System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving material; means for automatically analyzing the content of the material and assigning a classification tag; means for converting the analyzed content into a reusable format and storing it in a database; means for retrieving relevant information from the database for a question and generating an answer; and means for comparing different versions of the same material and identifying the latest version.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Knowledge management in companies faces challenges such as not knowing where documents are located, difficulty in identifying the latest documents, and difficulty in using the content of videos and slides with generative AI. Furthermore, while appropriate knowledge management and reuse are essential for generative AI responses, there is a lack of efficient ways to do this. To solve these problems, a system is needed that can consistently and automatically organize documents and videos, analyze knowledge, classify it, store it, update it, retrieve it, and respond to it. [Means for solving the problem]

[0005] The present invention provides a system that includes a means for receiving materials, a means for automatically analyzing the content of the materials and assigning classification tags, a means for converting the analyzed content into a reusable format and storing it in a database, a means for extracting relevant information from the database in response to a question and generating an answer, and a means for comparing different versions of the same material and identifying the latest version. This reduces the complexity of knowledge management within a company and enables the construction of a knowledge base that can be effectively used by generative AI. The system of the present invention also includes a means for automatically classifying and organizing materials from multiple sources and updating the database based on the classification, as well as a communication means for providing generated answers to users' devices, thereby achieving greater convenience and efficiency.

[0006] "Materials" refers to any form of information that contains corporate knowledge, such as videos, documents, and presentation files.

[0007] The "receiving means" refers to an interface through which users upload materials to the system and the process of taking the materials into the server through that interface.

[0008] "Means of analysis" refers to the process of using generative AI to automatically analyze the content of documents and videos and extract text and key points.

[0009] "Tagging" is the process of automatically tagging analyzed material with tags that indicate content-based categories or attributes.

[0010] "Means of converting into a reusable format" refers to the process of converting the analyzed information into a format that can be used repeatedly later, such as text data or structured data.

[0011] "Means for storing in a database" refers to the process of storing the parsed and transformed information in a structured database.

[0012] Retrieval is the process of searching for and extracting specific information or materials from a database.

[0013] "Means of generating answers" is the process of using generative AI to create appropriate responses to users' questions.

[0014] "Version comparison measures" are the process of analyzing different versions of the same material and identifying differences in content.

[0015] "Latest version identification" is the process of identifying the most recent of different versions of the same material and marking it in the database.

[0016] "Communication means" is a network communication interface for transmitting generated answers and materials to the user's terminal. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8]FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0018] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0019] First, the terms used in the following description will be explained.

[0020] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0021] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0022] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0023] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0024] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0028] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0029] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0030] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0031] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0032] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0035] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0036] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0037] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0038] The present invention is a system that efficiently manages materials and converts that knowledge into a reusable form using generative AI. A specific embodiment of the system and the processing flow are described below.

[0039] System configuration

[0040] 1. Uploading materials

[0041] Users access the Knowledge Mate interface using their devices and upload documents and video files, including document files (PDF, Word), presentation files (PPT), and video files (MP4).

[0042] 2. Receiving the file

[0043] The server receives the uploaded file, identifies the file type (document, presentation, video, etc.), and prepares it for appropriate processing.

[0044] 3. Automatic organization and tagging

[0045] The server automatically analyzes the content of the files and assigns relevant classification tags, such as marketing materials, technical documentation, annual reports, etc. This process organizes the materials into categories.

[0046] 4. Content Analysis and Transformation

[0047] The server uses the generated AI to analyze the contents of the uploaded file. For documents and presentations, elements such as text, images, and graphs are extracted. For video files, the audio is converted to text and the video content is analyzed to extract key points.

[0048] 5. Saving to the database

[0049] The server then formats the parsed and converted content into a reusable format (e.g., text data, JSON data, etc.) and stores it in a database, which serves as the foundation for future knowledge search and reuse.

[0050] 6. Knowledge Search and Retrieval

[0051] The user uses a device to input a question to the generation AI, for example, "Please show me materials related to last year's marketing strategy."

[0052] The server analyzes the user's question using generative AI and searches for relevant knowledge in the database using RAG (Retrieval-Augmented Generation).

[0053] 7. Generating Answers to Questions

[0054] Based on the knowledge extracted by the generative AI, it generates appropriate answers to user questions, such as "Last year's marketing strategy materials can be found here."

[0055] The server sends the generated response to the user's terminal, allowing the user to access the information.

[0056] 8. Versioning and Identifying the Latest Version

[0057] The server searches the database for different versions of the same document and uses generative AI to analyze the differences in content.

[0058] The server uses the analysis to identify the most up-to-date version and marks it in the database, which is then served whenever a user requests the most up-to-date document.

[0059] Specific examples

[0060] Example 1: Uploading and analyzing data

[0061] 1. User Action:

[0062] To upload "new presentation materials," a user logs into the Knowledge Mate portal site, selects the file, and then presses the upload button.

[0063] 2. Server processing:

[0064] The server receives the file, identifies its format, and, if it is a presentation, analyzes the slides and extracts text and image information.

[0065] The server assigns tags such as "Presentation Materials, Category: Marketing, Date: October 2023" and stores it in a database.

[0066] Example 2: Extracting knowledge using generative AI

[0067] 1. User Action:

[0068] The user inputs a question into the generation AI from their device, asking, "Do you have any materials related to last year's marketing strategy?"

[0069] 2. Server processing:

[0070] The server receives the user's question, analyzes it using generative AI, and uses RAG to search for relevant marketing strategy materials in the database.

[0071] The server identifies the relevant materials, and the generation AI generates a response such as, "We found last year's marketing strategy materials. This is the latest version."

[0072] The server sends this response to the user, who can then view the information on their terminal.

[0073] This invention is a system that significantly improves the efficiency of knowledge management and realizes effective use of knowledge within a company. This automates the organization, analysis, search, and version management of documents, making it easier for users to utilize knowledge.

[0074] The processing flow will be explained below.

[0075] Step 1:

[0076] A user logs in to the Knowledge Mate interface and clicks the upload button. The user selects and uploads a document file (PDF, Word), a presentation file (PPT), or a video file (MP4).

[0077] Step 2:

[0078] The server receives the uploaded file and identifies the file format, determining whether it is a document, presentation, video, etc.

[0079] Step 3:

[0080] The server automatically assigns classification tags to uploaded files, identifying the category to which the file belongs, such as "marketing materials," "technical documentation," or "annual reports."

[0081] Step 4:

[0082] The server sends the file contents to the generative AI engine. If it's a document or presentation, the generative AI extracts elements like text, images, and graphs. If it's a video file, it converts the audio into text and analyzes the video to extract key points.

[0083] Step 5:

[0084] The server formats the parsed and converted content into a reusable format (text data, JSON data, etc.) and saves the formatted data in a database.

[0085] Step 6:

[0086] The user uses a device to input a question to the generation AI, for example, "Please show me materials related to last year's marketing strategy."

[0087] Step 7:

[0088] The server sends the user's question to the generation AI, which uses Retrieval-Augmented Generation (RAG) to search for relevant knowledge in the database.

[0089] Step 8:

[0090] Generative AI generates answers based on search results, such as "Here are last year's marketing strategy documents."

[0091] Step 9:

[0092] The server sends the generated response to the user's terminal, allowing the user to access the information.

[0093] Step 10:

[0094] The server searches the database for different versions of the same document and uses generative AI to analyze the differences in content. Based on the analysis results, it identifies the most recent version with the most up-to-date content and marks it in the database. This latest version is always provided when a user requests the document.

[0095] Example 1

[0096] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0097] Current knowledge management systems require a great deal of time and effort to manually classify and organize documents and search and retrieve knowledge. Furthermore, when managing different versions of the same document, it is difficult to identify the latest version. This makes it difficult for users to quickly and accurately obtain the information they need.

[0098] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0099] In this invention, the server includes: means for users to upload materials using a computing device; means for receiving the materials and identifying their format; means for analyzing the file content and automatically assigning classification tags; means for converting the analyzed content into a reusable format and storing it in a database; means for users to input questions in natural language; means for using a generative AI model to extract relevant information from a database and generate an answer to the question; means for transmitting the generated answer to the user's computing device; and means for comparing different versions of the same material and identifying the latest version. This facilitates automatic organization, analysis, search, and version management of materials, enabling users to efficiently obtain the knowledge they need.

[0100] "User" means an individual or organization that uses the system to upload materials or search for knowledge.

[0101] A "computing device" is a device such as a calculator, PC, smartphone, or tablet that a user uses to upload materials or search for knowledge.

[0102] "Materials" refers to a collection of information, such as documents, presentations, and video files, that can be managed, analyzed, and searched.

[0103] A "generative AI model" is an artificial intelligence framework that uses deep learning technology to perform natural language processing and data analysis to generate and provide necessary information.

[0104] "Server" means a computer system that receives, analyzes, and stores uploaded materials and provides information in response to user requests.

[0105] "Means for identifying formats" refers to a function that automatically detects the file format of uploaded materials and performs appropriate processing.

[0106] "Tags" are keywords or labels that are assigned to indicate the classification or characteristics of materials.

[0107] "Knowledge" is a set of knowledge or information extracted from information stored in a database and provided in response to a user's question.

[0108] A "database" is a digital information repository for storing and managing analyzed materials and generated knowledge.

[0109] "Version control" is the process of distinguishing between different versions of the same material and identifying and managing the most recent version.

[0110] The present invention provides a system for efficiently managing materials and enabling knowledge reuse by generative AI. Specific embodiments of the system are described below.

[0111] Uploading materials

[0112] A user accesses the KnowledgeMate interface using a computing device, selects materials such as documents (PDF, Word), presentations (PPT), and video files (MP4), and presses the upload button, which sends the materials to the server.

[0113] Receiving files

[0114] The server receives the uploaded materials from the user and identifies the file format, for example, the server checks the file extension and automatically classifies the format as ".pdf", ".docx", ".pptx", ".mp4", etc.

[0115] Auto-organization and tagging

[0116] Once the file is received, the server automatically analyzes the file contents using natural language processing (NLP) technology, such as PyMuPDF for extracting text from PDF files, python-docx for extracting text from Word files, and python-pptx for extracting text from presentation files. Based on the analysis results, the server then assigns classification tags, such as marketing materials, technical documents, and annual reports, to make it easier to organize the documents.

[0117] Content analysis and transformation

[0118] The server uses generative AI models to further analyze the content of the uploaded files. For text files, topic modeling techniques are used to extract important keywords. For video files, a speech recognition engine (e.g., Google® Cloud Speech-to-Text) is used to convert the audio into text and identify important scenes and key points.

[0119] Saving to a database

[0120] The parsed and converted content is formatted into a reusable format (e.g., JSON, text data). The server creates a JSON object with fields such as "Document ID," "Title," "Content," and "Tags" as text data and stores it in a database. This database is built using, for example, "MongoDB" or "PostgreSQL."

[0121] Knowledge Search and Retrieval

[0122] A user inputs a prompt sentence into the generative AI model using a terminal. For example, if a request is made such as "Please tell me about materials related to last year's marketing strategy," the server receives the prompt sentence and analyzes it using the generative AI model. The server then uses Retrieval-Augmented Generation (RAG) technology to search for related knowledge in the database.

[0123] The generative AI model generates an appropriate answer to the user's question, such as "Here is last year's marketing strategy document," and the server sends this to the user's device.

[0124] Versioning and identifying the latest version

[0125] The server searches the database for different versions of the same document and uses a generative AI model to analyze the differences in content, particularly identifying differences and added information, and marks the version with the most recent content in the database, ensuring that the latest version is always served when a user requests the latest version.

[0126] Specific examples

[0127] Uploading and analyzing data

[0128] 1. User operation: To upload "new presentation materials," the user logs in to the Knowledge Mate portal site, selects the file "presentation_marketing_2023.pptx," and presses the upload button.

[0129] 2. Server processing: The server receives "presentation_marketing_2023.pptx" and identifies its format. The server uses the "python-pptx" library to analyze the slide content and extract text and image information. The server adds tags such as "Presentation Materials, Category: Marketing, Date: October 2023" and saves the analysis results in a database.

[0130] Extracting knowledge with generative AI

[0131] 1. User interaction: The user inputs a prompt into the generation AI from their device, asking, "Please tell me about materials related to last year's marketing strategy."

[0132] 2. Server processing: The server receives the user's prompt and uses a generative AI (e.g., OpenAI® GPT-4®) to analyze the prompt. It uses RAG technology to search for relevant marketing strategy materials in a database and generates an answer from the found materials, such as "We found last year's marketing strategy materials. This is the latest version." The server then sends this answer to the user, allowing them to view the materials on their device.

[0133] This allows users to efficiently obtain the knowledge they need, and makes it easier to manage materials and reuse information.

[0134] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0135] Step 1:

[0136] User operations

[0137] A user accesses the KnowledgeMate interface using a computing device, presses the upload button, and selects materials. The input materials can be documents (PDF, Word), presentations (PPT), video files (MP4), etc. The output is an upload request that is sent to the server.

[0138] Step 2:

[0139] Server Processing

[0140] The server processes the received upload request and receives the document. At this time, the server identifies the format of the document. It checks the extension of the received file and recognizes the format from ".pdf", ".docx", ".pptx", ".mp4", etc. The output is the format information of the document.

[0141] Step 3:

[0142] Server Processing

[0143] The server analyzes the file contents and automatically assigns classification tags. This process uses natural language processing (NLP) techniques and specific libraries (e.g., "PyMuPDF," "python-docx," "python-pptx"). The input is the received document itself, and the output is the analyzed text data and corresponding tag information. For example, in the case of a PDF file, the server generates tags such as "marketing material" and "technical document."

[0144] Step 4:

[0145] Server Processing

[0146] The server uses the generated AI model to further analyze and convert the content of the material. This process involves topic modeling and keyword extraction for text data, and in the case of video files, converting the audio into text using a speech recognition engine. The input is the analyzed text data, and the output is data formatted in a reusable format (e.g., JSON, text data). In the case of video files, the server extracts important scenes and key points.

[0147] Step 5:

[0148] Server Processing

[0149] The parsed and converted data is saved to a database. In this process, the data is stored in a reusable format in a database (such as MongoDB or PostgreSQL). The input is the formatted data, and the output is a response confirming that it has been saved to the database. The server creates a JSON object with fields such as "Document ID", "Title", "Content", and "Tags" that hold the text data.

[0150] Step 6:

[0151] User operations

[0152] A user inputs a question in natural language to the generative AI model using a terminal. For example, they input a prompt such as, "Please tell me about materials related to last year's marketing strategy." The input is the user's question, and the output is the generated prompt sentence.

[0153] Step 7:

[0154] Server Processing

[0155] The server uses a generative AI model to analyze the prompt and retrieve relevant information from the database. It then uses RAG technology to search for relevant knowledge in the database. The input is the prompt and the knowledge in the database, and the output is the retrieved relevant material.

[0156] Step 8:

[0157] Server Processing

[0158] The generative AI model generates an appropriate answer to the user's question. For example, it creates an answer text such as "Last year's marketing strategy materials are here." The input is the searched materials, and the output is the generated answer. The server sends this answer to the user's device, where it can be viewed by the user.

[0159] Step 9:

[0160] Server Processing

[0161] Different versions of the same document are searched for in the database and a generative AI model is used to analyze the differences in content. The input is the different versions of the document, and the output is a comparison of the old and new content based on that. The server then identifies the latest version and marks it in the database. Whenever a user requests the latest version, the latest version is always provided.

[0162] (Application example 1)

[0163] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0164] Currently, work knowledge and troubleshooting procedures within factories depend on the experience and knowledge of individual employees, making it difficult to share and reuse knowledge. Furthermore, in emergencies, appropriate information cannot be obtained quickly, which risks reducing the efficiency of the entire factory. Furthermore, when different versions of documents exist, it is time-consuming to identify which document is the latest version, which can lead to work being carried out based on incorrect information.

[0165] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0166] In this invention, the server includes means for receiving materials, means for automatically analyzing the content of the materials and assigning classification tags, means for converting the analyzed content into a reusable format and saving it in a database, means for retrieving related information from the database in response to a question and generating an answer using a generative AI model, means for comparing different versions of the same material and identifying the latest version, and means for managing work knowledge and troubleshooting procedures with an application installed in an automated device in a factory and providing quick answers based on prompts. This automates knowledge management within the factory and enables efficient and error-free information provision.

[0167] "Means for receiving materials" refers to the devices and functions that receive various materials from users and incorporate them into the system.

[0168] "Means for automatically analyzing the content of materials and assigning classification tags" refers to devices or functions that analyze the content of materials and automatically assign appropriate tags based on their characteristics and categories.

[0169] "Means for converting the analyzed content into a reusable format and storing it in a database" refers to a device or function that converts the content of the material into a reusable format (e.g., text data or JSON data) and stores it in a database.

[0170] "Means of retrieving relevant information from a database in response to a question and generating an answer using a generative AI model" refers to a device or function that searches for relevant information from a database in response to a user's question and generates an appropriate answer using a generative AI model.

[0171] "Means for comparing different versions of the same document and identifying the most recent version" refers to devices or functions that analyze and compare multiple versions of the same document and identify the most recent version.

[0172] "Applications installed on automated equipment in factories" refers to applications that are installed on automated equipment (e.g., robots) in factories and that manage work knowledge and troubleshooting procedures.

[0173] "Means for managing work knowledge and troubleshooting procedures and providing quick answers based on prompt statements" refers to devices and functions that organize and manage knowledge about factory work and problem-solving procedures, and generate quick answers based on user prompt statements.

[0174] This is a system for realizing applications for factory robots that efficiently manage work knowledge and troubleshooting procedures within factories and provide quick answers.

[0175] System configuration

[0176] 1. Technical configuration

[0177] The hardware used is a Linux (registered trademark)-based server (e.g., AWS (registered trademark) EC2, Google Cloud Compute Engine), Amazon S3 for file storage, and MongoDB for the database.

[0178] The software used is a web application server using Flask, various libraries for data analysis, and OpenAI GPT-3 (registered trademark) as a generative AI model.

[0179] 2. Uploading and analyzing materials

[0180] Users can upload documents via the factory robot's interface (e.g., screen or console). Document formats include PDF, Word, PPT, MP4, etc.

[0181] The server receives the uploaded materials, automatically analyzes their contents, and assigns appropriate classification tags, such as "factory maintenance" or "troubleshooting."

[0182] 3. Building a knowledge database

[0183] The server converts the parsed content into a reusable format (e.g., text data or JSON data) and stores it in a database, which will serve as the foundation for future knowledge searches.

[0184] If different versions of the same material are uploaded, the server automatically compares them and identifies the most recent version, marking it in the database.

[0185] 4. Question and Answer Generation

[0186] The user inputs a question to the factory robot, for example, using a prompt such as "Tell me about yesterday's maintenance procedures."

[0187] The server sends the question to a generative AI model (e.g., OpenAI GPT-3), which searches for relevant information in a database and generates an appropriate answer, which is then provided to the user through the factory robot's interface.

[0188] 5. Specific Examples

[0189] Uploading materials: A worker uploads a presentation (e.g., maintenance procedures) to the robot's interface. The system assigns a tag such as "Factory Maintenance" and stores it in a database.

[0190] Example of question-and-answering: A worker asks the robot, "What were the maintenance procedures yesterday?" The system searches the database for relevant documents and generates and provides an answer such as, "Here are the maintenance procedures yesterday."

[0191] Example prompt sentence:

[0192] Based on your knowledge database, generate an appropriate answer to the question, "What was the maintenance procedure yesterday?"

[0193] This system automates knowledge management within factories, facilitating the sharing and reuse of work knowledge and troubleshooting procedures. In addition, by utilizing generative AI models, it can quickly provide appropriate information in emergencies, improving factory efficiency and safety.

[0194] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0195] Step 1:

[0196] The user operates the factory robot's interface to upload a file (e.g., PDF, Word, PPT, MP4). The user presses the file selection button and selects the file they want to upload. This input sends the file to the server.

[0197] Step 2:

[0198] The server receives the uploaded material, identifies the format of the received material file, and prepares it for appropriate processing. Here, different analysis modules are selected depending on the file format (e.g., document file, presentation file, video file, etc.).

[0199] Step 3:

[0200] The server automatically analyzes the content of the file. For documents and presentations, it extracts text and image information. For video files, it converts audio to text and analyzes the video content to extract key points. The output of this process is the analyzed data.

[0201] Step 4:

[0202] The server automatically assigns relevant classification tags to the analyzed data, such as "factory maintenance" or "troubleshooting." This tagging allows the materials to be organized into categories.

[0203] Step 5:

[0204] The server analyzes and converts the tagged content into a reusable format (text data, JSON data, etc.), which is then stored in a database, making it easier to search for the material.

[0205] Step 6:

[0206] The user inputs a question into the factory robot's interface, for example, a prompt such as "Tell me about yesterday's maintenance procedures." This input is sent to the server.

