it. The analyzing means analyzes the text data using a natural language processing (NLP) model. Specifically, a generative AI model such as "GPT-4" is used. The evaluating means calculates the probability that the sentence is AI-generated based on the analysis results. Furthermore, the analysis reflecting means reflects the user's emotional data in the analysis results and adjusts the reliability of the evaluation. The result transmitting means transmits the evaluation results to the terminal.

[1295] System Operation

[1296] The system works as follows: The user uses the device's camera to take an image of a book report or assignment. The device's OCR software extracts text information from the image and formats this data. The formatted text data is sent to the server using a secure communications protocol. The server receives the text data and inputs it into a natural language processing model for analysis. Based on the analysis results, it calculates the probability that the text is AI-generated and sends this evaluation result to the device.

[1297] Furthermore, the device's emotional data collection means uses the user's facial recognition camera and microphone to capture facial expressions and voice data, generating emotional data. This emotional data is sent to the server, which then reflects it in the analysis results. Finally, the device displays the evaluation results and emotional feedback to the user. For example, it may display, "There is an 80% chance that this sentence is an AI-generated sentence. User's emotional state: Confused 45%."

[1298] Specific examples

[1299] For example, consider the case where a faculty member wants to evaluate a student's book report. The user launches a dedicated app and takes a photo of the report with their camera. The system extracts text from the image, formats it, and sends it to the server. Analysis is performed on the server side, and an evaluation is given, such as "There is an 80% chance that this sentence was generated by AI." Furthermore, when the user checks the evaluation results, the emotion engine analyzes the user's emotions and provides feedback such as "Confused 45%." This allows the user to objectively confirm the reliability of the evaluation results.

[1300] Prompt Sentence Examples

[1301] Here is an example prompt:

[1302] "The following text is a book report written by a student. Please rate whether this text was generated by AI. Also, please analyze the user's emotional state at the time of rating."

[1303] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1304] Step 1:

[1305] The user launches the app and uses the smartphone camera to take a photo of a book report or university assignment. The input is the image captured by the camera, and the output is the image stored in the device's temporary memory. Specifically, the user taps the "take a photo" button in the app, focuses the camera on the object, and takes a photo.

[1306] Step 2:

[1307] The device uses OCR technology to extract text information from a captured image. The input is image data from temporary memory, and the output is the extracted text data. Specifically, the device's OCR engine (e.g., "Tesseract OCR") analyzes the image and converts the text information into text format.

[1308] Step 3:

[1309] The terminal formats the extracted text data. The input is the text data obtained by OCR processing, and the output is the formatted text data. Specifically, a program is executed to correct unnecessary line breaks, spaces, and misrecognized characters.

[1310] Step 4:

[1311] Sends formatted text data to the server. The input is the formatted text data, and the output is the data sent to the server. Specifically, the text data is encrypted and sent using a secure communication protocol (e.g., HTTPS).

[1312] Step 5:

[1313] The server receives the text data and inputs it into a natural language processing model for analysis. The input is the received text data, and the output is the analyzed result. Specifically, the server receives the HTTPS request, saves it in a database, and then inputs the data into an NLP engine (e.g., "GPT-4") to perform analysis.

[1314] Step 6:

[1315] The server calculates the probability that the sentence is AI-generated based on the analysis results. The input is the analysis result of the NLP engine, and the output is a probability value. Specifically, the generative AI model evaluates the analysis result and calculates the probability in the form of "There is an 80% chance that this sentence is AI-generated."

[1316] Step 7:

[1317] The server sends the evaluation results to the terminal. The input is the evaluation results including the probability values, and the output is the data sent to the terminal. Specifically, the evaluation results are sent to the terminal in JSON format using a secure communication protocol.

[1318] Step 8:

[1319] The device's emotion data collection means acquires the user's facial expressions and voice. The input is the user's face and voice, and the output is the acquired emotion data. Specifically, the device's camera and microphone operate simultaneously to capture the facial expressions and voice.

[1320] Step 9:

[1321] The device sends emotion data to the server. The input is the acquired emotion data, and the output is the data sent to the server. Specifically, the emotion data is formatted in JSON format and sent using a secure protocol.

[1322] Step 10:

[1323] The server reflects the emotional data in the analysis results. The input is the received emotional data and the analysis results, and the output is the final evaluation result that reflects the emotions. Specifically, the server analyzes the emotional data and adjusts the reliability of the evaluation results.

