system

Abstract

We provide the system. [Solution] Means of receiving materials, A means for extracting text from received materials and performing preprocessing for analysis, A means for analyzing the style and terminology of preprocessed text and evaluating its potential generation, A means for comparing text with other existing data and calculating similarity, A means for outputting the generated probability and similarity evaluation results, A system that includes this.

Description

Technical Field

[0005] ,

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

Background Art

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

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In recent years, with the evolution of generative AI technology, the cases where documents such as resumes and reports are generated by AI have been increasing. Therefore, in enterprises and educational institutions, it is required to accurately determine whether the submitted documents are those by conventional manual work or by AI. Also, it is necessary to check whether the document content is similar to other existing data to reduce the risk of copyright infringement. However, the current tools have a problem that it is difficult to determine these efficiently and accurately.

Means for Solving the Problems

[0005] This invention provides means for receiving data, extracting text from that data, and performing preprocessing. Furthermore, it provides means for analyzing the preprocessed text and evaluating the likelihood of it being generated based on its style and terminology selection. In addition, it includes means for comparing the text with other existing data and calculating similarity to evaluate the likelihood of AI generation and the risk of copyright infringement. These evaluations are then output in report format and provided to the user, enabling companies and educational institutions to make accurate decisions.

[0006] "Documents" refer to documents or electronic files that contain information or data.

[0007] "Means of receiving" refers to functions for importing data into the system from external sources.

[0008] "Extracting text" refers to the process of taking textual information from a document and converting it into a format that can be analyzed.

[0009] "Preprocessing" is the process of standardizing text for analysis and removing unnecessary parts.

[0010] "Means of analysis" refers to functions for detecting and evaluating specific patterns and features within a document.

[0011] "Style and vocabulary selection" refers to the manner of language and word choices used within a text, and indicates characteristics related to style.

[0012] "Assessing the possibility of generation" means determining whether a particular document was generated using artificial intelligence technology.

[0013] "Calculating similarity" is the process of evaluating and quantifying how similar a document is to existing data.

[0014] "Scale and copyright infringement risk" refers to the potential legal issues that may arise from a document being similar to other existing works.

[0015] The "means for outputting" is a function that provides the evaluation result in a visual or digital format.

[0016] "Output in report format" means summarizing and organizing the analysis results in an understandable manner and finally documenting them.

[0017] "Providing to the user" refers to the act of notifying or distributing the processing result to the person using the system.

Brief Description of Drawings

[0018] [Figure 1] It is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] It is a conceptual diagram showing an example of the main functions of a data processing device and a smart device according to the first embodiment. [Figure 3] It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] It is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12]It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Embodiment 2 when combined with an emotion engine. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when combined with an emotion engine.

Mode for Carrying Out the Invention

[0019] Hereinafter, an example of an embodiment of the system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

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

[0021] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be one arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be one type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0022] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0023] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0024] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0025] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0026] [First Embodiment]

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

[0028] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0029] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0030] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0031] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0032] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0033] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

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

[0035] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0036] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0037] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0038] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0039] This invention relates to a system for appropriately evaluating the authenticity and copyright infringement risk of materials submitted by users. This system is primarily intended for use in businesses and educational institutions.

[0040] First, the user uploads the document to be evaluated to the system. The server receives this document and performs text extraction. This extracts the textual information from the document in a parseable format. The server then performs text preprocessing to standardize the content. This includes removing unnecessary line breaks and special characters.

[0041] Next, the server passes the pre-processed text to an AI agent for analysis. The AI ​​agent analyzes the style and terminology in the document in detail and assesses the likelihood that the document was created by a generative AI. To perform this assessment, the AI ​​agent uses statistical models and machine learning algorithms.

[0042] Furthermore, the server compares the text within the document with existing content in the database and calculates the similarity. This step allows it to determine how similar the document is to other documents. The risk of copyright infringement is also assessed during this process.

[0043] The terminal receives evaluation results from the server and provides them to the user. These results are presented in a detailed report format, which the user can use to decide whether to resubmit or to review the content.

[0044] As a concrete example, consider a research report submitted by an undergraduate student at a university. When a student uploads a report, the server receives it and uses an AI agent to calculate the AI ​​generation rate. Furthermore, it evaluates how similar the report is to other research literature and reports on the possibility of copyright infringement. Based on these results, the university can notify the student of the appropriateness of the report and areas for improvement. In this way, the reliability of submitted materials can be improved by utilizing the system of the present invention.

[0045] The following describes the processing flow.

[0046] Step 1:

[0047] The user uploads the documents to be evaluated to the system. The server saves the received document files to secure storage and prepares them for analysis.

[0048] Step 2:

[0049] The server extracts text from the document. This process uses analysis tools to extract text information from various formats, such as PDFs and Word documents.

[0050] Step 3:

[0051] The server performs preprocessing on the extracted text. This includes removing line breaks and extra spaces, and standardizing special characters. This prepares the text for parsing.

[0052] Step 4:

[0053] The server passes the pre-processed text to a specialized AI agent. The AI ​​agent begins analyzing the writing style and terminology, detecting features that indicate potential for generation by a generative AI. This analysis utilizes natural language processing techniques and machine learning algorithms.

[0054] Step 5:

[0055] The server compares the text in the document to existing documents in the database. During this process, a document similarity algorithm is used to calculate a score indicating how similar the document is to other documents.

[0056] Step 6:

[0057] The server generates a comprehensive evaluation report using the generation rate of the AI ​​and the similarity score. This report includes an assessment of the likelihood that the content was created by the AI, as well as a copyright infringement risk assessment.

[0058] Step 7:

[0059] The terminal provides the user with an evaluation report from the server. Based on this report, the user can consider whether they need to resubmit the materials or revise their content.

[0060] (Example 1)

[0061] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0062] In today's information environment, a vast amount of data is produced daily, but there is a lack of readily available means to determine whether this data is original or generated by AI. Furthermore, there is a need to reliably assess the potential for copyright infringement. This is especially crucial for educational institutions and businesses, where the reliability of submitted materials must be assessed quickly and accurately.

[0063] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0064] In this invention, the server includes means for receiving data, means for extracting and pre-processing textual information, means for analyzing writing style and vocabulary selection and evaluating generative feasibility, means for comparing with other existing data and calculating similarity, and means for reporting the results of the reliability evaluation. This makes it possible to quickly and accurately evaluate the generative feasibility of the data generation AI and the risk of copyright infringement.

[0065] "Documents" refer to documents or files that contain information or data.

[0066] "Textual information" refers to information composed of characters and symbols contained in a document.

[0067] "Analysis" refers to the process of thoroughly examining textual information to understand its structure and meaning.

[0068] "Preprocessing" is the process of removing unnecessary information from textual data and standardizing it to make it easier to analyze.

[0069] "Writing style" refers to the characteristics related to the way a text is expressed and its style.

[0070] "Vocabulary selection" refers to the characteristics related to the choice of words and phrases used in a text.

[0071] "Generability" is an assessment indicating the likelihood that the subject material was created by a generative AI.

[0072] "Existing data" refers to information and data that have been stored in the past and is used as a basis for comparison.

[0073] "Similarity" is a numerical representation of how similar the content of a document is to other existing data.

[0074] "Reliability assessment" refers to the process of determining and reporting whether a document is original or potentially infringes on copyright.

[0075] This invention provides a specific system for evaluating the reliability of data. It describes how hardware and software are used to process and compute the data.

[0076] Users upload the documents to be evaluated to the system using their own devices. The documents are assumed to be in common document formats such as PDF or Word. The server receives the documents and extracts the text information. Various character recognition software, including OCR technology, is used for text extraction.

[0077] Next, the server preprocesses the extracted text information. This preprocessing removes unnecessary line breaks and special characters from the text and standardizes the content. This process is characterized by string manipulation using regular expressions and automated filtering using scripts.

[0078] The processed text information is sent to an AI agent on the server. The AI ​​agent uses natural language processing techniques to analyze the writing style and vocabulary selection of the document. This analysis assesses the likelihood that the document was created by a generative AI model. The AI ​​agent utilizes statistical models and machine learning algorithms to perform a highly accurate assessment.

[0079] Furthermore, the server compares the textual information of the document with existing databases. Vector space models and data mining techniques are used in the process of calculating similarity. This quantifies the degree to which the document is similar to existing literature and data, and assesses the likelihood of copyright infringement.

[0080] Ultimately, the terminal receives the evaluation results from the server and provides the user with information in the form of a detailed report. Based on this report, the user can check the reliability and areas for improvement of the material and revise it as needed.

[0081] As a concrete example, consider a scenario where an undergraduate student submits a research report. When the student uploads the report to the system, the server uses OCR technology to transcribe it into text and analyzes whether it was likely created by a generative AI. It also evaluates the similarity by comparing it to previously submitted reports and publicly available academic literature. This allows the university to verify the originality and reliability of the report and provide feedback to the student.

[0082] An example of a prompt is: "Evaluate whether the following research report is similar to other documents and whether it was likely created by a generative AI."

[0083] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0084] Step 1:

[0085] Users upload documents to be evaluated to the system using their terminals. These documents are in common formats such as PDF and Word. The input is the document file submitted by the user. The output is the server receiving and saving the relevant documents. The server stores the user's documents for future analysis.

[0086] Step 2:

[0087] The server extracts text information from the uploaded documents. This process utilizes OCR technology, converting image formats into text. The input is the document file received by the server. The output provides analyzable text information. This extraction allows the text within the document to be treated as digital data.

[0088] Step 3:

[0089] The server preprocesses the extracted character information. It removes unnecessary line breaks and special characters and standardizes the text. The input is the initial extracted character information. The output is clean, standardized text data. This processing improves the accuracy of subsequent analysis.

[0090] Step 4:

[0091] The server passes pre-processed text to the AI ​​agent, which evaluates the likelihood that the document was created by a generative AI model. The input is standardized text data. The output provides an evaluation of the likelihood that the document was created by a generative AI. The AI ​​agent uses statistical models and machine learning algorithms to make highly accurate judgments.

[0092] Step 5:

[0093] The server compares the textual information of the document with existing content stored in the database and calculates the similarity. The input is standardized text data. The output provides the similarity evaluation result. This process determines how similar the document is to other documents.

[0094] Step 6:

[0095] The terminal receives evaluation results from the server and provides them to the user in the form of a detailed report. The input is evaluation result data sent from the server. The output is a detailed reliability evaluation report presented to the user. The user can use this report to improve or resubmit their materials.

[0096] (Application Example 1)

[0097] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0098] Modern educational institutions are required to quickly and accurately assess the reliability of student reports and papers. However, a related problem is the difficulty in effectively detecting similarities between artificially generated texts and existing literature. Such challenges hinder the maintenance of educational quality.

[0099] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0100] In this invention, the server includes means for receiving data information, means for extracting textual information from the received data information and performing preprocessing for analysis, means for analyzing the style and terminology selection of the preprocessed textual information and evaluating the generated ratio, means for comparing the textual information with other existing data sets and calculating similarity, means for outputting the generated ratio and the evaluation result of the similarity, and means for notifying the recipient of the evaluation result in data format. This enables a rapid and accurate evaluation of the reliability of submitted data.

[0101] "Document information" refers to data that shows the contents of submitted documents and papers.

[0102] "Textual information" refers to text data extracted from source material, and it forms the basis for analysis.

[0103] "Preprocessing" refers to the process of removing unnecessary line breaks and special characters from received text information in order to analyze it.

[0104] "Style and terminology selection" refers to the unique writing style and vocabulary used in a particular text.

[0105] "Generated rate" is a numerical value that indicates the degree to which the source information was likely created by artificial intelligence.

[0106] "Similarity" is a measure used to evaluate how similar textual information is to an existing data set.

[0107] "Evaluation results" refer to the overall analysis results output based on the analysis of the generated proportions and similarity.

[0108] "Data format" refers to the reporting format in which evaluation results are provided to recipients.

[0109] "Recipient" refers to the individual or organization that receives the evaluation results.

[0110] This system is designed to evaluate the reliability of reports submitted by educational institutions. The server receives document information using a cloud platform. The received document information is converted into the necessary textual information by an internal character information extraction module. In this process, SpaCy, a natural language processing tool, is used to remove unnecessary symbols and line breaks, creating a parseable data format.

[0111] Next, the server analyzes the pre-processed text information using machine learning algorithms, specifically TENSORFLOW® or PyTorch, to determine the likelihood that the material is artificially generated. Furthermore, the text information is compared with existing data sets to calculate similarity. A text comparison algorithm is used for this matching to accurately assess how similar the material is to other documents.

[0112] The evaluation results are provided to the user in the form of a detailed report. The user's device can visually review this report, allowing them to identify areas for improvement and assess the reliability of the materials. This application primarily operates on smartphones and connects with a cloud server to provide real-time evaluation results.

[0113] As a concrete example, suppose a student uploads an environmental studies report. The server then evaluates the received report, assessing the likelihood that its content was generated by a generative AI model (15%) and determining its similarity to existing academic literature (10%). Such evaluations serve as important indicators for educational institutions to provide appropriate feedback to students.

