System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving historical sample reports and current event data from a user; means for pre-processing and converting the received data into a format that is easy to analyze; means for matching the historical sample reports and the current event data to extract similarities and differences; means for instructing a generative AI to generate a draft of a report based on the extracted information; and means for presenting the report draft generated by the generative AI to the user.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] When a system failure occurs in a company or organization, it is necessary to quickly understand the impact, identify the cause, and take measures to prevent recurrence. However, creating these failure reports requires a great deal of effort and time, draining human resources. In particular, the task of generating a report based on new events while referencing past reports is cumbersome and prone to errors. For this reason, there is a need for an efficient system that can generate failure reports quickly and accurately. [Means for solving the problem]

[0005] The present invention includes a means for receiving past sample reports and current event data from a user, preprocessing the received data, and converting it into an easily analyzable format. It also includes a means for comparing the past sample reports with the current event data and extracting similarities and differences. It also includes a means for instructing a generation AI to generate a draft report based on the extracted information. Specifically, the draft report generated by the generation AI includes sections for a failure summary, scope of impact, cause, and countermeasures. In this way, a system is provided that allows users to easily review and efficiently create failure reports.

[0006] "User" means any person or entity that operates the system, enters data, and reviews and modifies reports.

[0007] "Sample Report" refers to a previously created incident report that serves as reference data for comparison with current events.

[0008] "Event data" refers to data that includes information about a currently occurring failure, and is input in the form of a log file, error message, or the like.

[0009] "Means for receiving" refers to the ability to receive user-provided sample reports and event data within the system.

[0010] "Preprocessing means" refers to the function of removing unnecessary information from received data and converting it into a format that is easy to analyze.

[0011] "Matching means" refers to the function of comparing past sample reports with current event data and extracting commonalities and differences.

[0012] "Generative AI" refers to the part of a system that uses artificial intelligence (AI) technology to automatically generate draft reports based on instructions.

[0013] A "draft report" is an incomplete fault report created by the generation AI, which will become the final report after being reviewed and revised by the user.

[0014] "Outline of the fault" is an item that describes basic information about the fault that has occurred, including the details of the fault and the date and time of occurrence.

[0015] "Scope of impact" is an item that describes the impact of the failure that occurred, and includes the range of systems and services that were affected.

[0016] "Cause" is an item that describes the factors that caused the failure, including technical reasons and root causes.

[0017] "Countermeasures" is an item that describes countermeasures taken in response to the failure that occurred and measures to prevent recurrence. [Brief explanation of the drawings]

[0018] [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

[0019] 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.

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

[0021] 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).

[0022] 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.

[0023] 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.

[0024] 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.

[0025] 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."

[0026] [First embodiment]

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

[0028] 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.

[0029] 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).

[0030] 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.

[0031] 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.

[0032] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form 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.

[0033] 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.

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

[0035] 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.

[0036] 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.

[0037] 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.

[0038] 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."

[0039] System Overview

[0040] The present invention is a system for automatically generating fault reports based on user-provided sample reports and current event data. The system includes the following major functions:

[0041] Receiving data

[0042] Data Preprocessing

[0043] Data collation and analysis

[0044] Automatic generation of report drafts

[0045] User confirmation and correction

[0046] Program processing overview

[0047] 1. User Input

[0048] Users log into the system from a terminal and upload past sample reports and current event data.

[0049] 2. Data Preprocessing

[0050] The server receives the uploaded data and performs text extraction and formatting, specifically removing unnecessary line breaks and spaces and converting the data into a format that is easy to analyze.

[0051] 3. Data Collation and Analysis

[0052] The server compares past sample reports with current event data to extract commonalities and differences, including characteristic features such as error codes, the time of the failure, and the scope of the impact.

[0053] 4. Instructions to the generating AI

[0054] Based on the extracted data, the server instructs the AI ​​to generate a draft report, which includes an overview of the problem, its scope of impact, causes, and countermeasures.

[0055] 5. Generate a draft report

[0056] Based on instructions from the server, the generation AI creates a draft report detailing the problem, which is then sent back to the server in a format that can be reviewed and modified by the user.

[0057] 6. Checking and correcting the report

[0058] The user reviews the generated draft report and makes any necessary corrections. Once the final report is complete, the user saves it and shares it with the appropriate departments and stakeholders.

[0059] Specific examples

[0060] 1. User Input

[0061] Example: A user uploads "Sample Report_2022.pdf" and "Fault Log_2023-10-10.txt" to the system.

[0062] 2. Data Preprocessing

[0063] Example: The server extracts text from "Sample Report_2022.pdf" and the necessary log entries from "Error Log_2023-10-10.txt." It then removes unnecessary line breaks and spaces from the extracted data and converts it to JSON format.

[0064] 3. Data Collation and Analysis

[0065] Example: The server compares the sample report from 2022 with the error log from 2023, extracting the differences between the new error code "503" and the past error code "403." At the same time, it identifies commonalities in the scope of impact and occurrence time.

[0066] 4. Instructions to the generating AI

[0067] Example: The server instructs the AI ​​generation system that "An error code 503 occurred on October 10, 2023, causing service C to stop for 30 minutes." This instruction also includes a detailed summary of the error, the scope of its impact, the cause, and countermeasures.

[0068] 5. Generate a draft report

[0069] Example: The generation AI generates a draft report stating, "Server A failed on October 10, 2023. Service C was stopped for 30 minutes due to error code 503. The cause was a memory leak, and as a countermeasure, improved memory management is required."

[0070] 6. Checking and correcting the report

[0071] Example: User reviews the generated draft, corrects technical details and specific terminology to complete the final report, saves the completed report and shares it with the appropriate stakeholders.

[0072] In this way, the system can generate trouble reports efficiently and accurately, reducing the workload of the user.

[0073] The processing flow will be explained below.

[0074] Step 1:

[0075] The user logs in to the system from a terminal and goes to the upload page. The user selects the files of past sample reports and current event data and clicks the upload button.

[0076] Step 2:

[0077] The server receives the uploaded sample reports and event data, which are temporarily stored and prepared for further processing.

[0078] Step 3:

[0079] The server performs a parsing process to extract text from the uploaded sample reports, specifically extracting the required text information from PDFs and other file formats.

[0080] Step 4:

[0081] The server extracts the necessary information from the event data (e.g., log data), specifically by parsing and extracting important log entries such as the target error code and timestamp.

[0082] Step 5:

[0083] The server removes unnecessary line breaks and spaces from the extracted data and converts the data into a format that is easy to parse (e.g., JSON format).

[0084] Step 6:

[0085] The server compares the current event data with past sample reports, extracting commonalities (e.g., error codes, scope of impact) and differences (e.g., new error codes, different scope of impact).

[0086] Step 7:

[0087] Based on the results of the comparison, the server creates instructions for the generation AI to generate a draft report, including an overview of the problem, its scope of impact, causes, and countermeasures.

[0088] Step 8:

[0089] The generation AI generates a draft report based on instructions received from the server. The generation AI automatically generates detailed text for these items and provides feedback.

[0090] Step 9:

[0091] The generated report draft is sent back to the server and presented to the user, who can then check the generated draft through the terminal.

[0092] Step 10:

[0093] The user reviews the draft and makes any necessary corrections, such as changing technical terminology, adding details, or correcting typos.

[0094] Step 11:

[0095] The user saves the final report and then shares it with the appropriate departments and stakeholders, completing the process of generating a problem report.

[0096] Example 1

[0097] 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."

[0098] Creating a failure report manually is extremely time-consuming and prone to errors and oversights. The process of creating a report by comparing and analyzing past reports with current event data is particularly labor-intensive, making it necessary to create reports efficiently and accurately. Furthermore, reports must include a detailed failure summary, scope of impact, cause, and countermeasures, but there is currently a lack of a means to automatically generate standardized reports that include all of these.

[0099] 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.

[0100] In this invention, the server includes means for receiving past sample reports and current event data from a user, means for converting the received data into a format that is easy to analyze by extracting text and formatting, means for comparing the past sample reports and the current event data using natural language processing and machine learning algorithms to extract similarities and differences, means for providing a prompt to the generative AI model to generate a draft report based on the extracted information, and means for presenting the draft report generated by the generative AI model to the user in a format that can be confirmed and modified, thereby enabling the user to generate and modify reports efficiently and accurately.

[0101] "User" refers to a person or organization that uses a terminal to access the system and provide various data.

[0102] "Sample Report" means a document that contains a detailed record of a past event.

[0103] "Event Data" refers to data relating to a particular event that is currently occurring or has recently occurred.

[0104] "Text extraction" refers to the process of extracting character data from a rich text format file.

[0105] "Formatting" refers to the process of removing unnecessary line breaks and spaces from data and converting it into a format that is easier to parse.

[0106] "Natural language processing (NLP)" refers to technology for analyzing text data and understanding its meaning and structure.

[0107] A "machine learning algorithm" refers to a method of learning patterns and rules from large amounts of data and using that knowledge to analyze new data.

[0108] "Matching" refers to the process of finding similarities and differences between different data sets.

[0109] A "generative AI model" refers to an artificial intelligence system that automatically generates text or other data based on instructions it is given.

[0110] A "prompt sentence" refers to a textual input sentence that provides instructions to a generative AI model.

[0111] "Draft Report" refers to the initial version of the report automatically generated by the generative AI model.

[0112] "Format" refers to the particular structure or format of data or documents.

[0113] "Verification" refers to the process in which a user reviews the generated report draft and checks its content.

[0114] "Modification" refers to the user making any necessary changes or additions to the draft report.

[0115] The present invention is a system for automatically generating fault reports based on user-provided sample reports and current event data. The system can be implemented using the following major hardware and software components:

[0116] Hardware and software used

[0117] Terminal: A device used by a user to access the system and enter data. This includes devices such as PCs and smartphones.

[0118] Server: A central computing unit that preprocesses data, analyzes it, and directs generative AI models.

[0119] A text analysis library, such as Apache Tika, is used to extract text from PDF files.

[0120] Use an NLP library such as spaCy to analyze text data and extract the necessary information.

[0121] Generative AI models: Generative AI models such as GPT-4® are used to generate draft reports.

[0122] Data processing and analysis

[0123] 1. User Input

[0124] A user logs in to the system from a terminal and uploads past sample reports and current event data, for example, uploading a report in PDF format and a fault log in text format.

[0125] Specifically, the user uploads "Sample Report_2022.pdf" and "Error Log_2023-10-10.txt" to the system.

[0126] 2. Data Preprocessing

[0127] The server receives the uploaded data and performs text extraction and formatting.

[0128] Specifically, we use Apache Tika to extract text from "SampleReport_2022.pdf", select the required log entries from the TXT file, remove unnecessary line breaks and spaces, and convert the data to JSON format.

[0129] 3. Data Collation and Analysis

[0130] The server uses the extracted data to compare and analyze past sample reports with current event data.

[0131] Specifically, spaCy is used to extract information such as error codes, dates and times, and the extent of the impact, allowing past and present data to be compared to identify commonalities and differences.

[0132] 4. Instructions to the generating AI

[0133] Based on the analysis results, the server instructs the generative AI model to generate a draft report.

[0134] An example of a specific prompt might be, "An error code 503 occurred on October 10, 2023, causing service C to be down for 30 minutes." This instruction also includes a detailed summary of the error, the scope of its impact, the cause, and countermeasures.

[0135] 5. Generate a draft report

[0136] The generative AI model generates a draft report detailing the fault based on instructions from the server, which is then sent back to the server in a format that can be reviewed and modified by the user.

[0137] Specifically, it generates a draft report stating something like, "Server A failed on October 10, 2023. Service C stopped for 30 minutes due to error code 503. The cause was a memory leak, and as a countermeasure, improved memory management is required."

[0138] 6. Checking and correcting the report

[0139] The user reviews the generated draft report, makes any necessary revisions, and finally saves the final report and shares it with the appropriate departments and agencies.

[0140] Specifically, users can access the draft from their devices, edit specific technical information and terminology, and then finalize the report, which is then saved and sent to stakeholders via email or a shared link.

[0141] In this way, the present invention can efficiently and accurately generate trouble reports, significantly reducing the user's workload. This system can be implemented using specific hardware and software, making it applicable to a variety of situations.

[0142] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0143] Step 1: User Input

[0144] A user logs in to the system from a terminal and uploads a past sample report (PDF format) and current event data (TXT format). The input data is "Sample Report_2022.pdf" and "Fault Log_2023-10-10.txt."

[0145] Specific behavior:

[0146] The user enters their username and password on the login page to access the system.

[0147] After logging in, select the "File Upload" option on your dashboard, select the required file and upload it.

[0148] If the upload is successful, the file is sent to the server.

[0149] Step 2: Preprocessing the data

[0150] The server receives the uploaded data, extracts text, and formats it. The input data is PDF and TXT files, and the output data is in an easy-to-parse JSON format.

[0151] Specific behavior:

[0152] The server uses the Apache Tika library to extract text from the PDF file "SampleReport_2022.pdf".

[0153] At the same time, the server selects the necessary log entries from the TXT file "Error Log_2023-10-10.txt".

[0154] Unnecessary line breaks and spaces are removed from this text data, and the formatted text data is converted into JSON format.

[0155] Step 3: Collate and analyze data

[0156] The server uses the extracted data to compare and analyze past sample reports with current event data. The input is text data in JSON format, and the output is information about notable similarities and differences.

[0157] Specific behavior:

[0158] The server uses an NLP library such as spaCy to extract information such as the error code, date and time, and scope of impact from the text data.

[0159] Based on the extracted information, past reports are compared with the current fault log to identify similarities and differences.

[0160] As a result, information such as differences in error codes and commonalities in the scope of impact is generated and passed on to the next step.

[0161] Step 4: Instructions to the Generative AI

[0162] The server instructs the generative AI model to generate a draft report based on the analysis results. The input is information about similarities and differences, and the output is a prompt statement for the generative AI model.

[0163] Specific behavior:

[0164] The server formalizes the analysis results as prompt sentences and provides them to the generative AI model.

[0165] As an example of a specific prompt sentence, a sentence such as "An error code 503 occurred on October 10, 2023, causing service C to be down for 30 minutes" is generated.

[0166] The prompt text includes a detailed overview of the problem, the scope of impact, the cause, and countermeasures.

[0167] Step 5: Generate a draft report

[0168] The generative AI model generates a draft report based on the prompt provided by the server. The input is the prompt to the generative AI, and the output is the draft report.

[0169] Specific behavior:

[0170] The generative AI model receives the prompt and creates a draft report.

[0171] The generated report draft is generated in Markdown, HTML, Word document format, etc. and sent back to the server.

[0172] The draft includes content such as, "Server A failed on October 10, 2023. Service C was stopped for 30 minutes due to error code 503. The cause was a memory leak, and as a countermeasure, improved memory management is required."

[0173] Step 6: Check and correct the report

[0174] The user reviews the generated report draft and makes any necessary corrections. The input is the report draft and the output is the final report.

[0175] Specific behavior:

[0176] The user accesses the draft from a terminal and checks its contents.

[0177] Correct specific technical information and terminology as needed to complete the final report.

[0178] Completed reports are stored within the system and then sent to the appropriate parties via email or shared link.

[0179] In this way, the system can automatically generate trouble reports efficiently and accurately based on user-supplied data.

[0180] (Application example 1)

[0181] 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."

[0182] Conventional factory robot monitoring systems require human intervention when a failure occurs, making it difficult to respond quickly and create accurate reports. Furthermore, manually comparing past failure reports with current event data and generating reports is time-consuming and labor-intensive. These issues can significantly impact factory operational efficiency and product quality, so a system that solves these problems is needed.

[0183] 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.

[0184] In this invention, the server includes means for receiving past sample reports and current event data from a user, means for preprocessing the received data and converting it into an easily analyzable format, means for comparing the past sample reports with the current event data and extracting similarities and differences, means for instructing the generation AI to generate a draft report based on the extracted information, means for presenting the draft report generated by the generation AI to the user, and means for monitoring the operating status of factory robots in real time and automatically generating a report when an abnormality occurs. This makes it possible to respond quickly and accurately to factory robot failures and improve factory operating efficiency.

[0185] "User" means a person or entity that uses the System to provide and query past sample reports and current event data.

[0186] A "Past Sample Report" is a document that contains detailed information about past events or failures.

[0187] "Current event data" refers to data relating to events or failures occurring at the present time.

[0188] "Preprocessing" is the process of deleting unnecessary line breaks and spaces and reformatting the data in order to convert the received data into a format that is easy to analyze.

[0189] "Matching" is the process of comparing past sample reports with current event data to extract similarities and differences.

[0190] "Generative AI" is an artificial intelligence system that automatically generates draft reports based on given data and instructions.

[0191] A "report draft" is a draft of a report automatically generated by the generation AI.

[0192] A "factory robot" is a mechanical device used to perform automated tasks in manufacturing.

[0193] "Operation status" is information that indicates the state in which a factory robot is operating.

[0194] "Real-time monitoring" refers to the continuous observation of a situation occurring at the present time.

[0195] An "abnormality" is a deviation from normal operation or failure of a system or machine.

[0196] A "report" is a formal document containing detailed information, causes, and actions regarding a specific incident or situation.

[0197] This invention is a system that monitors the operating status of factory robots in real time and automatically generates a fault report when an abnormality occurs. This system creates reports with high accuracy and efficiency based on past sample reports and current event data provided by the user.

[0198] 1. System Configuration

[0199] The system includes a server, user terminals, factory robots, generative AI models, and front-end and back-end processing. The server receives data, preprocesses it, collates it, and issues instructions to the generative AI. The user terminal uploads data and checks and edits reports via a smartphone application. The factory robots are monitored for their operating status and provide data in real time.

[0200] 2. Hardware / Software to be used

[0201] Hardware: smartphones, servers, factory robots

[0202] software:

[0203] Server: Flask (web framework), SQLite (database)

[0204] Preprocessing: Text extraction and data formatting using Python, NLTK (Natural Language Toolkit)

[0205] Generative AI: OpenAI (registered trademark) GPT-3 (registered trademark)

[0206] Front-end: HTML / CSS / JavaScript (registered trademark)

[0207] 3. Specific examples of program processing

[0208] Data reception and preprocessing

[0209] Users upload past sample reports and current event data from their devices, and the server receives the data, then uses Python to extract and format the text, removing unnecessary line breaks and spaces, and converting it into a format that is easier to parse.

[0210] Data collation and analysis

[0211] The preprocessed data is compared with past sample reports and current event data to extract commonalities and differences, including characteristic information such as error codes, time of occurrence, and scope of impact.

[0212] Instructions for generative AI models

[0213] The server instructs a generative AI model (e.g., OpenAI GPT-3) to generate a draft report based on the extracted data, including a summary of the problem, its scope of impact, causes, and countermeasures.

[0214] Draft report generation and presentation

[0215] The generative AI model generates a draft report based on the data and prompts received from the server, which is then sent back to the server and presented to the user, who can make any necessary corrections to produce a final fault report.

[0216] Examples of concrete examples and prompts

[0217] When a user uploads "Sample Report_2022.pdf" and "Fault Log_2023-10-10.txt" to the system, the system automatically analyzes these data, extracts commonalities and differences, and generates a draft report. The generated draft contains the following content:

[0218] Example prompt sentence:

[0219] "Based on past incident reports and the current incident log, we will provide the following information. Please use this information to generate a detailed incident report."

[0220] This system will enable rapid and accurate response to factory robot failures, significantly improving factory operational efficiency.

[0221] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0222] Step 1:

[0223] The user uploads past sample reports and current event data from their device. Specifically, the user uses a smartphone application to send "Sample Report_2022.pdf" and "Fault Log_2023-10-10.txt" to the system. The input is a PDF and text file, and these files are saved on the server as output.

[0224] Step 2:

[0225] The data received by the server is preprocessed. Specifically, text is extracted using Python, and unnecessary line breaks and spaces are removed to convert it into a format that is easy to parse. The input is the uploaded PDF and text file, and the output is formatted text data. For example, the text extracted from the PDF and the contents of the text file are converted into JSON format.

[0226] Step 3:

[0227] The server compares past sample reports with current event data to extract commonalities and differences. Specifically, it uses a scripting language (Python) to compare characteristic information such as error codes, the time of the failure, and the scope of impact. The input is preprocessed text data, and the output is data containing commonalities and differences. For example, it extracts the differences between "error code 503" and "error code 403."

[0228] Step 4:

[0229] Based on the information extracted by the server, the generative AI model is instructed to generate a draft report. Specifically, OpenAI GPT-3 is used to generate a prompt and make a request to generate a draft report. The input is data including similarities and differences, and the output is a prompt to the generative AI and a request to generate a draft report. For example, the command could be, "An error code 503 failure occurred on October 10, 2023, causing service C to be down for 30 minutes."

