system

Abstract

We provide the system. [Solution] A means of receiving data files from users and sending them to the server, A means for the server to receive a data file and convert it into text data, A server analyzes the content of the document using natural language processing technology and has a means to detect inconsistencies. A server provides a means for generating suggestions for improving the document based on the analysis results, A means for the terminal to display to the user a list of inconsistencies and suggested improvements received from the server, A system that includes this.

Description

Technical Field

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

Background Art

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

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In conventional document creation, content checking and evaluation of persuasive power are often performed manually, which not only takes time and effort, but also has a high possibility of contradictions and inconsistencies due to human errors. In addition, specialized knowledge is required to obtain effective improvement plans, which is considered difficult for general users. There is a need for a system that solves such problems and checks and improves documents efficiently and accurately.

[0005] ​The present invention is a system comprising: means for receiving a document file from a user and sending it to a server; means for the server to receive the document file and convert it into text data; means for the server to analyze the content of the document using natural language processing technology and detect inconsistencies; means for the server to generate proposed improvements to the document based on the analysis results; and means for a terminal to display a list of inconsistencies and proposed improvements received from the server to the user. This enables efficient and accurate checking of document content and improvement of its appeal. Furthermore, by using optical character recognition technology, the system can handle various document file formats and automatically extract text data. In addition, by extracting information section by section of the document and analyzing the content of each section, it becomes possible to detect detailed inconsistencies and provide specific improvement proposals.

[0006] "Document files" refer to documents and data files that users input into the system, including formats such as PDF, Word, and Text.

[0007] A "user" refers to an individual or group that uses the system to check and improve documents.

[0008] A "server" is a computer device that receives data files and performs analysis on them.

[0009] A "terminal" is a device such as a computer or smartphone that is operated by a user and communicates with a server.

[0010] "Text data" refers to data that has been converted into character information and is the subject of analysis and processing.

[0011] A "contradiction" refers to a section in a document file where there is a logical inconsistency or error.

[0012] An "improvement suggestion" is a specific proposal or advice for improving the content of a document.

[0013] "Natural language processing technology" refers to the technology that enables computers to understand, analyze, and generate human language.

[0014] "Optical character recognition technology" is a technology for extracting character information from images or scanned documents.

[0015] A "section" refers to a part of a document file that contains information on a specific topic or topic. [Brief explanation of the drawing]

[0016] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12]It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Embodiment 2 when the emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when the emotion engine is combined.

Mode for Carrying Out the Invention

[0017] Hereinafter, an example of an embodiment of the system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0018] First, the language used in the following description will be explained.

[0019] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be one arithmetic unit or a combination of a plurality of arithmetic units. Also, the processor may be one type of arithmetic unit or a combination of a plurality of types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0020] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor. <0000​​In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0022] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0024] [First Embodiment]

[0025] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0026] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0027] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0029] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0032] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0036] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0037] The present invention is a system comprising means for receiving a data file from a user and sending it to a server; means for the server to receive the data file and convert it into text data; means for the server to analyze the content of the data using natural language processing technology and detect inconsistencies; means for the server to generate proposed improvements to the data based on the analysis results; and means for a terminal to display a list of inconsistencies and proposed improvements received from the server to the user.

[0038] Program execution (no code)

[0039] Importing data

[0040] The terminal receives document files from the user. Users can upload document files in formats such as PDF, Word, and Text. These document files are then sent to the server by the terminal.

[0041] Receiving and converting data

[0042] The server receives the data files sent from the terminal. The received files are converted into text data using optical character recognition (OCR) or other conversion technologies.

[0043] Analysis of the materials

[0044] The server analyzes the converted text data using natural language processing (NLP) techniques. This analysis includes morphological analysis, syntactic analysis, and contextual analysis. The server identifies each section of the document (e.g., "Goals," "Strategy," "Budget," "Risks") and analyzes the content of each section.

[0045] Detection and listing of inconsistencies

[0046] The server detects inconsistencies in the document based on the analysis results. For example, if the target section states "increase market share by 10%" but the risk section states "there is a possibility that market share will not increase," this will be detected as an inconsistency. The detected inconsistencies are listed.

[0047] Generating improvement plans

[0048] The server generates specific improvement suggestions to enhance the appeal of the materials. For example, it might generate a suggestion such as, "Conduct regular market research after advertising campaigns to confirm the effectiveness of the plan while mitigating risks."

[0049] Displaying Results

[0050] The terminal receives a list of inconsistencies and suggested improvements from the server. The user can review this information displayed on the terminal and revise the document. Because the inconsistencies and suggestions are presented specifically and in detail, the user can effectively improve the document.

[0051] Specific example

[0052] Example document

[0053] The user uploads the following document files to the system:

[0054] ---

[0055] Title: Marketing Strategy for New Products

[0056] 1. Goal

[0057] Increase market share by 10%

[0058] Increase brand awareness by 50%

[0059] 2. Strategy

[0060] We will conduct an advertising campaign.

[0061] We will focus on adding new features.

[0062] 3. Budget

[0063] Allocate 5 million yen to the advertising campaign.

[0064] Allocate 3 million yen to feature development.

[0065] 4. Risk

[0066] The advertising campaign's effectiveness fell short of expectations.

[0067] There is a possibility that market share will not increase.

[0068] ---

[0069] Inconsistency detection results and proposed improvements

[0070] Examples of inconsistencies detected by the server:

[0071] The "Goals" section states "increase market share by 10%", but the "Risks" section contradicts this by stating "the possibility that market share may not increase."

[0072] Examples of improvement suggestions provided by the server:

[0073] "To mitigate risks, we regularly conduct market research after advertising campaigns and fine-tune our plans accordingly."

[0074] This allows users to receive specific and detailed feedback, enabling them to effectively improve the materials.

[0075] The following describes the processing flow.

[0076] Step 1:

[0077] The user uploads the document file to their device. The document file can be in PDF, Word, or Text format.

[0078] Step 2:

[0079] The terminal converts the data files received from the user into a format suitable for the system. For example, it performs pre-processing such as verifying the file format and compressing the files as needed.

[0080] Step 3:

[0081] The terminal sends the pre-processed data files to the server. A secure protocol is used for transmission to maintain data integrity.

[0082] Step 4:

[0083] The server receives the data file sent from the terminal. The received file is temporarily stored in preparation for subsequent processing.

[0084] Step 5:

[0085] The server converts the received document files into text data. For PDF and image files, optical character recognition (OCR) technology is used to extract the text information.

[0086] Step 6:

[0087] The server analyzes text data using natural language processing (NLP) techniques. It sequentially performs morphological analysis, syntactic analysis, and contextual analysis to understand the meaning of the material.

[0088] Step 7:

[0089] The server identifies the overall structure of the document. For example, it organizes information into sections such as "goals," "strategies," "budget," and "risks."

[0090] Step 8:

[0091] The server analyzes the contents of each identified section in detail to detect inconsistencies and discrepancies. For example, it might identify a contradiction between "the possibility of not being able to increase market share" and "increasing market share by 10%."

[0092] Step 9:

[0093] The server creates a list of inconsistencies based on the inconsistencies it detects. This list includes a description of the specific inconsistency and the section in which it exists.

[0094] Step 10:

[0095] The server generates specific improvement suggestions to enhance the appeal of the materials. For example, it proposes concrete action plans such as "conduct regular market research to mitigate risks."

[0096] Step 11:

[0097] The terminal receives a list of inconsistencies and suggested improvements from the server. The received data is displayed in a user-friendly format.

[0098] Step 12:

[0099] The user reviews the list of inconsistencies and suggested improvements displayed on the terminal. They can then revise the document as needed and re-upload it to the system.

[0100] In this way, the system efficiently analyzes the documents created by the user, identifies inconsistencies, and provides concrete suggestions for improvement.

[0101] (Example 1)

[0102] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0103] In traditional document creation and analysis, manually checking and correcting inconsistencies within documents is an extremely cumbersome and time-consuming process. Therefore, there is a need for a system that can efficiently and automatically detect inconsistencies in documents and suggest improvements.

[0104] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0105] In this invention, the server includes means for receiving a data file from a user and sending it to the server; means for the server to receive the data file and convert it into text data; means for the server to analyze the content of the data using natural language processing technology and detect inconsistencies; means for the server to generate proposed improvements to the data based on the analysis results; and means for the terminal to display a list of inconsistencies and proposed improvements received from the server to the user. This enables the user to efficiently detect inconsistencies in the data and revise the data based on detailed proposed improvements.

[0106] A "document file" refers to document data created in formats such as PDF, Word, or Text, which is uploaded by the user.

[0107] A "server" is a computer system that receives data files, converts them into text data, analyzes the content of the data using natural language processing technology, detects inconsistencies, and generates improvement proposals.

[0108] A "terminal" is a device used by users to upload document files, receive and display a list of inconsistencies and suggested improvements sent from the server.

[0109] "Text data" refers to character data converted from document files and analyzed using natural language processing techniques.

[0110] "Natural language processing technology" is a technology that interprets meaning through morphological analysis, syntactic analysis, and contextual analysis of text data, and extracts various kinds of information.

[0111] A "contradiction" refers to inconsistent descriptions or conflicting information found within a document.

[0112] An "improvement proposal" is a specific and detailed suggestion generated by the server based on the analysis results, aimed at enhancing the appeal of the document.

[0113] Optical Character Recognition (OCR) is a technology that reads characters from document files in formats such as images and PDFs and converts them into text data.

[0114] A "section" refers to individual divisions within a document, such as "goals," "strategies," "budgets," and "risks."

[0115] A "generative AI model" is an artificial intelligence technology that generates natural language text based on input data.

[0116] The "HTTP protocol" is a communication protocol used to send and receive data, such as document files and analysis results, between a server and a terminal.

[0117] This invention relates to a system in which a user sends a data file to a server via a terminal, the server analyzes the data file to detect inconsistencies, and generates improvement proposals. The following hardware and software are used to implement this system.

[0118] First, the user uploads document files such as PDFs, Word documents, and Text documents using their own device (PC, smartphone, etc.). Then, the device sends these document files to the server via the HTTP protocol. To ensure data security, it is desirable to use the HTTPS protocol.

[0119] The server temporarily stores the received data files in storage. The received files are converted into text data using optical character recognition (OCR) technology. Specific software such as Tesseract OCR can be used. The converted text data is temporarily stored using a text editor (e.g., Notepad).

[0120] Next, the server analyzes the text data using natural language processing (NLP) techniques. This analysis includes morphological analysis, syntactic analysis, and contextual analysis. Specifically, it uses NLP libraries such as SpaCy and the Natural Language Toolkit (NLTK). The server identifies each section of the document (e.g., goals, strategies, budget, risks) and analyzes the content of each in detail.

[0121] Based on the analysis results, the server detects inconsistencies within the document. For example, if the target section states "increase market share by 10%", but the risk section states "there is a possibility that market share will not increase," this will be detected as an inconsistency. The detected inconsistencies are listed and stored in the database.

[0122] Subsequently, the server uses a generative AI model (e.g., GPT-4®) to automatically generate specific improvement suggestions to enhance the appeal of the materials. For example, it might generate suggestions such as, "To mitigate risks, conduct regular market research after advertising campaigns and fine-tune the plan."

[0123] Finally, the terminal receives a list of inconsistencies and suggested improvements from the server. This information is displayed on the user's terminal screen, allowing the user to review and correct the document. The terminal has a suitable interface for displaying the information received via the HTTP protocol on its screen.

[0124] Specific example

[0125] The user uploads the following document files to the system:

[0126] ---

[0127] Title: Marketing Strategy for New Products

[0128] 1. Goal

[0129] Increase market share by 10%

[0130] Increase brand awareness by 50%

[0131] 2. Strategy

[0132] We will conduct an advertising campaign.

[0133] We will focus on adding new features.

[0134] 3. Budget

[0135] Allocate 5 million yen to the advertising campaign.

[0136] Allocate 3 million yen to feature development.

[0137] 4. Risk

[0138] The advertising campaign's effectiveness fell short of expectations.

[0139] There is a possibility that market share will not increase.

[0140] ---

[0141] Examples of inconsistencies detected by the server:

[0142] The target section states "increase market share by 10%", but the risk section contradicts this by stating "the possibility that market share may not increase."

[0143] Examples of improvement suggestions provided by the server:

[0144] "To mitigate risks, we regularly conduct market research after advertising campaigns and fine-tune our plans accordingly."

[0145] As a result, users can receive specific and detailed feedback, enabling them to effectively improve the materials.

[0146] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0147] Step 1: Upload your document files

[0148] Users access the system interface using their own devices and upload document files. Input files can be in PDF, Word, or Text format. The device sends the submitted files to the server using the HTTP protocol.

[0149] Step 2: Receiving the document files

[0150] The server receives data files sent from the terminal. The input is the data files sent from the terminal. The received files are temporarily stored in the server's storage.

[0151] Step 3: Conversion using Optical Character Recognition (OCR) technology

[0152] The server converts the received document file into text data using optical character recognition (OCR) technology. Specifically, it uses software such as Tesseract OCR. The input is the document file saved in step 2. The output is text data.

[0153] Step 4: Retaining text data

[0154] The server temporarily saves the converted text data to a text editor (e.g., Notepad) or a database. The input is the text data converted by OCR. The output is the temporarily saved text data.

[0155] Step 5: Analysis using Natural Language Processing (NLP)

[0156] The server analyzes text data through morphological analysis, syntactic analysis, and contextual analysis. Specifically, it uses NLP libraries such as SpaCy and Natural Language Toolkit (NLTK). The input is stored text data. The output is data that analyzes the content of each section (e.g., goals, strategies, budget, risks) in detail.

[0157] Step 6: Detecting inconsistencies

[0158] The server detects inconsistencies within the document based on the analysis results. For example, it detects a contradiction if the target section states "increase market share by 10%" while the risk section states "there is a possibility that market share will not increase." The input is the data from each analyzed section. The output is a list of the detected inconsistencies.

[0159] Step 7: Generating improvement plans

[0160] The server generates suggested improvements to the document based on a list of inconsistencies. It uses a generative AI model (e.g., GPT-4) to automatically generate specific improvement suggestions. The input is a list of inconsistencies. The output is specific improvement suggestions.

[0161] Step 8: Submit inconsistencies and suggestions for improvement

[0162] The server sends a list of generated inconsistencies and suggested improvements to the terminal. The input is the list of inconsistencies and suggested improvements. The output is the data sent to the terminal.

[0163] Step 9: Displaying the results

[0164] The terminal displays a list of inconsistencies and suggested improvements received from the server to the user. The input is the list of inconsistencies and suggested improvements sent from the server. The output is the information displayed on the terminal's screen. The user can review the displayed content and effectively revise the document.

[0165] (Application Example 1)

[0166] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0167] In autonomous vehicles, there is a need for a system that can automatically detect inconsistencies in pre-created driving plans and safety plans and quickly and accurately propose improvements. Traditional manual verification and revision processes are time-consuming and labor-intensive, and also carry a high risk of human error, thus increasing the need for automated systems. Furthermore, there is a lack of efficient means to integrate a large amount of relevant information and identify optimal driving routes and risk factors.

[0168] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0169] This invention includes a server that receives data files from a user and transmits them to the server; a server that receives data files and converts them into text data; a server that analyzes the content of the data using natural language processing technology and detects inconsistencies; a server that generates proposed improvements to the data based on the analysis results; and a system for an autonomous vehicle that receives driving plan data and proposes safe driving routes and identifies risk factors based on the analysis results. This enables rapid and accurate analysis and revision of driving plan and safety plan data, as well as the identification of risk factors and the proposal of optimal driving routes.

[0170] A "document file" refers to a document file uploaded by a user and saved in formats such as PDF, Word, or Text.

[0171] A "server" is a computer system that receives data files, converts them into text data, and analyzes the content of the data using natural language processing technology.

[0172] "Text data" refers to data in string format converted from a document file.

[0173] "Natural language processing technology" refers to techniques such as morphological analysis, syntactic analysis, and contextual analysis used to understand and analyze the content of documents.

[0174] A "contradiction" refers to a point in a document where different sections or items do not match or have logical inconsistencies.

[0175] A "proposal for improvement" refers to specific ideas for changes or modifications proposed to enhance the appeal and reliability of the document in response to detected inconsistencies.

[0176] A "terminal" is a device used by users to upload document files and receive and display a list of inconsistencies and suggested improvements sent from the server.

[0177] An "autonomous vehicle" is an automobile that can autonomously perform driving and safety planning and receive analysis results and suggestions through an external interface.

[0178] A "driving plan document" is a document that describes the route and safety measures planned for an autonomous vehicle.

[0179] A "safe driving route" is the optimal route proposed based on the analysis results, designed to ensure that an autonomous vehicle safely reaches its destination.

[0180] "Risk factors" are items or conditions in a driving plan or safety plan that could potentially cause accidents or malfunctions.

[0181] Optical Character Recognition (OCR) is a technology that converts characters stored as images into text data.

[0182] The system that realizes this invention is comprised of the following hardware and software.

[0183] Hardware and software configuration

[0184] Hardware:

[0185] Onboard computers for autonomous vehicles (e.g., NVIDIA DRIVE, Intel Movidius, etc.)

[0186] Vehicle internal interface (display, touchscreen, etc.)

[0187] Servers (those with high-performance computing resources)

[0188] software:

[0189] Optical character recognition technology (e.g., Tesseract OCR)

[0190] Natural language processing technologies (e.g., Google® Cloud Natural Language API, spaCy)

[0191] Database management systems (e.g., MySQL®)

[0192] Data processing and data calculation

[0193] 1. Importing data files: The autonomous vehicle's system receives driving plan data. Users upload data files through the onboard interface. The latest traffic information from in-vehicle sensors such as radar, LiDAR, and cameras is also integrated.

[0194] 2. Receiving and Converting Documents: The server converts received document files into text data using OCR technology. Tesseract OCR is used for OCR.

[0195] 3. Document Analysis: The server analyzes the converted text data using NLP technology. Morphological analysis, syntactic analysis, and contextual analysis are performed to identify each section of the driving plan and safety plan. Google Cloud Natural Language API and spaCy are used for this process.

[0196] 4. Detection and Listing of Inconsistencies: The server detects inconsistencies in the document based on the analysis results. For example, it might detect a discrepancy such as "driving on a highway, but planning to enter a restricted area." These inconsistencies are then listed.

[0197] 5. Generation of Improvement Plans: The server generates specific improvement plans to resolve the detected inconsistencies. For example, improvement plans may include "suggest an alternative route to avoid restricted areas."

[0198] 6. Display of Results: The autonomous vehicle's interface displays a list of detected inconsistencies and suggested improvements. The user can review this and modify the driving plan and safety plan.

[0199] Specific example

[0200] Prompt: Analyze the following planning document and propose inconsistencies and improvements.

[0201] Planning documents:

[0202] ---

[0203] Departure point: Shinjuku

[0204] Destination: Shibuya

[0205] Route: Expressway entrance

[0206] Estimated arrival time: 3:00 PM

[0207] Note: Construction is scheduled to begin soon at the highway entrance.

[0208] ---

[0209] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0210] Step 1:

[0211] The user uploads the driving plan documents via the autonomous vehicle's onboard interface. The user inputs the document file in a specific format (PDF, Word, Text, etc.) into the interface and then uploads it. At this stage, the document file is saved to the onboard computer.

[0212] Step 2:

[0213] The onboard computer sends the uploaded data file to the server. The server receives the data file and converts it into text data using optical character recognition (Tesseract OCR) technology. The input is a data file, and the output is text data. In this process, text information in image format is extracted as text.

[0214] Step 3:

[0215] The server analyzes the converted text data using natural language processing technologies (Google Cloud Natural Language API, spaCy). The input is text data, and the output is the analysis result. Specifically, morphological analysis, syntactic analysis, and contextual analysis are performed to identify each section of the travel plan and safety plan (starting point, destination, waypoints, notes, etc.).

[0216] Step 4:

[0217] The server detects inconsistencies in the document based on the analysis results. The input is the analysis results, and the output is a list of inconsistencies. For example, it identifies discrepancies such as "driving on a highway, but planning to enter a restricted area." The server uses a specific algorithm to identify logical inconsistencies and contradictions and lists them.

[0218] Step 5:

[0219] The server generates improvement suggestions based on the inconsistencies it detects. The input is a list of inconsistencies, and the output is improvement suggestions. Specifically, it generates suggestions for corrections such as "suggest an alternative route to avoid restricted areas." The generated improvement suggestions are specific and actionable so that users can actually take action.