[0114] An example of a prompt might be: "Check if this report is similar to other literature and assess whether it was created by a generative AI."

[0115] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0116] Step 1:

[0117] The user uploads document information to the server through the application. The input data is document information obtained from the user's local device and sent to the server. The output is the received document recorded on the server side.

[0118] Step 2:

[0119] The server extracts textual information from the received document information. During this process, SpaCy, a natural language processing tool, is used to remove unnecessary symbols and line breaks, generating parseable text data. The input is document information, and the output is pre-processed textual information.

[0120] Step 3:

[0121] The server applies machine learning algorithms to analyze pre-processed text information. Here, TensorFlow or PyTorch is used to analyze writing style and terminology selection, and the likelihood that the material was generated by a generative AI model is calculated. The input is pre-processed text information, and the output is a numerical evaluation of the generative likelihood.

[0122] Step 4:

[0123] The server compares the textual information with an existing data set. A text comparison algorithm is used to calculate similarity, evaluating how similar the document is to other documents. The input is pre-processed textual information, and the output is a similarity score.

[0124] Step 5:

[0125] The server integrates the results of the generative feasibility and similarity evaluations and sends them to the user terminal in the form of a detailed report. The input is the evaluation value, and the output is the integrated report data.

[0126] Step 6:

[0127] The user terminal visually displays the received report and provides the user with areas for improvement and reliability indicators for the document. The input is integrated report data, and the output is visual feedback provided to the user.

[0128] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0129] This invention relates to a system that combines a function to evaluate the likelihood and similarity of materials submitted by users using a generation AI with an emotion engine. This system is primarily intended for use in businesses and educational institutions, and aims to provide more appropriate feedback by taking user emotions into consideration.

[0130] First, the user uploads the document to be evaluated to the system. The server receives this document and extracts text from the file. The server then preprocesses the extracted text, standardizing its content and preparing it for analysis.

[0131] Next, the server passes the pre-processed text to the AI ​​agent, which begins analyzing the writing style and terminology. The AI ​​agent detects and evaluates features that indicate the potential of generative AI. Statistical models and machine learning algorithms are used in this process.

[0132] Furthermore, the server compares the text to existing content in the database and calculates a similarity score. This evaluates how similar the text is to other documents and analyzes the risk of copyright infringement.

[0133] In this system, the emotion engine plays a crucial role. The terminal analyzes user interaction data and recognizes the user's emotional state from it. For example, it can determine emotions from the user's input speed and word choice when operating the system.

[0134] Based on the perceived emotions, the server adjusts the content and presentation format of the evaluation results. For example, if the system determines that the user is disappointed, it can present the evaluation results in a way that includes more thoughtful support.

[0135] A concrete example of this system is a job seeker. When a job seeker uploads their resume, the system evaluates the potential of AI generation and further uses an emotion engine to understand the job seeker's feelings. If the job seeker shows signs of nervousness, the device provides the evaluation results along with an encouraging message, making it easier for the user to understand areas for improvement.

[0136] By combining emotion engines in this way, it is possible to provide services that are more tailored to individual users and improve the effectiveness of the system.

[0137] The following describes the processing flow.

[0138] Step 1:

[0139] Users upload the document files they wish to have evaluated to the system. The server receives these documents and stores them securely. This process also includes verifying that the files are in the correct format.

[0140] Step 2:

[0141] The server extracts text data from the documents. Even if the documents are in PDF, Word, or other formats, it uses appropriate tools to extract the text and prepare it for analysis.

[0142] Step 3:

[0143] The server performs preprocessing on the extracted text. This includes removing unnecessary line breaks and spaces, and standardizing special characters. This makes the text more suitable for analysis.

[0144] Step 4:

[0145] The server sends pre-processed text to the AI ​​agent. The AI ​​agent analyzes the writing style and terminology selection and evaluates the feasibility of generating the text using generative AI. Statistical models and machine learning algorithms are used in this process.

[0146] Step 5:

[0147] In parallel, the server compares the document text with existing data in the database and calculates similarity. It then calculates a similarity score and analyzes the risk of copyright infringement.

[0148] Step 6:

[0149] The device uses user interaction data to activate an emotion engine. The emotion engine analyzes data such as the user's input speed and word choice to recognize the user's emotional state.

[0150] Step 7:

[0151] The server determines how to present the evaluation results based on the user's emotions recognized by the emotion engine. For example, if the user appears anxious, the results will be presented in a calm tone, including advice.

[0152] Step 8:

[0153] The device provides the user with a report containing the adjusted evaluation results. Based on this report, the user can consider areas for improvement in their materials and the possibility of resubmission.

[0154] (Example 2)

[0155] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0156] Conventional document evaluation systems could assess the likelihood of document generation and similarity using AI, but they lacked the means to provide feedback that took user emotions into consideration. This resulted in problems such as evaluation results causing psychological burden on users and preventing them from taking appropriate corrective actions.

[0157] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0158] In this invention, the server includes means for receiving data, means for extracting and pre-processing text, means for analyzing the style and terminology of the text and evaluating its generative potential, means for comparing the text with existing data and calculating similarity, and means for analyzing emotions from user operation data and adjusting the presentation format of the evaluation results. This enables feedback tailored to the user's emotional state, reduces psychological burden, and promotes appropriate improvement actions.

[0159] "Documents" refers to the documents uploaded to the system for evaluation. This typically includes content provided as PDF or Word files.

[0160] "Text" refers to string data extracted from the document. This allows the content to be analyzed mechanically.

[0161] "Preprocessing" refers to the process of standardizing and formatting the extracted text in preparation for analysis. This includes removing unnecessary characters and unifying the format.

[0162] "Analysis" is a general term for technical operations performed to evaluate the content of text. This includes the detection of style and terminology, and the evaluation of generative potential.

[0163] "Generability" refers to characteristics that indicate the likelihood that the material was generated by a generative AI. This includes evaluation results based on stylistic and terminology patterns.

[0164] "Similarity" is a measure that indicates the degree to which text shares commonalities with existing data. This allows for an objective evaluation of similarity with other materials.

[0165] "Emotion" refers to the psychological state inferred from the user's operation data. This includes evaluation results based on input speed and operation history.

[0166] "Feedback" refers to the information that the system provides to the user based on its analysis results. This information is adjusted according to the user's emotional state.

[0167] This invention realizes a system that analyzes the material to be evaluated, assesses the possibility and similarity of generation by the generation AI, and provides feedback while taking into account the user's emotions. It is primarily intended for use in educational institutions and companies, and aims to analyze materials and provide psychological support to users.

[0168] Users first upload the documents to be evaluated to the system. This operation is performed via a web interface on their terminal, and the data is provided in PDF or Word file format. The server receives the documents and extracts the text data. This utilizes PDF parsing and OCR (optical character recognition) technology.

[0169] Next, the server performs preprocessing on the extracted text. This preprocessing removes unnecessary spaces and special characters, and formats the text, preparing it for analysis. The server then passes the preprocessed text to an AI agent, which uses a generative AI model to analyze the writing style and terminology selection. This AI agent combines natural language processing and machine learning algorithms to evaluate the potential for generation.

[0170] The server also compares the text against existing data in the database to assess its similarity. Numerical methods, such as cosine similarity, are used to calculate a similarity score. This allows for an assessment of the text's similarities to other materials and the associated copyright risks.

[0171] A key feature of this system is that the emotion engine analyzes the user's operation data to infer their psychological state. The terminal collects data such as input speed and word choice to identify the emotional state. Based on this recognition, the server adjusts the format in which the evaluation results are presented. For example, if the user is feeling anxious, the feedback can be provided in an encouraging way.

[0172] As a concrete example, consider a case where a job seeker uploads their resume to the system and the AI ​​evaluates its potential for generation. In this case, if the emotion engine detects the job seeker's anxiety, the terminal can provide feedback such as, "This evaluation result is intended to show areas for improvement. Let's think positively about the next steps."

[0173] A concrete example of a prompt is the instruction, "Evaluate the AI-generated potential of the uploaded document and adjust the feedback according to the user's sentiment." By using this prompt, the system can achieve multi-layered evaluation and feedback.

[0174] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0175] Step 1:

[0176] Users upload documents to the system. PDF and Word files are used as input. The terminal checks the file format and prepares to send it to the server.

[0177] Step 2:

[0178] The server extracts text from the received documents. The input is the uploaded file, and the output is the extracted raw text data. This process utilizes OCR technology and PDF parser libraries.

[0179] Step 3:

[0180] The server preprocesses the extracted text. The input is raw text data, and the output is formatted and standardized text. Unnecessary spaces and special characters are removed, and morphological analysis is performed to prepare the text for analysis.

[0181] Step 4:

[0182] The server passes pre-processed text to the AI ​​agent for evaluation of its generative potential. The input is formatted text, and the output is a generative potential evaluation score. The AI ​​agent analyzes the writing style and terminology selection using natural language processing models and machine learning algorithms.

[0183] Step 5:

[0184] The server calculates similarity by comparing the text with existing data. The input is pre-processed text, and the output is a similarity score. Vector comparison methods such as cosine similarity are used in this process to evaluate the similarity with other materials.

[0185] Step 6:

[0186] The device analyzes user interaction data to infer emotions. Inputs include user input speed and interaction history data, while output is an inferred emotion label. The device sends this data to an emotion engine for analysis of the user's psychological state.

[0187] Step 7:

[0188] The server adjusts the presentation format of the evaluation results based on the emotion label. The input is the evaluation results of generativeness and similarity, and the emotion label, and the output is an adjusted feedback message. For example, if the server determines that the user is stressed, the feedback will be provided in gentler language.

[0189] Step 8:

[0190] The device presents the user with refined feedback. The input is a refined feedback message sent from the server, and the output is the evaluation result and support message displayed on the user's screen. Information is provided to help the user understand areas for improvement and move forward to the next step.

[0191] (Application Example 2)

[0192] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0193] The problem that this invention aims to solve is how to provide feedback that takes into account the user's emotional state, rather than simply checking the possibility and similarity of materials submitted by the user using a generating AI. In particular, current systems are unable to reflect the user's psychological state, resulting in uniform evaluation results and a lack of individualized support.

[0194] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0195] In this invention, the server includes means for receiving data, means for extracting text from the received data and performing preprocessing for analysis, means for analyzing the style and terminology of the preprocessed text and evaluating the likelihood of generation, means for comparing the text with other existing data and calculating similarity, means for outputting the evaluation results of the likelihood and similarity of generation, and means for recognizing the user's emotional state and adjusting the evaluation results accordingly. This enables detailed feedback that responds to the user's emotional state.

[0196] "Documents" refer to the documents and data that users submit to the system for evaluation.

[0197] "Means of receiving" refers to the function for importing data from users into the server.

[0198] "Means for extraction and pre-processing for analysis" refers to a function that extracts necessary text data from a document and prepares it in a format suitable for analysis.

[0199] "Means for evaluating the likelihood of generation" refers to a function that analyzes whether preprocessed text was generated by AI and quantifies that likelihood.

[0200] "Methods for comparing with existing data and calculating similarity" refers to a function that compares the input text with existing data in the database and expresses the degree of similarity as a numerical value.

[0201] "Means for outputting evaluation results" refers to a function that provides the generated probability and similarity analysis results in a format that users can review.

[0202] "Means for recognizing emotional states and adjusting evaluation results" refers to a function that analyzes user interaction data to determine emotions and changes the content and method of feedback accordingly.

[0203] The system implementing this invention receives data from the user and provides feedback that takes into account the user's emotional state based on that data. The server first receives data submitted electronically by the user. The data is taken into the server as text data, and its contents are pre-processed for analysis. In the pre-processing, the text is standardized and prepared for evaluation.

[0204] Next, the server analyzes the pre-processed text and evaluates its potential for generation by the AI ​​based on its writing style and terminology. Statistical models and machine learning algorithms are used for this evaluation. The text is also compared with existing data in the database, and a similarity score is calculated.

[0205] One of the features of this invention is that the server recognizes the user's emotional state and reflects it in the evaluation results. The terminal analyzes the user's interaction data and estimates the emotional state from input speed and word choice. For example, if the user is confused, the server carefully explains the evaluation results and clearly communicates areas for improvement. As a result, the user can receive feedback tailored to their emotional state.

[0206] One concrete application example is when children studying at home use a device designed to support home learning. When a child submits an essay assignment, the device evaluates it using a generative AI model and also adds encouraging words, taking into account the child's emotional state. In this way, the user experience can be further improved.

[0207] An example of a prompt message is: "Based on the following evaluation results, please tell me what encouraging comments I should add if the user is confused: The evaluation results indicate a 50% probability of being AI-generated and a similarity score of 20%."

[0208] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0209] Step 1:

[0210] The server receives materials submitted electronically by users. The input is document data uploaded by the user, which the server receives and stores. The output is the material to be processed, which is stored within the server.

[0211] Step 2:

[0212] The server extracts text from received materials and performs preprocessing for analysis. The input is document data stored on the server, and the server extracts text data from the documents using a text extraction algorithm. Next, it performs normalization and removes unnecessary information, and outputs text ready for analysis.

