System

Abstract

A system is provided.SOLUTION: This system includes a means for analyzing an uploaded document by using a natural language processing technique, a means for using a generation model for generating a summary on the basis of an analysis result, and a means for distributing the generated summary and an extracted trend or keyword to the terminal of each department.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In companies, analyzing and summarizing the large volumes of daily business reports and information generated every day takes a great deal of time and effort. As a result, the early detection of new trends and important information is delayed, making it difficult to make quick and accurate decisions. Traditional manual analysis methods cannot handle the huge amount of data, and there is a risk of delays and missing information. To solve this problem, a system is needed that can efficiently and quickly analyze daily business reports and extract summaries and trends. [Means for solving the problem]

[0005] This invention provides a system that includes a means for analyzing uploaded documents using natural language processing technology, a means for using a generative model to generate summaries based on the analysis results, and a means for distributing the generated summaries and extracted trends and keywords to terminals in each department. The uploaded documents are decomposed using natural language processing technology, and tokenized and analyzed for parts of speech. The analysis results are then generated in a data frame format and passed to a summary generation model, resulting in efficient generation of summaries. A report is then created that includes important keywords and trends in the generated summaries, and this report is distributed to each department, enabling quick and accurate sharing of business insights.

[0006] "Natural language processing technology" is a technology that enables computers to understand, interpret, and generate language that humans naturally speak.

[0007] "Upload" is an operation in which a user sends data or files from their own device to a server or cloud.

[0008] "Documents" refer to text information that is created on a daily basis, such as a company's daily report or business report.

[0009] "Analysis" is the process of breaking down input data to understand its structure and meaning.

[0010] The "analysis results" are data indicating the divided information and structure obtained through the analysis of the document.

[0011] A "summary" is a short, condensed version of the main points or meaning of the entire document.

[0012] A "generative model" is an algorithm or machine learning model that is trained to generate a specific output (here, a summary) from given input data.

[0013] "Trends" refer to recurring themes or tendencies found within a document.

[0014] "Keywords" refer to words or phrases that are particularly important in a document.

[0015] "Distribution" is the act of sending the generated summary and related information to appropriate recipients.

[0016] "Terminal" refers to a device used by a user, such as a computer or smartphone.

[0017] "Decomposition" is the process of breaking down input text into smaller components (sentences, words, tokens, etc.).

[0018] "Tokenization" is the process of dividing text into smaller units such as words or phrases.

[0019] "Part of speech analysis" is the process of identifying and assigning a part of speech (verb, noun, adjective, etc.) to each word in a sentence. [Brief explanation of the drawings]

[0020] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0021] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0022] First, the terms used in the following description will be explained.

[0023] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0024] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0025] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0026] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0027] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0028] [First embodiment]

[0029] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0030] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0031] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0032] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0033] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0034] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0035] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0036] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0037] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0038] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0039] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0040] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0041] Overall system overview

[0042] This system analyzes business daily reports and information uploaded by users on a server, automatically generates summaries and trend information, and distributes them to each department, thereby streamlining information sharing. Specifically, it combines natural language processing technology, generative models, and keyword and trend extraction algorithms to achieve fast and accurate information analysis and distribution.

[0043] Program processing flow and detailed explanation

[0044] Uploading and Receiving Files

[0045] User: Upload daily reports and business information files through the device interface.

[0046] On your device: Automatically validates the file format and size to ensure it is safe before sending it to the server.

[0047] Calling the API for receiving data

[0048] Terminal: Call the API endpoint for sending a file and send the file contents to the server.

[0049] Server: Stores the received file in temporary storage and then passes it to the natural language processing module.

[0050] Analysis using natural language processing

[0051] Server: Using a natural language processing engine, the uploaded document is broken down into sentences and words, tokenized, and analyzed for parts of speech to determine the meaning and relationships of each word.

[0052] Server: Prepares the analysis results in a data frame format and passes them to the summary generation module.

[0053] Generate a summary

[0054] Server: Uses a generative model to generate a summary from the analysis results, evaluating the importance of each sentence to ensure that important information is not omitted.

[0055] Server: Formats the generated summary and presents it in a format that is easy for users to understand.

[0056] Keyword and trend extraction

[0057] Server: Uses text mining technology to extract important keywords and trends from the analysis results.

[0058] Server: Adds extracted keywords and trends to a summary and creates the final report.

[0059] Report distribution

[0060] Server: Prepares the generated reports for distribution to each department, customizing the format based on user settings.

[0061] Server: Distributes reports to each department's terminal and confirms receipt.

[0062] View and share your results

[0063] Terminal: Displays received reports to users, allowing each department to quickly access the information they need.

[0064] Users: Review reports and share them with other employees or departments as needed, supporting efficient information sharing and quick decision-making.

[0065] Specific examples

[0066] Sales department daily report analysis and summary

[0067] 1. User: Upload the sales report (e.g. sales_report_2023_10_15.pdf) to the system from the terminal.

[0068] 2. Terminal: Validate the file and send it to the server.

[0069] 3. Server: Receives the daily reports and performs analysis using a natural language processing engine.

[0070] 4. Server: Based on the analysis results, a generative model creates a summary, summarizing key selling points and customer trends.

[0071] 5. Server: Delivers reports containing summary and trend information to sales department terminals.

[0072] 6. Terminal: Receives the report and displays it for the sales team to view.

[0073] 7. User: Review the reports and make decisions that will influence your sales strategy.

[0074] This system allows users to efficiently analyze the large amount of information generated in daily business operations and use it to make quick decisions.

[0075] The processing flow will be explained below.

[0076] Step 1:

[0077] User: Uploads daily reports and business information files using the device interface. Through the file selection dialog, select the specific file and press the send button.

[0078] Step 2:

[0079] Terminal: Validates the selected file, checking that the file format is supported and that the size is within the regulated range. If there are no problems, generates an API request to send the file to the server.

[0080] Step 3:

[0081] Terminal: Sends the generated API request to the server's API endpoint, usually using a secure protocol such as HTTPS.

[0082] Step 4:

[0083] Server: Received files are stored in temporary storage, where file metadata and other information are recorded and prepared for subsequent processing.

[0084] Step 5:

[0085] Server: Calls the natural language processing (NLP) engine to analyze the contents of the uploaded file. First, it breaks down the file into sentences and words, and performs tokenization and part-of-speech analysis.

[0086] Step 6:

[0087] Server: Stores the results of tokenization and part-of-speech analysis. Specifically, it stores the analysis results in data frame format and passes them to the next summary generation process.

[0088] Step 7:

[0089] Server: Uses a generative model for summary generation to generate a summary from the analysis results. Extracts important points from the document and summarizes the information.

[0090] Step 8:

[0091] Server: Evaluates the generated summary and formats it in an appropriate format. Here, the summary is organized into a concise and easy-to-read format so that users can quickly grasp the information.

[0092] Step 9:

[0093] Server: Along with the summary, we use text mining techniques to extract important keywords and trends. We use TF-IDF and word clouds to extract highly relevant information.

[0094] Step 10:

[0095] Server: Adds the extracted results to the summary and creates a final report, which includes the summary, keywords, and trend information.

[0096] Step 11:

[0097] Server: Prepares the created report for distribution to each department, customizes the format according to the user's settings, and generates a distribution list.

[0098] Step 12:

[0099] Server: Delivers reports to terminals in each department. Sends reports to designated recipients using a distribution protocol.

[0100] Step 13:

[0101] Terminal: Stores received reports and displays them in the user interface, allowing users to easily view and review the reports.

[0102] Step 14:

[0103] Users: Check the displayed report and share it with other employees or departments as needed, including by email or by sending a link to an internal chat tool.

[0104] This series of steps allows users to quickly and accurately analyze business information, obtain summary and trend information, and make effective decisions.

[0105] Example 1

[0106] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0107] In today's business environment, a huge amount of information is generated every day, creating a need to analyze and summarize that information quickly and accurately, and share it efficiently. However, traditional methods require a significant amount of time and effort to analyze and share information. Furthermore, manual information analysis is prone to errors and carries a high risk of important information being overlooked. Furthermore, the wide variety of information formats makes it difficult to analyze information in a unified and efficient manner. There is a need for a system that solves these problems.

[0108] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0109] In this invention, the server includes: means for a user to upload documents to the system through an electronic device interface; means for a terminal to verify the format and size of the uploaded documents and transmit them to the server after confirming their safety; means for the server to call a data reception API to receive and temporarily store the files; means for the server to use natural language processing technology to break down the uploaded documents into sentence units and words and perform tokenization and part-of-speech analysis; means for the server to generate the analysis results in a data frame format and pass them to a summary generation model; means for the server to use the generation model to generate summaries from the analysis results; means for the server to extract important keywords and trends and add them to the summaries to create a final report; means for the server to distribute the generated summaries and the extracted trends and keywords to terminals in each department; and means for the terminal to display the received reports to the user and enable information to be shared with other departments, thereby enabling efficient analysis of information and rapid sharing.

[0110] "User" refers to the entity that uses the system to upload materials and receive analysis results.

[0111] "Electronic Device" means any device that allows a User to manipulate and upload materials to the System, including, for example, a computer, smartphone, or tablet.

[0112] An "interface" is a point of contact between a user and a system for exchanging information. Examples include web browsers and dedicated applications.

[0113] "Documents" refers to business daily reports and other written information uploaded to the system, including, for example, PDF, Excel, Word files, etc.

[0114] "Terminal" means an electronic device operated by a User that verifies the format and size of uploaded materials and ensures their security.

[0115] "Server" refers to the computer system that receives uploaded materials, generates analysis and summaries, and distributes them to each department.

[0116] "Data Reception API" refers to a program interface for sending materials from a terminal to a server.

[0117] "Temporary storage" refers to the temporary storage of received materials by a server, including, for example, cloud storage and databases.

[0118] "Natural language processing technology" is a technology for analyzing uploaded documents, and is primarily a function for tokenization and part-of-speech analysis.

[0119] "Tokenization" refers to the process of breaking down a document into words or sentences.

[0120] "Part-of-speech analysis" refers to the process of identifying the part of speech of each word in a document and analyzing its grammatical relationships.

[0121] "Data frame format" refers to a data structure for storing analysis results in a table-like format.

[0122] A "summary generation model" refers to an algorithm or machine learning model for generating a summary based on the analysis results.

[0123] "Keywords" refer to words that appear frequently in a document and have important meanings.

[0124] "Trends" refers to major themes or tendencies extracted from the information in a document.

[0125] The "final report" refers to a document that compiles the generated summary, keywords, and trend information.

[0126] "Delivery" refers to the process of sending the generated final report to the terminals of each department.

[0127] "Receiving" refers to the terminal receiving distribution from the server.

[0128] "Display" refers to the terminal visually indicating to the user the report received.

[0129] "Sharing" refers to the act of a user collaborating and communicating the received report with other employees or departments.

[0130] MODE FOR CARRYING OUT THE INVENTION

[0131] This invention is a system that improves the efficiency of information sharing by analyzing business daily reports and information uploaded by users on a server, automatically generating summaries and trend information, and distributing them to each department. The system includes the following main elements:

[0132] Uploading and Receiving Files

[0133] Users upload files such as daily reports and business information through the interface of their electronic device. Specifically, they use the browser's file selection dialog to select a document such as "sales_report_2023_10_15.pdf" and click the upload button. The device then verifies the file's format and size (e.g., PDF, Excel, Word), confirms its security, and sends it to the server.

[0134] Calling the API for receiving data

[0135] The device calls an API endpoint (e.g., https: / / api.example.com / upload) to upload the file to the server. The server temporarily stores the file received at the API endpoint (e.g., Amazon S3 or Google Cloud Storage) and passes the path to the natural language processing engine.

[0136] Analysis using natural language processing

[0137] The server uses a natural language processing (NLP) engine (e.g., SpaCy, NLTK) to break down the uploaded document into sentences and words. Specifically, it performs tokenization, part-of-speech analysis, and removes stop words. The analysis results are formatted as a data frame (e.g., Pandas DataFrame) and passed to the summary generation model.

[0138] Generate a summary

[0139] The server uses a generative AI model (e.g., GPT-3, BERT) to generate a summary from the analysis results. The summary is generated by extracting particularly important sentences based on their importance scores and assembling them into a sentence. The generated summary is then organized in a format that is easy for users to understand (e.g., Markdown or HTML).

[0140] Keyword and trend extraction

[0141] The server uses text mining techniques (e.g., TF-IDF, Word2Vec) to extract important keywords and trends from the analysis results, including frequency analysis and semantic relevance assessment. The extracted keywords and trend information are added to a summary to create the final report.

[0142] Report distribution

[0143] The server prepares the generated reports for distribution to each department, customizing the reports based on user settings and departmental formatting requirements, and delivering the reports via email or the company's internal portal, where receipt is acknowledged.

[0144] View and share your results

[0145] The device displays the received report to the user, for example, through a web interface or desktop notification. The user can review the report's contents and share the link in an internal chat tool to share key selling points with other departments.

[0146] Prompt Sentence Examples

[0147] Below are some example input prompts for the generative AI model that can be used in this system:

[0148] "Today, we uploaded the sales report for October 15, 2023. Please extract trending keywords and summarize them. Don't forget to include your key selling points."

[0149] This system allows users to efficiently analyze the large amount of information generated in daily business operations and use it to make quick decisions.

[0150] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0151] Step 1:

[0152] The user uploads the document using an electronic device. The user opens a file selection dialog using the interface of their device, selects the daily report or business information file (e.g., "sales_report_2023_10_15.pdf"), and clicks the upload button. The input is the file selected by the user, and the output is the file selection information sent to the device.

[0153] Step 2:

[0154] The terminal verifies the format and size of the uploaded file to confirm safety. The terminal checks the file format (e.g. PDF, Excel, Word) and whether the file size exceeds the specified limit. It also performs a virus check. The input is the file uploaded by the user, and the output is the verification result (safety confirmation).

[0155] Step 3:

[0156] The device sends the verified file to the server. The device calls an API endpoint (e.g., https: / / api.example.com / upload) to upload the verified file to the server via an HTTP POST request. The input is the verified file, and the output is a notification that the file was successfully uploaded.

[0157] Step 4:

[0158] The server temporarily stores the received file. The server stores the file received at the API endpoint in temporary storage (e.g., Amazon S3 or Google Cloud Storage) and records the path. The input is the file sent from the device, and the output is the path to the temporary storage.

[0159] Step 5:

[0160] The server analyzes the file using a natural language processing (NLP) engine. The server passes the saved file to an NLP engine (e.g., SpaCy, NLTK) and breaks it down into sentences and words. It performs tokenization, part-of-speech analysis, and removes stop words, and then organizes the analysis results in data frame format (e.g., Pandas DataFrame). The input is the path to the saved file, and the output is a data frame of the analysis results.

[0161] Step 6:

[0162] The server generates a summary using a generative AI model. After tokenization and part-of-speech analysis, the server passes the data frame to a generative AI model (e.g., GPT-3, BERT) to generate a summary. It extracts particularly important sentences based on their importance scores and assembles them into a summary. The input is the analysis result data frame, and the output is the generated summary.

[0163] Step 7:

[0164] The server uses text mining techniques to extract keywords and trends. The server uses TF-IDF models and Word2Vec to extract important keywords and trends from the analysis results. This includes word frequency analysis and semantic relevance assessment. The input is a data frame of the analysis results, and the output is the extracted keywords and trends.

[0165] Step 8:

[0166] The server creates a final report and adds keywords and trends to the summary. The server adds keyword and trend information to the generated summary and formats the final report. The report is prepared in a format that is easy for users to understand (e.g., Markdown or HTML). The input is the generated summary and extracted keywords and trends, and the output is the final report.

[0167] Step 9:

[0168] The server prepares the final report for delivery to each department. The server customizes the report based on user settings and departmental formatting requirements. The server delivers the report via email or through the company's internal portal. The input is the final report and the output is the customized report.

[0169] Step 10:

[0170] The server distributes the report and the terminal confirms receipt. The server distributes the customized report to each department's terminal and receives a receipt confirmation. The input is the customized report and the output is the receipt confirmation result.

[0171] Step 11:

[0172] The terminal displays the received reports to the user. The terminal makes the reports available to the user through a web interface or desktop notification. The input is the received report and the output is the report displayed to the user.

[0173] Step 12:

[0174] The user reviews the report and shares it with other departments. The user reviews the received report and shares important selling points and analysis results with other employees and departments using an internal chat tool, etc. The input is the report displayed to the user, and the output is the results shared with other departments.

[0175] This is the flow of processing for this system's program, which enables efficient analysis and rapid sharing of information.

[0176] (Application example 1)

[0177] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0178] Conventional autonomous driving systems have had difficulty efficiently analyzing massive amounts of sensor data and operational information, summarizing it, and providing it to drivers and managers. This can lead to delays in quickly identifying important trends and operational issues and taking appropriate measures. Furthermore, there is no established method for displaying or distributing the generated data, which can lead to delays in information sharing and decision-making.

[0179] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0180] In this invention, the server includes means for analyzing uploaded documents using natural language processing technology, means for using a generative model to generate summaries based on the analysis results, means for distributing the generated summaries and extracted trends and keywords to terminals in each department, means for uploading sensor data and operation information and analyzing and summarizing them, and means for extracting important information from the operation data of the autonomous vehicle and displaying it to the driver or manager. This allows the operation information of the autonomous vehicle to be analyzed efficiently and quickly, summarized, and provided, making it possible to immediately grasp important trends and issues and respond promptly.

[0181] "Natural language processing technology" is a technology that automatically analyzes and understands text data and processes human language using a computer.

[0182] A "generative model" is a machine learning algorithm that generates new data based on given data.

[0183] A "trend" refers to the tendency of fluctuations or patterns in data or information over a period of time.

[0184] "Keywords" refer to words or phrases that have particular significance in documents or data.

[0185] "Sensor data" refers to measurement data such as temperature, pressure, position, and speed obtained from various sensors.

[0186] "Operation information" is data relating to the operation status of a vehicle, including location information, speed, route, traffic conditions, and the like.

[0187] An "autonomous vehicle" refers to a vehicle that drives autonomously without the need for human operation.

[0188] An "abstract" is a short sentence or text that succinctly summarizes the important points of the original document or data.

[0189] An "API endpoint" refers to a specific address or operation that allows software to communicate with other software.

[0190] "Customized format" means data or reports that have been formatted in a specific way according to the user's requirements or purposes.

[0191] A "prompt sentence" is text containing instructions or questions to be provided as input to a generative model.

[0192] The system for implementing this invention analyzes autonomous vehicle operation data and sensor data uploaded by users on a server, automatically generates summaries and trend information, and distributes them to drivers and managers. A detailed description of the system is as follows.

[0193] Program processing overview

[0194] Uploading and Receiving Files

[0195] Users upload operational data for autonomous vehicles via their smartphones or on-board displays. The device automatically verifies the file format and size, confirms that it is safe, and then sends it to the server.

[0196] Calling the API for receiving data

[0197] The device calls the API endpoint for sending a file and sends the file contents to the server, which then stores the received file in temporary storage and passes it to the natural language processing module.

[0198] Analysis using natural language processing

[0199] The server uses a natural language processing engine (e.g., spaCy or NLTK) to break down the uploaded operation logs and error reports into sentences and words. It then performs tokenization and part-of-speech analysis to analyze the meaning and relationships of each word. The analysis results are prepared in data frame format and passed to a summary generation model (e.g., GPT series).

[0200] Generate a summary

[0201] The server uses a generative model to generate a summary from the analysis results. During this process, it evaluates the importance of each sentence to ensure that important information is not omitted. The generated summary is then formatted and presented in a format that is easy for users to understand.

[0202] Keyword and trend extraction

[0203] The server uses text mining techniques to extract important keywords and trends from the analysis results, adds them to a summary, and creates a final report.

[0204] Report distribution

[0205] The server prepares the generated report for distribution to the driver or manager, customizes the report format based on the user's settings, and delivers the report to a smartphone or in-vehicle display for confirmation of receipt.

[0206] View and share your results

[0207] The terminal displays the received reports to the user, allowing them to quickly access the information they need. Users can then review the reports and share them with other drivers and fleet managers, supporting efficient information sharing and quick decision-making.

[0208] Specific examples

[0209] We will provide a concrete example of uploading log data from a self-driving vehicle's daily operations, analyzing it in real time, and extracting trends and important issues.

[0210] Example prompt for a generative AI model:

[0211] "Please extract important information from the operational data log below and generate a written summary."

[0212] Data Log:

[0213] This system will enable drivers and managers of autonomous vehicles to efficiently analyze and summarize data, enabling them to make quick decisions.

[0214] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0215] Step 1:

[0216] Uploading and Receiving Files

[0217] Users use their smartphones or in-vehicle displays to upload operational and sensor data from their autonomous vehicles. The device automatically validates the file format and size to ensure the content is safe. The input data is an operational data file, and the output is a file ready to be sent to the server. Specifically, the device checks the file's header information to ensure it is in an acceptable format and size.

[0218] Step 2:

[0219] Calling the API for receiving data

[0220] The terminal calls the API endpoint for file transmission and sends the contents of the operation data file to the server. The input data is the verified operation data file, and the output is a file stored in the server's temporary storage. Specifically, the terminal generates an API request, includes the file data in the request body, and sends it to the server. The server stores the received data in temporary storage.

[0221] Step 3:

[0222] Analysis using natural language processing

[0223] The server uses a natural language processing engine (e.g., spaCy or NLTK) to analyze the received operation data file. The input data is the operation data file stored in temporary storage, and the output is a data frame that has been tokenized and analyzed for parts of speech. Specifically, the server reads the file, tokenizes it into sentences and words, analyzes the parts of speech for each, and converts it into a data frame format.