[0230] Step 5:

[0231] The generative AI model generates a draft report based on the instructions and sends it back to the server. Specifically, OpenAI GPT-3 generates a draft report based on a prompt and sends it back to the server. The input is the prompt, and the output is the generated draft report. For example, a draft report might be generated that states, "Server A failed on October 10, 2023. Service C was stopped for 30 minutes due to error code 503. The cause was a memory leak, and improved memory management is required as a countermeasure."

[0232] Step 6:

[0233] The server presents the generated draft report to the user, who makes any necessary revisions. Specifically, the report draft is displayed to the user through a smartphone application, allowing the user to modify technical details and specific terminology. The input is the generated draft report, and the output is the modified final report. The user saves the modified report and shares it with the appropriate parties.

[0234] In this way, the operating status of the factory robot can be monitored in real time, and a fault report can be generated quickly and accurately when an abnormality occurs.

[0235] 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.

[0236] System Overview

[0237] The present invention is a system that automatically generates trouble reports based on past sample reports provided by the user and current event data, and further combines it with an emotion engine that recognizes the user's emotions. The system has the following main functions:

[0238] Receiving data

[0239] Data Preprocessing

[0240] Data collation and analysis

[0241] Automatic generation of report drafts

[0242] User confirmation and correction

[0243] User Emotion Recognition and Feedback

[0244] Program processing overview

[0245] 1. User Input

[0246] Users log into the system from a terminal and upload past sample reports and current event data.

[0247] 2. Data Preprocessing

[0248] The server receives the uploaded data and performs text extraction and formatting, specifically removing unnecessary line breaks and spaces and converting the data into a format that is easy to analyze.

[0249] 3. Data Collation and Analysis

[0250] The server compares past sample reports with current event data to extract commonalities and differences, including characteristic features such as error codes, the time of the failure, and the scope of the impact.

[0251] 4. Instructions to the generating AI

[0252] Based on the extracted data, the server instructs the AI ​​to generate a draft report, which includes an overview of the problem, its scope of impact, causes, and countermeasures.

[0253] 5. Generate a draft report

[0254] The generation AI generates a draft report based on instructions from the server. The generation AI automatically generates detailed text for these items and sends it back to the server.

[0255] 6. Checking and correcting the report

[0256] The user reviews the generated draft report and makes any necessary corrections. Once the final report is complete, the user saves it and shares it with the appropriate departments and stakeholders.

[0257] 7. Emotion Recognition with Emotion Engine

[0258] When a user reviews or edits a draft report, the server uses an emotion engine to analyze the user's emotions. Emotion recognition is achieved by analyzing the user's facial expressions and tone of voice using camera footage and audio data.

[0259] 8. Emotional Feedback

[0260] Based on the results of the user's sentiment analysis, the server instructs the AI ​​to adjust the expression or generate additional comments. For example, if the user is feeling stressed, the AI ​​will simplify the explanation.

[0261] Specific examples

[0262] 1. User Input

[0263] Example: A user uploads "Sample Report_2022.pdf" and "Fault Log_2023-10-10.txt" to the system.

[0264] 2. Data Preprocessing

[0265] Example: The server extracts text from "Sample Report_2022.pdf" and the necessary log entries from "Error Log_2023-10-10.txt." It then removes unnecessary line breaks and spaces from the extracted data and converts it to JSON format.

[0266] 3. Data Collation and Analysis

[0267] Example: The server compares the sample report from 2022 with the error log from 2023, extracting the differences between the new error code "503" and the past error code "403." At the same time, it identifies commonalities in the scope of impact and occurrence time.

[0268] 4. Instructions to the generating AI

[0269] Example: The server instructs the AI ​​generation system that "An error code 503 occurred on October 10, 2023, causing service C to stop for 30 minutes." This instruction also includes a detailed summary of the error, the scope of its impact, the cause, and countermeasures.

[0270] 5. Generate a draft report

[0271] Example: The generation AI generates a draft report stating, "Server A failed on October 10, 2023. Service C was stopped for 30 minutes due to error code 503. The cause was a memory leak, and as a countermeasure, improved memory management is required."

[0272] 6. Checking and correcting the report

[0273] Example: User reviews the generated draft, corrects technical details and specific terminology to complete the final report, saves the completed report and shares it with the appropriate stakeholders.

[0274] 7. Emotion Recognition with Emotion Engine

[0275] Example: While a user is reviewing a report, the emotion engine analyzes the user's facial expressions from camera footage and recognizes that the user is feeling stressed.

[0276] 8. Emotional Feedback

[0277] Example: The server instructs the generated AI to make explanations more concise and emphasize important parts to reduce user stress. Feedback is provided based on the analysis of the emotion engine.

[0278] In this way, by utilizing the emotion engine, the system can reduce the burden on users and provide a more user-friendly process for creating trouble reports.

[0279] The processing flow will be explained below.

[0280] Step 1:

[0281] The user logs in to the system from a terminal and goes to the upload page. The user selects the files of past sample reports and current event data and clicks the upload button.

[0282] Step 2:

[0283] The server receives the uploaded sample reports and event data, which are temporarily stored and prepared for further processing.

[0284] Step 3:

[0285] The server runs a parsing process to extract text from the sample reports, specifically extracting the necessary text information from PDFs and other file formats and converting it into a format that is easier to parse.

[0286] Step 4:

[0287] The server extracts the necessary information from the event data (e.g., log data), specifically by parsing and extracting important log entries such as the target error code and timestamp.

[0288] Step 5:

[0289] The server removes unnecessary line breaks and spaces from the extracted data and converts the data into a unified format (e.g., JSON format).

[0290] Step 6:

[0291] The server compares the current event data with past sample reports, extracting commonalities (e.g., error codes, scope of impact) and differences (e.g., new error codes, different scope of impact).

[0292] Step 7:

[0293] Based on the results of the comparison, the server creates instructions for the generation AI to generate a draft report, including an overview of the problem, its scope of impact, causes, and countermeasures.

[0294] Step 8:

[0295] The generation AI generates a report draft based on instructions received from the server. The generation AI automatically generates detailed text for these items and sends the completed draft back to the server.

[0296] Step 9:

[0297] The server displays the generated report draft on the user's terminal, where the user can check the report draft and make corrections as necessary.

[0298] Step 10:

[0299] When a user checks a report draft, the server uses an emotion engine to recognize the user's emotions. The emotion engine analyzes the user's facial expressions and tone of voice, which are captured by a camera and microphone.

[0300] Step 11:

[0301] The server determines the user's emotional state (e.g., stress, dissatisfaction, relief) based on the analysis results of the emotion engine, and instructs the generation AI to adjust the content and expression of the report draft depending on the results.

[0302] Step 12:

[0303] Based on the analysis results of the emotion engine, the server displays feedback and alerts to the user. For example, if the user is feeling stressed, a comment is added to the report to simplify the explanation.

[0304] Step 13:

[0305] The user completes the final report and saves it on the server, where it is shared with the appropriate departments and stakeholders.

[0306] Step 14:

[0307] Users can effectively use the feedback from the emotion engine to improve the report creation process and identify areas for improvement. Through this process, adaptive report creation that takes emotions into consideration is realized.

[0308] Example 2

[0309] 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."

[0310] Conventional problem report creation systems often require a great deal of time and effort for users to check and edit reports, placing a heavy burden on users. Furthermore, because they do not take into account the user's emotional state, they can easily become stressed or fatigued. Furthermore, because the content of the generated reports is uniform, they sometimes do not function adequately even in situations where flexible responses according to the situation are required.

[0311] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0312] In this invention, the server includes means for receiving past sample reports and current event data from a user, means for preprocessing the received data and converting it into an easily analyzable format, means for comparing the past sample reports with the current event data and extracting similarities and differences, means for instructing a generation AI to generate a draft report based on the extracted information, means for presenting the report draft generated by the generation AI to the user, means for accepting user confirmation and correction of the report draft, means for recognizing the user's emotions, and means for adjusting the content of the report draft based on the recognized emotions. This reduces the burden on the user, streamlines the report creation process, and enables flexible report generation that corresponds to the user's emotional state.

[0313] "Users" refer to people who utilize the system to provide historical sample reports and current event data, and who are responsible for reviewing and modifying the final report.

[0314] "Sample report" refers to a previously created incident report that serves as a basis for comparison with current event data.

[0315] "Event Data" means data, including logs and information related to a current fault or problem, that is used to generate reports.

[0316] "Preprocessing" refers to the process of removing unnecessary line breaks and spaces from the received data and formatting it into a format that is easy to analyze.

[0317] "Matching" refers to the process of comparing past sample reports with current event data to identify similarities and differences.

[0318] "Generative AI" refers to an artificial intelligence model that automatically generates a draft report based on instructions from a server.

[0319] A "Draft Report" is an early-stage problem report created by the generation AI, before it has been reviewed and corrected by the user.

[0320] An "emotion engine" refers to software or hardware for recognizing a user's emotional state, for example by analyzing the user's facial expressions and tone of voice using camera or audio data.

[0321] "Feedback" refers to the process of providing instructions or information to the generative AI to adjust the content of the draft report based on the user's emotional state.

[0322] The present invention is a system that automatically generates trouble reports based on past sample reports and current event data provided by the user. Furthermore, this system combines an emotion engine that recognizes the user's emotions, aiming to reduce the burden on the user and streamline the report creation process.

[0323] Hardware and Software Configuration

[0324] The system is implemented using the following major hardware and software:

[0325] Server: A central device that preprocesses data, collates it, and issues instructions to the generation AI.

[0326] Terminal: Provides a user interface for users to upload data and review and modify generated draft reports.

[0327] Generative AI: An artificial intelligence model that automatically generates report drafts.

[0328] Emotion engine: Software or hardware for recognizing a user's emotions, including, for example, a facial recognition camera or voice analysis software.

[0329] Examples of data processing and data calculation

[0330] User Input

[0331] The user logs into the system from a terminal and uploads a past sample report (e.g., "Sample Report_2022.pdf") and current event data (e.g., "Fault Log_2023-10-10.txt").

[0332] Data Preprocessing

[0333] The server extracts text from the uploaded data and converts it into a format that is easy to parse. Specifically, it removes unnecessary line breaks and spaces and converts it into JSON format. For example, the server extracts text from "Sample Report_2022.pdf" and extracts the necessary log entries from "Failure Log_2023-10-10.txt."

[0334] Data collation and analysis

[0335] The server compares past sample reports with current event data to identify commonalities and differences. For example, it can extract the differences between the new error code "503" and the previous error code "403" and identify commonalities in the scope of impact and time of occurrence.

[0336] Instructions to the generating AI

[0337] The server instructs the generation AI to generate a draft report based on the extracted data. For example, it provides a prompt such as, "An error code 503 occurred on October 10, 2023, causing service C to be down for 30 minutes."

[0338] Generate a draft report

[0339] Based on the server's instructions, the generation AI generates a detailed draft report and sends it back to the server. Specifically, it generates a draft report with content such as, "Server A failed on October 10, 2023. Service C stopped for 30 minutes due to error code 503. The cause was a memory leak, and as a countermeasure, improved memory management is required."

[0340] Checking and correcting reports

[0341] The user reviews the generated draft report and makes any necessary corrections. Once the final report is complete, the user saves it and shares it with the appropriate departments and stakeholders.

[0342] Emotion recognition by emotion engine

[0343] When a user reviews or modifies a draft report, the server uses an emotion engine to analyze the user's emotions. For example, it can analyze the user's facial expressions from camera footage and recognize that the user is feeling stressed.

[0344] Emotion-based feedback

[0345] Based on the results of the user's sentiment analysis, the server instructs the AI ​​to adjust the expression or generate additional comments. For example, if the user is feeling stressed, the AI ​​is instructed to simplify the explanation.

[0346] In this way, by linking the server, terminal, and user and making full use of generative AI and an emotion engine, a system is realized that allows users to create fault reports efficiently and with little burden.

[0347] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0348] Step 1:

[0349] A user logs into the system and uploads data

[0350] Input: User ID and password, past sample report (e.g., "Sample Report_2022.pdf"), current event data (e.g., "Error Log_2023-10-10.txt")

[0351] Specific behavior:

[0352] A user enters a user ID and password to log in to the system from a terminal.

[0353] After successfully logging in, the user uploads "Sample Report_2022.pdf" and "Error Log_2023-10-10.txt" to the system.

[0354] Output: Uploaded sample report and event data

[0355] Step 2:

[0356] The server preprocesses the uploaded data

[0357] Input: Uploaded sample reports and event data

[0358] Specific behavior:

[0359] The server extracts text from "SampleReport_2022.pdf".

[0360] Similarly, extract the necessary log entries from "Fault Log_2023-10-10.txt".

[0361] To make the extracted text data easier to analyze, unnecessary line breaks and spaces are removed and the data is converted to JSON format.

[0362] Output: Formatted sample report data and event data (JSON format)

[0363] Step 3:

[0364] The server collates and analyzes past sample reports with current event data.

[0365] Input: Sample report data and event data in JSON format

[0366] Specific behavior:

[0367] The server extracts the difference between the new error code "503" and the previous error code "403".

[0368] The server identifies commonalities in the scope of impact and time of occurrence.

[0369] Output: Information on the difference between extracted error codes, the scope of impact, and the time of occurrence

[0370] Step 4:

[0371] The server instructs the generation AI to generate a draft report.

[0372] Input: Information on the difference between the extracted error codes, the scope of the impact, and the time of occurrence

[0373] Specific behavior:

[0374] The server provides the AI ​​with a specific prompt, such as "An error code 503 occurred on October 10, 2023, causing service C to be down for 30 minutes."

[0375] Output: Instructions to the generating AI

[0376] Step 5:

[0377] Generative AI generates a draft report

[0378] Input: Instructions to the generation AI

[0379] Specific behavior:

[0380] Based on the instructions, the generative AI generates a draft report containing detailed text.

[0381] Specifically, a draft report will be created stating, "Server A failed on October 10, 2023. Service C was stopped for 30 minutes due to error code 503. The cause was a memory leak, and strengthening memory management is required as a countermeasure."

[0382] Output: Draft report

[0383] Step 6:

[0384] The user checks the draft report and makes corrections

[0385] Input: Generated report draft

[0386] Specific behavior:

[0387] The user checks the report draft from the terminal.

[0388] The user corrects technical details and specific terminology to complete the final report.

[0389] Save the completed report and share it with appropriate parties.

[0390] Output: Final report

[0391] Step 7:

[0392] The server uses an emotion engine to recognize the user's emotions.

[0393] Input: Camera footage and audio data of the user reviewing the draft report

[0394] Specific behavior:

[0395] The server analyzes camera footage and audio data and recognizes emotions from the user's facial expressions and tone of voice.

[0396] For example, it determines whether the user is feeling stressed.

[0397] Output: User's emotional state

[0398] Step 8:

[0399] The server provides emotion-based feedback to the generative AI.

[0400] Input: User's emotional state

[0401] Specific behavior:

[0402] The server instructs the generation AI to adjust the expression of the report draft based on the user's emotional state.

[0403] For example, if the user is feeling stressed, the AI ​​generator can be instructed to keep explanations brief.

[0404] Output: Instructions to regenerate the draft report reflecting the adjusted feedback

[0405] The above is the specific processing flow of the system program.

[0406] (Application example 2)

[0407] 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."

[0408] In customer service at brick-and-mortar stores, employees are required to respond quickly and appropriately to a variety of customer requests and complaints. However, it can be difficult for employees to generate high-quality reports, especially during busy or stressful times, which can result in a decrease in customer satisfaction. Furthermore, responses that do not take employees' feelings into consideration can increase their own stress and, in the long term, lead to a decrease in job satisfaction and performance. A solution to these issues is needed.

[0409] 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.

[0410] In this invention, the server includes means for receiving past sample reports and current event data from a user, means for preprocessing the received data and converting it into an easily analyzable format, means for comparing the past sample reports with the current event data and extracting similarities and differences, means for instructing a generation AI to generate a draft report based on the extracted information, means for presenting the draft report generated by the generation AI to the user, means including an emotion engine for recognizing the user's emotions, and means for providing feedback based on the user's emotions. This enables employees to generate customer interaction reports quickly and with high quality, and also enables them to receive feedback that takes into consideration the employee's emotions.

[0411] "User" means a user who utilizes the System to provide past sample reports and current event data.

[0412] "Past sample reports" refer to reports that have been prepared and stored in the past, and are used as reference materials to guide responses to new events.

[0413] "Current Event Data" refers to data that includes information about ongoing events, new requests, complaints, etc.

[0414] "Preprocessing" refers to the process of removing unnecessary information from received data and converting it into a format suitable for analysis.

[0415] "Generative AI" refers to artificial intelligence that automatically generates draft reports based on specified instructions.

[0416] A "draft" is an initial report created by the generative AI, which is subject to user review and correction.

[0417] An "emotion engine" is a system that includes technology that recognizes a user's emotions by analyzing the user's facial expressions and tone of voice.

[0418] "Providing feedback" means presenting appropriate comments and advice that take into consideration the user's emotions based on the analysis results of the emotion engine.

[0419] This invention provides a system for providing fast and high-quality customer service in brick-and-mortar stores. The system receives past sample reports and current event data, analyzes the data, and automatically generates a report draft. It also recognizes the user's emotions and provides feedback based on the emotions.

[0420] System Overview

[0421] The system consists of the following main components:

[0422] User Input Method

[0423] Data preprocessing measures

[0424] Data collation method

[0425] Generation AI instruction means

[0426] Generation AI

[0427] Emotion Engine

[0428] Feedback methods

[0429] Hardware and software used

[0430] Hardware:

[0431] PC or smart device (with camera)

[0432] software:

[0433] Python

[0434] spaCy (natural language processing)

[0435] OpenAI API (generative AI model)

[0436] OpenCV (for emotion analysis)

[0437] Program processing overview

[0438] The system processes data in the following order, generates a draft report, checks and corrects it, and finally provides feedback based on the user's feelings.

[0439] User Input Method

[0440] Users input past sample reports and current event data into the system.

[0441] Data preprocessing measures

[0442] The server preprocesses the data it receives, removing unnecessary line breaks and spaces from the text and converting it into a format that is easier to parse.

[0443] Data collation method

[0444] The server compares past sample reports with current event data and extracts commonalities and differences, providing the basis for the generation AI to generate reports.

[0445] Generation AI instruction means

[0446] The server issues instructions to the generation AI based on the extracted data to generate a draft report. An example of the prompt used at this time is shown below.

[0447] Example prompt sentence:

[0448] Common points: {Common points}, Differences: {Differences}. Please generate a customer service response report that addresses the new request effectively and improves customer satisfaction.

[0449] Generation AI

[0450] The generation AI generates a draft report based on instructions from the server. Specifically, it documents appropriate responses and details of customer responses based on similarities and differences.

[0451] Emotion Engine

[0452] When a user reviews or modifies a draft report, the emotion engine analyzes camera footage and audio data to recognize emotions from the user's facial expressions and tone of voice.

[0453] Feedback methods

[0454] The feedback provision means instructs the generation AI to adjust expressions or generate additional comments based on the analysis results of the emotion engine, ultimately generating a report that takes the user's emotions into consideration.

[0455] Specific examples

[0456] 1. User input:

[0457] Example: A user uploads "Sample Report_2022.pdf" and "Customer Request_2023-10-10.txt" to the system.

[0458] 2. Data preprocessing:

[0459] Example: The server extracts text from "Sample Report_2022.pdf" and necessary information from "Customer Request_2023-10-10.txt." It then removes unnecessary line breaks and spaces from the extracted data and converts it into a format suitable for analysis.

[0460] 3. Data collation and analysis:

[0461] Example: The server compares the current customer request with past sample reports and extracts the differences between the new request and the past response history.

[0462] 4. Instructions to the generating AI:

[0463] Example: The server instructs the generation AI to generate a draft report based on "commonalities: maintaining past services, differences: new requests."

[0464] 5. Generate a draft report:

[0465] Example: The generation AI generates a draft report stating, "A new request was received from the customer on October 10, 2023. A response is required. Based on past response history and the new request, we suggest the following response."

[0466] 6. Emotion Recognition with Emotion Engine:

[0467] Example: While a user is reviewing and revising a draft report, the camera analyzes their facial expressions and recognizes that they are feeling stressed.

[0468] 7. Emotion-based feedback:

[0469] Example: The server instructs the generated AI to simplify the explanation and emphasize the important parts to reduce stress for the user.