[1324] Step 11:

[1325] The device displays the final evaluation result and emotional feedback. The input is the final evaluation result received from the server, and the output is the evaluation result and emotional feedback displayed to the user. Specifically, the device's UI displays "There is an 80% chance that this sentence is an AI-generated sentence. User's emotional state: Confused 45%."

[1326] (Application example 2)

[1327] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1328] Passengers in autonomous vehicles are likely to be in various emotional states, and safer and more comfortable driving is required. However, current autonomous driving systems have difficulty recognizing passengers' emotional states in real time and adjusting vehicle operating parameters accordingly. This can lead to inability to respond appropriately to passengers' anxiety or confusion, which could result in a decrease in safety and comfort.

[1329] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1330] In this invention, the server includes a camera means of the terminal that acquires input data, a processing means of the terminal that extracts text from images using optical character recognition (OCR) technology, an analysis means of the server that inputs the text data into a natural language processing model for analysis, an emotion analysis means of the terminal or server that analyzes the user's emotion using the camera and voice input means, and an operation adjustment means that adjusts the vehicle's operation parameters based on the analysis results of the emotion analysis means. This makes it possible to recognize the emotional state of passengers in real time and appropriately adjust the vehicle's operation parameters based on the emotional state.

[1331] "Input data" refers to data such as images and sounds that are acquired by the system and are the subject of analysis.

[1332] "Camera means of the terminal" refers to a means for acquiring image or video data using a camera mounted on the terminal.

[1333] "Optical character recognition (OCR) technology" is a technology that extracts character information from image data.

[1334] "Terminal processing means" refers to a means that has the function of analyzing and processing data within the terminal.

[1335] "Communication means" refers to a means for transmitting and receiving data between a terminal and a server.

[1336] A "server" is a central computer system that performs data analysis and processing.

[1337] The "receiving means" is a means by which the server receives data transmitted from the terminal.

[1338] A "natural language processing model" is an algorithm or technology for analyzing text data and understanding and generating its content.

[1339] The "analysis means" is a means for analyzing the data received by the server.

[1340] The "evaluation means" is a means for calculating the probability that a text is an AI-generated sentence based on the analysis results.

[1341] The "result transmission means" is a means by which the server transmits the analysis results to the terminal.

[1342] The "display means" is a means for visually presenting the evaluation results received by the terminal to the user.

[1343] The "emotion analysis means" is a means for analyzing the user's emotional state using a camera and a voice input means.

[1344] The "operation adjustment means" is a means for adjusting the vehicle operation parameters based on the analysis results of the emotion analysis means.

[1345] The present invention provides a system for recognizing a user's emotional state in real time and adjusting vehicle operating parameters accordingly. The system includes the following means:

[1346] Hardware and software used

[1347] The system requires a device equipped with a camera and microphone, and a cloud server for analyzing the data. The device includes a camera for capturing video data and a voice input for capturing audio data. The software includes optical character recognition (OCR) technology, natural language processing models, and DeepFace and SpeechRecognition libraries for emotion recognition.

[1348] Data processing and calculation

[1349] The server converts the image data sent from the device into text data using OCR technology. This allows it to analyze the content of the text data and calculate the probability that the text was generated by AI. Meanwhile, the device or server uses a camera and voice input means to capture the user's facial expressions and voice in real time, and analyzes the user's emotional state based on this data.

[1350] Emotion analysis uses the DeepFace library to analyze facial expression data and identify the user's dominant emotion (e.g., anger, sadness, surprise, etc.). For voice data, the SpeechRecognition library is used to first convert the voice to text, and then the text is analyzed for emotion. The results of the emotion analysis are sent from the device to a server, which uses this data to generate instructions for adjusting the vehicle's operating parameters (speed, safety distance, etc.).

[1351] Specific examples

[1352] For example, if a passenger expresses anxiety while using the autonomous taxi application, the system will detect the passenger's emotional state and reduce the vehicle's speed and increase the safety distance in real time. The in-car environment (music, lighting, etc.) will also be adjusted according to the passenger's emotions, allowing passengers to enjoy a safer and more comfortable riding experience.

[1353] Prompt Sentence Examples

[1354] A possible prompt would be:

[1355] Generate a Python program to recognize passenger emotions in real time and adjust the driving parameters of an autonomous vehicle based on the emotions. Use cameras and microphones installed in the vehicle to perform facial and voice emotion analysis using DeepFace and SpeechRecognition libraries. Obtain emotion data in real time and reduce the vehicle's speed and increase the safety distance if the passenger is anxious.