[0224] Step 4:

[0225] Generate a summary

[0226] The server generates a summary using a generative model (e.g., GPT series) based on the analysis results. The input data is a data frame of the analysis results, and the output is a summary. Specifically, the server passes the data frame as input to the generative model, and generates a summary while evaluating important information.

[0227] Step 5:

[0228] Keyword and trend extraction

[0229] The server uses text mining technology to extract important keywords and trends from the analysis results. The input data is a summary sentence and a data frame of the analysis results, and the output is important keywords and trend information. Specifically, the server applies text mining technology to detect frequently occurring keywords and trend patterns.

[0230] Step 6:

[0231] Report creation and distribution

[0232] The server adds the extracted keywords and trends to the summary and creates a final report. The input data is the summary, keywords, and trend information, and the output is a completed report. Specifically, the server integrates the keywords and trend information into the summary and generates a report in a customized format based on user settings. The final report is then distributed to the driver's or manager's device.

[0233] Step 7:

[0234] View and share your results

[0235] The terminal displays the received report to the user. The input data is the delivered report, and the output is the information displayed to the user. Specifically, the terminal analyzes the report data and displays it in a visually easy-to-understand format. The user can review the report and share it with other drivers or fleet managers.

[0236] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0237] Overall system overview

[0238] This system analyzes daily business reports and information uploaded by users on a server, automatically generates summaries and trend information, and distributes them to each department. Furthermore, by combining it with an emotion engine, it is possible to analyze the sentiment of documents and reflect this in summaries and trend information. This allows users to analyze information while taking into account the emotional aspects of documents, enabling them to make quick and accurate decisions.

[0239] Program processing flow and detailed explanation

[0240] Uploading and Receiving Files

[0241] User: Uploads daily reports and business information files through the device interface, selects a specific file through the file selection dialog, and presses the send button.

[0242] Calling the API for receiving data

[0243] Terminal: Call the API endpoint for sending the file and send the file contents to the server. The sending protocol is usually a secure one such as HTTPS.

[0244] Analysis using natural language processing

[0245] Server: Calls the natural language processing (NLP) engine to analyze the contents of the uploaded file. First, it breaks the file down into sentences and words, then performs tokenization and part-of-speech analysis.

[0246] Server: Stores the results of tokenization and part-of-speech analysis. Specifically, the analysis results are stored in data frame format and passed to the next summary generation process.

[0247] Emotion analysis

[0248] Server: Calls the sentiment engine to analyze the sentiment of the uploaded document, identifying positive and negative phrases and words in the text and assessing the type and intensity of the sentiment.

[0249] Server: Integrates the analysis results into a summary generation model to generate summaries that include emotional information.

[0250] Generate a summary

[0251] Server: Generates a summary from the analysis results using a generative model, extracting important points from the document and summarizing the information.

[0252] Server: Evaluates the generated summary and formats it appropriately. Here, it is formatted in a concise and easy-to-read format to allow users to quickly grasp the information.

[0253] Keyword and trend extraction

[0254] Server: Uses text mining techniques to extract important keywords and trends from the analysis results, and uses TF-IDF and word clouds to extract highly relevant information.

[0255] Server: Adds extracted keywords and trends to a summary and creates a final report.

[0256] Report distribution

[0257] Server: Prepares the generated reports for distribution to each department, customizes the format according to the user's settings, and generates the distribution list.

[0258] Server: Distributes reports to each department's terminal and confirms receipt.

[0259] View and share your results

[0260] Terminal: Stores the received reports and displays them in the user interface, allowing users to easily view and check the reports.

[0261] Users: Review the displayed report and share it with other employees or departments as needed, including by email or by sending a link to an internal chat tool.

[0262] Specific examples

[0263] Sales department daily report analysis and summary

[0264] 1. User: Upload the sales report (e.g. sales_report_2023_10_15.pdf) to the system from the terminal.

[0265] 2. Terminal: Validate the file and send it to the server.

[0266] 3. Server: Receives the daily reports and performs analysis using a natural language processing engine and sentiment engine.

[0267] 4. Server: Based on the analysis results, a generative model creates a summary, summarizing key selling points, customer trends, and sentiment information.

[0268] 5. Server: Delivers reports containing summaries, trend information, and sentiment information to sales department terminals.

[0269] 6. Terminal: Receives the report and displays it for the sales team to view.

[0270] 7. User: Review the reports and make decisions that will influence your sales strategy.

[0271] This system enables users to efficiently analyze the large amount of information generated in daily business operations, quickly grasp summary and trend information, as well as sentiment information, and make effective decisions.

[0272] The processing flow will be explained below.

[0273] Step 1:

[0274] User: Uploads daily reports and business information files using the device interface. Through the file selection dialog, select the specific file and press the send button.

[0275] Step 2:

[0276] Terminal: Validates the selected file, checking whether the file format is supported and whether the size is within the regulated range. If there are no problems, generates an API request to send the file to the server.

[0277] Step 3:

[0278] Terminal: Sends the generated API request to the server's API endpoint, usually using a secure protocol such as HTTPS.

[0279] Step 4:

[0280] Server: Received files are stored in temporary storage, where file metadata and other information are recorded and prepared for subsequent processing.

[0281] Step 5:

[0282] Server: Calls the natural language processing (NLP) engine to analyze the contents of the uploaded file. First, it breaks the file down into sentences and words, then performs tokenization and part-of-speech analysis.

[0283] Step 6:

[0284] Server: Stores the results of tokenization and part-of-speech analysis. Specifically, it stores the analysis results in data frame format and passes them to the next summary generation process.

[0285] Step 7:

[0286] Server: Calls the sentiment engine to analyze the sentiment of the uploaded document, identifying positive and negative phrases and words in the text and assessing the type and intensity of the sentiment.

[0287] Step 8:

[0288] Server: Stores the analyzed emotion data and integrates it into the summary generation model, thereby generating summaries that include emotion information.

[0289] Step 9:

[0290] Server: Using a generative model for summary generation, a summary is generated based on the analysis results and sentiment data. Important points of the document are extracted and the information is summarized.

[0291] Step 10:

[0292] Server: Evaluates the generated summary and formats it in an appropriate format. Here, it is organized into a concise and easy-to-read format so that users can quickly grasp the information.

[0293] Step 11:

[0294] Server: Uses text mining techniques to extract important keywords and trends from the analysis results. Uses TF-IDF and word clouds to extract highly relevant information.

[0295] Step 12:

[0296] Server: Adds the extracted keywords and trends to a summary and creates a final report, including sentiment information.

[0297] Step 13:

[0298] Server: Prepares the created report for distribution to each department, customizes the format according to the user's settings, and generates a distribution list.

[0299] Step 14:

[0300] Server: Delivers reports to terminals in each department. Sends reports to designated recipients using a distribution protocol.

[0301] Step 15:

[0302] Terminal: Stores received reports and displays them in the user interface, allowing users to easily view and check the reports.

[0303] Step 16:

[0304] Users: Check the displayed report and share it with other employees or departments as needed. Sharing methods include email and internal chat tool links.

[0305] This detailed processing step enables users to efficiently analyze the large amount of information generated in daily business operations, quickly grasp summary and trend information, as well as sentiment information, and make effective decisions.

[0306] Example 2

[0307] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0308] In today's business environment, a huge amount of business documents and reports are generated every day, and there is a need to efficiently analyze them and quickly grasp important information. Conventional methods require manually analyzing document content and creating summaries, which is time-consuming and labor-intensive. Furthermore, it is difficult to appropriately evaluate and reflect the emotional information contained in documents. This leads to issues such as overlooking information and delays in identifying important trends.

[0309] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: a means for a user to upload a file through an interface; a means for a terminal to call an API endpoint and send the contents of the file; a means for the server to analyze the uploaded document using natural language processing technology; a means for analyzing the sentiment of the document using an emotion engine; a means for using a generative model to generate a summary based on the analysis results; a means for distributing the generated summary and extracted trends and keywords to terminals in each department; and a means for saving and displaying reports received by the terminals. This makes it possible to efficiently analyze the content and sentiment information of business documents and to quickly and accurately generate and distribute summaries and trend information.

[0310] "User" refers to the user who uploads business documents and daily reports through the system interface.

[0311] A "terminal" is a device used by a user that has the functionality to upload files and send requests to API endpoints.

[0312] "Server" refers to a central computing device that analyzes received files, generates summaries and trend information, and distributes it to each department.

[0313] "Natural language processing technology" is a technology for analyzing documents, and has the ability to break them down into sentences or words and perform tokenization and part-of-speech analysis.

[0314] An "API endpoint" is a specific URL or URI that a device uses to communicate with a server and is intended to receive HTTPS requests.

[0315] The "Sentiment Engine" is an analytical tool for identifying emotive phrases in a document and assessing positive and negative sentiment.

[0316] A "generative model" is an algorithm or AI model used to generate a summary of a document based on the results of its analysis.

[0317] "Tokenization" is the process of dividing a document into smaller units such as words or phrases.

[0318] "Part-of-speech analysis" is the process of identifying the part of speech (noun, verb, adjective, etc.) for each tokenized word or phrase.

[0319] A "data frame" is a data structure for storing analysis results, and is tabular data represented by rows and columns.

[0320] "Keywords" are words or phrases that are considered to be particularly important within a document.

[0321] "Trend information" refers to significant trends and patterns extracted from analyzed documents.

[0322] "Report" means the final output document containing the generated summary and extracted keyword trend information.

[0323] "Format" refers to the style and structure in which a report or summary is presented.

[0324] "Distribution" is the process of sending the generated report from the server to the terminals of each department.

[0325] "Display" refers to presenting the report received at the terminal on the user interface.

[0326] A "department" is a group responsible for a specific task or function in a company or organization.

[0327] In the system according to the present invention, a server analyzes business documents and daily reports uploaded by users and automatically generates summaries and trend information. Furthermore, an emotion engine can be used to analyze the emotional information of the documents and reflect this in the generated summaries and trend information. The configuration and operation of the system are described in detail below.

[0328] First, the user uploads a business document or daily report (e.g., sales_report_2023_10_15.pdf) using the terminal by selecting the appropriate file from the file selection dialog and clicking the upload button.

[0329] The device then sends the selected file to an API endpoint (e.g., "https: / / example.com / upload") using the HTTPS protocol. This request is sent as multipart form data.

[0330] The server inputs the received file into a natural language processing engine, which tokenizes and analyzes parts of speech at the sentence and word level. The analysis results are stored in a data frame format using the pandas library. The sentiment engine then evaluates the sentiment within the document based on the analysis results. It identifies positive and negative phrases and quantifies the strength of the sentiment.

[0331] The generative AI model generates a summary based on the analysis results and sentiment information. For example, it generates a summary such as, "Sales increased 15% this month, but there were many returns for a specific product group." The generated summary is formatted in Markdown or HTML for easy reading.

[0332] The server then uses text mining techniques (e.g., TF-IDF and word cloud) to extract important keywords and trend information from the analysis results. Keywords such as "sales," "surge," and "returns" are extracted. These keywords and trend information are also incorporated into the summary text to create the final report.

[0333] The server customizes the generated report according to the user's settings and prepares it for distribution to each department. The report is sent to each department's terminal based on the distribution list. This can be done by email using the SMTP protocol or by sending it to an internal notification system via REST API.

[0334] Finally, the terminal saves the received report and displays it in the user interface. The user can then review the displayed report and share it with other employees or departments as needed. This can be done via email or a link to an internal chat tool. A specific example of this is to share a link to the generated report in an internal chat so that everyone can access it.

[0335] In this way, this invention makes it possible to efficiently analyze large volumes of business documents and daily reports, and quickly and accurately generate and distribute summaries, trend information, and sentiment information, enabling users to quickly grasp important business information and make appropriate decisions.

[0336] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0337] Step 1:

[0338] User: Upload the daily report or business information file through the device interface by selecting the appropriate file (e.g., sales_report_2023_10_15.pdf) from the file selection dialog and clicking the upload button.

[0339] Input: File (for example, sales_report_2023_10_15.pdf).

[0340] Output: The file is selected on the device and ready to be uploaded.

[0341] Step 2:

[0342] On the device: After selecting the file, send an HTTPS request to the API endpoint (e.g., "https: / / example.com / upload"), which sends the file to the server as multipart form data.

[0343] Input: The selected file.

[0344] Output: A request to send a file to the server.

[0345] Step 3:

[0346] Server: The received file is input into a natural language processing (NLP) engine, which performs tokenization and part-of-speech analysis on a sentence-by-sentence and word-by-word basis. Specifically, the document is broken down into multiple words and each word is tagged with a part of speech.

[0347] Input: The contents of the uploaded file.

[0348] Output: Tokenized and part-of-speech analyzed data in dataframe format.

[0349] Step 4:

[0350] Server: The analysis results are stored in a data frame format and passed to the emotion engine. Here, the analysis results are converted into a data frame using the pandas library.

[0351] Input: Tokenized and part-of-speech analyzed data.

[0352] Output: Analysis results in data frame format.

[0353] Step 5:

[0354] Server: Uses a sentiment engine to assess the sentiment in the document, identifying positive and negative phrases and quantifying the strength of each sentiment.

[0355] Input: Analysis results in data frame format.

[0356] Output: Sentiment-analyzed data with sentiment scores.

[0357] Step 6:

[0358] Server: Generates summaries using generative AI models that integrate emotional information. They combine key document points and emotional elements to generate concise summaries.

[0359] Input: Sentiment-analyzed data.

[0360] Output: The generated summary.

[0361] Step 7:

[0362] Server: Uses text mining techniques to extract important keywords and trend information, specifically using TF-IDF and word clouds to identify frequently occurring words.

[0363] Input: Sentiment-analyzed data.

[0364] Output: Extracted keywords and trend information.

[0365] Step 8:

[0366] Server: The generated summary is combined with keywords and trend information to create a final report, which is formatted in Markdown or HTML.

[0367] Input: Abstract, keywords, trend information.

[0368] Output: Final report in Markdown or HTML format.

[0369] Step 9:

[0370] Server: Customize the final report based on user settings and generate a distribution list to distribute it to each department. The distribution method can be email using the SMTP protocol or via REST API to the company's notification system.

[0371] Inputs: Final report, distribution list.

[0372] Output: Reports distributed to each department.

[0373] Step 10:

[0374] Terminal: Saves the received reports and displays them in a user interface, specifically in HTML format using the Report Viewer.

[0375] Input: The report received.

[0376] Output: The report displayed in the user interface.

[0377] Step 11:

[0378] Users can review the displayed report and share it with other employees or departments as needed. They can share the link via email or internal chat tools, or download it as a PDF and distribute it.

[0379] Input: The report displayed in the user interface.

[0380] Output: The shared report.

[0381] (Application example 2)

[0382] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0383] There is a need to efficiently manage factory daily work logs and equipment status, and to quickly detect anomalies and provide information on future trends. However, current methods require manual analysis of huge amounts of data, which is time-consuming and labor-intensive, and the accuracy of the analysis results is unstable. Furthermore, real-time monitoring and anomaly detection are difficult, making it difficult to respond quickly when problems occur.

[0384] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0385] In this invention, the server includes means for analyzing uploaded documents using natural language processing technology, means for using a generative model to generate summaries based on the analysis results, means for providing analysis results of equipment status and daily work reports to robots operating in the factory, means for the robots to monitor the equipment status in real time and detect anomalies, and means for distributing the generated summaries and extracted trends and keywords to terminals in each department.This enables automatic and efficient data analysis and anomaly detection, enabling rapid troubleshooting.

[0386] "Natural language processing technology" is a technology that allows computers to understand, analyze, and generate human language, and includes document tokenization, part-of-speech analysis, sentiment analysis, and more.

[0387] A "generative model" is a machine learning algorithm that learns patterns in knowledge and data and generates new data.

[0388] "Equipment status" is information that indicates the current operating state of machines and devices in a factory, and includes parameters such as temperature, vibration, and operating time.

[0389] A "daily work report" is a report that details the work performed and events that occurred on a specific work day.

[0390] "Real-time monitoring" refers to continuously monitoring the status of equipment and systems within a factory and obtaining that information instantly.

[0391] "Anomaly detection" is the process of identifying unusual equipment states or behaviors, helping to detect problems early.

[0392] "Trends and keywords" are important terms that appear frequently in documents or data, or information that indicates patterns or trends over time.

[0393] "Customized Format" means a data presentation and layout that is specifically designed and tailored based on the needs and requirements of the User.

[0394] This invention is a system that analyzes daily work reports and equipment status in a factory and distributes the results as summaries and trend information to each department. This system utilizes natural language processing technology and generative models to enable fast and accurate data analysis and anomaly detection.

[0395] Hardware and software used

[0396] To implement the present invention, the following hardware and software are used:

[0397] Hardware: Factory robots, servers

[0398] Software: Python, TensorFlow, Hugging Face Transformers, FastAPI, MySQL

[0399] Program processing overview

[0400] 1. Uploading and Receiving Files

[0401] User: Factory workers and managers upload daily factory work reports through the robot's dedicated interface.

[0402] Terminal: Validates the file through the provided interface and sends it to the server using the HTTPS protocol, during which process it is verified that the file has been uploaded correctly.

[0403] 2. Data Reception and Analysis

[0404] Server: After receiving the uploaded daily report data, the server uses a natural language processing engine (Hugging Face Transformers) to tokenize and analyze the parts of speech of the daily report content, and then uses a sentiment analysis engine to analyze emotions such as positive, negative, and neutral.

[0405] 3. Summary Generation and Trend Extraction

[0406] Server: The parsed data is passed to a generative model (TensorFlow) to generate summaries, and then text mining techniques (TF-IDF) are used to extract important keywords and trends.

[0407] 4. Condition Monitoring and Anomaly Detection

[0408] Robot: Monitors the status of equipment in the factory in real time and compares it with analyzed daily data. If an abnormality is detected, the information is quickly sent to a server and notified to each department along with trend information.

[0409] 5. Report distribution

[0410] Server: Integrates the generated summary and trend information, generates reports in a format suitable for each department's terminal, and immediately distributes the reports to each department, where they are acknowledged for receipt.

[0411] Specific examples

[0412] Example 1: Suppose a factory daily report contains the following statement: "Today's work status of machine A is reported. No abnormalities." When the system analyzes this daily report, it extracts the keywords "machine A," "work status," and "no abnormalities," and generates a summary statement: "The work status of machine A is normal." This is then distributed to each department along with related trend information.

[0413] Example prompt sentence:

[0414] "Analyze the following daily report for key keywords, summaries, and sentiment analysis:

[0415] I'll report on the work status of Machine A today. No abnormalities."

[0416] This invention enables automatic and efficient data analysis and anomaly detection in factories, enabling rapid troubleshooting. It also enables each department to make efficient decisions based on the generated summary and trend information.

[0417] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0418] Step 1:

[0419] Uploading and Receiving Files

[0420] The user uploads the factory's daily work report through the robot's dedicated interface, selects a specific file, and presses the send button.

[0421] Input: Factory work daily report file

[0422] Output: Uploaded files are sent to the device.

[0423] Specific operation: Use the file selection dialog on the user interface to select a file and click the Send button. The device will send the file to the server using the HTTPS protocol.

[0424] Step 2:

[0425] Data reception and analysis

[0426] The server receives the uploaded daily report data and uses a natural language processing engine (Hugging Face Transformers) to tokenize and analyze the parts of speech of the daily report content.

[0427] Input: Uploaded daily report data

[0428] Output: Tokenized and part-of-speech analyzed data

[0429] Specific operation: The server decodes the received data into UTF-8 and converts it into text format. This is then passed to a natural language processing engine, which tokenizes the document and analyzes the parts of speech.

[0430] Step 3:

[0431] Emotion analysis

[0432] The server passes the tokenized data to a sentiment analysis engine, which analyzes the sentiment for positive, negative, neutral, etc.

[0433] Input: Tokenized and part-of-speech analyzed data

[0434] Output: Emotion analysis results

[0435] How it works: The parsed data is fed into a sentiment analysis engine, which calculates a sentiment score for each sentence or phrase. Based on this, the sentiment of the entire document is evaluated.

[0436] Step 4:

[0437] Summary Generation

[0438] The server uses a generative model (TensorFlow) to generate a summary based on the analysis results.

[0439] Input: Sentiment analysis results and tokenized data

[0440] Output: Generated summary

[0441] How it works: The sentiment analysis results are combined with the tokenized data and passed to a generative model, which extracts key information and generates a summary.

[0442] Step 5:

[0443] Trend and Keyword Extraction

[0444] The server uses text mining technology (TF-IDF) to extract important keywords and trends.

[0445] Input: Tokenized and part-of-speech analyzed data

[0446] Output: Extracted keywords and trend information

[0447] Specific operation: Text data is input into the TF-IDF algorithm to calculate and extract important keywords and trends that frequently appear in the document.

[0448] Step 6:

[0449] Condition Monitoring and Anomaly Detection

[0450] The robot monitors the status of equipment in the factory in real time and detects abnormalities by comparing it with analyzed daily data.

[0451] Input: Real-time monitoring data and daily report analysis results

[0452] Output: Anomaly detection notification

[0453] Specific operation: The monitored data is analyzed in real time and compared with the saved daily data. If an abnormal value is detected, an alert is sent to the server.

[0454] Step 7:

[0455] Report Distribution

[0456] The server integrates the generated summary and trend information, generates and distributes reports in a format suitable for each department's terminal.