[0470] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0471] Step 1:

[0472] The user uploads past sample reports and current event data from a terminal to the system. The input is a past sample report file and a current event data file, and the output is a confirmation of receipt of the uploaded files. The server receives these files and proceeds to the next step.

[0473] Step 2:

[0474] The server preprocesses the uploaded data and converts it into a format that is easy to analyze. This processing includes extracting text and removing unnecessary line breaks and spaces. The input is the uploaded sample report file and event data file, and the output is formatted text data. Specifically, it uses the Python spaCy library to remove unnecessary information from the text and convert the data into a structured format such as JSON.

[0475] Step 3:

[0476] The server uses the preprocessed data to match past sample reports with the current event data and extract commonalities and differences. The input is the formatted past sample report data and the current event data, and the output is a list of extracted commonalities and differences. Specifically, it uses text mining techniques to identify common keywords and phrases, as well as new elements, between the past reports and the new data.

[0477] Step 4:

[0478] The server instructs the generative AI to generate a draft report based on the extracted data. The input is a list of commonalities and differences, and the output is a prompt for the generative AI. The prompt is sent to the generative AI model (for example, OpenAI's API) to generate a detailed draft report. An example of a specific prompt is "Common points: {Common points}, Differences: {Differences}. Please generate a customer service response report that addresses the new request effectively and improves the customer satisfaction."

[0479] Step 5:

[0480] The generation AI generates a draft report based on prompts from the server and sends it back to the server. The input is the prompt text passed from the server, and the output is the generated draft report. Specifically, the generation AI automatically creates a detailed report of the customer interaction based on similarities and differences.

[0481] Step 6:

[0482] The user checks and modifies the generated report draft. To do this, they open the draft using a terminal and modify the content as necessary. The input is the generated report draft, and the output is the modified final report. The user checks the technical details and terminology of the report and makes any necessary modifications based on their own expertise.

[0483] Step 7:

[0484] The server uses an emotion engine to analyze the user's emotions when reviewing and revising a report draft. The input is video and audio data of the user's face, and the output is the analysis result of the user's emotional state. Specifically, it uses image analysis libraries such as OpenCV to recognize emotions from the user's facial expressions and tone of voice.

[0485] Step 8:

[0486] Based on the analysis results of the emotion engine, the server instructs the generation AI to provide feedback. The input is the user's emotion analysis results, and the output is feedback on the adjusted report and additional comments. Specifically, the generation AI readjusts the content of the report according to the user's emotions, providing concise explanations and emphasizing important parts to reduce stress. This allows the user to more efficiently finalize the report.

[0487] 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.

[0488] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0489] 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.

[0490] [Second embodiment]

[0491] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0492] 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.

[0493] 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).

[0494] 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.

[0495] 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.

[0496] 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).

[0497] 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.

[0498] 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.

[0499] 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.

[0500] 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.

[0501] 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.

[0502] 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."

[0503] System Overview

[0504] The present invention is a system for automatically generating fault reports based on user-provided sample reports and current event data. The system includes the following major functions:

[0505] Receiving data

[0506] Data Preprocessing

[0507] Data collation and analysis

[0508] Automatic generation of report drafts

[0509] User confirmation and correction

[0510] Program processing overview

[0511] 1. User Input

[0512] Users log into the system from a terminal and upload past sample reports and current event data.

[0513] 2. Data Preprocessing

[0514] The server receives the uploaded data and performs text extraction and formatting, specifically removing unnecessary line breaks and spaces and converting the data into a format that is easy to analyze.

[0515] 3. Data Collation and Analysis

[0516] The server compares past sample reports with current event data to extract commonalities and differences, including characteristic features such as error codes, the time of the failure, and the scope of the impact.

[0517] 4. Instructions to the generating AI

[0518] Based on the extracted data, the server instructs the AI ​​to generate a draft report, which includes an overview of the problem, its scope of impact, causes, and countermeasures.

[0519] 5. Generate a draft report

[0520] Based on instructions from the server, the generation AI creates a draft report detailing the problem, which is then sent back to the server in a format that can be reviewed and modified by the user.

[0521] 6. Checking and correcting the report

[0522] The user reviews the generated draft report and makes any necessary corrections. Once the final report is complete, the user saves it and shares it with the appropriate departments and stakeholders.

[0523] Specific examples

[0524] 1. User Input

[0525] Example: A user uploads "Sample Report_2022.pdf" and "Fault Log_2023-10-10.txt" to the system.

[0526] 2. Data Preprocessing

[0527] Example: The server extracts text from "Sample Report_2022.pdf" and the necessary log entries from "Error Log_2023-10-10.txt." It then removes unnecessary line breaks and spaces from the extracted data and converts it to JSON format.

[0528] 3. Data Collation and Analysis

[0529] Example: The server compares the sample report from 2022 with the error log from 2023, extracting the differences between the new error code "503" and the past error code "403." At the same time, it identifies commonalities in the scope of impact and occurrence time.

[0530] 4. Instructions to the generating AI

[0531] Example: The server instructs the AI ​​generation system that "An error code 503 occurred on October 10, 2023, causing service C to stop for 30 minutes." This instruction also includes a detailed summary of the error, the scope of its impact, the cause, and countermeasures.

[0532] 5. Generate a draft report

[0533] Example: The generation AI generates a draft report stating, "Server A failed on October 10, 2023. Service C was stopped for 30 minutes due to error code 503. The cause was a memory leak, and as a countermeasure, improved memory management is required."

[0534] 6. Checking and correcting the report

[0535] Example: User reviews the generated draft, corrects technical details and specific terminology to complete the final report, saves the completed report and shares it with the appropriate stakeholders.

[0536] In this way, the system can generate trouble reports efficiently and accurately, reducing the workload of the user.

[0537] The processing flow will be explained below.

[0538] Step 1:

[0539] The user logs in to the system from a terminal and goes to the upload page. The user selects the files of past sample reports and current event data and clicks the upload button.

[0540] Step 2:

[0541] The server receives the uploaded sample reports and event data, which are temporarily stored and prepared for further processing.

[0542] Step 3:

[0543] The server performs a parsing process to extract text from the uploaded sample reports, specifically extracting the required text information from PDFs and other file formats.

[0544] Step 4:

[0545] The server extracts the necessary information from the event data (e.g., log data), specifically by parsing and extracting important log entries such as the target error code and timestamp.

[0546] Step 5:

[0547] The server removes unnecessary line breaks and spaces from the extracted data and converts the data into a format that is easy to parse (e.g., JSON format).

[0548] Step 6:

[0549] The server compares the current event data with past sample reports, extracting commonalities (e.g., error codes, scope of impact) and differences (e.g., new error codes, different scope of impact).

[0550] Step 7:

[0551] Based on the results of the comparison, the server creates instructions for the generation AI to generate a draft report, including an overview of the problem, its scope of impact, causes, and countermeasures.

[0552] Step 8:

[0553] The generation AI generates a draft report based on instructions received from the server. The generation AI automatically generates detailed text for these items and provides feedback.

[0554] Step 9:

[0555] The generated report draft is sent back to the server and presented to the user, who can then check the generated draft through the terminal.

[0556] Step 10:

[0557] The user reviews the draft and makes any necessary corrections, such as changing technical terminology, adding details, or correcting typos.

[0558] Step 11:

[0559] The user saves the final report and then shares it with the appropriate departments and stakeholders, completing the process of generating a problem report.

[0560] Example 1

[0561] 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."

[0562] Creating a failure report manually is extremely time-consuming and prone to errors and oversights. The process of creating a report by comparing and analyzing past reports with current event data is particularly labor-intensive, making it necessary to create reports efficiently and accurately. Furthermore, reports must include a detailed failure summary, scope of impact, cause, and countermeasures, but there is currently a lack of a means to automatically generate standardized reports that include all of these.

[0563] 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.

[0564] In this invention, the server includes means for receiving past sample reports and current event data from a user, means for converting the received data into a format that is easy to analyze by extracting text and formatting, means for comparing the past sample reports and the current event data using natural language processing and machine learning algorithms to extract similarities and differences, means for providing a prompt to the generative AI model to generate a draft report based on the extracted information, and means for presenting the draft report generated by the generative AI model to the user in a format that can be confirmed and modified, thereby enabling the user to generate and modify reports efficiently and accurately.

[0565] "User" refers to a person or organization that uses a terminal to access the system and provide various data.

[0566] "Sample Report" means a document that contains a detailed record of a past event.

[0567] "Event Data" refers to data relating to a particular event that is currently occurring or has recently occurred.

[0568] "Text extraction" refers to the process of extracting character data from a rich text format file.

[0569] "Formatting" refers to the process of removing unnecessary line breaks and spaces from data and converting it into a format that is easier to parse.

[0570] "Natural language processing (NLP)" refers to technology for analyzing text data and understanding its meaning and structure.

[0571] A "machine learning algorithm" refers to a method of learning patterns and rules from large amounts of data and using that knowledge to analyze new data.

[0572] "Matching" refers to the process of finding similarities and differences between different data sets.

[0573] A "generative AI model" refers to an artificial intelligence system that automatically generates text or other data based on instructions it is given.

[0574] A "prompt sentence" refers to a textual input sentence that provides instructions to a generative AI model.

[0575] "Draft Report" refers to the initial version of the report automatically generated by the generative AI model.

[0576] "Format" refers to the particular structure or format of data or documents.

[0577] "Verification" refers to the process in which a user reviews the generated report draft and checks its content.

[0578] "Modification" refers to the user making any necessary changes or additions to the draft report.

[0579] The present invention is a system for automatically generating fault reports based on user-provided sample reports and current event data. The system can be implemented using the following major hardware and software components:

[0580] Hardware and software used

[0581] Terminal: A device used by a user to access the system and enter data. This includes devices such as PCs and smartphones.

[0582] Server: A central computing unit that preprocesses data, analyzes it, and directs generative AI models.

[0583] A text analysis library, such as Apache Tika, can be used to extract text from PDF files.

[0584] Use an NLP library such as spaCy to analyze text data and extract the necessary information.

[0585] Generative AI models: Generative AI models such as GPT-4 are used to generate draft reports.

[0586] Data processing and analysis

[0587] 1. User Input

[0588] A user logs in to the system from a terminal and uploads past sample reports and current event data, for example, uploading a report in PDF format and a fault log in text format.

[0589] Specifically, the user uploads "Sample Report_2022.pdf" and "Error Log_2023-10-10.txt" to the system.

[0590] 2. Data Preprocessing

[0591] The server receives the uploaded data and performs text extraction and formatting.

[0592] Specifically, we use Apache Tika to extract text from "SampleReport_2022.pdf", select the required log entries from the TXT file, remove unnecessary line breaks and spaces, and convert the data to JSON format.

[0593] 3. Data Collation and Analysis

[0594] The server uses the extracted data to compare and analyze past sample reports with current event data.

[0595] Specifically, spaCy is used to extract information such as error codes, dates and times, and the extent of the impact, allowing past and present data to be compared to identify commonalities and differences.

[0596] 4. Instructions to the generating AI

[0597] Based on the analysis results, the server instructs the generative AI model to generate a draft report.

[0598] An example of a specific prompt might be, "An error code 503 occurred on October 10, 2023, causing service C to be down for 30 minutes." This instruction also includes a detailed summary of the error, the scope of its impact, the cause, and countermeasures.

[0599] 5. Generate a draft report

[0600] The generative AI model generates a draft report detailing the fault based on instructions from the server, which is then sent back to the server in a format that can be reviewed and modified by the user.

[0601] Specifically, it generates a draft report stating something like, "Server A failed on October 10, 2023. Service C stopped for 30 minutes due to error code 503. The cause was a memory leak, and as a countermeasure, improved memory management is required."

[0602] 6. Checking and correcting the report

[0603] The user reviews the generated draft report, makes any necessary revisions, and finally saves the final report and shares it with the appropriate departments and agencies.

[0604] Specifically, users can access the draft from their devices, edit specific technical information and terminology, and then finalize the report, which is then saved and sent to stakeholders via email or a shared link.

[0605] In this way, the present invention can efficiently and accurately generate trouble reports, significantly reducing the user's workload. This system can be implemented using specific hardware and software, making it applicable to a variety of situations.

[0606] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0607] Step 1: User Input

[0608] A user logs in to the system from a terminal and uploads a past sample report (PDF format) and current event data (TXT format). The input data is "Sample Report_2022.pdf" and "Fault Log_2023-10-10.txt."

[0609] Specific behavior:

[0610] The user enters their username and password on the login page to access the system.

[0611] After logging in, select the "File Upload" option on your dashboard, select the required file and upload it.

[0612] If the upload is successful, the file is sent to the server.

[0613] Step 2: Preprocessing the data

[0614] The server receives the uploaded data, extracts text, and formats it. The input data is PDF and TXT files, and the output data is in an easy-to-parse JSON format.

[0615] Specific behavior:

[0616] The server uses the Apache Tika library to extract text from the PDF file "SampleReport_2022.pdf".

[0617] At the same time, the server selects the necessary log entries from the TXT file "Error Log_2023-10-10.txt".

[0618] Unnecessary line breaks and spaces are removed from this text data, and the formatted text data is converted into JSON format.

[0619] Step 3: Collate and analyze data

[0620] The server uses the extracted data to compare and analyze past sample reports with current event data. The input is text data in JSON format, and the output is information about notable similarities and differences.

[0621] Specific behavior:

[0622] The server uses an NLP library such as spaCy to extract information such as the error code, date and time, and scope of impact from the text data.

[0623] Based on the extracted information, past reports are compared with the current fault log to identify similarities and differences.

[0624] As a result, information such as differences in error codes and commonalities in the scope of impact is generated and passed on to the next step.

[0625] Step 4: Instructions to the Generative AI

[0626] The server instructs the generative AI model to generate a draft report based on the analysis results. The input is information about similarities and differences, and the output is a prompt statement for the generative AI model.

[0627] Specific behavior:

[0628] The server formalizes the analysis results as prompt sentences and provides them to the generative AI model.

[0629] As an example of a specific prompt sentence, a sentence such as "An error code 503 occurred on October 10, 2023, causing service C to be down for 30 minutes" is generated.

[0630] The prompt text includes a detailed overview of the problem, the scope of impact, the cause, and countermeasures.

[0631] Step 5: Generate a draft report

[0632] The generative AI model generates a draft report based on the prompt provided by the server. The input is the prompt to the generative AI, and the output is the draft report.

[0633] Specific behavior:

[0634] The generative AI model receives the prompt and creates a draft report.

[0635] The generated report draft is generated in Markdown, HTML, Word document format, etc. and sent back to the server.

[0636] The draft includes content such as, "Server A failed on October 10, 2023. Service C was stopped for 30 minutes due to error code 503. The cause was a memory leak, and as a countermeasure, improved memory management is required."

[0637] Step 6: Check and correct the report

[0638] The user reviews the generated report draft and makes any necessary corrections. The input is the report draft and the output is the final report.

[0639] Specific behavior:

[0640] The user accesses the draft from a terminal and checks its contents.

[0641] Correct specific technical information and terminology as needed to complete the final report.

[0642] Completed reports are stored within the system and then sent to the appropriate parties via email or shared link.

[0643] In this way, the system can automatically generate trouble reports efficiently and accurately based on user-supplied data.

[0644] (Application example 1)

[0645] 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."

[0646] Conventional factory robot monitoring systems require human intervention when a failure occurs, making it difficult to respond quickly and create accurate reports. Furthermore, manually comparing past failure reports with current event data and generating reports is time-consuming and labor-intensive. These issues can significantly impact factory operational efficiency and product quality, so a system that solves these problems is needed.

[0647] 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.

[0648] In this invention, the server includes means for receiving past sample reports and current event data from a user, means for preprocessing the received data and converting it into an easily analyzable format, means for comparing the past sample reports with the current event data and extracting similarities and differences, means for instructing the generation AI to generate a draft report based on the extracted information, means for presenting the draft report generated by the generation AI to the user, and means for monitoring the operating status of factory robots in real time and automatically generating a report when an abnormality occurs. This makes it possible to respond quickly and accurately to factory robot failures and improve factory operating efficiency.

[0649] "User" means a person or entity that uses the System to provide and query past sample reports and current event data.

[0650] A "Past Sample Report" is a document that contains detailed information about past events or failures.

[0651] "Current event data" refers to data relating to events or failures occurring at the present time.

[0652] "Preprocessing" is the process of deleting unnecessary line breaks and spaces and reformatting the data in order to convert the received data into a format that is easy to analyze.

[0653] "Matching" is the process of comparing past sample reports with current event data to extract similarities and differences.

[0654] "Generative AI" is an artificial intelligence system that automatically generates draft reports based on given data and instructions.

[0655] A "report draft" is a draft of a report automatically generated by the generation AI.

[0656] A "factory robot" is a mechanical device used to perform automated tasks in manufacturing.

[0657] "Operation status" is information that indicates the state in which a factory robot is operating.

[0658] "Real-time monitoring" refers to the continuous observation of a situation occurring at the present time.

[0659] An "abnormality" is a deviation from normal operation or failure of a system or machine.

[0660] A "report" is a formal document containing detailed information, causes, and actions regarding a specific incident or situation.

[0661] This invention is a system that monitors the operating status of factory robots in real time and automatically generates a fault report when an abnormality occurs. This system creates reports with high accuracy and efficiency based on past sample reports and current event data provided by the user.

[0662] 1. System Configuration

[0663] The system includes a server, user terminals, factory robots, generative AI models, and front-end and back-end processing. The server receives data, preprocesses it, collates it, and issues instructions to the generative AI. The user terminal uploads data and checks and edits reports via a smartphone application. The factory robots are monitored for their operating status and provide data in real time.

[0664] 2. Hardware / Software to be used

[0665] Hardware: smartphones, servers, factory robots

[0666] software:

[0667] Server: Flask (web framework), SQLite (database)

[0668] Preprocessing: Text extraction and data formatting using Python, NLTK (Natural Language Toolkit)

[0669] Generation AI: OpenAI GPT-3

[0670] Frontend: HTML / CSS / JavaScript

[0671] 3. Specific examples of program processing

[0672] Data reception and preprocessing

[0673] Users upload past sample reports and current event data from their devices, and the server receives the data, then uses Python to extract and format the text, removing unnecessary line breaks and spaces, and converting it into a format that is easier to parse.

[0674] Data collation and analysis

[0675] The preprocessed data is compared with past sample reports and current event data to extract commonalities and differences, including characteristic information such as error codes, time of occurrence, and scope of impact.

[0676] Instructions for generative AI models

[0677] The server instructs a generative AI model (e.g., OpenAI GPT-3) to generate a draft report based on the extracted data, including a summary of the problem, its scope of impact, causes, and countermeasures.

[0678] Draft report generation and presentation

[0679] The generative AI model generates a draft report based on the data and prompts received from the server, which is then sent back to the server and presented to the user, who can make any necessary corrections to produce a final fault report.

[0680] Examples of concrete examples and prompts

[0681] When a user uploads "Sample Report_2022.pdf" and "Fault Log_2023-10-10.txt" to the system, the system automatically analyzes these data, extracts commonalities and differences, and generates a draft report. The generated draft contains the following content:

[0682] Example prompt sentence:

[0683] "Based on past incident reports and the current incident log, we will provide the following information. Please use this information to generate a detailed incident report."

[0684] This system will enable rapid and accurate response to factory robot failures, significantly improving factory operational efficiency.

[0685] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0686] Step 1:

[0687] The user uploads past sample reports and current event data from their device. Specifically, the user uses a smartphone application to send "Sample Report_2022.pdf" and "Fault Log_2023-10-10.txt" to the system. The input is a PDF and text file, and these files are saved on the server as output.

[0688] Step 2:

[0689] The data received by the server is preprocessed. Specifically, text is extracted using Python, and unnecessary line breaks and spaces are removed to convert it into a format that is easy to parse. The input is the uploaded PDF and text file, and the output is formatted text data. For example, the text extracted from the PDF and the contents of the text file are converted into JSON format.

[0690] Step 3:

[0691] The server compares past sample reports with current event data to extract commonalities and differences. Specifically, it uses a scripting language (Python) to compare characteristic information such as error codes, the time of the failure, and the scope of impact. The input is preprocessed text data, and the output is data containing commonalities and differences. For example, it extracts the differences between "error code 503" and "error code 403."

[0692] Step 4:

[0693] Based on the information extracted by the server, the generative AI model is instructed to generate a draft report. Specifically, OpenAI GPT-3 is used to generate a prompt and make a request to generate a draft report. The input is data including similarities and differences, and the output is a prompt to the generative AI and a request to generate a draft report. For example, the command could be, "An error code 503 failure occurred on October 10, 2023, causing service C to be down for 30 minutes."