[0213] Step 3:

[0214] The server analyzes pre-processed text using a generative AI model and evaluates the likelihood of it being generated. The input is pre-processed text data, and the server analyzes the style and terminology selection of this data using statistical models and machine learning algorithms. The output is a numerical value indicating the likelihood that the text was generated by a generative AI.

[0215] Step 4:

[0216] The server compares text with data in an existing database and calculates similarity. The input is pre-processed text data, and the server calculates similarity by comparing it with similar data in the database. The output is the similarity score obtained from the comparison.

[0217] Step 5:

[0218] The device analyzes user interaction data in real time to recognize emotional states. Input data includes user input speed and word choice, and the device analyzes this data to estimate the emotional state and output the emotion recognition result.

[0219] Step 6:

[0220] The server combines the generated probabilities, similarity scores, and the user's emotional state to refine and output feedback. The inputs are the generated probability values, similarity scores, and emotional recognition results, which the server integrates to generate feedback tailored to the user. The output is presented to the user as a refined evaluation result.

[0221] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0222] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0223] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0224] [Second Embodiment]

[0225] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0226] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0227] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0228] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0229] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0230] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0231] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0232] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0233] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0234] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0235] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0236] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0237] This invention relates to a system for appropriately evaluating the authenticity and copyright infringement risk of materials submitted by users. This system is primarily intended for use in businesses and educational institutions.

[0238] First, the user uploads the document to be evaluated to the system. The server receives this document and performs text extraction. This extracts the textual information from the document in a parseable format. The server then performs text preprocessing to standardize the content. This includes removing unnecessary line breaks and special characters.

[0239] Next, the server passes the pre-processed text to an AI agent for analysis. The AI ​​agent analyzes the style and terminology in the document in detail and assesses the likelihood that the document was created by a generative AI. To perform this assessment, the AI ​​agent uses statistical models and machine learning algorithms.

[0240] Furthermore, the server compares the text within the document with existing content in the database and calculates the similarity. This step allows it to determine how similar the document is to other documents. The risk of copyright infringement is also assessed during this process.

[0241] The terminal receives evaluation results from the server and provides them to the user. These results are presented in a detailed report format, which the user can use to decide whether to resubmit or to review the content.

[0242] As a concrete example, consider a research report submitted by an undergraduate student at a university. When a student uploads a report, the server receives it and uses an AI agent to calculate the AI ​​generation rate. Furthermore, it evaluates how similar the report is to other research literature and reports on the possibility of copyright infringement. Based on these results, the university can notify the student of the appropriateness of the report and areas for improvement. In this way, the reliability of submitted materials can be improved by utilizing the system of the present invention.

[0243] The following describes the processing flow.

[0244] Step 1:

[0245] The user uploads the documents to be evaluated to the system. The server saves the received document files to secure storage and prepares them for analysis.

[0246] Step 2:

[0247] The server extracts text from the document. This process uses analysis tools to extract text information from various formats, such as PDFs and Word documents.

[0248] Step 3:

[0249] The server performs preprocessing on the extracted text. This includes removing line breaks and extra spaces, and standardizing special characters. This prepares the text for parsing.

[0250] Step 4:

[0251] The server passes the pre-processed text to a specialized AI agent. The AI ​​agent begins analyzing the writing style and terminology, detecting features that indicate potential for generation by a generative AI. This analysis utilizes natural language processing techniques and machine learning algorithms.

[0252] Step 5:

[0253] The server compares the text in the document to existing documents in the database. During this process, a document similarity algorithm is used to calculate a score indicating how similar the document is to other documents.

[0254] Step 6:

[0255] The server generates a comprehensive evaluation report using the generation rate of the AI ​​and the similarity score. This report includes an assessment of the likelihood that the content was created by the AI, as well as a copyright infringement risk assessment.

[0256] Step 7:

[0257] The terminal provides the user with an evaluation report from the server. Based on this report, the user can consider whether they need to resubmit the materials or revise their content.

[0258] (Example 1)

[0259] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0260] In today's information environment, a vast amount of data is produced daily, but there is a lack of readily available means to determine whether this data is original or generated by AI. Furthermore, there is a need to reliably assess the potential for copyright infringement. This is especially crucial for educational institutions and businesses, where the reliability of submitted materials must be assessed quickly and accurately.

[0261] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0262] In this invention, the server includes means for receiving data, means for extracting and pre-processing textual information, means for analyzing writing style and vocabulary selection and evaluating generative feasibility, means for comparing with other existing data and calculating similarity, and means for reporting the results of the reliability evaluation. This makes it possible to quickly and accurately evaluate the generative feasibility of the data generation AI and the risk of copyright infringement.

[0263] "Documents" refer to documents or files that contain information or data.

[0264] "Textual information" refers to information composed of characters and symbols contained in a document.

[0265] "Analysis" refers to the process of thoroughly examining textual information to understand its structure and meaning.

[0266] "Preprocessing" is the process of removing unnecessary information from textual data and standardizing it to make it easier to analyze.

[0267] "Writing style" refers to the characteristics related to the way a text is expressed and its style.

[0268] "Vocabulary selection" refers to the characteristics related to the choice of words and phrases used in a text.

[0269] "Generability" is an assessment indicating the likelihood that the subject material was created by a generative AI.

[0270] "Existing data" refers to information and data that have been stored in the past and is used as a basis for comparison.

[0271] "Similarity" is a numerical representation of how similar the content of a document is to other existing data.

[0272] "Reliability assessment" refers to the process of determining and reporting whether a document is original or potentially infringes on copyright.

[0273] This invention provides a specific system for evaluating the reliability of data. It describes how hardware and software are used to process and compute the data.

[0274] Users upload the documents to be evaluated to the system using their own devices. The documents are assumed to be in common document formats such as PDF or Word. The server receives the documents and extracts the text information. Various character recognition software, including OCR technology, is used for text extraction.

[0275] Next, the server preprocesses the extracted text information. This preprocessing removes unnecessary line breaks and special characters from the text and standardizes the content. This process is characterized by string manipulation using regular expressions and automated filtering using scripts.

[0276] The processed text information is sent to an AI agent on the server. The AI ​​agent uses natural language processing techniques to analyze the writing style and vocabulary selection of the document. This analysis assesses the likelihood that the document was created by a generative AI model. The AI ​​agent utilizes statistical models and machine learning algorithms to perform a highly accurate assessment.

[0277] Furthermore, the server compares the textual information of the document with existing databases. Vector space models and data mining techniques are used in the process of calculating similarity. This quantifies the degree to which the document is similar to existing literature and data, and assesses the likelihood of copyright infringement.

[0278] Ultimately, the terminal receives the evaluation results from the server and provides the user with information in the form of a detailed report. Based on this report, the user can check the reliability and areas for improvement of the material and revise it as needed.

[0279] As a concrete example, consider a scenario where an undergraduate student submits a research report. When the student uploads the report to the system, the server uses OCR technology to transcribe it into text and analyzes whether it was likely created by a generative AI. It also evaluates the similarity by comparing it to previously submitted reports and publicly available academic literature. This allows the university to verify the originality and reliability of the report and provide feedback to the student.

[0280] An example of a prompt is: "Evaluate whether the following research report is similar to other documents and whether it was likely created by a generative AI."

[0281] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0282] Step 1:

[0283] The user uploads the material to be evaluated to the system using the terminal. This material is in a common format such as PDF or Word. The input is the material file submitted by the user. As output, the server receives and stores the corresponding material. The server stores the user's material for future analysis.

[0284] Step 2:

[0285] The server extracts character information from the uploaded material. In this process, OCR technology is used to convert it to text if it is in image format. The input is the material file received by the server. As output, analyzable character information is provided. By this extraction, the characters in the material can be treated as digital data.

[0286] Step 3:

[0287] The server performs preprocessing on the extracted character information. It removes unnecessary line breaks and special characters and standardizes the text. The input is the initially extracted character information. As output, clean and standardized text data is obtained. This processing improves the analysis accuracy in the subsequent stage.

[0288] Step 4:

[0289] The server passes the preprocessed text to the AI agent to evaluate the possibility that the material was created by the generation AI model. The input is the standardized text data. As output, the evaluation result of the possibility of being created by the generation AI is provided. The AI agent uses statistical models and machine learning algorithms to make a highly accurate determination.

[0290] Step 5:

[0291] The server compares the textual information of the document with existing content stored in the database and calculates the similarity. The input is standardized text data. The output provides the similarity evaluation result. This process determines how similar the document is to other documents.

[0292] Step 6:

[0293] The terminal receives evaluation results from the server and provides them to the user in the form of a detailed report. The input is evaluation result data sent from the server. The output is a detailed reliability evaluation report presented to the user. The user can use this report to improve or resubmit their materials.

[0294] (Application Example 1)

[0295] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0296] Modern educational institutions are required to quickly and accurately assess the reliability of student reports and papers. However, a related problem is the difficulty in effectively detecting similarities between artificially generated texts and existing literature. Such challenges hinder the maintenance of educational quality.

[0297] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0298] In this invention, the server includes means for receiving document information, means for extracting character information from the received document information and performing preprocessing for analysis, means for analyzing the writing style and term selection of the preprocessed character information and evaluating the generated ratio, means for collating the character information with other existing data sets and calculating the similarity, means for outputting the evaluation results of the generated ratio and similarity, and means for notifying the recipient of the evaluation results in a data format. Thereby, it becomes possible to quickly and accurately evaluate the reliability of the submitted document.

[0299] "Document information" refers to data indicating the content of the submitted documents or documents.

[0300] "Character information" is the text data extracted from the document information and is the basis for analysis.

[0301] "Preprocessing" refers to the processing step of removing unnecessary line breaks and special characters for analyzing the received character information.

[0302] "Writing style and term selection" refers to the characteristics of the writing style and word usage unique to the text.

[0303] "Generated ratio" is a numerical value indicating the degree of possibility that the document information was created by artificial intelligence.

[0304] "Similarity" is a measure for evaluating how similar the character information is to the existing data sets.

[0305] "Evaluation result" refers to the comprehensive analysis result output based on the analysis of the generated ratio and similarity.

[0306] "Data format" refers to the reporting format when the evaluation result is provided to the recipient.

[0307] "Recipient" refers to an individual or organization that is the subject of receiving the evaluation result.

[0308] This system is designed to evaluate the reliability of reports submitted by educational institutions. The server receives document information using a cloud platform. The received document information is converted into the necessary textual information by an internal character information extraction module. In this process, SpaCy, a natural language processing tool, is used to remove unnecessary symbols and line breaks, creating a parseable data format.

[0309] Next, the server analyzes the pre-processed text information using machine learning algorithms, specifically TensorFlow or PyTorch, to determine the style and terminology selection, and calculates the likelihood that the material is artificially generated. Furthermore, the text information is compared with existing data sets, and a similarity score is calculated. A text comparison algorithm is used for this matching to accurately assess how similar the material is to other documents.

[0310] The evaluation results are provided to the user in the form of a detailed report. The user's device can visually review this report, allowing them to identify areas for improvement and assess the reliability of the materials. This application primarily operates on smartphones and connects with a cloud server to provide real-time evaluation results.

[0311] As a concrete example, suppose a student uploads an environmental studies report. The server then evaluates the received report, assessing the likelihood that its content was generated by a generative AI model (15%) and determining its similarity to existing academic literature (10%). Such evaluations serve as important indicators for educational institutions to provide appropriate feedback to students.

[0312] An example of a prompt might be: "Check if this report is similar to other literature and assess whether it was created by a generative AI."

[0313] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0314] Step 1:

[0315] The user uploads document information to the server through the application. The input data is document information obtained from the user's local device and sent to the server. The output is the received document recorded on the server side.

[0316] Step 2:

[0317] The server extracts textual information from the received document information. During this process, SpaCy, a natural language processing tool, is used to remove unnecessary symbols and line breaks, generating parseable text data. The input is document information, and the output is pre-processed textual information.

[0318] Step 3:

[0319] The server applies machine learning algorithms to analyze pre-processed text information. Here, TensorFlow or PyTorch is used to analyze writing style and terminology selection, and the likelihood that the material was generated by a generative AI model is calculated. The input is pre-processed text information, and the output is a numerical evaluation of the generative likelihood.

[0320] Step 4:

[0321] The server compares the textual information with an existing data set. A text comparison algorithm is used to calculate similarity, evaluating how similar the document is to other documents. The input is pre-processed textual information, and the output is a similarity score.

[0322] Step 5:

[0323] The server integrates the results of the generative feasibility and similarity evaluations and sends them to the user terminal in the form of a detailed report. The input is the evaluation value, and the output is the integrated report data.

[0324] Step 6:

[0325] The user terminal visually displays the received report and provides the user with areas for improvement and reliability indicators for the document. The input is integrated report data, and the output is visual feedback provided to the user.

[0326] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0327] This invention relates to a system that combines a function to evaluate the likelihood and similarity of materials submitted by users using a generation AI with an emotion engine. This system is primarily intended for use in businesses and educational institutions, and aims to provide more appropriate feedback by taking user emotions into consideration.

[0328] First, the user uploads the document to be evaluated to the system. The server receives this document and extracts text from the file. The server then preprocesses the extracted text, standardizing its content and preparing it for analysis.