[0457] Input: Generated abstracts, keywords and trend information

[0458] Output: Reports delivered to departments

[0459] Specific operation: The generated summaries, keywords, and trend information are formatted and a report is created in a format that matches the settings of each department. This is then distributed to each department's terminal and receipt is confirmed.

[0460] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0461] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0462] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0463] [Second embodiment]

[0464] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0465] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0466] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0467] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0468] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0469] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0470] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0471] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0472] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0473] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0474] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0475] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0476] Overall system overview

[0477] This system analyzes business daily reports and information uploaded by users on a server, automatically generates summaries and trend information, and distributes them to each department, thereby streamlining information sharing. Specifically, it combines natural language processing technology, generative models, and keyword and trend extraction algorithms to achieve fast and accurate information analysis and distribution.

[0478] Program processing flow and detailed explanation

[0479] Uploading and Receiving Files

[0480] User: Upload daily reports and business information files through the device interface.

[0481] On your device: Automatically validates the file format and size to ensure it is safe before sending it to the server.

[0482] Calling the API for receiving data

[0483] Terminal: Call the API endpoint for sending a file and send the file contents to the server.

[0484] Server: Stores the received file in temporary storage and then passes it to the natural language processing module.

[0485] Analysis using natural language processing

[0486] Server: Using a natural language processing engine, the uploaded document is broken down into sentences and words, tokenized, and analyzed for parts of speech to determine the meaning and relationships of each word.

[0487] Server: Prepares the analysis results in a data frame format and passes them to the summary generation module.

[0488] Generate a summary

[0489] Server: Uses a generative model to generate a summary from the analysis results, evaluating the importance of each sentence to ensure that important information is not omitted.

[0490] Server: Formats the generated summary and presents it in a format that is easy for users to understand.

[0491] Keyword and trend extraction

[0492] Server: Uses text mining technology to extract important keywords and trends from the analysis results.

[0493] Server: Adds extracted keywords and trends to a summary and creates the final report.

[0494] Report distribution

[0495] Server: Prepares the generated reports for distribution to each department, customizing the format based on user settings.

[0496] Server: Distributes reports to each department's terminal and confirms receipt.

[0497] View and share your results

[0498] Terminal: Displays received reports to users, allowing each department to quickly access the information they need.

[0499] Users: Review reports and share them with other employees or departments as needed, supporting efficient information sharing and quick decision-making.

[0500] Specific examples

[0501] Sales department daily report analysis and summary

[0502] 1. User: Upload the sales report (e.g. sales_report_2023_10_15.pdf) to the system from the terminal.

[0503] 2. Terminal: Validate the file and send it to the server.

[0504] 3. Server: Receives the daily reports and performs analysis using a natural language processing engine.

[0505] 4. Server: Based on the analysis results, a generative model creates a summary, summarizing key selling points and customer trends.

[0506] 5. Server: Delivers reports containing summary and trend information to sales department terminals.

[0507] 6. Terminal: Receives the report and displays it for the sales team to view.

[0508] 7. User: Review the reports and make decisions that will influence your sales strategy.

[0509] This system allows users to efficiently analyze the large amount of information generated in daily business operations and use it to make quick decisions.

[0510] The processing flow will be explained below.

[0511] Step 1:

[0512] User: Uploads daily reports and business information files using the device interface. Through the file selection dialog, select the specific file and press the send button.

[0513] Step 2:

[0514] Terminal: Validates the selected file, checking that the file format is supported and that the size is within the regulated range. If there are no problems, generates an API request to send the file to the server.

[0515] Step 3:

[0516] Terminal: Sends the generated API request to the server's API endpoint, usually using a secure protocol such as HTTPS.

[0517] Step 4:

[0518] Server: Received files are stored in temporary storage, where file metadata and other information are recorded and prepared for subsequent processing.

[0519] Step 5:

[0520] Server: Calls the natural language processing (NLP) engine to analyze the contents of the uploaded file. First, it breaks down the file into sentences and words, and performs tokenization and part-of-speech analysis.

[0521] Step 6:

[0522] Server: Stores the results of tokenization and part-of-speech analysis. Specifically, it stores the analysis results in data frame format and passes them to the next summary generation process.

[0523] Step 7:

[0524] Server: Uses a generative model for summary generation to generate a summary from the analysis results. Extracts important points from the document and summarizes the information.

[0525] Step 8:

[0526] Server: Evaluates the generated summary and formats it in an appropriate format. Here, the summary is organized into a concise and easy-to-read format so that users can quickly grasp the information.

[0527] Step 9:

[0528] Server: Along with the summary, we use text mining techniques to extract important keywords and trends. We use TF-IDF and word clouds to extract highly relevant information.

[0529] Step 10:

[0530] Server: Adds the extracted results to the summary and creates a final report, which includes the summary, keywords, and trend information.

[0531] Step 11:

[0532] Server: Prepares the created report for distribution to each department, customizes the format according to the user's settings, and generates a distribution list.

[0533] Step 12:

[0534] Server: Delivers reports to terminals in each department. Sends reports to designated recipients using a distribution protocol.

[0535] Step 13:

[0536] Terminal: Stores received reports and displays them in the user interface, allowing users to easily view and review the reports.

[0537] Step 14:

[0538] Users: Check the displayed report and share it with other employees or departments as needed, including by email or by sending a link to an internal chat tool.

[0539] This series of steps allows users to quickly and accurately analyze business information, obtain summary and trend information, and make effective decisions.

[0540] Example 1

[0541] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0542] In today's business environment, a huge amount of information is generated every day, creating a need to analyze and summarize that information quickly and accurately, and share it efficiently. However, traditional methods require a significant amount of time and effort to analyze and share information. Furthermore, manual information analysis is prone to errors and carries a high risk of important information being overlooked. Furthermore, the wide variety of information formats makes it difficult to analyze information in a unified and efficient manner. There is a need for a system that solves these problems.

[0543] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0544] In this invention, the server includes: means for a user to upload documents to the system through an electronic device interface; means for a terminal to verify the format and size of the uploaded documents and transmit them to the server after confirming their safety; means for the server to call a data reception API to receive and temporarily store the files; means for the server to use natural language processing technology to break down the uploaded documents into sentence units and words and perform tokenization and part-of-speech analysis; means for the server to generate the analysis results in a data frame format and pass them to a summary generation model; means for the server to use the generation model to generate summaries from the analysis results; means for the server to extract important keywords and trends and add them to the summaries to create a final report; means for the server to distribute the generated summaries and the extracted trends and keywords to terminals in each department; and means for the terminal to display the received reports to the user and enable information to be shared with other departments, thereby enabling efficient analysis of information and rapid sharing.

[0545] "User" refers to the entity that uses the system to upload materials and receive analysis results.

[0546] "Electronic Device" means any device that allows a User to manipulate and upload materials to the System, including, for example, a computer, smartphone, or tablet.

[0547] An "interface" is a point of contact between a user and a system for exchanging information. Examples include web browsers and dedicated applications.

[0548] "Documents" refers to business daily reports and other written information uploaded to the system, including, for example, PDF, Excel, Word files, etc.

[0549] "Terminal" means an electronic device operated by a User that verifies the format and size of uploaded materials and ensures their security.

[0550] "Server" refers to the computer system that receives uploaded materials, generates analysis and summaries, and distributes them to each department.

[0551] "Data Reception API" refers to a program interface for sending materials from a terminal to a server.

[0552] "Temporary storage" refers to the temporary storage of received materials by a server, including, for example, cloud storage and databases.

[0553] "Natural language processing technology" is a technology for analyzing uploaded documents, and is primarily a function for tokenization and part-of-speech analysis.

[0554] "Tokenization" refers to the process of breaking down a document into words or sentences.

[0555] "Part-of-speech analysis" refers to the process of identifying the part of speech of each word in a document and analyzing its grammatical relationships.

[0556] "Data frame format" refers to a data structure for storing analysis results in a table-like format.

[0557] A "summary generation model" refers to an algorithm or machine learning model for generating a summary based on the analysis results.

[0558] "Keywords" refer to words that appear frequently in a document and have important meanings.

[0559] "Trends" refers to major themes or tendencies extracted from the information in a document.

[0560] The "final report" refers to a document that compiles the generated summary, keywords, and trend information.

[0561] "Delivery" refers to the process of sending the generated final report to the terminals of each department.

[0562] "Receiving" refers to the terminal receiving distribution from the server.

[0563] "Display" refers to the terminal visually indicating to the user the report received.

[0564] "Sharing" refers to the act of a user collaborating and communicating the received report with other employees or departments.

[0565] MODE FOR CARRYING OUT THE INVENTION

[0566] This invention is a system that improves the efficiency of information sharing by analyzing business daily reports and information uploaded by users on a server, automatically generating summaries and trend information, and distributing them to each department. The system includes the following main elements:

[0567] Uploading and Receiving Files

[0568] Users upload files such as daily reports and business information through the interface of their electronic device. Specifically, they use the browser's file selection dialog to select a document such as "sales_report_2023_10_15.pdf" and click the upload button. The device then verifies the file's format and size (e.g., PDF, Excel, Word), confirms its security, and sends it to the server.

[0569] Calling the API for receiving data

[0570] The device calls an API endpoint (e.g., https: / / api.example.com / upload) to upload the file to the server. The server temporarily stores the file received at the API endpoint (e.g., Amazon S3 or Google Cloud Storage) and passes the path to the natural language processing engine.

[0571] Analysis using natural language processing

[0572] The server uses a natural language processing (NLP) engine (e.g., SpaCy, NLTK) to break down the uploaded document into sentences and words. Specifically, it performs tokenization, part-of-speech analysis, and removes stop words. The analysis results are formatted as a data frame (e.g., Pandas DataFrame) and passed to the summary generation model.

[0573] Generate a summary

[0574] The server uses a generative AI model (e.g., GPT-3, BERT) to generate a summary from the analysis results. The summary is generated by extracting particularly important sentences based on their importance scores and assembling them into a sentence. The generated summary is then organized in a format that is easy for users to understand (e.g., Markdown or HTML).

[0575] Keyword and trend extraction

[0576] The server uses text mining techniques (e.g., TF-IDF, Word2Vec) to extract important keywords and trends from the analysis results, including frequency analysis and semantic relevance assessment. The extracted keywords and trend information are added to a summary to create the final report.

[0577] Report distribution

[0578] The server prepares the generated reports for distribution to each department, customizing the reports based on user settings and departmental formatting requirements, and delivering the reports via email or the company's internal portal, where receipt is acknowledged.

[0579] View and share your results

[0580] The device displays the received report to the user, for example, through a web interface or desktop notification. The user can review the report's contents and share the link in an internal chat tool to share key selling points with other departments.

[0581] Prompt Sentence Examples

[0582] Below are some example input prompts for the generative AI model that can be used in this system:

[0583] "Today, we uploaded the sales report for October 15, 2023. Please extract trending keywords and summarize them. Don't forget to include your key selling points."

[0584] This system allows users to efficiently analyze the large amount of information generated in daily business operations and use it to make quick decisions.

[0585] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0586] Step 1:

[0587] The user uploads the document using an electronic device. The user opens a file selection dialog using the interface of their device, selects the daily report or business information file (e.g., "sales_report_2023_10_15.pdf"), and clicks the upload button. The input is the file selected by the user, and the output is the file selection information sent to the device.

[0588] Step 2:

[0589] The terminal verifies the format and size of the uploaded file to confirm safety. The terminal checks the file format (e.g. PDF, Excel, Word) and whether the file size exceeds the specified limit. It also performs a virus check. The input is the file uploaded by the user, and the output is the verification result (safety confirmation).

[0590] Step 3:

[0591] The device sends the verified file to the server. The device calls an API endpoint (e.g., https: / / api.example.com / upload) to upload the verified file to the server via an HTTP POST request. The input is the verified file, and the output is a notification that the file was successfully uploaded.

[0592] Step 4:

[0593] The server temporarily stores the received file. The server stores the file received at the API endpoint in temporary storage (e.g., Amazon S3 or Google Cloud Storage) and records the path. The input is the file sent from the device, and the output is the path to the temporary storage.

[0594] Step 5:

[0595] The server analyzes the file using a natural language processing (NLP) engine. The server passes the saved file to an NLP engine (e.g., SpaCy, NLTK) and breaks it down into sentences and words. It performs tokenization, part-of-speech analysis, and removes stop words, and then organizes the analysis results in data frame format (e.g., Pandas DataFrame). The input is the path to the saved file, and the output is a data frame of the analysis results.

[0596] Step 6:

[0597] The server generates a summary using a generative AI model. After tokenization and part-of-speech analysis, the server passes the data frame to a generative AI model (e.g., GPT-3, BERT) to generate a summary. It extracts particularly important sentences based on their importance scores and assembles them into a summary. The input is the analysis result data frame, and the output is the generated summary.

[0598] Step 7:

[0599] The server uses text mining techniques to extract keywords and trends. The server uses TF-IDF models and Word2Vec to extract important keywords and trends from the analysis results. This includes word frequency analysis and semantic relevance assessment. The input is a data frame of the analysis results, and the output is the extracted keywords and trends.

[0600] Step 8:

[0601] The server creates a final report and adds keywords and trends to the summary. The server adds keyword and trend information to the generated summary and formats the final report. The report is prepared in a format that is easy for users to understand (e.g., Markdown or HTML). The input is the generated summary and extracted keywords and trends, and the output is the final report.

[0602] Step 9:

[0603] The server prepares the final report for delivery to each department. The server customizes the report based on user settings and departmental formatting requirements. The server delivers the report via email or through the company's internal portal. The input is the final report and the output is the customized report.

[0604] Step 10:

[0605] The server distributes the report and the terminal confirms receipt. The server distributes the customized report to each department's terminal and receives a receipt confirmation. The input is the customized report and the output is the receipt confirmation result.

[0606] Step 11:

[0607] The terminal displays the received reports to the user. The terminal makes the reports available to the user through a web interface or desktop notification. The input is the received report and the output is the report displayed to the user.

[0608] Step 12:

[0609] The user reviews the report and shares it with other departments. The user reviews the received report and shares important selling points and analysis results with other employees and departments using an internal chat tool, etc. The input is the report displayed to the user, and the output is the results shared with other departments.

[0610] This is the flow of processing for this system's program, which enables efficient analysis and rapid sharing of information.

[0611] (Application example 1)

[0612] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0613] Conventional autonomous driving systems have had difficulty efficiently analyzing massive amounts of sensor data and operational information, summarizing it, and providing it to drivers and managers. This can lead to delays in quickly identifying important trends and operational issues and taking appropriate measures. Furthermore, there is no established method for displaying or distributing the generated data, which can lead to delays in information sharing and decision-making.

[0614] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0615] In this invention, the server includes means for analyzing uploaded documents using natural language processing technology, means for using a generative model to generate summaries based on the analysis results, means for distributing the generated summaries and extracted trends and keywords to terminals in each department, means for uploading sensor data and operation information and analyzing and summarizing them, and means for extracting important information from the operation data of the autonomous vehicle and displaying it to the driver or manager. This allows the operation information of the autonomous vehicle to be analyzed efficiently and quickly, summarized, and provided, making it possible to immediately grasp important trends and issues and respond promptly.

[0616] "Natural language processing technology" is a technology that automatically analyzes and understands text data and processes human language using a computer.

[0617] A "generative model" is a machine learning algorithm that generates new data based on given data.

[0618] A "trend" refers to the tendency of fluctuations or patterns in data or information over a period of time.

[0619] "Keywords" refer to words or phrases that have particular significance in documents or data.

[0620] "Sensor data" refers to measurement data such as temperature, pressure, position, and speed obtained from various sensors.

[0621] "Operation information" is data relating to the operation status of a vehicle, including location information, speed, route, traffic conditions, and the like.

[0622] An "autonomous vehicle" refers to a vehicle that drives autonomously without the need for human operation.

[0623] An "abstract" is a short sentence or text that succinctly summarizes the important points of the original document or data.

[0624] An "API endpoint" refers to a specific address or operation that allows software to communicate with other software.

[0625] "Customized format" means data or reports that have been formatted in a specific way according to the user's requirements or purposes.

[0626] A "prompt sentence" is text containing instructions or questions to be provided as input to a generative model.

[0627] The system for implementing this invention analyzes autonomous vehicle operation data and sensor data uploaded by users on a server, automatically generates summaries and trend information, and distributes them to drivers and managers. A detailed description of the system is as follows.

[0628] Program processing overview

[0629] Uploading and Receiving Files

[0630] Users upload operational data for autonomous vehicles via their smartphones or on-board displays. The device automatically verifies the file format and size, confirms that it is safe, and then sends it to the server.

[0631] Calling the API for receiving data

[0632] The device calls the API endpoint for sending a file and sends the file contents to the server, which then stores the received file in temporary storage and passes it to the natural language processing module.

[0633] Analysis using natural language processing

[0634] The server uses a natural language processing engine (e.g., spaCy or NLTK) to break down the uploaded operation logs and error reports into sentences and words. It then performs tokenization and part-of-speech analysis to analyze the meaning and relationships of each word. The analysis results are prepared in data frame format and passed to a summary generation model (e.g., GPT series).

[0635] Generate a summary

[0636] The server uses a generative model to generate a summary from the analysis results. During this process, it evaluates the importance of each sentence to ensure that important information is not omitted. The generated summary is then formatted and presented in a format that is easy for users to understand.

[0637] Keyword and trend extraction

[0638] The server uses text mining techniques to extract important keywords and trends from the analysis results, adds them to a summary, and creates a final report.

[0639] Report distribution

[0640] The server prepares the generated report for distribution to the driver or manager, customizes the report format based on the user's settings, and delivers the report to a smartphone or in-vehicle display for confirmation of receipt.

[0641] View and share your results

[0642] The terminal displays the received reports to the user, allowing them to quickly access the information they need. Users can then review the reports and share them with other drivers and fleet managers, supporting efficient information sharing and quick decision-making.

[0643] Specific examples

[0644] We will provide a concrete example of uploading log data from a self-driving vehicle's daily operations, analyzing it in real time, and extracting trends and important issues.

[0645] Example prompt for a generative AI model:

[0646] "Please extract important information from the operational data log below and generate a written summary."

[0647] Data Log:

[0648] This system will enable drivers and managers of autonomous vehicles to efficiently analyze and summarize data, enabling them to make quick decisions.

[0649] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0650] Step 1:

[0651] Uploading and Receiving Files

[0652] Users use their smartphones or in-vehicle displays to upload operational and sensor data from their autonomous vehicles. The device automatically validates the file format and size to ensure the content is safe. The input data is an operational data file, and the output is a file ready to be sent to the server. Specifically, the device checks the file's header information to ensure it is in an acceptable format and size.

[0653] Step 2:

[0654] Calling the API for receiving data

[0655] The terminal calls the API endpoint for file transmission and sends the contents of the operation data file to the server. The input data is the verified operation data file, and the output is a file stored in the server's temporary storage. Specifically, the terminal generates an API request, includes the file data in the request body, and sends it to the server. The server stores the received data in temporary storage.

[0656] Step 3:

[0657] Analysis using natural language processing

[0658] The server uses a natural language processing engine (e.g., spaCy or NLTK) to analyze the received operation data file. The input data is the operation data file stored in temporary storage, and the output is a data frame that has been tokenized and analyzed for parts of speech. Specifically, the server reads the file, tokenizes it into sentences and words, analyzes the parts of speech for each, and converts it into a data frame format.

[0659] Step 4:

[0660] Generate a summary

[0661] The server generates a summary using a generative model (e.g., GPT series) based on the analysis results. The input data is a data frame of the analysis results, and the output is a summary. Specifically, the server passes the data frame as input to the generative model, and generates a summary while evaluating important information.

[0662] Step 5:

[0663] Keyword and trend extraction

[0664] The server uses text mining technology to extract important keywords and trends from the analysis results. The input data is a summary sentence and a data frame of the analysis results, and the output is important keywords and trend information. Specifically, the server applies text mining technology to detect frequently occurring keywords and trend patterns.

[0665] Step 6:

[0666] Report creation and distribution

[0667] The server adds the extracted keywords and trends to the summary and creates a final report. The input data is the summary, keywords, and trend information, and the output is a completed report. Specifically, the server integrates the keywords and trend information into the summary and generates a report in a customized format based on user settings. The final report is then distributed to the driver's or manager's device.

[0668] Step 7:

[0669] View and share your results

[0670] The terminal displays the received report to the user. The input data is the delivered report, and the output is the information displayed to the user. Specifically, the terminal analyzes the report data and displays it in a visually easy-to-understand format. The user can review the report and share it with other drivers or fleet managers.

[0671] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0672] Overall system overview

[0673] This system analyzes daily business reports and information uploaded by users on a server, automatically generates summaries and trend information, and distributes them to each department. Furthermore, by combining it with an emotion engine, it is possible to analyze the sentiment of documents and reflect this in summaries and trend information. This allows users to analyze information while taking into account the emotional aspects of documents, enabling them to make quick and accurate decisions.

[0674] Program processing flow and detailed explanation

[0675] Uploading and Receiving Files

[0676] User: Uploads daily reports and business information files through the device interface, selects a specific file through the file selection dialog, and presses the send button.

[0677] Calling the API for receiving data

[0678] Terminal: Call the API endpoint for sending the file and send the file contents to the server. The sending protocol is usually a secure one such as HTTPS.

[0679] Analysis using natural language processing

[0680] Server: Calls the natural language processing (NLP) engine to analyze the contents of the uploaded file. First, it breaks the file down into sentences and words, then performs tokenization and part-of-speech analysis.

[0681] Server: Stores the results of tokenization and part-of-speech analysis. Specifically, the analysis results are stored in data frame format and passed to the next summary generation process.