[0694] Step 5:

[0695] The generative AI model generates a draft report based on the instructions and sends it back to the server. Specifically, OpenAI GPT-3 generates a draft report based on a prompt and sends it back to the server. The input is the prompt, and the output is the generated draft report. For example, a draft report might be generated that states, "Server A failed on October 10, 2023. Service C was stopped for 30 minutes due to error code 503. The cause was a memory leak, and improved memory management is required as a countermeasure."

[0696] Step 6:

[0697] The server presents the generated draft report to the user, who makes any necessary revisions. Specifically, the report draft is displayed to the user through a smartphone application, allowing the user to modify technical details and specific terminology. The input is the generated draft report, and the output is the modified final report. The user saves the modified report and shares it with the appropriate parties.

[0698] In this way, the operating status of the factory robot can be monitored in real time, and a fault report can be generated quickly and accurately when an abnormality occurs.

[0699] 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.

[0700] System Overview

[0701] The present invention is a system that automatically generates trouble reports based on past sample reports provided by the user and current event data, and further combines it with an emotion engine that recognizes the user's emotions. The system has the following main functions:

[0702] Receiving data

[0703] Data Preprocessing

[0704] Data collation and analysis

[0705] Automatic generation of report drafts

[0706] User confirmation and correction

[0707] User Emotion Recognition and Feedback

[0708] Program processing overview

[0709] 1. User Input

[0710] Users log into the system from a terminal and upload past sample reports and current event data.

[0711] 2. Data Preprocessing

[0712] The server receives the uploaded data and performs text extraction and formatting, specifically removing unnecessary line breaks and spaces and converting the data into a format that is easy to analyze.

[0713] 3. Data Collation and Analysis

[0714] The server compares past sample reports with current event data to extract commonalities and differences, including characteristic features such as error codes, the time of the failure, and the scope of the impact.

[0715] 4. Instructions to the generating AI

[0716] Based on the extracted data, the server instructs the AI ​​to generate a draft report, which includes an overview of the problem, its scope of impact, causes, and countermeasures.

[0717] 5. Generate a draft report

[0718] The generation AI generates a draft report based on instructions from the server. The generation AI automatically generates detailed text for these items and sends it back to the server.

[0719] 6. Checking and correcting the report

[0720] The user reviews the generated draft report and makes any necessary corrections. Once the final report is complete, the user saves it and shares it with the appropriate departments and stakeholders.

[0721] 7. Emotion Recognition with Emotion Engine

[0722] When a user reviews or edits a draft report, the server uses an emotion engine to analyze the user's emotions. Emotion recognition is achieved by analyzing the user's facial expressions and tone of voice using camera footage and audio data.

[0723] 8. Emotional Feedback

[0724] Based on the results of the user's sentiment analysis, the server instructs the AI ​​to adjust the expression or generate additional comments. For example, if the user is feeling stressed, the AI ​​will simplify the explanation.

[0725] Specific examples

[0726] 1. User Input

[0727] Example: A user uploads "Sample Report_2022.pdf" and "Fault Log_2023-10-10.txt" to the system.

[0728] 2. Data Preprocessing

[0729] Example: The server extracts text from "Sample Report_2022.pdf" and the necessary log entries from "Error Log_2023-10-10.txt." It then removes unnecessary line breaks and spaces from the extracted data and converts it to JSON format.

[0730] 3. Data Collation and Analysis

[0731] Example: The server compares the sample report from 2022 with the error log from 2023, extracting the differences between the new error code "503" and the past error code "403." At the same time, it identifies commonalities in the scope of impact and occurrence time.

[0732] 4. Instructions to the generating AI

[0733] Example: The server instructs the AI ​​generation system that "An error code 503 occurred on October 10, 2023, causing service C to stop for 30 minutes." This instruction also includes a detailed summary of the error, the scope of its impact, the cause, and countermeasures.

[0734] 5. Generate a draft report

[0735] Example: The generation AI generates a draft report stating, "Server A failed on October 10, 2023. Service C was stopped for 30 minutes due to error code 503. The cause was a memory leak, and as a countermeasure, improved memory management is required."

[0736] 6. Checking and correcting the report

[0737] Example: User reviews the generated draft, corrects technical details and specific terminology to complete the final report, saves the completed report and shares it with the appropriate stakeholders.

[0738] 7. Emotion Recognition with Emotion Engine

[0739] Example: While a user is reviewing a report, the emotion engine analyzes the user's facial expressions from camera footage and recognizes that the user is feeling stressed.

[0740] 8. Emotional Feedback

[0741] Example: The server instructs the generated AI to make explanations more concise and emphasize important parts to reduce user stress. Feedback is provided based on the analysis of the emotion engine.

[0742] In this way, by utilizing the emotion engine, the system can reduce the burden on users and provide a more user-friendly process for creating trouble reports.

[0743] The processing flow will be explained below.

[0744] Step 1:

[0745] The user logs in to the system from a terminal and goes to the upload page. The user selects the files of past sample reports and current event data and clicks the upload button.

[0746] Step 2:

[0747] The server receives the uploaded sample reports and event data, which are temporarily stored and prepared for further processing.

[0748] Step 3:

[0749] The server runs a parsing process to extract text from the sample reports, specifically extracting the necessary text information from PDFs and other file formats and converting it into a format that is easier to parse.

[0750] Step 4:

[0751] The server extracts the necessary information from the event data (e.g., log data), specifically by parsing and extracting important log entries such as the target error code and timestamp.

[0752] Step 5:

[0753] The server removes unnecessary line breaks and spaces from the extracted data and converts the data into a unified format (e.g., JSON format).

[0754] Step 6:

[0755] The server compares the current event data with past sample reports, extracting commonalities (e.g., error codes, scope of impact) and differences (e.g., new error codes, different scope of impact).

[0756] Step 7:

[0757] Based on the results of the comparison, the server creates instructions for the generation AI to generate a draft report, including an overview of the problem, its scope of impact, causes, and countermeasures.

[0758] Step 8:

[0759] The generation AI generates a report draft based on instructions received from the server. The generation AI automatically generates detailed text for these items and sends the completed draft back to the server.

[0760] Step 9:

[0761] The server displays the generated report draft on the user's terminal, where the user can check the report draft and make corrections as necessary.

[0762] Step 10:

[0763] When a user checks a report draft, the server uses an emotion engine to recognize the user's emotions. The emotion engine analyzes the user's facial expressions and tone of voice, which are captured by a camera and microphone.

[0764] Step 11:

[0765] The server determines the user's emotional state (e.g., stress, dissatisfaction, relief) based on the analysis results of the emotion engine, and instructs the generation AI to adjust the content and expression of the report draft depending on the results.

[0766] Step 12:

[0767] Based on the analysis results of the emotion engine, the server displays feedback and alerts to the user. For example, if the user is feeling stressed, a comment is added to the report to simplify the explanation.

[0768] Step 13:

[0769] The user completes the final report and saves it on the server, where it is shared with the appropriate departments and stakeholders.

[0770] Step 14:

[0771] Users can effectively use the feedback from the emotion engine to improve the report creation process and identify areas for improvement. Through this process, adaptive report creation that takes emotions into consideration is realized.

[0772] Example 2

[0773] 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."

[0774] Conventional problem report creation systems often require a great deal of time and effort for users to check and edit reports, placing a heavy burden on users. Furthermore, because they do not take into account the user's emotional state, they can easily become stressed or fatigued. Furthermore, because the content of the generated reports is uniform, they sometimes do not function adequately even in situations where flexible responses according to the situation are required.

[0775] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0776] In this invention, the server includes means for receiving past sample reports and current event data from a user, means for preprocessing the received data and converting it into an easily analyzable format, means for comparing the past sample reports with the current event data and extracting similarities and differences, means for instructing a generation AI to generate a draft report based on the extracted information, means for presenting the report draft generated by the generation AI to the user, means for accepting user confirmation and correction of the report draft, means for recognizing the user's emotions, and means for adjusting the content of the report draft based on the recognized emotions. This reduces the burden on the user, streamlines the report creation process, and enables flexible report generation that corresponds to the user's emotional state.

[0777] "Users" refer to people who utilize the system to provide historical sample reports and current event data, and who are responsible for reviewing and modifying the final report.

[0778] "Sample report" refers to a previously created incident report that serves as a basis for comparison with current event data.

[0779] "Event Data" means data, including logs and information related to a current fault or problem, that is used to generate reports.

[0780] "Preprocessing" refers to the process of removing unnecessary line breaks and spaces from the received data and formatting it into a format that is easy to analyze.

[0781] "Matching" refers to the process of comparing past sample reports with current event data to identify similarities and differences.

[0782] "Generative AI" refers to an artificial intelligence model that automatically generates a draft report based on instructions from a server.

[0783] A "Draft Report" is an early-stage problem report created by the generation AI, before it has been reviewed and corrected by the user.

[0784] An "emotion engine" refers to software or hardware for recognizing a user's emotional state, for example by analyzing the user's facial expressions and tone of voice using camera or audio data.

[0785] "Feedback" refers to the process of providing instructions or information to the generative AI to adjust the content of the draft report based on the user's emotional state.

[0786] The present invention is a system that automatically generates trouble reports based on past sample reports and current event data provided by the user. Furthermore, this system combines an emotion engine that recognizes the user's emotions, aiming to reduce the burden on the user and streamline the report creation process.

[0787] Hardware and Software Configuration

[0788] The system is implemented using the following major hardware and software:

[0789] Server: A central device that preprocesses data, collates it, and issues instructions to the generation AI.

[0790] Terminal: Provides a user interface for users to upload data and review and modify generated draft reports.

[0791] Generative AI: An artificial intelligence model that automatically generates report drafts.

[0792] Emotion engine: Software or hardware for recognizing a user's emotions, including, for example, a facial recognition camera or voice analysis software.

[0793] Examples of data processing and data calculation

[0794] User Input

[0795] The user logs into the system from a terminal and uploads a past sample report (e.g., "Sample Report_2022.pdf") and current event data (e.g., "Fault Log_2023-10-10.txt").

[0796] Data Preprocessing

[0797] The server extracts text from the uploaded data and converts it into a format that is easy to parse. Specifically, it removes unnecessary line breaks and spaces and converts it into JSON format. For example, the server extracts text from "Sample Report_2022.pdf" and extracts the necessary log entries from "Failure Log_2023-10-10.txt."

[0798] Data collation and analysis

[0799] The server compares past sample reports with current event data to identify commonalities and differences. For example, it can extract the differences between the new error code "503" and the previous error code "403" and identify commonalities in the scope of impact and time of occurrence.

[0800] Instructions to the generating AI

[0801] The server instructs the generation AI to generate a draft report based on the extracted data. For example, it provides a prompt such as, "An error code 503 occurred on October 10, 2023, causing service C to be down for 30 minutes."

[0802] Generate a draft report

[0803] Based on the server's instructions, the generation AI generates a detailed draft report and sends it back to the server. Specifically, it generates a draft report with content such as, "Server A failed on October 10, 2023. Service C stopped for 30 minutes due to error code 503. The cause was a memory leak, and as a countermeasure, improved memory management is required."

[0804] Checking and correcting reports

[0805] The user reviews the generated draft report and makes any necessary corrections. Once the final report is complete, the user saves it and shares it with the appropriate departments and stakeholders.

[0806] Emotion recognition by emotion engine

[0807] When a user reviews or modifies a draft report, the server uses an emotion engine to analyze the user's emotions. For example, it can analyze the user's facial expressions from camera footage and recognize that the user is feeling stressed.

[0808] Emotion-based feedback

[0809] Based on the results of the user's sentiment analysis, the server instructs the AI ​​to adjust the expression or generate additional comments. For example, if the user is feeling stressed, the AI ​​is instructed to simplify the explanation.

[0810] In this way, by linking the server, terminal, and user and making full use of generative AI and an emotion engine, a system is realized that allows users to create fault reports efficiently and with little burden.

[0811] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0812] Step 1:

[0813] A user logs into the system and uploads data

[0814] Input: User ID and password, past sample report (e.g., "Sample Report_2022.pdf"), current event data (e.g., "Error Log_2023-10-10.txt")

[0815] Specific behavior:

[0816] A user enters a user ID and password to log in to the system from a terminal.

[0817] After successfully logging in, the user uploads "Sample Report_2022.pdf" and "Error Log_2023-10-10.txt" to the system.

[0818] Output: Uploaded sample report and event data

[0819] Step 2:

[0820] The server preprocesses the uploaded data

[0821] Input: Uploaded sample reports and event data

[0822] Specific behavior:

[0823] The server extracts text from "SampleReport_2022.pdf".

[0824] Similarly, extract the necessary log entries from "Fault Log_2023-10-10.txt".

[0825] To make the extracted text data easier to analyze, unnecessary line breaks and spaces are removed and the data is converted to JSON format.

[0826] Output: Formatted sample report data and event data (JSON format)

[0827] Step 3:

[0828] The server collates and analyzes past sample reports with current event data.

[0829] Input: Sample report data and event data in JSON format

[0830] Specific behavior:

[0831] The server extracts the difference between the new error code "503" and the previous error code "403".

[0832] The server identifies commonalities in the scope of impact and time of occurrence.

[0833] Output: Information on the difference between extracted error codes, the scope of impact, and the time of occurrence

[0834] Step 4:

[0835] The server instructs the generation AI to generate a draft report.

[0836] Input: Information on the difference between the extracted error codes, the scope of the impact, and the time of occurrence

[0837] Specific behavior:

[0838] The server provides the AI ​​with a specific prompt, such as "An error code 503 occurred on October 10, 2023, causing service C to be down for 30 minutes."

[0839] Output: Instructions to the generating AI

[0840] Step 5:

[0841] Generative AI generates a draft report

[0842] Input: Instructions to the generation AI

[0843] Specific behavior:

[0844] Based on the instructions, the generative AI generates a draft report containing detailed text.

[0845] Specifically, a draft report will be created stating, "Server A failed on October 10, 2023. Service C was stopped for 30 minutes due to error code 503. The cause was a memory leak, and strengthening memory management is required as a countermeasure."

[0846] Output: Draft report

[0847] Step 6:

[0848] The user checks the draft report and makes corrections

[0849] Input: Generated report draft

[0850] Specific behavior:

[0851] The user checks the report draft from the terminal.

[0852] The user corrects technical details and specific terminology to complete the final report.

[0853] Save the completed report and share it with appropriate parties.

[0854] Output: Final report

[0855] Step 7:

[0856] The server uses an emotion engine to recognize the user's emotions.

[0857] Input: Camera footage and audio data of the user reviewing the draft report

[0858] Specific behavior:

[0859] The server analyzes camera footage and audio data and recognizes emotions from the user's facial expressions and tone of voice.

[0860] For example, it determines whether the user is feeling stressed.

[0861] Output: User's emotional state

[0862] Step 8:

[0863] The server provides emotion-based feedback to the generative AI.

[0864] Input: User's emotional state

[0865] Specific behavior:

[0866] The server instructs the generation AI to adjust the expression of the report draft based on the user's emotional state.

[0867] For example, if the user is feeling stressed, the AI ​​generator can be instructed to keep explanations brief.

[0868] Output: Instructions to regenerate the draft report reflecting the adjusted feedback

[0869] The above is the specific processing flow of the system program.

[0870] (Application example 2)

[0871] 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."

[0872] In customer service at brick-and-mortar stores, employees are required to respond quickly and appropriately to a variety of customer requests and complaints. However, it can be difficult for employees to generate high-quality reports, especially during busy or stressful times, which can result in a decrease in customer satisfaction. Furthermore, responses that do not take employees' feelings into consideration can increase their own stress and, in the long term, lead to a decrease in job satisfaction and performance. A solution to these issues is needed.

[0873] 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.

[0874] In this invention, the server includes means for receiving past sample reports and current event data from a user, means for preprocessing the received data and converting it into an easily analyzable format, means for comparing the past sample reports with the current event data and extracting similarities and differences, means for instructing a generation AI to generate a draft report based on the extracted information, means for presenting the draft report generated by the generation AI to the user, means including an emotion engine for recognizing the user's emotions, and means for providing feedback based on the user's emotions. This enables employees to generate customer interaction reports quickly and with high quality, and also enables them to receive feedback that takes into consideration the employee's emotions.

[0875] "User" means a user who utilizes the System to provide past sample reports and current event data.

[0876] "Past sample reports" refer to reports that have been prepared and stored in the past, and are used as reference materials to guide responses to new events.

[0877] "Current Event Data" refers to data that includes information about ongoing events, new requests, complaints, etc.

[0878] "Preprocessing" refers to the process of removing unnecessary information from received data and converting it into a format suitable for analysis.

[0879] "Generative AI" refers to artificial intelligence that automatically generates draft reports based on specified instructions.

[0880] A "draft" is an initial report created by the generative AI, which is subject to user review and correction.

[0881] An "emotion engine" is a system that includes technology that recognizes a user's emotions by analyzing the user's facial expressions and tone of voice.

[0882] "Providing feedback" means presenting appropriate comments and advice that take into consideration the user's emotions based on the analysis results of the emotion engine.

[0883] This invention provides a system for providing fast and high-quality customer service in brick-and-mortar stores. The system receives past sample reports and current event data, analyzes the data, and automatically generates a report draft. It also recognizes the user's emotions and provides feedback based on the emotions.

[0884] System Overview

[0885] The system consists of the following main components:

[0886] User Input Method

[0887] Data preprocessing measures

[0888] Data collation method

[0889] Generation AI instruction means

[0890] Generation AI

[0891] Emotion Engine

[0892] Feedback methods

[0893] Hardware and software used

[0894] Hardware:

[0895] PC or smart device (with camera)

[0896] software:

[0897] Python

[0898] spaCy (natural language processing)

[0899] OpenAI API (generative AI model)

[0900] OpenCV (for emotion analysis)

[0901] Program processing overview

[0902] The system processes data in the following order, generates a draft report, checks and corrects it, and finally provides feedback based on the user's feelings.

[0903] User Input Method

[0904] Users input past sample reports and current event data into the system.

[0905] Data preprocessing measures

[0906] The server preprocesses the data it receives, removing unnecessary line breaks and spaces from the text and converting it into a format that is easier to parse.

[0907] Data collation method

[0908] The server compares past sample reports with current event data and extracts commonalities and differences, providing the basis for the generation AI to generate reports.

[0909] Generation AI instruction means

[0910] The server issues instructions to the generation AI based on the extracted data to generate a draft report. An example of the prompt used at this time is shown below.

[0911] Example prompt sentence:

[0912] Common points: {Common points}, Differences: {Differences}. Please generate a customer service response report that addresses the new request effectively and improves customer satisfaction.

[0913] Generation AI

[0914] The generation AI generates a draft report based on instructions from the server. Specifically, it documents appropriate responses and details of customer responses based on similarities and differences.

[0915] Emotion Engine

[0916] When a user reviews or modifies a draft report, the emotion engine analyzes camera footage and audio data to recognize emotions from the user's facial expressions and tone of voice.

[0917] Feedback methods

[0918] The feedback provision means instructs the generation AI to adjust expressions or generate additional comments based on the analysis results of the emotion engine, ultimately generating a report that takes the user's emotions into consideration.

[0919] Specific examples

[0920] 1. User input:

[0921] Example: A user uploads "Sample Report_2022.pdf" and "Customer Request_2023-10-10.txt" to the system.

[0922] 2. Data preprocessing:

[0923] Example: The server extracts text from "Sample Report_2022.pdf" and necessary information from "Customer Request_2023-10-10.txt." It then removes unnecessary line breaks and spaces from the extracted data and converts it into a format suitable for analysis.

[0924] 3. Data collation and analysis:

[0925] Example: The server compares the current customer request with past sample reports and extracts the differences between the new request and the past response history.

[0926] 4. Instructions to the generating AI:

[0927] Example: The server instructs the generation AI to generate a draft report based on "commonalities: maintaining past services, differences: new requests."

[0928] 5. Generate a draft report:

[0929] Example: The generation AI generates a draft report stating, "A new request was received from the customer on October 10, 2023. A response is required. Based on past response history and the new request, we suggest the following response."

[0930] 6. Emotion Recognition with Emotion Engine:

[0931] Example: While a user is reviewing and revising a draft report, the camera analyzes their facial expressions and recognizes that they are feeling stressed.

[0932] 7. Emotion-based feedback:

[0933] Example: The server instructs the generated AI to simplify the explanation and emphasize the important parts to reduce stress for the user.

[0934] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0935] Step 1:

[0936] The user uploads past sample reports and current event data from a terminal to the system. The input is a past sample report file and a current event data file, and the output is a confirmation of receipt of the uploaded files. The server receives these files and proceeds to the next step.