[0220] Step 6:

[0221] The server sends a list of inconsistencies and suggested improvements to the autonomous vehicle's interface. The terminal receives this information and displays it to the user. The input is the list of inconsistencies and suggested improvements, and the output is the information displayed on the interface. The user reviews this information and modifies the driving plan as needed. Specifically, information is conveyed to the user visually using a display or touchscreen.

[0222] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0223] The present invention is a system that combines means for receiving a data file from a user and sending it to a server, means for the server to receive the data file and convert it into text data, means for the server to analyze the content of the data using natural language processing technology and detect inconsistencies, means for the server to generate suggestions for improving the data based on the analysis results, means for a terminal to display a list of inconsistencies and suggestions for improvement received from the server to the user, and an emotion engine that recognizes the user's emotions.

[0224] Program execution (no code)

[0225] Importing data

[0226] The user uploads a document file to the device. The document file can be in PDF, Word, or Text format. During the upload process, the device analyzes the user's facial expressions and voice using an emotion engine to obtain the user's emotional data.

[0227] Receiving and converting data

[0228] The terminal sends the pre-processed data file to the server. A secure protocol is used for transmission. The server receives the data file sent from the terminal and converts the file into text data using optical character recognition (OCR) technology.

[0229] Analysis of the materials

[0230] The server analyzes text data using natural language processing (NLP) techniques. It sequentially performs morphological analysis, syntactic analysis, and contextual analysis to understand the meaning of the document. The analysis results are categorized into sections of the document (e.g., "Goals," "Strategy," "Budget," "Risks").

[0231] Detection and listing of inconsistencies

[0232] The server detects inconsistencies in the document based on the analysis results. For example, it identifies a contradiction between "the possibility that market share cannot be expected to increase" and "a 10% increase in market share." The server then lists these inconsistencies.

[0233] Generating improvement plans

[0234] The server generates specific improvement suggestions to enhance the appeal of the materials based on user sentiment data and analysis results. For example, if a user is dissatisfied, it provides proactive feedback to alleviate that dissatisfaction.

[0235] Displaying Results

[0236] The device receives a list of inconsistencies and suggested improvements from the server. The device displays these in a user-friendly format. When the user responds to the suggested improvements, the device analyzes the reaction using an emotion engine and adjusts the feedback accordingly.

[0237] Specific example

[0238] Example document

[0239] The user uploads the following document files to the system:

[0240] ---

[0241] Title: Marketing Strategy for New Products

[0242] 1. Goal

[0243] Increase market share by 10%

[0244] Increase brand awareness by 50%

[0245] 2. Strategy

[0246] We will conduct an advertising campaign.

[0247] We will focus on adding new features.

[0248] 3. Budget

[0249] Allocate 5 million yen to the advertising campaign.

[0250] Allocate 3 million yen to feature development.

[0251] 4. Risk

[0252] The advertising campaign's effectiveness fell short of expectations.

[0253] There is a possibility that market share will not increase.

[0254] ---

[0255] Inconsistency detection results and proposed improvements

[0256] Examples of inconsistencies detected by the server:

[0257] The "Goals" section states "increase market share by 10%", but the "Risks" section contradicts this by stating "the possibility that market share may not increase."

[0258] Examples of improvement suggestions provided by the server:

[0259] We propose concrete action plans, such as "conducting regular market research to mitigate risks."

[0260] Utilizing the Emotion Engine

[0261] If the device analyzes the user's facial expressions and voice and determines that the user is dissatisfied with the display of inconsistencies in the document, the server will provide proactive feedback to the user. For example, it might display an encouraging message such as, "Correcting this section will significantly improve the persuasiveness of your marketing strategy."

[0262] As a result, users can receive specific and detailed feedback, enabling them to effectively improve their materials. Furthermore, because appropriate feedback that takes user emotions into consideration is provided, the efficiency and quality of material creation improve.

[0263] The following describes the processing flow.

[0264] Step 1:

[0265] The user uploads document files to their device. During the upload process, the device's camera and microphone are used to record the user's facial expressions and voice, which are then sent to the emotion engine.

[0266] Step 2:

[0267] The terminal converts the data files received from the user into a format suitable for the system and verifies file integrity. If necessary, it performs preprocessing such as compressing or splitting the files.

[0268] Step 3:

[0269] The terminal sends pre-processed data files and user sentiment data to the server. An encrypted protocol is used for transmission to ensure security.

[0270] Step 4:

[0271] The server receives data files and sentiment data sent from the terminal. The data files are temporarily stored and used for subsequent processing.

[0272] Step 5:

[0273] The server converts the received document files into text data. For PDF and image files, optical character recognition (OCR) technology is used to extract the text information.

[0274] Step 6:

[0275] The server analyzes text data using natural language processing (NLP) techniques. It sequentially performs morphological analysis, syntactic analysis, and contextual analysis to understand the intent and content of the material.

[0276] Step 7:

[0277] The server categorizes the analyzed text data into sections. For example, it might divide it into sections such as "Goals," "Strategy," "Budget," and "Risks."

[0278] Step 8:

[0279] The server analyzes the contents of each section in detail to detect inconsistencies and contradictions. For example, it identifies contradictory statements such as "the possibility of not being able to increase market share" and "increasing market share by 10%."

[0280] Step 9:

[0281] The server analyzes emotional data and evaluates the user's emotional state. For example, it identifies emotions such as anxiety, excitement, and satisfaction.

[0282] Step 10:

[0283] The server generates improvement plans based on the contradictions in the materials. At this time, considering the user's emotional state, adjustments are made, for example, adding encouraging messages when the user feels anxious.

[0284] Step 11:

[0285] The server sends the list of contradictions and the improvement plans to the terminal. This also includes the specific procedures on how to implement the improvement measures.

[0286] Step 12:

[0287] The terminal displays the list of contradictions and the improvement plans received from the server. It monitors the user's reaction in real time and uses the emotion engine to collect feedback again.

[0288] Step 13:

[0289] The user checks the list of contradictions and the improvement plans displayed on the terminal and modifies the materials as needed. After modification, the materials can be uploaded to the system again for re-checking.

[0290] In this way, the system efficiently analyzes the materials created by the user, identifies contradictions, and provides specific improvement plans considering the user's emotions.

[0291] (Example 2)

[0292] Next, Example 2 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart device 14 is referred to as the "terminal".

[0293] Conventional document creation support systems converted document content into text data and analyzed the content using natural language processing technology. However, they lacked the ability to generate feedback and improvement suggestions that considered user emotions, making it difficult to provide concrete and effective suggestions for improving the quality of documents. Furthermore, they were not adequately able to detect inconsistencies in user-submitted documents and immediately suggest appropriate improvements. As a result, there was insufficient support for enhancing the persuasiveness and effectiveness of documents, leading to the problem of time-consuming document creation.

[0294] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0295] In this invention, the server includes means for receiving a data file and converting it into text data, means for analyzing the content of the data using natural language processing technology and detecting inconsistencies, means for generating proposed improvements to the data based on the analysis results, and means for adjusting the proposed improvements based on the user's sentiment data. This makes it possible to quickly detect inconsistencies in the data and provide specific and effective improvement proposals while taking the user's sentiment into consideration.

[0296] A "document file" is a document that a user uploads to the system, and it can be in PDF, Word, or Text format.

[0297] A "server" is a computer system that receives data files, converts them into text data, analyzes their content, detects inconsistencies, and generates improvement suggestions.

[0298] A "terminal" is a device used by the user, which is used for uploading document files, sending and receiving data with the server, displaying analysis results and improvement suggestions, and acquiring sentiment data.

[0299] "Text data" refers to character information converted from document files, which is then analyzed by the server.

[0300] "Natural language processing technology" refers to the technology that analyzes text data, understands the content of materials and detects contradictions, and includes technologies such as morphological analysis, syntactic analysis, and context analysis.

[0301] "Contradiction" refers to the content in the materials that lacks consistency. For example, it refers to the case where in the target section, it is stated that "increase the market share by 10%", and in the risk section, it is stated that "there is a possibility that the increase in market share cannot be expected".

[0302] "Improvement plan" refers to a proposal generated by the server based on the analysis results and the user's sentiment data to make the content of the materials more effective.

[0303] "Sentiment data" refers to the information obtained by the terminal analyzing the user's expressions and voices, and indicates the user's sentiment state.

[0304] "Sentiment recognition means" refers to the technology by which the terminal analyzes the user's expressions and voices to obtain sentiment data.

[0305] "Optical character recognition technology" refers to the technology that extracts character information from image or PDF-formatted material files and converts it into text data.

[0306] "Extracting information for each section" refers to the process in which the server organizes and analyzes the content for each part (e.g., goals, strategies, budgets, risks) of the materials.

[0307] "Generative AI model" refers to the artificial intelligence technology by which the server generates specific improvement plans based on the analysis results and sentiment data.

[0308] This invention is a system that receives data files from a user and transmits them to a server. This system converts the data files into text data, analyzes the content of the data using natural language processing technology to detect inconsistencies, and generates improvement suggestions for the data based on the analysis results. Furthermore, the terminal acquires the user's emotional data, and the server adjusts the improvement suggestions based on that emotional data, thereby providing specific and effective feedback tailored to the user.

[0309] First, the user logs into the system using their device and uploads a document file. The document file can be provided in formats such as PDF, Word, or Text. Once the user uploads the file, the device saves it to temporary storage. Simultaneously, the device analyzes the user's facial expressions and voice using an emotion engine (e.g., Microsoft® Azure® Emotion API) to obtain the user's emotion data.

[0310] Next, the terminal sends the data files stored in temporary storage and the acquired emotional data to the server using a secure protocol (e.g., HTTPS). To ensure data security, the data is encrypted before transmission (e.g., AES encryption).

[0311] The server receives the document file from the terminal and verifies its format. If the document file is a PDF or image, the server uses the OCR function of the Google Cloud Vision API to convert the document into text data. Word or text files are treated as text data as is. The converted text data is stored in the database.

[0312] The server uses stored text data to analyze the content of the documents using natural language processing (NLP) techniques. Specifically, it performs morphological analysis on the text, divides it into sentences, and performs syntactic analysis to classify it into sections such as "goals," "strategies," "budgets," and "risks." This analysis utilizes the TENSORFLOW® natural language processing library (e.g., the BERT model).

[0313] Based on the analyzed data, the server detects inconsistencies. For example, it identifies contradictory statements within the document, such as "there is a possibility that market share will not increase" and "market share will increase by 10%." These inconsistencies are listed and managed using the Python pandas library.

[0314] Next, the server generates specific improvement suggestions based on the list of inconsistencies and user sentiment data. This process uses generative AI models such as OpenAI's GPT-3®. For example, it might generate suggestions such as "conduct regular market research" as a risk mitigation measure.

[0315] The generated analysis results and improvement suggestions are sent from the server to the terminal. Secure protocols are used to encrypt the data during this process to ensure security. The terminal displays the received analysis results and improvement suggestions to the user in an easy-to-understand format. HTML5, CSS, and JavaScript (registered trademark) technologies are used for display. If the user reacts to the improvement suggestions, their reaction is analyzed again by the sentiment engine, and the server adjusts the feedback content as needed. Furthermore, if the user dislikes the improvement suggestions, the system also has a function to suggest more specific countermeasures.

[0316] As a concrete example, a user uploads the following document file to the system:

[0317] ---

[0318] Title: Marketing Strategy for New Products

[0319] 1. Goal

[0320] Increase market share by 10%

[0321] Increase brand awareness by 50%

[0322] 2. Strategy

[0323] We will conduct an advertising campaign.

[0324] We will focus on adding new features.

[0325] 3. Budget

[0326] Allocate 5 million yen to the advertising campaign.

[0327] Allocate 3 million yen to feature development.

[0328] 4. Risk

[0329] The advertising campaign's effectiveness fell short of expectations.

[0330] There is a possibility that market share will not increase.

[0331] ---

[0332] An example of a contradiction detected by the server is when the "Goals" section states "Increase market share by 10%", while the "Risks" section states "There is a possibility that market share may not increase." The server also proposes specific action plans as improvement suggestions, such as "Conduct regular market research to mitigate risks."

[0333] As described above, this system can provide users with specific and effective feedback, improving the efficiency and quality of document creation.

[0334] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0335] Step 1:

[0336] The user logs into the system and uploads a document file. The file format can be PDF, Word, or Text. When the user selects a file saved on their device and clicks the upload button, the terminal stores the file in a temporary storage area. Simultaneously, the user's facial expressions and voice are analyzed by an emotion engine (general emotion recognition software) to acquire emotion data. At this stage, the input consists of the document file selected by the user and the user's facial expressions and voice, while the output consists of the document file stored in temporary storage and the acquired emotion data.

[0337] Step 2:

[0338] The terminal sends document files and sentiment data stored in temporary storage to the server. The data is encrypted (e.g., AES encryption) and transmitted using a secure protocol (e.g., HTTPS). The input in this process is the files and sentiment data in temporary storage, and the output is the files and data sent to the server.

[0339] Step 3:

[0340] The server receives the document file from the terminal. If the document file is in PDF or image format, the server uses the Optical Character Recognition (OCR) technology of the Google Cloud Vision API to convert the file into text data. Word or Text format files skip this step and are treated as text data as is. The input for this step is the received document file, and the output is text data.

[0341] Step 4:

[0342] The server analyzes the converted text data using natural language processing (NLP) techniques. First, it performs morphological analysis to divide the text into sentences, then syntactically analyzes it to classify it into sections such as goals, strategies, budgets, and risks. TensorFlow's BERT model is used for this analysis. The input is text data, and the output is the analyzed result.

[0343] Step 5:

[0344] The server detects inconsistencies in the document based on the analysis results. For example, if the "Goals" section states "increase market share by 10%" while the "Risks" section states "there is a possibility that market share will not increase," the server will identify this inconsistency. A natural language understanding model is used for this detection. The input is the analyzed text data, and the output is a list of inconsistencies.

[0345] Step 6:

[0346] The server generates specific improvement suggestions based on a list of inconsistencies and user sentiment data. This generation uses generative AI models such as OpenAI GPT-3. For example, it creates specific action plans to mitigate risks (e.g., "Conduct regular market research to mitigate risks"). The input is a list of inconsistencies and sentiment data, and the output is the generated improvement suggestions.

[0347] Step 7:

[0348] The server sends the analysis results and suggested improvements to the terminal. The data is encrypted and transmitted using a secure protocol. The input to this process is the suggested improvements and analysis results, and the output is the data sent to the terminal.

[0349] Step 8:

[0350] The terminal displays analysis results and improvement suggestions received from the server to the user. HTML5, CSS, and JavaScript are used for display. If the user reacts to the improvement suggestions, the terminal analyzes the reaction using an emotion engine and adjusts the feedback content as needed. The input is data from the server, and the output is the analysis results and improvement suggestions displayed to the user.

[0351] This completes the overall processing flow of the system that efficiently creates documents and provides high-quality feedback to users.

[0352] (Application Example 2)

[0353] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0354] Traditional document management systems struggle to effectively detect inconsistencies in complex documents and generate improvement suggestions. Furthermore, the lack of feedback that takes user sentiment into account reduces the efficiency of document improvement. These issues can negatively impact operational efficiency and quality control, particularly in quality control reports and production plans used in factories.

[0355] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving and transmitting data files to the server, means for receiving data files and converting them into text data, means for analyzing the content of the data using natural language processing technology and detecting inconsistencies, means for generating improvement proposals for the data based on the analysis results, means equipped with an emotion recognition engine for analyzing the user's facial expressions and voice and acquiring emotion data, and means for generating appropriate feedback based on the analysis results and emotion data. This makes it possible to efficiently and accurately detect inconsistencies in the data and generate improvement proposals, and further improve the efficiency of data improvement by providing feedback that takes the user's emotions into consideration.

[0356] "Document files" refer to documents and papers saved in formats such as PDF, Word, or text.

[0357] A "server" refers to a computer system that receives data files via network communication, converts them into text data, and performs analysis on them.

[0358] "Text data" refers to string information obtained by converting document files using optical character recognition (OCR) or other methods.

[0359] "Natural language processing technology" refers to the technology that enables computers to understand, analyze, and extract information from human language.

[0360] A "contradiction" refers to a logically inconsistent section or discrepancy in information within a document.

[0361] An "emotion recognition engine" refers to a software or hardware system that analyzes a user's facial expressions and voice to identify their emotional state.

[0362] "Means by which the server generates suggested improvements to the document based on the analysis results" refers to the system component that has the function of generating specific corrections and methods to suggest to the user based on the analysis results and sentiment data.

[0363] "Appropriate feedback" refers to feedback that takes into account the user's emotional state and includes encouragement when suggesting improvements to the material.

[0364] A "terminal" refers to a device used by users to upload data files and receive and display analysis results and improvement suggestions.

[0365] This invention is a system that analyzes data files, detects inconsistencies, and generates improvement suggestions. When a user uploads a data file, the system receives it and performs analysis. Furthermore, it is characterized by its ability to recognize the user's emotional state and provide appropriate feedback.

[0366] System Configuration

[0367] 1. User terminal

[0368] The user terminal is a device used to send data files to the server, and can be a PC or tablet. The terminal is also equipped with an emotion recognition engine, which captures the user's facial expressions and voice using the camera and microphone. Specifically, OpenCV is used for camera analysis and TensorFlow for emotion recognition.

[0369] 2. Server

[0370] A server is a computer system that receives data files, converts them into text data, and performs analysis. It uses software such as the following:

[0371] Optical character recognition technology: Tesseract OCR

[0372] Natural Language Processing Technologies: spaCy, Hugging Face Transformers

[0373] Generative AI models: GPT-3, etc.

[0374] Program processing

[0375] The document files uploaded by the user are first sent from the terminal to the server. SSL / TLS, a secure communication protocol, is used for transmission. The server converts the received document files into text data using Tesseract OCR. Next, this text data is analyzed using spaCy and Hugging Face Transformers to detect inconsistencies.

[0376] Examples of data analysis

[0377] Consider the case where a user uploads the following documents:

[0378] Title: Marketing Strategy for New Products

[0379] 1. Goal

[0380] Increase market share by 10%

[0381] Increase brand awareness by 50%

[0382] 2. Strategy

[0383] We will conduct an advertising campaign.

[0384] We will focus on adding new features.

[0385] 3. Budget

[0386] Allocate 5 million yen to the advertising campaign.

[0387] Allocate 3 million yen to feature development.

[0388] 4. Risk

[0389] The advertising campaign's effectiveness fell short of expectations.

[0390] There is a possibility that market share will not increase.

[0391] The server analyzes this text data and detects, for example, inconsistencies between the "Goals" section and the "Risks" section. Next, it generates improvement suggestions based on these inconsistencies using a generative AI model (GPT-3).

[0392] Emotion recognition and feedback

[0393] When the device displays inconsistencies or suggestions for improvement in the document to the user, the emotion recognition engine analyzes the user's facial expressions and voice. If the user expresses dissatisfaction, the server generates positive feedback and displays it to the user through the device.

[0394] For example, the prompt statement is as follows:

[0395] "Please review the quality control report and point out any inconsistencies or problems. Furthermore, please propose solutions for improvement."

[0396] The above describes the embodiments for carrying out this invention.

[0397] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0398] Step 1:

[0399] The user uploads a document file. The input is a document file (PDF, Word, or text format), and the user's facial expressions and voice are also captured using the device's camera and microphone. This allows for the acquisition of both the document file and the user's emotional data.

[0400] Step 2:

[0401] The terminal preprocesses the data files and sentiment data and sends them to the server. The transmission uses a secure communication protocol (SSL / TLS). The input consists of the data files and sentiment data, and the output is the data files and sentiment data received by the server.

[0402] Step 3:

[0403] The server converts the received document file into text data using Optical Character Recognition (OCR) technology. Specifically, Tesseract OCR software is used. The input is a document file, and the output is text data.

[0404] Step 4:

[0405] The server analyzes text data using natural language processing (NLP) techniques. Morphological analysis, syntactic analysis, and contextual analysis are performed sequentially. The software used is spaCy and Hugging Face Transformers. The input is text data, and the output is the analysis result.

[0406] Step 5:

[0407] The server detects inconsistencies in the data based on the analysis results. For example, it checks for inconsistencies between objectives and risks. The input is the analysis results, and the output is a list of detected inconsistencies.

[0408] Step 6:

[0409] The server generates suggested improvements to the document based on inconsistencies. It uses a generative AI model (such as GPT-3) to propose specific corrections and methods. The input is a list of detected inconsistencies and user sentiment data, and the output is the suggested improvements.