[0682] Emotion analysis

[0683] Server: Calls the sentiment engine to analyze the sentiment of the uploaded document, identifying positive and negative phrases and words in the text and assessing the type and intensity of the sentiment.

[0684] Server: Integrates the analysis results into a summary generation model to generate summaries that include emotional information.

[0685] Generate a summary

[0686] Server: Generates a summary from the analysis results using a generative model, extracting important points from the document and summarizing the information.

[0687] Server: Evaluates the generated summary and formats it appropriately. Here, it is formatted in a concise and easy-to-read format to allow users to quickly grasp the information.

[0688] Keyword and trend extraction

[0689] Server: Uses text mining techniques to extract important keywords and trends from the analysis results, and uses TF-IDF and word clouds to extract highly relevant information.

[0690] Server: Adds extracted keywords and trends to a summary and creates a final report.

[0691] Report distribution

[0692] Server: Prepares the generated reports for distribution to each department, customizes the format according to the user's settings, and generates the distribution list.

[0693] Server: Distributes reports to each department's terminal and confirms receipt.

[0694] View and share your results

[0695] Terminal: Stores the received reports and displays them in the user interface, allowing users to easily view and check the reports.

[0696] Users: Review the displayed report and share it with other employees or departments as needed, including by email or by sending a link to an internal chat tool.

[0697] Specific examples

[0698] Sales department daily report analysis and summary

[0699] 1. User: Upload the sales report (e.g. sales_report_2023_10_15.pdf) to the system from the terminal.

[0700] 2. Terminal: Validate the file and send it to the server.

[0701] 3. Server: Receives the daily reports and performs analysis using a natural language processing engine and sentiment engine.

[0702] 4. Server: Based on the analysis results, a generative model creates a summary, summarizing key selling points, customer trends, and sentiment information.

[0703] 5. Server: Delivers reports containing summaries, trend information, and sentiment information to sales department terminals.

[0704] 6. Terminal: Receives the report and displays it for the sales team to view.

[0705] 7. User: Review the reports and make decisions that will influence your sales strategy.

[0706] This system enables users to efficiently analyze the large amount of information generated in daily business operations, quickly grasp summary and trend information, as well as sentiment information, and make effective decisions.

[0707] The processing flow will be explained below.

[0708] Step 1:

[0709] User: Uploads daily reports and business information files using the device interface. Through the file selection dialog, select the specific file and press the send button.

[0710] Step 2:

[0711] Terminal: Validates the selected file, checking whether the file format is supported and whether the size is within the regulated range. If there are no problems, generates an API request to send the file to the server.

[0712] Step 3:

[0713] Terminal: Sends the generated API request to the server's API endpoint, usually using a secure protocol such as HTTPS.

[0714] Step 4:

[0715] Server: Received files are stored in temporary storage, where file metadata and other information are recorded and prepared for subsequent processing.

[0716] Step 5:

[0717] Server: Calls the natural language processing (NLP) engine to analyze the contents of the uploaded file. First, it breaks the file down into sentences and words, then performs tokenization and part-of-speech analysis.

[0718] Step 6:

[0719] Server: Stores the results of tokenization and part-of-speech analysis. Specifically, it stores the analysis results in data frame format and passes them to the next summary generation process.

[0720] Step 7:

[0721] Server: Calls the sentiment engine to analyze the sentiment of the uploaded document, identifying positive and negative phrases and words in the text and assessing the type and intensity of the sentiment.

[0722] Step 8:

[0723] Server: Stores the analyzed emotion data and integrates it into the summary generation model, thereby generating summaries that include emotion information.

[0724] Step 9:

[0725] Server: Using a generative model for summary generation, a summary is generated based on the analysis results and sentiment data. Important points of the document are extracted and the information is summarized.

[0726] Step 10:

[0727] Server: Evaluates the generated summary and formats it in an appropriate format. Here, it is organized into a concise and easy-to-read format so that users can quickly grasp the information.

[0728] Step 11:

[0729] Server: Uses text mining techniques to extract important keywords and trends from the analysis results. Uses TF-IDF and word clouds to extract highly relevant information.

[0730] Step 12:

[0731] Server: Adds the extracted keywords and trends to a summary and creates a final report, including sentiment information.

[0732] Step 13:

[0733] Server: Prepares the created report for distribution to each department, customizes the format according to the user's settings, and generates a distribution list.

[0734] Step 14:

[0735] Server: Delivers reports to terminals in each department. Sends reports to designated recipients using a distribution protocol.

[0736] Step 15:

[0737] Terminal: Stores received reports and displays them in the user interface, allowing users to easily view and check the reports.

[0738] Step 16:

[0739] Users: Check the displayed report and share it with other employees or departments as needed. Sharing methods include email and internal chat tool links.

[0740] This detailed processing step enables users to efficiently analyze the large amount of information generated in daily business operations, quickly grasp summary and trend information, as well as sentiment information, and make effective decisions.

[0741] Example 2

[0742] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0743] In today's business environment, a huge amount of business documents and reports are generated every day, and there is a need to efficiently analyze them and quickly grasp important information. Conventional methods require manually analyzing document content and creating summaries, which is time-consuming and labor-intensive. Furthermore, it is difficult to appropriately evaluate and reflect the emotional information contained in documents. This leads to issues such as overlooking information and delays in identifying important trends.

[0744] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: a means for a user to upload a file through an interface; a means for a terminal to call an API endpoint and send the contents of the file; a means for the server to analyze the uploaded document using natural language processing technology; a means for analyzing the sentiment of the document using an emotion engine; a means for using a generative model to generate a summary based on the analysis results; a means for distributing the generated summary and extracted trends and keywords to terminals in each department; and a means for saving and displaying reports received by the terminals. This makes it possible to efficiently analyze the content and sentiment information of business documents and to quickly and accurately generate and distribute summaries and trend information.

[0745] "User" refers to the user who uploads business documents and daily reports through the system interface.

[0746] A "terminal" is a device used by a user that has the functionality to upload files and send requests to API endpoints.

[0747] "Server" refers to a central computing device that analyzes received files, generates summaries and trend information, and distributes it to each department.

[0748] "Natural language processing technology" is a technology for analyzing documents, and has the ability to break them down into sentences or words and perform tokenization and part-of-speech analysis.

[0749] An "API endpoint" is a specific URL or URI that a device uses to communicate with a server and is intended to receive HTTPS requests.

[0750] The "Sentiment Engine" is an analytical tool for identifying emotive phrases in a document and assessing positive and negative sentiment.

[0751] A "generative model" is an algorithm or AI model used to generate a summary of a document based on the results of its analysis.

[0752] "Tokenization" is the process of dividing a document into smaller units such as words or phrases.

[0753] "Part-of-speech analysis" is the process of identifying the part of speech (noun, verb, adjective, etc.) for each tokenized word or phrase.

[0754] A "data frame" is a data structure for storing analysis results, and is tabular data represented by rows and columns.

[0755] "Keywords" are words or phrases that are considered to be particularly important within a document.

[0756] "Trend information" refers to significant trends and patterns extracted from analyzed documents.

[0757] "Report" means the final output document containing the generated summary and extracted keyword trend information.

[0758] "Format" refers to the style and structure in which a report or summary is presented.

[0759] "Distribution" is the process of sending the generated report from the server to the terminals of each department.

[0760] "Display" refers to presenting the report received at the terminal on the user interface.

[0761] A "department" is a group responsible for a specific task or function in a company or organization.

[0762] In the system according to the present invention, a server analyzes business documents and daily reports uploaded by users and automatically generates summaries and trend information. Furthermore, an emotion engine can be used to analyze the emotional information of the documents and reflect this in the generated summaries and trend information. The configuration and operation of the system are described in detail below.

[0763] First, the user uploads a business document or daily report (e.g., sales_report_2023_10_15.pdf) using the terminal by selecting the appropriate file from the file selection dialog and clicking the upload button.

[0764] The device then sends the selected file to an API endpoint (e.g., "https: / / example.com / upload") using the HTTPS protocol. This request is sent as multipart form data.

[0765] The server inputs the received file into a natural language processing engine, which tokenizes and analyzes parts of speech at the sentence and word level. The analysis results are stored in a data frame format using the pandas library. The sentiment engine then evaluates the sentiment within the document based on the analysis results. It identifies positive and negative phrases and quantifies the strength of the sentiment.

[0766] The generative AI model generates a summary based on the analysis results and sentiment information. For example, it generates a summary such as, "Sales increased 15% this month, but there were many returns for a specific product group." The generated summary is formatted in Markdown or HTML for easy reading.

[0767] The server then uses text mining techniques (e.g., TF-IDF and word cloud) to extract important keywords and trend information from the analysis results. Keywords such as "sales," "surge," and "returns" are extracted. These keywords and trend information are also incorporated into the summary text to create the final report.

[0768] The server customizes the generated report according to the user's settings and prepares it for distribution to each department. The report is sent to each department's terminal based on the distribution list. This can be done by email using the SMTP protocol or by sending it to an internal notification system via REST API.

[0769] Finally, the terminal saves the received report and displays it in the user interface. The user can then review the displayed report and share it with other employees or departments as needed. This can be done via email or a link to an internal chat tool. A specific example of this is to share a link to the generated report in an internal chat so that everyone can access it.

[0770] In this way, this invention makes it possible to efficiently analyze large volumes of business documents and daily reports, and quickly and accurately generate and distribute summaries, trend information, and sentiment information, enabling users to quickly grasp important business information and make appropriate decisions.

[0771] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0772] Step 1:

[0773] User: Upload the daily report or business information file through the device interface by selecting the appropriate file (e.g., sales_report_2023_10_15.pdf) from the file selection dialog and clicking the upload button.

[0774] Input: File (for example, sales_report_2023_10_15.pdf).

[0775] Output: The file is selected on the device and ready to be uploaded.

[0776] Step 2:

[0777] On the device: After selecting the file, send an HTTPS request to the API endpoint (e.g., "https: / / example.com / upload"), which sends the file to the server as multipart form data.

[0778] Input: The selected file.

[0779] Output: A request to send a file to the server.

[0780] Step 3:

[0781] Server: The received file is input into a natural language processing (NLP) engine, which performs tokenization and part-of-speech analysis on a sentence-by-sentence and word-by-word basis. Specifically, the document is broken down into multiple words and each word is tagged with a part of speech.

[0782] Input: The contents of the uploaded file.

[0783] Output: Tokenized and part-of-speech analyzed data in dataframe format.

[0784] Step 4:

[0785] Server: The analysis results are stored in a data frame format and passed to the emotion engine. Here, the analysis results are converted into a data frame using the pandas library.

[0786] Input: Tokenized and part-of-speech analyzed data.

[0787] Output: Analysis results in data frame format.

[0788] Step 5:

[0789] Server: Uses a sentiment engine to assess the sentiment in the document, identifying positive and negative phrases and quantifying the strength of each sentiment.

[0790] Input: Analysis results in data frame format.

[0791] Output: Sentiment-analyzed data with sentiment scores.

[0792] Step 6:

[0793] Server: Generates summaries using generative AI models that integrate emotional information. They combine key document points and emotional elements to generate concise summaries.

[0794] Input: Sentiment-analyzed data.

[0795] Output: The generated summary.

[0796] Step 7:

[0797] Server: Uses text mining techniques to extract important keywords and trend information, specifically using TF-IDF and word clouds to identify frequently occurring words.

[0798] Input: Sentiment-analyzed data.

[0799] Output: Extracted keywords and trend information.

[0800] Step 8:

[0801] Server: The generated summary is combined with keywords and trend information to create a final report, which is formatted in Markdown or HTML.

[0802] Input: Abstract, keywords, trend information.

[0803] Output: Final report in Markdown or HTML format.

[0804] Step 9:

[0805] Server: Customize the final report based on user settings and generate a distribution list to distribute it to each department. The distribution method can be email using the SMTP protocol or via REST API to the company's notification system.

[0806] Inputs: Final report, distribution list.

[0807] Output: Reports distributed to each department.

[0808] Step 10:

[0809] Terminal: Saves the received reports and displays them in a user interface, specifically in HTML format using the Report Viewer.

[0810] Input: The report received.

[0811] Output: The report displayed in the user interface.

[0812] Step 11:

[0813] Users can review the displayed report and share it with other employees or departments as needed. They can share the link via email or internal chat tools, or download it as a PDF and distribute it.

[0814] Input: The report displayed in the user interface.

[0815] Output: The shared report.

[0816] (Application example 2)

[0817] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0818] There is a need to efficiently manage factory daily work logs and equipment status, and to quickly detect anomalies and provide information on future trends. However, current methods require manual analysis of huge amounts of data, which is time-consuming and labor-intensive, and the accuracy of the analysis results is unstable. Furthermore, real-time monitoring and anomaly detection are difficult, making it difficult to respond quickly when problems occur.

[0819] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0820] In this invention, the server includes means for analyzing uploaded documents using natural language processing technology, means for using a generative model to generate summaries based on the analysis results, means for providing analysis results of equipment status and daily work reports to robots operating in the factory, means for the robots to monitor the equipment status in real time and detect anomalies, and means for distributing the generated summaries and extracted trends and keywords to terminals in each department.This enables automatic and efficient data analysis and anomaly detection, enabling rapid troubleshooting.

[0821] "Natural language processing technology" is a technology that allows computers to understand, analyze, and generate human language, and includes document tokenization, part-of-speech analysis, sentiment analysis, and more.

[0822] A "generative model" is a machine learning algorithm that learns patterns in knowledge and data and generates new data.

[0823] "Equipment status" is information that indicates the current operating state of machines and devices in a factory, and includes parameters such as temperature, vibration, and operating time.

[0824] A "daily work report" is a report that details the work performed and events that occurred on a specific work day.

[0825] "Real-time monitoring" refers to continuously monitoring the status of equipment and systems within a factory and obtaining that information instantly.

[0826] "Anomaly detection" is the process of identifying unusual equipment states or behaviors, helping to detect problems early.

[0827] "Trends and keywords" are important terms that appear frequently in documents or data, or information that indicates patterns or trends over time.

[0828] "Customized Format" means a data presentation and layout that is specifically designed and tailored based on the needs and requirements of the User.

[0829] This invention is a system that analyzes daily work reports and equipment status in a factory and distributes the results as summaries and trend information to each department. This system utilizes natural language processing technology and generative models to enable fast and accurate data analysis and anomaly detection.

[0830] Hardware and software used

[0831] To implement the present invention, the following hardware and software are used:

[0832] Hardware: Factory robots, servers

[0833] Software: Python, TensorFlow, Hugging Face Transformers, FastAPI, MySQL

[0834] Program processing overview

[0835] 1. Uploading and Receiving Files

[0836] User: Factory workers and managers upload daily factory work reports through the robot's dedicated interface.

[0837] Terminal: Validates the file through the provided interface and sends it to the server using the HTTPS protocol, during which process it is verified that the file has been uploaded correctly.

[0838] 2. Data Reception and Analysis

[0839] Server: After receiving the uploaded daily report data, the server uses a natural language processing engine (Hugging Face Transformers) to tokenize and analyze the parts of speech of the daily report content, and then uses a sentiment analysis engine to analyze emotions such as positive, negative, and neutral.

[0840] 3. Summary Generation and Trend Extraction

[0841] Server: The parsed data is passed to a generative model (TensorFlow) to generate summaries, and then text mining techniques (TF-IDF) are used to extract important keywords and trends.

[0842] 4. Condition Monitoring and Anomaly Detection

[0843] Robot: Monitors the status of equipment in the factory in real time and compares it with analyzed daily data. If an abnormality is detected, the information is quickly sent to a server and notified to each department along with trend information.

[0844] 5. Report distribution

[0845] Server: Integrates the generated summary and trend information, generates reports in a format suitable for each department's terminal, and immediately distributes the reports to each department, where they are acknowledged for receipt.

[0846] Specific examples

[0847] Example 1: Suppose a factory daily report contains the following statement: "Today's work status of machine A is reported. No abnormalities." When the system analyzes this daily report, it extracts the keywords "machine A," "work status," and "no abnormalities," and generates a summary statement: "The work status of machine A is normal." This is then distributed to each department along with related trend information.

[0848] Example prompt sentence:

[0849] "Analyze the following daily report for key keywords, summaries, and sentiment analysis:

[0850] I'll report on the work status of Machine A today. No abnormalities."

[0851] This invention enables automatic and efficient data analysis and anomaly detection in factories, enabling rapid troubleshooting. It also enables each department to make efficient decisions based on the generated summary and trend information.

[0852] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0853] Step 1:

[0854] Uploading and Receiving Files

[0855] The user uploads the factory's daily work report through the robot's dedicated interface, selects a specific file, and presses the send button.

[0856] Input: Factory work daily report file

[0857] Output: Uploaded files are sent to the device.

[0858] Specific operation: Use the file selection dialog on the user interface to select a file and click the Send button. The device will send the file to the server using the HTTPS protocol.

[0859] Step 2:

[0860] Data reception and analysis

[0861] The server receives the uploaded daily report data and uses a natural language processing engine (Hugging Face Transformers) to tokenize and analyze the parts of speech of the daily report content.

[0862] Input: Uploaded daily report data

[0863] Output: Tokenized and part-of-speech analyzed data

[0864] Specific operation: The server decodes the received data into UTF-8 and converts it into text format. This is then passed to a natural language processing engine, which tokenizes the document and analyzes the parts of speech.

[0865] Step 3:

[0866] Emotion analysis

[0867] The server passes the tokenized data to a sentiment analysis engine, which analyzes the sentiment for positive, negative, neutral, etc.

[0868] Input: Tokenized and part-of-speech analyzed data

[0869] Output: Emotion analysis results

[0870] How it works: The parsed data is fed into a sentiment analysis engine, which calculates a sentiment score for each sentence or phrase. Based on this, the sentiment of the entire document is evaluated.

[0871] Step 4:

[0872] Summary Generation

[0873] The server uses a generative model (TensorFlow) to generate a summary based on the analysis results.

[0874] Input: Sentiment analysis results and tokenized data

[0875] Output: Generated summary

[0876] How it works: The sentiment analysis results are combined with the tokenized data and passed to a generative model, which extracts key information and generates a summary.

[0877] Step 5:

[0878] Trend and Keyword Extraction

[0879] The server uses text mining technology (TF-IDF) to extract important keywords and trends.

[0880] Input: Tokenized and part-of-speech analyzed data

[0881] Output: Extracted keywords and trend information

[0882] Specific operation: Text data is input into the TF-IDF algorithm to calculate and extract important keywords and trends that frequently appear in the document.

[0883] Step 6:

[0884] Condition Monitoring and Anomaly Detection

[0885] The robot monitors the status of equipment in the factory in real time and detects abnormalities by comparing it with analyzed daily data.

[0886] Input: Real-time monitoring data and daily report analysis results

[0887] Output: Anomaly detection notification

[0888] Specific operation: The monitored data is analyzed in real time and compared with the saved daily data. If an abnormal value is detected, an alert is sent to the server.

[0889] Step 7:

[0890] Report Distribution

[0891] The server integrates the generated summary and trend information, generates and distributes reports in a format suitable for each department's terminal.

[0892] Input: Generated abstracts, keywords and trend information

[0893] Output: Reports delivered to departments

[0894] Specific operation: The generated summaries, keywords, and trend information are formatted and a report is created in a format that matches the settings of each department. This is then distributed to each department's terminal and receipt is confirmed.

[0895] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0896] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0897] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0898] [Third embodiment]

[0899] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0900] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0901] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0902] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0903] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0904] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0905] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0906] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0907] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0908] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0909] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0910] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0911] Overall system overview

[0912] This system analyzes business daily reports and information uploaded by users on a server, automatically generates summaries and trend information, and distributes them to each department, thereby streamlining information sharing. Specifically, it combines natural language processing technology, generative models, and keyword and trend extraction algorithms to achieve fast and accurate information analysis and distribution.

[0913] Program processing flow and detailed explanation

[0914] Uploading and Receiving Files

[0915] User: Upload daily reports and business information files through the device interface.

[0916] On your device: Automatically validates the file format and size to ensure it is safe before sending it to the server.

[0917] Calling the API for receiving data

[0918] Terminal: Call the API endpoint for sending a file and send the file contents to the server.

[0919] Server: Stores the received file in temporary storage and then passes it to the natural language processing module.

[0920] Analysis using natural language processing

[0921] Server: Using a natural language processing engine, the uploaded document is broken down into sentences and words, tokenized, and analyzed for parts of speech to determine the meaning and relationships of each word.

[0922] Server: Prepares the analysis results in a data frame format and passes them to the summary generation module.

[0923] Generate a summary

[0924] Server: Uses a generative model to generate a summary from the analysis results, evaluating the importance of each sentence to ensure that important information is not omitted.

[0925] Server: Formats the generated summary and presents it in a format that is easy for users to understand.

[0926] Keyword and trend extraction

[0927] Server: Uses text mining technology to extract important keywords and trends from the analysis results.

[0928] Server: Adds extracted keywords and trends to a summary and creates the final report.

[0929] Report distribution

[0930] Server: Prepares the generated reports for distribution to each department, customizing the format based on user settings.

[0931] Server: Distributes reports to each department's terminal and confirms receipt.

[0932] View and share your results

[0933] Terminal: Displays received reports to users, allowing each department to quickly access the information they need.

[0934] Users: Review reports and share them with other employees or departments as needed, supporting efficient information sharing and quick decision-making.

[0935] Specific examples

[0936] Sales department daily report analysis and summary

[0937] 1. User: Upload the sales report (e.g. sales_report_2023_10_15.pdf) to the system from the terminal.

[0938] 2. Terminal: Validate the file and send it to the server.