[0937] Step 2:

[0938] The server preprocesses the uploaded data and converts it into a format that is easy to analyze. This processing includes extracting text and removing unnecessary line breaks and spaces. The input is the uploaded sample report file and event data file, and the output is formatted text data. Specifically, it uses the Python spaCy library to remove unnecessary information from the text and convert the data into a structured format such as JSON.

[0939] Step 3:

[0940] The server uses the preprocessed data to match past sample reports with the current event data and extract commonalities and differences. The input is the formatted past sample report data and the current event data, and the output is a list of extracted commonalities and differences. Specifically, it uses text mining techniques to identify common keywords and phrases, as well as new elements, between the past reports and the new data.

[0941] Step 4:

[0942] The server instructs the generative AI to generate a draft report based on the extracted data. The input is a list of commonalities and differences, and the output is a prompt for the generative AI. The prompt is sent to the generative AI model (for example, OpenAI's API) to generate a detailed draft report. An example of a specific prompt is "Common points: {Common points}, Differences: {Differences}. Please generate a customer service response report that addresses the new request effectively and improves the customer satisfaction."

[0943] Step 5:

[0944] The generation AI generates a draft report based on prompts from the server and sends it back to the server. The input is the prompt text passed from the server, and the output is the generated draft report. Specifically, the generation AI automatically creates a detailed report of the customer interaction based on similarities and differences.

[0945] Step 6:

[0946] The user checks and modifies the generated report draft. To do this, they open the draft using a terminal and modify the content as necessary. The input is the generated report draft, and the output is the modified final report. The user checks the technical details and terminology of the report and makes any necessary modifications based on their own expertise.

[0947] Step 7:

[0948] The server uses an emotion engine to analyze the user's emotions when reviewing and revising a report draft. The input is video and audio data of the user's face, and the output is the analysis result of the user's emotional state. Specifically, it uses image analysis libraries such as OpenCV to recognize emotions from the user's facial expressions and tone of voice.

[0949] Step 8:

[0950] Based on the analysis results of the emotion engine, the server instructs the generation AI to provide feedback. The input is the user's emotion analysis results, and the output is feedback on the adjusted report and additional comments. Specifically, the generation AI readjusts the content of the report according to the user's emotions, providing concise explanations and emphasizing important parts to reduce stress. This allows the user to more efficiently finalize the report.

[0951] 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.

[0952] 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.

[0953] 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.

[0954] [Third embodiment]

[0955] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0956] 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.

[0957] 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).

[0958] 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.

[0959] 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.

[0960] 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).

[0961] 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.

[0962] 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.

[0963] 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.

[0964] 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.

[0965] 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.

[0966] 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."

[0967] System Overview

[0968] The present invention is a system for automatically generating fault reports based on user-provided sample reports and current event data. The system includes the following major functions:

[0969] Receiving data

[0970] Data Preprocessing

[0971] Data collation and analysis

[0972] Automatic generation of report drafts

[0973] User confirmation and correction

[0974] Program processing overview

[0975] 1. User Input

[0976] Users log into the system from a terminal and upload past sample reports and current event data.

[0977] 2. Data Preprocessing

[0978] The server receives the uploaded data and performs text extraction and formatting, specifically removing unnecessary line breaks and spaces and converting the data into a format that is easy to analyze.

[0979] 3. Data Collation and Analysis

[0980] The server compares past sample reports with current event data to extract commonalities and differences, including characteristic features such as error codes, the time of the failure, and the scope of the impact.

[0981] 4. Instructions to the generating AI

[0982] Based on the extracted data, the server instructs the AI ​​to generate a draft report, which includes an overview of the problem, its scope of impact, causes, and countermeasures.

[0983] 5. Generate a draft report

[0984] Based on instructions from the server, the generation AI creates a draft report detailing the problem, which is then sent back to the server in a format that can be reviewed and modified by the user.

[0985] 6. Checking and correcting the report

[0986] The user reviews the generated draft report and makes any necessary corrections. Once the final report is complete, the user saves it and shares it with the appropriate departments and stakeholders.

[0987] Specific examples

[0988] 1. User Input

[0989] Example: A user uploads "Sample Report_2022.pdf" and "Fault Log_2023-10-10.txt" to the system.

[0990] 2. Data Preprocessing

[0991] Example: The server extracts text from "Sample Report_2022.pdf" and the necessary log entries from "Error Log_2023-10-10.txt." It then removes unnecessary line breaks and spaces from the extracted data and converts it to JSON format.

[0992] 3. Data Collation and Analysis

[0993] Example: The server compares the sample report from 2022 with the error log from 2023, extracting the differences between the new error code "503" and the past error code "403." At the same time, it identifies commonalities in the scope of impact and occurrence time.

[0994] 4. Instructions to the generating AI

[0995] Example: The server instructs the AI ​​generation system that "An error code 503 occurred on October 10, 2023, causing service C to stop for 30 minutes." This instruction also includes a detailed summary of the error, the scope of its impact, the cause, and countermeasures.

[0996] 5. Generate a draft report

[0997] Example: The generation AI generates a draft report stating, "Server A failed on October 10, 2023. Service C was stopped for 30 minutes due to error code 503. The cause was a memory leak, and as a countermeasure, improved memory management is required."

[0998] 6. Checking and correcting the report

[0999] Example: User reviews the generated draft, corrects technical details and specific terminology to complete the final report, saves the completed report and shares it with the appropriate stakeholders.

[1000] In this way, the system can generate trouble reports efficiently and accurately, reducing the workload of the user.

[1001] The processing flow will be explained below.

[1002] Step 1:

[1003] The user logs in to the system from a terminal and goes to the upload page. The user selects the files of past sample reports and current event data and clicks the upload button.

[1004] Step 2:

[1005] The server receives the uploaded sample reports and event data, which are temporarily stored and prepared for further processing.

[1006] Step 3:

[1007] The server performs a parsing process to extract text from the uploaded sample reports, specifically extracting the required text information from PDFs and other file formats.

[1008] Step 4:

[1009] The server extracts the necessary information from the event data (e.g., log data), specifically by parsing and extracting important log entries such as the target error code and timestamp.

[1010] Step 5:

[1011] The server removes unnecessary line breaks and spaces from the extracted data and converts the data into a format that is easy to parse (e.g., JSON format).

[1012] Step 6:

[1013] The server compares the current event data with past sample reports, extracting commonalities (e.g., error codes, scope of impact) and differences (e.g., new error codes, different scope of impact).

[1014] Step 7:

[1015] Based on the results of the comparison, the server creates instructions for the generation AI to generate a draft report, including an overview of the problem, its scope of impact, causes, and countermeasures.

[1016] Step 8:

[1017] The generation AI generates a draft report based on instructions received from the server. The generation AI automatically generates detailed text for these items and provides feedback.

[1018] Step 9:

[1019] The generated report draft is sent back to the server and presented to the user, who can then check the generated draft through the terminal.

[1020] Step 10:

[1021] The user reviews the draft and makes any necessary corrections, such as changing technical terminology, adding details, or correcting typos.

[1022] Step 11:

[1023] The user saves the final report and then shares it with the appropriate departments and stakeholders, completing the process of generating a problem report.

[1024] Example 1

[1025] 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."

[1026] Creating a failure report manually is extremely time-consuming and prone to errors and oversights. The process of creating a report by comparing and analyzing past reports with current event data is particularly labor-intensive, making it necessary to create reports efficiently and accurately. Furthermore, reports must include a detailed failure summary, scope of impact, cause, and countermeasures, but there is currently a lack of a means to automatically generate standardized reports that include all of these.

[1027] 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.

[1028] In this invention, the server includes means for receiving past sample reports and current event data from a user, means for converting the received data into a format that is easy to analyze by extracting text and formatting, means for comparing the past sample reports and the current event data using natural language processing and machine learning algorithms to extract similarities and differences, means for providing a prompt to the generative AI model to generate a draft report based on the extracted information, and means for presenting the draft report generated by the generative AI model to the user in a format that can be confirmed and modified, thereby enabling the user to generate and modify reports efficiently and accurately.

[1029] "User" refers to a person or organization that uses a terminal to access the system and provide various data.

[1030] "Sample Report" means a document that contains a detailed record of a past event.

[1031] "Event Data" refers to data relating to a particular event that is currently occurring or has recently occurred.

[1032] "Text extraction" refers to the process of extracting character data from a rich text format file.

[1033] "Formatting" refers to the process of removing unnecessary line breaks and spaces from data and converting it into a format that is easier to parse.

[1034] "Natural language processing (NLP)" refers to technology for analyzing text data and understanding its meaning and structure.

[1035] A "machine learning algorithm" refers to a method of learning patterns and rules from large amounts of data and using that knowledge to analyze new data.

[1036] "Matching" refers to the process of finding similarities and differences between different data sets.

[1037] A "generative AI model" refers to an artificial intelligence system that automatically generates text or other data based on instructions it is given.

[1038] A "prompt sentence" refers to a textual input sentence that provides instructions to a generative AI model.

[1039] "Draft Report" refers to the initial version of the report automatically generated by the generative AI model.

[1040] "Format" refers to the particular structure or format of data or documents.

[1041] "Verification" refers to the process in which a user reviews the generated report draft and checks its content.

[1042] "Modification" refers to the user making any necessary changes or additions to the draft report.

[1043] The present invention is a system for automatically generating fault reports based on user-provided sample reports and current event data. The system can be implemented using the following major hardware and software components:

[1044] Hardware and software used

[1045] Terminal: A device used by a user to access the system and enter data. This includes devices such as PCs and smartphones.

[1046] Server: A central computing unit that preprocesses data, analyzes it, and directs generative AI models.

[1047] A text analysis library, such as Apache Tika, can be used to extract text from PDF files.

[1048] Use an NLP library such as spaCy to analyze text data and extract the necessary information.

[1049] Generative AI models: Generative AI models such as GPT-4 are used to generate draft reports.

[1050] Data processing and analysis

[1051] 1. User Input

[1052] A user logs in to the system from a terminal and uploads past sample reports and current event data, for example, uploading a report in PDF format and a fault log in text format.

[1053] Specifically, the user uploads "Sample Report_2022.pdf" and "Error Log_2023-10-10.txt" to the system.

[1054] 2. Data Preprocessing

[1055] The server receives the uploaded data and performs text extraction and formatting.

[1056] Specifically, we use Apache Tika to extract text from "SampleReport_2022.pdf", select the required log entries from the TXT file, remove unnecessary line breaks and spaces, and convert the data to JSON format.

[1057] 3. Data Collation and Analysis

[1058] The server uses the extracted data to compare and analyze past sample reports with current event data.

[1059] Specifically, spaCy is used to extract information such as error codes, dates and times, and the extent of the impact, allowing past and present data to be compared to identify commonalities and differences.

[1060] 4. Instructions to the generating AI

[1061] Based on the analysis results, the server instructs the generative AI model to generate a draft report.

[1062] An example of a specific prompt might be, "An error code 503 occurred on October 10, 2023, causing service C to be down for 30 minutes." This instruction also includes a detailed summary of the error, the scope of its impact, the cause, and countermeasures.

[1063] 5. Generate a draft report

[1064] The generative AI model generates a draft report detailing the fault based on instructions from the server, which is then sent back to the server in a format that can be reviewed and modified by the user.

[1065] Specifically, it generates a draft report stating something like, "Server A failed on October 10, 2023. Service C stopped for 30 minutes due to error code 503. The cause was a memory leak, and as a countermeasure, improved memory management is required."

[1066] 6. Checking and correcting the report

[1067] The user reviews the generated draft report, makes any necessary revisions, and finally saves the final report and shares it with the appropriate departments and agencies.

[1068] Specifically, users can access the draft from their devices, edit specific technical information and terminology, and then finalize the report, which is then saved and sent to stakeholders via email or a shared link.

[1069] In this way, the present invention can efficiently and accurately generate trouble reports, significantly reducing the user's workload. This system can be implemented using specific hardware and software, making it applicable to a variety of situations.

[1070] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1071] Step 1: User Input

[1072] A user logs in to the system from a terminal and uploads a past sample report (PDF format) and current event data (TXT format). The input data is "Sample Report_2022.pdf" and "Fault Log_2023-10-10.txt."

[1073] Specific behavior:

[1074] The user enters their username and password on the login page to access the system.

[1075] After logging in, select the "File Upload" option on your dashboard, select the required file and upload it.

[1076] If the upload is successful, the file is sent to the server.

[1077] Step 2: Preprocessing the data

[1078] The server receives the uploaded data, extracts text, and formats it. The input data is PDF and TXT files, and the output data is in an easy-to-parse JSON format.

[1079] Specific behavior:

[1080] The server uses the Apache Tika library to extract text from the PDF file "SampleReport_2022.pdf".

[1081] At the same time, the server selects the necessary log entries from the TXT file "Error Log_2023-10-10.txt".

[1082] Unnecessary line breaks and spaces are removed from this text data, and the formatted text data is converted into JSON format.

[1083] Step 3: Collate and analyze data

[1084] The server uses the extracted data to compare and analyze past sample reports with current event data. The input is text data in JSON format, and the output is information about notable similarities and differences.

[1085] Specific behavior:

[1086] The server uses an NLP library such as spaCy to extract information such as the error code, date and time, and scope of impact from the text data.

[1087] Based on the extracted information, past reports are compared with the current fault log to identify similarities and differences.

[1088] As a result, information such as differences in error codes and commonalities in the scope of impact is generated and passed on to the next step.

[1089] Step 4: Instructions to the Generative AI

[1090] The server instructs the generative AI model to generate a draft report based on the analysis results. The input is information about similarities and differences, and the output is a prompt statement for the generative AI model.

[1091] Specific behavior:

[1092] The server formalizes the analysis results as prompt sentences and provides them to the generative AI model.

[1093] As an example of a specific prompt sentence, a sentence such as "An error code 503 occurred on October 10, 2023, causing service C to be down for 30 minutes" is generated.

[1094] The prompt text includes a detailed overview of the problem, the scope of impact, the cause, and countermeasures.

[1095] Step 5: Generate a draft report

[1096] The generative AI model generates a draft report based on the prompt provided by the server. The input is the prompt to the generative AI, and the output is the draft report.

[1097] Specific behavior:

[1098] The generative AI model receives the prompt and creates a draft report.

[1099] The generated report draft is generated in Markdown, HTML, Word document format, etc. and sent back to the server.

[1100] The draft includes content such as, "Server A failed on October 10, 2023. Service C was stopped for 30 minutes due to error code 503. The cause was a memory leak, and as a countermeasure, improved memory management is required."

[1101] Step 6: Check and correct the report

[1102] The user reviews the generated report draft and makes any necessary corrections. The input is the report draft and the output is the final report.

[1103] Specific behavior:

[1104] The user accesses the draft from a terminal and checks its contents.

[1105] Correct specific technical information and terminology as needed to complete the final report.

[1106] Completed reports are stored within the system and then sent to the appropriate parties via email or shared link.

[1107] In this way, the system can automatically generate trouble reports efficiently and accurately based on user-supplied data.

[1108] (Application example 1)

[1109] 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."

[1110] Conventional factory robot monitoring systems require human intervention when a failure occurs, making it difficult to respond quickly and create accurate reports. Furthermore, manually comparing past failure reports with current event data and generating reports is time-consuming and labor-intensive. These issues can significantly impact factory operational efficiency and product quality, so a system that solves these problems is needed.

[1111] 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.

[1112] In this invention, the server includes means for receiving past sample reports and current event data from a user, means for preprocessing the received data and converting it into an easily analyzable format, means for comparing the past sample reports with the current event data and extracting similarities and differences, means for instructing the generation AI to generate a draft report based on the extracted information, means for presenting the draft report generated by the generation AI to the user, and means for monitoring the operating status of factory robots in real time and automatically generating a report when an abnormality occurs. This makes it possible to respond quickly and accurately to factory robot failures and improve factory operating efficiency.

[1113] "User" means a person or entity that uses the System to provide and query past sample reports and current event data.

[1114] A "Past Sample Report" is a document that contains detailed information about past events or failures.

[1115] "Current event data" refers to data relating to events or failures occurring at the present time.

[1116] "Preprocessing" is the process of deleting unnecessary line breaks and spaces and reformatting the data in order to convert the received data into a format that is easy to analyze.

[1117] "Matching" is the process of comparing past sample reports with current event data to extract similarities and differences.

[1118] "Generative AI" is an artificial intelligence system that automatically generates draft reports based on given data and instructions.

[1119] A "report draft" is a draft of a report automatically generated by the generation AI.

[1120] A "factory robot" is a mechanical device used to perform automated tasks in manufacturing.

[1121] "Operation status" is information that indicates the state in which a factory robot is operating.

[1122] "Real-time monitoring" refers to the continuous observation of a situation occurring at the present time.

[1123] An "abnormality" is a deviation from normal operation or failure of a system or machine.

[1124] A "report" is a formal document containing detailed information, causes, and actions regarding a specific incident or situation.

[1125] This invention is a system that monitors the operating status of factory robots in real time and automatically generates a fault report when an abnormality occurs. This system creates reports with high accuracy and efficiency based on past sample reports and current event data provided by the user.

[1126] 1. System Configuration

[1127] The system includes a server, user terminals, factory robots, generative AI models, and front-end and back-end processing. The server receives data, preprocesses it, collates it, and issues instructions to the generative AI. The user terminal uploads data and checks and edits reports via a smartphone application. The factory robots are monitored for their operating status and provide data in real time.

[1128] 2. Hardware / Software to be used

[1129] Hardware: smartphones, servers, factory robots

[1130] software:

[1131] Server: Flask (web framework), SQLite (database)

[1132] Preprocessing: Text extraction and data formatting using Python, NLTK (Natural Language Toolkit)

[1133] Generation AI: OpenAI GPT-3

[1134] Frontend: HTML / CSS / JavaScript

[1135] 3. Specific examples of program processing

[1136] Data reception and preprocessing

[1137] Users upload past sample reports and current event data from their devices, and the server receives the data, then uses Python to extract and format the text, removing unnecessary line breaks and spaces, and converting it into a format that is easier to parse.

[1138] Data collation and analysis

[1139] The preprocessed data is compared with past sample reports and current event data to extract commonalities and differences, including characteristic information such as error codes, time of occurrence, and scope of impact.

[1140] Instructions for generative AI models

[1141] The server instructs a generative AI model (e.g., OpenAI GPT-3) to generate a draft report based on the extracted data, including a summary of the problem, its scope of impact, causes, and countermeasures.

[1142] Draft report generation and presentation

[1143] The generative AI model generates a draft report based on the data and prompts received from the server, which is then sent back to the server and presented to the user, who can make any necessary corrections to produce a final fault report.

[1144] Examples of concrete examples and prompts

[1145] When a user uploads "Sample Report_2022.pdf" and "Fault Log_2023-10-10.txt" to the system, the system automatically analyzes these data, extracts commonalities and differences, and generates a draft report. The generated draft contains the following content:

[1146] Example prompt sentence:

[1147] "Based on past incident reports and the current incident log, we will provide the following information. Please use this information to generate a detailed incident report."

[1148] This system will enable rapid and accurate response to factory robot failures, significantly improving factory operational efficiency.

[1149] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1150] Step 1:

[1151] The user uploads past sample reports and current event data from their device. Specifically, the user uses a smartphone application to send "Sample Report_2022.pdf" and "Fault Log_2023-10-10.txt" to the system. The input is a PDF and text file, and these files are saved on the server as output.

[1152] Step 2:

[1153] The data received by the server is preprocessed. Specifically, text is extracted using Python, and unnecessary line breaks and spaces are removed to convert it into a format that is easy to parse. The input is the uploaded PDF and text file, and the output is formatted text data. For example, the text extracted from the PDF and the contents of the text file are converted into JSON format.

[1154] Step 3:

[1155] The server compares past sample reports with current event data to extract commonalities and differences. Specifically, it uses a scripting language (Python) to compare characteristic information such as error codes, the time of the failure, and the scope of impact. The input is preprocessed text data, and the output is data containing commonalities and differences. For example, it extracts the differences between "error code 503" and "error code 403."

[1156] Step 4:

[1157] Based on the information extracted by the server, the generative AI model is instructed to generate a draft report. Specifically, OpenAI GPT-3 is used to generate a prompt and make a request to generate a draft report. The input is data including similarities and differences, and the output is a prompt to the generative AI and a request to generate a draft report. For example, the command could be, "An error code 503 failure occurred on October 10, 2023, causing service C to be down for 30 minutes."

[1158] Step 5:

[1159] The generative AI model generates a draft report based on the instructions and sends it back to the server. Specifically, OpenAI GPT-3 generates a draft report based on a prompt and sends it back to the server. The input is the prompt, and the output is the generated draft report. For example, a draft report might be generated that states, "Server A failed on October 10, 2023. Service C was stopped for 30 minutes due to error code 503. The cause was a memory leak, and improved memory management is required as a countermeasure."