[1356] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1357] Step 1:

[1358] The terminal uses a camera and microphone to capture passenger image and voice data. These data are captured in real time and used for subsequent processing. The input is image and voice data, and the output is a real-time capture of these data.

[1359] Step 2:

[1360] The DeepFace library is used to analyze the image data captured by the device to identify the passenger's emotional state. It analyzes facial patterns and expressions to identify the dominant emotion (e.g., joy, anger, sadness, etc.). The input is the image data, and the output is the identified emotional state.

[1361] Step 3:

[1362] The voice data acquired by the device is converted into text using the SpeechRecognition library, and then the text content is analyzed to infer the emotional state. The emotional nuances of the spoken content are evaluated. The input is voice data, and the output is text data containing the emotional state.

[1363] Step 4:

[1364] The device integrates the emotional states identified in steps 2 and 3 to generate an overall emotional assessment. The device outputs the overall emotional assessment from multiple emotional data. The input is emotional data from images and text, and the output is an overall emotional assessment.

[1365] Step 5:

[1366] The device sends a comprehensive emotional evaluation to the server, which then generates instructions to adjust the vehicle's operating parameters based on the evaluation results. Specifically, it adjusts the speed or changes the safety distance. The input is comprehensive emotional evaluation data, and the output is instructions to adjust the vehicle's operating parameters.

[1367] Step 6:

[1368] The server generates an instruction to adjust the operating parameters and sends it to the terminal, which then adjusts the vehicle's operating parameters in real time based on the instruction received. Specific actions include reducing the vehicle's speed, maintaining distance, etc. The input is the instruction to adjust the operating parameters, and the output is the adjusted vehicle's operating parameters.

[1369] Step 7:

[1370] The user (passenger) checks the displayed evaluation results and the vehicle's operating status. The terminal displays the analyzed emotional state and the operating parameters based on it to the user. The input is the analyzed emotional state and operating parameters, and the output is visual feedback to the user.

[1371] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1372] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1373] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1374] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1375] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1376] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1377] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1378] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1379] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1380] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1381] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1382] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1383] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1384] 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.

[1385] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1386] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1387] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1388] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1389] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1390] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1391] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1392] The following is further disclosed regarding the above embodiment.

[1393] (Claim 1)

[1394] a camera means of the terminal for acquiring input data;

[1395] processing means in the terminal for extracting text from the image using optical character recognition (OCR) techniques;

[1396] a communication means for transmitting the extracted text data to a server;

[1397] a receiving means for receiving text data by the server;

[1398] an analysis means of a server that inputs text data into a natural language processing model for analysis;

[1399] An evaluation means for calculating the probability that the sentence is an AI-generated sentence based on the analysis result;

[1400] a result transmission means for transmitting the evaluation result to the terminal via a communication means;

[1401] Display means for displaying evaluation results on the terminal

[1402] A system including:

[1403] (Claim 2)

[1404] the terminal further includes a formatting means for formatting the extracted text data;

[1405] 10. The system of claim 1.

[1406] (Claim 3)

[1407] The server further includes a rating generation means for organizing the analysis results and generating ratings in a format that is easy for the user to understand.

[1408] 10. The system of claim 1.

[1409] "Example 1"

[1410] (Claim 1)

[1411] A photographing means of the terminal for acquiring input data;

[1412] an analyzing means for the terminal that extracts characters from the image using optical character recognition technology;

[1413] a communication means for transmitting the extracted character data to the information processing device;

[1414] a receiving means for receiving character data by the information processing device;

[1415] an analysis means of an information processing device that inputs character data into a natural language processing model for analysis;

[1416] evaluation means for calculating the probability that the sentence is an AI-generated sentence based on the analysis result;

[1417] a result transmission means for transmitting the evaluation result to the terminal via a communication means;

[1418] Display means for displaying evaluation results on the terminal

[1419] A system including:

[1420] (Claim 2)

[1421] The terminal further includes a formatting means for formatting the extracted character data.

[1422] 10. The system of claim 1.

[1423] (Claim 3)

[1424] The information processing device further includes an evaluation generation means for organizing the analysis results and generating an evaluation in a format that is easy for a user to understand.

[1425] 10. The system of claim 1.