[0207] Step 7:

[0208] The server sends the user's question to a generative AI model (e.g., OpenAI GPT-3), which analyzes the question. The server searches for relevant information in a knowledge database and uses the generative AI model to generate an appropriate answer. The output of this process is an answer to the user's question.

[0209] Step 8:

[0210] The server sends the generated answer to the user's device, which receives the answer and displays it to the user through the factory robot's interface. The user can refer to the displayed answer to proceed with the work.

[0211] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0212] The present invention is a system for managing, analyzing, and reusing materials, and recognizing user emotions. Specific embodiments for carrying out the present invention and the processing contents thereof will be described below.

[0213] System configuration

[0214] 1. Uploading materials

[0215] Users access the Knowledge Mate interface using their devices and upload documents and video files, including document files (PDF, Word), presentation files (PPT), and video files (MP4).

[0216] 2. Receiving the file

[0217] The server receives the uploaded file, identifies the file type (document, presentation, video, etc.), and prepares it for appropriate processing.

[0218] 3. Automatic organization and tagging

[0219] The server automatically analyzes the content of the files and assigns relevant classification tags, such as marketing materials, technical documentation, annual reports, etc. This process organizes the materials into categories.

[0220] 4. Content Analysis and Transformation

[0221] The server uses a generative AI engine to analyze the contents of the uploaded file. For documents and presentations, elements such as text, images, and graphs are extracted. For video files, the audio is converted to text and the video content is analyzed to extract key points.

[0222] 5. Saving to the database

[0223] The server then formats the parsed and converted content into a reusable format (e.g., text data, JSON data, etc.) and stores it in a database, which serves as the foundation for future knowledge search and reuse.

[0224] 6. User Emotion Recognition

[0225] The device sends the user's input or voice to the emotion engine, which analyzes it and recognizes the user's emotion (e.g., joy, anger, sadness, surprise, etc.).

[0226] 7. Knowledge Search and Retrieval

[0227] The user uses a device to input a question to the generation AI, for example, "Please show me materials related to last year's marketing strategy."

[0228] 8. Emotion-Based Response Modulation

[0229] The server sends the user's question to the generation AI, which then uses Retrieval-Augmented Generation (RAG) to search for relevant knowledge in the database. At the same time, the emotion engine recognizes the user's emotional state and adjusts the response accordingly.

[0230] 9. Generating Answers to Questions

[0231] Based on the knowledge extracted by the generative AI and the analysis results of the emotion engine, the system generates an appropriate answer to the user's question. For example, it creates a response such as, "Here are last year's marketing strategy documents. Please let me know if you have any additional questions."

[0232] 10. Versioning and Identifying the Latest Version

[0233] The server searches the database for different versions of the same document and uses generative AI to analyze the differences in content.

[0234] The server uses the analysis to identify the most up-to-date version and marks it in the database, which is then served whenever a user requests the most up-to-date document.

[0235] Specific examples

[0236] Example 1: Uploading and analyzing data

[0237] 1. User Action:

[0238] To upload "new presentation materials," a user logs into the Knowledge Mate portal site, selects the file, and then presses the upload button.

[0239] 2. Server processing:

[0240] The server receives the file, identifies its format, and, if it is a presentation, analyzes the slides and extracts text and image information.

[0241] The server assigns tags such as "Presentation Materials, Category: Marketing, Date: October 2023" and stores it in a database.

[0242] Example 2: Knowledge extraction using generative AI and emotion engines

[0243] 1. User Action:

[0244] The user inputs a question into the generation AI from their device, such as, "Do you have any materials related to last year's marketing strategy?" At the same time, the emotion engine recognizes the user's "confused" state from their tone of voice and text.

[0245] 2. Server processing:

[0246] The server receives the user's question, analyzes it using generative AI, and uses RAG to search for relevant marketing strategy materials in the database.

[0247] Based on the emotion of "confusion" recognized by the emotion engine, the generative AI generates a more polite and understandable response, such as, "Here are last year's marketing strategy documents. If you have any questions, please let us know as we have more detailed documents available."

[0248] 3. Response to the user:

[0249] The server sends this response to the user's terminal, allowing the user to access the information.

[0250] This invention is a system that significantly improves the efficiency of knowledge management and enables effective use of knowledge within a company. By incorporating an emotion engine, it provides appropriate responses that take into account the user's emotional state, creating a more user-friendly interface. This automates the organization, analysis, search, and version management of documents, making it even easier for users to utilize knowledge.

[0251] The processing flow will be explained below.

[0252] Step 1:

[0253] A user logs in to the Knowledge Mate interface using a terminal and clicks the upload button. The user selects and uploads a document file (e.g., PDF, Word), a presentation file (e.g., PPT), or a video file (e.g., MP4).

[0254] Step 2:

[0255] The server receives the uploaded file. The server identifies the file type (document, presentation, video, etc.) and prepares it for appropriate processing.

[0256] Step 3:

[0257] The server automatically assigns classification tags to files it receives, automatically determining which category a file belongs to (e.g., "marketing materials," "technical documentation," "annual reports," etc.) and assigning the tag.

[0258] Step 4:

[0259] The server uses a generated AI engine to analyze the contents of the uploaded file. For documents and presentations, elements such as text, images, and graphs are extracted. For video files, audio is converted to text and the video content is analyzed to extract key points.

[0260] Step 5:

[0261] The server formats the parsed content into a reusable format (text data, JSON data, etc.) and saves the formatted data in a database.

[0262] Step 6:

[0263] The user inputs a question into the generation AI from their device, for example, "Please show me materials related to last year's marketing strategy."

[0264] Step 7:

[0265] The server receives the user's question and sends it to the generation AI. At the same time, the device sends the emotion information from the user's tone of voice and input sentences to the emotion engine, which then analyzes the user's emotion (e.g., "confusion," "joy," "anger," etc.).

[0266] Step 8:

[0267] The generation AI uses Retrieval-Augmented Generation (RAG) to search for relevant knowledge in the database, and the server adjusts the tone and content of responses based on the analysis results from the emotion engine.

[0268] Step 9:

[0269] The generative AI generates appropriate answers to user questions based on the extracted knowledge and the analysis results of the emotion engine. For example, it creates a response such as, "Last year's marketing strategy materials are here. If you have any questions, please let us know as we have more detailed materials available."

[0270] Step 10:

[0271] The server sends the generated response to the user's terminal, allowing the user to access the information.

[0272] Step 11:

[0273] The server searches the database for different versions of the same document and uses generative AI to analyze the differences in content. Based on the analysis results, it identifies the most recent version with the most up-to-date content and marks it in the database. This latest version is always provided when a user requests the document.

[0274] Example 2

[0275] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0276] Conventional document management systems required a great deal of time and effort to organize and search documents, and it was difficult to provide appropriate responses that took user sentiment into account. As a result, documents could not be reused or information provided promptly, leading to problems that reduced work efficiency. Furthermore, the mixing of different versions of documents created the problem of cumbersome management of the latest versions.

[0277] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0278] In this invention, the server includes means for receiving materials, means for automatically analyzing the content of the materials and assigning classification tags, means for converting the analyzed content into a reusable format and storing it in a database, means for extracting relevant information from the database in response to a question and generating an answer using a generative AI model, means for comparing different versions of the same material and identifying the latest version, means for analyzing user sentiment and adjusting answers based on the analysis results, and means for converting voice to text and extracting key points from video when analyzing the content of the materials. This enables efficient material management, rapid information provision, user-friendly responses, and accurate provision of the latest materials.

[0279] "Means for receiving materials" refers to the function by which the server receives and stores materials uploaded by users, including identifying file formats and temporarily storing them.

[0280] "Means for automatically analyzing the content of materials and assigning classification tags" is a function that analyzes the content of materials and automatically assigns tags according to their characteristics and categories. This allows materials to be properly classified and managed.

[0281] "Means of converting the analyzed content into a reusable format and saving it in a database" refers to the function of formatting the information extracted from the materials into a format that is easy to reuse and saving that content in a database, making it easier to search and reuse in the future.

[0282] "Means of extracting relevant information from a database in response to a question and generating an answer using a generative AI model" refers to a function that searches for relevant information in a database in response to a question from a user and generates an appropriate answer using generative AI.

[0283] "Means for comparing different versions of the same document and identifying the latest version" refers to a function that compares different versions of the same document, analyzes the changes, and identifies the latest version. This allows the latest version to be provided at all times.

[0284] "Means for analyzing the user's emotions and adjusting responses based on the analysis results" refers to a function that analyzes the user's input and voice data to recognize their emotional state and generates a response based on that emotion, thereby providing the user with an appropriate and empathetic response.

[0285] "Means for converting audio to text and extracting key points from video when analyzing the content of materials" refers to a function that converts video and audio data into text and automatically extracts important points from the video content.

[0286] The present invention is a system for managing, analyzing, and reusing materials, and for recognizing user emotions. The system aims to improve user convenience by automatically organizing materials and generating appropriate responses. Specific embodiments for implementing the present invention are described below.

[0287] System configuration

[0288] Uploading materials

[0289] The user uploads documents using the terminal. Specifically, the user selects files such as PDF, Word, PPT, MP4, etc. through the web interface and presses the upload button. At this time, the terminal organizes the uploaded files into a specific format.

[0290] Receiving files

[0291] The server receives the uploaded files from the device, automatically identifies their format, analyzes the metadata of each file, and prepares them for temporary storage.

[0292] Auto-organization and tagging

[0293] The server automatically analyzes the content of the documents and assigns relevant classification tags, such as marketing materials, technical documents, annual reports, etc. This process is performed using a generative AI engine (e.g., GPT-4). The documents are organized into categories.

[0294] Content analysis and transformation

[0295] The server uses a generative AI engine to perform a detailed analysis of the uploaded file's contents. It extracts text, images, and graphs from document files, converts audio to text from video files, and analyzes the video content to extract key points. This information is then formatted for reuse.

[0296] Saving to a database

[0297] The server stores the parsed and converted content in a database in a reusable format (e.g. text data, JSON data), which serves as the basis for future knowledge search and reuse.

[0298] User Emotion Recognition

[0299] The device sends the user's input and voice to an emotion engine (e.g., EmotionAI) that analyzes the user's emotions. The emotion engine recognizes emotions such as joy, anger, sadness, and surprise.

[0300] Knowledge Search and Retrieval

[0301] The user uses a device to input a question to the generation AI, for example, "Please show me materials related to last year's marketing strategy."

[0302] Emotion-Based Response Modulation

[0303] The server receives the user's question and emotional information, and requests the generation AI to analyze it. It uses Retrieval-Augmented Generation (RAG) to search for relevant knowledge in the database. At the same time, it adjusts an appropriate response based on the analysis results of the emotion engine.

[0304] Generate answers to questions

[0305] Based on the knowledge extracted by the generative AI and the analysis results of the emotion engine, the system generates an appropriate answer to the user's question. For example, it creates a response such as, "Here are last year's marketing strategy documents. Please let me know if you have any additional questions."

[0306] Versioning and identifying the latest version

[0307] The server searches the database for different versions of the same document and uses generative AI to analyze the differences in content. Based on the analysis results, it identifies the most recent version with the most up-to-date content and marks it in the database. This latest version is always provided when a user requests the document.

[0308] Specific examples

[0309] Example 1: Uploading and analyzing data

[0310] 1. User operation: To upload "new presentation materials," the user logs in to the portal site, selects the file, and then presses the upload button.

[0311] 2. Server processing: The server receives the file, identifies its format, and if it is a presentation, analyzes the slides and extracts text and image information.

[0312] 3. Tagging: The server assigns tags such as "Presentation Materials, Category: Marketing, Date: October 2023" and stores them in the database.

[0313] Example 2: Knowledge extraction using generative AI and emotion engines

[0314] 1. User operation: The user inputs a question into the generation AI from their device, asking, "Do you have any materials about last year's marketing strategy?" At the same time, the emotion engine recognizes the user's "confused" state from their tone of voice and text.

[0315] 2. Server processing: The server receives the user's question, analyzes it using the generative AI, and uses RAG to search for relevant marketing strategy materials in the database.

[0316] 3. Emotional response: Based on the emotion of "confusion" recognized by the emotion engine, the generative AI generates a more polite and understandable response, such as, "Here are last year's marketing strategy documents. If you have any questions, please let us know as we have more detailed documents available."

[0317] 4. Reply to the user: The server sends this answer to the user's device, allowing the user to access the information.

[0318] Hardware and software used

[0319] Hardware: Servers, terminals (PCs, smartphones)

[0320] Software: Generative AI engine (e.g., GPT-4), emotion engine (e.g., EmotionAI), database management system (e.g., MySQL (registered trademark)), Knowledge Mate interface (web application)

[0321] Prompt Sentence Examples

[0322] 1. "Please find materials related to last year's marketing strategy."

[0323] 2. "Upload a new presentation."

[0324] 3. "I'm having trouble and need help. What resources should I look at?"

[0325] This system not only significantly improves the efficiency of knowledge management, but also realizes a user-friendly interface by providing responses that respond to the user's emotions. By automating the organization, analysis, search, and version management of materials, users can utilize knowledge more easily and effectively.

[0326] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0327] Step 1:

[0328] The user uploads the document using a terminal. Specifically, the user accesses the web interface, selects the file, and then presses the upload button. The input is a file in a format such as PDF, Word, PPT, or MP4, and the output is the file data sent to the server.

[0329] Step 2:

[0330] The server receives the file uploaded by the user. The server automatically identifies the file format and temporarily stores it. In this step, the input is the file data sent by the user, and the output is the identified file format and the temporarily stored file.

[0331] Step 3:

[0332] The server automatically analyzes the content of the received material and assigns classification tags. It uses a generative AI engine (e.g., GPT-4) to analyze the text, images, and other elements in the file. In this step, the input is the temporarily saved file data, and the output is the assigned classification tag (e.g., marketing material, technical document, etc.).

[0333] Step 4:

[0334] The server converts the content into a reusable format, for example extracting text data from a document or converting audio to text from a video. In this step, the input is the parsed content and the output is formatted data in a reusable format (e.g. text data, JSON data).

[0335] Step 5:

[0336] The server stores the data in a reusable format in a database. This process involves database manipulation using SQL queries. In this step, the input is the data in a reusable format and the output is the data stored in the database.

[0337] Step 6:

[0338] The device sends the user's input or voice to the emotion engine, which uses a model like EmotionAI. In this step, the input is the user's input text or voice data, and the output is the analyzed user's emotion (e.g., joy, anger, sadness).

[0339] Step 7:

[0340] The user inputs a question to the generative AI, for example, "Do you have any materials about last year's marketing strategy?" In this step, the input is the question text and the output is the request data sent to the server.

[0341] Step 8:

[0342] The server sends the question to the generation AI, which uses Retrieval-Augmented Generation (RAG) to search for relevant information in the database. It also adjusts the response based on information from the emotion engine. In this step, the input is the user's question and emotion information, and the output is the answer adjusted by the generation AI.

[0343] Step 9:

[0344] The generative AI generates an answer based on the extracted knowledge and sentiment analysis. Specifically, it generates a response such as, "Here are last year's marketing strategy documents. Please let me know if you have any additional questions." In this step, the input is information from the database and the results of sentiment analysis, and the output is the generated answer text.

[0345] Step 10:

[0346] The server compares different versions of the material and identifies the latest version. Generative AI is used to analyze the content of the different versions, identify the latest version with the most up-to-date content, and mark it in the database. In this step, the input is the material data of the different versions, and the output is the identified latest version of the material.

[0347] (Application example 2)

[0348] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0349] Conventional document management systems lack functionality such as navigation assistance and real-time analysis of the driver's emotional state, and technologies to improve driving safety and convenience are needed, especially for autonomous vehicles. There is also a need for technologies to properly grasp the user's emotional state while driving and provide appropriate feedback.

[0350] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving materials, means for automatically analyzing the content of the materials and assigning classification tags, means for converting the analyzed content into a reusable format and saving it in a database, means for retrieving related information from the database in response to a question and generating an answer, means for comparing different versions of the same material and identifying the latest version, means for uploading map data and destination information, means for analyzing the user's emotional state in real time, and means for providing navigation assistance information based on the analysis results. This makes it possible to appropriately grasp the driver's emotional state in an autonomous vehicle and provide feedback and warnings accordingly.

[0351] "Means for receiving materials" refers to devices or software that allow the server to receive materials in various formats (e.g., documents, presentations, video files) provided by users.

[0352] The "means for assigning classification tags" refers to a device or software that automatically analyzes the content of received materials and assigns tags indicating appropriate categories or attributes to the materials.

[0353] "Means for converting into a reusable format and storing in a database" refers to a device or software that converts the content of the analyzed material into a reusable format (e.g., text data, JSON data) and stores it in a database.

[0354] "Means for retrieving relevant information from a database in response to a question and generating an answer" refers to a device or software that searches a database for relevant information in response to a question from a user and generates an appropriate answer.

[0355] A "means for comparing different versions of the same material and identifying the most recent version" is a device or software that compares multiple versions of the same material and identifies the version with the most recent content.

[0356] The "means for uploading map data and destination information" refers to a device or software that uploads map data and destination information entered by the user to the system.

[0357] The "means for analyzing the user's emotional state in real time" refers to a device or software that uses a sensor, camera, microphone, etc. to analyze the user's emotional state in real time.

[0358] The "means for providing navigation assistance information" is a device or software that provides appropriate navigation assistance information (e.g., route directions, warning messages) to the driver based on the analyzed information and the user's emotional state.

[0359] To put the present invention into practice, a document management system and a navigation support system are integrated, as will be described in detail below.

[0360] System configuration

[0361] The system mainly consists of a server, a user terminal, smart glasses, and a head-mounted display (HMD).

[0362] User operations

[0363] Users input destination information using smart glasses or an HMD and upload it to the vehicle system, where it is received and analyzed by a server.

[0364] Server Processing

[0365] The server receives map data and destination information uploaded by users using a means for receiving data. Next, the server analyzes this data using a means for assigning classification tags, and assigns tags according to its characteristics. The analyzed content is then converted into a reusable format and managed using a means for storing it in a database. For example, tags such as "long-distance drive" and "intra-city travel" are assigned.

[0366] Real-time analytics

[0367] To analyze the user's emotional state, sensors, cameras, and microphones built into the smart glasses or HMD are used. This allows the user's emotional state to be transmitted to a server in real time for analysis. Depending on the emotional state, a means for providing navigation assistance information is activated, and appropriate feedback and warnings are provided to the user.

[0368] Navigation Aids

[0369] The server generates navigation assistance information based on map data and destination information. Based on the analysis results and the user's emotional state, specific driving instructions such as "Turn right at the next intersection" or emotional feedback such as "Drive calmly and without rushing" are displayed on the smart glasses or HMD.

[0370] Specific examples

[0371] For example, if a user inputs "Drive to Tokyo Station," the server receives this data and tags it as "Long-distance drive." At the same time, if the server detects an emotion of "impatience" from the user's tone of voice or facial expression, it displays emotional feedback such as "Please drive calmly."

[0372] Example prompts

[0373] Here is an example prompt:

[0374] Analyze the following map data and tag it appropriately:

[0375] "Destination: Tokyo Station, looking for a long drive"

[0376] Also, analyze the emotional state from the following audio data and provide appropriate feedback:

[0377] "You seem anxious."

[0378] Hardware and Software Examples

[0379] The hardware used includes smart glasses (e.g., Google Glass®), head-mounted displays (e.g., Microsoft® HoloLens®), and in-vehicle servers (equipped with high-performance processors).The software used includes Python scripts for document analysis and emotion recognition, and Tensorflow® or PyTorch for generative AI and emotion analysis.

[0380] As described above, the present invention integrates a document management system and a navigation assistance system, allowing users to receive real-time emotion recognition and navigation assistance while driving.

[0381] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0382] Step 1:

[0383] The user uses smart glasses or a head-mounted display (HMD) to input destination information and upload it to the vehicle system. The input data is in the form of, for example, "I would like to drive to Tokyo Station." This information is then sent to the server via the terminal.

[0384] Step 2:

[0385] The server analyzes the received map data and destination information. The input data is "I would like to drive to Tokyo Station." The server uses a generative AI model to analyze the input data and tag it (e.g., "long-distance drive"). The tagged data is obtained as output.

[0386] Step 3:

[0387] The server converts the parsed data into a reusable format (e.g. text data, JSON data) and stores it in the database. The tagged data as input is processed and converted into a format that can be stored in the database. The converted data as output is stored in the database.

[0388] Step 4:

[0389] The smart glasses and HMD use built-in sensors and microphones to analyze the user's emotional state in real time. Voice and facial expressions are sent as input data to the server, which then analyzes them using an emotion recognition algorithm. The analysis results (e.g., "anxiety") are obtained as output.

[0390] Step 5:

[0391] The server generates appropriate feedback or warnings based on the analysis results. It processes the emotion analysis results (e.g., "impatience") as input and generates a feedback message (e.g., "Please drive calmly"). The generated feedback is obtained as output.