[0329] Next, the server passes the pre-processed text to the AI ​​agent, which begins analyzing the writing style and terminology. The AI ​​agent detects and evaluates features that indicate the potential of generative AI. Statistical models and machine learning algorithms are used in this process.

[0330] Furthermore, the server compares the text to existing content in the database and calculates a similarity score. This evaluates how similar the text is to other documents and analyzes the risk of copyright infringement.

[0331] In this system, the emotion engine plays a crucial role. The terminal analyzes user interaction data and recognizes the user's emotional state from it. For example, it can determine emotions from the user's input speed and word choice when operating the system.

[0332] Based on the perceived emotions, the server adjusts the content and presentation format of the evaluation results. For example, if the system determines that the user is disappointed, it can present the evaluation results in a way that includes more thoughtful support.

[0333] A concrete example of this system is a job seeker. When a job seeker uploads their resume, the system evaluates the potential of AI generation and further uses an emotion engine to understand the job seeker's feelings. If the job seeker shows signs of nervousness, the device provides the evaluation results along with an encouraging message, making it easier for the user to understand areas for improvement.

[0334] By combining emotion engines in this way, it is possible to provide services that are more tailored to individual users and improve the effectiveness of the system.

[0335] The following describes the processing flow.

[0336] Step 1:

[0337] Users upload the document files they wish to have evaluated to the system. The server receives these documents and stores them securely. This process also includes verifying that the files are in the correct format.

[0338] Step 2:

[0339] The server extracts text data from the documents. Even if the documents are in PDF, Word, or other formats, it uses appropriate tools to extract the text and prepare it for analysis.

[0340] Step 3:

[0341] The server performs preprocessing on the extracted text. This includes removing unnecessary line breaks and spaces, and standardizing special characters. This makes the text more suitable for analysis.

[0342] Step 4:

[0343] The server sends pre-processed text to the AI ​​agent. The AI ​​agent analyzes the writing style and terminology selection and evaluates the feasibility of generating the text using generative AI. Statistical models and machine learning algorithms are used in this process.

[0344] Step 5:

[0345] In parallel, the server compares the document text with existing data in the database and calculates similarity. It then calculates a similarity score and analyzes the risk of copyright infringement.

[0346] Step 6:

[0347] The device uses user interaction data to activate an emotion engine. The emotion engine analyzes data such as the user's input speed and word choice to recognize the user's emotional state.

[0348] Step 7:

[0349] The server determines how to present the evaluation results based on the user's emotions recognized by the emotion engine. For example, if the user appears anxious, the results will be presented in a calm tone, including advice.

[0350] Step 8:

[0351] The device provides the user with a report containing the adjusted evaluation results. Based on this report, the user can consider areas for improvement in their materials and the possibility of resubmission.

[0352] (Example 2)

[0353] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0354] Conventional document evaluation systems could assess the likelihood of document generation and similarity using AI, but they lacked the means to provide feedback that took user emotions into consideration. This resulted in problems such as evaluation results causing psychological burden on users and preventing them from taking appropriate corrective actions.

[0355] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0356] In this invention, the server includes means for receiving data, means for extracting and pre-processing text, means for analyzing the style and terminology of the text and evaluating its generative potential, means for comparing the text with existing data and calculating similarity, and means for analyzing emotions from user operation data and adjusting the presentation format of the evaluation results. This enables feedback tailored to the user's emotional state, reduces psychological burden, and promotes appropriate improvement actions.

[0357] "Documents" refers to the documents uploaded to the system for evaluation. This typically includes content provided as PDF or Word files.

[0358] "Text" refers to string data extracted from the document. This allows the content to be analyzed mechanically.

[0359] "Preprocessing" refers to the process of standardizing and formatting the extracted text in preparation for analysis. This includes removing unnecessary characters and unifying the format.

[0360] "Analysis" is a general term for technical operations performed to evaluate the content of text. This includes the detection of style and terminology, and the evaluation of generative potential.

[0361] "Generability" refers to characteristics that indicate the likelihood that the material was generated by a generative AI. This includes evaluation results based on stylistic and terminology patterns.

[0362] "Similarity" is a measure that indicates the degree to which text shares commonalities with existing data. This allows for an objective evaluation of similarity with other materials.

[0363] "Emotion" refers to the psychological state inferred from the user's operation data. This includes evaluation results based on input speed and operation history.

[0364] "Feedback" refers to the information that the system provides to the user based on its analysis results. This information is adjusted according to the user's emotional state.

[0365] This invention realizes a system that analyzes the material to be evaluated, assesses the possibility and similarity of generation by the generation AI, and provides feedback while taking into account the user's emotions. It is primarily intended for use in educational institutions and companies, and aims to analyze materials and provide psychological support to users.

[0366] Users first upload the documents to be evaluated to the system. This operation is performed via a web interface on their terminal, and the data is provided in PDF or Word file format. The server receives the documents and extracts the text data. This utilizes PDF parsing and OCR (optical character recognition) technology.

[0367] Next, the server performs preprocessing on the extracted text. This preprocessing removes unnecessary spaces and special characters, and formats the text, preparing it for analysis. The server then passes the preprocessed text to an AI agent, which uses a generative AI model to analyze the writing style and terminology selection. This AI agent combines natural language processing and machine learning algorithms to evaluate the potential for generation.

[0368] The server also compares the text against existing data in the database to assess its similarity. Numerical methods, such as cosine similarity, are used to calculate a similarity score. This allows for an assessment of the text's similarities to other materials and the associated copyright risks.

[0369] A key feature of this system is that the emotion engine analyzes the user's operation data to infer their psychological state. The terminal collects data such as input speed and word choice to identify the emotional state. Based on this recognition, the server adjusts the format in which the evaluation results are presented. For example, if the user is feeling anxious, the feedback can be provided in an encouraging way.

[0370] As a concrete example, consider a case where a job seeker uploads their resume to the system and the AI ​​evaluates its potential for generation. In this case, if the emotion engine detects the job seeker's anxiety, the terminal can provide feedback such as, "This evaluation result is intended to show areas for improvement. Let's think positively about the next steps."

[0371] A concrete example of a prompt is the instruction, "Evaluate the AI-generated potential of the uploaded document and adjust the feedback according to the user's sentiment." By using this prompt, the system can achieve multi-layered evaluation and feedback.

[0372] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0373] Step 1:

[0374] Users upload documents to the system. PDF and Word files are used as input. The terminal checks the file format and prepares to send it to the server.

[0375] Step 2:

[0376] The server extracts text from the received documents. The input is the uploaded file, and the output is the extracted raw text data. This process utilizes OCR technology and PDF parser libraries.

[0377] Step 3:

[0378] The server preprocesses the extracted text. The input is raw text data, and the output is formatted and standardized text. Unnecessary spaces and special characters are removed, and morphological analysis is performed to prepare the text for analysis.

[0379] Step 4:

[0380] The server passes pre-processed text to the AI ​​agent for evaluation of its generative potential. The input is formatted text, and the output is a generative potential evaluation score. The AI ​​agent analyzes the writing style and terminology selection using natural language processing models and machine learning algorithms.

[0381] Step 5:

[0382] The server calculates similarity by comparing the text with existing data. The input is pre-processed text, and the output is a similarity score. Vector comparison methods such as cosine similarity are used in this process to evaluate the similarity with other materials.

[0383] Step 6:

[0384] The device analyzes user interaction data to infer emotions. Inputs include user input speed and interaction history data, while output is an inferred emotion label. The device sends this data to an emotion engine for analysis of the user's psychological state.

[0385] Step 7:

[0386] The server adjusts the presentation format of the evaluation results based on the emotion label. The input is the evaluation results of generativeness and similarity, and the emotion label, and the output is an adjusted feedback message. For example, if the server determines that the user is stressed, the feedback will be provided in gentler language.

[0387] Step 8:

[0388] The device presents the user with refined feedback. The input is a refined feedback message sent from the server, and the output is the evaluation result and support message displayed on the user's screen. Information is provided to help the user understand areas for improvement and move forward to the next step.

[0389] (Application Example 2)

[0390] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0391] The problem that this invention aims to solve is how to provide feedback that takes into account the user's emotional state, rather than simply checking the possibility and similarity of materials submitted by the user using a generating AI. In particular, current systems are unable to reflect the user's psychological state, resulting in uniform evaluation results and a lack of individualized support.

[0392] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0393] In this invention, the server includes means for receiving data, means for extracting text from the received data and performing preprocessing for analysis, means for analyzing the style and terminology of the preprocessed text and evaluating the likelihood of generation, means for comparing the text with other existing data and calculating similarity, means for outputting the evaluation results of the likelihood and similarity of generation, and means for recognizing the user's emotional state and adjusting the evaluation results accordingly. This enables detailed feedback that responds to the user's emotional state.

[0394] "Documents" refer to the documents and data that users submit to the system for evaluation.

[0395] "Means of receiving" refers to the function for importing data from users into the server.

[0396] "Means for extraction and pre-processing for analysis" refers to a function that extracts necessary text data from a document and prepares it in a format suitable for analysis.

[0397] "Means for evaluating the likelihood of generation" refers to a function that analyzes whether preprocessed text was generated by AI and quantifies that likelihood.

[0398] "Methods for comparing with existing data and calculating similarity" refers to a function that compares the input text with existing data in the database and expresses the degree of similarity as a numerical value.

[0399] "Means for outputting evaluation results" refers to a function that provides the generated probability and similarity analysis results in a format that users can review.

[0400] "Means for recognizing emotional states and adjusting evaluation results" refers to a function that analyzes user interaction data to determine emotions and changes the content and method of feedback accordingly.

[0401] The system implementing this invention receives data from the user and provides feedback that takes into account the user's emotional state based on that data. The server first receives data submitted electronically by the user. The data is taken into the server as text data, and its contents are pre-processed for analysis. In the pre-processing, the text is standardized and prepared for evaluation.

[0402] Next, the server analyzes the pre-processed text and evaluates its potential for generation by the AI ​​based on its writing style and terminology. Statistical models and machine learning algorithms are used for this evaluation. The text is also compared with existing data in the database, and a similarity score is calculated.

[0403] One of the features of this invention is that the server recognizes the user's emotional state and reflects it in the evaluation results. The terminal analyzes the user's interaction data and estimates the emotional state from input speed and word choice. For example, if the user is confused, the server carefully explains the evaluation results and clearly communicates areas for improvement. As a result, the user can receive feedback tailored to their emotional state.

[0404] One concrete application example is when children studying at home use a device designed to support home learning. When a child submits an essay assignment, the device evaluates it using a generative AI model and also adds encouraging words, taking into account the child's emotional state. In this way, the user experience can be further improved.

[0405] An example of a prompt message is: "Based on the following evaluation results, please tell me what encouraging comments I should add if the user is confused: The evaluation results indicate a 50% probability of being AI-generated and a similarity score of 20%."

[0406] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0407] Step 1:

[0408] The server receives materials submitted electronically by users. The input is document data uploaded by the user, which the server receives and stores. The output is the material to be processed, which is stored within the server.

[0409] Step 2:

[0410] The server extracts text from received materials and performs preprocessing for analysis. The input is document data stored on the server, and the server extracts text data from the documents using a text extraction algorithm. Next, it performs normalization and removes unnecessary information, and outputs text ready for analysis.

[0411] Step 3:

[0412] The server analyzes pre-processed text using a generative AI model and evaluates the likelihood of it being generated. The input is pre-processed text data, and the server analyzes the style and terminology selection of this data using statistical models and machine learning algorithms. The output is a numerical value indicating the likelihood that the text was generated by a generative AI.

[0413] Step 4:

[0414] The server compares text with data in an existing database and calculates similarity. The input is pre-processed text data, and the server calculates similarity by comparing it with similar data in the database. The output is the similarity score obtained from the comparison.

[0415] Step 5:

[0416] The device analyzes user interaction data in real time to recognize emotional states. Input data includes user input speed and word choice, and the device analyzes this data to estimate the emotional state and output the emotion recognition result.

[0417] Step 6:

[0418] The server combines the generated probabilities, similarity scores, and the user's emotional state to refine and output feedback. The inputs are the generated probability values, similarity scores, and emotional recognition results, which the server integrates to generate feedback tailored to the user. The output is presented to the user as a refined evaluation result.

[0419] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0420] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0421] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0422] [Third Embodiment]

[0423] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0424] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0425] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0426] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0427] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0428] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0429] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0430] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0431] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0432] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0433] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0434] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0435] This invention relates to a system for appropriately evaluating the authenticity and copyright infringement risk of materials submitted by users. This system is primarily intended for use in businesses and educational institutions.

[0436] First, the user uploads the document to be evaluated to the system. The server receives this document and performs text extraction. This extracts the textual information from the document in a parseable format. The server then performs text preprocessing to standardize the content. This includes removing unnecessary line breaks and special characters.

[0437] Next, the server passes the pre-processed text to an AI agent for analysis. The AI ​​agent analyzes the style and terminology in the document in detail and assesses the likelihood that the document was created by a generative AI. To perform this assessment, the AI ​​agent uses statistical models and machine learning algorithms.