[0939] 3. Server: Receives the daily reports and performs analysis using a natural language processing engine.

[0940] 4. Server: Based on the analysis results, a generative model creates a summary, summarizing key selling points and customer trends.

[0941] 5. Server: Delivers reports containing summary and trend information to sales department terminals.

[0942] 6. Terminal: Receives the report and displays it for the sales team to view.

[0943] 7. User: Review the reports and make decisions that will influence your sales strategy.

[0944] This system allows users to efficiently analyze the large amount of information generated in daily business operations and use it to make quick decisions.

[0945] The processing flow will be explained below.

[0946] Step 1:

[0947] User: Uploads daily reports and business information files using the device interface. Through the file selection dialog, select the specific file and press the send button.

[0948] Step 2:

[0949] Terminal: Validates the selected file, checking that the file format is supported and that the size is within the regulated range. If there are no problems, generates an API request to send the file to the server.

[0950] Step 3:

[0951] Terminal: Sends the generated API request to the server's API endpoint, usually using a secure protocol such as HTTPS.

[0952] Step 4:

[0953] Server: Received files are stored in temporary storage, where file metadata and other information are recorded and prepared for subsequent processing.

[0954] Step 5:

[0955] Server: Calls the natural language processing (NLP) engine to analyze the contents of the uploaded file. First, it breaks down the file into sentences and words, and performs tokenization and part-of-speech analysis.

[0956] Step 6:

[0957] Server: Stores the results of tokenization and part-of-speech analysis. Specifically, it stores the analysis results in data frame format and passes them to the next summary generation process.

[0958] Step 7:

[0959] Server: Uses a generative model for summary generation to generate a summary from the analysis results. Extracts important points from the document and summarizes the information.

[0960] Step 8:

[0961] Server: Evaluates the generated summary and formats it in an appropriate format. Here, the summary is organized into a concise and easy-to-read format so that users can quickly grasp the information.

[0962] Step 9:

[0963] Server: Along with the summary, we use text mining techniques to extract important keywords and trends. We use TF-IDF and word clouds to extract highly relevant information.

[0964] Step 10:

[0965] Server: Adds the extracted results to the summary and creates a final report, which includes the summary, keywords, and trend information.

[0966] Step 11:

[0967] Server: Prepares the created report for distribution to each department, customizes the format according to the user's settings, and generates a distribution list.

[0968] Step 12:

[0969] Server: Delivers reports to terminals in each department. Sends reports to designated recipients using a distribution protocol.

[0970] Step 13:

[0971] Terminal: Stores received reports and displays them in the user interface, allowing users to easily view and review the reports.

[0972] Step 14:

[0973] Users: Check the displayed report and share it with other employees or departments as needed, including by email or by sending a link to an internal chat tool.

[0974] This series of steps allows users to quickly and accurately analyze business information, obtain summary and trend information, and make effective decisions.

[0975] Example 1

[0976] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0977] In today's business environment, a huge amount of information is generated every day, creating a need to analyze and summarize that information quickly and accurately, and share it efficiently. However, traditional methods require a significant amount of time and effort to analyze and share information. Furthermore, manual information analysis is prone to errors and carries a high risk of important information being overlooked. Furthermore, the wide variety of information formats makes it difficult to analyze information in a unified and efficient manner. There is a need for a system that solves these problems.

[0978] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0979] In this invention, the server includes: means for a user to upload documents to the system through an electronic device interface; means for a terminal to verify the format and size of the uploaded documents and transmit them to the server after confirming their safety; means for the server to call a data reception API to receive and temporarily store the files; means for the server to use natural language processing technology to break down the uploaded documents into sentence units and words and perform tokenization and part-of-speech analysis; means for the server to generate the analysis results in a data frame format and pass them to a summary generation model; means for the server to use the generation model to generate summaries from the analysis results; means for the server to extract important keywords and trends and add them to the summaries to create a final report; means for the server to distribute the generated summaries and the extracted trends and keywords to terminals in each department; and means for the terminal to display the received reports to the user and enable information to be shared with other departments, thereby enabling efficient analysis of information and rapid sharing.

[0980] "User" refers to the entity that uses the system to upload materials and receive analysis results.

[0981] "Electronic Device" means any device that allows a User to manipulate and upload materials to the System, including, for example, a computer, smartphone, or tablet.

[0982] An "interface" is a point of contact between a user and a system for exchanging information. Examples include web browsers and dedicated applications.

[0983] "Documents" refers to business daily reports and other written information uploaded to the system, including, for example, PDF, Excel, Word files, etc.

[0984] "Terminal" means an electronic device operated by a User that verifies the format and size of uploaded materials and ensures their security.

[0985] "Server" refers to the computer system that receives uploaded materials, generates analysis and summaries, and distributes them to each department.

[0986] "Data Reception API" refers to a program interface for sending materials from a terminal to a server.

[0987] "Temporary storage" refers to the temporary storage of received materials by a server, including, for example, cloud storage and databases.

[0988] "Natural language processing technology" is a technology for analyzing uploaded documents, and is primarily a function for tokenization and part-of-speech analysis.

[0989] "Tokenization" refers to the process of breaking down a document into words or sentences.

[0990] "Part-of-speech analysis" refers to the process of identifying the part of speech of each word in a document and analyzing its grammatical relationships.

[0991] "Data frame format" refers to a data structure for storing analysis results in a table-like format.

[0992] A "summary generation model" refers to an algorithm or machine learning model for generating a summary based on the analysis results.

[0993] "Keywords" refer to words that appear frequently in a document and have important meanings.

[0994] "Trends" refers to major themes or tendencies extracted from the information in a document.

[0995] The "final report" refers to a document that compiles the generated summary, keywords, and trend information.

[0996] "Delivery" refers to the process of sending the generated final report to the terminals of each department.

[0997] "Receiving" refers to the terminal receiving distribution from the server.

[0998] "Display" refers to the terminal visually indicating to the user the report received.

[0999] "Sharing" refers to the act of a user collaborating and communicating the received report with other employees or departments.

[1000] MODE FOR CARRYING OUT THE INVENTION

[1001] This invention is a system that improves the efficiency of information sharing by analyzing business daily reports and information uploaded by users on a server, automatically generating summaries and trend information, and distributing them to each department. The system includes the following main elements:

[1002] Uploading and Receiving Files

[1003] Users upload files such as daily reports and business information through the interface of their electronic device. Specifically, they use the browser's file selection dialog to select a document such as "sales_report_2023_10_15.pdf" and click the upload button. The device then verifies the file's format and size (e.g., PDF, Excel, Word), confirms its security, and sends it to the server.

[1004] Calling the API for receiving data

[1005] The device calls an API endpoint (e.g., https: / / api.example.com / upload) to upload the file to the server. The server temporarily stores the file received at the API endpoint (e.g., Amazon S3 or Google Cloud Storage) and passes the path to the natural language processing engine.

[1006] Analysis using natural language processing

[1007] The server uses a natural language processing (NLP) engine (e.g., SpaCy, NLTK) to break down the uploaded document into sentences and words. Specifically, it performs tokenization, part-of-speech analysis, and removes stop words. The analysis results are formatted as a data frame (e.g., Pandas DataFrame) and passed to the summary generation model.

[1008] Generate a summary

[1009] The server uses a generative AI model (e.g., GPT-3, BERT) to generate a summary from the analysis results. The summary is generated by extracting particularly important sentences based on their importance scores and assembling them into a sentence. The generated summary is then organized in a format that is easy for users to understand (e.g., Markdown or HTML).

[1010] Keyword and trend extraction

[1011] The server uses text mining techniques (e.g., TF-IDF, Word2Vec) to extract important keywords and trends from the analysis results, including frequency analysis and semantic relevance assessment. The extracted keywords and trend information are added to a summary to create the final report.

[1012] Report distribution

[1013] The server prepares the generated reports for distribution to each department, customizing the reports based on user settings and departmental formatting requirements, and delivering the reports via email or the company's internal portal, where receipt is acknowledged.

[1014] View and share your results

[1015] The device displays the received report to the user, for example, through a web interface or desktop notification. The user can review the report's contents and share the link in an internal chat tool to share key selling points with other departments.

[1016] Prompt Sentence Examples

[1017] Below are some example input prompts for the generative AI model that can be used in this system:

[1018] "Today, we uploaded the sales report for October 15, 2023. Please extract trending keywords and summarize them. Don't forget to include your key selling points."

[1019] This system allows users to efficiently analyze the large amount of information generated in daily business operations and use it to make quick decisions.

[1020] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1021] Step 1:

[1022] The user uploads the document using an electronic device. The user opens a file selection dialog using the interface of their device, selects the daily report or business information file (e.g., "sales_report_2023_10_15.pdf"), and clicks the upload button. The input is the file selected by the user, and the output is the file selection information sent to the device.

[1023] Step 2:

[1024] The terminal verifies the format and size of the uploaded file to confirm safety. The terminal checks the file format (e.g. PDF, Excel, Word) and whether the file size exceeds the specified limit. It also performs a virus check. The input is the file uploaded by the user, and the output is the verification result (safety confirmation).

[1025] Step 3:

[1026] The device sends the verified file to the server. The device calls an API endpoint (e.g., https: / / api.example.com / upload) to upload the verified file to the server via an HTTP POST request. The input is the verified file, and the output is a notification that the file was successfully uploaded.

[1027] Step 4:

[1028] The server temporarily stores the received file. The server stores the file received at the API endpoint in temporary storage (e.g., Amazon S3 or Google Cloud Storage) and records the path. The input is the file sent from the device, and the output is the path to the temporary storage.

[1029] Step 5:

[1030] The server analyzes the file using a natural language processing (NLP) engine. The server passes the saved file to an NLP engine (e.g., SpaCy, NLTK) and breaks it down into sentences and words. It performs tokenization, part-of-speech analysis, and removes stop words, and then organizes the analysis results in data frame format (e.g., Pandas DataFrame). The input is the path to the saved file, and the output is a data frame of the analysis results.

[1031] Step 6:

[1032] The server generates a summary using a generative AI model. After tokenization and part-of-speech analysis, the server passes the data frame to a generative AI model (e.g., GPT-3, BERT) to generate a summary. It extracts particularly important sentences based on their importance scores and assembles them into a summary. The input is the analysis result data frame, and the output is the generated summary.

[1033] Step 7:

[1034] The server uses text mining techniques to extract keywords and trends. The server uses TF-IDF models and Word2Vec to extract important keywords and trends from the analysis results. This includes word frequency analysis and semantic relevance assessment. The input is a data frame of the analysis results, and the output is the extracted keywords and trends.

[1035] Step 8:

[1036] The server creates a final report and adds keywords and trends to the summary. The server adds keyword and trend information to the generated summary and formats the final report. The report is prepared in a format that is easy for users to understand (e.g., Markdown or HTML). The input is the generated summary and extracted keywords and trends, and the output is the final report.

[1037] Step 9:

[1038] The server prepares the final report for delivery to each department. The server customizes the report based on user settings and departmental formatting requirements. The server delivers the report via email or through the company's internal portal. The input is the final report and the output is the customized report.

[1039] Step 10:

[1040] The server distributes the report and the terminal confirms receipt. The server distributes the customized report to each department's terminal and receives a receipt confirmation. The input is the customized report and the output is the receipt confirmation result.

[1041] Step 11:

[1042] The terminal displays the received reports to the user. The terminal makes the reports available to the user through a web interface or desktop notification. The input is the received report and the output is the report displayed to the user.

[1043] Step 12:

[1044] The user reviews the report and shares it with other departments. The user reviews the received report and shares important selling points and analysis results with other employees and departments using an internal chat tool, etc. The input is the report displayed to the user, and the output is the results shared with other departments.

[1045] This is the flow of processing for this system's program, which enables efficient analysis and rapid sharing of information.

[1046] (Application example 1)

[1047] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1048] Conventional autonomous driving systems have had difficulty efficiently analyzing massive amounts of sensor data and operational information, summarizing it, and providing it to drivers and managers. This can lead to delays in quickly identifying important trends and operational issues and taking appropriate measures. Furthermore, there is no established method for displaying or distributing the generated data, which can lead to delays in information sharing and decision-making.

[1049] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1050] In this invention, the server includes means for analyzing uploaded documents using natural language processing technology, means for using a generative model to generate summaries based on the analysis results, means for distributing the generated summaries and extracted trends and keywords to terminals in each department, means for uploading sensor data and operation information and analyzing and summarizing them, and means for extracting important information from the operation data of the autonomous vehicle and displaying it to the driver or manager. This allows the operation information of the autonomous vehicle to be analyzed efficiently and quickly, summarized, and provided, making it possible to immediately grasp important trends and issues and respond promptly.

[1051] "Natural language processing technology" is a technology that automatically analyzes and understands text data and processes human language using a computer.

[1052] A "generative model" is a machine learning algorithm that generates new data based on given data.

[1053] A "trend" refers to the tendency of fluctuations or patterns in data or information over a period of time.

[1054] "Keywords" refer to words or phrases that have particular significance in documents or data.

[1055] "Sensor data" refers to measurement data such as temperature, pressure, position, and speed obtained from various sensors.

[1056] "Operation information" is data relating to the operation status of a vehicle, including location information, speed, route, traffic conditions, and the like.

[1057] An "autonomous vehicle" refers to a vehicle that drives autonomously without the need for human operation.

[1058] An "abstract" is a short sentence or text that succinctly summarizes the important points of the original document or data.

[1059] An "API endpoint" refers to a specific address or operation that allows software to communicate with other software.

[1060] "Customized format" means data or reports that have been formatted in a specific way according to the user's requirements or purposes.

[1061] A "prompt sentence" is text containing instructions or questions to be provided as input to a generative model.

[1062] The system for implementing this invention analyzes autonomous vehicle operation data and sensor data uploaded by users on a server, automatically generates summaries and trend information, and distributes them to drivers and managers. A detailed description of the system is as follows.

[1063] Program processing overview

[1064] Uploading and Receiving Files

[1065] Users upload operational data for autonomous vehicles via their smartphones or on-board displays. The device automatically verifies the file format and size, confirms that it is safe, and then sends it to the server.

[1066] Calling the API for receiving data

[1067] The device calls the API endpoint for sending a file and sends the file contents to the server, which then stores the received file in temporary storage and passes it to the natural language processing module.

[1068] Analysis using natural language processing

[1069] The server uses a natural language processing engine (e.g., spaCy or NLTK) to break down the uploaded operation logs and error reports into sentences and words. It then performs tokenization and part-of-speech analysis to analyze the meaning and relationships of each word. The analysis results are prepared in data frame format and passed to a summary generation model (e.g., GPT series).

[1070] Generate a summary

[1071] The server uses a generative model to generate a summary from the analysis results. During this process, it evaluates the importance of each sentence to ensure that important information is not omitted. The generated summary is then formatted and presented in a format that is easy for users to understand.

[1072] Keyword and trend extraction

[1073] The server uses text mining techniques to extract important keywords and trends from the analysis results, adds them to a summary, and creates a final report.

[1074] Report distribution

[1075] The server prepares the generated report for distribution to the driver or manager, customizes the report format based on the user's settings, and delivers the report to a smartphone or in-vehicle display for confirmation of receipt.

[1076] View and share your results

[1077] The terminal displays the received reports to the user, allowing them to quickly access the information they need. Users can then review the reports and share them with other drivers and fleet managers, supporting efficient information sharing and quick decision-making.

[1078] Specific examples

[1079] We will provide a concrete example of uploading log data from a self-driving vehicle's daily operations, analyzing it in real time, and extracting trends and important issues.

[1080] Example prompt for a generative AI model:

[1081] "Please extract important information from the operational data log below and generate a written summary."

[1082] Data Log:

[1083] This system will enable drivers and managers of autonomous vehicles to efficiently analyze and summarize data, enabling them to make quick decisions.

[1084] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1085] Step 1:

[1086] Uploading and Receiving Files

[1087] Users use their smartphones or in-vehicle displays to upload operational and sensor data from their autonomous vehicles. The device automatically validates the file format and size to ensure the content is safe. The input data is an operational data file, and the output is a file ready to be sent to the server. Specifically, the device checks the file's header information to ensure it is in an acceptable format and size.

[1088] Step 2:

[1089] Calling the API for receiving data

[1090] The terminal calls the API endpoint for file transmission and sends the contents of the operation data file to the server. The input data is the verified operation data file, and the output is a file stored in the server's temporary storage. Specifically, the terminal generates an API request, includes the file data in the request body, and sends it to the server. The server stores the received data in temporary storage.

[1091] Step 3:

[1092] Analysis using natural language processing

[1093] The server uses a natural language processing engine (e.g., spaCy or NLTK) to analyze the received operation data file. The input data is the operation data file stored in temporary storage, and the output is a data frame that has been tokenized and analyzed for parts of speech. Specifically, the server reads the file, tokenizes it into sentences and words, analyzes the parts of speech for each, and converts it into a data frame format.

[1094] Step 4:

[1095] Generate a summary

[1096] The server generates a summary using a generative model (e.g., GPT series) based on the analysis results. The input data is a data frame of the analysis results, and the output is a summary. Specifically, the server passes the data frame as input to the generative model, and generates a summary while evaluating important information.

[1097] Step 5:

[1098] Keyword and trend extraction

[1099] The server uses text mining technology to extract important keywords and trends from the analysis results. The input data is a summary sentence and a data frame of the analysis results, and the output is important keywords and trend information. Specifically, the server applies text mining technology to detect frequently occurring keywords and trend patterns.

[1100] Step 6:

[1101] Report creation and distribution

[1102] The server adds the extracted keywords and trends to the summary and creates a final report. The input data is the summary, keywords, and trend information, and the output is a completed report. Specifically, the server integrates the keywords and trend information into the summary and generates a report in a customized format based on user settings. The final report is then distributed to the driver's or manager's device.

[1103] Step 7:

[1104] View and share your results

[1105] The terminal displays the received report to the user. The input data is the delivered report, and the output is the information displayed to the user. Specifically, the terminal analyzes the report data and displays it in a visually easy-to-understand format. The user can review the report and share it with other drivers or fleet managers.

[1106] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1107] Overall system overview

[1108] This system analyzes daily business reports and information uploaded by users on a server, automatically generates summaries and trend information, and distributes them to each department. Furthermore, by combining it with an emotion engine, it is possible to analyze the sentiment of documents and reflect this in summaries and trend information. This allows users to analyze information while taking into account the emotional aspects of documents, enabling them to make quick and accurate decisions.

[1109] Program processing flow and detailed explanation

[1110] Uploading and Receiving Files

[1111] User: Uploads daily reports and business information files through the device interface, selects a specific file through the file selection dialog, and presses the send button.

[1112] Calling the API for receiving data

[1113] Terminal: Call the API endpoint for sending the file and send the file contents to the server. The sending protocol is usually a secure one such as HTTPS.

[1114] Analysis using natural language processing

[1115] Server: Calls the natural language processing (NLP) engine to analyze the contents of the uploaded file. First, it breaks the file down into sentences and words, then performs tokenization and part-of-speech analysis.

[1116] Server: Stores the results of tokenization and part-of-speech analysis. Specifically, the analysis results are stored in data frame format and passed to the next summary generation process.

[1117] Emotion analysis

[1118] Server: Calls the sentiment engine to analyze the sentiment of the uploaded document, identifying positive and negative phrases and words in the text and assessing the type and intensity of the sentiment.

[1119] Server: Integrates the analysis results into a summary generation model to generate summaries that include emotional information.

[1120] Generate a summary

[1121] Server: Generates a summary from the analysis results using a generative model, extracting important points from the document and summarizing the information.

[1122] Server: Evaluates the generated summary and formats it appropriately. Here, it is formatted in a concise and easy-to-read format to allow users to quickly grasp the information.

[1123] Keyword and trend extraction

[1124] Server: Uses text mining techniques to extract important keywords and trends from the analysis results, and uses TF-IDF and word clouds to extract highly relevant information.

[1125] Server: Adds extracted keywords and trends to a summary and creates a final report.

[1126] Report distribution

[1127] Server: Prepares the generated reports for distribution to each department, customizes the format according to the user's settings, and generates the distribution list.

[1128] Server: Distributes reports to each department's terminal and confirms receipt.

[1129] View and share your results

[1130] Terminal: Stores the received reports and displays them in the user interface, allowing users to easily view and check the reports.

[1131] Users: Review the displayed report and share it with other employees or departments as needed, including by email or by sending a link to an internal chat tool.

[1132] Specific examples

[1133] Sales department daily report analysis and summary

[1134] 1. User: Upload the sales report (e.g. sales_report_2023_10_15.pdf) to the system from the terminal.

[1135] 2. Terminal: Validate the file and send it to the server.

[1136] 3. Server: Receives the daily reports and performs analysis using a natural language processing engine and sentiment engine.

[1137] 4. Server: Based on the analysis results, a generative model creates a summary, summarizing key selling points, customer trends, and sentiment information.

[1138] 5. Server: Delivers reports containing summaries, trend information, and sentiment information to sales department terminals.

[1139] 6. Terminal: Receives the report and displays it for the sales team to view.

[1140] 7. User: Review the reports and make decisions that will influence your sales strategy.

[1141] This system enables users to efficiently analyze the large amount of information generated in daily business operations, quickly grasp summary and trend information, as well as sentiment information, and make effective decisions.

[1142] The processing flow will be explained below.

[1143] Step 1:

[1144] User: Uploads daily reports and business information files using the device interface. Through the file selection dialog, select the specific file and press the send button.

[1145] Step 2:

[1146] Terminal: Validates the selected file, checking whether the file format is supported and whether the size is within the regulated range. If there are no problems, generates an API request to send the file to the server.