[0410] Step 7:

[0411] The server sends the analysis results and improvement suggestions to the terminal. The input is the analysis results and improvement suggestions, and the output is the analysis results and improvement suggestions sent by the server.

[0412] Step 8:

[0413] The terminal displays a list of inconsistencies and suggested improvements received from the server to the user. When the user responds to the suggested improvements, the terminal re-analyzes that reaction using an emotion recognition engine. The input is the analysis results and suggested improvements, while the output is the display data for the user to see and the re-acquired emotion data.

[0414] Step 9:

[0415] The device adjusts the feedback content according to the user's emotions and resends it to the server. The input is the user's reaction data, and the output is the adjusted feedback and its transmission data.

[0416] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0417] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0418] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0419] [Second Embodiment]

[0420] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0421] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0422] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0423] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0424] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0425] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0426] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0427] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0428] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0429] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0430] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0431] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0432] The present invention is a system comprising means for receiving a data file from a user and sending it to a server; means for the server to receive the data file and convert it into text data; means for the server to analyze the content of the data using natural language processing technology and detect inconsistencies; means for the server to generate proposed improvements to the data based on the analysis results; and means for a terminal to display a list of inconsistencies and proposed improvements received from the server to the user.

[0433] Program execution (no code)

[0434] Importing data

[0435] The terminal receives document files from the user. Users can upload document files in formats such as PDF, Word, and Text. These document files are then sent to the server by the terminal.

[0436] Receiving and converting data

[0437] The server receives the data files sent from the terminal. The received files are converted into text data using optical character recognition (OCR) or other conversion technologies.

[0438] Analysis of the materials

[0439] The server analyzes the converted text data using natural language processing (NLP) techniques. This analysis includes morphological analysis, syntactic analysis, and contextual analysis. The server identifies each section of the document (e.g., "Goals," "Strategy," "Budget," "Risks") and analyzes the content of each section.

[0440] Detection and listing of inconsistencies

[0441] The server detects inconsistencies in the document based on the analysis results. For example, if the target section states "increase market share by 10%" but the risk section states "there is a possibility that market share will not increase," this will be detected as an inconsistency. The detected inconsistencies are listed.

[0442] Generating improvement plans

[0443] The server generates specific improvement suggestions to enhance the appeal of the materials. For example, it might generate a suggestion such as, "Conduct regular market research after advertising campaigns to confirm the effectiveness of the plan while mitigating risks."

[0444] Displaying Results

[0445] The terminal receives a list of inconsistencies and suggested improvements from the server. The user can review this information displayed on the terminal and revise the document. Because the inconsistencies and suggestions are presented specifically and in detail, the user can effectively improve the document.

[0446] Specific example

[0447] Example document

[0448] The user uploads the following document files to the system:

[0449] ---

[0450] Title: Marketing Strategy for New Products

[0451] 1. Goal

[0452] Increase market share by 10%

[0453] Increase brand awareness by 50%

[0454] 2. Strategy

[0455] We will conduct an advertising campaign.

[0456] We will focus on adding new features.

[0457] 3. Budget

[0458] Allocate 5 million yen to the advertising campaign.

[0459] Allocate 3 million yen to feature development.

[0460] 4. Risk

[0461] The advertising campaign's effectiveness fell short of expectations.

[0462] There is a possibility that market share will not increase.

[0463] ---

[0464] Inconsistency detection results and proposed improvements

[0465] Examples of inconsistencies detected by the server:

[0466] The "Goals" section states "increase market share by 10%", but the "Risks" section contradicts this by stating "the possibility that market share may not increase."

[0467] Examples of improvement suggestions provided by the server:

[0468] "To mitigate risks, we regularly conduct market research after advertising campaigns and fine-tune our plans accordingly."

[0469] This allows users to receive specific and detailed feedback, enabling them to effectively improve the materials.

[0470] The following describes the processing flow.

[0471] Step 1:

[0472] The user uploads the document file to their device. The document file can be in PDF, Word, or Text format.

[0473] Step 2:

[0474] The terminal converts the data files received from the user into a format suitable for the system. For example, it performs pre-processing such as verifying the file format and compressing the files as needed.

[0475] Step 3:

[0476] The terminal sends the pre-processed data files to the server. A secure protocol is used for transmission to maintain data integrity.

[0477] Step 4:

[0478] The server receives the data file sent from the terminal. The received file is temporarily stored in preparation for subsequent processing.

[0479] Step 5:

[0480] The server converts the received document files into text data. For PDF and image files, optical character recognition (OCR) technology is used to extract the text information.

[0481] Step 6:

[0482] The server analyzes text data using natural language processing (NLP) techniques. It sequentially performs morphological analysis, syntactic analysis, and contextual analysis to understand the meaning of the material.

[0483] Step 7:

[0484] The server identifies the overall structure of the document. For example, it organizes information into sections such as "goals," "strategies," "budget," and "risks."

[0485] Step 8:

[0486] The server analyzes the contents of each identified section in detail to detect inconsistencies and discrepancies. For example, it might identify a contradiction between "the possibility of not being able to increase market share" and "increasing market share by 10%."

[0487] Step 9:

[0488] The server creates a list of inconsistencies based on the inconsistencies it detects. This list includes a description of the specific inconsistency and the section in which it exists.

[0489] Step 10:

[0490] The server generates specific improvement suggestions to enhance the appeal of the materials. For example, it proposes concrete action plans such as "conduct regular market research to mitigate risks."

[0491] Step 11:

[0492] The terminal receives a list of inconsistencies and suggested improvements from the server. The received data is displayed in a user-friendly format.

[0493] Step 12:

[0494] The user reviews the list of inconsistencies and suggested improvements displayed on the terminal. They can then revise the document as needed and re-upload it to the system.

[0495] In this way, the system efficiently analyzes the documents created by the user, identifies inconsistencies, and provides concrete suggestions for improvement.

[0496] (Example 1)

[0497] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0498] In traditional document creation and analysis, manually checking and correcting inconsistencies within documents is an extremely cumbersome and time-consuming process. Therefore, there is a need for a system that can efficiently and automatically detect inconsistencies in documents and suggest improvements.

[0499] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0500] In this invention, the server includes means for receiving a data file from a user and sending it to the server; means for the server to receive the data file and convert it into text data; means for the server to analyze the content of the data using natural language processing technology and detect inconsistencies; means for the server to generate proposed improvements to the data based on the analysis results; and means for the terminal to display a list of inconsistencies and proposed improvements received from the server to the user. This enables the user to efficiently detect inconsistencies in the data and revise the data based on detailed proposed improvements.

[0501] A "document file" refers to document data created in formats such as PDF, Word, or Text, which is uploaded by the user.

[0502] A "server" is a computer system that receives data files, converts them into text data, analyzes the content of the data using natural language processing technology, detects inconsistencies, and generates improvement proposals.

[0503] A "terminal" is a device used by users to upload document files, receive and display a list of inconsistencies and suggested improvements sent from the server.

[0504] "Text data" refers to character data converted from document files and analyzed using natural language processing techniques.

[0505] "Natural language processing technology" is a technology that interprets meaning through morphological analysis, syntactic analysis, and contextual analysis of text data, and extracts various kinds of information.

[0506] A "contradiction" refers to inconsistent descriptions or conflicting information found within a document.

[0507] An "improvement proposal" is a specific and detailed suggestion generated by the server based on the analysis results, aimed at enhancing the appeal of the document.

[0508] Optical Character Recognition (OCR) is a technology that reads characters from document files in formats such as images and PDFs and converts them into text data.

[0509] A "section" refers to individual divisions within a document, such as "goals," "strategies," "budgets," and "risks."

[0510] A "generative AI model" is an artificial intelligence technology that generates natural language text based on input data.

[0511] The "HTTP protocol" is a communication protocol used to send and receive data, such as document files and analysis results, between a server and a terminal.

[0512] This invention relates to a system in which a user sends a data file to a server via a terminal, the server analyzes the data file to detect inconsistencies, and generates improvement proposals. The following hardware and software are used to implement this system.

[0513] First, the user uploads document files such as PDFs, Word documents, and Text documents using their own device (PC, smartphone, etc.). Then, the device sends these document files to the server via the HTTP protocol. To ensure data security, it is desirable to use the HTTPS protocol.

[0514] The server temporarily stores the received data files in storage. The received files are converted into text data using optical character recognition (OCR) technology. Specific software such as Tesseract OCR can be used. The converted text data is temporarily stored using a text editor (e.g., Notepad).

[0515] Next, the server analyzes the text data using natural language processing (NLP) techniques. This analysis includes morphological analysis, syntactic analysis, and contextual analysis. Specifically, it uses NLP libraries such as SpaCy and the Natural Language Toolkit (NLTK). The server identifies each section of the document (e.g., goals, strategies, budget, risks) and analyzes the content of each in detail.

[0516] Based on the analysis results, the server detects inconsistencies within the document. For example, if the target section states "increase market share by 10%", but the risk section states "there is a possibility that market share will not increase," this will be detected as an inconsistency. The detected inconsistencies are listed and stored in the database.

[0517] The server then uses a generative AI model (e.g., GPT-4) to automatically generate specific improvement suggestions to enhance the appeal of the materials. For example, it might generate suggestions such as, "To mitigate risks, conduct regular market research after advertising campaigns and fine-tune the plan."

[0518] Finally, the terminal receives a list of inconsistencies and suggested improvements from the server. This information is displayed on the user's terminal screen, allowing the user to review and correct the document. The terminal has a suitable interface for displaying the information received via the HTTP protocol on its screen.

[0519] Specific example

[0520] The user uploads the following document files to the system:

[0521] ---

[0522] Title: Marketing Strategy for New Products

[0523] 1. Goal

[0524] Increase market share by 10%

[0525] Increase brand awareness by 50%

[0526] 2. Strategy

[0527] We will conduct an advertising campaign.

[0528] We will focus on adding new features.

[0529] 3. Budget

[0530] Allocate 5 million yen to the advertising campaign.

[0531] Allocate 3 million yen to feature development.

[0532] 4. Risk

[0533] The advertising campaign's effectiveness fell short of expectations.

[0534] There is a possibility that market share will not increase.

[0535] ---

[0536] Examples of inconsistencies detected by the server:

[0537] The target section states "increase market share by 10%", but the risk section contradicts this by stating "the possibility that market share may not increase."

[0538] Examples of improvement suggestions provided by the server:

[0539] "To mitigate risks, we regularly conduct market research after advertising campaigns and fine-tune our plans accordingly."

[0540] As a result, users can receive specific and detailed feedback, enabling them to effectively improve the materials.

[0541] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0542] Step 1: Upload your document files

[0543] Users access the system interface using their own devices and upload document files. Input files can be in PDF, Word, or Text format. The device sends the submitted files to the server using the HTTP protocol.

[0544] Step 2: Receiving the document files

[0545] The server receives data files sent from the terminal. The input is the data files sent from the terminal. The received files are temporarily stored in the server's storage.

[0546] Step 3: Conversion using Optical Character Recognition (OCR) technology

[0547] The server converts the received document file into text data using optical character recognition (OCR) technology. Specifically, it uses software such as Tesseract OCR. The input is the document file saved in step 2. The output is text data.

[0548] Step 4: Retaining text data

[0549] The server temporarily saves the converted text data to a text editor (e.g., Notepad) or a database. The input is the text data converted by OCR. The output is the temporarily saved text data.

[0550] Step 5: Analysis using Natural Language Processing (NLP)

[0551] The server analyzes text data through morphological analysis, syntactic analysis, and contextual analysis. Specifically, it uses NLP libraries such as SpaCy and Natural Language Toolkit (NLTK). The input is stored text data. The output is data that analyzes the content of each section (e.g., goals, strategies, budget, risks) in detail.

[0552] Step 6: Detecting inconsistencies

[0553] The server detects inconsistencies within the document based on the analysis results. For example, it detects a contradiction if the target section states "increase market share by 10%" while the risk section states "there is a possibility that market share will not increase." The input is the data from each analyzed section. The output is a list of the detected inconsistencies.

[0554] Step 7: Generating improvement plans

[0555] The server generates suggested improvements to the document based on a list of inconsistencies. It uses a generative AI model (e.g., GPT-4) to automatically generate specific improvement suggestions. The input is a list of inconsistencies. The output is specific improvement suggestions.

[0556] Step 8: Submit inconsistencies and suggestions for improvement

[0557] The server sends a list of generated inconsistencies and suggested improvements to the terminal. The input is the list of inconsistencies and suggested improvements. The output is the data sent to the terminal.

[0558] Step 9: Displaying the results

[0559] The terminal displays a list of inconsistencies and suggested improvements received from the server to the user. The input is the list of inconsistencies and suggested improvements sent from the server. The output is the information displayed on the terminal's screen. The user can review the displayed content and effectively revise the document.

[0560] (Application Example 1)

[0561] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0562] In autonomous vehicles, there is a need for a system that can automatically detect inconsistencies in pre-created driving plans and safety plans and quickly and accurately propose improvements. Traditional manual verification and revision processes are time-consuming and labor-intensive, and also carry a high risk of human error, thus increasing the need for automated systems. Furthermore, there is a lack of efficient means to integrate a large amount of relevant information and identify optimal driving routes and risk factors.

[0563] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0564] This invention includes a server that receives data files from a user and transmits them to the server; a server that receives data files and converts them into text data; a server that analyzes the content of the data using natural language processing technology and detects inconsistencies; a server that generates proposed improvements to the data based on the analysis results; and a system for an autonomous vehicle that receives driving plan data and proposes safe driving routes and identifies risk factors based on the analysis results. This enables rapid and accurate analysis and revision of driving plan and safety plan data, as well as the identification of risk factors and the proposal of optimal driving routes.

[0565] A "document file" refers to a document file uploaded by a user and saved in formats such as PDF, Word, or Text.

[0566] A "server" is a computer system that receives data files, converts them into text data, and analyzes the content of the data using natural language processing technology.

[0567] "Text data" refers to data in string format converted from a document file.

[0568] "Natural language processing technology" refers to techniques such as morphological analysis, syntactic analysis, and contextual analysis used to understand and analyze the content of documents.

[0569] A "contradiction" refers to a point in a document where different sections or items do not match or have logical inconsistencies.

[0570] A "proposal for improvement" refers to specific ideas for changes or modifications proposed to enhance the appeal and reliability of the document in response to detected inconsistencies.

[0571] A "terminal" is a device used by users to upload document files and receive and display a list of inconsistencies and suggested improvements sent from the server.

[0572] An "autonomous vehicle" is an automobile that can autonomously perform driving and safety planning and receive analysis results and suggestions through an external interface.

[0573] A "driving plan document" is a document that describes the route and safety measures planned for an autonomous vehicle.

[0574] A "safe driving route" is the optimal route proposed based on the analysis results, designed to ensure that an autonomous vehicle safely reaches its destination.

[0575] "Risk factors" are items or conditions in a driving plan or safety plan that could potentially cause accidents or malfunctions.

[0576] Optical Character Recognition (OCR) is a technology that converts characters stored as images into text data.

[0577] The system that realizes this invention is comprised of the following hardware and software.

[0578] Hardware and software configuration

[0579] Hardware:

[0580] Onboard computers for autonomous vehicles (e.g., NVIDIA DRIVE, Intel Movidius, etc.)

[0581] Vehicle internal interface (display, touchscreen, etc.)

[0582] Servers (those with high-performance computing resources)

[0583] software:

[0584] Optical character recognition technology (e.g., Tesseract OCR)

[0585] Natural language processing technologies (e.g., Google Cloud Natural Language API, spaCy)

[0586] Database management system (e.g., MySQL)

[0587] Data processing and data calculation

[0588] 1. Importing data files: The autonomous vehicle's system receives driving plan data. Users upload data files through the onboard interface. The latest traffic information from in-vehicle sensors such as radar, LiDAR, and cameras is also integrated.

[0589] 2. Receiving and Converting Documents: The server converts received document files into text data using OCR technology. Tesseract OCR is used for OCR.

[0590] 3. Document Analysis: The server analyzes the converted text data using NLP technology. Morphological analysis, syntactic analysis, and contextual analysis are performed to identify each section of the driving plan and safety plan. Google Cloud Natural Language API and spaCy are used for this process.

[0591] 4. Detection and Listing of Inconsistencies: The server detects inconsistencies in the document based on the analysis results. For example, it might detect a discrepancy such as "driving on a highway, but planning to enter a restricted area." These inconsistencies are then listed.

[0592] 5. Generation of Improvement Plans: The server generates specific improvement plans to resolve the detected inconsistencies. For example, improvement plans may include "suggest an alternative route to avoid restricted areas."

[0593] 6. Display of Results: The autonomous vehicle's interface displays a list of detected inconsistencies and suggested improvements. The user can review this and modify the driving plan and safety plan.

[0594] Specific example

[0595] Prompt: Analyze the following planning document and propose inconsistencies and improvements.

[0596] Planning documents:

[0597] ---

[0598] Departure point: Shinjuku

[0599] Destination: Shibuya

[0600] Route: Expressway entrance

[0601] Estimated arrival time: 3:00 PM

[0602] Note: Construction is scheduled to begin soon at the highway entrance.

[0603] ---

[0604] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0605] Step 1:

[0606] The user uploads the driving plan documents via the autonomous vehicle's onboard interface. The user inputs the document file in a specific format (PDF, Word, Text, etc.) into the interface and then uploads it. At this stage, the document file is saved to the onboard computer.

[0607] Step 2:

[0608] The onboard computer sends the uploaded data file to the server. The server receives the data file and converts it into text data using optical character recognition (Tesseract OCR) technology. The input is a data file, and the output is text data. In this process, text information in image format is extracted as text.

[0609] Step 3:

[0610] The server analyzes the converted text data using natural language processing technologies (Google Cloud Natural Language API, spaCy). The input is text data, and the output is the analysis result. Specifically, morphological analysis, syntactic analysis, and contextual analysis are performed to identify each section of the travel plan and safety plan (starting point, destination, waypoints, notes, etc.).

[0611] Step 4:

[0612] The server detects inconsistencies in the document based on the analysis results. The input is the analysis results, and the output is a list of inconsistencies. For example, it identifies discrepancies such as "driving on a highway, but planning to enter a restricted area." The server uses a specific algorithm to identify logical inconsistencies and contradictions and lists them.

[0613] Step 5:

[0614] The server generates improvement suggestions based on the inconsistencies it detects. The input is a list of inconsistencies, and the output is improvement suggestions. Specifically, it generates suggestions for corrections such as "suggest an alternative route to avoid restricted areas." The generated improvement suggestions are specific and actionable so that users can actually take action.

[0615] Step 6:

[0616] The server sends a list of inconsistencies and suggested improvements to the autonomous vehicle's interface. The terminal receives this information and displays it to the user. The input is the list of inconsistencies and suggested improvements, and the output is the information displayed on the interface. The user reviews this information and modifies the driving plan as needed. Specifically, information is conveyed to the user visually using a display or touchscreen.

[0617] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0618] The present invention is a system that combines means for receiving a data file from a user and sending it to a server, means for the server to receive the data file and convert it into text data, means for the server to analyze the content of the data using natural language processing technology and detect inconsistencies, means for the server to generate suggestions for improving the data based on the analysis results, means for a terminal to display a list of inconsistencies and suggestions for improvement received from the server to the user, and an emotion engine that recognizes the user's emotions.

[0619] Program execution (no code)

[0620] Importing data

[0621] The user uploads a document file to the device. The document file can be in PDF, Word, or Text format. During the upload process, the device analyzes the user's facial expressions and voice using an emotion engine to obtain the user's emotional data.

[0622] Receiving and converting data

[0623] The terminal sends the pre-processed data file to the server. A secure protocol is used for transmission. The server receives the data file sent from the terminal and converts the file into text data using optical character recognition (OCR) technology.

[0624] Analysis of the materials

[0625] The server analyzes text data using natural language processing (NLP) techniques. It sequentially performs morphological analysis, syntactic analysis, and contextual analysis to understand the meaning of the document. The analysis results are categorized into sections of the document (e.g., "Goals," "Strategy," "Budget," "Risks").

[0626] Detection and listing of inconsistencies

[0627] The server detects inconsistencies in the document based on the analysis results. For example, it identifies a contradiction between "the possibility that market share cannot be expected to increase" and "a 10% increase in market share." The server then lists these inconsistencies.

[0628] Generating improvement plans

[0629] The server generates specific improvement suggestions to enhance the appeal of the materials based on user sentiment data and analysis results. For example, if a user is dissatisfied, it provides proactive feedback to alleviate that dissatisfaction.