[0438] Furthermore, the server compares the text within the document with existing content in the database and calculates the similarity. This step allows it to determine how similar the document is to other documents. The risk of copyright infringement is also assessed during this process.

[0439] The terminal receives evaluation results from the server and provides them to the user. These results are presented in a detailed report format, which the user can use to decide whether to resubmit or to review the content.

[0440] As a concrete example, consider a research report submitted by an undergraduate student at a university. When a student uploads a report, the server receives it and uses an AI agent to calculate the AI ​​generation rate. Furthermore, it evaluates how similar the report is to other research literature and reports on the possibility of copyright infringement. Based on these results, the university can notify the student of the appropriateness of the report and areas for improvement. In this way, the reliability of submitted materials can be improved by utilizing the system of the present invention.

[0441] The following describes the processing flow.

[0442] Step 1:

[0443] The user uploads the documents to be evaluated to the system. The server saves the received document files to secure storage and prepares them for analysis.

[0444] Step 2:

[0445] The server extracts text from the document. This process uses analysis tools to extract text information from various formats, such as PDFs and Word documents.

[0446] Step 3:

[0447] The server performs preprocessing on the extracted text. This includes removing line breaks and extra spaces, and standardizing special characters. This prepares the text for parsing.

[0448] Step 4:

[0449] The server passes the pre-processed text to a specialized AI agent. The AI ​​agent begins analyzing the writing style and terminology, detecting features that indicate potential for generation by a generative AI. This analysis utilizes natural language processing techniques and machine learning algorithms.

[0450] Step 5:

[0451] The server compares the text in the document to existing documents in the database. During this process, a document similarity algorithm is used to calculate a score indicating how similar the document is to other documents.

[0452] Step 6:

[0453] The server generates a comprehensive evaluation report using the generation rate of the AI ​​and the similarity score. This report includes an assessment of the likelihood that the content was created by the AI, as well as a copyright infringement risk assessment.

[0454] Step 7:

[0455] The terminal provides the user with an evaluation report from the server. Based on this report, the user can consider whether they need to resubmit the materials or revise their content.

[0456] (Example 1)

[0457] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0458] In today's information environment, a vast amount of data is produced daily, but there is a lack of readily available means to determine whether this data is original or generated by AI. Furthermore, there is a need to reliably assess the potential for copyright infringement. This is especially crucial for educational institutions and businesses, where the reliability of submitted materials must be assessed quickly and accurately.

[0459] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0460] In this invention, the server includes means for receiving data, means for extracting and pre-processing textual information, means for analyzing writing style and vocabulary selection and evaluating generative feasibility, means for comparing with other existing data and calculating similarity, and means for reporting the results of the reliability evaluation. This makes it possible to quickly and accurately evaluate the generative feasibility of the data generation AI and the risk of copyright infringement.

[0461] "Documents" refer to documents or files that contain information or data.

[0462] "Textual information" refers to information composed of characters and symbols contained in a document.

[0463] "Analysis" refers to the process of thoroughly examining textual information to understand its structure and meaning.

[0464] "Preprocessing" is the process of removing unnecessary information from textual data and standardizing it to make it easier to analyze.

[0465] "Writing style" refers to the characteristics related to the way a text is expressed and its style.

[0466] "Vocabulary selection" refers to the characteristics related to the choice of words and phrases used in a text.

[0467] "Generability" is an assessment indicating the likelihood that the subject material was created by a generative AI.

[0468] "Existing data" refers to information and data that have been stored in the past and is used as a basis for comparison.

[0469] "Similarity" is a numerical representation of how similar the content of a document is to other existing data.

[0470] "Reliability assessment" refers to the process of determining and reporting whether a document is original or potentially infringes on copyright.

[0471] This invention provides a specific system for evaluating the reliability of data. It describes how hardware and software are used to process and compute the data.

[0472] Users upload the documents to be evaluated to the system using their own devices. The documents are assumed to be in common document formats such as PDF or Word. The server receives the documents and extracts the text information. Various character recognition software, including OCR technology, is used for text extraction.

[0473] Next, the server preprocesses the extracted text information. This preprocessing removes unnecessary line breaks and special characters from the text and standardizes the content. This process is characterized by string manipulation using regular expressions and automated filtering using scripts.

[0474] The processed text information is sent to an AI agent on the server. The AI ​​agent uses natural language processing techniques to analyze the writing style and vocabulary selection of the document. This analysis assesses the likelihood that the document was created by a generative AI model. The AI ​​agent utilizes statistical models and machine learning algorithms to perform a highly accurate assessment.

[0475] Furthermore, the server compares the textual information of the document with existing databases. Vector space models and data mining techniques are used in the process of calculating similarity. This quantifies the degree to which the document is similar to existing literature and data, and assesses the likelihood of copyright infringement.

[0476] Ultimately, the terminal receives the evaluation results from the server and provides the user with information in the form of a detailed report. Based on this report, the user can check the reliability and areas for improvement of the material and revise it as needed.

[0477] As a concrete example, consider a scenario where an undergraduate student submits a research report. When the student uploads the report to the system, the server uses OCR technology to transcribe it into text and analyzes whether it was likely created by a generative AI. It also evaluates the similarity by comparing it to previously submitted reports and publicly available academic literature. This allows the university to verify the originality and reliability of the report and provide feedback to the student.

[0478] An example of a prompt is: "Evaluate whether the following research report is similar to other documents and whether it was likely created by a generative AI."

[0479] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0480] Step 1:

[0481] Users upload documents to be evaluated to the system using their terminals. These documents are in common formats such as PDF and Word. The input is the document file submitted by the user. The output is the server receiving and saving the relevant documents. The server stores the user's documents for future analysis.

[0482] Step 2:

[0483] The server extracts text information from the uploaded documents. This process utilizes OCR technology, converting image formats into text. The input is the document file received by the server. The output provides analyzable text information. This extraction allows the text within the document to be treated as digital data.

[0484] Step 3:

[0485] The server preprocesses the extracted character information. It removes unnecessary line breaks and special characters and standardizes the text. The input is the initial extracted character information. The output is clean, standardized text data. This processing improves the accuracy of subsequent analysis.

[0486] Step 4:

[0487] The server passes pre-processed text to the AI ​​agent, which evaluates the likelihood that the document was created by a generative AI model. The input is standardized text data. The output provides an evaluation of the likelihood that the document was created by a generative AI. The AI ​​agent uses statistical models and machine learning algorithms to make highly accurate judgments.

[0488] Step 5:

[0489] The server compares the textual information of the document with existing content stored in the database and calculates the similarity. The input is standardized text data. The output provides the similarity evaluation result. This process determines how similar the document is to other documents.

[0490] Step 6:

[0491] The terminal receives evaluation results from the server and provides them to the user in the form of a detailed report. The input is evaluation result data sent from the server. The output is a detailed reliability evaluation report presented to the user. The user can use this report to improve or resubmit their materials.

[0492] (Application Example 1)

[0493] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0494] Modern educational institutions are required to quickly and accurately assess the reliability of student reports and papers. However, a related problem is the difficulty in effectively detecting similarities between artificially generated texts and existing literature. Such challenges hinder the maintenance of educational quality.

[0495] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0496] In this invention, the server includes means for receiving data information, means for extracting textual information from the received data information and performing preprocessing for analysis, means for analyzing the style and terminology selection of the preprocessed textual information and evaluating the generated ratio, means for comparing the textual information with other existing data sets and calculating similarity, means for outputting the generated ratio and the evaluation result of the similarity, and means for notifying the recipient of the evaluation result in data format. This enables a rapid and accurate evaluation of the reliability of submitted data.

[0497] "Document information" refers to data that shows the contents of submitted documents and papers.

[0498] "Textual information" refers to text data extracted from source material, and it forms the basis for analysis.

[0499] "Preprocessing" refers to the process of removing unnecessary line breaks and special characters from received text information in order to analyze it.

[0500] "Style and terminology selection" refers to the unique writing style and vocabulary used in a particular text.

[0501] "Generated rate" is a numerical value that indicates the degree to which the source information was likely created by artificial intelligence.

[0502] "Similarity" is a measure used to evaluate how similar textual information is to an existing data set.

[0503] "Evaluation results" refer to the overall analysis results output based on the analysis of the generated proportions and similarity.

[0504] "Data format" refers to the reporting format in which evaluation results are provided to recipients.

[0505] "Recipient" refers to the individual or organization that receives the evaluation results.

[0506] This system is designed to evaluate the reliability of reports submitted by educational institutions. The server receives document information using a cloud platform. The received document information is converted into the necessary textual information by an internal character information extraction module. In this process, SpaCy, a natural language processing tool, is used to remove unnecessary symbols and line breaks, creating a parseable data format.

[0507] Next, the server analyzes the pre-processed text information using machine learning algorithms, specifically TensorFlow or PyTorch, to determine the style and terminology selection, and calculates the likelihood that the material is artificially generated. Furthermore, the text information is compared with existing data sets, and a similarity score is calculated. A text comparison algorithm is used for this matching to accurately assess how similar the material is to other documents.

[0508] The evaluation results are provided to the user in the form of a detailed report. The user's device can visually review this report, allowing them to identify areas for improvement and assess the reliability of the materials. This application primarily operates on smartphones and connects with a cloud server to provide real-time evaluation results.

[0509] As a concrete example, suppose a student uploads an environmental studies report. The server then evaluates the received report, assessing the likelihood that its content was generated by a generative AI model (15%) and determining its similarity to existing academic literature (10%). Such evaluations serve as important indicators for educational institutions to provide appropriate feedback to students.

[0510] An example of a prompt might be: "Check if this report is similar to other literature and assess whether it was created by a generative AI."

[0511] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0512] Step 1:

[0513] The user uploads document information to the server through the application. The input data is document information obtained from the user's local device and sent to the server. The output is the received document recorded on the server side.

[0514] Step 2:

[0515] The server extracts textual information from the received document information. During this process, SpaCy, a natural language processing tool, is used to remove unnecessary symbols and line breaks, generating parseable text data. The input is document information, and the output is pre-processed textual information.

[0516] Step 3:

[0517] The server applies machine learning algorithms to analyze pre-processed text information. Here, TensorFlow or PyTorch is used to analyze writing style and terminology selection, and the likelihood that the material was generated by a generative AI model is calculated. The input is pre-processed text information, and the output is a numerical evaluation of the generative likelihood.

[0518] Step 4:

[0519] The server compares the textual information with an existing data set. A text comparison algorithm is used to calculate similarity, evaluating how similar the document is to other documents. The input is pre-processed textual information, and the output is a similarity score.

[0520] Step 5:

[0521] The server integrates the results of the generative feasibility and similarity evaluations and sends them to the user terminal in the form of a detailed report. The input is the evaluation value, and the output is the integrated report data.

[0522] Step 6:

[0523] The user terminal visually displays the received report and provides the user with areas for improvement and reliability indicators for the document. The input is integrated report data, and the output is visual feedback provided to the user.

[0524] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0525] This invention relates to a system that combines a function to evaluate the likelihood and similarity of materials submitted by users using a generation AI with an emotion engine. This system is primarily intended for use in businesses and educational institutions, and aims to provide more appropriate feedback by taking user emotions into consideration.

[0526] First, the user uploads the document to be evaluated to the system. The server receives this document and extracts text from the file. The server then preprocesses the extracted text, standardizing its content and preparing it for analysis.

[0527] Next, the server passes the pre-processed text to the AI ​​agent, which begins analyzing the writing style and terminology. The AI ​​agent detects and evaluates features that indicate the potential of generative AI. Statistical models and machine learning algorithms are used in this process.

[0528] Furthermore, the server compares the text to existing content in the database and calculates a similarity score. This evaluates how similar the text is to other documents and analyzes the risk of copyright infringement.

[0529] In this system, the emotion engine plays a crucial role. The terminal analyzes user interaction data and recognizes the user's emotional state from it. For example, it can determine emotions from the user's input speed and word choice when operating the system.

[0530] Based on the perceived emotions, the server adjusts the content and presentation format of the evaluation results. For example, if the system determines that the user is disappointed, it can present the evaluation results in a way that includes more thoughtful support.

[0531] A concrete example of this system is a job seeker. When a job seeker uploads their resume, the system evaluates the potential of AI generation and further uses an emotion engine to understand the job seeker's feelings. If the job seeker shows signs of nervousness, the device provides the evaluation results along with an encouraging message, making it easier for the user to understand areas for improvement.

[0532] By combining emotion engines in this way, it is possible to provide services that are more tailored to individual users and improve the effectiveness of the system.

[0533] The following describes the processing flow.

[0534] Step 1:

[0535] Users upload the document files they wish to have evaluated to the system. The server receives these documents and stores them securely. This process also includes verifying that the files are in the correct format.

[0536] Step 2:

[0537] The server extracts text data from the documents. Even if the documents are in PDF, Word, or other formats, it uses appropriate tools to extract the text and prepare it for analysis.

[0538] Step 3:

[0539] The server performs preprocessing on the extracted text. This includes removing unnecessary line breaks and spaces, and standardizing special characters. This makes the text more suitable for analysis.