[1147] Step 3:

[1148] Terminal: Sends the generated API request to the server's API endpoint, usually using a secure protocol such as HTTPS.

[1149] Step 4:

[1150] Server: Received files are stored in temporary storage, where file metadata and other information are recorded and prepared for subsequent processing.

[1151] Step 5:

[1152] Server: Calls the natural language processing (NLP) engine to analyze the contents of the uploaded file. First, it breaks the file down into sentences and words, then performs tokenization and part-of-speech analysis.

[1153] Step 6:

[1154] Server: Stores the results of tokenization and part-of-speech analysis. Specifically, it stores the analysis results in data frame format and passes them to the next summary generation process.

[1155] Step 7:

[1156] Server: Calls the sentiment engine to analyze the sentiment of the uploaded document, identifying positive and negative phrases and words in the text and assessing the type and intensity of the sentiment.

[1157] Step 8:

[1158] Server: Stores the analyzed emotion data and integrates it into the summary generation model, thereby generating summaries that include emotion information.

[1159] Step 9:

[1160] Server: Using a generative model for summary generation, a summary is generated based on the analysis results and sentiment data. Important points of the document are extracted and the information is summarized.

[1161] Step 10:

[1162] Server: Evaluates the generated summary and formats it in an appropriate format. Here, it is organized into a concise and easy-to-read format so that users can quickly grasp the information.

[1163] Step 11:

[1164] Server: Uses text mining techniques to extract important keywords and trends from the analysis results. Uses TF-IDF and word clouds to extract highly relevant information.

[1165] Step 12:

[1166] Server: Adds the extracted keywords and trends to a summary and creates a final report, including sentiment information.

[1167] Step 13:

[1168] Server: Prepares the created report for distribution to each department, customizes the format according to the user's settings, and generates a distribution list.

[1169] Step 14:

[1170] Server: Delivers reports to terminals in each department. Sends reports to designated recipients using a distribution protocol.

[1171] Step 15:

[1172] Terminal: Stores received reports and displays them in the user interface, allowing users to easily view and check the reports.

[1173] Step 16:

[1174] Users: Check the displayed report and share it with other employees or departments as needed. Sharing methods include email and internal chat tool links.

[1175] This detailed processing step enables users to efficiently analyze the large amount of information generated in daily business operations, quickly grasp summary and trend information, as well as sentiment information, and make effective decisions.

[1176] Example 2

[1177] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1178] In today's business environment, a huge amount of business documents and reports are generated every day, and there is a need to efficiently analyze them and quickly grasp important information. Conventional methods require manually analyzing document content and creating summaries, which is time-consuming and labor-intensive. Furthermore, it is difficult to appropriately evaluate and reflect the emotional information contained in documents. This leads to issues such as overlooking information and delays in identifying important trends.

[1179] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: a means for a user to upload a file through an interface; a means for a terminal to call an API endpoint and send the contents of the file; a means for the server to analyze the uploaded document using natural language processing technology; a means for analyzing the sentiment of the document using an emotion engine; a means for using a generative model to generate a summary based on the analysis results; a means for distributing the generated summary and extracted trends and keywords to terminals in each department; and a means for saving and displaying reports received by the terminals. This makes it possible to efficiently analyze the content and sentiment information of business documents and to quickly and accurately generate and distribute summaries and trend information.

[1180] "User" refers to the user who uploads business documents and daily reports through the system interface.

[1181] A "terminal" is a device used by a user that has the functionality to upload files and send requests to API endpoints.

[1182] "Server" refers to a central computing device that analyzes received files, generates summaries and trend information, and distributes it to each department.

[1183] "Natural language processing technology" is a technology for analyzing documents, and has the ability to break them down into sentences or words and perform tokenization and part-of-speech analysis.

[1184] An "API endpoint" is a specific URL or URI that a device uses to communicate with a server and is intended to receive HTTPS requests.

[1185] The "Sentiment Engine" is an analytical tool for identifying emotive phrases in a document and assessing positive and negative sentiment.

[1186] A "generative model" is an algorithm or AI model used to generate a summary of a document based on the results of its analysis.

[1187] "Tokenization" is the process of dividing a document into smaller units such as words or phrases.

[1188] "Part-of-speech analysis" is the process of identifying the part of speech (noun, verb, adjective, etc.) for each tokenized word or phrase.

[1189] A "data frame" is a data structure for storing analysis results, and is tabular data represented by rows and columns.

[1190] "Keywords" are words or phrases that are considered to be particularly important within a document.

[1191] "Trend information" refers to significant trends and patterns extracted from analyzed documents.

[1192] "Report" means the final output document containing the generated summary and extracted keyword trend information.

[1193] "Format" refers to the style and structure in which a report or summary is presented.

[1194] "Distribution" is the process of sending the generated report from the server to the terminals of each department.

[1195] "Display" refers to presenting the report received at the terminal on the user interface.

[1196] A "department" is a group responsible for a specific task or function in a company or organization.

[1197] In the system according to the present invention, a server analyzes business documents and daily reports uploaded by users and automatically generates summaries and trend information. Furthermore, an emotion engine can be used to analyze the emotional information of the documents and reflect this in the generated summaries and trend information. The configuration and operation of the system are described in detail below.

[1198] First, the user uploads a business document or daily report (e.g., sales_report_2023_10_15.pdf) using the terminal by selecting the appropriate file from the file selection dialog and clicking the upload button.

[1199] The device then sends the selected file to an API endpoint (e.g., "https: / / example.com / upload") using the HTTPS protocol. This request is sent as multipart form data.

[1200] The server inputs the received file into a natural language processing engine, which tokenizes and analyzes parts of speech at the sentence and word level. The analysis results are stored in a data frame format using the pandas library. The sentiment engine then evaluates the sentiment within the document based on the analysis results. It identifies positive and negative phrases and quantifies the strength of the sentiment.

[1201] The generative AI model generates a summary based on the analysis results and sentiment information. For example, it generates a summary such as, "Sales increased 15% this month, but there were many returns for a specific product group." The generated summary is formatted in Markdown or HTML for easy reading.

[1202] The server then uses text mining techniques (e.g., TF-IDF and word cloud) to extract important keywords and trend information from the analysis results. Keywords such as "sales," "surge," and "returns" are extracted. These keywords and trend information are also incorporated into the summary text to create the final report.

[1203] The server customizes the generated report according to the user's settings and prepares it for distribution to each department. The report is sent to each department's terminal based on the distribution list. This can be done by email using the SMTP protocol or by sending it to an internal notification system via REST API.

[1204] Finally, the terminal saves the received report and displays it in the user interface. The user can then review the displayed report and share it with other employees or departments as needed. This can be done via email or a link to an internal chat tool. A specific example of this is to share a link to the generated report in an internal chat so that everyone can access it.

[1205] In this way, this invention makes it possible to efficiently analyze large volumes of business documents and daily reports, and quickly and accurately generate and distribute summaries, trend information, and sentiment information, enabling users to quickly grasp important business information and make appropriate decisions.

[1206] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1207] Step 1:

[1208] User: Upload the daily report or business information file through the device interface by selecting the appropriate file (e.g., sales_report_2023_10_15.pdf) from the file selection dialog and clicking the upload button.

[1209] Input: File (for example, sales_report_2023_10_15.pdf).

[1210] Output: The file is selected on the device and ready to be uploaded.

[1211] Step 2:

[1212] On the device: After selecting the file, send an HTTPS request to the API endpoint (e.g., "https: / / example.com / upload"), which sends the file to the server as multipart form data.

[1213] Input: The selected file.

[1214] Output: A request to send a file to the server.

[1215] Step 3:

[1216] Server: The received file is input into a natural language processing (NLP) engine, which performs tokenization and part-of-speech analysis on a sentence-by-sentence and word-by-word basis. Specifically, the document is broken down into multiple words and each word is tagged with a part of speech.

[1217] Input: The contents of the uploaded file.

[1218] Output: Tokenized and part-of-speech analyzed data in dataframe format.

[1219] Step 4:

[1220] Server: The analysis results are stored in a data frame format and passed to the emotion engine. Here, the analysis results are converted into a data frame using the pandas library.

[1221] Input: Tokenized and part-of-speech analyzed data.

[1222] Output: Analysis results in data frame format.

[1223] Step 5:

[1224] Server: Uses a sentiment engine to assess the sentiment in the document, identifying positive and negative phrases and quantifying the strength of each sentiment.

[1225] Input: Analysis results in data frame format.

[1226] Output: Sentiment-analyzed data with sentiment scores.

[1227] Step 6:

[1228] Server: Generates summaries using generative AI models that integrate emotional information. They combine key document points and emotional elements to generate concise summaries.

[1229] Input: Sentiment-analyzed data.

[1230] Output: The generated summary.

[1231] Step 7:

[1232] Server: Uses text mining techniques to extract important keywords and trend information, specifically using TF-IDF and word clouds to identify frequently occurring words.

[1233] Input: Sentiment-analyzed data.

[1234] Output: Extracted keywords and trend information.

[1235] Step 8:

[1236] Server: The generated summary is combined with keywords and trend information to create a final report, which is formatted in Markdown or HTML.

[1237] Input: Abstract, keywords, trend information.

[1238] Output: Final report in Markdown or HTML format.

[1239] Step 9:

[1240] Server: Customize the final report based on user settings and generate a distribution list to distribute it to each department. The distribution method can be email using the SMTP protocol or via REST API to the company's notification system.

[1241] Inputs: Final report, distribution list.

[1242] Output: Reports distributed to each department.

[1243] Step 10:

[1244] Terminal: Saves the received reports and displays them in a user interface, specifically in HTML format using the Report Viewer.

[1245] Input: The report received.

[1246] Output: The report displayed in the user interface.

[1247] Step 11:

[1248] Users can review the displayed report and share it with other employees or departments as needed. They can share the link via email or internal chat tools, or download it as a PDF and distribute it.

[1249] Input: The report displayed in the user interface.

[1250] Output: The shared report.

[1251] (Application example 2)

[1252] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1253] There is a need to efficiently manage factory daily work logs and equipment status, and to quickly detect anomalies and provide information on future trends. However, current methods require manual analysis of huge amounts of data, which is time-consuming and labor-intensive, and the accuracy of the analysis results is unstable. Furthermore, real-time monitoring and anomaly detection are difficult, making it difficult to respond quickly when problems occur.

[1254] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1255] In this invention, the server includes means for analyzing uploaded documents using natural language processing technology, means for using a generative model to generate summaries based on the analysis results, means for providing analysis results of equipment status and daily work reports to robots operating in the factory, means for the robots to monitor the equipment status in real time and detect anomalies, and means for distributing the generated summaries and extracted trends and keywords to terminals in each department.This enables automatic and efficient data analysis and anomaly detection, enabling rapid troubleshooting.

[1256] "Natural language processing technology" is a technology that allows computers to understand, analyze, and generate human language, and includes document tokenization, part-of-speech analysis, sentiment analysis, and more.

[1257] A "generative model" is a machine learning algorithm that learns patterns in knowledge and data and generates new data.

[1258] "Equipment status" is information that indicates the current operating state of machines and devices in a factory, and includes parameters such as temperature, vibration, and operating time.

[1259] A "daily work report" is a report that details the work performed and events that occurred on a specific work day.

[1260] "Real-time monitoring" refers to continuously monitoring the status of equipment and systems within a factory and obtaining that information instantly.

[1261] "Anomaly detection" is the process of identifying unusual equipment states or behaviors, helping to detect problems early.

[1262] "Trends and keywords" are important terms that appear frequently in documents or data, or information that indicates patterns or trends over time.

[1263] "Customized Format" means a data presentation and layout that is specifically designed and tailored based on the needs and requirements of the User.

[1264] This invention is a system that analyzes daily work reports and equipment status in a factory and distributes the results as summaries and trend information to each department. This system utilizes natural language processing technology and generative models to enable fast and accurate data analysis and anomaly detection.

[1265] Hardware and software used

[1266] To implement the present invention, the following hardware and software are used:

[1267] Hardware: Factory robots, servers

[1268] Software: Python, TensorFlow, Hugging Face Transformers, FastAPI, MySQL

[1269] Program processing overview

[1270] 1. Uploading and Receiving Files

[1271] User: Factory workers and managers upload daily factory work reports through the robot's dedicated interface.

[1272] Terminal: Validates the file through the provided interface and sends it to the server using the HTTPS protocol, during which process it is verified that the file has been uploaded correctly.

[1273] 2. Data Reception and Analysis

[1274] Server: After receiving the uploaded daily report data, the server uses a natural language processing engine (Hugging Face Transformers) to tokenize and analyze the parts of speech of the daily report content, and then uses a sentiment analysis engine to analyze emotions such as positive, negative, and neutral.

[1275] 3. Summary Generation and Trend Extraction

[1276] Server: The parsed data is passed to a generative model (TensorFlow) to generate summaries, and then text mining techniques (TF-IDF) are used to extract important keywords and trends.

[1277] 4. Condition Monitoring and Anomaly Detection

[1278] Robot: Monitors the status of equipment in the factory in real time and compares it with analyzed daily data. If an abnormality is detected, the information is quickly sent to a server and notified to each department along with trend information.

[1279] 5. Report distribution

[1280] Server: Integrates the generated summary and trend information, generates reports in a format suitable for each department's terminal, and immediately distributes the reports to each department, where they are acknowledged for receipt.

[1281] Specific examples

[1282] Example 1: Suppose a factory daily report contains the following statement: "Today's work status of machine A is reported. No abnormalities." When the system analyzes this daily report, it extracts the keywords "machine A," "work status," and "no abnormalities," and generates a summary statement: "The work status of machine A is normal." This is then distributed to each department along with related trend information.

[1283] Example prompt sentence:

[1284] "Analyze the following daily report for key keywords, summaries, and sentiment analysis:

[1285] I'll report on the work status of Machine A today. No abnormalities."

[1286] This invention enables automatic and efficient data analysis and anomaly detection in factories, enabling rapid troubleshooting. It also enables each department to make efficient decisions based on the generated summary and trend information.

[1287] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1288] Step 1:

[1289] Uploading and Receiving Files

[1290] The user uploads the factory's daily work report through the robot's dedicated interface, selects a specific file, and presses the send button.

[1291] Input: Factory work daily report file

[1292] Output: Uploaded files are sent to the device.

[1293] Specific operation: Use the file selection dialog on the user interface to select a file and click the Send button. The device will send the file to the server using the HTTPS protocol.

[1294] Step 2:

[1295] Data reception and analysis

[1296] The server receives the uploaded daily report data and uses a natural language processing engine (Hugging Face Transformers) to tokenize and analyze the parts of speech of the daily report content.

[1297] Input: Uploaded daily report data

[1298] Output: Tokenized and part-of-speech analyzed data

[1299] Specific operation: The server decodes the received data into UTF-8 and converts it into text format. This is then passed to a natural language processing engine, which tokenizes the document and analyzes the parts of speech.

[1300] Step 3:

[1301] Emotion analysis

[1302] The server passes the tokenized data to a sentiment analysis engine, which analyzes the sentiment for positive, negative, neutral, etc.

[1303] Input: Tokenized and part-of-speech analyzed data

[1304] Output: Emotion analysis results

[1305] How it works: The parsed data is fed into a sentiment analysis engine, which calculates a sentiment score for each sentence or phrase. Based on this, the sentiment of the entire document is evaluated.

[1306] Step 4:

[1307] Summary Generation

[1308] The server uses a generative model (TensorFlow) to generate a summary based on the analysis results.

[1309] Input: Sentiment analysis results and tokenized data

[1310] Output: Generated summary

[1311] How it works: The sentiment analysis results are combined with the tokenized data and passed to a generative model, which extracts key information and generates a summary.

[1312] Step 5:

[1313] Trend and Keyword Extraction

[1314] The server uses text mining technology (TF-IDF) to extract important keywords and trends.

[1315] Input: Tokenized and part-of-speech analyzed data

[1316] Output: Extracted keywords and trend information

[1317] Specific operation: Text data is input into the TF-IDF algorithm to calculate and extract important keywords and trends that frequently appear in the document.

[1318] Step 6:

[1319] Condition Monitoring and Anomaly Detection

[1320] The robot monitors the status of equipment in the factory in real time and detects abnormalities by comparing it with analyzed daily data.

[1321] Input: Real-time monitoring data and daily report analysis results

[1322] Output: Anomaly detection notification

[1323] Specific operation: The monitored data is analyzed in real time and compared with the saved daily data. If an abnormal value is detected, an alert is sent to the server.

[1324] Step 7:

[1325] Report Distribution

[1326] The server integrates the generated summary and trend information, generates and distributes reports in a format suitable for each department's terminal.

[1327] Input: Generated abstracts, keywords and trend information

[1328] Output: Reports delivered to departments

[1329] Specific operation: The generated summaries, keywords, and trend information are formatted and a report is created in a format that matches the settings of each department. This is then distributed to each department's terminal and receipt is confirmed.

[1330] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1331] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1332] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1333] [Fourth embodiment]

[1334] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1335] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1336] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1337] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1338] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1339] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1340] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1341] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1342] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1343] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1344] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1345] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1346] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1347] Overall system overview

[1348] This system analyzes business daily reports and information uploaded by users on a server, automatically generates summaries and trend information, and distributes them to each department, thereby streamlining information sharing. Specifically, it combines natural language processing technology, generative models, and keyword and trend extraction algorithms to achieve fast and accurate information analysis and distribution.

[1349] Program processing flow and detailed explanation

[1350] Uploading and Receiving Files

[1351] User: Upload daily reports and business information files through the device interface.

[1352] On your device: Automatically validates the file format and size to ensure it is safe before sending it to the server.

[1353] Calling the API for receiving data

[1354] Terminal: Call the API endpoint for sending a file and send the file contents to the server.

[1355] Server: Stores the received file in temporary storage and then passes it to the natural language processing module.

[1356] Analysis using natural language processing

[1357] Server: Using a natural language processing engine, the uploaded document is broken down into sentences and words, tokenized, and analyzed for parts of speech to determine the meaning and relationships of each word.

[1358] Server: Prepares the analysis results in a data frame format and passes them to the summary generation module.

[1359] Generate a summary

[1360] Server: Uses a generative model to generate a summary from the analysis results, evaluating the importance of each sentence to ensure that important information is not omitted.

[1361] Server: Formats the generated summary and presents it in a format that is easy for users to understand.

[1362] Keyword and trend extraction

[1363] Server: Uses text mining technology to extract important keywords and trends from the analysis results.

[1364] Server: Adds extracted keywords and trends to a summary and creates the final report.

[1365] Report distribution

[1366] Server: Prepares the generated reports for distribution to each department, customizing the format based on user settings.

[1367] Server: Distributes reports to each department's terminal and confirms receipt.

[1368] View and share your results

[1369] Terminal: Displays received reports to users, allowing each department to quickly access the information they need.

[1370] Users: Review reports and share them with other employees or departments as needed, supporting efficient information sharing and quick decision-making.

[1371] Specific examples

[1372] Sales department daily report analysis and summary

[1373] 1. User: Upload the sales report (e.g. sales_report_2023_10_15.pdf) to the system from the terminal.

[1374] 2. Terminal: Validate the file and send it to the server.

[1375] 3. Server: Receives the daily reports and performs analysis using a natural language processing engine.

[1376] 4. Server: Based on the analysis results, a generative model creates a summary, summarizing key selling points and customer trends.

[1377] 5. Server: Delivers reports containing summary and trend information to sales department terminals.

[1378] 6. Terminal: Receives the report and displays it for the sales team to view.

[1379] 7. User: Review the reports and make decisions that will influence your sales strategy.

[1380] This system allows users to efficiently analyze the large amount of information generated in daily business operations and use it to make quick decisions.

[1381] The processing flow will be explained below.

[1382] Step 1:

[1383] User: Uploads daily reports and business information files using the device interface. Through the file selection dialog, select the specific file and press the send button.

[1384] Step 2:

[1385] Terminal: Validates the selected file, checking that the file format is supported and that the size is within the regulated range. If there are no problems, generates an API request to send the file to the server.

[1386] Step 3:

[1387] Terminal: Sends the generated API request to the server's API endpoint, usually using a secure protocol such as HTTPS.

[1388] Step 4:

[1389] Server: Received files are stored in temporary storage, where file metadata and other information are recorded and prepared for subsequent processing.

[1390] Step 5:

[1391] Server: Calls the natural language processing (NLP) engine to analyze the contents of the uploaded file. First, it breaks down the file into sentences and words, and performs tokenization and part-of-speech analysis.

[1392] Step 6:

[1393] Server: Stores the results of tokenization and part-of-speech analysis. Specifically, it stores the analysis results in data frame format and passes them to the next summary generation process.

[1394] Step 7:

[1395] Server: Uses a generative model for summary generation to generate a summary from the analysis results. Extracts important points from the document and summarizes the information.

[1396] Step 8:

[1397] Server: Evaluates the generated summary and formats it in an appropriate format. Here, the summary is organized into a concise and easy-to-read format so that users can quickly grasp the information.

[1398] Step 9:

[1399] Server: Along with the summary, we use text mining techniques to extract important keywords and trends. We use TF-IDF and word clouds to extract highly relevant information.

[1400] Step 10:

[1401] Server: Adds the extracted results to the summary and creates a final report, which includes the summary, keywords, and trend information.

[1402] Step 11:

[1403] Server: Prepares the created report for distribution to each department, customizes the format according to the user's settings, and generates a distribution list.

[1404] Step 12:

[1405] Server: Delivers reports to terminals in each department. Sends reports to designated recipients using a distribution protocol.