[1426] "Application Example 1"

[1427] (Claim 1)

[1428] a camera means of the terminal for acquiring input data;

[1429] processing means in the terminal for extracting text from the image using optical character recognition (OCR) techniques;

[1430] a communication means for transmitting the extracted text data to a server;

[1431] a receiving means for receiving text data by the server;

[1432] an analysis means of a server that inputs text data into a natural language processing model for analysis;

[1433] An evaluation means for calculating the probability that the sentence is an AI-generated sentence based on the analysis result;

[1434] a result transmission means for transmitting the evaluation result to the terminal via a communication means;

[1435] a display means for displaying the evaluation results on the terminal;

[1436] A means of capturing documents shared on the internet or internally to identify whether they are generated by AI;

[1437] a wearable device that displays the assessment results to the user in real time;

[1438] A system including:

[1439] (Claim 2)

[1440] the terminal further includes a formatting means for formatting the extracted text data;

[1441] 10. The system of claim 1.

[1442] (Claim 3)

[1443] The server further includes a rating generation means for organizing the analysis results and generating ratings in a format that is easy for the user to understand.

[1444] 10. The system of claim 1.

[1445] "Example 2: Combining Emotion Engines"

[1446] (Claim 1)

[1447] a camera means of the terminal for acquiring input data;

[1448] processing means at the terminal for extracting text from the image using optical character recognition techniques;

[1449] a communication means for transmitting the extracted text data to a server;

[1450] a receiving means for receiving text data by the server;

[1451] an analysis means of a server that inputs text data into a natural language processing model for analysis;

[1452] An evaluation means for calculating the probability that the sentence is an AI-generated sentence based on the analysis result;

[1453] a result transmission means for transmitting the evaluation result to the terminal via a communication means;

[1454] An emotion data collection means for the terminal to acquire facial expressions and voice of the user;

[1455] emotion data transmission means for transmitting emotion data from the terminal to the server;

[1456] an analysis reflection means for the server to reflect emotion data in the analysis result;

[1457] A display means for the device to display the evaluation results and emotional feedback

[1458] A system including:

[1459] (Claim 2)

[1460] the terminal further includes a formatting means for formatting the extracted text data;

[1461] 10. The system of claim 1.

[1462] (Claim 3)

[1463] The server further includes a rating generation means for organizing the analysis results and generating ratings in a format that is easy for the user to understand.

[1464] 10. The system of claim 1.

[1465] "Application example 2 when combining emotion engines"

[1466] (Claim 1)

[1467] a camera means of the terminal for acquiring input data;

[1468] processing means in the terminal for extracting text from the image using optical character recognition (OCR) techniques;

[1469] a communication means for transmitting the extracted text data to a server;

[1470] a receiving means for receiving text data by the server;

[1471] an analysis means of a server that inputs text data into a natural language processing model for analysis;

[1472] An evaluation means for calculating the probability that the sentence is an AI-generated sentence based on the analysis result;

[1473] a result transmission means for transmitting the evaluation result to the terminal via a communication means;

[1474] a display means for displaying the evaluation results on the terminal;

[1475] An emotion analysis means for analyzing the emotion of a user using a camera and a voice input means in the terminal or the server;

[1476] an operation adjustment means for adjusting the operation parameters of the vehicle based on the analysis results of the emotion analysis means;

[1477] A system including:

[1478] (Claim 2)

[1479] 10. The system of claim 1, wherein the terminal further comprises formatting means for formatting the extracted text data.

[1480] (Claim 3)

[1481] 10. The system of claim 1, further comprising a rating generation means, wherein the server organizes the analysis results and generates a rating in a user-understandable format. [Explanation of symbols]

[1482] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a camera means of the terminal for acquiring input data; processing means at the terminal for extracting text from the image using optical character recognition techniques; a communication means for transmitting the extracted text data to a server; a receiving means for receiving text data by the server; an analysis means of a server that inputs text data into a natural language processing model for analysis; An evaluation means for calculating the probability that the sentence is an AI-generated sentence based on the analysis result; a result transmission means for transmitting the evaluation result to the terminal via a communication means; Display means for displaying evaluation results on the terminal A system including:

2. the terminal further includes a formatting means for formatting the extracted text data; The system of claim 1 .

3. The server further includes a rating generation means for organizing the analysis results and generating ratings in a format that is easy for the user to understand. The system of claim 1 .

Images