[0540] Step 4:

[0541] The server sends pre-processed text to the AI ​​agent. The AI ​​agent analyzes the writing style and terminology selection and evaluates the feasibility of generating the text using generative AI. Statistical models and machine learning algorithms are used in this process.

[0542] Step 5:

[0543] In parallel, the server compares the document text with existing data in the database and calculates similarity. It then calculates a similarity score and analyzes the risk of copyright infringement.

[0544] Step 6:

[0545] The device uses user interaction data to activate an emotion engine. The emotion engine analyzes data such as the user's input speed and word choice to recognize the user's emotional state.

[0546] Step 7:

[0547] The server determines how to present the evaluation results based on the user's emotions recognized by the emotion engine. For example, if the user appears anxious, the results will be presented in a calm tone, including advice.

[0548] Step 8:

[0549] The device provides the user with a report containing the adjusted evaluation results. Based on this report, the user can consider areas for improvement in their materials and the possibility of resubmission.

[0550] (Example 2)

[0551] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0552] Conventional document evaluation systems could assess the likelihood of document generation and similarity using AI, but they lacked the means to provide feedback that took user emotions into consideration. This resulted in problems such as evaluation results causing psychological burden on users and preventing them from taking appropriate corrective actions.

[0553] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0554] In this invention, the server includes means for receiving data, means for extracting and pre-processing text, means for analyzing the style and terminology of the text and evaluating its generative potential, means for comparing the text with existing data and calculating similarity, and means for analyzing emotions from user operation data and adjusting the presentation format of the evaluation results. This enables feedback tailored to the user's emotional state, reduces psychological burden, and promotes appropriate improvement actions.

[0555] "Documents" refers to the documents uploaded to the system for evaluation. This typically includes content provided as PDF or Word files.

[0556] "Text" refers to string data extracted from the document. This allows the content to be analyzed mechanically.

[0557] "Preprocessing" refers to the process of standardizing and formatting the extracted text in preparation for analysis. This includes removing unnecessary characters and unifying the format.

[0558] "Analysis" is a general term for technical operations performed to evaluate the content of text. This includes the detection of style and terminology, and the evaluation of generative potential.

[0559] "Generability" refers to characteristics that indicate the likelihood that the material was generated by a generative AI. This includes evaluation results based on stylistic and terminology patterns.

[0560] "Similarity" is a measure that indicates the degree to which text shares commonalities with existing data. This allows for an objective evaluation of similarity with other materials.

[0561] "Emotion" refers to the psychological state inferred from the user's operation data. This includes evaluation results based on input speed and operation history.

[0562] "Feedback" refers to the information that the system provides to the user based on its analysis results. This information is adjusted according to the user's emotional state.

[0563] This invention realizes a system that analyzes the material to be evaluated, assesses the possibility and similarity of generation by the generation AI, and provides feedback while taking into account the user's emotions. It is primarily intended for use in educational institutions and companies, and aims to analyze materials and provide psychological support to users.

[0564] Users first upload the documents to be evaluated to the system. This operation is performed via a web interface on their terminal, and the data is provided in PDF or Word file format. The server receives the documents and extracts the text data. This utilizes PDF parsing and OCR (optical character recognition) technology.

[0565] Next, the server performs preprocessing on the extracted text. This preprocessing removes unnecessary spaces and special characters, and formats the text, preparing it for analysis. The server then passes the preprocessed text to an AI agent, which uses a generative AI model to analyze the writing style and terminology selection. This AI agent combines natural language processing and machine learning algorithms to evaluate the potential for generation.

[0566] The server also compares the text against existing data in the database to assess its similarity. Numerical methods, such as cosine similarity, are used to calculate a similarity score. This allows for an assessment of the text's similarities to other materials and the associated copyright risks.

[0567] A key feature of this system is that the emotion engine analyzes the user's operation data to infer their psychological state. The terminal collects data such as input speed and word choice to identify the emotional state. Based on this recognition, the server adjusts the format in which the evaluation results are presented. For example, if the user is feeling anxious, the feedback can be provided in an encouraging way.

[0568] As a concrete example, consider a case where a job seeker uploads their resume to the system and the AI ​​evaluates its potential for generation. In this case, if the emotion engine detects the job seeker's anxiety, the terminal can provide feedback such as, "This evaluation result is intended to show areas for improvement. Let's think positively about the next steps."

[0569] A concrete example of a prompt is the instruction, "Evaluate the AI-generated potential of the uploaded document and adjust the feedback according to the user's sentiment." By using this prompt, the system can achieve multi-layered evaluation and feedback.

[0570] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0571] Step 1:

[0572] Users upload documents to the system. PDF and Word files are used as input. The terminal checks the file format and prepares to send it to the server.

[0573] Step 2:

[0574] The server extracts text from the received documents. The input is the uploaded file, and the output is the extracted raw text data. This process utilizes OCR technology and PDF parser libraries.

[0575] Step 3:

[0576] The server preprocesses the extracted text. The input is raw text data, and the output is formatted and standardized text. Unnecessary spaces and special characters are removed, and morphological analysis is performed to prepare the text for analysis.

[0577] Step 4:

[0578] The server passes pre-processed text to the AI ​​agent for evaluation of its generative potential. The input is formatted text, and the output is a generative potential evaluation score. The AI ​​agent analyzes the writing style and terminology selection using natural language processing models and machine learning algorithms.

[0579] Step 5:

[0580] The server calculates similarity by comparing the text with existing data. The input is pre-processed text, and the output is a similarity score. Vector comparison methods such as cosine similarity are used in this process to evaluate the similarity with other materials.

[0581] Step 6:

[0582] The device analyzes user interaction data to infer emotions. Inputs include user input speed and interaction history data, while output is an inferred emotion label. The device sends this data to an emotion engine for analysis of the user's psychological state.

[0583] Step 7:

[0584] The server adjusts the presentation format of the evaluation results based on the emotion label. The input is the evaluation results of generativeness and similarity, and the emotion label, and the output is an adjusted feedback message. For example, if the server determines that the user is stressed, the feedback will be provided in gentler language.

[0585] Step 8:

[0586] The device presents the user with refined feedback. The input is a refined feedback message sent from the server, and the output is the evaluation result and support message displayed on the user's screen. Information is provided to help the user understand areas for improvement and move forward to the next step.

[0587] (Application Example 2)

[0588] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0589] The problem that this invention aims to solve is how to provide feedback that takes into account the user's emotional state, rather than simply checking the possibility and similarity of materials submitted by the user using a generating AI. In particular, current systems are unable to reflect the user's psychological state, resulting in uniform evaluation results and a lack of individualized support.

[0590] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0591] In this invention, the server includes means for receiving data, means for extracting text from the received data and performing preprocessing for analysis, means for analyzing the style and terminology of the preprocessed text and evaluating the likelihood of generation, means for comparing the text with other existing data and calculating similarity, means for outputting the evaluation results of the likelihood and similarity of generation, and means for recognizing the user's emotional state and adjusting the evaluation results accordingly. This enables detailed feedback that responds to the user's emotional state.

[0592] "Documents" refer to the documents and data that users submit to the system for evaluation.

[0593] "Means of receiving" refers to the function for importing data from users into the server.

[0594] "Means for extraction and pre-processing for analysis" refers to a function that extracts necessary text data from a document and prepares it in a format suitable for analysis.

[0595] "Means for evaluating the likelihood of generation" refers to a function that analyzes whether preprocessed text was generated by AI and quantifies that likelihood.

[0596] "Methods for comparing with existing data and calculating similarity" refers to a function that compares the input text with existing data in the database and expresses the degree of similarity as a numerical value.

[0597] "Means for outputting evaluation results" refers to a function that provides the generated probability and similarity analysis results in a format that users can review.

[0598] "Means for recognizing emotional states and adjusting evaluation results" refers to a function that analyzes user interaction data to determine emotions and changes the content and method of feedback accordingly.

[0599] The system implementing this invention receives data from the user and provides feedback that takes into account the user's emotional state based on that data. The server first receives data submitted electronically by the user. The data is taken into the server as text data, and its contents are pre-processed for analysis. In the pre-processing, the text is standardized and prepared for evaluation.

[0600] Next, the server analyzes the pre-processed text and evaluates its potential for generation by the AI ​​based on its writing style and terminology. Statistical models and machine learning algorithms are used for this evaluation. The text is also compared with existing data in the database, and a similarity score is calculated.

[0601] One of the features of this invention is that the server recognizes the user's emotional state and reflects it in the evaluation results. The terminal analyzes the user's interaction data and estimates the emotional state from input speed and word choice. For example, if the user is confused, the server carefully explains the evaluation results and clearly communicates areas for improvement. As a result, the user can receive feedback tailored to their emotional state.

[0602] One concrete application example is when children studying at home use a device designed to support home learning. When a child submits an essay assignment, the device evaluates it using a generative AI model and also adds encouraging words, taking into account the child's emotional state. In this way, the user experience can be further improved.

[0603] An example of a prompt message is: "Based on the following evaluation results, please tell me what encouraging comments I should add if the user is confused: The evaluation results indicate a 50% probability of being AI-generated and a similarity score of 20%."

[0604] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0605] Step 1:

[0606] The server receives materials submitted electronically by users. The input is document data uploaded by the user, which the server receives and stores. The output is the material to be processed, which is stored within the server.

[0607] Step 2:

[0608] The server extracts text from received materials and performs preprocessing for analysis. The input is document data stored on the server, and the server extracts text data from the documents using a text extraction algorithm. Next, it performs normalization and removes unnecessary information, and outputs text ready for analysis.

[0609] Step 3:

[0610] The server analyzes pre-processed text using a generative AI model and evaluates the likelihood of it being generated. The input is pre-processed text data, and the server analyzes the style and terminology selection of this data using statistical models and machine learning algorithms. The output is a numerical value indicating the likelihood that the text was generated by a generative AI.

[0611] Step 4:

[0612] The server compares text with data in an existing database and calculates similarity. The input is pre-processed text data, and the server calculates similarity by comparing it with similar data in the database. The output is the similarity score obtained from the comparison.

[0613] Step 5:

[0614] The device analyzes user interaction data in real time to recognize emotional states. Input data includes user input speed and word choice, and the device analyzes this data to estimate the emotional state and output the emotion recognition result.

[0615] Step 6:

[0616] The server combines the generated probabilities, similarity scores, and the user's emotional state to refine and output feedback. The inputs are the generated probability values, similarity scores, and emotional recognition results, which the server integrates to generate feedback tailored to the user. The output is presented to the user as a refined evaluation result.

[0617] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0618] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0619] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0620] [Fourth Embodiment]

[0621] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0622] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0623] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0624] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0625] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0626] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0627] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0628] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0629] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0630] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0631] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0632] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0633] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0634] This invention relates to a system for appropriately evaluating the authenticity and copyright infringement risk of materials submitted by users. This system is primarily intended for use in businesses and educational institutions.

[0635] First, the user uploads the document to be evaluated to the system. The server receives this document and performs text extraction. This extracts the textual information from the document in a parseable format. The server then performs text preprocessing to standardize the content. This includes removing unnecessary line breaks and special characters.

[0636] Next, the server passes the pre-processed text to an AI agent for analysis. The AI ​​agent analyzes the style and terminology in the document in detail and assesses the likelihood that the document was created by a generative AI. To perform this assessment, the AI ​​agent uses statistical models and machine learning algorithms.

[0637] Furthermore, the server compares the text within the document with existing content in the database and calculates the similarity. This step allows it to determine how similar the document is to other documents. The risk of copyright infringement is also assessed during this process.

[0638] The terminal receives evaluation results from the server and provides them to the user. These results are presented in a detailed report format, which the user can use to decide whether to resubmit or to review the content.

[0639] As a concrete example, consider a research report submitted by an undergraduate student at a university. When a student uploads a report, the server receives it and uses an AI agent to calculate the AI ​​generation rate. Furthermore, it evaluates how similar the report is to other research literature and reports on the possibility of copyright infringement. Based on these results, the university can notify the student of the appropriateness of the report and areas for improvement. In this way, the reliability of submitted materials can be improved by utilizing the system of the present invention.

[0640] The following describes the processing flow.

[0641] Step 1:

[0642] The user uploads the documents to be evaluated to the system. The server saves the received document files to secure storage and prepares them for analysis.

[0643] Step 2:

[0644] The server extracts text from the document. This process uses analysis tools to extract text information from various formats, such as PDFs and Word documents.

[0645] Step 3:

[0646] The server performs preprocessing on the extracted text. This includes removing line breaks and extra spaces, and standardizing special characters. This prepares the text for parsing.

[0647] Step 4:

[0648] The server passes the pre-processed text to a specialized AI agent. The AI ​​agent begins analyzing the writing style and terminology, detecting features that indicate potential for generation by a generative AI. This analysis utilizes natural language processing techniques and machine learning algorithms.

[0649] Step 5:

[0650] The server compares the text in the document to existing documents in the database. During this process, a document similarity algorithm is used to calculate a score indicating how similar the document is to other documents.

[0651] Step 6:

[0652] The server generates a comprehensive evaluation report using the generation rate of the AI ​​and the similarity score. This report includes an assessment of the likelihood that the content was created by the AI, as well as a copyright infringement risk assessment.