[0630] Displaying Results

[0631] The device receives a list of inconsistencies and suggested improvements from the server. The device displays these in a user-friendly format. When the user responds to the suggested improvements, the device analyzes the reaction using an emotion engine and adjusts the feedback accordingly.

[0632] Specific example

[0633] Example document

[0634] The user uploads the following document files to the system:

[0635] ---

[0636] Title: Marketing Strategy for New Products

[0637] 1. Goal

[0638] Increase market share by 10%

[0639] Increase brand awareness by 50%

[0640] 2. Strategy

[0641] We will conduct an advertising campaign.

[0642] We will focus on adding new features.

[0643] 3. Budget

[0644] Allocate 5 million yen to the advertising campaign.

[0645] Allocate 3 million yen to feature development.

[0646] 4. Risk

[0647] The advertising campaign's effectiveness fell short of expectations.

[0648] There is a possibility that market share will not increase.

[0649] ---

[0650] Inconsistency detection results and proposed improvements

[0651] Examples of inconsistencies detected by the server:

[0652] The "Goals" section states "increase market share by 10%", but the "Risks" section contradicts this by stating "the possibility that market share may not increase."

[0653] Examples of improvement suggestions provided by the server:

[0654] We propose concrete action plans, such as "conducting regular market research to mitigate risks."

[0655] Utilizing the Emotion Engine

[0656] If the device analyzes the user's facial expressions and voice and determines that the user is dissatisfied with the display of inconsistencies in the document, the server will provide proactive feedback to the user. For example, it might display an encouraging message such as, "Correcting this section will significantly improve the persuasiveness of your marketing strategy."

[0657] As a result, users can receive specific and detailed feedback, enabling them to effectively improve their materials. Furthermore, because appropriate feedback that takes user emotions into consideration is provided, the efficiency and quality of material creation improve.

[0658] The following describes the processing flow.

[0659] Step 1:

[0660] The user uploads document files to their device. During the upload process, the device's camera and microphone are used to record the user's facial expressions and voice, which are then sent to the emotion engine.

[0661] Step 2:

[0662] The terminal converts the data files received from the user into a format suitable for the system and verifies file integrity. If necessary, it performs preprocessing such as compressing or splitting the files.

[0663] Step 3:

[0664] The terminal sends pre-processed data files and user sentiment data to the server. An encrypted protocol is used for transmission to ensure security.

[0665] Step 4:

[0666] The server receives data files and sentiment data sent from the terminal. The data files are temporarily stored and used for subsequent processing.

[0667] Step 5:

[0668] The server converts the received document files into text data. For PDF and image files, optical character recognition (OCR) technology is used to extract the text information.

[0669] Step 6:

[0670] The server analyzes text data using natural language processing (NLP) techniques. It sequentially performs morphological analysis, syntactic analysis, and contextual analysis to understand the intent and content of the material.

[0671] Step 7:

[0672] The server categorizes the analyzed text data into sections. For example, it might divide it into sections such as "Goals," "Strategy," "Budget," and "Risks."

[0673] Step 8:

[0674] The server analyzes the contents of each section in detail to detect inconsistencies and contradictions. For example, it identifies contradictory statements such as "the possibility of not being able to increase market share" and "increasing market share by 10%."

[0675] Step 9:

[0676] The server analyzes emotional data and evaluates the user's emotional state. For example, it identifies emotions such as anxiety, excitement, and satisfaction.

[0677] Step 10:

[0678] The server generates improvement suggestions based on inconsistencies in the document. In doing so, it takes into account the user's emotional state, making adjustments such as adding encouraging messages if the user is feeling anxious.

[0679] Step 11:

[0680] The server sends a list of inconsistencies and suggested improvements to the terminal, including specific steps on how to implement them.

[0681] Step 12:

[0682] The terminal displays a list of inconsistencies and suggested improvements received from the server. It monitors user reactions in real time and collects further feedback using an emotion engine.

[0683] Step 13:

[0684] Users can review the list of inconsistencies and suggested improvements displayed on their device and revise the document as needed. After revisions, they can re-upload the document to the system for further checks.

[0685] This allows the system to efficiently analyze user-generated materials, identify inconsistencies, and provide concrete improvement suggestions that take user sentiment into consideration.

[0686] (Example 2)

[0687] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0688] Conventional document creation support systems converted document content into text data and analyzed the content using natural language processing technology. However, they lacked the ability to generate feedback and improvement suggestions that considered user emotions, making it difficult to provide concrete and effective suggestions for improving the quality of documents. Furthermore, they were not adequately able to detect inconsistencies in user-submitted documents and immediately suggest appropriate improvements. As a result, there was insufficient support for enhancing the persuasiveness and effectiveness of documents, leading to the problem of time-consuming document creation.

[0689] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0690] In this invention, the server includes means for receiving a data file and converting it into text data, means for analyzing the content of the data using natural language processing technology and detecting inconsistencies, means for generating proposed improvements to the data based on the analysis results, and means for adjusting the proposed improvements based on the user's sentiment data. This makes it possible to quickly detect inconsistencies in the data and provide specific and effective improvement proposals while taking the user's sentiment into consideration.

[0691] A "document file" is a document that a user uploads to the system, and it can be in PDF, Word, or Text format.

[0692] A "server" is a computer system that receives data files, converts them into text data, analyzes their content, detects inconsistencies, and generates improvement suggestions.

[0693] A "terminal" is a device used by the user, which is used for uploading document files, sending and receiving data with the server, displaying analysis results and improvement suggestions, and acquiring sentiment data.

[0694] "Text data" refers to character information converted from document files, which is then analyzed by the server.

[0695] "Natural language processing technology" refers to techniques that analyze text data, understand the content of materials, and detect inconsistencies, and includes morphological analysis, syntactic analysis, and contextual analysis.

[0696] A "contradiction" refers to inconsistencies within a document. For example, it might state in the objectives section that the goal is to "increase market share by 10%" while in the risks section that there is a "possibility that market share may not increase."

[0697] An "improvement suggestion" is a proposal to make the content of the materials generated by the server based on analysis results and user sentiment data more effective.

[0698] "Emotional data" refers to information acquired by a device by analyzing the user's facial expressions and voice, and it indicates the user's emotional state.

[0699] "Emotion recognition means" refers to technology that allows a device to analyze the user's facial expressions and voice to acquire emotional data.

[0700] "Optical character recognition technology" is a technology that extracts character information from images and PDF document files and converts it into text data.

[0701] "Extracting information section by section" refers to the process where the server organizes and analyzes the content of each part of the document (e.g., goals, strategies, budget, risks).

[0702] A "generative AI model" refers to artificial intelligence technology in which a server generates specific improvement suggestions based on analysis results and sentiment data.

[0703] This invention is a system that receives data files from a user and transmits them to a server. This system converts the data files into text data, analyzes the content of the data using natural language processing technology to detect inconsistencies, and generates improvement suggestions for the data based on the analysis results. Furthermore, the terminal acquires the user's emotional data, and the server adjusts the improvement suggestions based on that emotional data, thereby providing specific and effective feedback tailored to the user.

[0704] First, the user logs into the system using their device and uploads a document file. The document file can be provided in formats such as PDF, Word, or Text. Once the user uploads the file, the device saves it to temporary storage. Simultaneously, the device analyzes the user's facial expressions and voice using an emotion engine (e.g., Microsoft Azure Emotion API) to obtain the user's emotional data.

[0705] Next, the terminal sends the data files stored in temporary storage and the acquired emotional data to the server using a secure protocol (e.g., HTTPS). To ensure data security, the data is encrypted before transmission (e.g., AES encryption).

[0706] The server receives the document file from the terminal and verifies its format. If the document file is a PDF or image, the server uses the OCR function of the Google Cloud Vision API to convert the document into text data. Word or text files are treated as text data as is. The converted text data is stored in the database.

[0707] The server uses stored text data to analyze the content of the documents using natural language processing (NLP) techniques. Specifically, it performs morphological analysis on the text, divides it into sentences, and performs syntactic analysis to classify it into sections such as "goals," "strategies," "budgets," and "risks." This analysis utilizes TensorFlow's natural language processing library (for example, the BERT model).

[0708] Based on the analyzed data, the server detects inconsistencies. For example, it identifies contradictory statements within the document, such as "there is a possibility that market share will not increase" and "market share will increase by 10%." These inconsistencies are listed and managed using the Python pandas library.

[0709] Next, the server generates specific improvement suggestions based on the list of inconsistencies and user sentiment data. This process uses generative AI models such as OpenAI's GPT-3. For example, it might generate suggestions such as "conduct regular market research" as a risk mitigation measure.

[0710] The generated analysis results and improvement suggestions are sent from the server to the terminal. Secure protocols are used to encrypt the data during this process to ensure security. The terminal displays the received analysis results and improvement suggestions to the user in an easy-to-understand format. HTML5, CSS, and JavaScript are used for the display. If the user reacts to the improvement suggestions, their reaction is analyzed again by the sentiment engine, and the server adjusts the feedback content as needed. Furthermore, if the user dislikes the improvement suggestions, the system also includes a function to suggest more specific countermeasures.

[0711] As a concrete example, a user uploads the following document file to the system:

[0712] ---

[0713] Title: Marketing Strategy for New Products

[0714] 1. Goal

[0715] Increase market share by 10%

[0716] Increase brand awareness by 50%

[0717] 2. Strategy

[0718] We will conduct an advertising campaign.

[0719] We will focus on adding new features.

[0720] 3. Budget

[0721] Allocate 5 million yen to the advertising campaign.

[0722] Allocate 3 million yen to feature development.

[0723] 4. Risk

[0724] The advertising campaign's effectiveness fell short of expectations.

[0725] There is a possibility that market share will not increase.

[0726] ---

[0727] An example of a contradiction detected by the server is when the "Goals" section states "Increase market share by 10%", while the "Risks" section states "There is a possibility that market share may not increase." The server also proposes specific action plans as improvement suggestions, such as "Conduct regular market research to mitigate risks."

[0728] As described above, this system can provide users with specific and effective feedback, improving the efficiency and quality of document creation.

[0729] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0730] Step 1:

[0731] The user logs into the system and uploads a document file. The file format can be PDF, Word, or Text. When the user selects a file saved on their device and clicks the upload button, the terminal stores the file in a temporary storage area. Simultaneously, the user's facial expressions and voice are analyzed by an emotion engine (general emotion recognition software) to acquire emotion data. At this stage, the input consists of the document file selected by the user and the user's facial expressions and voice, while the output consists of the document file stored in temporary storage and the acquired emotion data.

[0732] Step 2:

[0733] The terminal sends document files and sentiment data stored in temporary storage to the server. The data is encrypted (e.g., AES encryption) and transmitted using a secure protocol (e.g., HTTPS). The input in this process is the files and sentiment data in temporary storage, and the output is the files and data sent to the server.

[0734] Step 3:

[0735] The server receives the document file from the terminal. If the document file is in PDF or image format, the server uses the Optical Character Recognition (OCR) technology of the Google Cloud Vision API to convert the file into text data. Word or Text format files skip this step and are treated as text data as is. The input for this step is the received document file, and the output is text data.

[0736] Step 4:

[0737] The server analyzes the converted text data using natural language processing (NLP) techniques. First, it performs morphological analysis to divide the text into sentences, then syntactically analyzes it to classify it into sections such as goals, strategies, budgets, and risks. TensorFlow's BERT model is used for this analysis. The input is text data, and the output is the analyzed result.

[0738] Step 5:

[0739] The server detects inconsistencies in the document based on the analysis results. For example, if the "Goals" section states "increase market share by 10%" while the "Risks" section states "there is a possibility that market share will not increase," the server will identify this inconsistency. A natural language understanding model is used for this detection. The input is the analyzed text data, and the output is a list of inconsistencies.

[0740] Step 6:

[0741] The server generates specific improvement suggestions based on a list of inconsistencies and user sentiment data. This generation uses generative AI models such as OpenAI GPT-3. For example, it creates specific action plans to mitigate risks (e.g., "Conduct regular market research to mitigate risks"). The input is a list of inconsistencies and sentiment data, and the output is the generated improvement suggestions.

[0742] Step 7:

[0743] The server sends the analysis results and suggested improvements to the terminal. The data is encrypted and transmitted using a secure protocol. The input to this process is the suggested improvements and analysis results, and the output is the data sent to the terminal.

[0744] Step 8:

[0745] The terminal displays analysis results and improvement suggestions received from the server to the user. HTML5, CSS, and JavaScript are used for display. If the user reacts to the improvement suggestions, the terminal analyzes the reaction using an emotion engine and adjusts the feedback content as needed. The input is data from the server, and the output is the analysis results and improvement suggestions displayed to the user.

[0746] This completes the overall processing flow of the system that efficiently creates documents and provides high-quality feedback to users.

[0747] (Application Example 2)

[0748] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0749] Traditional document management systems struggle to effectively detect inconsistencies in complex documents and generate improvement suggestions. Furthermore, the lack of feedback that takes user sentiment into account reduces the efficiency of document improvement. These issues can negatively impact operational efficiency and quality control, particularly in quality control reports and production plans used in factories.

[0750] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving and transmitting data files to the server, means for receiving data files and converting them into text data, means for analyzing the content of the data using natural language processing technology and detecting inconsistencies, means for generating improvement proposals for the data based on the analysis results, means equipped with an emotion recognition engine for analyzing the user's facial expressions and voice and acquiring emotion data, and means for generating appropriate feedback based on the analysis results and emotion data. This makes it possible to efficiently and accurately detect inconsistencies in the data and generate improvement proposals, and further improve the efficiency of data improvement by providing feedback that takes the user's emotions into consideration.

[0751] "Document files" refer to documents and papers saved in formats such as PDF, Word, or text.

[0752] A "server" refers to a computer system that receives data files via network communication, converts them into text data, and performs analysis on them.

[0753] "Text data" refers to string information obtained by converting document files using optical character recognition (OCR) or other methods.

[0754] "Natural language processing technology" refers to the technology that enables computers to understand, analyze, and extract information from human language.

[0755] A "contradiction" refers to a logically inconsistent section or discrepancy in information within a document.

[0756] An "emotion recognition engine" refers to a software or hardware system that analyzes a user's facial expressions and voice to identify their emotional state.

[0757] "Means by which the server generates suggested improvements to the document based on the analysis results" refers to the system component that has the function of generating specific corrections and methods to suggest to the user based on the analysis results and sentiment data.

[0758] "Appropriate feedback" refers to feedback that takes into account the user's emotional state and includes encouragement when suggesting improvements to the material.

[0759] A "terminal" refers to a device used by users to upload data files and receive and display analysis results and improvement suggestions.

[0760] This invention is a system that analyzes data files, detects inconsistencies, and generates improvement suggestions. When a user uploads a data file, the system receives it and performs analysis. Furthermore, it is characterized by its ability to recognize the user's emotional state and provide appropriate feedback.

[0761] System Configuration

[0762] 1. User terminal

[0763] The user terminal is a device used to send data files to the server, and can be a PC or tablet. The terminal is also equipped with an emotion recognition engine, which captures the user's facial expressions and voice using the camera and microphone. Specifically, OpenCV is used for camera analysis and TensorFlow for emotion recognition.

[0764] 2. Server

[0765] A server is a computer system that receives data files, converts them into text data, and performs analysis. It uses software such as the following:

[0766] Optical character recognition technology: Tesseract OCR

[0767] Natural Language Processing Technologies: spaCy, Hugging Face Transformers

[0768] Generative AI models: GPT-3, etc.

[0769] Program processing

[0770] The document files uploaded by the user are first sent from the terminal to the server. SSL / TLS, a secure communication protocol, is used for transmission. The server converts the received document files into text data using Tesseract OCR. Next, this text data is analyzed using spaCy and Hugging Face Transformers to detect inconsistencies.

[0771] Examples of data analysis

[0772] Consider the case where a user uploads the following documents:

[0773] Title: Marketing Strategy for New Products

[0774] 1. Goal

[0775] Increase market share by 10%

[0776] Increase brand awareness by 50%

[0777] 2. Strategy

[0778] We will conduct an advertising campaign.

[0779] We will focus on adding new features.

[0780] 3. Budget

[0781] Allocate 5 million yen to the advertising campaign.

[0782] Allocate 3 million yen to feature development.

[0783] 4. Risk

[0784] The advertising campaign's effectiveness fell short of expectations.

[0785] There is a possibility that market share will not increase.

[0786] The server analyzes this text data and detects, for example, inconsistencies between the "Goals" section and the "Risks" section. Next, it generates improvement suggestions based on these inconsistencies using a generative AI model (GPT-3).

[0787] Emotion recognition and feedback

[0788] When the device displays inconsistencies or suggestions for improvement in the document to the user, the emotion recognition engine analyzes the user's facial expressions and voice. If the user expresses dissatisfaction, the server generates positive feedback and displays it to the user through the device.

[0789] For example, the prompt statement is as follows:

[0790] "Please review the quality control report and point out any inconsistencies or problems. Furthermore, please propose solutions for improvement."

[0791] The above describes the embodiments for carrying out this invention.

[0792] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0793] Step 1:

[0794] The user uploads a document file. The input is a document file (PDF, Word, or text format), and the user's facial expressions and voice are also captured using the device's camera and microphone. This allows for the acquisition of both the document file and the user's emotional data.

[0795] Step 2:

[0796] The terminal preprocesses the data files and sentiment data and sends them to the server. The transmission uses a secure communication protocol (SSL / TLS). The input consists of the data files and sentiment data, and the output is the data files and sentiment data received by the server.

[0797] Step 3:

[0798] The server converts the received document file into text data using Optical Character Recognition (OCR) technology. Specifically, Tesseract OCR software is used. The input is a document file, and the output is text data.

[0799] Step 4:

[0800] The server analyzes text data using natural language processing (NLP) techniques. Morphological analysis, syntactic analysis, and contextual analysis are performed sequentially. The software used is spaCy and Hugging Face Transformers. The input is text data, and the output is the analysis result.

[0801] Step 5:

[0802] The server detects inconsistencies in the data based on the analysis results. For example, it checks for inconsistencies between objectives and risks. The input is the analysis results, and the output is a list of detected inconsistencies.

[0803] Step 6:

[0804] The server generates suggested improvements to the document based on inconsistencies. It uses a generative AI model (such as GPT-3) to propose specific corrections and methods. The input is a list of detected inconsistencies and user sentiment data, and the output is the suggested improvements.

[0805] Step 7:

[0806] The server sends the analysis results and improvement suggestions to the terminal. The input is the analysis results and improvement suggestions, and the output is the analysis results and improvement suggestions sent by the server.

[0807] Step 8:

[0808] The terminal displays a list of inconsistencies and suggested improvements received from the server to the user. When the user responds to the suggested improvements, the terminal re-analyzes that reaction using an emotion recognition engine. The input is the analysis results and suggested improvements, while the output is the display data for the user to see and the re-acquired emotion data.

[0809] Step 9:

[0810] The device adjusts the feedback content according to the user's emotions and resends it to the server. The input is the user's reaction data, and the output is the adjusted feedback and its transmission data.

[0811] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0812] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0813] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0814] [Third Embodiment]

[0815] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0816] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0817] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0818] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0819] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0820] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0821] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0822] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0823] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0824] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0825] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0826] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0827] The present invention is a system comprising means for receiving a data file from a user and sending it to a server; means for the server to receive the data file and convert it into text data; means for the server to analyze the content of the data using natural language processing technology and detect inconsistencies; means for the server to generate proposed improvements to the data based on the analysis results; and means for a terminal to display a list of inconsistencies and proposed improvements received from the server to the user.

[0828] Program execution (no code)

[0829] Importing data

[0830] The terminal receives document files from the user. Users can upload document files in formats such as PDF, Word, and Text. These document files are then sent to the server by the terminal.

[0831] Receiving and converting data

[0832] The server receives the data files sent from the terminal. The received files are converted into text data using optical character recognition (OCR) or other conversion technologies.

[0833] Analysis of the materials

[0834] The server analyzes the converted text data using natural language processing (NLP) techniques. This analysis includes morphological analysis, syntactic analysis, and contextual analysis. The server identifies each section of the document (e.g., "Goals," "Strategy," "Budget," "Risks") and analyzes the content of each section.

[0835] Detection and listing of inconsistencies

[0836] The server detects inconsistencies in the document based on the analysis results. For example, if the target section states "increase market share by 10%" but the risk section states "there is a possibility that market share will not increase," this will be detected as an inconsistency. The detected inconsistencies are listed.