[0392] Step 6:

[0393] The server provides navigation assistance information. Based on input data such as map data, destination information, and emotion analysis results, driving assistance information (e.g., "Please turn right at the next intersection") is generated. The output is displayed on the user's smart glasses or HMD.

[0394] Step 7:

[0395] The server stores the generated navigation assistance information and emotional feedback in a database. The server converts the generated information as input data into a reusable format and stores it in the database. The stored data is obtained as output.

[0396] Through the above steps, the present invention provides navigation assistance and emotional feedback based on the destination information entered by the user and the emotional state analyzed in real time, and this information is stored in a database for future analysis and improvement.

[0397] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0398] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0399] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0400] [Second embodiment]

[0401] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0402] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0403] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0404] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0405] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0406] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0407] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0408] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0409] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0410] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0411] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0412] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0413] The present invention is a system that efficiently manages materials and converts that knowledge into a reusable form using generative AI. A specific embodiment of the system and the processing flow are described below.

[0414] System configuration

[0415] 1. Uploading materials

[0416] Users access the Knowledge Mate interface using their devices and upload documents and video files, including document files (PDF, Word), presentation files (PPT), and video files (MP4).

[0417] 2. Receiving the file

[0418] The server receives the uploaded file, identifies the file type (document, presentation, video, etc.), and prepares it for appropriate processing.

[0419] 3. Automatic organization and tagging

[0420] The server automatically analyzes the content of the files and assigns relevant classification tags, such as marketing materials, technical documentation, annual reports, etc. This process organizes the materials into categories.

[0421] 4. Content Analysis and Transformation

[0422] The server uses the generated AI to analyze the contents of the uploaded file. For documents and presentations, elements such as text, images, and graphs are extracted. For video files, the audio is converted to text and the video content is analyzed to extract key points.

[0423] 5. Saving to the database

[0424] The server then formats the parsed and converted content into a reusable format (e.g., text data, JSON data, etc.) and stores it in a database, which serves as the foundation for future knowledge search and reuse.

[0425] 6. Knowledge Search and Retrieval

[0426] The user uses a device to input a question to the generation AI, for example, "Please show me materials related to last year's marketing strategy."

[0427] The server analyzes the user's question using generative AI and searches for relevant knowledge in the database using RAG (Retrieval-Augmented Generation).

[0428] 7. Generating Answers to Questions

[0429] Based on the knowledge extracted by the generative AI, it generates appropriate answers to user questions, such as "Last year's marketing strategy materials can be found here."

[0430] The server sends the generated response to the user's terminal, allowing the user to access the information.

[0431] 8. Versioning and Identifying the Latest Version

[0432] The server searches the database for different versions of the same document and uses generative AI to analyze the differences in content.

[0433] The server uses the analysis to identify the most up-to-date version and marks it in the database, which is then served whenever a user requests the most up-to-date document.

[0434] Specific examples

[0435] Example 1: Uploading and analyzing data

[0436] 1. User Action:

[0437] To upload "new presentation materials," a user logs into the Knowledge Mate portal site, selects the file, and then presses the upload button.

[0438] 2. Server processing:

[0439] The server receives the file, identifies its format, and, if it is a presentation, analyzes the slides and extracts text and image information.

[0440] The server assigns tags such as "Presentation Materials, Category: Marketing, Date: October 2023" and stores it in a database.

[0441] Example 2: Extracting knowledge using generative AI

[0442] 1. User Action:

[0443] The user inputs a question into the generation AI from their device, asking, "Do you have any materials related to last year's marketing strategy?"

[0444] 2. Server processing:

[0445] The server receives the user's question, analyzes it using generative AI, and uses RAG to search for relevant marketing strategy materials in the database.

[0446] The server identifies the relevant materials, and the generation AI generates a response such as, "We found last year's marketing strategy materials. This is the latest version."

[0447] The server sends this response to the user, who can then view the information on their terminal.

[0448] This invention is a system that significantly improves the efficiency of knowledge management and realizes effective use of knowledge within a company. This automates the organization, analysis, search, and version management of documents, making it easier for users to utilize knowledge.

[0449] The processing flow will be explained below.

[0450] Step 1:

[0451] A user logs in to the Knowledge Mate interface and clicks the upload button. The user selects and uploads a document file (PDF, Word), a presentation file (PPT), or a video file (MP4).

[0452] Step 2:

[0453] The server receives the uploaded file and identifies the file format, determining whether it is a document, presentation, video, etc.

[0454] Step 3:

[0455] The server automatically assigns classification tags to uploaded files, identifying the category to which the file belongs, such as "marketing materials," "technical documentation," or "annual reports."

[0456] Step 4:

[0457] The server sends the file contents to the generative AI engine. If it's a document or presentation, the generative AI extracts elements like text, images, and graphs. If it's a video file, it converts the audio into text and analyzes the video to extract key points.

[0458] Step 5:

[0459] The server formats the parsed and converted content into a reusable format (text data, JSON data, etc.) and saves the formatted data in a database.

[0460] Step 6:

[0461] The user uses a device to input a question to the generation AI, for example, "Please show me materials related to last year's marketing strategy."

[0462] Step 7:

[0463] The server sends the user's question to the generation AI, which uses Retrieval-Augmented Generation (RAG) to search for relevant knowledge in the database.

[0464] Step 8:

[0465] Generative AI generates answers based on search results, such as "Here are last year's marketing strategy documents."

[0466] Step 9:

[0467] The server sends the generated response to the user's terminal, allowing the user to access the information.

[0468] Step 10:

[0469] The server searches the database for different versions of the same document and uses generative AI to analyze the differences in content. Based on the analysis results, it identifies the most recent version with the most up-to-date content and marks it in the database. This latest version is always provided when a user requests the document.

[0470] Example 1

[0471] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0472] Current knowledge management systems require a great deal of time and effort to manually classify and organize documents and search and retrieve knowledge. Furthermore, when managing different versions of the same document, it is difficult to identify the latest version. This makes it difficult for users to quickly and accurately obtain the information they need.

[0473] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0474] In this invention, the server includes: means for users to upload materials using a computing device; means for receiving the materials and identifying their format; means for analyzing the file content and automatically assigning classification tags; means for converting the analyzed content into a reusable format and storing it in a database; means for users to input questions in natural language; means for using a generative AI model to extract relevant information from a database and generate an answer to the question; means for transmitting the generated answer to the user's computing device; and means for comparing different versions of the same material and identifying the latest version. This facilitates automatic organization, analysis, search, and version management of materials, enabling users to efficiently obtain the knowledge they need.

[0475] "User" means an individual or organization that uses the system to upload materials or search for knowledge.

[0476] A "computing device" is a device such as a calculator, PC, smartphone, or tablet that a user uses to upload materials or search for knowledge.

[0477] "Materials" refers to a collection of information, such as documents, presentations, and video files, that can be managed, analyzed, and searched.

[0478] A "generative AI model" is an artificial intelligence framework that uses deep learning technology to perform natural language processing and data analysis to generate and provide necessary information.

[0479] "Server" means a computer system that receives, analyzes, and stores uploaded materials and provides information in response to user requests.

[0480] "Means for identifying formats" refers to a function that automatically detects the file format of uploaded materials and performs appropriate processing.

[0481] "Tags" are keywords or labels that are assigned to indicate the classification or characteristics of materials.

[0482] "Knowledge" is a set of knowledge or information extracted from information stored in a database and provided in response to a user's question.

[0483] A "database" is a digital information repository for storing and managing analyzed materials and generated knowledge.

[0484] "Version control" is the process of distinguishing between different versions of the same material and identifying and managing the most recent version.

[0485] The present invention provides a system for efficiently managing materials and enabling knowledge reuse by generative AI. Specific embodiments of the system are described below.

[0486] Uploading materials

[0487] A user accesses the KnowledgeMate interface using a computing device, selects materials such as documents (PDF, Word), presentations (PPT), and video files (MP4), and presses the upload button, which sends the materials to the server.

[0488] Receiving files

[0489] The server receives the uploaded materials from the user and identifies the file format, for example, the server checks the file extension and automatically classifies the format as ".pdf", ".docx", ".pptx", ".mp4", etc.

[0490] Auto-organization and tagging

[0491] Once the file is received, the server automatically analyzes the file contents using natural language processing (NLP) technology, such as PyMuPDF for extracting text from PDF files, python-docx for extracting text from Word files, and python-pptx for extracting text from presentation files. Based on the analysis results, the server then assigns classification tags, such as marketing materials, technical documents, and annual reports, to make it easier to organize the documents.

[0492] Content analysis and transformation

[0493] The server then uses generative AI models to further analyze the content of the uploaded files. For text files, topic modeling techniques are used to extract important keywords. For video files, a speech recognition engine (e.g., Google Cloud Speech-to-Text) is used to convert the audio into text and identify important scenes and key points.

[0494] Saving to a database

[0495] The parsed and converted content is formatted into a reusable format (e.g., JSON, text data). The server creates a JSON object with fields such as "Document ID," "Title," "Content," and "Tags" as text data and stores it in a database. This database is built using, for example, "MongoDB" or "PostgreSQL."

[0496] Knowledge Search and Retrieval

[0497] A user inputs a prompt sentence into the generative AI model using a terminal. For example, if a request is made such as "Please tell me about materials related to last year's marketing strategy," the server receives the prompt sentence and analyzes it using the generative AI model. The server then uses Retrieval-Augmented Generation (RAG) technology to search for related knowledge in the database.

[0498] The generative AI model generates an appropriate answer to the user's question, such as "Here is last year's marketing strategy document," and the server sends this to the user's device.

[0499] Versioning and identifying the latest version

[0500] The server searches the database for different versions of the same document and uses a generative AI model to analyze the differences in content, particularly identifying differences and added information, and marks the version with the most recent content in the database, ensuring that the latest version is always served when a user requests the latest version.

[0501] Specific examples

[0502] Uploading and analyzing data

[0503] 1. User operation: To upload "new presentation materials," the user logs in to the Knowledge Mate portal site, selects the file "presentation_marketing_2023.pptx," and presses the upload button.

[0504] 2. Server processing: The server receives "presentation_marketing_2023.pptx" and identifies its format. The server uses the "python-pptx" library to analyze the slide content and extract text and image information. The server adds tags such as "Presentation Materials, Category: Marketing, Date: October 2023" and saves the analysis results in a database.

[0505] Extracting knowledge with generative AI

[0506] 1. User interaction: The user inputs a prompt into the generation AI from their device, asking, "Please tell me about materials related to last year's marketing strategy."

[0507] 2. Server processing: The server receives the user's prompt and uses a generative AI (e.g., OpenAI GPT-4) to analyze the prompt. It then uses RAG technology to search for relevant marketing strategy documents in a database and generates an answer from the documents found, such as "We've found last year's marketing strategy documents. This is the latest version." The server then sends this answer to the user, allowing them to view the documents on their device.

[0508] This allows users to efficiently obtain the knowledge they need, and makes it easier to manage materials and reuse information.

[0509] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0510] Step 1:

[0511] User operations

[0512] A user accesses the KnowledgeMate interface using a computing device, presses the upload button, and selects materials. The input materials can be documents (PDF, Word), presentations (PPT), video files (MP4), etc. The output is an upload request that is sent to the server.

[0513] Step 2:

[0514] Server Processing

[0515] The server processes the received upload request and receives the document. At this time, the server identifies the format of the document. It checks the extension of the received file and recognizes the format from ".pdf", ".docx", ".pptx", ".mp4", etc. The output is the format information of the document.

[0516] Step 3:

[0517] Server Processing

[0518] The server analyzes the file contents and automatically assigns classification tags. This process uses natural language processing (NLP) techniques and specific libraries (e.g., "PyMuPDF," "python-docx," "python-pptx"). The input is the received document itself, and the output is the analyzed text data and corresponding tag information. For example, in the case of a PDF file, the server generates tags such as "marketing material" and "technical document."

[0519] Step 4:

[0520] Server Processing

[0521] The server uses the generated AI model to further analyze and convert the content of the material. This process involves topic modeling and keyword extraction for text data, and in the case of video files, converting the audio into text using a speech recognition engine. The input is the analyzed text data, and the output is data formatted in a reusable format (e.g., JSON, text data). In the case of video files, the server extracts important scenes and key points.

[0522] Step 5:

[0523] Server Processing

[0524] The parsed and converted data is saved to a database. In this process, the data is stored in a reusable format in a database (such as MongoDB or PostgreSQL). The input is the formatted data, and the output is a response confirming that it has been saved to the database. The server creates a JSON object with fields such as "Document ID", "Title", "Content", and "Tags" that hold the text data.

[0525] Step 6:

[0526] User operations

[0527] A user inputs a question in natural language to the generative AI model using a terminal. For example, they input a prompt such as, "Please tell me about materials related to last year's marketing strategy." The input is the user's question, and the output is the generated prompt sentence.

[0528] Step 7:

[0529] Server Processing

[0530] The server uses a generative AI model to analyze the prompt and retrieve relevant information from the database. It then uses RAG technology to search for relevant knowledge in the database. The input is the prompt and the knowledge in the database, and the output is the retrieved relevant material.

[0531] Step 8:

[0532] Server Processing

[0533] The generative AI model generates an appropriate answer to the user's question. For example, it creates an answer text such as "Last year's marketing strategy materials are here." The input is the searched materials, and the output is the generated answer. The server sends this answer to the user's device, where it can be viewed by the user.

[0534] Step 9:

[0535] Server Processing

[0536] Different versions of the same document are searched for in the database and a generative AI model is used to analyze the differences in content. The input is the different versions of the document, and the output is a comparison of the old and new content based on that. The server then identifies the latest version and marks it in the database. Whenever a user requests the latest version, the latest version is always provided.

[0537] (Application example 1)

[0538] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0539] Currently, work knowledge and troubleshooting procedures within factories depend on the experience and knowledge of individual employees, making it difficult to share and reuse knowledge. Furthermore, in emergencies, appropriate information cannot be obtained quickly, which risks reducing the efficiency of the entire factory. Furthermore, when different versions of documents exist, it is time-consuming to identify which document is the latest version, which can lead to work being carried out based on incorrect information.

[0540] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0541] In this invention, the server includes means for receiving materials, means for automatically analyzing the content of the materials and assigning classification tags, means for converting the analyzed content into a reusable format and saving it in a database, means for retrieving related information from the database in response to a question and generating an answer using a generative AI model, means for comparing different versions of the same material and identifying the latest version, and means for managing work knowledge and troubleshooting procedures with an application installed in an automated device in a factory and providing quick answers based on prompts. This automates knowledge management within the factory and enables efficient and error-free information provision.

[0542] "Means for receiving materials" refers to the devices and functions that receive various materials from users and incorporate them into the system.

[0543] "Means for automatically analyzing the content of materials and assigning classification tags" refers to devices or functions that analyze the content of materials and automatically assign appropriate tags based on their characteristics and categories.

[0544] "Means for converting the analyzed content into a reusable format and storing it in a database" refers to a device or function that converts the content of the material into a reusable format (e.g., text data or JSON data) and stores it in a database.

[0545] "Means of retrieving relevant information from a database in response to a question and generating an answer using a generative AI model" refers to a device or function that searches for relevant information from a database in response to a user's question and generates an appropriate answer using a generative AI model.

[0546] "Means for comparing different versions of the same document and identifying the most recent version" refers to devices or functions that analyze and compare multiple versions of the same document and identify the most recent version.

[0547] "Applications installed on automated equipment in factories" refers to applications that are installed on automated equipment (e.g., robots) in factories and that manage work knowledge and troubleshooting procedures.

[0548] "Means for managing work knowledge and troubleshooting procedures and providing quick answers based on prompt statements" refers to devices and functions that organize and manage knowledge about factory work and problem-solving procedures, and generate quick answers based on user prompt statements.

[0549] This is a system for realizing applications for factory robots that efficiently manage work knowledge and troubleshooting procedures within factories and provide quick answers.

[0550] System configuration

[0551] 1. Technical configuration

[0552] The hardware used is a Linux-based server (e.g., AWS EC2, Google Cloud Compute Engine), Amazon S3 for file storage, and MongoDB for the database.

[0553] The software used is a web application server using Flask, various libraries for data analysis, and OpenAI GPT-3 as a generative AI model.

[0554] 2. Uploading and analyzing materials

[0555] Users can upload documents via the factory robot's interface (e.g., screen or console). Document formats include PDF, Word, PPT, MP4, etc.

[0556] The server receives the uploaded materials, automatically analyzes their contents, and assigns appropriate classification tags, such as "factory maintenance" or "troubleshooting."

[0557] 3. Building a knowledge database

[0558] The server converts the parsed content into a reusable format (e.g., text data or JSON data) and stores it in a database, which will serve as the foundation for future knowledge searches.

[0559] If different versions of the same material are uploaded, the server automatically compares them and identifies the most recent version, marking it in the database.

[0560] 4. Question and Answer Generation

[0561] The user inputs a question to the factory robot, for example, using a prompt such as "Tell me about yesterday's maintenance procedures."

[0562] The server sends the question to a generative AI model (e.g., OpenAI GPT-3), which searches for relevant information in a database and generates an appropriate answer, which is then provided to the user through the factory robot's interface.

[0563] 5. Specific Examples

[0564] Uploading materials: A worker uploads a presentation (e.g., maintenance procedures) to the robot's interface. The system assigns a tag such as "Factory Maintenance" and stores it in a database.

[0565] Example of question-and-answering: A worker asks the robot, "What were the maintenance procedures yesterday?" The system searches the database for relevant documents and generates and provides an answer such as, "Here are the maintenance procedures yesterday."

[0566] Example prompt sentence:

[0567] Based on your knowledge database, generate an appropriate answer to the question, "What was the maintenance procedure yesterday?"

[0568] This system automates knowledge management within factories, facilitating the sharing and reuse of work knowledge and troubleshooting procedures. In addition, by utilizing generative AI models, it can quickly provide appropriate information in emergencies, improving factory efficiency and safety.

[0569] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0570] Step 1:

[0571] The user operates the factory robot's interface to upload a file (e.g., PDF, Word, PPT, MP4). The user presses the file selection button and selects the file they want to upload. This input sends the file to the server.

[0572] Step 2:

[0573] The server receives the uploaded material, identifies the format of the received material file, and prepares it for appropriate processing. Here, different analysis modules are selected depending on the file format (e.g., document file, presentation file, video file, etc.).

[0574] Step 3:

[0575] The server automatically analyzes the content of the file. For documents and presentations, it extracts text and image information. For video files, it converts audio to text and analyzes the video content to extract key points. The output of this process is the analyzed data.

[0576] Step 4:

[0577] The server automatically assigns relevant classification tags to the analyzed data, such as "factory maintenance" or "troubleshooting." This tagging allows the materials to be organized into categories.

[0578] Step 5:

[0579] The server analyzes and converts the tagged content into a reusable format (text data, JSON data, etc.), which is then stored in a database, making it easier to search for the material.

[0580] Step 6:

[0581] The user inputs a question into the factory robot's interface, for example, a prompt such as "Tell me about yesterday's maintenance procedures." This input is sent to the server.

[0582] Step 7:

[0583] The server sends the user's question to a generative AI model (e.g., OpenAI GPT-3), which analyzes the question. The server searches for relevant information in a knowledge database and uses the generative AI model to generate an appropriate answer. The output of this process is an answer to the user's question.

[0584] Step 8:

[0585] The server sends the generated answer to the user's device, which receives the answer and displays it to the user through the factory robot's interface. The user can refer to the displayed answer to proceed with the work.

[0586] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0587] The present invention is a system for managing, analyzing, and reusing materials, and recognizing user emotions. Specific embodiments for carrying out the present invention and the processing contents thereof will be described below.

[0588] System configuration

[0589] 1. Uploading materials

[0590] Users access the Knowledge Mate interface using their devices and upload documents and video files, including document files (PDF, Word), presentation files (PPT), and video files (MP4).

[0591] 2. Receiving the file

[0592] The server receives the uploaded file, identifies the file type (document, presentation, video, etc.), and prepares it for appropriate processing.

[0593] 3. Automatic organization and tagging

[0594] The server automatically analyzes the content of the files and assigns relevant classification tags, such as marketing materials, technical documentation, annual reports, etc. This process organizes the materials into categories.

[0595] 4. Content Analysis and Transformation

[0596] The server uses a generative AI engine to analyze the contents of the uploaded file. For documents and presentations, elements such as text, images, and graphs are extracted. For video files, the audio is converted to text and the video content is analyzed to extract key points.

[0597] 5. Saving to the database

[0598] The server then formats the parsed and converted content into a reusable format (e.g., text data, JSON data, etc.) and stores it in a database, which serves as the foundation for future knowledge search and reuse.

[0599] 6. User Emotion Recognition

[0600] The device sends the user's input or voice to the emotion engine, which analyzes it and recognizes the user's emotion (e.g., joy, anger, sadness, surprise, etc.).