[0653] Step 7:

[0654] The terminal provides the user with an evaluation report from the server. Based on this report, the user can consider whether they need to resubmit the materials or revise their content.

[0655] (Example 1)

[0656] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0657] In today's information environment, a vast amount of data is produced daily, but there is a lack of readily available means to determine whether this data is original or generated by AI. Furthermore, there is a need to reliably assess the potential for copyright infringement. This is especially crucial for educational institutions and businesses, where the reliability of submitted materials must be assessed quickly and accurately.

[0658] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0659] In this invention, the server includes means for receiving data, means for extracting and pre-processing textual information, means for analyzing writing style and vocabulary selection and evaluating generative feasibility, means for comparing with other existing data and calculating similarity, and means for reporting the results of the reliability evaluation. This makes it possible to quickly and accurately evaluate the generative feasibility of the data generation AI and the risk of copyright infringement.

[0660] "Documents" refer to documents or files that contain information or data.

[0661] "Textual information" refers to information composed of characters and symbols contained in a document.

[0662] "Analysis" refers to the process of thoroughly examining textual information to understand its structure and meaning.

[0663] "Preprocessing" is the process of removing unnecessary information from textual data and standardizing it to make it easier to analyze.

[0664] "Writing style" refers to the characteristics related to the way a text is expressed and its style.

[0665] "Vocabulary selection" refers to the characteristics related to the choice of words and phrases used in a text.

[0666] "Generability" is an assessment indicating the likelihood that the subject material was created by a generative AI.

[0667] "Existing data" refers to information and data that have been stored in the past and is used as a basis for comparison.

[0668] "Similarity" is a numerical representation of how similar the content of a document is to other existing data.

[0669] "Reliability assessment" refers to the process of determining and reporting whether a document is original or potentially infringes on copyright.

[0670] This invention provides a specific system for evaluating the reliability of data. It describes how hardware and software are used to process and compute the data.

[0671] Users upload the documents to be evaluated to the system using their own devices. The documents are assumed to be in common document formats such as PDF or Word. The server receives the documents and extracts the text information. Various character recognition software, including OCR technology, is used for text extraction.

[0672] Next, the server preprocesses the extracted text information. This preprocessing removes unnecessary line breaks and special characters from the text and standardizes the content. This process is characterized by string manipulation using regular expressions and automated filtering using scripts.

[0673] The processed text information is sent to an AI agent on the server. The AI ​​agent uses natural language processing techniques to analyze the writing style and vocabulary selection of the document. This analysis assesses the likelihood that the document was created by a generative AI model. The AI ​​agent utilizes statistical models and machine learning algorithms to perform a highly accurate assessment.

[0674] Furthermore, the server compares the textual information of the document with existing databases. Vector space models and data mining techniques are used in the process of calculating similarity. This quantifies the degree to which the document is similar to existing literature and data, and assesses the likelihood of copyright infringement.

[0675] Ultimately, the terminal receives the evaluation results from the server and provides the user with information in the form of a detailed report. Based on this report, the user can check the reliability and areas for improvement of the material and revise it as needed.

[0676] As a concrete example, consider a scenario where an undergraduate student submits a research report. When the student uploads the report to the system, the server uses OCR technology to transcribe it into text and analyzes whether it was likely created by a generative AI. It also evaluates the similarity by comparing it to previously submitted reports and publicly available academic literature. This allows the university to verify the originality and reliability of the report and provide feedback to the student.

[0677] An example of a prompt is: "Evaluate whether the following research report is similar to other documents and whether it was likely created by a generative AI."

[0678] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0679] Step 1:

[0680] Users upload documents to be evaluated to the system using their terminals. These documents are in common formats such as PDF and Word. The input is the document file submitted by the user. The output is the server receiving and saving the relevant documents. The server stores the user's documents for future analysis.

[0681] Step 2:

[0682] The server extracts text information from the uploaded documents. This process utilizes OCR technology, converting image formats into text. The input is the document file received by the server. The output provides analyzable text information. This extraction allows the text within the document to be treated as digital data.

[0683] Step 3:

[0684] The server preprocesses the extracted character information. It removes unnecessary line breaks and special characters and standardizes the text. The input is the initial extracted character information. The output is clean, standardized text data. This processing improves the accuracy of subsequent analysis.

[0685] Step 4:

[0686] The server passes pre-processed text to the AI ​​agent, which evaluates the likelihood that the document was created by a generative AI model. The input is standardized text data. The output provides an evaluation of the likelihood that the document was created by a generative AI. The AI ​​agent uses statistical models and machine learning algorithms to make highly accurate judgments.

[0687] Step 5:

[0688] The server compares the textual information of the document with existing content stored in the database and calculates the similarity. The input is standardized text data. The output provides the similarity evaluation result. This process determines how similar the document is to other documents.

[0689] Step 6:

[0690] The terminal receives evaluation results from the server and provides them to the user in the form of a detailed report. The input is evaluation result data sent from the server. The output is a detailed reliability evaluation report presented to the user. The user can use this report to improve or resubmit their materials.

[0691] (Application Example 1)

[0692] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0693] Modern educational institutions are required to quickly and accurately assess the reliability of student reports and papers. However, a related problem is the difficulty in effectively detecting similarities between artificially generated texts and existing literature. Such challenges hinder the maintenance of educational quality.

[0694] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0695] In this invention, the server includes means for receiving data information, means for extracting textual information from the received data information and performing preprocessing for analysis, means for analyzing the style and terminology selection of the preprocessed textual information and evaluating the generated ratio, means for comparing the textual information with other existing data sets and calculating similarity, means for outputting the generated ratio and the evaluation result of the similarity, and means for notifying the recipient of the evaluation result in data format. This enables a rapid and accurate evaluation of the reliability of submitted data.

[0696] "Document information" refers to data that shows the contents of submitted documents and papers.

[0697] "Textual information" refers to text data extracted from source material, and it forms the basis for analysis.

[0698] "Preprocessing" refers to the process of removing unnecessary line breaks and special characters from received text information in order to analyze it.

[0699] "Style and terminology selection" refers to the unique writing style and vocabulary used in a particular text.

[0700] "Generated rate" is a numerical value that indicates the degree to which the source information was likely created by artificial intelligence.

[0701] "Similarity" is a measure used to evaluate how similar textual information is to an existing data set.

[0702] "Evaluation results" refer to the overall analysis results output based on the analysis of the generated proportions and similarity.

[0703] "Data format" refers to the reporting format in which evaluation results are provided to recipients.

[0704] "Recipient" refers to the individual or organization that receives the evaluation results.

[0705] This system is designed to evaluate the reliability of reports submitted by educational institutions. The server receives document information using a cloud platform. The received document information is converted into the necessary textual information by an internal character information extraction module. In this process, SpaCy, a natural language processing tool, is used to remove unnecessary symbols and line breaks, creating a parseable data format.

[0706] Next, the server analyzes the pre-processed text information using machine learning algorithms, specifically TensorFlow or PyTorch, to determine the style and terminology selection, and calculates the likelihood that the material is artificially generated. Furthermore, the text information is compared with existing data sets, and a similarity score is calculated. A text comparison algorithm is used for this matching to accurately assess how similar the material is to other documents.

[0707] The evaluation results are provided to the user in the form of a detailed report. The user's device can visually review this report, allowing them to identify areas for improvement and assess the reliability of the materials. This application primarily operates on smartphones and connects with a cloud server to provide real-time evaluation results.

[0708] As a concrete example, suppose a student uploads an environmental studies report. The server then evaluates the received report, assessing the likelihood that its content was generated by a generative AI model (15%) and determining its similarity to existing academic literature (10%). Such evaluations serve as important indicators for educational institutions to provide appropriate feedback to students.

[0709] An example of a prompt might be: "Check if this report is similar to other literature and assess whether it was created by a generative AI."

[0710] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0711] Step 1:

[0712] The user uploads document information to the server through the application. The input data is document information obtained from the user's local device and sent to the server. The output is the received document recorded on the server side.

[0713] Step 2:

[0714] The server extracts textual information from the received document information. During this process, SpaCy, a natural language processing tool, is used to remove unnecessary symbols and line breaks, generating parseable text data. The input is document information, and the output is pre-processed textual information.

[0715] Step 3:

[0716] The server applies machine learning algorithms to analyze pre-processed text information. Here, TensorFlow or PyTorch is used to analyze writing style and terminology selection, and the likelihood that the material was generated by a generative AI model is calculated. The input is pre-processed text information, and the output is a numerical evaluation of the generative likelihood.

[0717] Step 4:

[0718] The server compares the textual information with an existing data set. A text comparison algorithm is used to calculate similarity, evaluating how similar the document is to other documents. The input is pre-processed textual information, and the output is a similarity score.

[0719] Step 5:

[0720] The server integrates the results of the generative feasibility and similarity evaluations and sends them to the user terminal in the form of a detailed report. The input is the evaluation value, and the output is the integrated report data.

[0721] Step 6:

[0722] The user terminal visually displays the received report and provides the user with areas for improvement and reliability indicators for the document. The input is integrated report data, and the output is visual feedback provided to the user.

[0723] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0724] This invention relates to a system that combines a function to evaluate the likelihood and similarity of materials submitted by users using a generation AI with an emotion engine. This system is primarily intended for use in businesses and educational institutions, and aims to provide more appropriate feedback by taking user emotions into consideration.

[0725] First, the user uploads the document to be evaluated to the system. The server receives this document and extracts text from the file. The server then preprocesses the extracted text, standardizing its content and preparing it for analysis.

[0726] Next, the server passes the pre-processed text to the AI ​​agent, which begins analyzing the writing style and terminology. The AI ​​agent detects and evaluates features that indicate the potential of generative AI. Statistical models and machine learning algorithms are used in this process.

[0727] Furthermore, the server compares the text to existing content in the database and calculates a similarity score. This evaluates how similar the text is to other documents and analyzes the risk of copyright infringement.

[0728] In this system, the emotion engine plays a crucial role. The terminal analyzes user interaction data and recognizes the user's emotional state from it. For example, it can determine emotions from the user's input speed and word choice when operating the system.

[0729] Based on the perceived emotions, the server adjusts the content and presentation format of the evaluation results. For example, if the system determines that the user is disappointed, it can present the evaluation results in a way that includes more thoughtful support.

[0730] A concrete example of this system is a job seeker. When a job seeker uploads their resume, the system evaluates the potential of AI generation and further uses an emotion engine to understand the job seeker's feelings. If the job seeker shows signs of nervousness, the device provides the evaluation results along with an encouraging message, making it easier for the user to understand areas for improvement.

[0731] By combining emotion engines in this way, it is possible to provide services that are more tailored to individual users and improve the effectiveness of the system.

[0732] The following describes the processing flow.

[0733] Step 1:

[0734] Users upload the document files they wish to have evaluated to the system. The server receives these documents and stores them securely. This process also includes verifying that the files are in the correct format.

[0735] Step 2:

[0736] The server extracts text data from the documents. Even if the documents are in PDF, Word, or other formats, it uses appropriate tools to extract the text and prepare it for analysis.

[0737] Step 3:

[0738] The server performs preprocessing on the extracted text. This includes removing unnecessary line breaks and spaces, and standardizing special characters. This makes the text more suitable for analysis.

[0739] Step 4:

[0740] The server sends pre-processed text to the AI ​​agent. The AI ​​agent analyzes the writing style and terminology selection and evaluates the feasibility of generating the text using generative AI. Statistical models and machine learning algorithms are used in this process.

[0741] Step 5:

[0742] In parallel, the server compares the document text with existing data in the database and calculates similarity. It then calculates a similarity score and analyzes the risk of copyright infringement.

[0743] Step 6:

[0744] The device uses user interaction data to activate an emotion engine. The emotion engine analyzes data such as the user's input speed and word choice to recognize the user's emotional state.

[0745] Step 7:

[0746] The server determines how to present the evaluation results based on the user's emotions recognized by the emotion engine. For example, if the user appears anxious, the results will be presented in a calm tone, including advice.

[0747] Step 8:

[0748] The device provides the user with a report containing the adjusted evaluation results. Based on this report, the user can consider areas for improvement in their materials and the possibility of resubmission.

[0749] (Example 2)

[0750] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0751] Conventional document evaluation systems could assess the likelihood of document generation and similarity using AI, but they lacked the means to provide feedback that took user emotions into consideration. This resulted in problems such as evaluation results causing psychological burden on users and preventing them from taking appropriate corrective actions.

[0752] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0753] In this invention, the server includes means for receiving data, means for extracting and pre-processing text, means for analyzing the style and terminology of the text and evaluating its generative potential, means for comparing the text with existing data and calculating similarity, and means for analyzing emotions from user operation data and adjusting the presentation format of the evaluation results. This enables feedback tailored to the user's emotional state, reduces psychological burden, and promotes appropriate improvement actions.

[0754] "Documents" refers to the documents uploaded to the system for evaluation. This typically includes content provided as PDF or Word files.

[0755] "Text" refers to string data extracted from the document. This allows the content to be analyzed mechanically.

[0756] "Preprocessing" refers to the process of standardizing and formatting the extracted text in preparation for analysis. This includes removing unnecessary characters and unifying the format.