[1160] Step 6:

[1161] The server presents the generated draft report to the user, who makes any necessary revisions. Specifically, the report draft is displayed to the user through a smartphone application, allowing the user to modify technical details and specific terminology. The input is the generated draft report, and the output is the modified final report. The user saves the modified report and shares it with the appropriate parties.

[1162] In this way, the operating status of the factory robot can be monitored in real time, and a fault report can be generated quickly and accurately when an abnormality occurs.

[1163] 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.

[1164] System Overview

[1165] The present invention is a system that automatically generates trouble reports based on past sample reports provided by the user and current event data, and further combines it with an emotion engine that recognizes the user's emotions. The system has the following main functions:

[1166] Receiving data

[1167] Data Preprocessing

[1168] Data collation and analysis

[1169] Automatic generation of report drafts

[1170] User confirmation and correction

[1171] User Emotion Recognition and Feedback

[1172] Program processing overview

[1173] 1. User Input

[1174] Users log into the system from a terminal and upload past sample reports and current event data.

[1175] 2. Data Preprocessing

[1176] The server receives the uploaded data and performs text extraction and formatting, specifically removing unnecessary line breaks and spaces and converting the data into a format that is easy to analyze.

[1177] 3. Data Collation and Analysis

[1178] The server compares past sample reports with current event data to extract commonalities and differences, including characteristic features such as error codes, the time of the failure, and the scope of the impact.

[1179] 4. Instructions to the generating AI

[1180] Based on the extracted data, the server instructs the AI ​​to generate a draft report, which includes an overview of the problem, its scope of impact, causes, and countermeasures.

[1181] 5. Generate a draft report

[1182] The generation AI generates a draft report based on instructions from the server. The generation AI automatically generates detailed text for these items and sends it back to the server.

[1183] 6. Checking and correcting the report

[1184] The user reviews the generated draft report and makes any necessary corrections. Once the final report is complete, the user saves it and shares it with the appropriate departments and stakeholders.

[1185] 7. Emotion Recognition with Emotion Engine

[1186] When a user reviews or edits a draft report, the server uses an emotion engine to analyze the user's emotions. Emotion recognition is achieved by analyzing the user's facial expressions and tone of voice using camera footage and audio data.

[1187] 8. Emotional Feedback

[1188] Based on the results of the user's sentiment analysis, the server instructs the AI ​​to adjust the expression or generate additional comments. For example, if the user is feeling stressed, the AI ​​will simplify the explanation.

[1189] Specific examples

[1190] 1. User Input

[1191] Example: A user uploads "Sample Report_2022.pdf" and "Fault Log_2023-10-10.txt" to the system.

[1192] 2. Data Preprocessing

[1193] Example: The server extracts text from "Sample Report_2022.pdf" and the necessary log entries from "Error Log_2023-10-10.txt." It then removes unnecessary line breaks and spaces from the extracted data and converts it to JSON format.

[1194] 3. Data Collation and Analysis

[1195] Example: The server compares the sample report from 2022 with the error log from 2023, extracting the differences between the new error code "503" and the past error code "403." At the same time, it identifies commonalities in the scope of impact and occurrence time.

[1196] 4. Instructions to the generating AI

[1197] Example: The server instructs the AI ​​generation system that "An error code 503 occurred on October 10, 2023, causing service C to stop for 30 minutes." This instruction also includes a detailed summary of the error, the scope of its impact, the cause, and countermeasures.

[1198] 5. Generate a draft report

[1199] Example: The generation AI generates a draft report stating, "Server A failed on October 10, 2023. Service C was stopped for 30 minutes due to error code 503. The cause was a memory leak, and as a countermeasure, improved memory management is required."

[1200] 6. Checking and correcting the report

[1201] Example: User reviews the generated draft, corrects technical details and specific terminology to complete the final report, saves the completed report and shares it with the appropriate stakeholders.

[1202] 7. Emotion Recognition with Emotion Engine

[1203] Example: While a user is reviewing a report, the emotion engine analyzes the user's facial expressions from camera footage and recognizes that the user is feeling stressed.

[1204] 8. Emotional Feedback

[1205] Example: The server instructs the generated AI to make explanations more concise and emphasize important parts to reduce user stress. Feedback is provided based on the analysis of the emotion engine.

[1206] In this way, by utilizing the emotion engine, the system can reduce the burden on users and provide a more user-friendly process for creating trouble reports.

[1207] The processing flow will be explained below.

[1208] Step 1:

[1209] The user logs in to the system from a terminal and goes to the upload page. The user selects the files of past sample reports and current event data and clicks the upload button.

[1210] Step 2:

[1211] The server receives the uploaded sample reports and event data, which are temporarily stored and prepared for further processing.

[1212] Step 3:

[1213] The server runs a parsing process to extract text from the sample reports, specifically extracting the necessary text information from PDFs and other file formats and converting it into a format that is easier to parse.

[1214] Step 4:

[1215] The server extracts the necessary information from the event data (e.g., log data), specifically by parsing and extracting important log entries such as the target error code and timestamp.

[1216] Step 5:

[1217] The server removes unnecessary line breaks and spaces from the extracted data and converts the data into a unified format (e.g., JSON format).

[1218] Step 6:

[1219] The server compares the current event data with past sample reports, extracting commonalities (e.g., error codes, scope of impact) and differences (e.g., new error codes, different scope of impact).

[1220] Step 7:

[1221] Based on the results of the comparison, the server creates instructions for the generation AI to generate a draft report, including an overview of the problem, its scope of impact, causes, and countermeasures.

[1222] Step 8:

[1223] The generation AI generates a report draft based on instructions received from the server. The generation AI automatically generates detailed text for these items and sends the completed draft back to the server.

[1224] Step 9:

[1225] The server displays the generated report draft on the user's terminal, where the user can check the report draft and make corrections as necessary.

[1226] Step 10:

[1227] When a user checks a report draft, the server uses an emotion engine to recognize the user's emotions. The emotion engine analyzes the user's facial expressions and tone of voice, which are captured by a camera and microphone.

[1228] Step 11:

[1229] The server determines the user's emotional state (e.g., stress, dissatisfaction, relief) based on the analysis results of the emotion engine, and instructs the generation AI to adjust the content and expression of the report draft depending on the results.

[1230] Step 12:

[1231] Based on the analysis results of the emotion engine, the server displays feedback and alerts to the user. For example, if the user is feeling stressed, a comment is added to the report to simplify the explanation.

[1232] Step 13:

[1233] The user completes the final report and saves it on the server, where it is shared with the appropriate departments and stakeholders.

[1234] Step 14:

[1235] Users can effectively use the feedback from the emotion engine to improve the report creation process and identify areas for improvement. Through this process, adaptive report creation that takes emotions into consideration is realized.

[1236] Example 2

[1237] 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."

[1238] Conventional problem report creation systems often require a great deal of time and effort for users to check and edit reports, placing a heavy burden on users. Furthermore, because they do not take into account the user's emotional state, they can easily become stressed or fatigued. Furthermore, because the content of the generated reports is uniform, they sometimes do not function adequately even in situations where flexible responses according to the situation are required.

[1239] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1240] In this invention, the server includes means for receiving past sample reports and current event data from a user, means for preprocessing the received data and converting it into an easily analyzable format, means for comparing the past sample reports with the current event data and extracting similarities and differences, means for instructing a generation AI to generate a draft report based on the extracted information, means for presenting the report draft generated by the generation AI to the user, means for accepting user confirmation and correction of the report draft, means for recognizing the user's emotions, and means for adjusting the content of the report draft based on the recognized emotions. This reduces the burden on the user, streamlines the report creation process, and enables flexible report generation that corresponds to the user's emotional state.

[1241] "Users" refer to people who utilize the system to provide historical sample reports and current event data, and who are responsible for reviewing and modifying the final report.

[1242] "Sample report" refers to a previously created incident report that serves as a basis for comparison with current event data.

[1243] "Event Data" means data, including logs and information related to a current fault or problem, that is used to generate reports.

[1244] "Preprocessing" refers to the process of removing unnecessary line breaks and spaces from the received data and formatting it into a format that is easy to analyze.

[1245] "Matching" refers to the process of comparing past sample reports with current event data to identify similarities and differences.

[1246] "Generative AI" refers to an artificial intelligence model that automatically generates a draft report based on instructions from a server.

[1247] A "Draft Report" is an early-stage problem report created by the generation AI, before it has been reviewed and corrected by the user.

[1248] An "emotion engine" refers to software or hardware for recognizing a user's emotional state, for example by analyzing the user's facial expressions and tone of voice using camera or audio data.

[1249] "Feedback" refers to the process of providing instructions or information to the generative AI to adjust the content of the draft report based on the user's emotional state.

[1250] The present invention is a system that automatically generates trouble reports based on past sample reports and current event data provided by the user. Furthermore, this system combines an emotion engine that recognizes the user's emotions, aiming to reduce the burden on the user and streamline the report creation process.

[1251] Hardware and Software Configuration

[1252] The system is implemented using the following major hardware and software:

[1253] Server: A central device that preprocesses data, collates it, and issues instructions to the generation AI.

[1254] Terminal: Provides a user interface for users to upload data and review and modify generated draft reports.

[1255] Generative AI: An artificial intelligence model that automatically generates report drafts.

[1256] Emotion engine: Software or hardware for recognizing a user's emotions, including, for example, a facial recognition camera or voice analysis software.

[1257] Examples of data processing and data calculation

[1258] User Input

[1259] The user logs into the system from a terminal and uploads a past sample report (e.g., "Sample Report_2022.pdf") and current event data (e.g., "Fault Log_2023-10-10.txt").

[1260] Data Preprocessing

[1261] The server extracts text from the uploaded data and converts it into a format that is easy to parse. Specifically, it removes unnecessary line breaks and spaces and converts it into JSON format. For example, the server extracts text from "Sample Report_2022.pdf" and extracts the necessary log entries from "Failure Log_2023-10-10.txt."

[1262] Data collation and analysis

[1263] The server compares past sample reports with current event data to identify commonalities and differences. For example, it can extract the differences between the new error code "503" and the previous error code "403" and identify commonalities in the scope of impact and time of occurrence.

[1264] Instructions to the generating AI

[1265] The server instructs the generation AI to generate a draft report based on the extracted data. For example, it provides a prompt such as, "An error code 503 occurred on October 10, 2023, causing service C to be down for 30 minutes."

[1266] Generate a draft report

[1267] Based on the server's instructions, the generation AI generates a detailed draft report and sends it back to the server. Specifically, it generates a draft report with content such as, "Server A failed on October 10, 2023. Service C stopped for 30 minutes due to error code 503. The cause was a memory leak, and as a countermeasure, improved memory management is required."

[1268] Checking and correcting reports

[1269] The user reviews the generated draft report and makes any necessary corrections. Once the final report is complete, the user saves it and shares it with the appropriate departments and stakeholders.

[1270] Emotion recognition by emotion engine

[1271] When a user reviews or modifies a draft report, the server uses an emotion engine to analyze the user's emotions. For example, it can analyze the user's facial expressions from camera footage and recognize that the user is feeling stressed.

[1272] Emotion-based feedback

[1273] Based on the results of the user's sentiment analysis, the server instructs the AI ​​to adjust the expression or generate additional comments. For example, if the user is feeling stressed, the AI ​​is instructed to simplify the explanation.

[1274] In this way, by linking the server, terminal, and user and making full use of generative AI and an emotion engine, a system is realized that allows users to create fault reports efficiently and with little burden.

[1275] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1276] Step 1:

[1277] A user logs into the system and uploads data

[1278] Input: User ID and password, past sample report (e.g., "Sample Report_2022.pdf"), current event data (e.g., "Error Log_2023-10-10.txt")

[1279] Specific behavior:

[1280] A user enters a user ID and password to log in to the system from a terminal.

[1281] After successfully logging in, the user uploads "Sample Report_2022.pdf" and "Error Log_2023-10-10.txt" to the system.

[1282] Output: Uploaded sample report and event data

[1283] Step 2:

[1284] The server preprocesses the uploaded data

[1285] Input: Uploaded sample reports and event data

[1286] Specific behavior:

[1287] The server extracts text from "SampleReport_2022.pdf".

[1288] Similarly, extract the necessary log entries from "Fault Log_2023-10-10.txt".

[1289] To make the extracted text data easier to analyze, unnecessary line breaks and spaces are removed and the data is converted to JSON format.

[1290] Output: Formatted sample report data and event data (JSON format)

[1291] Step 3:

[1292] The server collates and analyzes past sample reports with current event data.

[1293] Input: Sample report data and event data in JSON format

[1294] Specific behavior:

[1295] The server extracts the difference between the new error code "503" and the previous error code "403".

[1296] The server identifies commonalities in the scope of impact and time of occurrence.

[1297] Output: Information on the difference between extracted error codes, the scope of impact, and the time of occurrence

[1298] Step 4:

[1299] The server instructs the generation AI to generate a draft report.

[1300] Input: Information on the difference between the extracted error codes, the scope of the impact, and the time of occurrence

[1301] Specific behavior:

[1302] The server provides the AI ​​with a specific prompt, such as "An error code 503 occurred on October 10, 2023, causing service C to be down for 30 minutes."

[1303] Output: Instructions to the generating AI

[1304] Step 5:

[1305] Generative AI generates a draft report

[1306] Input: Instructions to the generation AI

[1307] Specific behavior:

[1308] Based on the instructions, the generative AI generates a draft report containing detailed text.

[1309] Specifically, a draft report will be created stating, "Server A failed on October 10, 2023. Service C was stopped for 30 minutes due to error code 503. The cause was a memory leak, and strengthening memory management is required as a countermeasure."

[1310] Output: Draft report

[1311] Step 6:

[1312] The user checks the draft report and makes corrections

[1313] Input: Generated report draft

[1314] Specific behavior:

[1315] The user checks the report draft from the terminal.

[1316] The user corrects technical details and specific terminology to complete the final report.

[1317] Save the completed report and share it with appropriate parties.

[1318] Output: Final report

[1319] Step 7:

[1320] The server uses an emotion engine to recognize the user's emotions.

[1321] Input: Camera footage and audio data of the user reviewing the draft report

[1322] Specific behavior:

[1323] The server analyzes camera footage and audio data and recognizes emotions from the user's facial expressions and tone of voice.

[1324] For example, it determines whether the user is feeling stressed.

[1325] Output: User's emotional state

[1326] Step 8:

[1327] The server provides emotion-based feedback to the generative AI.

[1328] Input: User's emotional state

[1329] Specific behavior:

[1330] The server instructs the generation AI to adjust the expression of the report draft based on the user's emotional state.

[1331] For example, if the user is feeling stressed, the AI ​​generator can be instructed to keep explanations brief.

[1332] Output: Instructions to regenerate the draft report reflecting the adjusted feedback

[1333] The above is the specific processing flow of the system program.

[1334] (Application example 2)

[1335] 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."

[1336] In customer service at brick-and-mortar stores, employees are required to respond quickly and appropriately to a variety of customer requests and complaints. However, it can be difficult for employees to generate high-quality reports, especially during busy or stressful times, which can result in a decrease in customer satisfaction. Furthermore, responses that do not take employees' feelings into consideration can increase their own stress and, in the long term, lead to a decrease in job satisfaction and performance. A solution to these issues is needed.

[1337] 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.

[1338] In this invention, the server includes means for receiving past sample reports and current event data from a user, means for preprocessing the received data and converting it into an easily analyzable format, means for comparing the past sample reports with the current event data and extracting similarities and differences, means for instructing a generation AI to generate a draft report based on the extracted information, means for presenting the draft report generated by the generation AI to the user, means including an emotion engine for recognizing the user's emotions, and means for providing feedback based on the user's emotions. This enables employees to generate customer interaction reports quickly and with high quality, and also enables them to receive feedback that takes into consideration the employee's emotions.

[1339] "User" means a user who utilizes the System to provide past sample reports and current event data.

[1340] "Past sample reports" refer to reports that have been prepared and stored in the past, and are used as reference materials to guide responses to new events.

[1341] "Current Event Data" refers to data that includes information about ongoing events, new requests, complaints, etc.

[1342] "Preprocessing" refers to the process of removing unnecessary information from received data and converting it into a format suitable for analysis.

[1343] "Generative AI" refers to artificial intelligence that automatically generates draft reports based on specified instructions.

[1344] A "draft" is an initial report created by the generative AI, which is subject to user review and correction.

[1345] An "emotion engine" is a system that includes technology that recognizes a user's emotions by analyzing the user's facial expressions and tone of voice.

[1346] "Providing feedback" means presenting appropriate comments and advice that take into consideration the user's emotions based on the analysis results of the emotion engine.

[1347] This invention provides a system for providing fast and high-quality customer service in brick-and-mortar stores. The system receives past sample reports and current event data, analyzes the data, and automatically generates a report draft. It also recognizes the user's emotions and provides feedback based on the emotions.

[1348] System Overview

[1349] The system consists of the following main components:

[1350] User Input Method

[1351] Data preprocessing measures

[1352] Data collation method

[1353] Generation AI instruction means

[1354] Generation AI

[1355] Emotion Engine

[1356] Feedback methods

[1357] Hardware and software used

[1358] Hardware:

[1359] PC or smart device (with camera)

[1360] software:

[1361] Python

[1362] spaCy (natural language processing)

[1363] OpenAI API (generative AI model)

[1364] OpenCV (for emotion analysis)

[1365] Program processing overview

[1366] The system processes data in the following order, generates a draft report, checks and corrects it, and finally provides feedback based on the user's feelings.

[1367] User Input Method

[1368] Users input past sample reports and current event data into the system.

[1369] Data preprocessing measures

[1370] The server preprocesses the data it receives, removing unnecessary line breaks and spaces from the text and converting it into a format that is easier to parse.

[1371] Data collation method

[1372] The server compares past sample reports with current event data and extracts commonalities and differences, providing the basis for the generation AI to generate reports.

[1373] Generation AI instruction means

[1374] The server issues instructions to the generation AI based on the extracted data to generate a draft report. An example of the prompt used at this time is shown below.

[1375] Example prompt sentence:

[1376] Common points: {Common points}, Differences: {Differences}. Please generate a customer service response report that addresses the new request effectively and improves customer satisfaction.

[1377] Generation AI

[1378] The generation AI generates a draft report based on instructions from the server. Specifically, it documents appropriate responses and details of customer responses based on similarities and differences.

[1379] Emotion Engine

[1380] When a user reviews or modifies a draft report, the emotion engine analyzes camera footage and audio data to recognize emotions from the user's facial expressions and tone of voice.

[1381] Feedback methods

[1382] The feedback provision means instructs the generation AI to adjust expressions or generate additional comments based on the analysis results of the emotion engine, ultimately generating a report that takes the user's emotions into consideration.

[1383] Specific examples

[1384] 1. User input:

[1385] Example: A user uploads "Sample Report_2022.pdf" and "Customer Request_2023-10-10.txt" to the system.

[1386] 2. Data preprocessing:

[1387] Example: The server extracts text from "Sample Report_2022.pdf" and necessary information from "Customer Request_2023-10-10.txt." It then removes unnecessary line breaks and spaces from the extracted data and converts it into a format suitable for analysis.

[1388] 3. Data collation and analysis:

[1389] Example: The server compares the current customer request with past sample reports and extracts the differences between the new request and the past response history.

[1390] 4. Instructions to the generating AI:

[1391] Example: The server instructs the generation AI to generate a draft report based on "commonalities: maintaining past services, differences: new requests."

[1392] 5. Generate a draft report:

[1393] Example: The generation AI generates a draft report stating, "A new request was received from the customer on October 10, 2023. A response is required. Based on past response history and the new request, we suggest the following response."

[1394] 6. Emotion Recognition with Emotion Engine:

[1395] Example: While a user is reviewing and revising a draft report, the camera analyzes their facial expressions and recognizes that they are feeling stressed.

[1396] 7. Emotion-based feedback:

[1397] Example: The server instructs the generated AI to simplify the explanation and emphasize the important parts to reduce stress for the user.

[1398] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1399] Step 1:

[1400] The user uploads past sample reports and current event data from a terminal to the system. The input is a past sample report file and a current event data file, and the output is a confirmation of receipt of the uploaded files. The server receives these files and proceeds to the next step.

[1401] Step 2:

[1402] The server preprocesses the uploaded data and converts it into a format that is easy to analyze. This processing includes extracting text and removing unnecessary line breaks and spaces. The input is the uploaded sample report file and event data file, and the output is formatted text data. Specifically, it uses the Python spaCy library to remove unnecessary information from the text and convert the data into a structured format such as JSON.