[0601] 7. Knowledge Search and Retrieval

[0602] The user uses a device to input a question to the generation AI, for example, "Please show me materials related to last year's marketing strategy."

[0603] 8. Emotion-Based Response Modulation

[0604] The server sends the user's question to the generation AI, which then uses Retrieval-Augmented Generation (RAG) to search for relevant knowledge in the database. At the same time, the emotion engine recognizes the user's emotional state and adjusts the response accordingly.

[0605] 9. Generating Answers to Questions

[0606] Based on the knowledge extracted by the generative AI and the analysis results of the emotion engine, the system generates an appropriate answer to the user's question. For example, it creates a response such as, "Here are last year's marketing strategy documents. Please let me know if you have any additional questions."

[0607] 10. Versioning and Identifying the Latest Version

[0608] The server searches the database for different versions of the same document and uses generative AI to analyze the differences in content.

[0609] The server uses the analysis to identify the most up-to-date version and marks it in the database, which is then served whenever a user requests the most up-to-date document.

[0610] Specific examples

[0611] Example 1: Uploading and analyzing data

[0612] 1. User Action:

[0613] To upload "new presentation materials," a user logs into the Knowledge Mate portal site, selects the file, and then presses the upload button.

[0614] 2. Server processing:

[0615] The server receives the file, identifies its format, and, if it is a presentation, analyzes the slides and extracts text and image information.

[0616] The server assigns tags such as "Presentation Materials, Category: Marketing, Date: October 2023" and stores it in a database.

[0617] Example 2: Knowledge extraction using generative AI and emotion engines

[0618] 1. User Action:

[0619] The user inputs a question into the generation AI from their device, such as, "Do you have any materials related to last year's marketing strategy?" At the same time, the emotion engine recognizes the user's "confused" state from their tone of voice and text.

[0620] 2. Server processing:

[0621] The server receives the user's question, analyzes it using generative AI, and uses RAG to search for relevant marketing strategy materials in the database.

[0622] Based on the emotion of "confusion" recognized by the emotion engine, the generative AI generates a more polite and understandable response, such as, "Here are last year's marketing strategy documents. If you have any questions, please let us know as we have more detailed documents available."

[0623] 3. Response to the user:

[0624] The server sends this response to the user's terminal, allowing the user to access the information.

[0625] This invention is a system that significantly improves the efficiency of knowledge management and enables effective use of knowledge within a company. By incorporating an emotion engine, it provides appropriate responses that take into account the user's emotional state, creating a more user-friendly interface. This automates the organization, analysis, search, and version management of documents, making it even easier for users to utilize knowledge.

[0626] The processing flow will be explained below.

[0627] Step 1:

[0628] A user logs in to the Knowledge Mate interface using a terminal and clicks the upload button. The user selects and uploads a document file (e.g., PDF, Word), a presentation file (e.g., PPT), or a video file (e.g., MP4).

[0629] Step 2:

[0630] The server receives the uploaded file. The server identifies the file type (document, presentation, video, etc.) and prepares it for appropriate processing.

[0631] Step 3:

[0632] The server automatically assigns classification tags to files it receives, automatically determining which category a file belongs to (e.g., "marketing materials," "technical documentation," "annual reports," etc.) and assigning the tag.

[0633] Step 4:

[0634] The server uses a generated AI engine to analyze the contents of the uploaded file. For documents and presentations, elements such as text, images, and graphs are extracted. For video files, audio is converted to text and the video content is analyzed to extract key points.

[0635] Step 5:

[0636] The server formats the parsed content into a reusable format (text data, JSON data, etc.) and saves the formatted data in a database.

[0637] Step 6:

[0638] The user inputs a question into the generation AI from their device, for example, "Please show me materials related to last year's marketing strategy."

[0639] Step 7:

[0640] The server receives the user's question and sends it to the generation AI. At the same time, the device sends the emotion information from the user's tone of voice and input sentences to the emotion engine, which then analyzes the user's emotion (e.g., "confusion," "joy," "anger," etc.).

[0641] Step 8:

[0642] The generation AI uses Retrieval-Augmented Generation (RAG) to search for relevant knowledge in the database, and the server adjusts the tone and content of responses based on the analysis results from the emotion engine.

[0643] Step 9:

[0644] The generative AI generates appropriate answers to user questions based on the extracted knowledge and the analysis results of the emotion engine. For example, it creates a response such as, "Last year's marketing strategy materials are here. If you have any questions, please let us know as we have more detailed materials available."

[0645] Step 10:

[0646] The server sends the generated response to the user's terminal, allowing the user to access the information.

[0647] Step 11:

[0648] The server searches the database for different versions of the same document and uses generative AI to analyze the differences in content. Based on the analysis results, it identifies the most recent version with the most up-to-date content and marks it in the database. This latest version is always provided when a user requests the document.

[0649] Example 2

[0650] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0651] Conventional document management systems required a great deal of time and effort to organize and search documents, and it was difficult to provide appropriate responses that took user sentiment into account. As a result, documents could not be reused or information provided promptly, leading to problems that reduced work efficiency. Furthermore, the mixing of different versions of documents created the problem of cumbersome management of the latest versions.

[0652] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0653] In this invention, the server includes means for receiving materials, means for automatically analyzing the content of the materials and assigning classification tags, means for converting the analyzed content into a reusable format and storing it in a database, means for extracting relevant information from the database in response to a question and generating an answer using a generative AI model, means for comparing different versions of the same material and identifying the latest version, means for analyzing user sentiment and adjusting answers based on the analysis results, and means for converting voice to text and extracting key points from video when analyzing the content of the materials. This enables efficient material management, rapid information provision, user-friendly responses, and accurate provision of the latest materials.

[0654] "Means for receiving materials" refers to the function by which the server receives and stores materials uploaded by users, including identifying file formats and temporarily storing them.

[0655] "Means for automatically analyzing the content of materials and assigning classification tags" is a function that analyzes the content of materials and automatically assigns tags according to their characteristics and categories. This allows materials to be properly classified and managed.

[0656] "Means of converting the analyzed content into a reusable format and saving it in a database" refers to the function of formatting the information extracted from the materials into a format that is easy to reuse and saving that content in a database, making it easier to search and reuse in the future.

[0657] "Means of extracting relevant information from a database in response to a question and generating an answer using a generative AI model" refers to a function that searches for relevant information in a database in response to a question from a user and generates an appropriate answer using generative AI.

[0658] "Means for comparing different versions of the same document and identifying the latest version" refers to a function that compares different versions of the same document, analyzes the changes, and identifies the latest version. This allows the latest version to be provided at all times.

[0659] "Means for analyzing the user's emotions and adjusting responses based on the analysis results" refers to a function that analyzes the user's input and voice data to recognize their emotional state and generates a response based on that emotion, thereby providing the user with an appropriate and empathetic response.

[0660] "Means for converting audio to text and extracting key points from video when analyzing the content of materials" refers to a function that converts video and audio data into text and automatically extracts important points from the video content.

[0661] The present invention is a system for managing, analyzing, and reusing materials, and for recognizing user emotions. The system aims to improve user convenience by automatically organizing materials and generating appropriate responses. Specific embodiments for implementing the present invention are described below.

[0662] System configuration

[0663] Uploading materials

[0664] The user uploads documents using the terminal. Specifically, the user selects files such as PDF, Word, PPT, MP4, etc. through the web interface and presses the upload button. At this time, the terminal organizes the uploaded files into a specific format.

[0665] Receiving files

[0666] The server receives the uploaded files from the device, automatically identifies their format, analyzes the metadata of each file, and prepares them for temporary storage.

[0667] Auto-organization and tagging

[0668] The server automatically analyzes the content of the documents and assigns relevant classification tags, such as marketing materials, technical documents, annual reports, etc. This process is performed using a generative AI engine (e.g., GPT-4). The documents are organized into categories.

[0669] Content analysis and transformation

[0670] The server uses a generative AI engine to perform a detailed analysis of the uploaded file's contents. It extracts text, images, and graphs from document files, converts audio to text from video files, and analyzes the video content to extract key points. This information is then formatted for reuse.

[0671] Saving to a database

[0672] The server stores the parsed and converted content in a database in a reusable format (e.g. text data, JSON data), which serves as the basis for future knowledge search and reuse.

[0673] User Emotion Recognition

[0674] The device sends the user's input and voice to an emotion engine (e.g., EmotionAI) that analyzes the user's emotions. The emotion engine recognizes emotions such as joy, anger, sadness, and surprise.

[0675] Knowledge Search and Retrieval

[0676] The user uses a device to input a question to the generation AI, for example, "Please show me materials related to last year's marketing strategy."

[0677] Emotion-Based Response Modulation

[0678] The server receives the user's question and emotional information, and requests the generation AI to analyze it. It uses Retrieval-Augmented Generation (RAG) to search for relevant knowledge in the database. At the same time, it adjusts an appropriate response based on the analysis results of the emotion engine.

[0679] Generate answers to questions

[0680] Based on the knowledge extracted by the generative AI and the analysis results of the emotion engine, the system generates an appropriate answer to the user's question. For example, it creates a response such as, "Here are last year's marketing strategy documents. Please let me know if you have any additional questions."

[0681] Versioning and identifying the latest version

[0682] The server searches the database for different versions of the same document and uses generative AI to analyze the differences in content. Based on the analysis results, it identifies the most recent version with the most up-to-date content and marks it in the database. This latest version is always provided when a user requests the document.

[0683] Specific examples

[0684] Example 1: Uploading and analyzing data

[0685] 1. User operation: To upload "new presentation materials," the user logs in to the portal site, selects the file, and then presses the upload button.

[0686] 2. Server processing: The server receives the file, identifies its format, and if it is a presentation, analyzes the slides and extracts text and image information.

[0687] 3. Tagging: The server assigns tags such as "Presentation Materials, Category: Marketing, Date: October 2023" and stores them in the database.

[0688] Example 2: Knowledge extraction using generative AI and emotion engines

[0689] 1. User operation: The user inputs a question into the generation AI from their device, asking, "Do you have any materials about last year's marketing strategy?" At the same time, the emotion engine recognizes the user's "confused" state from their tone of voice and text.

[0690] 2. Server processing: The server receives the user's question, analyzes it using the generative AI, and uses RAG to search for relevant marketing strategy materials in the database.

[0691] 3. Emotional response: Based on the emotion of "confusion" recognized by the emotion engine, the generative AI generates a more polite and understandable response, such as, "Here are last year's marketing strategy documents. If you have any questions, please let us know as we have more detailed documents available."

[0692] 4. Reply to the user: The server sends this answer to the user's device, allowing the user to access the information.

[0693] Hardware and software used

[0694] Hardware: Servers, terminals (PCs, smartphones)

[0695] Software: Generative AI engine (e.g., GPT-4), emotion engine (e.g., EmotionAI), database management system (e.g., MySQL), Knowledge Mate interface (web application)

[0696] Prompt Sentence Examples

[0697] 1. "Please find materials related to last year's marketing strategy."

[0698] 2. "Upload a new presentation."

[0699] 3. "I'm having trouble and need help. What resources should I look at?"

[0700] This system not only significantly improves the efficiency of knowledge management, but also realizes a user-friendly interface by providing responses that respond to the user's emotions. By automating the organization, analysis, search, and version management of materials, users can utilize knowledge more easily and effectively.

[0701] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0702] Step 1:

[0703] The user uploads the document using a terminal. Specifically, the user accesses the web interface, selects the file, and then presses the upload button. The input is a file in a format such as PDF, Word, PPT, or MP4, and the output is the file data sent to the server.

[0704] Step 2:

[0705] The server receives the file uploaded by the user. The server automatically identifies the file format and temporarily stores it. In this step, the input is the file data sent by the user, and the output is the identified file format and the temporarily stored file.

[0706] Step 3:

[0707] The server automatically analyzes the content of the received material and assigns classification tags. It uses a generative AI engine (e.g., GPT-4) to analyze the text, images, and other elements in the file. In this step, the input is the temporarily saved file data, and the output is the assigned classification tag (e.g., marketing material, technical document, etc.).

[0708] Step 4:

[0709] The server converts the content into a reusable format, for example extracting text data from a document or converting audio to text from a video. In this step, the input is the parsed content and the output is formatted data in a reusable format (e.g. text data, JSON data).

[0710] Step 5:

[0711] The server stores the data in a reusable format in a database. This process involves database manipulation using SQL queries. In this step, the input is the data in a reusable format and the output is the data stored in the database.

[0712] Step 6:

[0713] The device sends the user's input or voice to the emotion engine, which uses a model like EmotionAI. In this step, the input is the user's input text or voice data, and the output is the analyzed user's emotion (e.g., joy, anger, sadness).

[0714] Step 7:

[0715] The user inputs a question to the generative AI, for example, "Do you have any materials about last year's marketing strategy?" In this step, the input is the question text and the output is the request data sent to the server.

[0716] Step 8:

[0717] The server sends the question to the generation AI, which uses Retrieval-Augmented Generation (RAG) to search for relevant information in the database. It also adjusts the response based on information from the emotion engine. In this step, the input is the user's question and emotion information, and the output is the answer adjusted by the generation AI.

[0718] Step 9:

[0719] The generative AI generates an answer based on the extracted knowledge and sentiment analysis. Specifically, it generates a response such as, "Here are last year's marketing strategy documents. Please let me know if you have any additional questions." In this step, the input is information from the database and the results of sentiment analysis, and the output is the generated answer text.

[0720] Step 10:

[0721] The server compares different versions of the material and identifies the latest version. Generative AI is used to analyze the content of the different versions, identify the latest version with the most up-to-date content, and mark it in the database. In this step, the input is the material data of the different versions, and the output is the identified latest version of the material.

[0722] (Application example 2)

[0723] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0724] Conventional document management systems lack functionality such as navigation assistance and real-time analysis of the driver's emotional state, and technologies to improve driving safety and convenience are needed, especially for autonomous vehicles. There is also a need for technologies to properly grasp the user's emotional state while driving and provide appropriate feedback.

[0725] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving materials, means for automatically analyzing the content of the materials and assigning classification tags, means for converting the analyzed content into a reusable format and saving it in a database, means for retrieving related information from the database in response to a question and generating an answer, means for comparing different versions of the same material and identifying the latest version, means for uploading map data and destination information, means for analyzing the user's emotional state in real time, and means for providing navigation assistance information based on the analysis results. This makes it possible to appropriately grasp the driver's emotional state in an autonomous vehicle and provide feedback and warnings accordingly.

[0726] "Means for receiving materials" refers to devices or software that allow the server to receive materials in various formats (e.g., documents, presentations, video files) provided by users.

[0727] The "means for assigning classification tags" refers to a device or software that automatically analyzes the content of received materials and assigns tags indicating appropriate categories or attributes to the materials.

[0728] "Means for converting into a reusable format and storing in a database" refers to a device or software that converts the content of the analyzed material into a reusable format (e.g., text data, JSON data) and stores it in a database.

[0729] "Means for retrieving relevant information from a database in response to a question and generating an answer" refers to a device or software that searches a database for relevant information in response to a question from a user and generates an appropriate answer.

[0730] A "means for comparing different versions of the same material and identifying the most recent version" is a device or software that compares multiple versions of the same material and identifies the version with the most recent content.

[0731] The "means for uploading map data and destination information" refers to a device or software that uploads map data and destination information entered by the user to the system.

[0732] The "means for analyzing the user's emotional state in real time" refers to a device or software that uses a sensor, camera, microphone, etc. to analyze the user's emotional state in real time.

[0733] The "means for providing navigation assistance information" is a device or software that provides appropriate navigation assistance information (e.g., route directions, warning messages) to the driver based on the analyzed information and the user's emotional state.

[0734] To put the present invention into practice, a document management system and a navigation support system are integrated, as will be described in detail below.

[0735] System configuration

[0736] The system mainly consists of a server, a user terminal, smart glasses, and a head-mounted display (HMD).

[0737] User operations

[0738] Users input destination information using smart glasses or an HMD and upload it to the vehicle system, where it is received and analyzed by a server.

[0739] Server Processing

[0740] The server receives map data and destination information uploaded by users using a means for receiving data. Next, the server analyzes this data using a means for assigning classification tags, and assigns tags according to its characteristics. The analyzed content is then converted into a reusable format and managed using a means for storing it in a database. For example, tags such as "long-distance drive" and "intra-city travel" are assigned.

[0741] Real-time analytics

[0742] To analyze the user's emotional state, sensors, cameras, and microphones built into the smart glasses or HMD are used. This allows the user's emotional state to be transmitted to a server in real time for analysis. Depending on the emotional state, a means for providing navigation assistance information is activated, and appropriate feedback and warnings are provided to the user.

[0743] Navigation Aids

[0744] The server generates navigation assistance information based on map data and destination information. Based on the analysis results and the user's emotional state, specific driving instructions such as "Turn right at the next intersection" or emotional feedback such as "Drive calmly and without rushing" are displayed on the smart glasses or HMD.

[0745] Specific examples

[0746] For example, if a user inputs "Drive to Tokyo Station," the server receives this data and tags it as "Long-distance drive." At the same time, if the server detects an emotion of "impatience" from the user's tone of voice or facial expression, it displays emotional feedback such as "Please drive calmly."

[0747] Example prompts

[0748] Here is an example prompt:

[0749] Analyze the following map data and tag it appropriately:

[0750] "Destination: Tokyo Station, looking for a long drive"

[0751] Also, analyze the emotional state from the following audio data and provide appropriate feedback:

[0752] "You seem anxious."

[0753] Hardware and Software Examples

[0754] The hardware used includes smart glasses (e.g., Google Glass), head-mounted displays (e.g., Microsoft HoloLens), and in-vehicle servers (with high-performance processors).The software used includes Python scripts for data analysis and emotion recognition, and TensorFlow or PyTorch for generative AI and emotion analysis.

[0755] As described above, the present invention integrates a document management system and a navigation assistance system, allowing users to receive real-time emotion recognition and navigation assistance while driving.

[0756] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0757] Step 1:

[0758] The user uses smart glasses or a head-mounted display (HMD) to input destination information and upload it to the vehicle system. The input data is in the form of, for example, "I would like to drive to Tokyo Station." This information is then sent to the server via the terminal.

[0759] Step 2:

[0760] The server analyzes the received map data and destination information. The input data is "I would like to drive to Tokyo Station." The server uses a generative AI model to analyze the input data and tag it (e.g., "long-distance drive"). The tagged data is obtained as output.

[0761] Step 3:

[0762] The server converts the parsed data into a reusable format (e.g. text data, JSON data) and stores it in the database. The tagged data as input is processed and converted into a format that can be stored in the database. The converted data as output is stored in the database.

[0763] Step 4:

[0764] The smart glasses and HMD use built-in sensors and microphones to analyze the user's emotional state in real time. Voice and facial expressions are sent as input data to the server, which then analyzes them using an emotion recognition algorithm. The analysis results (e.g., "anxiety") are obtained as output.

[0765] Step 5:

[0766] The server generates appropriate feedback or warnings based on the analysis results. It processes the emotion analysis results (e.g., "impatience") as input and generates a feedback message (e.g., "Please drive calmly"). The generated feedback is obtained as output.

[0767] Step 6:

[0768] The server provides navigation assistance information. Based on input data such as map data, destination information, and emotion analysis results, driving assistance information (e.g., "Please turn right at the next intersection") is generated. The output is displayed on the user's smart glasses or HMD.

[0769] Step 7:

[0770] The server stores the generated navigation assistance information and emotional feedback in a database. The server converts the generated information as input data into a reusable format and stores it in the database. The stored data is obtained as output.

[0771] Through the above steps, the present invention provides navigation assistance and emotional feedback based on the destination information entered by the user and the emotional state analyzed in real time, and this information is stored in a database for future analysis and improvement.

[0772] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0773] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0774] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0775] [Third embodiment]

[0776] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0777] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0778] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0779] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0780] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0781] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0782] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0783] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0784] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0785] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0786] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0787] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0788] The present invention is a system that efficiently manages materials and converts that knowledge into a reusable form using generative AI. A specific embodiment of the system and the processing flow are described below.

[0789] System configuration

[0790] 1. Uploading materials

[0791] Users access the Knowledge Mate interface using their devices and upload documents and video files, including document files (PDF, Word), presentation files (PPT), and video files (MP4).

[0792] 2. Receiving the file

[0793] The server receives the uploaded file, identifies the file type (document, presentation, video, etc.), and prepares it for appropriate processing.

[0794] 3. Automatic organization and tagging

[0795] The server automatically analyzes the content of the files and assigns relevant classification tags, such as marketing materials, technical documentation, annual reports, etc. This process organizes the materials into categories.

[0796] 4. Content Analysis and Transformation

[0797] The server uses the generated AI to analyze the contents of the uploaded file. For documents and presentations, elements such as text, images, and graphs are extracted. For video files, the audio is converted to text and the video content is analyzed to extract key points.

[0798] 5. Saving to the database

[0799] The server then formats the parsed and converted content into a reusable format (e.g., text data, JSON data, etc.) and stores it in a database, which serves as the foundation for future knowledge search and reuse.