[0837] Generating improvement plans

[0838] The server generates specific improvement suggestions to enhance the appeal of the materials. For example, it might generate a suggestion such as, "Conduct regular market research after advertising campaigns to confirm the effectiveness of the plan while mitigating risks."

[0839] Displaying Results

[0840] The terminal receives a list of inconsistencies and suggested improvements from the server. The user can review this information displayed on the terminal and revise the document. Because the inconsistencies and suggestions are presented specifically and in detail, the user can effectively improve the document.

[0841] Specific example

[0842] Example document

[0843] The user uploads the following document files to the system:

[0844] ---

[0845] Title: Marketing Strategy for New Products

[0846] 1. Goal

[0847] Increase market share by 10%

[0848] Increase brand awareness by 50%

[0849] 2. Strategy

[0850] We will conduct an advertising campaign.

[0851] We will focus on adding new features.

[0852] 3. Budget

[0853] Allocate 5 million yen to the advertising campaign.

[0854] Allocate 3 million yen to feature development.

[0855] 4. Risk

[0856] The advertising campaign's effectiveness fell short of expectations.

[0857] There is a possibility that market share will not increase.

[0858] ---

[0859] Inconsistency detection results and proposed improvements

[0860] Examples of inconsistencies detected by the server:

[0861] The "Goals" section states "increase market share by 10%", but the "Risks" section contradicts this by stating "the possibility that market share may not increase."

[0862] Examples of improvement suggestions provided by the server:

[0863] "To mitigate risks, we regularly conduct market research after advertising campaigns and fine-tune our plans accordingly."

[0864] This allows users to receive specific and detailed feedback, enabling them to effectively improve the materials.

[0865] The following describes the processing flow.

[0866] Step 1:

[0867] The user uploads the document file to their device. The document file can be in PDF, Word, or Text format.

[0868] Step 2:

[0869] The terminal converts the data files received from the user into a format suitable for the system. For example, it performs pre-processing such as verifying the file format and compressing the files as needed.

[0870] Step 3:

[0871] The terminal sends the pre-processed data files to the server. A secure protocol is used for transmission to maintain data integrity.

[0872] Step 4:

[0873] The server receives the data file sent from the terminal. The received file is temporarily stored in preparation for subsequent processing.

[0874] Step 5:

[0875] The server converts the received document files into text data. For PDF and image files, optical character recognition (OCR) technology is used to extract the text information.

[0876] Step 6:

[0877] The server analyzes text data using natural language processing (NLP) techniques. It sequentially performs morphological analysis, syntactic analysis, and contextual analysis to understand the meaning of the material.

[0878] Step 7:

[0879] The server identifies the overall structure of the document. For example, it organizes information into sections such as "goals," "strategies," "budget," and "risks."

[0880] Step 8:

[0881] The server analyzes the contents of each identified section in detail to detect inconsistencies and discrepancies. For example, it might identify a contradiction between "the possibility of not being able to increase market share" and "increasing market share by 10%."

[0882] Step 9:

[0883] The server creates a list of inconsistencies based on the inconsistencies it detects. This list includes a description of the specific inconsistency and the section in which it exists.

[0884] Step 10:

[0885] The server generates specific improvement suggestions to enhance the appeal of the materials. For example, it proposes concrete action plans such as "conduct regular market research to mitigate risks."

[0886] Step 11:

[0887] The terminal receives a list of inconsistencies and suggested improvements from the server. The received data is displayed in a user-friendly format.

[0888] Step 12:

[0889] The user reviews the list of inconsistencies and suggested improvements displayed on the terminal. They can then revise the document as needed and re-upload it to the system.

[0890] In this way, the system efficiently analyzes the documents created by the user, identifies inconsistencies, and provides concrete suggestions for improvement.

[0891] (Example 1)

[0892] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0893] In traditional document creation and analysis, manually checking and correcting inconsistencies within documents is an extremely cumbersome and time-consuming process. Therefore, there is a need for a system that can efficiently and automatically detect inconsistencies in documents and suggest improvements.

[0894] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0895] In this invention, the server includes means for receiving a data file from a user and sending it to the server; means for the server to receive the data file and convert it into text data; means for the server to analyze the content of the data using natural language processing technology and detect inconsistencies; means for the server to generate proposed improvements to the data based on the analysis results; and means for the terminal to display a list of inconsistencies and proposed improvements received from the server to the user. This enables the user to efficiently detect inconsistencies in the data and revise the data based on detailed proposed improvements.

[0896] A "document file" refers to document data created in formats such as PDF, Word, or Text, which is uploaded by the user.

[0897] A "server" is a computer system that receives data files, converts them into text data, analyzes the content of the data using natural language processing technology, detects inconsistencies, and generates improvement proposals.

[0898] A "terminal" is a device used by users to upload document files, receive and display a list of inconsistencies and suggested improvements sent from the server.

[0899] "Text data" refers to character data converted from document files and analyzed using natural language processing techniques.

[0900] "Natural language processing technology" is a technology that interprets meaning through morphological analysis, syntactic analysis, and contextual analysis of text data, and extracts various kinds of information.

[0901] A "contradiction" refers to inconsistent descriptions or conflicting information found within a document.

[0902] An "improvement proposal" is a specific and detailed suggestion generated by the server based on the analysis results, aimed at enhancing the appeal of the document.

[0903] Optical Character Recognition (OCR) is a technology that reads characters from document files in formats such as images and PDFs and converts them into text data.

[0904] A "section" refers to individual divisions within a document, such as "goals," "strategies," "budgets," and "risks."

[0905] A "generative AI model" is an artificial intelligence technology that generates natural language text based on input data.

[0906] The "HTTP protocol" is a communication protocol used to send and receive data, such as document files and analysis results, between a server and a terminal.

[0907] This invention relates to a system in which a user sends a data file to a server via a terminal, the server analyzes the data file to detect inconsistencies, and generates improvement proposals. The following hardware and software are used to implement this system.

[0908] First, the user uploads document files such as PDFs, Word documents, and Text documents using their own device (PC, smartphone, etc.). Then, the device sends these document files to the server via the HTTP protocol. To ensure data security, it is desirable to use the HTTPS protocol.

[0909] The server temporarily stores the received data files in storage. The received files are converted into text data using optical character recognition (OCR) technology. Specific software such as Tesseract OCR can be used. The converted text data is temporarily stored using a text editor (e.g., Notepad).

[0910] Next, the server analyzes the text data using natural language processing (NLP) techniques. This analysis includes morphological analysis, syntactic analysis, and contextual analysis. Specifically, it uses NLP libraries such as SpaCy and the Natural Language Toolkit (NLTK). The server identifies each section of the document (e.g., goals, strategies, budget, risks) and analyzes the content of each in detail.

[0911] Based on the analysis results, the server detects inconsistencies within the document. For example, if the target section states "increase market share by 10%", but the risk section states "there is a possibility that market share will not increase," this will be detected as an inconsistency. The detected inconsistencies are listed and stored in the database.

[0912] The server then uses a generative AI model (e.g., GPT-4) to automatically generate specific improvement suggestions to enhance the appeal of the materials. For example, it might generate suggestions such as, "To mitigate risks, conduct regular market research after advertising campaigns and fine-tune the plan."

[0913] Finally, the terminal receives a list of inconsistencies and suggested improvements from the server. This information is displayed on the user's terminal screen, allowing the user to review and correct the document. The terminal has a suitable interface for displaying the information received via the HTTP protocol on its screen.

[0914] Specific example

[0915] The user uploads the following document files to the system:

[0916] ---

[0917] Title: Marketing Strategy for New Products

[0918] 1. Goal

[0919] Increase market share by 10%

[0920] Increase brand awareness by 50%

[0921] 2. Strategy

[0922] We will conduct an advertising campaign.

[0923] We will focus on adding new features.

[0924] 3. Budget

[0925] Allocate 5 million yen to the advertising campaign.

[0926] Allocate 3 million yen to feature development.

[0927] 4. Risk

[0928] The advertising campaign's effectiveness fell short of expectations.

[0929] There is a possibility that market share will not increase.

[0930] ---

[0931] Examples of inconsistencies detected by the server:

[0932] The target section states "increase market share by 10%", but the risk section contradicts this by stating "the possibility that market share may not increase."

[0933] Examples of improvement suggestions provided by the server:

[0934] "To mitigate risks, we regularly conduct market research after advertising campaigns and fine-tune our plans accordingly."

[0935] As a result, users can receive specific and detailed feedback, enabling them to effectively improve the materials.

[0936] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0937] Step 1: Upload your document files

[0938] Users access the system interface using their own devices and upload document files. Input files can be in PDF, Word, or Text format. The device sends the submitted files to the server using the HTTP protocol.

[0939] Step 2: Receiving the document files

[0940] The server receives data files sent from the terminal. The input is the data files sent from the terminal. The received files are temporarily stored in the server's storage.

[0941] Step 3: Conversion using Optical Character Recognition (OCR) technology

[0942] The server converts the received document file into text data using optical character recognition (OCR) technology. Specifically, it uses software such as Tesseract OCR. The input is the document file saved in step 2. The output is text data.

[0943] Step 4: Retaining text data

[0944] The server temporarily saves the converted text data to a text editor (e.g., Notepad) or a database. The input is the text data converted by OCR. The output is the temporarily saved text data.

[0945] Step 5: Analysis using Natural Language Processing (NLP)

[0946] The server analyzes text data through morphological analysis, syntactic analysis, and contextual analysis. Specifically, it uses NLP libraries such as SpaCy and Natural Language Toolkit (NLTK). The input is stored text data. The output is data that analyzes the content of each section (e.g., goals, strategies, budget, risks) in detail.

[0947] Step 6: Detecting inconsistencies

[0948] The server detects inconsistencies within the document based on the analysis results. For example, it detects a contradiction if the target section states "increase market share by 10%" while the risk section states "there is a possibility that market share will not increase." The input is the data from each analyzed section. The output is a list of the detected inconsistencies.

[0949] Step 7: Generating improvement plans

[0950] The server generates suggested improvements to the document based on a list of inconsistencies. It uses a generative AI model (e.g., GPT-4) to automatically generate specific improvement suggestions. The input is a list of inconsistencies. The output is specific improvement suggestions.

[0951] Step 8: Submit inconsistencies and suggestions for improvement

[0952] The server sends a list of generated inconsistencies and suggested improvements to the terminal. The input is the list of inconsistencies and suggested improvements. The output is the data sent to the terminal.

[0953] Step 9: Displaying the results

[0954] The terminal displays a list of inconsistencies and suggested improvements received from the server to the user. The input is the list of inconsistencies and suggested improvements sent from the server. The output is the information displayed on the terminal's screen. The user can review the displayed content and effectively revise the document.

[0955] (Application Example 1)

[0956] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0957] In autonomous vehicles, there is a need for a system that can automatically detect inconsistencies in pre-created driving plans and safety plans and quickly and accurately propose improvements. Traditional manual verification and revision processes are time-consuming and labor-intensive, and also carry a high risk of human error, thus increasing the need for automated systems. Furthermore, there is a lack of efficient means to integrate a large amount of relevant information and identify optimal driving routes and risk factors.

[0958] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0959] This invention includes a server that receives data files from a user and transmits them to the server; a server that receives data files and converts them into text data; a server that analyzes the content of the data using natural language processing technology and detects inconsistencies; a server that generates proposed improvements to the data based on the analysis results; and a system for an autonomous vehicle that receives driving plan data and proposes safe driving routes and identifies risk factors based on the analysis results. This enables rapid and accurate analysis and revision of driving plan and safety plan data, as well as the identification of risk factors and the proposal of optimal driving routes.

[0960] A "document file" refers to a document file uploaded by a user and saved in formats such as PDF, Word, or Text.

[0961] A "server" is a computer system that receives data files, converts them into text data, and analyzes the content of the data using natural language processing technology.

[0962] "Text data" refers to data in string format converted from a document file.

[0963] "Natural language processing technology" refers to techniques such as morphological analysis, syntactic analysis, and contextual analysis used to understand and analyze the content of documents.

[0964] A "contradiction" refers to a point in a document where different sections or items do not match or have logical inconsistencies.

[0965] A "proposal for improvement" refers to specific ideas for changes or modifications proposed to enhance the appeal and reliability of the document in response to detected inconsistencies.

[0966] A "terminal" is a device used by users to upload document files and receive and display a list of inconsistencies and suggested improvements sent from the server.

[0967] An "autonomous vehicle" is an automobile that can autonomously perform driving and safety planning and receive analysis results and suggestions through an external interface.

[0968] A "driving plan document" is a document that describes the route and safety measures planned for an autonomous vehicle.

[0969] A "safe driving route" is the optimal route proposed based on the analysis results, designed to ensure that an autonomous vehicle safely reaches its destination.

[0970] "Risk factors" are items or conditions in a driving plan or safety plan that could potentially cause accidents or malfunctions.

[0971] Optical Character Recognition (OCR) is a technology that converts characters stored as images into text data.

[0972] The system that realizes this invention is comprised of the following hardware and software.

[0973] Hardware and software configuration

[0974] Hardware:

[0975] Onboard computers for autonomous vehicles (e.g., NVIDIA DRIVE, Intel Movidius, etc.)

[0976] Vehicle internal interface (display, touchscreen, etc.)

[0977] Servers (those with high-performance computing resources)

[0978] software:

[0979] Optical character recognition technology (e.g., Tesseract OCR)

[0980] Natural language processing technologies (e.g., Google Cloud Natural Language API, spaCy)

[0981] Database management system (e.g., MySQL)

[0982] Data processing and data calculation

[0983] 1. Importing data files: The autonomous vehicle's system receives driving plan data. Users upload data files through the onboard interface. The latest traffic information from in-vehicle sensors such as radar, LiDAR, and cameras is also integrated.

[0984] 2. Receiving and Converting Documents: The server converts received document files into text data using OCR technology. Tesseract OCR is used for OCR.

[0985] 3. Document Analysis: The server analyzes the converted text data using NLP technology. Morphological analysis, syntactic analysis, and contextual analysis are performed to identify each section of the driving plan and safety plan. Google Cloud Natural Language API and spaCy are used for this process.

[0986] 4. Detection and Listing of Inconsistencies: The server detects inconsistencies in the document based on the analysis results. For example, it might detect a discrepancy such as "driving on a highway, but planning to enter a restricted area." These inconsistencies are then listed.

[0987] 5. Generation of Improvement Plans: The server generates specific improvement plans to resolve the detected inconsistencies. For example, improvement plans may include "suggest an alternative route to avoid restricted areas."

[0988] 6. Display of Results: The autonomous vehicle's interface displays a list of detected inconsistencies and suggested improvements. The user can review this and modify the driving plan and safety plan.

[0989] Specific example

[0990] Prompt: Analyze the following planning document and propose inconsistencies and improvements.

[0991] Planning documents:

[0992] ---

[0993] Departure point: Shinjuku

[0994] Destination: Shibuya

[0995] Route: Expressway entrance

[0996] Estimated arrival time: 3:00 PM

[0997] Note: Construction is scheduled to begin soon at the highway entrance.

[0998] ---

[0999] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1000] Step 1:

[1001] The user uploads the driving plan documents via the autonomous vehicle's onboard interface. The user inputs the document file in a specific format (PDF, Word, Text, etc.) into the interface and then uploads it. At this stage, the document file is saved to the onboard computer.

[1002] Step 2:

[1003] The onboard computer sends the uploaded data file to the server. The server receives the data file and converts it into text data using optical character recognition (Tesseract OCR) technology. The input is a data file, and the output is text data. In this process, text information in image format is extracted as text.

[1004] Step 3:

[1005] The server analyzes the converted text data using natural language processing technologies (Google Cloud Natural Language API, spaCy). The input is text data, and the output is the analysis result. Specifically, morphological analysis, syntactic analysis, and contextual analysis are performed to identify each section of the travel plan and safety plan (starting point, destination, waypoints, notes, etc.).

[1006] Step 4:

[1007] The server detects inconsistencies in the document based on the analysis results. The input is the analysis results, and the output is a list of inconsistencies. For example, it identifies discrepancies such as "driving on a highway, but planning to enter a restricted area." The server uses a specific algorithm to identify logical inconsistencies and contradictions and lists them.

[1008] Step 5:

[1009] The server generates improvement suggestions based on the inconsistencies it detects. The input is a list of inconsistencies, and the output is improvement suggestions. Specifically, it generates suggestions for corrections such as "suggest an alternative route to avoid restricted areas." The generated improvement suggestions are specific and actionable so that users can actually take action.

[1010] Step 6:

[1011] The server sends a list of inconsistencies and suggested improvements to the autonomous vehicle's interface. The terminal receives this information and displays it to the user. The input is the list of inconsistencies and suggested improvements, and the output is the information displayed on the interface. The user reviews this information and modifies the driving plan as needed. Specifically, information is conveyed to the user visually using a display or touchscreen.

[1012] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1013] The present invention is a system that combines means for receiving a data file from a user and sending it to a server, means for the server to receive the data file and convert it into text data, means for the server to analyze the content of the data using natural language processing technology and detect inconsistencies, means for the server to generate suggestions for improving the data based on the analysis results, means for a terminal to display a list of inconsistencies and suggestions for improvement received from the server to the user, and an emotion engine that recognizes the user's emotions.

[1014] Program execution (no code)

[1015] Importing data

[1016] The user uploads a document file to the device. The document file can be in PDF, Word, or Text format. During the upload process, the device analyzes the user's facial expressions and voice using an emotion engine to obtain the user's emotional data.

[1017] Receiving and converting data

[1018] The terminal sends the pre-processed data file to the server. A secure protocol is used for transmission. The server receives the data file sent from the terminal and converts the file into text data using optical character recognition (OCR) technology.

[1019] Analysis of the materials

[1020] The server analyzes text data using natural language processing (NLP) techniques. It sequentially performs morphological analysis, syntactic analysis, and contextual analysis to understand the meaning of the document. The analysis results are categorized into sections of the document (e.g., "Goals," "Strategy," "Budget," "Risks").

[1021] Detection and listing of inconsistencies

[1022] The server detects inconsistencies in the document based on the analysis results. For example, it identifies a contradiction between "the possibility that market share cannot be expected to increase" and "a 10% increase in market share." The server then lists these inconsistencies.

[1023] Generating improvement plans

[1024] The server generates specific improvement suggestions to enhance the appeal of the materials based on user sentiment data and analysis results. For example, if a user is dissatisfied, it provides proactive feedback to alleviate that dissatisfaction.

[1025] Displaying Results

[1026] The device receives a list of inconsistencies and suggested improvements from the server. The device displays these in a user-friendly format. When the user responds to the suggested improvements, the device analyzes the reaction using an emotion engine and adjusts the feedback accordingly.

[1027] Specific example

[1028] Example document

[1029] The user uploads the following document files to the system:

[1030] ---

[1031] Title: Marketing Strategy for New Products

[1032] 1. Goal

[1033] Increase market share by 10%

[1034] Increase brand awareness by 50%

[1035] 2. Strategy

[1036] We will conduct an advertising campaign.

[1037] We will focus on adding new features.

[1038] 3. Budget

[1039] Allocate 5 million yen to the advertising campaign.

[1040] Allocate 3 million yen to feature development.

[1041] 4. Risk

[1042] The advertising campaign's effectiveness fell short of expectations.

[1043] There is a possibility that market share will not increase.

[1044] ---

[1045] Inconsistency detection results and proposed improvements

[1046] Examples of inconsistencies detected by the server:

[1047] The "Goals" section states "increase market share by 10%", but the "Risks" section contradicts this by stating "the possibility that market share may not increase."

[1048] Examples of improvement suggestions provided by the server:

[1049] We propose concrete action plans, such as "conducting regular market research to mitigate risks."

[1050] Utilizing the Emotion Engine

[1051] If the device analyzes the user's facial expressions and voice and determines that the user is dissatisfied with the display of inconsistencies in the document, the server will provide proactive feedback to the user. For example, it might display an encouraging message such as, "Correcting this section will significantly improve the persuasiveness of your marketing strategy."

[1052] As a result, users can receive specific and detailed feedback, enabling them to effectively improve their materials. Furthermore, because appropriate feedback that takes user emotions into consideration is provided, the efficiency and quality of material creation improve.

[1053] The following describes the processing flow.

[1054] Step 1:

[1055] The user uploads document files to their device. During the upload process, the device's camera and microphone are used to record the user's facial expressions and voice, which are then sent to the emotion engine.

[1056] Step 2:

[1057] The terminal converts the data files received from the user into a format suitable for the system and verifies file integrity. If necessary, it performs preprocessing such as compressing or splitting the files.