[1403] Step 3:

[1404] The server uses the preprocessed data to match past sample reports with the current event data and extract commonalities and differences. The input is the formatted past sample report data and the current event data, and the output is a list of extracted commonalities and differences. Specifically, it uses text mining techniques to identify common keywords and phrases, as well as new elements, between the past reports and the new data.

[1405] Step 4:

[1406] The server instructs the generative AI to generate a draft report based on the extracted data. The input is a list of commonalities and differences, and the output is a prompt for the generative AI. The prompt is sent to the generative AI model (for example, OpenAI's API) to generate a detailed draft report. An example of a specific prompt is "Common points: {Common points}, Differences: {Differences}. Please generate a customer service response report that addresses the new request effectively and improves the customer satisfaction."

[1407] Step 5:

[1408] The generation AI generates a draft report based on prompts from the server and sends it back to the server. The input is the prompt text passed from the server, and the output is the generated draft report. Specifically, the generation AI automatically creates a detailed report of the customer interaction based on similarities and differences.

[1409] Step 6:

[1410] The user checks and modifies the generated report draft. To do this, they open the draft using a terminal and modify the content as necessary. The input is the generated report draft, and the output is the modified final report. The user checks the technical details and terminology of the report and makes any necessary modifications based on their own expertise.

[1411] Step 7:

[1412] The server uses an emotion engine to analyze the user's emotions when reviewing and revising a report draft. The input is video and audio data of the user's face, and the output is the analysis result of the user's emotional state. Specifically, it uses image analysis libraries such as OpenCV to recognize emotions from the user's facial expressions and tone of voice.

[1413] Step 8:

[1414] Based on the analysis results of the emotion engine, the server instructs the generation AI to provide feedback. The input is the user's emotion analysis results, and the output is feedback on the adjusted report and additional comments. Specifically, the generation AI readjusts the content of the report according to the user's emotions, providing concise explanations and emphasizing important parts to reduce stress. This allows the user to more efficiently finalize the report.

[1415] 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.

[1416] 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.

[1417] 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.

[1418] [Fourth embodiment]

[1419] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1420] 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.

[1421] 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).

[1422] 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.

[1423] 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.

[1424] 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).

[1425] 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.

[1426] 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.

[1427] 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.

[1428] 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.

[1429] 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.

[1430] 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.

[1431] 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."

[1432] System Overview

[1433] The present invention is a system for automatically generating fault reports based on user-provided sample reports and current event data. The system includes the following major functions:

[1434] Receiving data

[1435] Data Preprocessing

[1436] Data collation and analysis

[1437] Automatic generation of report drafts

[1438] User confirmation and correction

[1439] Program processing overview

[1440] 1. User Input

[1441] Users log into the system from a terminal and upload past sample reports and current event data.

[1442] 2. Data Preprocessing

[1443] The server receives the uploaded data and performs text extraction and formatting, specifically removing unnecessary line breaks and spaces and converting the data into a format that is easy to analyze.

[1444] 3. Data Collation and Analysis

[1445] The server compares past sample reports with current event data to extract commonalities and differences, including characteristic features such as error codes, the time of the failure, and the scope of the impact.

[1446] 4. Instructions to the generating AI

[1447] Based on the extracted data, the server instructs the AI ​​to generate a draft report, which includes an overview of the problem, its scope of impact, causes, and countermeasures.

[1448] 5. Generate a draft report

[1449] Based on instructions from the server, the generation AI creates a draft report detailing the problem, which is then sent back to the server in a format that can be reviewed and modified by the user.

[1450] 6. Checking and correcting the report

[1451] The user reviews the generated draft report and makes any necessary corrections. Once the final report is complete, the user saves it and shares it with the appropriate departments and stakeholders.

[1452] Specific examples

[1453] 1. User Input

[1454] Example: A user uploads "Sample Report_2022.pdf" and "Fault Log_2023-10-10.txt" to the system.

[1455] 2. Data Preprocessing

[1456] Example: The server extracts text from "Sample Report_2022.pdf" and the necessary log entries from "Error Log_2023-10-10.txt." It then removes unnecessary line breaks and spaces from the extracted data and converts it to JSON format.

[1457] 3. Data Collation and Analysis

[1458] Example: The server compares the sample report from 2022 with the error log from 2023, extracting the differences between the new error code "503" and the past error code "403." At the same time, it identifies commonalities in the scope of impact and occurrence time.

[1459] 4. Instructions to the generating AI

[1460] Example: The server instructs the AI ​​generation system that "An error code 503 occurred on October 10, 2023, causing service C to stop for 30 minutes." This instruction also includes a detailed summary of the error, the scope of its impact, the cause, and countermeasures.

[1461] 5. Generate a draft report

[1462] Example: The generation AI generates a draft report stating, "Server A failed on October 10, 2023. Service C was stopped for 30 minutes due to error code 503. The cause was a memory leak, and as a countermeasure, improved memory management is required."

[1463] 6. Checking and correcting the report

[1464] Example: User reviews the generated draft, corrects technical details and specific terminology to complete the final report, saves the completed report and shares it with the appropriate stakeholders.

[1465] In this way, the system can generate trouble reports efficiently and accurately, reducing the workload of the user.

[1466] The processing flow will be explained below.

[1467] Step 1:

[1468] The user logs in to the system from a terminal and goes to the upload page. The user selects the files of past sample reports and current event data and clicks the upload button.

[1469] Step 2:

[1470] The server receives the uploaded sample reports and event data, which are temporarily stored and prepared for further processing.

[1471] Step 3:

[1472] The server performs a parsing process to extract text from the uploaded sample reports, specifically extracting the required text information from PDFs and other file formats.

[1473] Step 4:

[1474] The server extracts the necessary information from the event data (e.g., log data), specifically by parsing and extracting important log entries such as the target error code and timestamp.

[1475] Step 5:

[1476] The server removes unnecessary line breaks and spaces from the extracted data and converts the data into a format that is easy to parse (e.g., JSON format).

[1477] Step 6:

[1478] The server compares the current event data with past sample reports, extracting commonalities (e.g., error codes, scope of impact) and differences (e.g., new error codes, different scope of impact).

[1479] Step 7:

[1480] Based on the results of the comparison, the server creates instructions for the generation AI to generate a draft report, including an overview of the problem, its scope of impact, causes, and countermeasures.

[1481] Step 8:

[1482] The generation AI generates a draft report based on instructions received from the server. The generation AI automatically generates detailed text for these items and provides feedback.

[1483] Step 9:

[1484] The generated report draft is sent back to the server and presented to the user, who can then check the generated draft through the terminal.

[1485] Step 10:

[1486] The user reviews the draft and makes any necessary corrections, such as changing technical terminology, adding details, or correcting typos.

[1487] Step 11:

[1488] The user saves the final report and then shares it with the appropriate departments and stakeholders, completing the process of generating a problem report.

[1489] Example 1

[1490] 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."

[1491] Creating a failure report manually is extremely time-consuming and prone to errors and oversights. The process of creating a report by comparing and analyzing past reports with current event data is particularly labor-intensive, making it necessary to create reports efficiently and accurately. Furthermore, reports must include a detailed failure summary, scope of impact, cause, and countermeasures, but there is currently a lack of a means to automatically generate standardized reports that include all of these.

[1492] 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.

[1493] In this invention, the server includes means for receiving past sample reports and current event data from a user, means for converting the received data into a format that is easy to analyze by extracting text and formatting, means for comparing the past sample reports and the current event data using natural language processing and machine learning algorithms to extract similarities and differences, means for providing a prompt to the generative AI model to generate a draft report based on the extracted information, and means for presenting the draft report generated by the generative AI model to the user in a format that can be confirmed and modified, thereby enabling the user to generate and modify reports efficiently and accurately.

[1494] "User" refers to a person or organization that uses a terminal to access the system and provide various data.

[1495] "Sample Report" means a document that contains a detailed record of a past event.

[1496] "Event Data" refers to data relating to a particular event that is currently occurring or has recently occurred.

[1497] "Text extraction" refers to the process of extracting character data from a rich text format file.

[1498] "Formatting" refers to the process of removing unnecessary line breaks and spaces from data and converting it into a format that is easier to parse.

[1499] "Natural language processing (NLP)" refers to technology for analyzing text data and understanding its meaning and structure.

[1500] A "machine learning algorithm" refers to a method of learning patterns and rules from large amounts of data and using that knowledge to analyze new data.

[1501] "Matching" refers to the process of finding similarities and differences between different data sets.

[1502] A "generative AI model" refers to an artificial intelligence system that automatically generates text or other data based on instructions it is given.

[1503] A "prompt sentence" refers to a textual input sentence that provides instructions to a generative AI model.

[1504] "Draft Report" refers to the initial version of the report automatically generated by the generative AI model.

[1505] "Format" refers to the particular structure or format of data or documents.

[1506] "Verification" refers to the process in which a user reviews the generated report draft and checks its content.

[1507] "Modification" refers to the user making any necessary changes or additions to the draft report.

[1508] The present invention is a system for automatically generating fault reports based on user-provided sample reports and current event data. The system can be implemented using the following major hardware and software components:

[1509] Hardware and software used

[1510] Terminal: A device used by a user to access the system and enter data. This includes devices such as PCs and smartphones.

[1511] Server: A central computing unit that preprocesses data, analyzes it, and directs generative AI models.

[1512] A text analysis library, such as Apache Tika, can be used to extract text from PDF files.

[1513] Use an NLP library such as spaCy to analyze text data and extract the necessary information.

[1514] Generative AI models: Generative AI models such as GPT-4 are used to generate draft reports.

[1515] Data processing and analysis

[1516] 1. User Input

[1517] A user logs in to the system from a terminal and uploads past sample reports and current event data, for example, uploading a report in PDF format and a fault log in text format.

[1518] Specifically, the user uploads "Sample Report_2022.pdf" and "Error Log_2023-10-10.txt" to the system.

[1519] 2. Data Preprocessing

[1520] The server receives the uploaded data and performs text extraction and formatting.

[1521] Specifically, we use Apache Tika to extract text from "SampleReport_2022.pdf", select the required log entries from the TXT file, remove unnecessary line breaks and spaces, and convert the data to JSON format.

[1522] 3. Data Collation and Analysis

[1523] The server uses the extracted data to compare and analyze past sample reports with current event data.

[1524] Specifically, spaCy is used to extract information such as error codes, dates and times, and the extent of the impact, allowing past and present data to be compared to identify commonalities and differences.

[1525] 4. Instructions to the generating AI

[1526] Based on the analysis results, the server instructs the generative AI model to generate a draft report.

[1527] An example of a specific prompt might be, "An error code 503 occurred on October 10, 2023, causing service C to be down for 30 minutes." This instruction also includes a detailed summary of the error, the scope of its impact, the cause, and countermeasures.

[1528] 5. Generate a draft report

[1529] The generative AI model generates a draft report detailing the fault based on instructions from the server, which is then sent back to the server in a format that can be reviewed and modified by the user.

[1530] Specifically, it generates a draft report stating something like, "Server A failed on October 10, 2023. Service C stopped for 30 minutes due to error code 503. The cause was a memory leak, and as a countermeasure, improved memory management is required."

[1531] 6. Checking and correcting the report

[1532] The user reviews the generated draft report, makes any necessary revisions, and finally saves the final report and shares it with the appropriate departments and agencies.

[1533] Specifically, users can access the draft from their devices, edit specific technical information and terminology, and then finalize the report, which is then saved and sent to stakeholders via email or a shared link.

[1534] In this way, the present invention can efficiently and accurately generate trouble reports, significantly reducing the user's workload. This system can be implemented using specific hardware and software, making it applicable to a variety of situations.

[1535] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1536] Step 1: User Input

[1537] A user logs in to the system from a terminal and uploads a past sample report (PDF format) and current event data (TXT format). The input data is "Sample Report_2022.pdf" and "Fault Log_2023-10-10.txt."

[1538] Specific behavior:

[1539] The user enters their username and password on the login page to access the system.

[1540] After logging in, select the "File Upload" option on your dashboard, select the required file and upload it.

[1541] If the upload is successful, the file is sent to the server.

[1542] Step 2: Preprocessing the data

[1543] The server receives the uploaded data, extracts text, and formats it. The input data is PDF and TXT files, and the output data is in an easy-to-parse JSON format.

[1544] Specific behavior:

[1545] The server uses the Apache Tika library to extract text from the PDF file "SampleReport_2022.pdf".

[1546] At the same time, the server selects the necessary log entries from the TXT file "Error Log_2023-10-10.txt".

[1547] Unnecessary line breaks and spaces are removed from this text data, and the formatted text data is converted into JSON format.

[1548] Step 3: Collate and analyze data

[1549] The server uses the extracted data to compare and analyze past sample reports with current event data. The input is text data in JSON format, and the output is information about notable similarities and differences.

[1550] Specific behavior:

[1551] The server uses an NLP library such as spaCy to extract information such as the error code, date and time, and scope of impact from the text data.

[1552] Based on the extracted information, past reports are compared with the current fault log to identify similarities and differences.

[1553] As a result, information such as differences in error codes and commonalities in the scope of impact is generated and passed on to the next step.

[1554] Step 4: Instructions to the Generative AI

[1555] The server instructs the generative AI model to generate a draft report based on the analysis results. The input is information about similarities and differences, and the output is a prompt statement for the generative AI model.

[1556] Specific behavior:

[1557] The server formalizes the analysis results as prompt sentences and provides them to the generative AI model.

[1558] As an example of a specific prompt sentence, a sentence such as "An error code 503 occurred on October 10, 2023, causing service C to be down for 30 minutes" is generated.

[1559] The prompt text includes a detailed overview of the problem, the scope of impact, the cause, and countermeasures.

[1560] Step 5: Generate a draft report

[1561] The generative AI model generates a draft report based on the prompt provided by the server. The input is the prompt to the generative AI, and the output is the draft report.

[1562] Specific behavior:

[1563] The generative AI model receives the prompt and creates a draft report.

[1564] The generated report draft is generated in Markdown, HTML, Word document format, etc. and sent back to the server.

[1565] The draft includes content such as, "Server A failed on October 10, 2023. Service C was stopped for 30 minutes due to error code 503. The cause was a memory leak, and as a countermeasure, improved memory management is required."

[1566] Step 6: Check and correct the report

[1567] The user reviews the generated report draft and makes any necessary corrections. The input is the report draft and the output is the final report.

[1568] Specific behavior:

[1569] The user accesses the draft from a terminal and checks its contents.

[1570] Correct specific technical information and terminology as needed to complete the final report.

[1571] Completed reports are stored within the system and then sent to the appropriate parties via email or shared link.

[1572] In this way, the system can automatically generate trouble reports efficiently and accurately based on user-supplied data.

[1573] (Application example 1)

[1574] 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."

[1575] Conventional factory robot monitoring systems require human intervention when a failure occurs, making it difficult to respond quickly and create accurate reports. Furthermore, manually comparing past failure reports with current event data and generating reports is time-consuming and labor-intensive. These issues can significantly impact factory operational efficiency and product quality, so a system that solves these problems is needed.

[1576] 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.

[1577] In this invention, the server includes means for receiving past sample reports and current event data from a user, means for preprocessing the received data and converting it into an easily analyzable format, means for comparing the past sample reports with the current event data and extracting similarities and differences, means for instructing the generation AI to generate a draft report based on the extracted information, means for presenting the draft report generated by the generation AI to the user, and means for monitoring the operating status of factory robots in real time and automatically generating a report when an abnormality occurs. This makes it possible to respond quickly and accurately to factory robot failures and improve factory operating efficiency.

[1578] "User" means a person or entity that uses the System to provide and query past sample reports and current event data.

[1579] A "Past Sample Report" is a document that contains detailed information about past events or failures.

[1580] "Current event data" refers to data relating to events or failures occurring at the present time.

[1581] "Preprocessing" is the process of deleting unnecessary line breaks and spaces and reformatting the data in order to convert the received data into a format that is easy to analyze.

[1582] "Matching" is the process of comparing past sample reports with current event data to extract similarities and differences.

[1583] "Generative AI" is an artificial intelligence system that automatically generates draft reports based on given data and instructions.

[1584] A "report draft" is a draft of a report automatically generated by the generation AI.

[1585] A "factory robot" is a mechanical device used to perform automated tasks in manufacturing.

[1586] "Operation status" is information that indicates the state in which a factory robot is operating.

[1587] "Real-time monitoring" refers to the continuous observation of a situation occurring at the present time.

[1588] An "abnormality" is a deviation from normal operation or failure of a system or machine.

[1589] A "report" is a formal document containing detailed information, causes, and actions regarding a specific incident or situation.

[1590] This invention is a system that monitors the operating status of factory robots in real time and automatically generates a fault report when an abnormality occurs. This system creates reports with high accuracy and efficiency based on past sample reports and current event data provided by the user.

[1591] 1. System Configuration

[1592] The system includes a server, user terminals, factory robots, generative AI models, and front-end and back-end processing. The server receives data, preprocesses it, collates it, and issues instructions to the generative AI. The user terminal uploads data and checks and edits reports via a smartphone application. The factory robots are monitored for their operating status and provide data in real time.

[1593] 2. Hardware / Software to be used

[1594] Hardware: smartphones, servers, factory robots

[1595] software:

[1596] Server: Flask (web framework), SQLite (database)

[1597] Preprocessing: Text extraction and data formatting using Python, NLTK (Natural Language Toolkit)

[1598] Generation AI: OpenAI GPT-3

[1599] Frontend: HTML / CSS / JavaScript

[1600] 3. Specific examples of program processing

[1601] Data reception and preprocessing

[1602] Users upload past sample reports and current event data from their devices, and the server receives the data, then uses Python to extract and format the text, removing unnecessary line breaks and spaces, and converting it into a format that is easier to parse.

[1603] Data collation and analysis

[1604] The preprocessed data is compared with past sample reports and current event data to extract commonalities and differences, including characteristic information such as error codes, time of occurrence, and scope of impact.

[1605] Instructions for generative AI models

[1606] The server instructs a generative AI model (e.g., OpenAI GPT-3) to generate a draft report based on the extracted data, including a summary of the problem, its scope of impact, causes, and countermeasures.

[1607] Draft report generation and presentation

[1608] The generative AI model generates a draft report based on the data and prompts received from the server, which is then sent back to the server and presented to the user, who can make any necessary corrections to produce a final fault report.

[1609] Examples of concrete examples and prompts

[1610] When a user uploads "Sample Report_2022.pdf" and "Fault Log_2023-10-10.txt" to the system, the system automatically analyzes these data, extracts commonalities and differences, and generates a draft report. The generated draft contains the following content:

[1611] Example prompt sentence:

[1612] "Based on past incident reports and the current incident log, we will provide the following information. Please use this information to generate a detailed incident report."

[1613] This system will enable rapid and accurate response to factory robot failures, significantly improving factory operational efficiency.

[1614] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1615] Step 1:

[1616] The user uploads past sample reports and current event data from their device. Specifically, the user uses a smartphone application to send "Sample Report_2022.pdf" and "Fault Log_2023-10-10.txt" to the system. The input is a PDF and text file, and these files are saved on the server as output.

[1617] Step 2:

[1618] The data received by the server is preprocessed. Specifically, text is extracted using Python, and unnecessary line breaks and spaces are removed to convert it into a format that is easy to parse. The input is the uploaded PDF and text file, and the output is formatted text data. For example, the text extracted from the PDF and the contents of the text file are converted into JSON format.

[1619] Step 3:

[1620] The server compares past sample reports with current event data to extract commonalities and differences. Specifically, it uses a scripting language (Python) to compare characteristic information such as error codes, the time of the failure, and the scope of impact. The input is preprocessed text data, and the output is data containing commonalities and differences. For example, it extracts the differences between "error code 503" and "error code 403."

[1621] Step 4:

[1622] Based on the information extracted by the server, the generative AI model is instructed to generate a draft report. Specifically, OpenAI GPT-3 is used to generate a prompt and make a request to generate a draft report. The input is data including similarities and differences, and the output is a prompt to the generative AI and a request to generate a draft report. For example, the command could be, "An error code 503 failure occurred on October 10, 2023, causing service C to be down for 30 minutes."

[1623] Step 5:

[1624] The generative AI model generates a draft report based on the instructions and sends it back to the server. Specifically, OpenAI GPT-3 generates a draft report based on a prompt and sends it back to the server. The input is the prompt, and the output is the generated draft report. For example, a draft report might be generated that states, "Server A failed on October 10, 2023. Service C was stopped for 30 minutes due to error code 503. The cause was a memory leak, and improved memory management is required as a countermeasure."