[0800] 6. Knowledge Search and Retrieval

[0801] The user uses a device to input a question to the generation AI, for example, "Please show me materials related to last year's marketing strategy."

[0802] The server analyzes the user's question using generative AI and searches for relevant knowledge in the database using RAG (Retrieval-Augmented Generation).

[0803] 7. Generating Answers to Questions

[0804] Based on the knowledge extracted by the generative AI, it generates appropriate answers to user questions, such as "Last year's marketing strategy materials can be found here."

[0805] The server sends the generated response to the user's terminal, allowing the user to access the information.

[0806] 8. Versioning and Identifying the Latest Version

[0807] The server searches the database for different versions of the same document and uses generative AI to analyze the differences in content.

[0808] The server uses the analysis to identify the most up-to-date version and marks it in the database, which is then served whenever a user requests the most up-to-date document.

[0809] Specific examples

[0810] Example 1: Uploading and analyzing data

[0811] 1. User Action:

[0812] To upload "new presentation materials," a user logs into the Knowledge Mate portal site, selects the file, and then presses the upload button.

[0813] 2. Server processing:

[0814] The server receives the file, identifies its format, and, if it is a presentation, analyzes the slides and extracts text and image information.

[0815] The server assigns tags such as "Presentation Materials, Category: Marketing, Date: October 2023" and stores it in a database.

[0816] Example 2: Extracting knowledge using generative AI

[0817] 1. User Action:

[0818] The user inputs a question into the generation AI from their device, asking, "Do you have any materials related to last year's marketing strategy?"

[0819] 2. Server processing:

[0820] The server receives the user's question, analyzes it using generative AI, and uses RAG to search for relevant marketing strategy materials in the database.

[0821] The server identifies the relevant materials, and the generation AI generates a response such as, "We found last year's marketing strategy materials. This is the latest version."

[0822] The server sends this response to the user, who can then view the information on their terminal.

[0823] This invention is a system that significantly improves the efficiency of knowledge management and realizes effective use of knowledge within a company. This automates the organization, analysis, search, and version management of documents, making it easier for users to utilize knowledge.

[0824] The processing flow will be explained below.

[0825] Step 1:

[0826] A user logs in to the Knowledge Mate interface and clicks the upload button. The user selects and uploads a document file (PDF, Word), a presentation file (PPT), or a video file (MP4).

[0827] Step 2:

[0828] The server receives the uploaded file and identifies the file format, determining whether it is a document, presentation, video, etc.

[0829] Step 3:

[0830] The server automatically assigns classification tags to uploaded files, identifying the category to which the file belongs, such as "marketing materials," "technical documentation," or "annual reports."

[0831] Step 4:

[0832] The server sends the file contents to the generative AI engine. If it's a document or presentation, the generative AI extracts elements like text, images, and graphs. If it's a video file, it converts the audio into text and analyzes the video to extract key points.

[0833] Step 5:

[0834] The server formats the parsed and converted content into a reusable format (text data, JSON data, etc.) and saves the formatted data in a database.

[0835] Step 6:

[0836] The user uses a device to input a question to the generation AI, for example, "Please show me materials related to last year's marketing strategy."

[0837] Step 7:

[0838] The server sends the user's question to the generation AI, which uses Retrieval-Augmented Generation (RAG) to search for relevant knowledge in the database.

[0839] Step 8:

[0840] Generative AI generates answers based on search results, such as "Here are last year's marketing strategy documents."

[0841] Step 9:

[0842] The server sends the generated response to the user's terminal, allowing the user to access the information.

[0843] Step 10:

[0844] The server searches the database for different versions of the same document and uses generative AI to analyze the differences in content. Based on the analysis results, it identifies the most recent version with the most up-to-date content and marks it in the database. This latest version is always provided when a user requests the document.

[0845] Example 1

[0846] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0847] Current knowledge management systems require a great deal of time and effort to manually classify and organize documents and search and retrieve knowledge. Furthermore, when managing different versions of the same document, it is difficult to identify the latest version. This makes it difficult for users to quickly and accurately obtain the information they need.

[0848] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0849] In this invention, the server includes: means for users to upload materials using a computing device; means for receiving the materials and identifying their format; means for analyzing the file content and automatically assigning classification tags; means for converting the analyzed content into a reusable format and storing it in a database; means for users to input questions in natural language; means for using a generative AI model to extract relevant information from a database and generate an answer to the question; means for transmitting the generated answer to the user's computing device; and means for comparing different versions of the same material and identifying the latest version. This facilitates automatic organization, analysis, search, and version management of materials, enabling users to efficiently obtain the knowledge they need.

[0850] "User" means an individual or organization that uses the system to upload materials or search for knowledge.

[0851] A "computing device" is a device such as a calculator, PC, smartphone, or tablet that a user uses to upload materials or search for knowledge.

[0852] "Materials" refers to a collection of information, such as documents, presentations, and video files, that can be managed, analyzed, and searched.

[0853] A "generative AI model" is an artificial intelligence framework that uses deep learning technology to perform natural language processing and data analysis to generate and provide necessary information.

[0854] "Server" means a computer system that receives, analyzes, and stores uploaded materials and provides information in response to user requests.

[0855] "Means for identifying formats" refers to a function that automatically detects the file format of uploaded materials and performs appropriate processing.

[0856] "Tags" are keywords or labels that are assigned to indicate the classification or characteristics of materials.

[0857] "Knowledge" is a set of knowledge or information extracted from information stored in a database and provided in response to a user's question.

[0858] A "database" is a digital information repository for storing and managing analyzed materials and generated knowledge.

[0859] "Version control" is the process of distinguishing between different versions of the same material and identifying and managing the most recent version.

[0860] The present invention provides a system for efficiently managing materials and enabling knowledge reuse by generative AI. Specific embodiments of the system are described below.

[0861] Uploading materials

[0862] A user accesses the KnowledgeMate interface using a computing device, selects materials such as documents (PDF, Word), presentations (PPT), and video files (MP4), and presses the upload button, which sends the materials to the server.

[0863] Receiving files

[0864] The server receives the uploaded materials from the user and identifies the file format, for example, the server checks the file extension and automatically classifies the format as ".pdf", ".docx", ".pptx", ".mp4", etc.

[0865] Auto-organization and tagging

[0866] Once the file is received, the server automatically analyzes the file contents using natural language processing (NLP) technology, such as PyMuPDF for extracting text from PDF files, python-docx for extracting text from Word files, and python-pptx for extracting text from presentation files. Based on the analysis results, the server then assigns classification tags, such as marketing materials, technical documents, and annual reports, to make it easier to organize the documents.

[0867] Content analysis and transformation

[0868] The server then uses generative AI models to further analyze the content of the uploaded files. For text files, topic modeling techniques are used to extract important keywords. For video files, a speech recognition engine (e.g., Google Cloud Speech-to-Text) is used to convert the audio into text and identify important scenes and key points.

[0869] Saving to a database

[0870] The parsed and converted content is formatted into a reusable format (e.g., JSON, text data). The server creates a JSON object with fields such as "Document ID," "Title," "Content," and "Tags" as text data and stores it in a database. This database is built using, for example, "MongoDB" or "PostgreSQL."

[0871] Knowledge Search and Retrieval

[0872] A user inputs a prompt sentence into the generative AI model using a terminal. For example, if a request is made such as "Please tell me about materials related to last year's marketing strategy," the server receives the prompt sentence and analyzes it using the generative AI model. The server then uses Retrieval-Augmented Generation (RAG) technology to search for related knowledge in the database.

[0873] The generative AI model generates an appropriate answer to the user's question, such as "Here is last year's marketing strategy document," and the server sends this to the user's device.

[0874] Versioning and identifying the latest version

[0875] The server searches the database for different versions of the same document and uses a generative AI model to analyze the differences in content, particularly identifying differences and added information, and marks the version with the most recent content in the database, ensuring that the latest version is always served when a user requests the latest version.

[0876] Specific examples

[0877] Uploading and analyzing data

[0878] 1. User operation: To upload "new presentation materials," the user logs in to the Knowledge Mate portal site, selects the file "presentation_marketing_2023.pptx," and presses the upload button.

[0879] 2. Server processing: The server receives "presentation_marketing_2023.pptx" and identifies its format. The server uses the "python-pptx" library to analyze the slide content and extract text and image information. The server adds tags such as "Presentation Materials, Category: Marketing, Date: October 2023" and saves the analysis results in a database.

[0880] Extracting knowledge with generative AI

[0881] 1. User interaction: The user inputs a prompt into the generation AI from their device, asking, "Please tell me about materials related to last year's marketing strategy."

[0882] 2. Server processing: The server receives the user's prompt and uses a generative AI (e.g., OpenAI GPT-4) to analyze the prompt. It then uses RAG technology to search for relevant marketing strategy documents in a database and generates an answer from the documents found, such as "We've found last year's marketing strategy documents. This is the latest version." The server then sends this answer to the user, allowing them to view the documents on their device.

[0883] This allows users to efficiently obtain the knowledge they need, and makes it easier to manage materials and reuse information.

[0884] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0885] Step 1:

[0886] User operations

[0887] A user accesses the KnowledgeMate interface using a computing device, presses the upload button, and selects materials. The input materials can be documents (PDF, Word), presentations (PPT), video files (MP4), etc. The output is an upload request that is sent to the server.

[0888] Step 2:

[0889] Server Processing

[0890] The server processes the received upload request and receives the document. At this time, the server identifies the format of the document. It checks the extension of the received file and recognizes the format from ".pdf", ".docx", ".pptx", ".mp4", etc. The output is the format information of the document.

[0891] Step 3:

[0892] Server Processing

[0893] The server analyzes the file contents and automatically assigns classification tags. This process uses natural language processing (NLP) techniques and specific libraries (e.g., "PyMuPDF," "python-docx," "python-pptx"). The input is the received document itself, and the output is the analyzed text data and corresponding tag information. For example, in the case of a PDF file, the server generates tags such as "marketing material" and "technical document."

[0894] Step 4:

[0895] Server Processing

[0896] The server uses the generated AI model to further analyze and convert the content of the material. This process involves topic modeling and keyword extraction for text data, and in the case of video files, converting the audio into text using a speech recognition engine. The input is the analyzed text data, and the output is data formatted in a reusable format (e.g., JSON, text data). In the case of video files, the server extracts important scenes and key points.

[0897] Step 5:

[0898] Server Processing

[0899] The parsed and converted data is saved to a database. In this process, the data is stored in a reusable format in a database (such as MongoDB or PostgreSQL). The input is the formatted data, and the output is a response confirming that it has been saved to the database. The server creates a JSON object with fields such as "Document ID", "Title", "Content", and "Tags" that hold the text data.

[0900] Step 6:

[0901] User operations

[0902] A user inputs a question in natural language to the generative AI model using a terminal. For example, they input a prompt such as, "Please tell me about materials related to last year's marketing strategy." The input is the user's question, and the output is the generated prompt sentence.

[0903] Step 7:

[0904] Server Processing

[0905] The server uses a generative AI model to analyze the prompt and retrieve relevant information from the database. It then uses RAG technology to search for relevant knowledge in the database. The input is the prompt and the knowledge in the database, and the output is the retrieved relevant material.

[0906] Step 8:

[0907] Server Processing

[0908] The generative AI model generates an appropriate answer to the user's question. For example, it creates an answer text such as "Last year's marketing strategy materials are here." The input is the searched materials, and the output is the generated answer. The server sends this answer to the user's device, where it can be viewed by the user.

[0909] Step 9:

[0910] Server Processing

[0911] Different versions of the same document are searched for in the database and a generative AI model is used to analyze the differences in content. The input is the different versions of the document, and the output is a comparison of the old and new content based on that. The server then identifies the latest version and marks it in the database. Whenever a user requests the latest version, the latest version is always provided.

[0912] (Application example 1)

[0913] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0914] Currently, work knowledge and troubleshooting procedures within factories depend on the experience and knowledge of individual employees, making it difficult to share and reuse knowledge. Furthermore, in emergencies, appropriate information cannot be obtained quickly, which risks reducing the efficiency of the entire factory. Furthermore, when different versions of documents exist, it is time-consuming to identify which document is the latest version, which can lead to work being carried out based on incorrect information.

[0915] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0916] In this invention, the server includes means for receiving materials, means for automatically analyzing the content of the materials and assigning classification tags, means for converting the analyzed content into a reusable format and saving it in a database, means for retrieving related information from the database in response to a question and generating an answer using a generative AI model, means for comparing different versions of the same material and identifying the latest version, and means for managing work knowledge and troubleshooting procedures with an application installed in an automated device in a factory and providing quick answers based on prompts. This automates knowledge management within the factory and enables efficient and error-free information provision.

[0917] "Means for receiving materials" refers to the devices and functions that receive various materials from users and incorporate them into the system.

[0918] "Means for automatically analyzing the content of materials and assigning classification tags" refers to devices or functions that analyze the content of materials and automatically assign appropriate tags based on their characteristics and categories.

[0919] "Means for converting the analyzed content into a reusable format and storing it in a database" refers to a device or function that converts the content of the material into a reusable format (e.g., text data or JSON data) and stores it in a database.

[0920] "Means of retrieving relevant information from a database in response to a question and generating an answer using a generative AI model" refers to a device or function that searches for relevant information from a database in response to a user's question and generates an appropriate answer using a generative AI model.

[0921] "Means for comparing different versions of the same document and identifying the most recent version" refers to devices or functions that analyze and compare multiple versions of the same document and identify the most recent version.

[0922] "Applications installed on automated equipment in factories" refers to applications that are installed on automated equipment (e.g., robots) in factories and that manage work knowledge and troubleshooting procedures.

[0923] "Means for managing work knowledge and troubleshooting procedures and providing quick answers based on prompt statements" refers to devices and functions that organize and manage knowledge about factory work and problem-solving procedures, and generate quick answers based on user prompt statements.

[0924] This is a system for realizing applications for factory robots that efficiently manage work knowledge and troubleshooting procedures within factories and provide quick answers.

[0925] System configuration

[0926] 1. Technical configuration

[0927] The hardware used is a Linux-based server (e.g., AWS EC2, Google Cloud Compute Engine), Amazon S3 for file storage, and MongoDB for the database.

[0928] The software used is a web application server using Flask, various libraries for data analysis, and OpenAI GPT-3 as a generative AI model.

[0929] 2. Uploading and analyzing materials

[0930] Users can upload documents via the factory robot's interface (e.g., screen or console). Document formats include PDF, Word, PPT, MP4, etc.

[0931] The server receives the uploaded materials, automatically analyzes their contents, and assigns appropriate classification tags, such as "factory maintenance" or "troubleshooting."

[0932] 3. Building a knowledge database

[0933] The server converts the parsed content into a reusable format (e.g., text data or JSON data) and stores it in a database, which will serve as the foundation for future knowledge searches.

[0934] If different versions of the same material are uploaded, the server automatically compares them and identifies the most recent version, marking it in the database.

[0935] 4. Question and Answer Generation

[0936] The user inputs a question to the factory robot, for example, using a prompt such as "Tell me about yesterday's maintenance procedures."

[0937] The server sends the question to a generative AI model (e.g., OpenAI GPT-3), which searches for relevant information in a database and generates an appropriate answer, which is then provided to the user through the factory robot's interface.

[0938] 5. Specific Examples

[0939] Uploading materials: A worker uploads a presentation (e.g., maintenance procedures) to the robot's interface. The system assigns a tag such as "Factory Maintenance" and stores it in a database.

[0940] Example of question-and-answering: A worker asks the robot, "What were the maintenance procedures yesterday?" The system searches the database for relevant documents and generates and provides an answer such as, "Here are the maintenance procedures yesterday."

[0941] Example prompt sentence:

[0942] Based on your knowledge database, generate an appropriate answer to the question, "What was the maintenance procedure yesterday?"

[0943] This system automates knowledge management within factories, facilitating the sharing and reuse of work knowledge and troubleshooting procedures. In addition, by utilizing generative AI models, it can quickly provide appropriate information in emergencies, improving factory efficiency and safety.

[0944] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0945] Step 1:

[0946] The user operates the factory robot's interface to upload a file (e.g., PDF, Word, PPT, MP4). The user presses the file selection button and selects the file they want to upload. This input sends the file to the server.

[0947] Step 2:

[0948] The server receives the uploaded material, identifies the format of the received material file, and prepares it for appropriate processing. Here, different analysis modules are selected depending on the file format (e.g., document file, presentation file, video file, etc.).

[0949] Step 3:

[0950] The server automatically analyzes the content of the file. For documents and presentations, it extracts text and image information. For video files, it converts audio to text and analyzes the video content to extract key points. The output of this process is the analyzed data.

[0951] Step 4:

[0952] The server automatically assigns relevant classification tags to the analyzed data, such as "factory maintenance" or "troubleshooting." This tagging allows the materials to be organized into categories.

[0953] Step 5:

[0954] The server analyzes and converts the tagged content into a reusable format (text data, JSON data, etc.), which is then stored in a database, making it easier to search for the material.

[0955] Step 6:

[0956] The user inputs a question into the factory robot's interface, for example, a prompt such as "Tell me about yesterday's maintenance procedures." This input is sent to the server.

[0957] Step 7:

[0958] The server sends the user's question to a generative AI model (e.g., OpenAI GPT-3), which analyzes the question. The server searches for relevant information in a knowledge database and uses the generative AI model to generate an appropriate answer. The output of this process is an answer to the user's question.

[0959] Step 8:

[0960] The server sends the generated answer to the user's device, which receives the answer and displays it to the user through the factory robot's interface. The user can refer to the displayed answer to proceed with the work.

[0961] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0962] The present invention is a system for managing, analyzing, and reusing materials, and recognizing user emotions. Specific embodiments for carrying out the present invention and the processing contents thereof will be described below.

[0963] System configuration

[0964] 1. Uploading materials

[0965] Users access the Knowledge Mate interface using their devices and upload documents and video files, including document files (PDF, Word), presentation files (PPT), and video files (MP4).

[0966] 2. Receiving the file

[0967] The server receives the uploaded file, identifies the file type (document, presentation, video, etc.), and prepares it for appropriate processing.

[0968] 3. Automatic organization and tagging

[0969] The server automatically analyzes the content of the files and assigns relevant classification tags, such as marketing materials, technical documentation, annual reports, etc. This process organizes the materials into categories.

[0970] 4. Content Analysis and Transformation

[0971] The server uses a generative AI engine to analyze the contents of the uploaded file. For documents and presentations, elements such as text, images, and graphs are extracted. For video files, the audio is converted to text and the video content is analyzed to extract key points.

[0972] 5. Saving to the database

[0973] The server then formats the parsed and converted content into a reusable format (e.g., text data, JSON data, etc.) and stores it in a database, which serves as the foundation for future knowledge search and reuse.

[0974] 6. User Emotion Recognition

[0975] The device sends the user's input or voice to the emotion engine, which analyzes it and recognizes the user's emotion (e.g., joy, anger, sadness, surprise, etc.).

[0976] 7. Knowledge Search and Retrieval

[0977] The user uses a device to input a question to the generation AI, for example, "Please show me materials related to last year's marketing strategy."

[0978] 8. Emotion-Based Response Modulation

[0979] The server sends the user's question to the generation AI, which then uses Retrieval-Augmented Generation (RAG) to search for relevant knowledge in the database. At the same time, the emotion engine recognizes the user's emotional state and adjusts the response accordingly.

[0980] 9. Generating Answers to Questions

[0981] Based on the knowledge extracted by the generative AI and the analysis results of the emotion engine, the system generates an appropriate answer to the user's question. For example, it creates a response such as, "Here are last year's marketing strategy documents. Please let me know if you have any additional questions."

[0982] 10. Versioning and Identifying the Latest Version

[0983] The server searches the database for different versions of the same document and uses generative AI to analyze the differences in content.

[0984] The server uses the analysis to identify the most up-to-date version and marks it in the database, which is then served whenever a user requests the most up-to-date document.

[0985] Specific examples

[0986] Example 1: Uploading and analyzing data

[0987] 1. User Action:

[0988] To upload "new presentation materials," a user logs into the Knowledge Mate portal site, selects the file, and then presses the upload button.

[0989] 2. Server processing:

[0990] The server receives the file, identifies its format, and, if it is a presentation, analyzes the slides and extracts text and image information.

[0991] The server assigns tags such as "Presentation Materials, Category: Marketing, Date: October 2023" and stores it in a database.

[0992] Example 2: Knowledge extraction using generative AI and emotion engines

[0993] 1. User Action:

[0994] The user inputs a question into the generation AI from their device, such as, "Do you have any materials related to last year's marketing strategy?" At the same time, the emotion engine recognizes the user's "confused" state from their tone of voice and text.

[0995] 2. Server processing:

[0996] The server receives the user's question, analyzes it using generative AI, and uses RAG to search for relevant marketing strategy materials in the database.