[1058] Step 3:

[1059] The terminal sends pre-processed data files and user sentiment data to the server. An encrypted protocol is used for transmission to ensure security.

[1060] Step 4:

[1061] The server receives data files and sentiment data sent from the terminal. The data files are temporarily stored and used for subsequent processing.

[1062] Step 5:

[1063] The server converts the received document files into text data. For PDF and image files, optical character recognition (OCR) technology is used to extract the text information.

[1064] Step 6:

[1065] The server analyzes text data using natural language processing (NLP) techniques. It sequentially performs morphological analysis, syntactic analysis, and contextual analysis to understand the intent and content of the material.

[1066] Step 7:

[1067] The server categorizes the analyzed text data into sections. For example, it might divide it into sections such as "Goals," "Strategy," "Budget," and "Risks."

[1068] Step 8:

[1069] The server analyzes the contents of each section in detail to detect inconsistencies and contradictions. For example, it identifies contradictory statements such as "the possibility of not being able to increase market share" and "increasing market share by 10%."

[1070] Step 9:

[1071] The server analyzes emotional data and evaluates the user's emotional state. For example, it identifies emotions such as anxiety, excitement, and satisfaction.

[1072] Step 10:

[1073] The server generates improvement suggestions based on inconsistencies in the document. In doing so, it takes into account the user's emotional state, making adjustments such as adding encouraging messages if the user is feeling anxious.

[1074] Step 11:

[1075] The server sends a list of inconsistencies and suggested improvements to the terminal, including specific steps on how to implement them.

[1076] Step 12:

[1077] The terminal displays a list of inconsistencies and suggested improvements received from the server. It monitors user reactions in real time and collects further feedback using an emotion engine.

[1078] Step 13:

[1079] Users can review the list of inconsistencies and suggested improvements displayed on their device and revise the document as needed. After revisions, they can re-upload the document to the system for further checks.

[1080] This allows the system to efficiently analyze user-generated materials, identify inconsistencies, and provide concrete improvement suggestions that take user sentiment into consideration.

[1081] (Example 2)

[1082] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1083] Conventional document creation support systems converted document content into text data and analyzed the content using natural language processing technology. However, they lacked the ability to generate feedback and improvement suggestions that considered user emotions, making it difficult to provide concrete and effective suggestions for improving the quality of documents. Furthermore, they were not adequately able to detect inconsistencies in user-submitted documents and immediately suggest appropriate improvements. As a result, there was insufficient support for enhancing the persuasiveness and effectiveness of documents, leading to the problem of time-consuming document creation.

[1084] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1085] In this invention, the server includes means for receiving a data file and converting it into text data, means for analyzing the content of the data using natural language processing technology and detecting inconsistencies, means for generating proposed improvements to the data based on the analysis results, and means for adjusting the proposed improvements based on the user's sentiment data. This makes it possible to quickly detect inconsistencies in the data and provide specific and effective improvement proposals while taking the user's sentiment into consideration.

[1086] A "document file" is a document that a user uploads to the system, and it can be in PDF, Word, or Text format.

[1087] A "server" is a computer system that receives data files, converts them into text data, analyzes their content, detects inconsistencies, and generates improvement suggestions.

[1088] A "terminal" is a device used by the user, which is used for uploading document files, sending and receiving data with the server, displaying analysis results and improvement suggestions, and acquiring sentiment data.

[1089] "Text data" refers to character information converted from document files, which is then analyzed by the server.

[1090] "Natural language processing technology" refers to techniques that analyze text data, understand the content of materials, and detect inconsistencies, and includes morphological analysis, syntactic analysis, and contextual analysis.

[1091] A "contradiction" refers to inconsistencies within a document. For example, it might state in the objectives section that the goal is to "increase market share by 10%" while in the risks section that there is a "possibility that market share may not increase."

[1092] An "improvement suggestion" is a proposal to make the content of the materials generated by the server based on analysis results and user sentiment data more effective.

[1093] "Emotional data" refers to information acquired by a device by analyzing the user's facial expressions and voice, and it indicates the user's emotional state.

[1094] "Emotion recognition means" refers to technology that allows a device to analyze the user's facial expressions and voice to acquire emotional data.

[1095] "Optical character recognition technology" is a technology that extracts character information from images and PDF document files and converts it into text data.

[1096] "Extracting information section by section" refers to the process where the server organizes and analyzes the content of each part of the document (e.g., goals, strategies, budget, risks).

[1097] A "generative AI model" refers to artificial intelligence technology in which a server generates specific improvement suggestions based on analysis results and sentiment data.

[1098] This invention is a system that receives data files from a user and transmits them to a server. This system converts the data files into text data, analyzes the content of the data using natural language processing technology to detect inconsistencies, and generates improvement suggestions for the data based on the analysis results. Furthermore, the terminal acquires the user's emotional data, and the server adjusts the improvement suggestions based on that emotional data, thereby providing specific and effective feedback tailored to the user.

[1099] First, the user logs into the system using their device and uploads a document file. The document file can be provided in formats such as PDF, Word, or Text. Once the user uploads the file, the device saves it to temporary storage. Simultaneously, the device analyzes the user's facial expressions and voice using an emotion engine (e.g., Microsoft Azure Emotion API) to obtain the user's emotional data.

[1100] Next, the terminal sends the data files stored in temporary storage and the acquired emotional data to the server using a secure protocol (e.g., HTTPS). To ensure data security, the data is encrypted before transmission (e.g., AES encryption).

[1101] The server receives the document file from the terminal and verifies its format. If the document file is a PDF or image, the server uses the OCR function of the Google Cloud Vision API to convert the document into text data. Word or text files are treated as text data as is. The converted text data is stored in the database.

[1102] The server uses stored text data to analyze the content of the documents using natural language processing (NLP) techniques. Specifically, it performs morphological analysis on the text, divides it into sentences, and performs syntactic analysis to classify it into sections such as "goals," "strategies," "budgets," and "risks." This analysis utilizes TensorFlow's natural language processing library (for example, the BERT model).

[1103] Based on the analyzed data, the server detects inconsistencies. For example, it identifies contradictory statements within the document, such as "there is a possibility that market share will not increase" and "market share will increase by 10%." These inconsistencies are listed and managed using the Python pandas library.

[1104] Next, the server generates specific improvement suggestions based on the list of inconsistencies and user sentiment data. This process uses generative AI models such as OpenAI's GPT-3. For example, it might generate suggestions such as "conduct regular market research" as a risk mitigation measure.

[1105] The generated analysis results and improvement suggestions are sent from the server to the terminal. Secure protocols are used to encrypt the data during this process to ensure security. The terminal displays the received analysis results and improvement suggestions to the user in an easy-to-understand format. HTML5, CSS, and JavaScript are used for the display. If the user reacts to the improvement suggestions, their reaction is analyzed again by the sentiment engine, and the server adjusts the feedback content as needed. Furthermore, if the user dislikes the improvement suggestions, the system also includes a function to suggest more specific countermeasures.

[1106] As a concrete example, a user uploads the following document file to the system:

[1107] ---

[1108] Title: Marketing Strategy for New Products

[1109] 1. Goal

[1110] Increase market share by 10%

[1111] Increase brand awareness by 50%

[1112] 2. Strategy

[1113] We will conduct an advertising campaign.

[1114] We will focus on adding new features.

[1115] 3. Budget

[1116] Allocate 5 million yen to the advertising campaign.

[1117] Allocate 3 million yen to feature development.

[1118] 4. Risk

[1119] The advertising campaign's effectiveness fell short of expectations.

[1120] There is a possibility that market share will not increase.

[1121] ---

[1122] An example of a contradiction detected by the server is when the "Goals" section states "Increase market share by 10%", while the "Risks" section states "There is a possibility that market share may not increase." The server also proposes specific action plans as improvement suggestions, such as "Conduct regular market research to mitigate risks."

[1123] As described above, this system can provide users with specific and effective feedback, improving the efficiency and quality of document creation.

[1124] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1125] Step 1:

[1126] The user logs into the system and uploads a document file. The file format can be PDF, Word, or Text. When the user selects a file saved on their device and clicks the upload button, the terminal stores the file in a temporary storage area. Simultaneously, the user's facial expressions and voice are analyzed by an emotion engine (general emotion recognition software) to acquire emotion data. At this stage, the input consists of the document file selected by the user and the user's facial expressions and voice, while the output consists of the document file stored in temporary storage and the acquired emotion data.

[1127] Step 2:

[1128] The terminal sends document files and sentiment data stored in temporary storage to the server. The data is encrypted (e.g., AES encryption) and transmitted using a secure protocol (e.g., HTTPS). The input in this process is the files and sentiment data in temporary storage, and the output is the files and data sent to the server.

[1129] Step 3:

[1130] The server receives the document file from the terminal. If the document file is in PDF or image format, the server uses the Optical Character Recognition (OCR) technology of the Google Cloud Vision API to convert the file into text data. Word or Text format files skip this step and are treated as text data as is. The input for this step is the received document file, and the output is text data.

[1131] Step 4:

[1132] The server analyzes the converted text data using natural language processing (NLP) techniques. First, it performs morphological analysis to divide the text into sentences, then syntactically analyzes it to classify it into sections such as goals, strategies, budgets, and risks. TensorFlow's BERT model is used for this analysis. The input is text data, and the output is the analyzed result.

[1133] Step 5:

[1134] The server detects inconsistencies in the document based on the analysis results. For example, if the "Goals" section states "increase market share by 10%" while the "Risks" section states "there is a possibility that market share will not increase," the server will identify this inconsistency. A natural language understanding model is used for this detection. The input is the analyzed text data, and the output is a list of inconsistencies.

[1135] Step 6:

[1136] The server generates specific improvement suggestions based on a list of inconsistencies and user sentiment data. This generation uses generative AI models such as OpenAI GPT-3. For example, it creates specific action plans to mitigate risks (e.g., "Conduct regular market research to mitigate risks"). The input is a list of inconsistencies and sentiment data, and the output is the generated improvement suggestions.

[1137] Step 7:

[1138] The server sends the analysis results and suggested improvements to the terminal. The data is encrypted and transmitted using a secure protocol. The input to this process is the suggested improvements and analysis results, and the output is the data sent to the terminal.

[1139] Step 8:

[1140] The terminal displays analysis results and improvement suggestions received from the server to the user. HTML5, CSS, and JavaScript are used for display. If the user reacts to the improvement suggestions, the terminal analyzes the reaction using an emotion engine and adjusts the feedback content as needed. The input is data from the server, and the output is the analysis results and improvement suggestions displayed to the user.

[1141] This completes the overall processing flow of the system that efficiently creates documents and provides high-quality feedback to users.

[1142] (Application Example 2)

[1143] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1144] Traditional document management systems struggle to effectively detect inconsistencies in complex documents and generate improvement suggestions. Furthermore, the lack of feedback that takes user sentiment into account reduces the efficiency of document improvement. These issues can negatively impact operational efficiency and quality control, particularly in quality control reports and production plans used in factories.

[1145] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving and transmitting data files to the server, means for receiving data files and converting them into text data, means for analyzing the content of the data using natural language processing technology and detecting inconsistencies, means for generating improvement proposals for the data based on the analysis results, means equipped with an emotion recognition engine for analyzing the user's facial expressions and voice and acquiring emotion data, and means for generating appropriate feedback based on the analysis results and emotion data. This makes it possible to efficiently and accurately detect inconsistencies in the data and generate improvement proposals, and further improve the efficiency of data improvement by providing feedback that takes the user's emotions into consideration.

[1146] "Document files" refer to documents and papers saved in formats such as PDF, Word, or text.

[1147] A "server" refers to a computer system that receives data files via network communication, converts them into text data, and performs analysis on them.

[1148] "Text data" refers to string information obtained by converting document files using optical character recognition (OCR) or other methods.

[1149] "Natural language processing technology" refers to the technology that enables computers to understand, analyze, and extract information from human language.

[1150] A "contradiction" refers to a logically inconsistent section or discrepancy in information within a document.

[1151] An "emotion recognition engine" refers to a software or hardware system that analyzes a user's facial expressions and voice to identify their emotional state.

[1152] "Means by which the server generates suggested improvements to the document based on the analysis results" refers to the system component that has the function of generating specific corrections and methods to suggest to the user based on the analysis results and sentiment data.

[1153] "Appropriate feedback" refers to feedback that takes into account the user's emotional state and includes encouragement when suggesting improvements to the material.

[1154] A "terminal" refers to a device used by users to upload data files and receive and display analysis results and improvement suggestions.

[1155] This invention is a system that analyzes data files, detects inconsistencies, and generates improvement suggestions. When a user uploads a data file, the system receives it and performs analysis. Furthermore, it is characterized by its ability to recognize the user's emotional state and provide appropriate feedback.

[1156] System Configuration

[1157] 1. User terminal

[1158] The user terminal is a device used to send data files to the server, and can be a PC or tablet. The terminal is also equipped with an emotion recognition engine, which captures the user's facial expressions and voice using the camera and microphone. Specifically, OpenCV is used for camera analysis and TensorFlow for emotion recognition.

[1159] 2. Server

[1160] A server is a computer system that receives data files, converts them into text data, and performs analysis. It uses software such as the following:

[1161] Optical character recognition technology: Tesseract OCR

[1162] Natural Language Processing Technologies: spaCy, Hugging Face Transformers

[1163] Generative AI models: GPT-3, etc.

[1164] Program processing

[1165] The document files uploaded by the user are first sent from the terminal to the server. SSL / TLS, a secure communication protocol, is used for transmission. The server converts the received document files into text data using Tesseract OCR. Next, this text data is analyzed using spaCy and Hugging Face Transformers to detect inconsistencies.

[1166] Examples of data analysis

[1167] Consider the case where a user uploads the following documents:

[1168] Title: Marketing Strategy for New Products

[1169] 1. Goal

[1170] Increase market share by 10%

[1171] Increase brand awareness by 50%

[1172] 2. Strategy

[1173] We will conduct an advertising campaign.

[1174] We will focus on adding new features.

[1175] 3. Budget

[1176] Allocate 5 million yen to the advertising campaign.

[1177] Allocate 3 million yen to feature development.

[1178] 4. Risk

[1179] The advertising campaign's effectiveness fell short of expectations.

[1180] There is a possibility that market share will not increase.

[1181] The server analyzes this text data and detects, for example, inconsistencies between the "Goals" section and the "Risks" section. Next, it generates improvement suggestions based on these inconsistencies using a generative AI model (GPT-3).

[1182] Emotion recognition and feedback

[1183] When the device displays inconsistencies or suggestions for improvement in the document to the user, the emotion recognition engine analyzes the user's facial expressions and voice. If the user expresses dissatisfaction, the server generates positive feedback and displays it to the user through the device.

[1184] For example, the prompt statement is as follows:

[1185] "Please review the quality control report and point out any inconsistencies or problems. Furthermore, please propose solutions for improvement."

[1186] The above describes the embodiments for carrying out this invention.

[1187] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1188] Step 1:

[1189] The user uploads a document file. The input is a document file (PDF, Word, or text format), and the user's facial expressions and voice are also captured using the device's camera and microphone. This allows for the acquisition of both the document file and the user's emotional data.

[1190] Step 2:

[1191] The terminal preprocesses the data files and sentiment data and sends them to the server. The transmission uses a secure communication protocol (SSL / TLS). The input consists of the data files and sentiment data, and the output is the data files and sentiment data received by the server.

[1192] Step 3:

[1193] The server converts the received document file into text data using Optical Character Recognition (OCR) technology. Specifically, Tesseract OCR software is used. The input is a document file, and the output is text data.

[1194] Step 4:

[1195] The server analyzes text data using natural language processing (NLP) techniques. Morphological analysis, syntactic analysis, and contextual analysis are performed sequentially. The software used is spaCy and Hugging Face Transformers. The input is text data, and the output is the analysis result.

[1196] Step 5:

[1197] The server detects inconsistencies in the data based on the analysis results. For example, it checks for inconsistencies between objectives and risks. The input is the analysis results, and the output is a list of detected inconsistencies.

[1198] Step 6:

[1199] The server generates suggested improvements to the document based on inconsistencies. It uses a generative AI model (such as GPT-3) to propose specific corrections and methods. The input is a list of detected inconsistencies and user sentiment data, and the output is the suggested improvements.

[1200] Step 7:

[1201] The server sends the analysis results and improvement suggestions to the terminal. The input is the analysis results and improvement suggestions, and the output is the analysis results and improvement suggestions sent by the server.

[1202] Step 8:

[1203] The terminal displays a list of inconsistencies and suggested improvements received from the server to the user. When the user responds to the suggested improvements, the terminal re-analyzes that reaction using an emotion recognition engine. The input is the analysis results and suggested improvements, while the output is the display data for the user to see and the re-acquired emotion data.

[1204] Step 9:

[1205] The device adjusts the feedback content according to the user's emotions and resends it to the server. The input is the user's reaction data, and the output is the adjusted feedback and its transmission data.

[1206] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1207] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1208] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1209] [Fourth Embodiment]

[1210] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1211] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1212] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1213] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1214] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1215] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1216] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1217] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1218] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1219] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1220] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1221] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1222] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1223] The present invention is a system comprising means for receiving a data file from a user and sending it to a server; means for the server to receive the data file and convert it into text data; means for the server to analyze the content of the data using natural language processing technology and detect inconsistencies; means for the server to generate proposed improvements to the data based on the analysis results; and means for a terminal to display a list of inconsistencies and proposed improvements received from the server to the user.

[1224] Program execution (no code)

[1225] Importing data

[1226] The terminal receives document files from the user. Users can upload document files in formats such as PDF, Word, and Text. These document files are then sent to the server by the terminal.

[1227] Receiving and converting data

[1228] The server receives the data files sent from the terminal. The received files are converted into text data using optical character recognition (OCR) or other conversion technologies.

[1229] Analysis of the materials

[1230] The server analyzes the converted text data using natural language processing (NLP) techniques. This analysis includes morphological analysis, syntactic analysis, and contextual analysis. The server identifies each section of the document (e.g., "Goals," "Strategy," "Budget," "Risks") and analyzes the content of each section.

[1231] Detection and listing of inconsistencies

[1232] The server detects inconsistencies in the document based on the analysis results. For example, if the target section states "increase market share by 10%" but the risk section states "there is a possibility that market share will not increase," this will be detected as an inconsistency. The detected inconsistencies are listed.

[1233] Generating improvement plans

[1234] The server generates specific improvement suggestions to enhance the appeal of the materials. For example, it might generate a suggestion such as, "Conduct regular market research after advertising campaigns to confirm the effectiveness of the plan while mitigating risks."

[1235] Displaying Results

[1236] The terminal receives a list of inconsistencies and suggested improvements from the server. The user can review this information displayed on the terminal and revise the document. Because the inconsistencies and suggestions are presented specifically and in detail, the user can effectively improve the document.

[1237] Specific example

[1238] Example document

[1239] The user uploads the following document files to the system:

[1240] ---

[1241] Title: Marketing Strategy for New Products

[1242] 1. Goal

[1243] Increase market share by 10%

[1244] Increase brand awareness by 50%

[1245] 2. Strategy

[1246] We will conduct an advertising campaign.

[1247] We will focus on adding new features.

[1248] 3. Budget

[1249] Allocate 5 million yen to the advertising campaign.

[1250] Allocate 3 million yen to feature development.

[1251] 4. Risk

[1252] The advertising campaign's effectiveness fell short of expectations.

[1253] There is a possibility that market share will not increase.

[1254] ---

[1255] Inconsistency detection results and proposed improvements

[1256] Examples of inconsistencies detected by the server:

[1257] The "Goals" section states "increase market share by 10%", but the "Risks" section contradicts this by stating "the possibility that market share may not increase."

[1258] Examples of improvement suggestions provided by the server:

[1259] "To mitigate risks, we regularly conduct market research after advertising campaigns and fine-tune our plans accordingly."

[1260] This allows users to receive specific and detailed feedback, enabling them to effectively improve the materials.

[1261] The following describes the processing flow.

[1262] Step 1:

[1263] The user uploads the document file to their device. The document file can be in PDF, Word, or Text format.

[1264] Step 2:

[1265] The terminal converts the data files received from the user into a format suitable for the system. For example, it performs pre-processing such as verifying the file format and compressing the files as needed.

[1266] Step 3:

[1267] The terminal sends the pre-processed data files to the server. A secure protocol is used for transmission to maintain data integrity.