[0757] "Analysis" is a general term for technical operations performed to evaluate the content of text. This includes the detection of style and terminology, and the evaluation of generative potential.

[0758] "Generability" refers to characteristics that indicate the likelihood that the material was generated by a generative AI. This includes evaluation results based on stylistic and terminology patterns.

[0759] "Similarity" is a measure that indicates the degree to which text shares commonalities with existing data. This allows for an objective evaluation of similarity with other materials.

[0760] "Emotion" refers to the psychological state inferred from the user's operation data. This includes evaluation results based on input speed and operation history.

[0761] "Feedback" refers to the information that the system provides to the user based on its analysis results. This information is adjusted according to the user's emotional state.

[0762] This invention realizes a system that analyzes the material to be evaluated, assesses the possibility and similarity of generation by the generation AI, and provides feedback while taking into account the user's emotions. It is primarily intended for use in educational institutions and companies, and aims to analyze materials and provide psychological support to users.

[0763] Users first upload the documents to be evaluated to the system. This operation is performed via a web interface on their terminal, and the data is provided in PDF or Word file format. The server receives the documents and extracts the text data. This utilizes PDF parsing and OCR (optical character recognition) technology.

[0764] Next, the server performs preprocessing on the extracted text. This preprocessing removes unnecessary spaces and special characters, and formats the text, preparing it for analysis. The server then passes the preprocessed text to an AI agent, which uses a generative AI model to analyze the writing style and terminology selection. This AI agent combines natural language processing and machine learning algorithms to evaluate the potential for generation.

[0765] The server also compares the text against existing data in the database to assess its similarity. Numerical methods, such as cosine similarity, are used to calculate a similarity score. This allows for an assessment of the text's similarities to other materials and the associated copyright risks.

[0766] A key feature of this system is that the emotion engine analyzes the user's operation data to infer their psychological state. The terminal collects data such as input speed and word choice to identify the emotional state. Based on this recognition, the server adjusts the format in which the evaluation results are presented. For example, if the user is feeling anxious, the feedback can be provided in an encouraging way.

[0767] As a concrete example, consider a case where a job seeker uploads their resume to the system and the AI ​​evaluates its potential for generation. In this case, if the emotion engine detects the job seeker's anxiety, the terminal can provide feedback such as, "This evaluation result is intended to show areas for improvement. Let's think positively about the next steps."

[0768] A concrete example of a prompt is the instruction, "Evaluate the AI-generated potential of the uploaded document and adjust the feedback according to the user's sentiment." By using this prompt, the system can achieve multi-layered evaluation and feedback.

[0769] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0770] Step 1:

[0771] Users upload documents to the system. PDF and Word files are used as input. The terminal checks the file format and prepares to send it to the server.

[0772] Step 2:

[0773] The server extracts text from the received documents. The input is the uploaded file, and the output is the extracted raw text data. This process utilizes OCR technology and PDF parser libraries.

[0774] Step 3:

[0775] The server preprocesses the extracted text. The input is raw text data, and the output is formatted and standardized text. Unnecessary spaces and special characters are removed, and morphological analysis is performed to prepare the text for analysis.

[0776] Step 4:

[0777] The server passes pre-processed text to the AI ​​agent for evaluation of its generative potential. The input is formatted text, and the output is a generative potential evaluation score. The AI ​​agent analyzes the writing style and terminology selection using natural language processing models and machine learning algorithms.

[0778] Step 5:

[0779] The server calculates similarity by comparing the text with existing data. The input is pre-processed text, and the output is a similarity score. Vector comparison methods such as cosine similarity are used in this process to evaluate the similarity with other materials.

[0780] Step 6:

[0781] The device analyzes user interaction data to infer emotions. Inputs include user input speed and interaction history data, while output is an inferred emotion label. The device sends this data to an emotion engine for analysis of the user's psychological state.

[0782] Step 7:

[0783] The server adjusts the presentation format of the evaluation results based on the emotion label. The input is the evaluation results of generativeness and similarity, and the emotion label, and the output is an adjusted feedback message. For example, if the server determines that the user is stressed, the feedback will be provided in gentler language.

[0784] Step 8:

[0785] The device presents the user with refined feedback. The input is a refined feedback message sent from the server, and the output is the evaluation result and support message displayed on the user's screen. Information is provided to help the user understand areas for improvement and move forward to the next step.

[0786] (Application Example 2)

[0787] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0788] The problem that this invention aims to solve is how to provide feedback that takes into account the user's emotional state, rather than simply checking the possibility and similarity of materials submitted by the user using a generating AI. In particular, current systems are unable to reflect the user's psychological state, resulting in uniform evaluation results and a lack of individualized support.

[0789] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0790] In this invention, the server includes means for receiving data, means for extracting text from the received data and performing preprocessing for analysis, means for analyzing the style and terminology of the preprocessed text and evaluating the likelihood of generation, means for comparing the text with other existing data and calculating similarity, means for outputting the evaluation results of the likelihood and similarity of generation, and means for recognizing the user's emotional state and adjusting the evaluation results accordingly. This enables detailed feedback that responds to the user's emotional state.

[0791] "Documents" refer to the documents and data that users submit to the system for evaluation.

[0792] "Means of receiving" refers to the function for importing data from users into the server.

[0793] "Means for extraction and pre-processing for analysis" refers to a function that extracts necessary text data from a document and prepares it in a format suitable for analysis.

[0794] "Means for evaluating the likelihood of generation" refers to a function that analyzes whether preprocessed text was generated by AI and quantifies that likelihood.

[0795] "Methods for comparing with existing data and calculating similarity" refers to a function that compares the input text with existing data in the database and expresses the degree of similarity as a numerical value.

[0796] "Means for outputting evaluation results" refers to a function that provides the generated probability and similarity analysis results in a format that users can review.

[0797] "Means for recognizing emotional states and adjusting evaluation results" refers to a function that analyzes user interaction data to determine emotions and changes the content and method of feedback accordingly.

[0798] The system implementing this invention receives data from the user and provides feedback that takes into account the user's emotional state based on that data. The server first receives data submitted electronically by the user. The data is taken into the server as text data, and its contents are pre-processed for analysis. In the pre-processing, the text is standardized and prepared for evaluation.

[0799] Next, the server analyzes the pre-processed text and evaluates its potential for generation by the AI ​​based on its writing style and terminology. Statistical models and machine learning algorithms are used for this evaluation. The text is also compared with existing data in the database, and a similarity score is calculated.

[0800] One of the features of this invention is that the server recognizes the user's emotional state and reflects it in the evaluation results. The terminal analyzes the user's interaction data and estimates the emotional state from input speed and word choice. For example, if the user is confused, the server carefully explains the evaluation results and clearly communicates areas for improvement. As a result, the user can receive feedback tailored to their emotional state.

[0801] One concrete application example is when children studying at home use a device designed to support home learning. When a child submits an essay assignment, the device evaluates it using a generative AI model and also adds encouraging words, taking into account the child's emotional state. In this way, the user experience can be further improved.

[0802] An example of a prompt message is: "Based on the following evaluation results, please tell me what encouraging comments I should add if the user is confused: The evaluation results indicate a 50% probability of being AI-generated and a similarity score of 20%."

[0803] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0804] Step 1:

[0805] The server receives materials submitted electronically by users. The input is document data uploaded by the user, which the server receives and stores. The output is the material to be processed, which is stored within the server.

[0806] Step 2:

[0807] The server extracts text from received materials and performs preprocessing for analysis. The input is document data stored on the server, and the server extracts text data from the documents using a text extraction algorithm. Next, it performs normalization and removes unnecessary information, and outputs text ready for analysis.

[0808] Step 3:

[0809] The server analyzes pre-processed text using a generative AI model and evaluates the likelihood of it being generated. The input is pre-processed text data, and the server analyzes the style and terminology selection of this data using statistical models and machine learning algorithms. The output is a numerical value indicating the likelihood that the text was generated by a generative AI.

[0810] Step 4:

[0811] The server compares text with data in an existing database and calculates similarity. The input is pre-processed text data, and the server calculates similarity by comparing it with similar data in the database. The output is the similarity score obtained from the comparison.

[0812] Step 5:

[0813] The device analyzes user interaction data in real time to recognize emotional states. Input data includes user input speed and word choice, and the device analyzes this data to estimate the emotional state and output the emotion recognition result.

[0814] Step 6:

[0815] The server combines the generated probabilities, similarity scores, and the user's emotional state to refine and output feedback. The inputs are the generated probability values, similarity scores, and emotional recognition results, which the server integrates to generate feedback tailored to the user. The output is presented to the user as a refined evaluation result.

[0816] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0817] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0818] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0819] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0820] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0821] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0822] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0823] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0824] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0825] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0826] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0827] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0828] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0829] 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.

[0830] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0831] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0832] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0833] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0834] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0835] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0836] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0837] The following is further disclosed regarding the embodiments described above.

[0838] (Claim 1)

[0839] Means of receiving materials,

[0840] A means for extracting text from received materials and performing preprocessing for analysis,

[0841] A means for analyzing the style and terminology of preprocessed text and evaluating its potential generation,

[0842] A means for comparing text with other existing data and calculating similarity,

[0843] A means for outputting the generated probability and similarity evaluation results,

[0844] A system that includes this.

[0845] (Claim 2)

[0846] The system according to claim 1, which calculates the generated probabilities using a statistical model or machine learning algorithm.

[0847] (Claim 3)

[0848] The system according to claim 1, which outputs evaluation results in report format and provides them to the user.

[0849] "Example 1"

[0850] (Claim 1)

[0851] Means of receiving materials,

[0852] A means for extracting textual information from received materials and performing preprocessing for analysis,

[0853] A means for analyzing the style and vocabulary selection of preprocessed textual information and evaluating the likelihood of its generation,

[0854] A method for comparing textual information with other existing data and calculating similarity,

[0855] A means of reporting the results of reliability evaluation,

[0856] Information processing device including

[0857] (Claim 2)

[0858] The information processing apparatus according to claim 1, which calculates the generated probabilities using a computational model or machine learning method.

[0859] (Claim 3)

[0860] The information processing device according to claim 1, which provides the results of a reliability evaluation in a record format and reports them to the user.

[0861] "Application Example 1"

[0862] (Claim 1)

[0863] Means for receiving information and

[0864] A means for extracting textual information from received document information and performing preprocessing for analysis,

[0865] A means for analyzing the style and terminology selection of preprocessed text information and evaluating the proportion of generated text,

[0866] A means for comparing textual information with other existing data sets and calculating similarity,

[0867] A means for outputting the generated proportion and the evaluation result of similarity,

[0868] A means of notifying recipients of evaluation results in data format,

[0869] A system that includes this.

[0870] (Claim 2)

[0871] The system according to claim 1, wherein the generated proportion is calculated using a statistical method or a machine learning algorithm.

[0872] (Claim 3)

[0873] The system according to claim 1, which outputs evaluation results in an informational format and provides them to the user.

[0874] "Example 2 of combining an emotion engine"

[0875] (Claim 1)

[0876] Means of receiving materials,

[0877] A means for extracting text from received materials and performing preprocessing for analysis,

[0878] A means for analyzing the style and terminology of preprocessed text and evaluating its potential generation,

[0879] A means for comparing text with other existing data and calculating similarity,

[0880] A means of analyzing emotions from user operation data and adjusting the format of presentation of evaluation results,

[0881] A means for outputting the generated probability and similarity evaluation results,

[0882] A system that includes this.

[0883] (Claim 2)

[0884] The system according to claim 1, which calculates the generated probabilities using a statistical model or machine learning algorithm and includes the results of adjustments made by sentiment analysis.

[0885] (Claim 3)

[0886] The system according to claim 1, which outputs the evaluation results and adjustment results based on sentiment analysis in report format and provides them to the user.

[0887] "Application example 2 when combining with an emotional engine"

[0888] (Claim 1)

[0889] Means of receiving materials,

[0890] A means for extracting text from received materials and performing preprocessing for analysis,

[0891] A means for analyzing the style and terminology of preprocessed text and evaluating its potential generation,

[0892] A means for comparing text with other existing data and calculating similarity,

[0893] A means for outputting the generated probability and similarity evaluation results,

[0894] A means of recognizing the user's emotional state and adjusting the evaluation results based on that,

[0895] A system that includes this.

[0896] (Claim 2)

[0897] The system according to claim 1, which calculates generated probabilities using a statistical model or machine learning algorithm and detects emotions from user interaction data.

[0898] (Claim 3)

[0899] The system according to claim 1, which outputs evaluation results in report format and provides feedback according to the user's emotional state. [Explanation of symbols]

[0900] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. Means of receiving materials, A means for extracting text from received materials and performing preprocessing for analysis, A means for analyzing the style and terminology of preprocessed text and evaluating its potential generation, A means for comparing text with other existing data and calculating similarity, A means for outputting the generated probability and similarity evaluation results, A system that includes this.

2. The system according to claim 1, which calculates the generated probabilities using a statistical model or machine learning algorithm.

3. The system according to claim 1, which outputs evaluation results in report format and provides them to the user.

Images