[0997] Based on the emotion of "confusion" recognized by the emotion engine, the generative AI generates a more polite and understandable response, such as, "Here are last year's marketing strategy documents. If you have any questions, please let us know as we have more detailed documents available."

[0998] 3. Response to the user:

[0999] The server sends this response to the user's terminal, allowing the user to access the information.

[1000] This invention is a system that significantly improves the efficiency of knowledge management and enables effective use of knowledge within a company. By incorporating an emotion engine, it provides appropriate responses that take into account the user's emotional state, creating a more user-friendly interface. This automates the organization, analysis, search, and version management of documents, making it even easier for users to utilize knowledge.

[1001] The processing flow will be explained below.

[1002] Step 1:

[1003] A user logs in to the Knowledge Mate interface using a terminal and clicks the upload button. The user selects and uploads a document file (e.g., PDF, Word), a presentation file (e.g., PPT), or a video file (e.g., MP4).

[1004] Step 2:

[1005] The server receives the uploaded file. The server identifies the file type (document, presentation, video, etc.) and prepares it for appropriate processing.

[1006] Step 3:

[1007] The server automatically assigns classification tags to files it receives, automatically determining which category a file belongs to (e.g., "marketing materials," "technical documentation," "annual reports," etc.) and assigning the tag.

[1008] Step 4:

[1009] The server uses a generated AI engine to analyze the contents of the uploaded file. For documents and presentations, elements such as text, images, and graphs are extracted. For video files, audio is converted to text and the video content is analyzed to extract key points.

[1010] Step 5:

[1011] The server formats the parsed content into a reusable format (text data, JSON data, etc.) and saves the formatted data in a database.

[1012] Step 6:

[1013] The user inputs a question into the generation AI from their device, for example, "Please show me materials related to last year's marketing strategy."

[1014] Step 7:

[1015] The server receives the user's question and sends it to the generation AI. At the same time, the device sends the emotion information from the user's tone of voice and input sentences to the emotion engine, which then analyzes the user's emotion (e.g., "confusion," "joy," "anger," etc.).

[1016] Step 8:

[1017] The generation AI uses Retrieval-Augmented Generation (RAG) to search for relevant knowledge in the database, and the server adjusts the tone and content of responses based on the analysis results from the emotion engine.

[1018] Step 9:

[1019] The generative AI generates appropriate answers to user questions based on the extracted knowledge and the analysis results of the emotion engine. For example, it creates a response such as, "Last year's marketing strategy materials are here. If you have any questions, please let us know as we have more detailed materials available."

[1020] Step 10:

[1021] The server sends the generated response to the user's terminal, allowing the user to access the information.

[1022] Step 11:

[1023] The server searches the database for different versions of the same document and uses generative AI to analyze the differences in content. Based on the analysis results, it identifies the most recent version with the most up-to-date content and marks it in the database. This latest version is always provided when a user requests the document.

[1024] Example 2

[1025] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1026] Conventional document management systems required a great deal of time and effort to organize and search documents, and it was difficult to provide appropriate responses that took user sentiment into account. As a result, documents could not be reused or information provided promptly, leading to problems that reduced work efficiency. Furthermore, the mixing of different versions of documents created the problem of cumbersome management of the latest versions.

[1027] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1028] In this invention, the server includes means for receiving materials, means for automatically analyzing the content of the materials and assigning classification tags, means for converting the analyzed content into a reusable format and storing it in a database, means for extracting relevant information from the database in response to a question and generating an answer using a generative AI model, means for comparing different versions of the same material and identifying the latest version, means for analyzing user sentiment and adjusting answers based on the analysis results, and means for converting voice to text and extracting key points from video when analyzing the content of the materials. This enables efficient material management, rapid information provision, user-friendly responses, and accurate provision of the latest materials.

[1029] "Means for receiving materials" refers to the function by which the server receives and stores materials uploaded by users, including identifying file formats and temporarily storing them.

[1030] "Means for automatically analyzing the content of materials and assigning classification tags" is a function that analyzes the content of materials and automatically assigns tags according to their characteristics and categories. This allows materials to be properly classified and managed.

[1031] "Means of converting the analyzed content into a reusable format and saving it in a database" refers to the function of formatting the information extracted from the materials into a format that is easy to reuse and saving that content in a database, making it easier to search and reuse in the future.

[1032] "Means of extracting relevant information from a database in response to a question and generating an answer using a generative AI model" refers to a function that searches for relevant information in a database in response to a question from a user and generates an appropriate answer using generative AI.

[1033] "Means for comparing different versions of the same document and identifying the latest version" refers to a function that compares different versions of the same document, analyzes the changes, and identifies the latest version. This allows the latest version to be provided at all times.

[1034] "Means for analyzing the user's emotions and adjusting responses based on the analysis results" refers to a function that analyzes the user's input and voice data to recognize their emotional state and generates a response based on that emotion, thereby providing the user with an appropriate and empathetic response.

[1035] "Means for converting audio to text and extracting key points from video when analyzing the content of materials" refers to a function that converts video and audio data into text and automatically extracts important points from the video content.

[1036] The present invention is a system for managing, analyzing, and reusing materials, and for recognizing user emotions. The system aims to improve user convenience by automatically organizing materials and generating appropriate responses. Specific embodiments for implementing the present invention are described below.

[1037] System configuration

[1038] Uploading materials

[1039] The user uploads documents using the terminal. Specifically, the user selects files such as PDF, Word, PPT, MP4, etc. through the web interface and presses the upload button. At this time, the terminal organizes the uploaded files into a specific format.

[1040] Receiving files

[1041] The server receives the uploaded files from the device, automatically identifies their format, analyzes the metadata of each file, and prepares them for temporary storage.

[1042] Auto-organization and tagging

[1043] The server automatically analyzes the content of the documents and assigns relevant classification tags, such as marketing materials, technical documents, annual reports, etc. This process is performed using a generative AI engine (e.g., GPT-4). The documents are organized into categories.

[1044] Content analysis and transformation

[1045] The server uses a generative AI engine to perform a detailed analysis of the uploaded file's contents. It extracts text, images, and graphs from document files, converts audio to text from video files, and analyzes the video content to extract key points. This information is then formatted for reuse.

[1046] Saving to a database

[1047] The server stores the parsed and converted content in a database in a reusable format (e.g. text data, JSON data), which serves as the basis for future knowledge search and reuse.

[1048] User Emotion Recognition

[1049] The device sends the user's input and voice to an emotion engine (e.g., EmotionAI) that analyzes the user's emotions. The emotion engine recognizes emotions such as joy, anger, sadness, and surprise.

[1050] Knowledge Search and Retrieval

[1051] The user uses a device to input a question to the generation AI, for example, "Please show me materials related to last year's marketing strategy."

[1052] Emotion-Based Response Modulation

[1053] The server receives the user's question and emotional information, and requests the generation AI to analyze it. It uses Retrieval-Augmented Generation (RAG) to search for relevant knowledge in the database. At the same time, it adjusts an appropriate response based on the analysis results of the emotion engine.

[1054] Generate answers to questions

[1055] Based on the knowledge extracted by the generative AI and the analysis results of the emotion engine, the system generates an appropriate answer to the user's question. For example, it creates a response such as, "Here are last year's marketing strategy documents. Please let me know if you have any additional questions."

[1056] Versioning and identifying the latest version

[1057] The server searches the database for different versions of the same document and uses generative AI to analyze the differences in content. Based on the analysis results, it identifies the most recent version with the most up-to-date content and marks it in the database. This latest version is always provided when a user requests the document.

[1058] Specific examples

[1059] Example 1: Uploading and analyzing data

[1060] 1. User operation: To upload "new presentation materials," the user logs in to the portal site, selects the file, and then presses the upload button.

[1061] 2. Server processing: The server receives the file, identifies its format, and if it is a presentation, analyzes the slides and extracts text and image information.

[1062] 3. Tagging: The server assigns tags such as "Presentation Materials, Category: Marketing, Date: October 2023" and stores them in the database.

[1063] Example 2: Knowledge extraction using generative AI and emotion engines

[1064] 1. User operation: The user inputs a question into the generation AI from their device, asking, "Do you have any materials about last year's marketing strategy?" At the same time, the emotion engine recognizes the user's "confused" state from their tone of voice and text.

[1065] 2. Server processing: The server receives the user's question, analyzes it using the generative AI, and uses RAG to search for relevant marketing strategy materials in the database.

[1066] 3. Emotional response: Based on the emotion of "confusion" recognized by the emotion engine, the generative AI generates a more polite and understandable response, such as, "Here are last year's marketing strategy documents. If you have any questions, please let us know as we have more detailed documents available."

[1067] 4. Reply to the user: The server sends this answer to the user's device, allowing the user to access the information.

[1068] Hardware and software used

[1069] Hardware: Servers, terminals (PCs, smartphones)

[1070] Software: Generative AI engine (e.g., GPT-4), emotion engine (e.g., EmotionAI), database management system (e.g., MySQL), Knowledge Mate interface (web application)

[1071] Prompt Sentence Examples

[1072] 1. "Please find materials related to last year's marketing strategy."

[1073] 2. "Upload a new presentation."

[1074] 3. "I'm having trouble and need help. What resources should I look at?"

[1075] This system not only significantly improves the efficiency of knowledge management, but also realizes a user-friendly interface by providing responses that respond to the user's emotions. By automating the organization, analysis, search, and version management of materials, users can utilize knowledge more easily and effectively.

[1076] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1077] Step 1:

[1078] The user uploads the document using a terminal. Specifically, the user accesses the web interface, selects the file, and then presses the upload button. The input is a file in a format such as PDF, Word, PPT, or MP4, and the output is the file data sent to the server.

[1079] Step 2:

[1080] The server receives the file uploaded by the user. The server automatically identifies the file format and temporarily stores it. In this step, the input is the file data sent by the user, and the output is the identified file format and the temporarily stored file.

[1081] Step 3:

[1082] The server automatically analyzes the content of the received material and assigns classification tags. It uses a generative AI engine (e.g., GPT-4) to analyze the text, images, and other elements in the file. In this step, the input is the temporarily saved file data, and the output is the assigned classification tag (e.g., marketing material, technical document, etc.).

[1083] Step 4:

[1084] The server converts the content into a reusable format, for example extracting text data from a document or converting audio to text from a video. In this step, the input is the parsed content and the output is formatted data in a reusable format (e.g. text data, JSON data).

[1085] Step 5:

[1086] The server stores the data in a reusable format in a database. This process involves database manipulation using SQL queries. In this step, the input is the data in a reusable format and the output is the data stored in the database.

[1087] Step 6:

[1088] The device sends the user's input or voice to the emotion engine, which uses a model like EmotionAI. In this step, the input is the user's input text or voice data, and the output is the analyzed user's emotion (e.g., joy, anger, sadness).

[1089] Step 7:

[1090] The user inputs a question to the generative AI, for example, "Do you have any materials about last year's marketing strategy?" In this step, the input is the question text and the output is the request data sent to the server.

[1091] Step 8:

[1092] The server sends the question to the generation AI, which uses Retrieval-Augmented Generation (RAG) to search for relevant information in the database. It also adjusts the response based on information from the emotion engine. In this step, the input is the user's question and emotion information, and the output is the answer adjusted by the generation AI.

[1093] Step 9:

[1094] The generative AI generates an answer based on the extracted knowledge and sentiment analysis. Specifically, it generates a response such as, "Here are last year's marketing strategy documents. Please let me know if you have any additional questions." In this step, the input is information from the database and the results of sentiment analysis, and the output is the generated answer text.

[1095] Step 10:

[1096] The server compares different versions of the material and identifies the latest version. Generative AI is used to analyze the content of the different versions, identify the latest version with the most up-to-date content, and mark it in the database. In this step, the input is the material data of the different versions, and the output is the identified latest version of the material.

[1097] (Application example 2)

[1098] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1099] Conventional document management systems lack functionality such as navigation assistance and real-time analysis of the driver's emotional state, and technologies to improve driving safety and convenience are needed, especially for autonomous vehicles. There is also a need for technologies to properly grasp the user's emotional state while driving and provide appropriate feedback.

[1100] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving materials, means for automatically analyzing the content of the materials and assigning classification tags, means for converting the analyzed content into a reusable format and saving it in a database, means for retrieving related information from the database in response to a question and generating an answer, means for comparing different versions of the same material and identifying the latest version, means for uploading map data and destination information, means for analyzing the user's emotional state in real time, and means for providing navigation assistance information based on the analysis results. This makes it possible to appropriately grasp the driver's emotional state in an autonomous vehicle and provide feedback and warnings accordingly.

[1101] "Means for receiving materials" refers to devices or software that allow the server to receive materials in various formats (e.g., documents, presentations, video files) provided by users.

[1102] The "means for assigning classification tags" refers to a device or software that automatically analyzes the content of received materials and assigns tags indicating appropriate categories or attributes to the materials.

[1103] "Means for converting into a reusable format and storing in a database" refers to a device or software that converts the content of the analyzed material into a reusable format (e.g., text data, JSON data) and stores it in a database.

[1104] "Means for retrieving relevant information from a database in response to a question and generating an answer" refers to a device or software that searches a database for relevant information in response to a question from a user and generates an appropriate answer.

[1105] A "means for comparing different versions of the same material and identifying the most recent version" is a device or software that compares multiple versions of the same material and identifies the version with the most recent content.

[1106] The "means for uploading map data and destination information" refers to a device or software that uploads map data and destination information entered by the user to the system.

[1107] The "means for analyzing the user's emotional state in real time" refers to a device or software that uses a sensor, camera, microphone, etc. to analyze the user's emotional state in real time.

[1108] The "means for providing navigation assistance information" is a device or software that provides appropriate navigation assistance information (e.g., route directions, warning messages) to the driver based on the analyzed information and the user's emotional state.

[1109] To put the present invention into practice, a document management system and a navigation support system are integrated, as will be described in detail below.

[1110] System configuration

[1111] The system mainly consists of a server, a user terminal, smart glasses, and a head-mounted display (HMD).

[1112] User operations

[1113] Users input destination information using smart glasses or an HMD and upload it to the vehicle system, where it is received and analyzed by a server.

[1114] Server Processing

[1115] The server receives map data and destination information uploaded by users using a means for receiving data. Next, the server analyzes this data using a means for assigning classification tags, and assigns tags according to its characteristics. The analyzed content is then converted into a reusable format and managed using a means for storing it in a database. For example, tags such as "long-distance drive" and "intra-city travel" are assigned.

[1116] Real-time analytics

[1117] To analyze the user's emotional state, sensors, cameras, and microphones built into the smart glasses or HMD are used. This allows the user's emotional state to be transmitted to a server in real time for analysis. Depending on the emotional state, a means for providing navigation assistance information is activated, and appropriate feedback and warnings are provided to the user.

[1118] Navigation Aids

[1119] The server generates navigation assistance information based on map data and destination information. Based on the analysis results and the user's emotional state, specific driving instructions such as "Turn right at the next intersection" or emotional feedback such as "Drive calmly and without rushing" are displayed on the smart glasses or HMD.

[1120] Specific examples

[1121] For example, if a user inputs "Drive to Tokyo Station," the server receives this data and tags it as "Long-distance drive." At the same time, if the server detects an emotion of "impatience" from the user's tone of voice or facial expression, it displays emotional feedback such as "Please drive calmly."

[1122] Example prompts

[1123] Here is an example prompt:

[1124] Analyze the following map data and tag it appropriately:

[1125] "Destination: Tokyo Station, looking for a long drive"

[1126] Also, analyze the emotional state from the following audio data and provide appropriate feedback:

[1127] "You seem anxious."

[1128] Hardware and Software Examples

[1129] The hardware used includes smart glasses (e.g., Google Glass), head-mounted displays (e.g., Microsoft HoloLens), and in-vehicle servers (with high-performance processors).The software used includes Python scripts for data analysis and emotion recognition, and TensorFlow or PyTorch for generative AI and emotion analysis.

[1130] As described above, the present invention integrates a document management system and a navigation assistance system, allowing users to receive real-time emotion recognition and navigation assistance while driving.

[1131] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1132] Step 1:

[1133] The user uses smart glasses or a head-mounted display (HMD) to input destination information and upload it to the vehicle system. The input data is in the form of, for example, "I would like to drive to Tokyo Station." This information is then sent to the server via the terminal.

[1134] Step 2:

[1135] The server analyzes the received map data and destination information. The input data is "I would like to drive to Tokyo Station." The server uses a generative AI model to analyze the input data and tag it (e.g., "long-distance drive"). The tagged data is obtained as output.

[1136] Step 3:

[1137] The server converts the parsed data into a reusable format (e.g. text data, JSON data) and stores it in the database. The tagged data as input is processed and converted into a format that can be stored in the database. The converted data as output is stored in the database.

[1138] Step 4:

[1139] The smart glasses and HMD use built-in sensors and microphones to analyze the user's emotional state in real time. Voice and facial expressions are sent as input data to the server, which then analyzes them using an emotion recognition algorithm. The analysis results (e.g., "anxiety") are obtained as output.

[1140] Step 5:

[1141] The server generates appropriate feedback or warnings based on the analysis results. It processes the emotion analysis results (e.g., "impatience") as input and generates a feedback message (e.g., "Please drive calmly"). The generated feedback is obtained as output.

[1142] Step 6:

[1143] The server provides navigation assistance information. Based on input data such as map data, destination information, and emotion analysis results, driving assistance information (e.g., "Please turn right at the next intersection") is generated. The output is displayed on the user's smart glasses or HMD.

[1144] Step 7:

[1145] The server stores the generated navigation assistance information and emotional feedback in a database. The server converts the generated information as input data into a reusable format and stores it in the database. The stored data is obtained as output.

[1146] Through the above steps, the present invention provides navigation assistance and emotional feedback based on the destination information entered by the user and the emotional state analyzed in real time, and this information is stored in a database for future analysis and improvement.

[1147] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1148] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1149] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1150] [Fourth embodiment]

[1151] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1152] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1153] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1154] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1155] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1156] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1157] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1158] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1159] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1160] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1161] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1162] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1163] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1164] The present invention is a system that efficiently manages materials and converts that knowledge into a reusable form using generative AI. A specific embodiment of the system and the processing flow are described below.

[1165] System configuration

[1166] 1. Uploading materials

[1167] Users access the Knowledge Mate interface using their devices and upload documents and video files, including document files (PDF, Word), presentation files (PPT), and video files (MP4).

[1168] 2. Receiving the file

[1169] The server receives the uploaded file, identifies the file type (document, presentation, video, etc.), and prepares it for appropriate processing.

[1170] 3. Automatic organization and tagging

[1171] The server automatically analyzes the content of the files and assigns relevant classification tags, such as marketing materials, technical documentation, annual reports, etc. This process organizes the materials into categories.

[1172] 4. Content Analysis and Transformation

[1173] The server uses the generated AI to analyze the contents of the uploaded file. For documents and presentations, elements such as text, images, and graphs are extracted. For video files, the audio is converted to text and the video content is analyzed to extract key points.

[1174] 5. Saving to the database

[1175] The server then formats the parsed and converted content into a reusable format (e.g., text data, JSON data, etc.) and stores it in a database, which serves as the foundation for future knowledge search and reuse.

[1176] 6. Knowledge Search and Retrieval

[1177] The user uses a device to input a question to the generation AI, for example, "Please show me materials related to last year's marketing strategy."

[1178] The server analyzes the user's question using generative AI and searches for relevant knowledge in the database using RAG (Retrieval-Augmented Generation).

[1179] 7. Generating Answers to Questions

[1180] Based on the knowledge extracted by the generative AI, it generates appropriate answers to user questions, such as "Last year's marketing strategy materials can be found here."

[1181] The server sends the generated response to the user's terminal, allowing the user to access the information.

[1182] 8. Versioning and Identifying the Latest Version

[1183] The server searches the database for different versions of the same document and uses generative AI to analyze the differences in content.

[1184] The server uses the analysis to identify the most up-to-date version and marks it in the database, which is then served whenever a user requests the most up-to-date document.

[1185] Specific examples

[1186] Example 1: Uploading and analyzing data

[1187] 1. User Action:

[1188] To upload "new presentation materials," a user logs into the Knowledge Mate portal site, selects the file, and then presses the upload button.

[1189] 2. Server processing:

[1190] The server receives the file, identifies its format, and, if it is a presentation, analyzes the slides and extracts text and image information.

[1191] The server assigns tags such as "Presentation Materials, Category: Marketing, Date: October 2023" and stores it in a database.

[1192] Example 2: Extracting knowledge using generative AI

[1193] 1. User Action:

[1194] The user inputs a question into the generation AI from their device, asking, "Do you have any materials related to last year's marketing strategy?"

[1195] 2. Server processing:

[1196] The server receives the user's question, analyzes it using generative AI, and uses RAG to search for relevant marketing strategy materials in the database.

[1197] The server identifies the relevant materials, and the generation AI generates a response such as, "We found last year's marketing strategy materials. This is the latest version."

[1198] The server sends this response to the user, who can then view the information on their terminal.