[1268] Step 4:

[1269] The server receives the data file sent from the terminal. The received file is temporarily stored in preparation for subsequent processing.

[1270] Step 5:

[1271] The server converts the received document files into text data. For PDF and image files, optical character recognition (OCR) technology is used to extract the text information.

[1272] Step 6:

[1273] The server analyzes text data using natural language processing (NLP) techniques. It sequentially performs morphological analysis, syntactic analysis, and contextual analysis to understand the meaning of the material.

[1274] Step 7:

[1275] The server identifies the overall structure of the document. For example, it organizes information into sections such as "goals," "strategies," "budget," and "risks."

[1276] Step 8:

[1277] The server analyzes the contents of each identified section in detail to detect inconsistencies and discrepancies. For example, it might identify a contradiction between "the possibility of not being able to increase market share" and "increasing market share by 10%."

[1278] Step 9:

[1279] The server creates a list of inconsistencies based on the inconsistencies it detects. This list includes a description of the specific inconsistency and the section in which it exists.

[1280] Step 10:

[1281] The server generates specific improvement suggestions to enhance the appeal of the materials. For example, it proposes concrete action plans such as "conduct regular market research to mitigate risks."

[1282] Step 11:

[1283] The terminal receives a list of inconsistencies and suggested improvements from the server. The received data is displayed in a user-friendly format.

[1284] Step 12:

[1285] The user reviews the list of inconsistencies and suggested improvements displayed on the terminal. They can then revise the document as needed and re-upload it to the system.

[1286] In this way, the system efficiently analyzes the documents created by the user, identifies inconsistencies, and provides concrete suggestions for improvement.

[1287] (Example 1)

[1288] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1289] In traditional document creation and analysis, manually checking and correcting inconsistencies within documents is an extremely cumbersome and time-consuming process. Therefore, there is a need for a system that can efficiently and automatically detect inconsistencies in documents and suggest improvements.

[1290] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1291] In this invention, the server includes means for receiving a data file from a user and sending it to the server; means for the server to receive the data file and convert it into text data; means for the server to analyze the content of the data using natural language processing technology and detect inconsistencies; means for the server to generate proposed improvements to the data based on the analysis results; and means for the terminal to display a list of inconsistencies and proposed improvements received from the server to the user. This enables the user to efficiently detect inconsistencies in the data and revise the data based on detailed proposed improvements.

[1292] A "document file" refers to document data created in formats such as PDF, Word, or Text, which is uploaded by the user.

[1293] A "server" is a computer system that receives data files, converts them into text data, analyzes the content of the data using natural language processing technology, detects inconsistencies, and generates improvement proposals.

[1294] A "terminal" is a device used by users to upload document files, receive and display a list of inconsistencies and suggested improvements sent from the server.

[1295] "Text data" refers to character data converted from document files and analyzed using natural language processing techniques.

[1296] "Natural language processing technology" is a technology that interprets meaning through morphological analysis, syntactic analysis, and contextual analysis of text data, and extracts various kinds of information.

[1297] A "contradiction" refers to inconsistent descriptions or conflicting information found within a document.

[1298] An "improvement proposal" is a specific and detailed suggestion generated by the server based on the analysis results, aimed at enhancing the appeal of the document.

[1299] Optical Character Recognition (OCR) is a technology that reads characters from document files in formats such as images and PDFs and converts them into text data.

[1300] A "section" refers to individual divisions within a document, such as "goals," "strategies," "budgets," and "risks."

[1301] A "generative AI model" is an artificial intelligence technology that generates natural language text based on input data.

[1302] The "HTTP protocol" is a communication protocol used to send and receive data, such as document files and analysis results, between a server and a terminal.

[1303] This invention relates to a system in which a user sends a data file to a server via a terminal, the server analyzes the data file to detect inconsistencies, and generates improvement proposals. The following hardware and software are used to implement this system.

[1304] First, the user uploads document files such as PDFs, Word documents, and Text documents using their own device (PC, smartphone, etc.). Then, the device sends these document files to the server via the HTTP protocol. To ensure data security, it is desirable to use the HTTPS protocol.

[1305] The server temporarily stores the received data files in storage. The received files are converted into text data using optical character recognition (OCR) technology. Specific software such as Tesseract OCR can be used. The converted text data is temporarily stored using a text editor (e.g., Notepad).

[1306] Next, the server analyzes the text data using natural language processing (NLP) techniques. This analysis includes morphological analysis, syntactic analysis, and contextual analysis. Specifically, it uses NLP libraries such as SpaCy and the Natural Language Toolkit (NLTK). The server identifies each section of the document (e.g., goals, strategies, budget, risks) and analyzes the content of each in detail.

[1307] Based on the analysis results, the server detects inconsistencies within the document. For example, if the target section states "increase market share by 10%", but the risk section states "there is a possibility that market share will not increase," this will be detected as an inconsistency. The detected inconsistencies are listed and stored in the database.

[1308] The server then uses a generative AI model (e.g., GPT-4) to automatically generate specific improvement suggestions to enhance the appeal of the materials. For example, it might generate suggestions such as, "To mitigate risks, conduct regular market research after advertising campaigns and fine-tune the plan."

[1309] Finally, the terminal receives a list of inconsistencies and suggested improvements from the server. This information is displayed on the user's terminal screen, allowing the user to review and correct the document. The terminal has a suitable interface for displaying the information received via the HTTP protocol on its screen.

[1310] Specific example

[1311] The user uploads the following document files to the system:

[1312] ---

[1313] Title: Marketing Strategy for New Products

[1314] 1. Goal

[1315] Increase market share by 10%

[1316] Increase brand awareness by 50%

[1317] 2. Strategy

[1318] We will conduct an advertising campaign.

[1319] We will focus on adding new features.

[1320] 3. Budget

[1321] Allocate 5 million yen to the advertising campaign.

[1322] Allocate 3 million yen to feature development.

[1323] 4. Risk

[1324] The advertising campaign's effectiveness fell short of expectations.

[1325] There is a possibility that market share will not increase.

[1326] ---

[1327] Examples of inconsistencies detected by the server:

[1328] The target section states "increase market share by 10%", but the risk section contradicts this by stating "the possibility that market share may not increase."

[1329] Examples of improvement suggestions provided by the server:

[1330] "To mitigate risks, we regularly conduct market research after advertising campaigns and fine-tune our plans accordingly."

[1331] As a result, users can receive specific and detailed feedback, enabling them to effectively improve the materials.

[1332] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1333] Step 1: Upload your document files

[1334] Users access the system interface using their own devices and upload document files. Input files can be in PDF, Word, or Text format. The device sends the submitted files to the server using the HTTP protocol.

[1335] Step 2: Receiving the document files

[1336] The server receives data files sent from the terminal. The input is the data files sent from the terminal. The received files are temporarily stored in the server's storage.

[1337] Step 3: Conversion using Optical Character Recognition (OCR) technology

[1338] The server converts the received document file into text data using optical character recognition (OCR) technology. Specifically, it uses software such as Tesseract OCR. The input is the document file saved in step 2. The output is text data.

[1339] Step 4: Retaining text data

[1340] The server temporarily saves the converted text data to a text editor (e.g., Notepad) or a database. The input is the text data converted by OCR. The output is the temporarily saved text data.

[1341] Step 5: Analysis using Natural Language Processing (NLP)

[1342] The server analyzes text data through morphological analysis, syntactic analysis, and contextual analysis. Specifically, it uses NLP libraries such as SpaCy and Natural Language Toolkit (NLTK). The input is stored text data. The output is data that analyzes the content of each section (e.g., goals, strategies, budget, risks) in detail.

[1343] Step 6: Detecting inconsistencies

[1344] The server detects inconsistencies within the document based on the analysis results. For example, it detects a contradiction if the target section states "increase market share by 10%" while the risk section states "there is a possibility that market share will not increase." The input is the data from each analyzed section. The output is a list of the detected inconsistencies.

[1345] Step 7: Generating improvement plans

[1346] The server generates suggested improvements to the document based on a list of inconsistencies. It uses a generative AI model (e.g., GPT-4) to automatically generate specific improvement suggestions. The input is a list of inconsistencies. The output is specific improvement suggestions.

[1347] Step 8: Submit inconsistencies and suggestions for improvement

[1348] The server sends a list of generated inconsistencies and suggested improvements to the terminal. The input is the list of inconsistencies and suggested improvements. The output is the data sent to the terminal.

[1349] Step 9: Displaying the results

[1350] The terminal displays a list of inconsistencies and suggested improvements received from the server to the user. The input is the list of inconsistencies and suggested improvements sent from the server. The output is the information displayed on the terminal's screen. The user can review the displayed content and effectively revise the document.

[1351] (Application Example 1)

[1352] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1353] In autonomous vehicles, there is a need for a system that can automatically detect inconsistencies in pre-created driving plans and safety plans and quickly and accurately propose improvements. Traditional manual verification and revision processes are time-consuming and labor-intensive, and also carry a high risk of human error, thus increasing the need for automated systems. Furthermore, there is a lack of efficient means to integrate a large amount of relevant information and identify optimal driving routes and risk factors.

[1354] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1355] This invention includes a server that receives data files from a user and transmits them to the server; a server that receives data files and converts them into text data; a server that analyzes the content of the data using natural language processing technology and detects inconsistencies; a server that generates proposed improvements to the data based on the analysis results; and a system for an autonomous vehicle that receives driving plan data and proposes safe driving routes and identifies risk factors based on the analysis results. This enables rapid and accurate analysis and revision of driving plan and safety plan data, as well as the identification of risk factors and the proposal of optimal driving routes.

[1356] A "document file" refers to a document file uploaded by a user and saved in formats such as PDF, Word, or Text.

[1357] A "server" is a computer system that receives data files, converts them into text data, and analyzes the content of the data using natural language processing technology.

[1358] "Text data" refers to data in string format converted from a document file.

[1359] "Natural language processing technology" refers to techniques such as morphological analysis, syntactic analysis, and contextual analysis used to understand and analyze the content of documents.

[1360] A "contradiction" refers to a point in a document where different sections or items do not match or have logical inconsistencies.

[1361] A "proposal for improvement" refers to specific ideas for changes or modifications proposed to enhance the appeal and reliability of the document in response to detected inconsistencies.

[1362] A "terminal" is a device used by users to upload document files and receive and display a list of inconsistencies and suggested improvements sent from the server.

[1363] An "autonomous vehicle" is an automobile that can autonomously perform driving and safety planning and receive analysis results and suggestions through an external interface.

[1364] A "driving plan document" is a document that describes the route and safety measures planned for an autonomous vehicle.

[1365] A "safe driving route" is the optimal route proposed based on the analysis results, designed to ensure that an autonomous vehicle safely reaches its destination.

[1366] "Risk factors" are items or conditions in a driving plan or safety plan that could potentially cause accidents or malfunctions.

[1367] Optical Character Recognition (OCR) is a technology that converts characters stored as images into text data.

[1368] The system that realizes this invention is comprised of the following hardware and software.

[1369] Hardware and software configuration

[1370] Hardware:

[1371] Onboard computers for autonomous vehicles (e.g., NVIDIA DRIVE, Intel Movidius, etc.)

[1372] Vehicle internal interface (display, touchscreen, etc.)

[1373] Servers (those with high-performance computing resources)

[1374] software:

[1375] Optical character recognition technology (e.g., Tesseract OCR)

[1376] Natural language processing technologies (e.g., Google Cloud Natural Language API, spaCy)

[1377] Database management system (e.g., MySQL)

[1378] Data processing and data calculation

[1379] 1. Importing data files: The autonomous vehicle's system receives driving plan data. Users upload data files through the onboard interface. The latest traffic information from in-vehicle sensors such as radar, LiDAR, and cameras is also integrated.

[1380] 2. Receiving and Converting Documents: The server converts received document files into text data using OCR technology. Tesseract OCR is used for OCR.

[1381] 3. Document Analysis: The server analyzes the converted text data using NLP technology. Morphological analysis, syntactic analysis, and contextual analysis are performed to identify each section of the driving plan and safety plan. Google Cloud Natural Language API and spaCy are used for this process.

[1382] 4. Detection and Listing of Inconsistencies: The server detects inconsistencies in the document based on the analysis results. For example, it might detect a discrepancy such as "driving on a highway, but planning to enter a restricted area." These inconsistencies are then listed.

[1383] 5. Generation of Improvement Plans: The server generates specific improvement plans to resolve the detected inconsistencies. For example, improvement plans may include "suggest an alternative route to avoid restricted areas."

[1384] 6. Display of Results: The autonomous vehicle's interface displays a list of detected inconsistencies and suggested improvements. The user can review this and modify the driving plan and safety plan.

[1385] Specific example

[1386] Prompt: Analyze the following planning document and propose inconsistencies and improvements.

[1387] Planning documents:

[1388] ---

[1389] Departure point: Shinjuku

[1390] Destination: Shibuya

[1391] Route: Expressway entrance

[1392] Estimated arrival time: 3:00 PM

[1393] Note: Construction is scheduled to begin soon at the highway entrance.

[1394] ---

[1395] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1396] Step 1:

[1397] The user uploads the driving plan documents via the autonomous vehicle's onboard interface. The user inputs the document file in a specific format (PDF, Word, Text, etc.) into the interface and then uploads it. At this stage, the document file is saved to the onboard computer.

[1398] Step 2:

[1399] The onboard computer sends the uploaded data file to the server. The server receives the data file and converts it into text data using optical character recognition (Tesseract OCR) technology. The input is a data file, and the output is text data. In this process, text information in image format is extracted as text.

[1400] Step 3:

[1401] The server analyzes the converted text data using natural language processing technologies (Google Cloud Natural Language API, spaCy). The input is text data, and the output is the analysis result. Specifically, morphological analysis, syntactic analysis, and contextual analysis are performed to identify each section of the travel plan and safety plan (starting point, destination, waypoints, notes, etc.).

[1402] Step 4:

[1403] The server detects inconsistencies in the document based on the analysis results. The input is the analysis results, and the output is a list of inconsistencies. For example, it identifies discrepancies such as "driving on a highway, but planning to enter a restricted area." The server uses a specific algorithm to identify logical inconsistencies and contradictions and lists them.

[1404] Step 5:

[1405] The server generates improvement suggestions based on the inconsistencies it detects. The input is a list of inconsistencies, and the output is improvement suggestions. Specifically, it generates suggestions for corrections such as "suggest an alternative route to avoid restricted areas." The generated improvement suggestions are specific and actionable so that users can actually take action.

[1406] Step 6:

[1407] The server sends a list of inconsistencies and suggested improvements to the autonomous vehicle's interface. The terminal receives this information and displays it to the user. The input is the list of inconsistencies and suggested improvements, and the output is the information displayed on the interface. The user reviews this information and modifies the driving plan as needed. Specifically, information is conveyed to the user visually using a display or touchscreen.

[1408] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1409] The present invention is a system that combines means for receiving a data file from a user and sending it to a server, means for the server to receive the data file and convert it into text data, means for the server to analyze the content of the data using natural language processing technology and detect inconsistencies, means for the server to generate suggestions for improving the data based on the analysis results, means for a terminal to display a list of inconsistencies and suggestions for improvement received from the server to the user, and an emotion engine that recognizes the user's emotions.

[1410] Program execution (no code)

[1411] Importing data

[1412] The user uploads a document file to the device. The document file can be in PDF, Word, or Text format. During the upload process, the device analyzes the user's facial expressions and voice using an emotion engine to obtain the user's emotional data.

[1413] Receiving and converting data

[1414] The terminal sends the pre-processed data file to the server. A secure protocol is used for transmission. The server receives the data file sent from the terminal and converts the file into text data using optical character recognition (OCR) technology.

[1415] Analysis of the materials

[1416] The server analyzes text data using natural language processing (NLP) techniques. It sequentially performs morphological analysis, syntactic analysis, and contextual analysis to understand the meaning of the document. The analysis results are categorized into sections of the document (e.g., "Goals," "Strategy," "Budget," "Risks").

[1417] Detection and listing of inconsistencies

[1418] The server detects inconsistencies in the document based on the analysis results. For example, it identifies a contradiction between "the possibility that market share cannot be expected to increase" and "a 10% increase in market share." The server then lists these inconsistencies.

[1419] Generating improvement plans

[1420] The server generates specific improvement suggestions to enhance the appeal of the materials based on user sentiment data and analysis results. For example, if a user is dissatisfied, it provides proactive feedback to alleviate that dissatisfaction.

[1421] Displaying Results

[1422] The device receives a list of inconsistencies and suggested improvements from the server. The device displays these in a user-friendly format. When the user responds to the suggested improvements, the device analyzes the reaction using an emotion engine and adjusts the feedback accordingly.

[1423] Specific example

[1424] Example document

[1425] The user uploads the following document files to the system:

[1426] ---

[1427] Title: Marketing Strategy for New Products

[1428] 1. Goal

[1429] Increase market share by 10%

[1430] Increase brand awareness by 50%

[1431] 2. Strategy

[1432] We will conduct an advertising campaign.

[1433] We will focus on adding new features.

[1434] 3. Budget

[1435] Allocate 5 million yen to the advertising campaign.

[1436] Allocate 3 million yen to feature development.

[1437] 4. Risk

[1438] The advertising campaign's effectiveness fell short of expectations.

[1439] There is a possibility that market share will not increase.

[1440] ---

[1441] Inconsistency detection results and proposed improvements

[1442] Examples of inconsistencies detected by the server:

[1443] The "Goals" section states "increase market share by 10%", but the "Risks" section contradicts this by stating "the possibility that market share may not increase."

[1444] Examples of improvement suggestions provided by the server:

[1445] We propose concrete action plans, such as "conducting regular market research to mitigate risks."

[1446] Utilizing the Emotion Engine

[1447] If the device analyzes the user's facial expressions and voice and determines that the user is dissatisfied with the display of inconsistencies in the document, the server will provide proactive feedback to the user. For example, it might display an encouraging message such as, "Correcting this section will significantly improve the persuasiveness of your marketing strategy."

[1448] As a result, users can receive specific and detailed feedback, enabling them to effectively improve their materials. Furthermore, because appropriate feedback that takes user emotions into consideration is provided, the efficiency and quality of material creation improve.

[1449] The following describes the processing flow.

[1450] Step 1:

[1451] The user uploads document files to their device. During the upload process, the device's camera and microphone are used to record the user's facial expressions and voice, which are then sent to the emotion engine.

[1452] Step 2:

[1453] The terminal converts the data files received from the user into a format suitable for the system and verifies file integrity. If necessary, it performs preprocessing such as compressing or splitting the files.

[1454] Step 3:

[1455] The terminal sends pre-processed data files and user sentiment data to the server. An encrypted protocol is used for transmission to ensure security.

[1456] Step 4:

[1457] The server receives data files and sentiment data sent from the terminal. The data files are temporarily stored and used for subsequent processing.

[1458] Step 5:

[1459] The server converts the received document files into text data. For PDF and image files, optical character recognition (OCR) technology is used to extract the text information.

[1460] Step 6:

[1461] The server analyzes text data using natural language processing (NLP) techniques. It sequentially performs morphological analysis, syntactic analysis, and contextual analysis to understand the intent and content of the material.

[1462] Step 7:

[1463] The server categorizes the analyzed text data into sections. For example, it might divide it into sections such as "Goals," "Strategy," "Budget," and "Risks."

[1464] Step 8:

[1465] The server analyzes the contents of each section in detail to detect inconsistencies and contradictions. For example, it identifies contradictory statements such as "the possibility of not being able to increase market share" and "increasing market share by 10%."

[1466] Step 9:

[1467] The server analyzes emotional data and evaluates the user's emotional state. For example, it identifies emotions such as anxiety, excitement, and satisfaction.

[1468] Step 10:

[1469] The server generates improvement suggestions based on inconsistencies in the document. In doing so, it takes into account the user's emotional state, making adjustments such as adding encouraging messages if the user is feeling anxious.

[1470] Step 11:

[1471] The server sends a list of inconsistencies and suggested improvements to the terminal, including specific steps on how to implement them.

[1472] Step 12:

[1473] The terminal displays a list of inconsistencies and suggested improvements received from the server. It monitors user reactions in real time and collects further feedback using an emotion engine.

[1474] Step 13:

[1475] Users can review the list of inconsistencies and suggested improvements displayed on their device and revise the document as needed. After revisions, they can re-upload the document to the system for further checks.

[1476] This allows the system to efficiently analyze user-generated materials, identify inconsistencies, and provide concrete improvement suggestions that take user sentiment into consideration.