[1406] Step 13:

[1407] Terminal: Stores received reports and displays them in the user interface, allowing users to easily view and review the reports.

[1408] Step 14:

[1409] Users: Check the displayed report and share it with other employees or departments as needed, including by email or by sending a link to an internal chat tool.

[1410] This series of steps allows users to quickly and accurately analyze business information, obtain summary and trend information, and make effective decisions.

[1411] Example 1

[1412] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1413] In today's business environment, a huge amount of information is generated every day, creating a need to analyze and summarize that information quickly and accurately, and share it efficiently. However, traditional methods require a significant amount of time and effort to analyze and share information. Furthermore, manual information analysis is prone to errors and carries a high risk of important information being overlooked. Furthermore, the wide variety of information formats makes it difficult to analyze information in a unified and efficient manner. There is a need for a system that solves these problems.

[1414] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1415] In this invention, the server includes: means for a user to upload documents to the system through an electronic device interface; means for a terminal to verify the format and size of the uploaded documents and transmit them to the server after confirming their safety; means for the server to call a data reception API to receive and temporarily store the files; means for the server to use natural language processing technology to break down the uploaded documents into sentence units and words and perform tokenization and part-of-speech analysis; means for the server to generate the analysis results in a data frame format and pass them to a summary generation model; means for the server to use the generation model to generate summaries from the analysis results; means for the server to extract important keywords and trends and add them to the summaries to create a final report; means for the server to distribute the generated summaries and the extracted trends and keywords to terminals in each department; and means for the terminal to display the received reports to the user and enable information to be shared with other departments, thereby enabling efficient analysis of information and rapid sharing.

[1416] "User" refers to the entity that uses the system to upload materials and receive analysis results.

[1417] "Electronic Device" means any device that allows a User to manipulate and upload materials to the System, including, for example, a computer, smartphone, or tablet.

[1418] An "interface" is a point of contact between a user and a system for exchanging information. Examples include web browsers and dedicated applications.

[1419] "Documents" refers to business daily reports and other written information uploaded to the system, including, for example, PDF, Excel, Word files, etc.

[1420] "Terminal" means an electronic device operated by a User that verifies the format and size of uploaded materials and ensures their security.

[1421] "Server" refers to the computer system that receives uploaded materials, generates analysis and summaries, and distributes them to each department.

[1422] "Data Reception API" refers to a program interface for sending materials from a terminal to a server.

[1423] "Temporary storage" refers to the temporary storage of received materials by a server, including, for example, cloud storage and databases.

[1424] "Natural language processing technology" is a technology for analyzing uploaded documents, and is primarily a function for tokenization and part-of-speech analysis.

[1425] "Tokenization" refers to the process of breaking down a document into words or sentences.

[1426] "Part-of-speech analysis" refers to the process of identifying the part of speech of each word in a document and analyzing its grammatical relationships.

[1427] "Data frame format" refers to a data structure for storing analysis results in a table-like format.

[1428] A "summary generation model" refers to an algorithm or machine learning model for generating a summary based on the analysis results.

[1429] "Keywords" refer to words that appear frequently in a document and have important meanings.

[1430] "Trends" refers to major themes or tendencies extracted from the information in a document.

[1431] The "final report" refers to a document that compiles the generated summary, keywords, and trend information.

[1432] "Delivery" refers to the process of sending the generated final report to the terminals of each department.

[1433] "Receiving" refers to the terminal receiving distribution from the server.

[1434] "Display" refers to the terminal visually indicating to the user the report received.

[1435] "Sharing" refers to the act of a user collaborating and communicating the received report with other employees or departments.

[1436] MODE FOR CARRYING OUT THE INVENTION

[1437] This invention is a system that improves the efficiency of information sharing by analyzing business daily reports and information uploaded by users on a server, automatically generating summaries and trend information, and distributing them to each department. The system includes the following main elements:

[1438] Uploading and Receiving Files

[1439] Users upload files such as daily reports and business information through the interface of their electronic device. Specifically, they use the browser's file selection dialog to select a document such as "sales_report_2023_10_15.pdf" and click the upload button. The device then verifies the file's format and size (e.g., PDF, Excel, Word), confirms its security, and sends it to the server.

[1440] Calling the API for receiving data

[1441] The device calls an API endpoint (e.g., https: / / api.example.com / upload) to upload the file to the server. The server temporarily stores the file received at the API endpoint (e.g., Amazon S3 or Google Cloud Storage) and passes the path to the natural language processing engine.

[1442] Analysis using natural language processing

[1443] The server uses a natural language processing (NLP) engine (e.g., SpaCy, NLTK) to break down the uploaded document into sentences and words. Specifically, it performs tokenization, part-of-speech analysis, and removes stop words. The analysis results are formatted as a data frame (e.g., Pandas DataFrame) and passed to the summary generation model.

[1444] Generate a summary

[1445] The server uses a generative AI model (e.g., GPT-3, BERT) to generate a summary from the analysis results. The summary is generated by extracting particularly important sentences based on their importance scores and assembling them into a sentence. The generated summary is then organized in a format that is easy for users to understand (e.g., Markdown or HTML).

[1446] Keyword and trend extraction

[1447] The server uses text mining techniques (e.g., TF-IDF, Word2Vec) to extract important keywords and trends from the analysis results, including frequency analysis and semantic relevance assessment. The extracted keywords and trend information are added to a summary to create the final report.

[1448] Report distribution

[1449] The server prepares the generated reports for distribution to each department, customizing the reports based on user settings and departmental formatting requirements, and delivering the reports via email or the company's internal portal, where receipt is acknowledged.

[1450] View and share your results

[1451] The device displays the received report to the user, for example, through a web interface or desktop notification. The user can review the report's contents and share the link in an internal chat tool to share key selling points with other departments.

[1452] Prompt Sentence Examples

[1453] Below are some example input prompts for the generative AI model that can be used in this system:

[1454] "Today, we uploaded the sales report for October 15, 2023. Please extract trending keywords and summarize them. Don't forget to include your key selling points."

[1455] This system allows users to efficiently analyze the large amount of information generated in daily business operations and use it to make quick decisions.

[1456] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1457] Step 1:

[1458] The user uploads the document using an electronic device. The user opens a file selection dialog using the interface of their device, selects the daily report or business information file (e.g., "sales_report_2023_10_15.pdf"), and clicks the upload button. The input is the file selected by the user, and the output is the file selection information sent to the device.

[1459] Step 2:

[1460] The terminal verifies the format and size of the uploaded file to confirm safety. The terminal checks the file format (e.g. PDF, Excel, Word) and whether the file size exceeds the specified limit. It also performs a virus check. The input is the file uploaded by the user, and the output is the verification result (safety confirmation).

[1461] Step 3:

[1462] The device sends the verified file to the server. The device calls an API endpoint (e.g., https: / / api.example.com / upload) to upload the verified file to the server via an HTTP POST request. The input is the verified file, and the output is a notification that the file was successfully uploaded.

[1463] Step 4:

[1464] The server temporarily stores the received file. The server stores the file received at the API endpoint in temporary storage (e.g., Amazon S3 or Google Cloud Storage) and records the path. The input is the file sent from the device, and the output is the path to the temporary storage.

[1465] Step 5:

[1466] The server analyzes the file using a natural language processing (NLP) engine. The server passes the saved file to an NLP engine (e.g., SpaCy, NLTK) and breaks it down into sentences and words. It performs tokenization, part-of-speech analysis, and removes stop words, and then organizes the analysis results in data frame format (e.g., Pandas DataFrame). The input is the path to the saved file, and the output is a data frame of the analysis results.

[1467] Step 6:

[1468] The server generates a summary using a generative AI model. After tokenization and part-of-speech analysis, the server passes the data frame to a generative AI model (e.g., GPT-3, BERT) to generate a summary. It extracts particularly important sentences based on their importance scores and assembles them into a summary. The input is the analysis result data frame, and the output is the generated summary.

[1469] Step 7:

[1470] The server uses text mining techniques to extract keywords and trends. The server uses TF-IDF models and Word2Vec to extract important keywords and trends from the analysis results. This includes word frequency analysis and semantic relevance assessment. The input is a data frame of the analysis results, and the output is the extracted keywords and trends.

[1471] Step 8:

[1472] The server creates a final report and adds keywords and trends to the summary. The server adds keyword and trend information to the generated summary and formats the final report. The report is prepared in a format that is easy for users to understand (e.g., Markdown or HTML). The input is the generated summary and extracted keywords and trends, and the output is the final report.

[1473] Step 9:

[1474] The server prepares the final report for delivery to each department. The server customizes the report based on user settings and departmental formatting requirements. The server delivers the report via email or through the company's internal portal. The input is the final report and the output is the customized report.

[1475] Step 10:

[1476] The server distributes the report and the terminal confirms receipt. The server distributes the customized report to each department's terminal and receives a receipt confirmation. The input is the customized report and the output is the receipt confirmation result.

[1477] Step 11:

[1478] The terminal displays the received reports to the user. The terminal makes the reports available to the user through a web interface or desktop notification. The input is the received report and the output is the report displayed to the user.

[1479] Step 12:

[1480] The user reviews the report and shares it with other departments. The user reviews the received report and shares important selling points and analysis results with other employees and departments using an internal chat tool, etc. The input is the report displayed to the user, and the output is the results shared with other departments.

[1481] This is the flow of processing for this system's program, which enables efficient analysis and rapid sharing of information.

[1482] (Application example 1)

[1483] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1484] Conventional autonomous driving systems have had difficulty efficiently analyzing massive amounts of sensor data and operational information, summarizing it, and providing it to drivers and managers. This can lead to delays in quickly identifying important trends and operational issues and taking appropriate measures. Furthermore, there is no established method for displaying or distributing the generated data, which can lead to delays in information sharing and decision-making.

[1485] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1486] In this invention, the server includes means for analyzing uploaded documents using natural language processing technology, means for using a generative model to generate summaries based on the analysis results, means for distributing the generated summaries and extracted trends and keywords to terminals in each department, means for uploading sensor data and operation information and analyzing and summarizing them, and means for extracting important information from the operation data of the autonomous vehicle and displaying it to the driver or manager. This allows the operation information of the autonomous vehicle to be analyzed efficiently and quickly, summarized, and provided, making it possible to immediately grasp important trends and issues and respond promptly.

[1487] "Natural language processing technology" is a technology that automatically analyzes and understands text data and processes human language using a computer.

[1488] A "generative model" is a machine learning algorithm that generates new data based on given data.

[1489] A "trend" refers to the tendency of fluctuations or patterns in data or information over a period of time.

[1490] "Keywords" refer to words or phrases that have particular significance in documents or data.

[1491] "Sensor data" refers to measurement data such as temperature, pressure, position, and speed obtained from various sensors.

[1492] "Operation information" is data relating to the operation status of a vehicle, including location information, speed, route, traffic conditions, and the like.

[1493] An "autonomous vehicle" refers to a vehicle that drives autonomously without the need for human operation.

[1494] An "abstract" is a short sentence or text that succinctly summarizes the important points of the original document or data.

[1495] An "API endpoint" refers to a specific address or operation that allows software to communicate with other software.

[1496] "Customized format" means data or reports that have been formatted in a specific way according to the user's requirements or purposes.

[1497] A "prompt sentence" is text containing instructions or questions to be provided as input to a generative model.

[1498] The system for implementing this invention analyzes autonomous vehicle operation data and sensor data uploaded by users on a server, automatically generates summaries and trend information, and distributes them to drivers and managers. A detailed description of the system is as follows.

[1499] Program processing overview

[1500] Uploading and Receiving Files

[1501] Users upload operational data for autonomous vehicles via their smartphones or on-board displays. The device automatically verifies the file format and size, confirms that it is safe, and then sends it to the server.

[1502] Calling the API for receiving data

[1503] The device calls the API endpoint for sending a file and sends the file contents to the server, which then stores the received file in temporary storage and passes it to the natural language processing module.

[1504] Analysis using natural language processing

[1505] The server uses a natural language processing engine (e.g., spaCy or NLTK) to break down the uploaded operation logs and error reports into sentences and words. It then performs tokenization and part-of-speech analysis to analyze the meaning and relationships of each word. The analysis results are prepared in data frame format and passed to a summary generation model (e.g., GPT series).

[1506] Generate a summary

[1507] The server uses a generative model to generate a summary from the analysis results. During this process, it evaluates the importance of each sentence to ensure that important information is not omitted. The generated summary is then formatted and presented in a format that is easy for users to understand.

[1508] Keyword and trend extraction

[1509] The server uses text mining techniques to extract important keywords and trends from the analysis results, adds them to a summary, and creates a final report.

[1510] Report distribution

[1511] The server prepares the generated report for distribution to the driver or manager, customizes the report format based on the user's settings, and delivers the report to a smartphone or in-vehicle display for confirmation of receipt.

[1512] View and share your results

[1513] The terminal displays the received reports to the user, allowing them to quickly access the information they need. Users can then review the reports and share them with other drivers and fleet managers, supporting efficient information sharing and quick decision-making.

[1514] Specific examples

[1515] We will provide a concrete example of uploading log data from a self-driving vehicle's daily operations, analyzing it in real time, and extracting trends and important issues.

[1516] Example prompt for a generative AI model:

[1517] "Please extract important information from the operational data log below and generate a written summary."

[1518] Data Log:

[1519] This system will enable drivers and managers of autonomous vehicles to efficiently analyze and summarize data, enabling them to make quick decisions.

[1520] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1521] Step 1:

[1522] Uploading and Receiving Files

[1523] Users use their smartphones or in-vehicle displays to upload operational and sensor data from their autonomous vehicles. The device automatically validates the file format and size to ensure the content is safe. The input data is an operational data file, and the output is a file ready to be sent to the server. Specifically, the device checks the file's header information to ensure it is in an acceptable format and size.

[1524] Step 2:

[1525] Calling the API for receiving data

[1526] The terminal calls the API endpoint for file transmission and sends the contents of the operation data file to the server. The input data is the verified operation data file, and the output is a file stored in the server's temporary storage. Specifically, the terminal generates an API request, includes the file data in the request body, and sends it to the server. The server stores the received data in temporary storage.

[1527] Step 3:

[1528] Analysis using natural language processing

[1529] The server uses a natural language processing engine (e.g., spaCy or NLTK) to analyze the received operation data file. The input data is the operation data file stored in temporary storage, and the output is a data frame that has been tokenized and analyzed for parts of speech. Specifically, the server reads the file, tokenizes it into sentences and words, analyzes the parts of speech for each, and converts it into a data frame format.

[1530] Step 4:

[1531] Generate a summary

[1532] The server generates a summary using a generative model (e.g., GPT series) based on the analysis results. The input data is a data frame of the analysis results, and the output is a summary. Specifically, the server passes the data frame as input to the generative model, and generates a summary while evaluating important information.

[1533] Step 5:

[1534] Keyword and trend extraction

[1535] The server uses text mining technology to extract important keywords and trends from the analysis results. The input data is a summary sentence and a data frame of the analysis results, and the output is important keywords and trend information. Specifically, the server applies text mining technology to detect frequently occurring keywords and trend patterns.

[1536] Step 6:

[1537] Report creation and distribution

[1538] The server adds the extracted keywords and trends to the summary and creates a final report. The input data is the summary, keywords, and trend information, and the output is a completed report. Specifically, the server integrates the keywords and trend information into the summary and generates a report in a customized format based on user settings. The final report is then distributed to the driver's or manager's device.

[1539] Step 7:

[1540] View and share your results

[1541] The terminal displays the received report to the user. The input data is the delivered report, and the output is the information displayed to the user. Specifically, the terminal analyzes the report data and displays it in a visually easy-to-understand format. The user can review the report and share it with other drivers or fleet managers.

[1542] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1543] Overall system overview

[1544] This system analyzes daily business reports and information uploaded by users on a server, automatically generates summaries and trend information, and distributes them to each department. Furthermore, by combining it with an emotion engine, it is possible to analyze the sentiment of documents and reflect this in summaries and trend information. This allows users to analyze information while taking into account the emotional aspects of documents, enabling them to make quick and accurate decisions.

[1545] Program processing flow and detailed explanation

[1546] Uploading and Receiving Files

[1547] User: Uploads daily reports and business information files through the device interface, selects a specific file through the file selection dialog, and presses the send button.

[1548] Calling the API for receiving data

[1549] Terminal: Call the API endpoint for sending the file and send the file contents to the server. The sending protocol is usually a secure one such as HTTPS.

[1550] Analysis using natural language processing

[1551] Server: Calls the natural language processing (NLP) engine to analyze the contents of the uploaded file. First, it breaks the file down into sentences and words, then performs tokenization and part-of-speech analysis.

[1552] Server: Stores the results of tokenization and part-of-speech analysis. Specifically, the analysis results are stored in data frame format and passed to the next summary generation process.

[1553] Emotion analysis

[1554] Server: Calls the sentiment engine to analyze the sentiment of the uploaded document, identifying positive and negative phrases and words in the text and assessing the type and intensity of the sentiment.

[1555] Server: Integrates the analysis results into a summary generation model to generate summaries that include emotional information.

[1556] Generate a summary

[1557] Server: Generates a summary from the analysis results using a generative model, extracting important points from the document and summarizing the information.

[1558] Server: Evaluates the generated summary and formats it appropriately. Here, it is formatted in a concise and easy-to-read format to allow users to quickly grasp the information.

[1559] Keyword and trend extraction

[1560] Server: Uses text mining techniques to extract important keywords and trends from the analysis results, and uses TF-IDF and word clouds to extract highly relevant information.

[1561] Server: Adds extracted keywords and trends to a summary and creates a final report.

[1562] Report distribution

[1563] Server: Prepares the generated reports for distribution to each department, customizes the format according to the user's settings, and generates the distribution list.

[1564] Server: Distributes reports to each department's terminal and confirms receipt.

[1565] View and share your results

[1566] Terminal: Stores the received reports and displays them in the user interface, allowing users to easily view and check the reports.

[1567] Users: Review the displayed report and share it with other employees or departments as needed, including by email or by sending a link to an internal chat tool.

[1568] Specific examples

[1569] Sales department daily report analysis and summary

[1570] 1. User: Upload the sales report (e.g. sales_report_2023_10_15.pdf) to the system from the terminal.

[1571] 2. Terminal: Validate the file and send it to the server.

[1572] 3. Server: Receives the daily reports and performs analysis using a natural language processing engine and sentiment engine.

[1573] 4. Server: Based on the analysis results, a generative model creates a summary, summarizing key selling points, customer trends, and sentiment information.

[1574] 5. Server: Delivers reports containing summaries, trend information, and sentiment information to sales department terminals.

[1575] 6. Terminal: Receives the report and displays it for the sales team to view.

[1576] 7. User: Review the reports and make decisions that will influence your sales strategy.

[1577] This system enables users to efficiently analyze the large amount of information generated in daily business operations, quickly grasp summary and trend information, as well as sentiment information, and make effective decisions.

[1578] The processing flow will be explained below.

[1579] Step 1:

[1580] User: Uploads daily reports and business information files using the device interface. Through the file selection dialog, select the specific file and press the send button.

[1581] Step 2:

[1582] Terminal: Validates the selected file, checking whether the file format is supported and whether the size is within the regulated range. If there are no problems, generates an API request to send the file to the server.

[1583] Step 3:

[1584] Terminal: Sends the generated API request to the server's API endpoint, usually using a secure protocol such as HTTPS.

[1585] Step 4:

[1586] Server: Received files are stored in temporary storage, where file metadata and other information are recorded and prepared for subsequent processing.

[1587] Step 5:

[1588] Server: Calls the natural language processing (NLP) engine to analyze the contents of the uploaded file. First, it breaks the file down into sentences and words, then performs tokenization and part-of-speech analysis.

[1589] Step 6:

[1590] Server: Stores the results of tokenization and part-of-speech analysis. Specifically, it stores the analysis results in data frame format and passes them to the next summary generation process.

[1591] Step 7:

[1592] Server: Calls the sentiment engine to analyze the sentiment of the uploaded document, identifying positive and negative phrases and words in the text and assessing the type and intensity of the sentiment.

[1593] Step 8:

[1594] Server: Stores the analyzed emotion data and integrates it into the summary generation model, thereby generating summaries that include emotion information.

[1595] Step 9:

[1596] Server: Using a generative model for summary generation, a summary is generated based on the analysis results and sentiment data. Important points of the document are extracted and the information is summarized.

[1597] Step 10:

[1598] Server: Evaluates the generated summary and formats it in an appropriate format. Here, it is organized into a concise and easy-to-read format so that users can quickly grasp the information.

[1599] Step 11:

[1600] Server: Uses text mining techniques to extract important keywords and trends from the analysis results. Uses TF-IDF and word clouds to extract highly relevant information.

[1601] Step 12:

[1602] Server: Adds the extracted keywords and trends to a summary and creates a final report, including sentiment information.

[1603] Step 13:

[1604] Server: Prepares the created report for distribution to each department, customizes the format according to the user's settings, and generates a distribution list.

[1605] Step 14:

[1606] Server: Delivers reports to terminals in each department. Sends reports to designated recipients using a distribution protocol.

[1607] Step 15:

[1608] Terminal: Stores received reports and displays them in the user interface, allowing users to easily view and check the reports.

[1609] Step 16:

[1610] Users: Check the displayed report and share it with other employees or departments as needed. Sharing methods include email and internal chat tool links.

[1611] This detailed processing step enables users to efficiently analyze the large amount of information generated in daily business operations, quickly grasp summary and trend information, as well as sentiment information, and make effective decisions.