[1199] This invention is a system that significantly improves the efficiency of knowledge management and realizes effective use of knowledge within a company. This automates the organization, analysis, search, and version management of documents, making it easier for users to utilize knowledge.

[1200] The processing flow will be explained below.

[1201] Step 1:

[1202] A user logs in to the Knowledge Mate interface and clicks the upload button. The user selects and uploads a document file (PDF, Word), a presentation file (PPT), or a video file (MP4).

[1203] Step 2:

[1204] The server receives the uploaded file and identifies the file format, determining whether it is a document, presentation, video, etc.

[1205] Step 3:

[1206] The server automatically assigns classification tags to uploaded files, identifying the category to which the file belongs, such as "marketing materials," "technical documentation," or "annual reports."

[1207] Step 4:

[1208] The server sends the file contents to the generative AI engine. If it's a document or presentation, the generative AI extracts elements like text, images, and graphs. If it's a video file, it converts the audio into text and analyzes the video to extract key points.

[1209] Step 5:

[1210] The server formats the parsed and converted content into a reusable format (text data, JSON data, etc.) and saves the formatted data in a database.

[1211] Step 6:

[1212] The user uses a device to input a question to the generation AI, for example, "Please show me materials related to last year's marketing strategy."

[1213] Step 7:

[1214] The server sends the user's question to the generation AI, which uses Retrieval-Augmented Generation (RAG) to search for relevant knowledge in the database.

[1215] Step 8:

[1216] Generative AI generates answers based on search results, such as "Here are last year's marketing strategy documents."

[1217] Step 9:

[1218] The server sends the generated response to the user's terminal, allowing the user to access the information.

[1219] Step 10:

[1220] The server searches the database for different versions of the same document and uses generative AI to analyze the differences in content. Based on the analysis results, it identifies the most recent version with the most up-to-date content and marks it in the database. This latest version is always provided when a user requests the document.

[1221] Example 1

[1222] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1223] Current knowledge management systems require a great deal of time and effort to manually classify and organize documents and search and retrieve knowledge. Furthermore, when managing different versions of the same document, it is difficult to identify the latest version. This makes it difficult for users to quickly and accurately obtain the information they need.

[1224] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1225] In this invention, the server includes: means for users to upload materials using a computing device; means for receiving the materials and identifying their format; means for analyzing the file content and automatically assigning classification tags; means for converting the analyzed content into a reusable format and storing it in a database; means for users to input questions in natural language; means for using a generative AI model to extract relevant information from a database and generate an answer to the question; means for transmitting the generated answer to the user's computing device; and means for comparing different versions of the same material and identifying the latest version. This facilitates automatic organization, analysis, search, and version management of materials, enabling users to efficiently obtain the knowledge they need.

[1226] "User" means an individual or organization that uses the system to upload materials or search for knowledge.

[1227] A "computing device" is a device such as a calculator, PC, smartphone, or tablet that a user uses to upload materials or search for knowledge.

[1228] "Materials" refers to a collection of information, such as documents, presentations, and video files, that can be managed, analyzed, and searched.

[1229] A "generative AI model" is an artificial intelligence framework that uses deep learning technology to perform natural language processing and data analysis to generate and provide necessary information.

[1230] "Server" means a computer system that receives, analyzes, and stores uploaded materials and provides information in response to user requests.

[1231] "Means for identifying formats" refers to a function that automatically detects the file format of uploaded materials and performs appropriate processing.

[1232] "Tags" are keywords or labels that are assigned to indicate the classification or characteristics of materials.

[1233] "Knowledge" is a set of knowledge or information extracted from information stored in a database and provided in response to a user's question.

[1234] A "database" is a digital information repository for storing and managing analyzed materials and generated knowledge.

[1235] "Version control" is the process of distinguishing between different versions of the same material and identifying and managing the most recent version.

[1236] The present invention provides a system for efficiently managing materials and enabling knowledge reuse by generative AI. Specific embodiments of the system are described below.

[1237] Uploading materials

[1238] A user accesses the KnowledgeMate interface using a computing device, selects materials such as documents (PDF, Word), presentations (PPT), and video files (MP4), and presses the upload button, which sends the materials to the server.

[1239] Receiving files

[1240] The server receives the uploaded materials from the user and identifies the file format, for example, the server checks the file extension and automatically classifies the format as ".pdf", ".docx", ".pptx", ".mp4", etc.

[1241] Auto-organization and tagging

[1242] Once the file is received, the server automatically analyzes the file contents using natural language processing (NLP) technology, such as PyMuPDF for extracting text from PDF files, python-docx for extracting text from Word files, and python-pptx for extracting text from presentation files. Based on the analysis results, the server then assigns classification tags, such as marketing materials, technical documents, and annual reports, to make it easier to organize the documents.

[1243] Content analysis and transformation

[1244] The server then uses generative AI models to further analyze the content of the uploaded files. For text files, topic modeling techniques are used to extract important keywords. For video files, a speech recognition engine (e.g., Google Cloud Speech-to-Text) is used to convert the audio into text and identify important scenes and key points.

[1245] Saving to a database

[1246] The parsed and converted content is formatted into a reusable format (e.g., JSON, text data). The server creates a JSON object with fields such as "Document ID," "Title," "Content," and "Tags" as text data and stores it in a database. This database is built using, for example, "MongoDB" or "PostgreSQL."

[1247] Knowledge Search and Retrieval

[1248] A user inputs a prompt sentence into the generative AI model using a terminal. For example, if a request is made such as "Please tell me about materials related to last year's marketing strategy," the server receives the prompt sentence and analyzes it using the generative AI model. The server then uses Retrieval-Augmented Generation (RAG) technology to search for related knowledge in the database.

[1249] The generative AI model generates an appropriate answer to the user's question, such as "Here is last year's marketing strategy document," and the server sends this to the user's device.

[1250] Versioning and identifying the latest version

[1251] The server searches the database for different versions of the same document and uses a generative AI model to analyze the differences in content, particularly identifying differences and added information, and marks the version with the most recent content in the database, ensuring that the latest version is always served when a user requests the latest version.

[1252] Specific examples

[1253] Uploading and analyzing data

[1254] 1. User operation: To upload "new presentation materials," the user logs in to the Knowledge Mate portal site, selects the file "presentation_marketing_2023.pptx," and presses the upload button.

[1255] 2. Server processing: The server receives "presentation_marketing_2023.pptx" and identifies its format. The server uses the "python-pptx" library to analyze the slide content and extract text and image information. The server adds tags such as "Presentation Materials, Category: Marketing, Date: October 2023" and saves the analysis results in a database.

[1256] Extracting knowledge with generative AI

[1257] 1. User interaction: The user inputs a prompt into the generation AI from their device, asking, "Please tell me about materials related to last year's marketing strategy."

[1258] 2. Server processing: The server receives the user's prompt and uses a generative AI (e.g., OpenAI GPT-4) to analyze the prompt. It then uses RAG technology to search for relevant marketing strategy documents in a database and generates an answer from the documents found, such as "We've found last year's marketing strategy documents. This is the latest version." The server then sends this answer to the user, allowing them to view the documents on their device.

[1259] This allows users to efficiently obtain the knowledge they need, and makes it easier to manage materials and reuse information.

[1260] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1261] Step 1:

[1262] User operations

[1263] A user accesses the KnowledgeMate interface using a computing device, presses the upload button, and selects materials. The input materials can be documents (PDF, Word), presentations (PPT), video files (MP4), etc. The output is an upload request that is sent to the server.

[1264] Step 2:

[1265] Server Processing

[1266] The server processes the received upload request and receives the document. At this time, the server identifies the format of the document. It checks the extension of the received file and recognizes the format from ".pdf", ".docx", ".pptx", ".mp4", etc. The output is the format information of the document.

[1267] Step 3:

[1268] Server Processing

[1269] The server analyzes the file contents and automatically assigns classification tags. This process uses natural language processing (NLP) techniques and specific libraries (e.g., "PyMuPDF," "python-docx," "python-pptx"). The input is the received document itself, and the output is the analyzed text data and corresponding tag information. For example, in the case of a PDF file, the server generates tags such as "marketing material" and "technical document."

[1270] Step 4:

[1271] Server Processing

[1272] The server uses the generated AI model to further analyze and convert the content of the material. This process involves topic modeling and keyword extraction for text data, and in the case of video files, converting the audio into text using a speech recognition engine. The input is the analyzed text data, and the output is data formatted in a reusable format (e.g., JSON, text data). In the case of video files, the server extracts important scenes and key points.

[1273] Step 5:

[1274] Server Processing

[1275] The parsed and converted data is saved to a database. In this process, the data is stored in a reusable format in a database (such as MongoDB or PostgreSQL). The input is the formatted data, and the output is a response confirming that it has been saved to the database. The server creates a JSON object with fields such as "Document ID", "Title", "Content", and "Tags" that hold the text data.

[1276] Step 6:

[1277] User operations

[1278] A user inputs a question in natural language to the generative AI model using a terminal. For example, they input a prompt such as, "Please tell me about materials related to last year's marketing strategy." The input is the user's question, and the output is the generated prompt sentence.

[1279] Step 7:

[1280] Server Processing

[1281] The server uses a generative AI model to analyze the prompt and retrieve relevant information from the database. It then uses RAG technology to search for relevant knowledge in the database. The input is the prompt and the knowledge in the database, and the output is the retrieved relevant material.

[1282] Step 8:

[1283] Server Processing

[1284] The generative AI model generates an appropriate answer to the user's question. For example, it creates an answer text such as "Last year's marketing strategy materials are here." The input is the searched materials, and the output is the generated answer. The server sends this answer to the user's device, where it can be viewed by the user.

[1285] Step 9:

[1286] Server Processing

[1287] Different versions of the same document are searched for in the database and a generative AI model is used to analyze the differences in content. The input is the different versions of the document, and the output is a comparison of the old and new content based on that. The server then identifies the latest version and marks it in the database. Whenever a user requests the latest version, the latest version is always provided.

[1288] (Application example 1)

[1289] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1290] Currently, work knowledge and troubleshooting procedures within factories depend on the experience and knowledge of individual employees, making it difficult to share and reuse knowledge. Furthermore, in emergencies, appropriate information cannot be obtained quickly, which risks reducing the efficiency of the entire factory. Furthermore, when different versions of documents exist, it is time-consuming to identify which document is the latest version, which can lead to work being carried out based on incorrect information.

[1291] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1292] In this invention, the server includes means for receiving materials, means for automatically analyzing the content of the materials and assigning classification tags, means for converting the analyzed content into a reusable format and saving it in a database, means for retrieving related information from the database in response to a question and generating an answer using a generative AI model, means for comparing different versions of the same material and identifying the latest version, and means for managing work knowledge and troubleshooting procedures with an application installed in an automated device in a factory and providing quick answers based on prompts. This automates knowledge management within the factory and enables efficient and error-free information provision.

[1293] "Means for receiving materials" refers to the devices and functions that receive various materials from users and incorporate them into the system.

[1294] "Means for automatically analyzing the content of materials and assigning classification tags" refers to devices or functions that analyze the content of materials and automatically assign appropriate tags based on their characteristics and categories.

[1295] "Means for converting the analyzed content into a reusable format and storing it in a database" refers to a device or function that converts the content of the material into a reusable format (e.g., text data or JSON data) and stores it in a database.

[1296] "Means of retrieving relevant information from a database in response to a question and generating an answer using a generative AI model" refers to a device or function that searches for relevant information from a database in response to a user's question and generates an appropriate answer using a generative AI model.

[1297] "Means for comparing different versions of the same document and identifying the most recent version" refers to devices or functions that analyze and compare multiple versions of the same document and identify the most recent version.

[1298] "Applications installed on automated equipment in factories" refers to applications that are installed on automated equipment (e.g., robots) in factories and that manage work knowledge and troubleshooting procedures.

[1299] "Means for managing work knowledge and troubleshooting procedures and providing quick answers based on prompt statements" refers to devices and functions that organize and manage knowledge about factory work and problem-solving procedures, and generate quick answers based on user prompt statements.

[1300] This is a system for realizing applications for factory robots that efficiently manage work knowledge and troubleshooting procedures within factories and provide quick answers.

[1301] System configuration

[1302] 1. Technical configuration

[1303] The hardware used is a Linux-based server (e.g., AWS EC2, Google Cloud Compute Engine), Amazon S3 for file storage, and MongoDB for the database.

[1304] The software used is a web application server using Flask, various libraries for data analysis, and OpenAI GPT-3 as a generative AI model.

[1305] 2. Uploading and analyzing materials

[1306] Users can upload documents via the factory robot's interface (e.g., screen or console). Document formats include PDF, Word, PPT, MP4, etc.

[1307] The server receives the uploaded materials, automatically analyzes their contents, and assigns appropriate classification tags, such as "factory maintenance" or "troubleshooting."

[1308] 3. Building a knowledge database

[1309] The server converts the parsed content into a reusable format (e.g., text data or JSON data) and stores it in a database, which will serve as the foundation for future knowledge searches.

[1310] If different versions of the same material are uploaded, the server automatically compares them and identifies the most recent version, marking it in the database.

[1311] 4. Question and Answer Generation

[1312] The user inputs a question to the factory robot, for example, using a prompt such as "Tell me about yesterday's maintenance procedures."

[1313] The server sends the question to a generative AI model (e.g., OpenAI GPT-3), which searches for relevant information in a database and generates an appropriate answer, which is then provided to the user through the factory robot's interface.

[1314] 5. Specific Examples

[1315] Uploading materials: A worker uploads a presentation (e.g., maintenance procedures) to the robot's interface. The system assigns a tag such as "Factory Maintenance" and stores it in a database.

[1316] Example of question-and-answering: A worker asks the robot, "What were the maintenance procedures yesterday?" The system searches the database for relevant documents and generates and provides an answer such as, "Here are the maintenance procedures yesterday."

[1317] Example prompt sentence:

[1318] Based on your knowledge database, generate an appropriate answer to the question, "What was the maintenance procedure yesterday?"

[1319] This system automates knowledge management within factories, facilitating the sharing and reuse of work knowledge and troubleshooting procedures. In addition, by utilizing generative AI models, it can quickly provide appropriate information in emergencies, improving factory efficiency and safety.

[1320] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1321] Step 1:

[1322] The user operates the factory robot's interface to upload a file (e.g., PDF, Word, PPT, MP4). The user presses the file selection button and selects the file they want to upload. This input sends the file to the server.

[1323] Step 2:

[1324] The server receives the uploaded material, identifies the format of the received material file, and prepares it for appropriate processing. Here, different analysis modules are selected depending on the file format (e.g., document file, presentation file, video file, etc.).

[1325] Step 3:

[1326] The server automatically analyzes the content of the file. For documents and presentations, it extracts text and image information. For video files, it converts audio to text and analyzes the video content to extract key points. The output of this process is the analyzed data.

[1327] Step 4:

[1328] The server automatically assigns relevant classification tags to the analyzed data, such as "factory maintenance" or "troubleshooting." This tagging allows the materials to be organized into categories.

[1329] Step 5:

[1330] The server analyzes and converts the tagged content into a reusable format (text data, JSON data, etc.), which is then stored in a database, making it easier to search for the material.

[1331] Step 6:

[1332] The user inputs a question into the factory robot's interface, for example, a prompt such as "Tell me about yesterday's maintenance procedures." This input is sent to the server.

[1333] Step 7:

[1334] The server sends the user's question to a generative AI model (e.g., OpenAI GPT-3), which analyzes the question. The server searches for relevant information in a knowledge database and uses the generative AI model to generate an appropriate answer. The output of this process is an answer to the user's question.

[1335] Step 8:

[1336] The server sends the generated answer to the user's device, which receives the answer and displays it to the user through the factory robot's interface. The user can refer to the displayed answer to proceed with the work.

[1337] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1338] The present invention is a system for managing, analyzing, and reusing materials, and recognizing user emotions. Specific embodiments for carrying out the present invention and the processing contents thereof will be described below.

[1339] System configuration

[1340] 1. Uploading materials

[1341] Users access the Knowledge Mate interface using their devices and upload documents and video files, including document files (PDF, Word), presentation files (PPT), and video files (MP4).

[1342] 2. Receiving the file

[1343] The server receives the uploaded file, identifies the file type (document, presentation, video, etc.), and prepares it for appropriate processing.

[1344] 3. Automatic organization and tagging

[1345] The server automatically analyzes the content of the files and assigns relevant classification tags, such as marketing materials, technical documentation, annual reports, etc. This process organizes the materials into categories.

[1346] 4. Content Analysis and Transformation

[1347] The server uses a generative AI engine to analyze the contents of the uploaded file. For documents and presentations, elements such as text, images, and graphs are extracted. For video files, the audio is converted to text and the video content is analyzed to extract key points.

[1348] 5. Saving to the database

[1349] The server then formats the parsed and converted content into a reusable format (e.g., text data, JSON data, etc.) and stores it in a database, which serves as the foundation for future knowledge search and reuse.

[1350] 6. User Emotion Recognition

[1351] The device sends the user's input or voice to the emotion engine, which analyzes it and recognizes the user's emotion (e.g., joy, anger, sadness, surprise, etc.).

[1352] 7. Knowledge Search and Retrieval

[1353] The user uses a device to input a question to the generation AI, for example, "Please show me materials related to last year's marketing strategy."

[1354] 8. Emotion-Based Response Modulation

[1355] The server sends the user's question to the generation AI, which then uses Retrieval-Augmented Generation (RAG) to search for relevant knowledge in the database. At the same time, the emotion engine recognizes the user's emotional state and adjusts the response accordingly.

[1356] 9. Generating Answers to Questions

[1357] Based on the knowledge extracted by the generative AI and the analysis results of the emotion engine, the system generates an appropriate answer to the user's question. For example, it creates a response such as, "Here are last year's marketing strategy documents. Please let me know if you have any additional questions."

[1358] 10. Versioning and Identifying the Latest Version

[1359] The server searches the database for different versions of the same document and uses generative AI to analyze the differences in content.

[1360] The server uses the analysis to identify the most up-to-date version and marks it in the database, which is then served whenever a user requests the most up-to-date document.

[1361] Specific examples

[1362] Example 1: Uploading and analyzing data

[1363] 1. User Action:

[1364] To upload "new presentation materials," a user logs into the Knowledge Mate portal site, selects the file, and then presses the upload button.

[1365] 2. Server processing:

[1366] The server receives the file, identifies its format, and, if it is a presentation, analyzes the slides and extracts text and image information.

[1367] The server assigns tags such as "Presentation Materials, Category: Marketing, Date: October 2023" and stores it in a database.

[1368] Example 2: Knowledge extraction using generative AI and emotion engines

[1369] 1. User Action:

[1370] The user inputs a question into the generation AI from their device, such as, "Do you have any materials related to last year's marketing strategy?" At the same time, the emotion engine recognizes the user's "confused" state from their tone of voice and text.

[1371] 2. Server processing:

[1372] The server receives the user's question, analyzes it using generative AI, and uses RAG to search for relevant marketing strategy materials in the database.

[1373] Based on the emotion of "confusion" recognized by the emotion engine, the generative AI generates a more polite and understandable response, such as, "Here are last year's marketing strategy documents. If you have any questions, please let us know as we have more detailed documents available."

[1374] 3. Response to the user:

[1375] The server sends this response to the user's terminal, allowing the user to access the information.

[1376] This invention is a system that significantly improves the efficiency of knowledge management and enables effective use of knowledge within a company. By incorporating an emotion engine, it provides appropriate responses that take into account the user's emotional state, creating a more user-friendly interface. This automates the organization, analysis, search, and version management of documents, making it even easier for users to utilize knowledge.

[1377] The processing flow will be explained below.

[1378] Step 1:

[1379] A user logs in to the Knowledge Mate interface using a terminal and clicks the upload button. The user selects and uploads a document file (e.g., PDF, Word), a presentation file (e.g., PPT), or a video file (e.g., MP4).

[1380] Step 2:

[1381] The server receives the uploaded file. The server identifies the file type (document, presentation, video, etc.) and prepares it for appropriate processing.

[1382] Step 3:

[1383] The server automatically assigns classification tags to files it receives, automatically determining which category a file belongs to (e.g., "marketing materials," "technical documentation," "annual reports," etc.) and assigning the tag.

[1384] Step 4:

[1385] The server uses a generated AI engine to analyze the contents of the uploaded file. For documents and presentations, elements such as text, images, and graphs are extracted. For video files, audio is converted to text and the video content is analyzed to extract key points.

[1386] Step 5:

[1387] The server formats the parsed content into a reusable format (text data, JSON data, etc.) and saves the formatted data in a database.

[1388] Step 6:

[1389] The user inputs a question into the generation AI from their device, for example, "Please show me materials related to last year's marketing strategy."

[1390] Step 7:

[1391] The server receives the user's question and sends it to the generation AI. At the same time, the device sends the emotion information from the user's tone of voice and input sentences to the emotion engine, which then analyzes the user's emotion (e.g., "confusion," "joy," "anger," etc.).