[1477] (Example 2)

[1478] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1479] Conventional document creation support systems converted document content into text data and analyzed the content using natural language processing technology. However, they lacked the ability to generate feedback and improvement suggestions that considered user emotions, making it difficult to provide concrete and effective suggestions for improving the quality of documents. Furthermore, they were not adequately able to detect inconsistencies in user-submitted documents and immediately suggest appropriate improvements. As a result, there was insufficient support for enhancing the persuasiveness and effectiveness of documents, leading to the problem of time-consuming document creation.

[1480] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1481] In this invention, the server includes means for receiving a data file and converting it into text data, means for analyzing the content of the data using natural language processing technology and detecting inconsistencies, means for generating proposed improvements to the data based on the analysis results, and means for adjusting the proposed improvements based on the user's sentiment data. This makes it possible to quickly detect inconsistencies in the data and provide specific and effective improvement proposals while taking the user's sentiment into consideration.

[1482] A "document file" is a document that a user uploads to the system, and it can be in PDF, Word, or Text format.

[1483] A "server" is a computer system that receives data files, converts them into text data, analyzes their content, detects inconsistencies, and generates improvement suggestions.

[1484] A "terminal" is a device used by the user, which is used for uploading document files, sending and receiving data with the server, displaying analysis results and improvement suggestions, and acquiring sentiment data.

[1485] "Text data" refers to character information converted from document files, which is then analyzed by the server.

[1486] "Natural language processing technology" refers to techniques that analyze text data, understand the content of materials, and detect inconsistencies, and includes morphological analysis, syntactic analysis, and contextual analysis.

[1487] A "contradiction" refers to inconsistencies within a document. For example, it might state in the objectives section that the goal is to "increase market share by 10%" while in the risks section that there is a "possibility that market share may not increase."

[1488] An "improvement suggestion" is a proposal to make the content of the materials generated by the server based on analysis results and user sentiment data more effective.

[1489] "Emotional data" refers to information acquired by a device by analyzing the user's facial expressions and voice, and it indicates the user's emotional state.

[1490] "Emotion recognition means" refers to technology that allows a device to analyze the user's facial expressions and voice to acquire emotional data.

[1491] "Optical character recognition technology" is a technology that extracts character information from images and PDF document files and converts it into text data.

[1492] "Extracting information section by section" refers to the process where the server organizes and analyzes the content of each part of the document (e.g., goals, strategies, budget, risks).

[1493] A "generative AI model" refers to artificial intelligence technology in which a server generates specific improvement suggestions based on analysis results and sentiment data.

[1494] This invention is a system that receives data files from a user and transmits them to a server. This system converts the data files into text data, analyzes the content of the data using natural language processing technology to detect inconsistencies, and generates improvement suggestions for the data based on the analysis results. Furthermore, the terminal acquires the user's emotional data, and the server adjusts the improvement suggestions based on that emotional data, thereby providing specific and effective feedback tailored to the user.

[1495] First, the user logs into the system using their device and uploads a document file. The document file can be provided in formats such as PDF, Word, or Text. Once the user uploads the file, the device saves it to temporary storage. Simultaneously, the device analyzes the user's facial expressions and voice using an emotion engine (e.g., Microsoft Azure Emotion API) to obtain the user's emotional data.

[1496] Next, the terminal sends the data files stored in temporary storage and the acquired emotional data to the server using a secure protocol (e.g., HTTPS). To ensure data security, the data is encrypted before transmission (e.g., AES encryption).

[1497] The server receives the document file from the terminal and verifies its format. If the document file is a PDF or image, the server uses the OCR function of the Google Cloud Vision API to convert the document into text data. Word or text files are treated as text data as is. The converted text data is stored in the database.

[1498] The server uses stored text data to analyze the content of the documents using natural language processing (NLP) techniques. Specifically, it performs morphological analysis on the text, divides it into sentences, and performs syntactic analysis to classify it into sections such as "goals," "strategies," "budgets," and "risks." This analysis utilizes TensorFlow's natural language processing library (for example, the BERT model).

[1499] Based on the analyzed data, the server detects inconsistencies. For example, it identifies contradictory statements within the document, such as "there is a possibility that market share will not increase" and "market share will increase by 10%." These inconsistencies are listed and managed using the Python pandas library.

[1500] Next, the server generates specific improvement suggestions based on the list of inconsistencies and user sentiment data. This process uses generative AI models such as OpenAI's GPT-3. For example, it might generate suggestions such as "conduct regular market research" as a risk mitigation measure.

[1501] The generated analysis results and improvement suggestions are sent from the server to the terminal. Secure protocols are used to encrypt the data during this process to ensure security. The terminal displays the received analysis results and improvement suggestions to the user in an easy-to-understand format. HTML5, CSS, and JavaScript are used for the display. If the user reacts to the improvement suggestions, their reaction is analyzed again by the sentiment engine, and the server adjusts the feedback content as needed. Furthermore, if the user dislikes the improvement suggestions, the system also includes a function to suggest more specific countermeasures.

[1502] As a concrete example, a user uploads the following document file to the system:

[1503] ---

[1504] Title: Marketing Strategy for New Products

[1505] 1. Goal

[1506] Increase market share by 10%

[1507] Increase brand awareness by 50%

[1508] 2. Strategy

[1509] We will conduct an advertising campaign.

[1510] We will focus on adding new features.

[1511] 3. Budget

[1512] Allocate 5 million yen to the advertising campaign.

[1513] Allocate 3 million yen to feature development.

[1514] 4. Risk

[1515] The advertising campaign's effectiveness fell short of expectations.

[1516] There is a possibility that market share will not increase.

[1517] ---

[1518] An example of a contradiction detected by the server is when the "Goals" section states "Increase market share by 10%", while the "Risks" section states "There is a possibility that market share may not increase." The server also proposes specific action plans as improvement suggestions, such as "Conduct regular market research to mitigate risks."

[1519] As described above, this system can provide users with specific and effective feedback, improving the efficiency and quality of document creation.

[1520] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1521] Step 1:

[1522] The user logs into the system and uploads a document file. The file format can be PDF, Word, or Text. When the user selects a file saved on their device and clicks the upload button, the terminal stores the file in a temporary storage area. Simultaneously, the user's facial expressions and voice are analyzed by an emotion engine (general emotion recognition software) to acquire emotion data. At this stage, the input consists of the document file selected by the user and the user's facial expressions and voice, while the output consists of the document file stored in temporary storage and the acquired emotion data.

[1523] Step 2:

[1524] The terminal sends document files and sentiment data stored in temporary storage to the server. The data is encrypted (e.g., AES encryption) and transmitted using a secure protocol (e.g., HTTPS). The input in this process is the files and sentiment data in temporary storage, and the output is the files and data sent to the server.

[1525] Step 3:

[1526] The server receives the document file from the terminal. If the document file is in PDF or image format, the server uses the Optical Character Recognition (OCR) technology of the Google Cloud Vision API to convert the file into text data. Word or Text format files skip this step and are treated as text data as is. The input for this step is the received document file, and the output is text data.

[1527] Step 4:

[1528] The server analyzes the converted text data using natural language processing (NLP) techniques. First, it performs morphological analysis to divide the text into sentences, then syntactically analyzes it to classify it into sections such as goals, strategies, budgets, and risks. TensorFlow's BERT model is used for this analysis. The input is text data, and the output is the analyzed result.

[1529] Step 5:

[1530] The server detects inconsistencies in the document based on the analysis results. For example, if the "Goals" section states "increase market share by 10%" while the "Risks" section states "there is a possibility that market share will not increase," the server will identify this inconsistency. A natural language understanding model is used for this detection. The input is the analyzed text data, and the output is a list of inconsistencies.

[1531] Step 6:

[1532] The server generates specific improvement suggestions based on a list of inconsistencies and user sentiment data. This generation uses generative AI models such as OpenAI GPT-3. For example, it creates specific action plans to mitigate risks (e.g., "Conduct regular market research to mitigate risks"). The input is a list of inconsistencies and sentiment data, and the output is the generated improvement suggestions.

[1533] Step 7:

[1534] The server sends the analysis results and suggested improvements to the terminal. The data is encrypted and transmitted using a secure protocol. The input to this process is the suggested improvements and analysis results, and the output is the data sent to the terminal.

[1535] Step 8:

[1536] The terminal displays analysis results and improvement suggestions received from the server to the user. HTML5, CSS, and JavaScript are used for display. If the user reacts to the improvement suggestions, the terminal analyzes the reaction using an emotion engine and adjusts the feedback content as needed. The input is data from the server, and the output is the analysis results and improvement suggestions displayed to the user.

[1537] This completes the overall processing flow of the system that efficiently creates documents and provides high-quality feedback to users.

[1538] (Application Example 2)

[1539] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1540] Traditional document management systems struggle to effectively detect inconsistencies in complex documents and generate improvement suggestions. Furthermore, the lack of feedback that takes user sentiment into account reduces the efficiency of document improvement. These issues can negatively impact operational efficiency and quality control, particularly in quality control reports and production plans used in factories.

[1541] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving and transmitting data files to the server, means for receiving data files and converting them into text data, means for analyzing the content of the data using natural language processing technology and detecting inconsistencies, means for generating improvement proposals for the data based on the analysis results, means equipped with an emotion recognition engine for analyzing the user's facial expressions and voice and acquiring emotion data, and means for generating appropriate feedback based on the analysis results and emotion data. This makes it possible to efficiently and accurately detect inconsistencies in the data and generate improvement proposals, and further improve the efficiency of data improvement by providing feedback that takes the user's emotions into consideration.

[1542] "Document files" refer to documents and papers saved in formats such as PDF, Word, or text.

[1543] A "server" refers to a computer system that receives data files via network communication, converts them into text data, and performs analysis on them.

[1544] "Text data" refers to string information obtained by converting document files using optical character recognition (OCR) or other methods.

[1545] "Natural language processing technology" refers to the technology that enables computers to understand, analyze, and extract information from human language.

[1546] A "contradiction" refers to a logically inconsistent section or discrepancy in information within a document.

[1547] An "emotion recognition engine" refers to a software or hardware system that analyzes a user's facial expressions and voice to identify their emotional state.

[1548] "Means by which the server generates suggested improvements to the document based on the analysis results" refers to the system component that has the function of generating specific corrections and methods to suggest to the user based on the analysis results and sentiment data.

[1549] "Appropriate feedback" refers to feedback that takes into account the user's emotional state and includes encouragement when suggesting improvements to the material.

[1550] A "terminal" refers to a device used by users to upload data files and receive and display analysis results and improvement suggestions.

[1551] This invention is a system that analyzes data files, detects inconsistencies, and generates improvement suggestions. When a user uploads a data file, the system receives it and performs analysis. Furthermore, it is characterized by its ability to recognize the user's emotional state and provide appropriate feedback.

[1552] System Configuration

[1553] 1. User terminal

[1554] The user terminal is a device used to send data files to the server, and can be a PC or tablet. The terminal is also equipped with an emotion recognition engine, which captures the user's facial expressions and voice using the camera and microphone. Specifically, OpenCV is used for camera analysis and TensorFlow for emotion recognition.

[1555] 2. Server

[1556] A server is a computer system that receives data files, converts them into text data, and performs analysis. It uses software such as the following:

[1557] Optical character recognition technology: Tesseract OCR

[1558] Natural Language Processing Technologies: spaCy, Hugging Face Transformers

[1559] Generative AI models: GPT-3, etc.

[1560] Program processing

[1561] The document files uploaded by the user are first sent from the terminal to the server. SSL / TLS, a secure communication protocol, is used for transmission. The server converts the received document files into text data using Tesseract OCR. Next, this text data is analyzed using spaCy and Hugging Face Transformers to detect inconsistencies.

[1562] Examples of data analysis

[1563] Consider the case where a user uploads the following documents:

[1564] Title: Marketing Strategy for New Products

[1565] 1. Goal

[1566] Increase market share by 10%

[1567] Increase brand awareness by 50%

[1568] 2. Strategy

[1569] We will conduct an advertising campaign.

[1570] We will focus on adding new features.

[1571] 3. Budget

[1572] Allocate 5 million yen to the advertising campaign.

[1573] Allocate 3 million yen to feature development.

[1574] 4. Risk

[1575] The advertising campaign's effectiveness fell short of expectations.

[1576] There is a possibility that market share will not increase.

[1577] The server analyzes this text data and detects, for example, inconsistencies between the "Goals" section and the "Risks" section. Next, it generates improvement suggestions based on these inconsistencies using a generative AI model (GPT-3).

[1578] Emotion recognition and feedback

[1579] When the device displays inconsistencies or suggestions for improvement in the document to the user, the emotion recognition engine analyzes the user's facial expressions and voice. If the user expresses dissatisfaction, the server generates positive feedback and displays it to the user through the device.

[1580] For example, the prompt statement is as follows:

[1581] "Please review the quality control report and point out any inconsistencies or problems. Furthermore, please propose solutions for improvement."

[1582] The above describes the embodiments for carrying out this invention.

[1583] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1584] Step 1:

[1585] The user uploads a document file. The input is a document file (PDF, Word, or text format), and the user's facial expressions and voice are also captured using the device's camera and microphone. This allows for the acquisition of both the document file and the user's emotional data.

[1586] Step 2:

[1587] The terminal preprocesses the data files and sentiment data and sends them to the server. The transmission uses a secure communication protocol (SSL / TLS). The input consists of the data files and sentiment data, and the output is the data files and sentiment data received by the server.

[1588] Step 3:

[1589] The server converts the received document file into text data using Optical Character Recognition (OCR) technology. Specifically, Tesseract OCR software is used. The input is a document file, and the output is text data.

[1590] Step 4:

[1591] The server analyzes text data using natural language processing (NLP) techniques. Morphological analysis, syntactic analysis, and contextual analysis are performed sequentially. The software used is spaCy and Hugging Face Transformers. The input is text data, and the output is the analysis result.

[1592] Step 5:

[1593] The server detects inconsistencies in the data based on the analysis results. For example, it checks for inconsistencies between objectives and risks. The input is the analysis results, and the output is a list of detected inconsistencies.

[1594] Step 6:

[1595] The server generates suggested improvements to the document based on inconsistencies. It uses a generative AI model (such as GPT-3) to propose specific corrections and methods. The input is a list of detected inconsistencies and user sentiment data, and the output is the suggested improvements.

[1596] Step 7:

[1597] The server sends the analysis results and improvement suggestions to the terminal. The input is the analysis results and improvement suggestions, and the output is the analysis results and improvement suggestions sent by the server.

[1598] Step 8:

[1599] The terminal displays a list of inconsistencies and suggested improvements received from the server to the user. When the user responds to the suggested improvements, the terminal re-analyzes that reaction using an emotion recognition engine. The input is the analysis results and suggested improvements, while the output is the display data for the user to see and the re-acquired emotion data.

[1600] Step 9:

[1601] The device adjusts the feedback content according to the user's emotions and resends it to the server. The input is the user's reaction data, and the output is the adjusted feedback and its transmission data.

[1602] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1603] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1604] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1605] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1606] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1607] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1608] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1609] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1610] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1611] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1612] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1613] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1614] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1615] 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.

[1616] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1617] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1618] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1619] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1620] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1621] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1622] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[1623] The following is further disclosed regarding the embodiments described above.

[1624] (Claim 1)

[1625] A means of receiving data files from users and sending them to the server,

[1626] A means for the server to receive a data file and convert it into text data,

[1627] A server analyzes the content of the document using natural language processing technology and has a means to detect inconsistencies.

[1628] A server provides a means for generating suggestions for improving the document based on the analysis results,

[1629] A means for the terminal to display to the user a list of inconsistencies and suggested improvements received from the server,

[1630] A system that includes this.

[1631] (Claim 2)

[1632] The system according to claim 1, which uses optical character recognition technology to convert a data file into text data.

[1633] (Claim 3)

[1634] The system according to claim 1, wherein the server extracts information for each section of the document and analyzes the contents of each section.

[1635] "Example 1"

[1636] (Claim 1)

[1637] A means of receiving data files from users and sending them to the server,

[1638] A means for the server to receive a data file and convert it into text data,

[1639] A server analyzes the content of the document using natural language processing technology and has a means to detect inconsistencies.

[1640] A server provides a means for generating suggestions for improving the document based on the analysis results,

[1641] A means for the terminal to display to the user a list of inconsistencies and suggested improvements received from the server,

[1642] A system that includes this.

[1643] (Claim 2)

[1644] The system according to claim 1, which uses optical character recognition technology to convert a data file into text data.

[1645] (Claim 3)

[1646] The system according to claim 1, wherein the server extracts information for each section of the document and analyzes the contents of each section.

[1647] (Claim 4)

[1648] The system according to claim 1, wherein the user's terminal uploads a document file and sends the document file to a server using the HTTP protocol.

[1649] (Claim 5)

[1650] The system according to claim 1, which uses a natural language processing library that analyzes text data received by the server through morphological analysis, syntactic analysis, and contextual analysis.

[1651] (Claim 6)

[1652] The system according to claim 1, wherein when the server detects inconsistencies, it identifies and lists inconsistencies based on descriptions of different directions between the objective and the risk.

[1653] (Claim 7)

[1654] The system according to claim 1, wherein the server automatically generates specific improvement proposals using an AI model.

[1655] (Claim 8)

[1656] The system according to claim 1, further comprising means for displaying information received via the HTTP protocol on the screen when the terminal displays inconsistencies and suggested improvements to the user.

[1657] "Application Example 1"

[1658] (Claim 1)

[1659] A means of receiving data files from users and sending them to the server,

[1660] A means for the server to receive a data file and convert it into text data,

[1661] A server analyzes the content of the document using natural language processing technology and has a means to detect inconsistencies.

[1662] A server provides a means for generating suggestions for improving the document based on the analysis results,

[1663] A means for the terminal to display to the user a list of inconsistencies and suggested improvements received from the server,

[1664] A means by which the autonomous vehicle system receives driving plan data and, based on the analysis results, proposes a safe driving route and identifies risk factors,

[1665] A system that includes this.

[1666] (Claim 2)

[1667] The system according to claim 1, which uses optical character recognition technology to convert a data file into text data.

[1668] (Claim 3)

[1669] The system according to claim 1, wherein the server extracts information for each section of the document and analyzes the contents of each section.

[1670] "Example 2 of combining an emotion engine"

[1671] (Claim 1)

[1672] A means of receiving data files from users and sending them to the server,

[1673] A means for the server to receive a data file and convert it into text data,

[1674] A server analyzes the content of the document using natural language processing technology and has a means to detect inconsistencies.

[1675] A server provides a means for generating suggestions for improving the document based on the analysis results,

[1676] A means for the terminal to display to the user a list of inconsistencies and suggested improvements received from the server,

[1677] A means for recognizing emotions in which a terminal acquires user emotion data,

[1678] A means by which the server adjusts improvement suggestions based on user sentiment data,

[1679] A system that includes this.

[1680] (Claim 2)

[1681] The system according to claim 1, which uses optical character recognition technology to convert a data file into text data.

[1682] (Claim 3)

[1683] The system according to claim 1, wherein the server extracts information for each section of the document and analyzes the contents of each section.

[1684] "Application example 2 when combining with an emotional engine"

[1685] (Claim 1)

[1686] A means of receiving data files from users and sending them to the server,

[1687] A means for the server to receive a data file and convert it into text data,

[1688] A server analyzes the content of the document using natural language processing technology and has a means to detect inconsistencies.

[1689] A server provides a means for generating suggestions for improving the document based on the analysis results,

[1690] A means for the terminal to display to the user a list of inconsistencies and suggested improvements received from the server,

[1691] A means for a device to have an emotion recognition engine that analyzes the user's facial expressions and voice and acquires emotion data,

[1692] A server provides means for generating appropriate feedback based on analysis results and sentiment data,

[1693] A system that includes this.

[1694] (Claim 2)

[1695] The system according to claim 1, which uses optical character recognition technology to convert a data file into text data.

[1696] (Claim 3)

[1697] The system according to claim 1, wherein the server extracts information for each section of the document and analyzes the contents of each section. [Explanation of Symbols]

[1698] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of receiving data files from users and sending them to the server, A means for the server to receive a data file and convert it into text data, A server analyzes the content of the document using natural language processing technology and has a means to detect inconsistencies. A server provides a means for generating suggestions for improving the document based on the analysis results, A means for the terminal to display to the user a list of inconsistencies and suggested improvements received from the server, A system that includes this.

2. The system according to claim 1, which uses optical character recognition technology to convert a data file into text data.

3. The system according to claim 1, wherein the server extracts information for each section of the document and analyzes the contents of each section.

Images