[1612] Example 2

[1613] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1614] In today's business environment, a huge amount of business documents and reports are generated every day, and there is a need to efficiently analyze them and quickly grasp important information. Conventional methods require manually analyzing document content and creating summaries, which is time-consuming and labor-intensive. Furthermore, it is difficult to appropriately evaluate and reflect the emotional information contained in documents. This leads to issues such as overlooking information and delays in identifying important trends.

[1615] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: a means for a user to upload a file through an interface; a means for a terminal to call an API endpoint and send the contents of the file; a means for the server to analyze the uploaded document using natural language processing technology; a means for analyzing the sentiment of the document using an emotion engine; a means for using a generative model to generate a summary based on the analysis results; a means for distributing the generated summary and extracted trends and keywords to terminals in each department; and a means for saving and displaying reports received by the terminals. This makes it possible to efficiently analyze the content and sentiment information of business documents and to quickly and accurately generate and distribute summaries and trend information.

[1616] "User" refers to the user who uploads business documents and daily reports through the system interface.

[1617] A "terminal" is a device used by a user that has the functionality to upload files and send requests to API endpoints.

[1618] "Server" refers to a central computing device that analyzes received files, generates summaries and trend information, and distributes it to each department.

[1619] "Natural language processing technology" is a technology for analyzing documents, and has the ability to break them down into sentences or words and perform tokenization and part-of-speech analysis.

[1620] An "API endpoint" is a specific URL or URI that a device uses to communicate with a server and is intended to receive HTTPS requests.

[1621] The "Sentiment Engine" is an analytical tool for identifying emotive phrases in a document and assessing positive and negative sentiment.

[1622] A "generative model" is an algorithm or AI model used to generate a summary of a document based on the results of its analysis.

[1623] "Tokenization" is the process of dividing a document into smaller units such as words or phrases.

[1624] "Part-of-speech analysis" is the process of identifying the part of speech (noun, verb, adjective, etc.) for each tokenized word or phrase.

[1625] A "data frame" is a data structure for storing analysis results, and is tabular data represented by rows and columns.

[1626] "Keywords" are words or phrases that are considered to be particularly important within a document.

[1627] "Trend information" refers to significant trends and patterns extracted from analyzed documents.

[1628] "Report" means the final output document containing the generated summary and extracted keyword trend information.

[1629] "Format" refers to the style and structure in which a report or summary is presented.

[1630] "Distribution" is the process of sending the generated report from the server to the terminals of each department.

[1631] "Display" refers to presenting the report received at the terminal on the user interface.

[1632] A "department" is a group responsible for a specific task or function in a company or organization.

[1633] In the system according to the present invention, a server analyzes business documents and daily reports uploaded by users and automatically generates summaries and trend information. Furthermore, an emotion engine can be used to analyze the emotional information of the documents and reflect this in the generated summaries and trend information. The configuration and operation of the system are described in detail below.

[1634] First, the user uploads a business document or daily report (e.g., sales_report_2023_10_15.pdf) using the terminal by selecting the appropriate file from the file selection dialog and clicking the upload button.

[1635] The device then sends the selected file to an API endpoint (e.g., "https: / / example.com / upload") using the HTTPS protocol. This request is sent as multipart form data.

[1636] The server inputs the received file into a natural language processing engine, which tokenizes and analyzes parts of speech at the sentence and word level. The analysis results are stored in a data frame format using the pandas library. The sentiment engine then evaluates the sentiment within the document based on the analysis results. It identifies positive and negative phrases and quantifies the strength of the sentiment.

[1637] The generative AI model generates a summary based on the analysis results and sentiment information. For example, it generates a summary such as, "Sales increased 15% this month, but there were many returns for a specific product group." The generated summary is formatted in Markdown or HTML for easy reading.

[1638] The server then uses text mining techniques (e.g., TF-IDF and word cloud) to extract important keywords and trend information from the analysis results. Keywords such as "sales," "surge," and "returns" are extracted. These keywords and trend information are also incorporated into the summary text to create the final report.

[1639] The server customizes the generated report according to the user's settings and prepares it for distribution to each department. The report is sent to each department's terminal based on the distribution list. This can be done by email using the SMTP protocol or by sending it to an internal notification system via REST API.

[1640] Finally, the terminal saves the received report and displays it in the user interface. The user can then review the displayed report and share it with other employees or departments as needed. This can be done via email or a link to an internal chat tool. A specific example of this is to share a link to the generated report in an internal chat so that everyone can access it.

[1641] In this way, this invention makes it possible to efficiently analyze large volumes of business documents and daily reports, and quickly and accurately generate and distribute summaries, trend information, and sentiment information, enabling users to quickly grasp important business information and make appropriate decisions.

[1642] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1643] Step 1:

[1644] User: Upload the daily report or business information file through the device interface by selecting the appropriate file (e.g., sales_report_2023_10_15.pdf) from the file selection dialog and clicking the upload button.

[1645] Input: File (for example, sales_report_2023_10_15.pdf).

[1646] Output: The file is selected on the device and ready to be uploaded.

[1647] Step 2:

[1648] On the device: After selecting the file, send an HTTPS request to the API endpoint (e.g., "https: / / example.com / upload"), which sends the file to the server as multipart form data.

[1649] Input: The selected file.

[1650] Output: A request to send a file to the server.

[1651] Step 3:

[1652] Server: The received file is input into a natural language processing (NLP) engine, which performs tokenization and part-of-speech analysis on a sentence-by-sentence and word-by-word basis. Specifically, the document is broken down into multiple words and each word is tagged with a part of speech.

[1653] Input: The contents of the uploaded file.

[1654] Output: Tokenized and part-of-speech analyzed data in dataframe format.

[1655] Step 4:

[1656] Server: The analysis results are stored in a data frame format and passed to the emotion engine. Here, the analysis results are converted into a data frame using the pandas library.

[1657] Input: Tokenized and part-of-speech analyzed data.

[1658] Output: Analysis results in data frame format.

[1659] Step 5:

[1660] Server: Uses a sentiment engine to assess the sentiment in the document, identifying positive and negative phrases and quantifying the strength of each sentiment.

[1661] Input: Analysis results in data frame format.

[1662] Output: Sentiment-analyzed data with sentiment scores.

[1663] Step 6:

[1664] Server: Generates summaries using generative AI models that integrate emotional information. They combine key document points and emotional elements to generate concise summaries.

[1665] Input: Sentiment-analyzed data.

[1666] Output: The generated summary.

[1667] Step 7:

[1668] Server: Uses text mining techniques to extract important keywords and trend information, specifically using TF-IDF and word clouds to identify frequently occurring words.

[1669] Input: Sentiment-analyzed data.

[1670] Output: Extracted keywords and trend information.

[1671] Step 8:

[1672] Server: The generated summary is combined with keywords and trend information to create a final report, which is formatted in Markdown or HTML.

[1673] Input: Abstract, keywords, trend information.

[1674] Output: Final report in Markdown or HTML format.

[1675] Step 9:

[1676] Server: Customize the final report based on user settings and generate a distribution list to distribute it to each department. The distribution method can be email using the SMTP protocol or via REST API to the company's notification system.

[1677] Inputs: Final report, distribution list.

[1678] Output: Reports distributed to each department.

[1679] Step 10:

[1680] Terminal: Saves the received reports and displays them in a user interface, specifically in HTML format using the Report Viewer.

[1681] Input: The report received.

[1682] Output: The report displayed in the user interface.

[1683] Step 11:

[1684] Users can review the displayed report and share it with other employees or departments as needed. They can share the link via email or internal chat tools, or download it as a PDF and distribute it.

[1685] Input: The report displayed in the user interface.

[1686] Output: The shared report.

[1687] (Application example 2)

[1688] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1689] There is a need to efficiently manage factory daily work logs and equipment status, and to quickly detect anomalies and provide information on future trends. However, current methods require manual analysis of huge amounts of data, which is time-consuming and labor-intensive, and the accuracy of the analysis results is unstable. Furthermore, real-time monitoring and anomaly detection are difficult, making it difficult to respond quickly when problems occur.

[1690] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1691] In this invention, the server includes means for analyzing uploaded documents using natural language processing technology, means for using a generative model to generate summaries based on the analysis results, means for providing analysis results of equipment status and daily work reports to robots operating in the factory, means for the robots to monitor the equipment status in real time and detect anomalies, and means for distributing the generated summaries and extracted trends and keywords to terminals in each department.This enables automatic and efficient data analysis and anomaly detection, enabling rapid troubleshooting.

[1692] "Natural language processing technology" is a technology that allows computers to understand, analyze, and generate human language, and includes document tokenization, part-of-speech analysis, sentiment analysis, and more.

[1693] A "generative model" is a machine learning algorithm that learns patterns in knowledge and data and generates new data.

[1694] "Equipment status" is information that indicates the current operating state of machines and devices in a factory, and includes parameters such as temperature, vibration, and operating time.

[1695] A "daily work report" is a report that details the work performed and events that occurred on a specific work day.

[1696] "Real-time monitoring" refers to continuously monitoring the status of equipment and systems within a factory and obtaining that information instantly.

[1697] "Anomaly detection" is the process of identifying unusual equipment states or behaviors, helping to detect problems early.

[1698] "Trends and keywords" are important terms that appear frequently in documents or data, or information that indicates patterns or trends over time.

[1699] "Customized Format" means a data presentation and layout that is specifically designed and tailored based on the needs and requirements of the User.

[1700] This invention is a system that analyzes daily work reports and equipment status in a factory and distributes the results as summaries and trend information to each department. This system utilizes natural language processing technology and generative models to enable fast and accurate data analysis and anomaly detection.

[1701] Hardware and software used

[1702] To implement the present invention, the following hardware and software are used:

[1703] Hardware: Factory robots, servers

[1704] Software: Python, TensorFlow, Hugging Face Transformers, FastAPI, MySQL

[1705] Program processing overview

[1706] 1. Uploading and Receiving Files

[1707] User: Factory workers and managers upload daily factory work reports through the robot's dedicated interface.

[1708] Terminal: Validates the file through the provided interface and sends it to the server using the HTTPS protocol, during which process it is verified that the file has been uploaded correctly.

[1709] 2. Data Reception and Analysis

[1710] Server: After receiving the uploaded daily report data, the server uses a natural language processing engine (Hugging Face Transformers) to tokenize and analyze the parts of speech of the daily report content, and then uses a sentiment analysis engine to analyze emotions such as positive, negative, and neutral.

[1711] 3. Summary Generation and Trend Extraction

[1712] Server: The parsed data is passed to a generative model (TensorFlow) to generate summaries, and then text mining techniques (TF-IDF) are used to extract important keywords and trends.

[1713] 4. Condition Monitoring and Anomaly Detection

[1714] Robot: Monitors the status of equipment in the factory in real time and compares it with analyzed daily data. If an abnormality is detected, the information is quickly sent to a server and notified to each department along with trend information.

[1715] 5. Report distribution

[1716] Server: Integrates the generated summary and trend information, generates reports in a format suitable for each department's terminal, and immediately distributes the reports to each department, where they are acknowledged for receipt.

[1717] Specific examples

[1718] Example 1: Suppose a factory daily report contains the following statement: "Today's work status of machine A is reported. No abnormalities." When the system analyzes this daily report, it extracts the keywords "machine A," "work status," and "no abnormalities," and generates a summary statement: "The work status of machine A is normal." This is then distributed to each department along with related trend information.

[1719] Example prompt sentence:

[1720] "Analyze the following daily report for key keywords, summaries, and sentiment analysis:

[1721] I'll report on the work status of Machine A today. No abnormalities."

[1722] This invention enables automatic and efficient data analysis and anomaly detection in factories, enabling rapid troubleshooting. It also enables each department to make efficient decisions based on the generated summary and trend information.

[1723] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1724] Step 1:

[1725] Uploading and Receiving Files

[1726] The user uploads the factory's daily work report through the robot's dedicated interface, selects a specific file, and presses the send button.

[1727] Input: Factory work daily report file

[1728] Output: Uploaded files are sent to the device.

[1729] Specific operation: Use the file selection dialog on the user interface to select a file and click the Send button. The device will send the file to the server using the HTTPS protocol.

[1730] Step 2:

[1731] Data reception and analysis

[1732] The server receives the uploaded daily report data and uses a natural language processing engine (Hugging Face Transformers) to tokenize and analyze the parts of speech of the daily report content.

[1733] Input: Uploaded daily report data

[1734] Output: Tokenized and part-of-speech analyzed data

[1735] Specific operation: The server decodes the received data into UTF-8 and converts it into text format. This is then passed to a natural language processing engine, which tokenizes the document and analyzes the parts of speech.

[1736] Step 3:

[1737] Emotion analysis

[1738] The server passes the tokenized data to a sentiment analysis engine, which analyzes the sentiment for positive, negative, neutral, etc.

[1739] Input: Tokenized and part-of-speech analyzed data

[1740] Output: Emotion analysis results

[1741] How it works: The parsed data is fed into a sentiment analysis engine, which calculates a sentiment score for each sentence or phrase. Based on this, the sentiment of the entire document is evaluated.

[1742] Step 4:

[1743] Summary Generation

[1744] The server uses a generative model (TensorFlow) to generate a summary based on the analysis results.

[1745] Input: Sentiment analysis results and tokenized data

[1746] Output: Generated summary

[1747] How it works: The sentiment analysis results are combined with the tokenized data and passed to a generative model, which extracts key information and generates a summary.

[1748] Step 5:

[1749] Trend and Keyword Extraction

[1750] The server uses text mining technology (TF-IDF) to extract important keywords and trends.

[1751] Input: Tokenized and part-of-speech analyzed data

[1752] Output: Extracted keywords and trend information

[1753] Specific operation: Text data is input into the TF-IDF algorithm to calculate and extract important keywords and trends that frequently appear in the document.

[1754] Step 6:

[1755] Condition Monitoring and Anomaly Detection

[1756] The robot monitors the status of equipment in the factory in real time and detects abnormalities by comparing it with analyzed daily data.

[1757] Input: Real-time monitoring data and daily report analysis results

[1758] Output: Anomaly detection notification

[1759] Specific operation: The monitored data is analyzed in real time and compared with the saved daily data. If an abnormal value is detected, an alert is sent to the server.

[1760] Step 7:

[1761] Report Distribution

[1762] The server integrates the generated summary and trend information, generates and distributes reports in a format suitable for each department's terminal.

[1763] Input: Generated abstracts, keywords and trend information

[1764] Output: Reports delivered to departments

[1765] Specific operation: The generated summaries, keywords, and trend information are formatted and a report is created in a format that matches the settings of each department. This is then distributed to each department's terminal and receipt is confirmed.

[1766] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1767] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1768] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1769] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1770] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1771] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1772] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1773] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1774] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1775] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1776] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1777] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1778] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1779] 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.

[1780] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1781] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1782] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1783] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1784] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1785] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1786] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1787] The following is further disclosed regarding the above embodiment.

[1788] (Claim 1)

[1789] means for analyzing the uploaded document using natural language processing techniques;

[1790] a means for using a generative model to generate a summary based on the analysis results;

[1791] and means for distributing the generated summary and extracted trends and keywords to terminals in each department.

[1792] (Claim 2)

[1793] A means for analyzing the content of the uploaded document based on natural language processing technology, and performing tokenization and part-of-speech analysis;

[1794] A means to generate the analysis results in a data frame format and pass them to the summary generation model;

[1795] and means for delivering the automatically generated summaries to each department in a customized format.

[1796] (Claim 3)

[1797] In addition to the generated summaries, a means of extracting important keywords and trends and including them in the summary results;

[1798] A means to generate reports using the extracted keywords and trends;

[1799] 2. The system according to claim 1, further comprising: means for generating and distributing reports in a format specified by each department.

[1800] "Example 1"

[1801] (Claim 1)

[1802] means for a user to upload material to the system through an interface of an electronic device;

[1803] The terminal verifies the format and size of the uploaded material, and sends it to the server after confirming its safety;

[1804] The server calls the data reception API to receive the file and temporarily store it.

[1805] The server uses natural language processing technology to break down uploaded documents into sentences and words, and performs tokenization and part-of-speech analysis.

[1806] A means for the server to generate the analysis results in a data frame format and pass them to the summary generation model;

[1807] a means for the server to generate a summary from the analysis results using the generative model;

[1808] A means for the server to extract key keywords and trends and add them to the summary to produce the final report;

[1809] A means for the server to distribute the generated summaries and extracted trends and keywords to terminals in each department;

[1810] A means for displaying the report received by the terminal to the user and enabling information to be shared with other departments;

[1811] A system including:

[1812] (Claim 2)

[1813] A means for analyzing the content of the uploaded document based on natural language processing technology, and performing tokenization and part-of-speech analysis;

[1814] A means to generate the analysis results in a data frame format and pass them to the summary generation model;

[1815] A means to distribute automatically generated summaries to departments in customized formats;

[1816] 10. The system of claim 1, comprising:

[1817] (Claim 3)

[1818] In addition to the generated summaries, a means of extracting important keywords and trends and including them in the summary results;

[1819] A means to generate reports using the extracted keywords and trends;

[1820] A means to generate and distribute reports in the format specified by each department,

[1821] 10. The system of claim 1, comprising:

[1822] "Application Example 1"

[1823] (Claim 1)

[1824] means for analyzing the uploaded document using natural language processing techniques;

[1825] a means for using a generative model to generate a summary based on the analysis results;

[1826] A means for distributing the generated summary and extracted trends and keywords to terminals in each department;

[1827] A means to upload, analyze, and summarize sensor data and operational information;

[1828] A means to extract important information from the operation data of autonomous vehicles and display it to drivers and managers,

[1829] A system including:

[1830] (Claim 2)

[1831] A means for analyzing the content of the uploaded document based on natural language processing technology, and performing tokenization and part-of-speech analysis;

[1832] A means to generate the analysis results in a data frame format and pass them to the summary generation model;

[1833] A means to distribute automatically generated summaries to departments in customized formats;

[1834] A means of analyzing operation log data and extracting trend information and important issues,

[1835] means for delivering a report including the analysis results to a terminal in the vehicle;

[1836] 10. The system of claim 1, comprising:

[1837] (Claim 3)

[1838] In addition to the generated summaries, a means of extracting important keywords and trends and including them in the summary results;

[1839] A means to generate reports using the extracted keywords and trends;

[1840] A means to generate and distribute reports in the format specified by each department,

[1841] A means of providing real-time analysis results of autonomous vehicle operation data, and

[1842] A means of extracting key information from operational data using a generative AI model in prompt sentences;

[1843] 10. The system of claim 1, comprising:

[1844] "Example 2: Combining Emotion Engines"

[1845] (Claim 1)

[1846] a means for a user to upload files through the interface;

[1847] A means for the device to call an API endpoint and send the file contents;

[1848] means for the server to analyze the uploaded document using natural language processing techniques;

[1849] means for analyzing the sentiment of the document using a sentiment engine;

[1850] a means for using a generative model to generate a summary based on the analysis results;

[1851] A means for distributing the generated summary and extracted trends and keywords to terminals in each department;

[1852] and means for storing and displaying the reports received by the terminal.

[1853] (Claim 2)

[1854] A means for analyzing the content of the uploaded document based on natural language processing technology, and performing tokenization and part-of-speech analysis;

[1855] A means to generate the analysis results in a data frame format and pass them to the summary generation model;

[1856] a means for integrating the sentiment analysis results to generate a summary;

[1857] and means for delivering the automatically generated summaries to each department in a customized format.

[1858] (Claim 3)

[1859] In addition to the generated summaries, a means of extracting important keywords and trends and including them in the summary results;

[1860] A means to generate reports using the extracted keywords and trends;

[1861] 2. The system according to claim 1, further comprising: means for generating and distributing reports in a format specified by each department.

[1862] "Application example 2 when combining emotion engines"

[1863] (Claim 1)

[1864] means for analyzing the uploaded document using natural language processing techniques;

[1865] a means for using a generative model to generate a summary based on the analysis results;

[1866] A means of providing analysis results of equipment status and daily work reports to robots operating in factories,

[1867] A means for the robot to monitor the status of the equipment in real time and detect abnormalities;

[1868] A means for distributing the generated summary and extracted trends and keywords to terminals in each department;

[1869] A system including:

[1870] (Claim 2)

[1871] A means for analyzing the content of the uploaded document based on natural language processing technology, and performing tokenization and part-of-speech analysis;

[1872] A means to generate the analysis results in a data frame format and pass them to the summary generation model;

[1873] A means for comparing and analyzing the status information of equipment monitored in real time by factory robots with the results of daily report analysis;

[1874] A means to distribute automatically generated summaries to departments in customized formats;

[1875] 10. The system of claim 1, comprising:

[1876] (Claim 3)

[1877] In addition to the generated summaries, a means of extracting important keywords and trends and including them in the summary results;

[1878] a means for providing anomaly detection and future trend information;

[1879] A means to generate and distribute reports in the format specified by each department,

[1880] 10. The system of claim 1, comprising: [Explanation of symbols]

[1881] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for analyzing the uploaded document using natural language processing techniques; a means for using a generative model to generate a summary based on the analysis results; and means for distributing the generated summary and extracted trends and keywords to terminals in each department.

2. A means for analyzing the content of the uploaded document based on natural language processing technology, and performing tokenization and part-of-speech analysis; A means to generate the analysis results in a data frame format and pass them to the summary generation model; and means for distributing the automatically generated summaries to each department in a customized format.

3. In addition to the generated summaries, a means of extracting important keywords and trends and including them in the summary results; A means to generate reports using the extracted keywords and trends; 2. The system according to claim 1, further comprising: means for generating and distributing reports in a format designated by each department.

Images