[1392] Step 8:

[1393] The generation AI uses Retrieval-Augmented Generation (RAG) to search for relevant knowledge in the database, and the server adjusts the tone and content of responses based on the analysis results from the emotion engine.

[1394] Step 9:

[1395] The generative AI generates appropriate answers to user questions based on the extracted knowledge and the analysis results of the emotion engine. For example, it creates a response such as, "Last year's marketing strategy materials are here. If you have any questions, please let us know as we have more detailed materials available."

[1396] Step 10:

[1397] The server sends the generated response to the user's terminal, allowing the user to access the information.

[1398] Step 11:

[1399] The server searches the database for different versions of the same document and uses generative AI to analyze the differences in content. Based on the analysis results, it identifies the most recent version with the most up-to-date content and marks it in the database. This latest version is always provided when a user requests the document.

[1400] Example 2

[1401] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1402] Conventional document management systems required a great deal of time and effort to organize and search documents, and it was difficult to provide appropriate responses that took user sentiment into account. As a result, documents could not be reused or information provided promptly, leading to problems that reduced work efficiency. Furthermore, the mixing of different versions of documents created the problem of cumbersome management of the latest versions.

[1403] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1404] In this invention, the server includes means for receiving materials, means for automatically analyzing the content of the materials and assigning classification tags, means for converting the analyzed content into a reusable format and storing it in a database, means for extracting relevant information from the database in response to a question and generating an answer using a generative AI model, means for comparing different versions of the same material and identifying the latest version, means for analyzing user sentiment and adjusting answers based on the analysis results, and means for converting voice to text and extracting key points from video when analyzing the content of the materials. This enables efficient material management, rapid information provision, user-friendly responses, and accurate provision of the latest materials.

[1405] "Means for receiving materials" refers to the function by which the server receives and stores materials uploaded by users, including identifying file formats and temporarily storing them.

[1406] "Means for automatically analyzing the content of materials and assigning classification tags" is a function that analyzes the content of materials and automatically assigns tags according to their characteristics and categories. This allows materials to be properly classified and managed.

[1407] "Means of converting the analyzed content into a reusable format and saving it in a database" refers to the function of formatting the information extracted from the materials into a format that is easy to reuse and saving that content in a database, making it easier to search and reuse in the future.

[1408] "Means of extracting relevant information from a database in response to a question and generating an answer using a generative AI model" refers to a function that searches for relevant information in a database in response to a question from a user and generates an appropriate answer using generative AI.

[1409] "Means for comparing different versions of the same document and identifying the latest version" refers to a function that compares different versions of the same document, analyzes the changes, and identifies the latest version. This allows the latest version to be provided at all times.

[1410] "Means for analyzing the user's emotions and adjusting responses based on the analysis results" refers to a function that analyzes the user's input and voice data to recognize their emotional state and generates a response based on that emotion, thereby providing the user with an appropriate and empathetic response.

[1411] "Means for converting audio to text and extracting key points from video when analyzing the content of materials" refers to a function that converts video and audio data into text and automatically extracts important points from the video content.

[1412] The present invention is a system for managing, analyzing, and reusing materials, and for recognizing user emotions. The system aims to improve user convenience by automatically organizing materials and generating appropriate responses. Specific embodiments for implementing the present invention are described below.

[1413] System configuration

[1414] Uploading materials

[1415] The user uploads documents using the terminal. Specifically, the user selects files such as PDF, Word, PPT, MP4, etc. through the web interface and presses the upload button. At this time, the terminal organizes the uploaded files into a specific format.

[1416] Receiving files

[1417] The server receives the uploaded files from the device, automatically identifies their format, analyzes the metadata of each file, and prepares them for temporary storage.

[1418] Auto-organization and tagging

[1419] The server automatically analyzes the content of the documents and assigns relevant classification tags, such as marketing materials, technical documents, annual reports, etc. This process is performed using a generative AI engine (e.g., GPT-4). The documents are organized into categories.

[1420] Content analysis and transformation

[1421] The server uses a generative AI engine to perform a detailed analysis of the uploaded file's contents. It extracts text, images, and graphs from document files, converts audio to text from video files, and analyzes the video content to extract key points. This information is then formatted for reuse.

[1422] Saving to a database

[1423] The server stores the parsed and converted content in a database in a reusable format (e.g. text data, JSON data), which serves as the basis for future knowledge search and reuse.

[1424] User Emotion Recognition

[1425] The device sends the user's input and voice to an emotion engine (e.g., EmotionAI) that analyzes the user's emotions. The emotion engine recognizes emotions such as joy, anger, sadness, and surprise.

[1426] Knowledge Search and Retrieval

[1427] The user uses a device to input a question to the generation AI, for example, "Please show me materials related to last year's marketing strategy."

[1428] Emotion-Based Response Modulation

[1429] The server receives the user's question and emotional information, and requests the generation AI to analyze it. It uses Retrieval-Augmented Generation (RAG) to search for relevant knowledge in the database. At the same time, it adjusts an appropriate response based on the analysis results of the emotion engine.

[1430] Generate answers to questions

[1431] Based on the knowledge extracted by the generative AI and the analysis results of the emotion engine, the system generates an appropriate answer to the user's question. For example, it creates a response such as, "Here are last year's marketing strategy documents. Please let me know if you have any additional questions."

[1432] Versioning and identifying the latest version

[1433] The server searches the database for different versions of the same document and uses generative AI to analyze the differences in content. Based on the analysis results, it identifies the most recent version with the most up-to-date content and marks it in the database. This latest version is always provided when a user requests the document.

[1434] Specific examples

[1435] Example 1: Uploading and analyzing data

[1436] 1. User operation: To upload "new presentation materials," the user logs in to the portal site, selects the file, and then presses the upload button.

[1437] 2. Server processing: The server receives the file, identifies its format, and if it is a presentation, analyzes the slides and extracts text and image information.

[1438] 3. Tagging: The server assigns tags such as "Presentation Materials, Category: Marketing, Date: October 2023" and stores them in the database.

[1439] Example 2: Knowledge extraction using generative AI and emotion engines

[1440] 1. User operation: The user inputs a question into the generation AI from their device, asking, "Do you have any materials about last year's marketing strategy?" At the same time, the emotion engine recognizes the user's "confused" state from their tone of voice and text.

[1441] 2. Server processing: The server receives the user's question, analyzes it using the generative AI, and uses RAG to search for relevant marketing strategy materials in the database.

[1442] 3. Emotional response: Based on the emotion of "confusion" recognized by the emotion engine, the generative AI generates a more polite and understandable response, such as, "Here are last year's marketing strategy documents. If you have any questions, please let us know as we have more detailed documents available."

[1443] 4. Reply to the user: The server sends this answer to the user's device, allowing the user to access the information.

[1444] Hardware and software used

[1445] Hardware: Servers, terminals (PCs, smartphones)

[1446] Software: Generative AI engine (e.g., GPT-4), emotion engine (e.g., EmotionAI), database management system (e.g., MySQL), Knowledge Mate interface (web application)

[1447] Prompt Sentence Examples

[1448] 1. "Please find materials related to last year's marketing strategy."

[1449] 2. "Upload a new presentation."

[1450] 3. "I'm having trouble and need help. What resources should I look at?"

[1451] This system not only significantly improves the efficiency of knowledge management, but also realizes a user-friendly interface by providing responses that respond to the user's emotions. By automating the organization, analysis, search, and version management of materials, users can utilize knowledge more easily and effectively.

[1452] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1453] Step 1:

[1454] The user uploads the document using a terminal. Specifically, the user accesses the web interface, selects the file, and then presses the upload button. The input is a file in a format such as PDF, Word, PPT, or MP4, and the output is the file data sent to the server.

[1455] Step 2:

[1456] The server receives the file uploaded by the user. The server automatically identifies the file format and temporarily stores it. In this step, the input is the file data sent by the user, and the output is the identified file format and the temporarily stored file.

[1457] Step 3:

[1458] The server automatically analyzes the content of the received material and assigns classification tags. It uses a generative AI engine (e.g., GPT-4) to analyze the text, images, and other elements in the file. In this step, the input is the temporarily saved file data, and the output is the assigned classification tag (e.g., marketing material, technical document, etc.).

[1459] Step 4:

[1460] The server converts the content into a reusable format, for example extracting text data from a document or converting audio to text from a video. In this step, the input is the parsed content and the output is formatted data in a reusable format (e.g. text data, JSON data).

[1461] Step 5:

[1462] The server stores the data in a reusable format in a database. This process involves database manipulation using SQL queries. In this step, the input is the data in a reusable format and the output is the data stored in the database.

[1463] Step 6:

[1464] The device sends the user's input or voice to the emotion engine, which uses a model like EmotionAI. In this step, the input is the user's input text or voice data, and the output is the analyzed user's emotion (e.g., joy, anger, sadness).

[1465] Step 7:

[1466] The user inputs a question to the generative AI, for example, "Do you have any materials about last year's marketing strategy?" In this step, the input is the question text and the output is the request data sent to the server.

[1467] Step 8:

[1468] The server sends the question to the generation AI, which uses Retrieval-Augmented Generation (RAG) to search for relevant information in the database. It also adjusts the response based on information from the emotion engine. In this step, the input is the user's question and emotion information, and the output is the answer adjusted by the generation AI.

[1469] Step 9:

[1470] The generative AI generates an answer based on the extracted knowledge and sentiment analysis. Specifically, it generates a response such as, "Here are last year's marketing strategy documents. Please let me know if you have any additional questions." In this step, the input is information from the database and the results of sentiment analysis, and the output is the generated answer text.

[1471] Step 10:

[1472] The server compares different versions of the material and identifies the latest version. Generative AI is used to analyze the content of the different versions, identify the latest version with the most up-to-date content, and mark it in the database. In this step, the input is the material data of the different versions, and the output is the identified latest version of the material.

[1473] (Application example 2)

[1474] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1475] Conventional document management systems lack functionality such as navigation assistance and real-time analysis of the driver's emotional state, and technologies to improve driving safety and convenience are needed, especially for autonomous vehicles. There is also a need for technologies to properly grasp the user's emotional state while driving and provide appropriate feedback.

[1476] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving materials, means for automatically analyzing the content of the materials and assigning classification tags, means for converting the analyzed content into a reusable format and saving it in a database, means for retrieving related information from the database in response to a question and generating an answer, means for comparing different versions of the same material and identifying the latest version, means for uploading map data and destination information, means for analyzing the user's emotional state in real time, and means for providing navigation assistance information based on the analysis results. This makes it possible to appropriately grasp the driver's emotional state in an autonomous vehicle and provide feedback and warnings accordingly.

[1477] "Means for receiving materials" refers to devices or software that allow the server to receive materials in various formats (e.g., documents, presentations, video files) provided by users.

[1478] The "means for assigning classification tags" refers to a device or software that automatically analyzes the content of received materials and assigns tags indicating appropriate categories or attributes to the materials.

[1479] "Means for converting into a reusable format and storing in a database" refers to a device or software that converts the content of the analyzed material into a reusable format (e.g., text data, JSON data) and stores it in a database.

[1480] "Means for retrieving relevant information from a database in response to a question and generating an answer" refers to a device or software that searches a database for relevant information in response to a question from a user and generates an appropriate answer.

[1481] A "means for comparing different versions of the same material and identifying the most recent version" is a device or software that compares multiple versions of the same material and identifies the version with the most recent content.

[1482] The "means for uploading map data and destination information" refers to a device or software that uploads map data and destination information entered by the user to the system.

[1483] The "means for analyzing the user's emotional state in real time" refers to a device or software that uses a sensor, camera, microphone, etc. to analyze the user's emotional state in real time.

[1484] The "means for providing navigation assistance information" is a device or software that provides appropriate navigation assistance information (e.g., route directions, warning messages) to the driver based on the analyzed information and the user's emotional state.

[1485] To put the present invention into practice, a document management system and a navigation support system are integrated, as will be described in detail below.

[1486] System configuration

[1487] The system mainly consists of a server, a user terminal, smart glasses, and a head-mounted display (HMD).

[1488] User operations

[1489] Users input destination information using smart glasses or an HMD and upload it to the vehicle system, where it is received and analyzed by a server.

[1490] Server Processing

[1491] The server receives map data and destination information uploaded by users using a means for receiving data. Next, the server analyzes this data using a means for assigning classification tags, and assigns tags according to its characteristics. The analyzed content is then converted into a reusable format and managed using a means for storing it in a database. For example, tags such as "long-distance drive" and "intra-city travel" are assigned.

[1492] Real-time analytics

[1493] To analyze the user's emotional state, sensors, cameras, and microphones built into the smart glasses or HMD are used. This allows the user's emotional state to be transmitted to a server in real time for analysis. Depending on the emotional state, a means for providing navigation assistance information is activated, and appropriate feedback and warnings are provided to the user.

[1494] Navigation Aids

[1495] The server generates navigation assistance information based on map data and destination information. Based on the analysis results and the user's emotional state, specific driving instructions such as "Turn right at the next intersection" or emotional feedback such as "Drive calmly and without rushing" are displayed on the smart glasses or HMD.

[1496] Specific examples

[1497] For example, if a user inputs "Drive to Tokyo Station," the server receives this data and tags it as "Long-distance drive." At the same time, if the server detects an emotion of "impatience" from the user's tone of voice or facial expression, it displays emotional feedback such as "Please drive calmly."

[1498] Example prompts

[1499] Here is an example prompt:

[1500] Analyze the following map data and tag it appropriately:

[1501] "Destination: Tokyo Station, looking for a long drive"

[1502] Also, analyze the emotional state from the following audio data and provide appropriate feedback:

[1503] "You seem anxious."

[1504] Hardware and Software Examples

[1505] The hardware used includes smart glasses (e.g., Google Glass), head-mounted displays (e.g., Microsoft HoloLens), and in-vehicle servers (with high-performance processors).The software used includes Python scripts for data analysis and emotion recognition, and TensorFlow or PyTorch for generative AI and emotion analysis.

[1506] As described above, the present invention integrates a document management system and a navigation assistance system, allowing users to receive real-time emotion recognition and navigation assistance while driving.

[1507] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1508] Step 1:

[1509] The user uses smart glasses or a head-mounted display (HMD) to input destination information and upload it to the vehicle system. The input data is in the form of, for example, "I would like to drive to Tokyo Station." This information is then sent to the server via the terminal.

[1510] Step 2:

[1511] The server analyzes the received map data and destination information. The input data is "I would like to drive to Tokyo Station." The server uses a generative AI model to analyze the input data and tag it (e.g., "long-distance drive"). The tagged data is obtained as output.

[1512] Step 3:

[1513] The server converts the parsed data into a reusable format (e.g. text data, JSON data) and stores it in the database. The tagged data as input is processed and converted into a format that can be stored in the database. The converted data as output is stored in the database.

[1514] Step 4:

[1515] The smart glasses and HMD use built-in sensors and microphones to analyze the user's emotional state in real time. Voice and facial expressions are sent as input data to the server, which then analyzes them using an emotion recognition algorithm. The analysis results (e.g., "anxiety") are obtained as output.

[1516] Step 5:

[1517] The server generates appropriate feedback or warnings based on the analysis results. It processes the emotion analysis results (e.g., "impatience") as input and generates a feedback message (e.g., "Please drive calmly"). The generated feedback is obtained as output.

[1518] Step 6:

[1519] The server provides navigation assistance information. Based on input data such as map data, destination information, and emotion analysis results, driving assistance information (e.g., "Please turn right at the next intersection") is generated. The output is displayed on the user's smart glasses or HMD.

[1520] Step 7:

[1521] The server stores the generated navigation assistance information and emotional feedback in a database. The server converts the generated information as input data into a reusable format and stores it in the database. The stored data is obtained as output.

[1522] Through the above steps, the present invention provides navigation assistance and emotional feedback based on the destination information entered by the user and the emotional state analyzed in real time, and this information is stored in a database for future analysis and improvement.

[1523] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1524] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1525] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1526] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1527] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1528] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1529] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1530] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1531] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1532] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1533] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1534] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1535] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1536] 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.

[1537] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1538] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1539] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1540] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1541] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1542] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1543] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1544] The following is further disclosed regarding the above embodiment.

[1545] (Claim 1)

[1546] a means for receiving the materials;

[1547] A means to automatically analyze the content of materials and assign classification tags;

[1548] A means for converting the parsed content into a reusable format and storing it in a database;

[1549] a means for retrieving relevant information from a database in response to a question and generating an answer;

[1550] A system that includes a means to compare different versions of the same material and identify the most recent version.

[1551] (Claim 2)

[1552] 2. The system according to claim 1, further comprising means for automatically classifying and organizing materials received from a plurality of sources and updating the database accordingly.

[1553] (Claim 3)

[1554] 2. The system according to claim 1, further comprising a communication means for providing the generated answer to the user's terminal.

[1555] "Example 1"

[1556] (Claim 1)

[1557] a means for users to upload materials using a computing device;

[1558] means for receiving the material and identifying its format;

[1559] A means to analyze the contents of files and automatically assign classification tags;

[1560] A means for converting the parsed content into a reusable format and storing it in a database;

[1561] a means for a user to input a question in natural language;

[1562] A means for generating an answer by using a generative AI model to retrieve relevant information from a database in response to a question;

[1563] means for transmitting the generated answer to the user's computing device;

[1564] A system that includes a means to compare different versions of the same material and identify the most recent version.

[1565] (Claim 2)

[1566] 2. The system according to claim 1, further comprising means for automatically classifying and organizing materials received from a plurality of sources and updating the database accordingly.

[1567] (Claim 3)

[1568] 10. The system of claim 1, further comprising a communication means for providing the generated answer to a user's computing device.

[1569] "Application Example 1"

[1570] (Claim 1)

[1571] a means for receiving the materials;

[1572] A means to automatically analyze the content of materials and assign classification tags;

[1573] A means for converting the parsed content into a reusable format and storing it in a database;

[1574] A means for retrieving relevant information from a database in response to a question and generating an answer using a generative AI model;

[1575] A means of comparing different versions of the same material and identifying the most recent version;

[1576] An application installed on automated equipment in a factory that manages work knowledge and troubleshooting procedures and provides quick answers based on prompts.

[1577] A system including:

[1578] (Claim 2)

[1579] 2. The system of claim 1, further comprising a means for automatically classifying and organizing materials received from multiple sources and updating the database accordingly, thereby minimizing manual maintenance and support required by users.

[1580] (Claim 3)

[1581] The system according to claim 1, further comprising a communication means for providing the generated answers to a user's terminal, allowing factory workers to quickly access the answers.

[1582] "Example 2: Combining Emotion Engines"

[1583] (Claim 1)

[1584] a means for receiving the materials;

[1585] A means to automatically analyze the content of materials and assign classification tags;

[1586] A means for converting the parsed content into a reusable format and storing it in a database;

[1587] A means for retrieving relevant information from a database in response to a question and generating an answer using a generative AI model;

[1588] A means of comparing different versions of the same material and identifying the most recent version;

[1589] means for analyzing user sentiment and adjusting responses based on the analysis results;

[1590] The system includes means for converting audio to text and extracting key points from video when analyzing the content of the material.

[1591] (Claim 2)

[1592] 2. The system according to claim 1, further comprising means for automatically classifying and organizing materials received from a plurality of sources and updating the database accordingly.

[1593] (Claim 3)

[1594] 2. The system according to claim 1, further comprising a communication means for providing the generated answer to the user's terminal.

[1595] "Application example 2 when combining emotion engines"

[1596] (Claim 1)

[1597] a means for receiving the materials;

[1598] A means to automatically analyze the content of materials and assign classification tags;

[1599] A means for converting the parsed content into a reusable format and storing it in a database;

[1600] a means for retrieving relevant information from a database in response to a question and generating an answer;

[1601] A means of comparing different versions of the same material and identifying the most recent version;

[1602] means for uploading map data and destination information;

[1603] means for analyzing the user's emotional state in real time;

[1604] The system includes means for providing navigation assistance information based on the analysis results.

[1605] (Claim 2)

[1606] 2. The system according to claim 1, further comprising means for automatically classifying and organizing materials received from a plurality of sources and updating the database accordingly.

[1607] (Claim 3)

[1608] 2. The system according to claim 1, further comprising a communication means for providing the generated answer to the user's terminal.

[1609] (Claim 4)

[1610] 10. The system of claim 1, further comprising means for providing feedback and / or alerts based on emotional state. [Explanation of symbols]

[1611] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for receiving the materials; A means to automatically analyze the content of materials and assign classification tags; A means for converting the parsed content into a reusable format and storing it in a database; a means for retrieving relevant information from a database in response to a question and generating an answer; A system that includes a means to compare different versions of the same material and identify the most recent version.

2. 2. The system according to claim 1, further comprising means for automatically classifying and organizing materials received from a plurality of information sources and updating the database accordingly.

3. 2. The system according to claim 1, further comprising a communication means for providing the generated answer to a user terminal.

Images