[1625] Step 6:

[1626] The server presents the generated draft report to the user, who makes any necessary revisions. Specifically, the report draft is displayed to the user through a smartphone application, allowing the user to modify technical details and specific terminology. The input is the generated draft report, and the output is the modified final report. The user saves the modified report and shares it with the appropriate parties.

[1627] In this way, the operating status of the factory robot can be monitored in real time, and a fault report can be generated quickly and accurately when an abnormality occurs.

[1628] 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.

[1629] System Overview

[1630] The present invention is a system that automatically generates trouble reports based on past sample reports provided by the user and current event data, and further combines it with an emotion engine that recognizes the user's emotions. The system has the following main functions:

[1631] Receiving data

[1632] Data Preprocessing

[1633] Data collation and analysis

[1634] Automatic generation of report drafts

[1635] User confirmation and correction

[1636] User Emotion Recognition and Feedback

[1637] Program processing overview

[1638] 1. User Input

[1639] Users log into the system from a terminal and upload past sample reports and current event data.

[1640] 2. Data Preprocessing

[1641] The server receives the uploaded data and performs text extraction and formatting, specifically removing unnecessary line breaks and spaces and converting the data into a format that is easy to analyze.

[1642] 3. Data Collation and Analysis

[1643] The server compares past sample reports with current event data to extract commonalities and differences, including characteristic features such as error codes, the time of the failure, and the scope of the impact.

[1644] 4. Instructions to the generating AI

[1645] Based on the extracted data, the server instructs the AI ​​to generate a draft report, which includes an overview of the problem, its scope of impact, causes, and countermeasures.

[1646] 5. Generate a draft report

[1647] The generation AI generates a draft report based on instructions from the server. The generation AI automatically generates detailed text for these items and sends it back to the server.

[1648] 6. Checking and correcting the report

[1649] The user reviews the generated draft report and makes any necessary corrections. Once the final report is complete, the user saves it and shares it with the appropriate departments and stakeholders.

[1650] 7. Emotion Recognition with Emotion Engine

[1651] When a user reviews or edits a draft report, the server uses an emotion engine to analyze the user's emotions. Emotion recognition is achieved by analyzing the user's facial expressions and tone of voice using camera footage and audio data.

[1652] 8. Emotional Feedback

[1653] Based on the results of the user's sentiment analysis, the server instructs the AI ​​to adjust the expression or generate additional comments. For example, if the user is feeling stressed, the AI ​​will simplify the explanation.

[1654] Specific examples

[1655] 1. User Input

[1656] Example: A user uploads "Sample Report_2022.pdf" and "Fault Log_2023-10-10.txt" to the system.

[1657] 2. Data Preprocessing

[1658] Example: The server extracts text from "Sample Report_2022.pdf" and the necessary log entries from "Error Log_2023-10-10.txt." It then removes unnecessary line breaks and spaces from the extracted data and converts it to JSON format.

[1659] 3. Data Collation and Analysis

[1660] Example: The server compares the sample report from 2022 with the error log from 2023, extracting the differences between the new error code "503" and the past error code "403." At the same time, it identifies commonalities in the scope of impact and occurrence time.

[1661] 4. Instructions to the generating AI

[1662] Example: The server instructs the AI ​​generation system that "An error code 503 occurred on October 10, 2023, causing service C to stop for 30 minutes." This instruction also includes a detailed summary of the error, the scope of its impact, the cause, and countermeasures.

[1663] 5. Generate a draft report

[1664] Example: The generation AI generates a draft report stating, "Server A failed on October 10, 2023. Service C was stopped for 30 minutes due to error code 503. The cause was a memory leak, and as a countermeasure, improved memory management is required."

[1665] 6. Checking and correcting the report

[1666] Example: User reviews the generated draft, corrects technical details and specific terminology to complete the final report, saves the completed report and shares it with the appropriate stakeholders.

[1667] 7. Emotion Recognition with Emotion Engine

[1668] Example: While a user is reviewing a report, the emotion engine analyzes the user's facial expressions from camera footage and recognizes that the user is feeling stressed.

[1669] 8. Emotional Feedback

[1670] Example: The server instructs the generated AI to make explanations more concise and emphasize important parts to reduce user stress. Feedback is provided based on the analysis of the emotion engine.

[1671] In this way, by utilizing the emotion engine, the system can reduce the burden on users and provide a more user-friendly process for creating trouble reports.

[1672] The processing flow will be explained below.

[1673] Step 1:

[1674] The user logs in to the system from a terminal and goes to the upload page. The user selects the files of past sample reports and current event data and clicks the upload button.

[1675] Step 2:

[1676] The server receives the uploaded sample reports and event data, which are temporarily stored and prepared for further processing.

[1677] Step 3:

[1678] The server runs a parsing process to extract text from the sample reports, specifically extracting the necessary text information from PDFs and other file formats and converting it into a format that is easier to parse.

[1679] Step 4:

[1680] The server extracts the necessary information from the event data (e.g., log data), specifically by parsing and extracting important log entries such as the target error code and timestamp.

[1681] Step 5:

[1682] The server removes unnecessary line breaks and spaces from the extracted data and converts the data into a unified format (e.g., JSON format).

[1683] Step 6:

[1684] The server compares the current event data with past sample reports, extracting commonalities (e.g., error codes, scope of impact) and differences (e.g., new error codes, different scope of impact).

[1685] Step 7:

[1686] Based on the results of the comparison, the server creates instructions for the generation AI to generate a draft report, including an overview of the problem, its scope of impact, causes, and countermeasures.

[1687] Step 8:

[1688] The generation AI generates a report draft based on instructions received from the server. The generation AI automatically generates detailed text for these items and sends the completed draft back to the server.

[1689] Step 9:

[1690] The server displays the generated report draft on the user's terminal, where the user can check the report draft and make corrections as necessary.

[1691] Step 10:

[1692] When a user checks a report draft, the server uses an emotion engine to recognize the user's emotions. The emotion engine analyzes the user's facial expressions and tone of voice, which are captured by a camera and microphone.

[1693] Step 11:

[1694] The server determines the user's emotional state (e.g., stress, dissatisfaction, relief) based on the analysis results of the emotion engine, and instructs the generation AI to adjust the content and expression of the report draft depending on the results.

[1695] Step 12:

[1696] Based on the analysis results of the emotion engine, the server displays feedback and alerts to the user. For example, if the user is feeling stressed, a comment is added to the report to simplify the explanation.

[1697] Step 13:

[1698] The user completes the final report and saves it on the server, where it is shared with the appropriate departments and stakeholders.

[1699] Step 14:

[1700] Users can effectively use the feedback from the emotion engine to improve the report creation process and identify areas for improvement. Through this process, adaptive report creation that takes emotions into consideration is realized.

[1701] Example 2

[1702] 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."

[1703] Conventional problem report creation systems often require a great deal of time and effort for users to check and edit reports, placing a heavy burden on users. Furthermore, because they do not take into account the user's emotional state, they can easily become stressed or fatigued. Furthermore, because the content of the generated reports is uniform, they sometimes do not function adequately even in situations where flexible responses according to the situation are required.

[1704] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1705] In this invention, the server includes means for receiving past sample reports and current event data from a user, means for preprocessing the received data and converting it into an easily analyzable format, means for comparing the past sample reports with the current event data and extracting similarities and differences, means for instructing a generation AI to generate a draft report based on the extracted information, means for presenting the report draft generated by the generation AI to the user, means for accepting user confirmation and correction of the report draft, means for recognizing the user's emotions, and means for adjusting the content of the report draft based on the recognized emotions. This reduces the burden on the user, streamlines the report creation process, and enables flexible report generation that corresponds to the user's emotional state.

[1706] "Users" refer to people who utilize the system to provide historical sample reports and current event data, and who are responsible for reviewing and modifying the final report.

[1707] "Sample report" refers to a previously created incident report that serves as a basis for comparison with current event data.

[1708] "Event Data" means data, including logs and information related to a current fault or problem, that is used to generate reports.

[1709] "Preprocessing" refers to the process of removing unnecessary line breaks and spaces from the received data and formatting it into a format that is easy to analyze.

[1710] "Matching" refers to the process of comparing past sample reports with current event data to identify similarities and differences.

[1711] "Generative AI" refers to an artificial intelligence model that automatically generates a draft report based on instructions from a server.

[1712] A "Draft Report" is an early-stage problem report created by the generation AI, before it has been reviewed and corrected by the user.

[1713] An "emotion engine" refers to software or hardware for recognizing a user's emotional state, for example by analyzing the user's facial expressions and tone of voice using camera or audio data.

[1714] "Feedback" refers to the process of providing instructions or information to the generative AI to adjust the content of the draft report based on the user's emotional state.

[1715] The present invention is a system that automatically generates trouble reports based on past sample reports and current event data provided by the user. Furthermore, this system combines an emotion engine that recognizes the user's emotions, aiming to reduce the burden on the user and streamline the report creation process.

[1716] Hardware and Software Configuration

[1717] The system is implemented using the following major hardware and software:

[1718] Server: A central device that preprocesses data, collates it, and issues instructions to the generation AI.

[1719] Terminal: Provides a user interface for users to upload data and review and modify generated draft reports.

[1720] Generative AI: An artificial intelligence model that automatically generates report drafts.

[1721] Emotion engine: Software or hardware for recognizing a user's emotions, including, for example, a facial recognition camera or voice analysis software.

[1722] Examples of data processing and data calculation

[1723] User Input

[1724] The user logs into the system from a terminal and uploads a past sample report (e.g., "Sample Report_2022.pdf") and current event data (e.g., "Fault Log_2023-10-10.txt").

[1725] Data Preprocessing

[1726] The server extracts text from the uploaded data and converts it into a format that is easy to parse. Specifically, it removes unnecessary line breaks and spaces and converts it into JSON format. For example, the server extracts text from "Sample Report_2022.pdf" and extracts the necessary log entries from "Failure Log_2023-10-10.txt."

[1727] Data collation and analysis

[1728] The server compares past sample reports with current event data to identify commonalities and differences. For example, it can extract the differences between the new error code "503" and the previous error code "403" and identify commonalities in the scope of impact and time of occurrence.

[1729] Instructions to the generating AI

[1730] The server instructs the generation AI to generate a draft report based on the extracted data. For example, it provides a prompt such as, "An error code 503 occurred on October 10, 2023, causing service C to be down for 30 minutes."

[1731] Generate a draft report

[1732] Based on the server's instructions, the generation AI generates a detailed draft report and sends it back to the server. Specifically, it generates a draft report with content such as, "Server A failed on October 10, 2023. Service C stopped for 30 minutes due to error code 503. The cause was a memory leak, and as a countermeasure, improved memory management is required."

[1733] Checking and correcting reports

[1734] The user reviews the generated draft report and makes any necessary corrections. Once the final report is complete, the user saves it and shares it with the appropriate departments and stakeholders.

[1735] Emotion recognition by emotion engine

[1736] When a user reviews or modifies a draft report, the server uses an emotion engine to analyze the user's emotions. For example, it can analyze the user's facial expressions from camera footage and recognize that the user is feeling stressed.

[1737] Emotion-based feedback

[1738] Based on the results of the user's sentiment analysis, the server instructs the AI ​​to adjust the expression or generate additional comments. For example, if the user is feeling stressed, the AI ​​is instructed to simplify the explanation.

[1739] In this way, by linking the server, terminal, and user and making full use of generative AI and an emotion engine, a system is realized that allows users to create fault reports efficiently and with little burden.

[1740] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1741] Step 1:

[1742] A user logs into the system and uploads data

[1743] Input: User ID and password, past sample report (e.g., "Sample Report_2022.pdf"), current event data (e.g., "Error Log_2023-10-10.txt")

[1744] Specific behavior:

[1745] A user enters a user ID and password to log in to the system from a terminal.

[1746] After successfully logging in, the user uploads "Sample Report_2022.pdf" and "Error Log_2023-10-10.txt" to the system.

[1747] Output: Uploaded sample report and event data

[1748] Step 2:

[1749] The server preprocesses the uploaded data

[1750] Input: Uploaded sample reports and event data

[1751] Specific behavior:

[1752] The server extracts text from "SampleReport_2022.pdf".

[1753] Similarly, extract the necessary log entries from "Fault Log_2023-10-10.txt".

[1754] To make the extracted text data easier to analyze, unnecessary line breaks and spaces are removed and the data is converted to JSON format.

[1755] Output: Formatted sample report data and event data (JSON format)

[1756] Step 3:

[1757] The server collates and analyzes past sample reports with current event data.

[1758] Input: Sample report data and event data in JSON format

[1759] Specific behavior:

[1760] The server extracts the difference between the new error code "503" and the previous error code "403".

[1761] The server identifies commonalities in the scope of impact and time of occurrence.

[1762] Output: Information on the difference between extracted error codes, the scope of impact, and the time of occurrence

[1763] Step 4:

[1764] The server instructs the generation AI to generate a draft report.

[1765] Input: Information on the difference between the extracted error codes, the scope of the impact, and the time of occurrence

[1766] Specific behavior:

[1767] The server provides the AI ​​with a specific prompt, such as "An error code 503 occurred on October 10, 2023, causing service C to be down for 30 minutes."

[1768] Output: Instructions to the generating AI

[1769] Step 5:

[1770] Generative AI generates a draft report

[1771] Input: Instructions to the generation AI

[1772] Specific behavior:

[1773] Based on the instructions, the generative AI generates a draft report containing detailed text.

[1774] Specifically, a draft report will be created stating, "Server A failed on October 10, 2023. Service C was stopped for 30 minutes due to error code 503. The cause was a memory leak, and strengthening memory management is required as a countermeasure."

[1775] Output: Draft report

[1776] Step 6:

[1777] The user checks the draft report and makes corrections

[1778] Input: Generated report draft

[1779] Specific behavior:

[1780] The user checks the report draft from the terminal.

[1781] The user corrects technical details and specific terminology to complete the final report.

[1782] Save the completed report and share it with appropriate parties.

[1783] Output: Final report

[1784] Step 7:

[1785] The server uses an emotion engine to recognize the user's emotions.

[1786] Input: Camera footage and audio data of the user reviewing the draft report

[1787] Specific behavior:

[1788] The server analyzes camera footage and audio data and recognizes emotions from the user's facial expressions and tone of voice.

[1789] For example, it determines whether the user is feeling stressed.

[1790] Output: User's emotional state

[1791] Step 8:

[1792] The server provides emotion-based feedback to the generative AI.

[1793] Input: User's emotional state

[1794] Specific behavior:

[1795] The server instructs the generation AI to adjust the expression of the report draft based on the user's emotional state.

[1796] For example, if the user is feeling stressed, the AI ​​generator can be instructed to keep explanations brief.

[1797] Output: Instructions to regenerate the draft report reflecting the adjusted feedback

[1798] The above is the specific processing flow of the system program.

[1799] (Application example 2)

[1800] 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."

[1801] In customer service at brick-and-mortar stores, employees are required to respond quickly and appropriately to a variety of customer requests and complaints. However, it can be difficult for employees to generate high-quality reports, especially during busy or stressful times, which can result in a decrease in customer satisfaction. Furthermore, responses that do not take employees' feelings into consideration can increase their own stress and, in the long term, lead to a decrease in job satisfaction and performance. A solution to these issues is needed.

[1802] 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.

[1803] In this invention, the server includes means for receiving past sample reports and current event data from a user, means for preprocessing the received data and converting it into an easily analyzable format, means for comparing the past sample reports with the current event data and extracting similarities and differences, means for instructing a generation AI to generate a draft report based on the extracted information, means for presenting the draft report generated by the generation AI to the user, means including an emotion engine for recognizing the user's emotions, and means for providing feedback based on the user's emotions. This enables employees to generate customer interaction reports quickly and with high quality, and also enables them to receive feedback that takes into consideration the employee's emotions.

[1804] "User" means a user who utilizes the System to provide past sample reports and current event data.

[1805] "Past sample reports" refer to reports that have been prepared and stored in the past, and are used as reference materials to guide responses to new events.

[1806] "Current Event Data" refers to data that includes information about ongoing events, new requests, complaints, etc.

[1807] "Preprocessing" refers to the process of removing unnecessary information from received data and converting it into a format suitable for analysis.

[1808] "Generative AI" refers to artificial intelligence that automatically generates draft reports based on specified instructions.

[1809] A "draft" is an initial report created by the generative AI, which is subject to user review and correction.

[1810] An "emotion engine" is a system that includes technology that recognizes a user's emotions by analyzing the user's facial expressions and tone of voice.

[1811] "Providing feedback" means presenting appropriate comments and advice that take into consideration the user's emotions based on the analysis results of the emotion engine.

[1812] This invention provides a system for providing fast and high-quality customer service in brick-and-mortar stores. The system receives past sample reports and current event data, analyzes the data, and automatically generates a report draft. It also recognizes the user's emotions and provides feedback based on the emotions.

[1813] System Overview

[1814] The system consists of the following main components:

[1815] User Input Method

[1816] Data preprocessing measures

[1817] Data collation method

[1818] Generation AI instruction means

[1819] Generation AI

[1820] Emotion Engine

[1821] Feedback methods

[1822] Hardware and software used

[1823] Hardware:

[1824] PC or smart device (with camera)

[1825] software:

[1826] Python

[1827] spaCy (natural language processing)

[1828] OpenAI API (generative AI model)

[1829] OpenCV (for emotion analysis)

[1830] Program processing overview

[1831] The system processes data in the following order, generates a draft report, checks and corrects it, and finally provides feedback based on the user's feelings.

[1832] User Input Method

[1833] Users input past sample reports and current event data into the system.

[1834] Data preprocessing measures

[1835] The server preprocesses the data it receives, removing unnecessary line breaks and spaces from the text and converting it into a format that is easier to parse.

[1836] Data collation method

[1837] The server compares past sample reports with current event data and extracts commonalities and differences, providing the basis for the generation AI to generate reports.

[1838] Generation AI instruction means

[1839] The server issues instructions to the generation AI based on the extracted data to generate a draft report. An example of the prompt used at this time is shown below.

[1840] Example prompt sentence:

[1841] Common points: {Common points}, Differences: {Differences}. Please generate a customer service response report that addresses the new request effectively and improves customer satisfaction.

[1842] Generation AI

[1843] The generation AI generates a draft report based on instructions from the server. Specifically, it documents appropriate responses and details of customer responses based on similarities and differences.

[1844] Emotion Engine

[1845] When a user reviews or modifies a draft report, the emotion engine analyzes camera footage and audio data to recognize emotions from the user's facial expressions and tone of voice.

[1846] Feedback methods

[1847] The feedback provision means instructs the generation AI to adjust expressions or generate additional comments based on the analysis results of the emotion engine, ultimately generating a report that takes the user's emotions into consideration.

[1848] Specific examples

[1849] 1. User input:

[1850] Example: A user uploads "Sample Report_2022.pdf" and "Customer Request_2023-10-10.txt" to the system.

[1851] 2. Data preprocessing:

[1852] Example: The server extracts text from "Sample Report_2022.pdf" and necessary information from "Customer Request_2023-10-10.txt." It then removes unnecessary line breaks and spaces from the extracted data and converts it into a format suitable for analysis.

[1853] 3. Data collation and analysis:

[1854] Example: The server compares the current customer request with past sample reports and extracts the differences between the new request and the past response history.

[1855] 4. Instructions to the generating AI:

[1856] Example: The server instructs the generation AI to generate a draft report based on "commonalities: maintaining past services, differences: new requests."

[1857] 5. Generate a draft report:

[1858] Example: The generation AI generates a draft report stating, "A new request was received from the customer on October 10, 2023. A response is required. Based on past response history and the new request, we suggest the following response."

[1859] 6. Emotion Recognition with Emotion Engine:

[1860] Example: While a user is reviewing and revising a draft report, the camera analyzes their facial expressions and recognizes that they are feeling stressed.

[1861] 7. Emotion-based feedback:

[1862] Example: The server instructs the generated AI to simplify the explanation and emphasize the important parts to reduce stress for the user.

[1863] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1864] Step 1:

[1865] The user uploads past sample reports and current event data from a terminal to the system. The input is a past sample report file and a current event data file, and the output is a confirmation of receipt of the uploaded files. The server receives these files and proceeds to the next step.

[1866] Step 2:

[1867] The server preprocesses the uploaded data and converts it into a format that is easy to analyze. This processing includes extracting text and removing unnecessary line breaks and spaces. The input is the uploaded sample report file and event data file, and the output is formatted text data. Specifically, it uses the Python spaCy library to remove unnecessary information from the text and convert the data into a structured format such as JSON.

[1868] Step 3:

[1869] The server uses the preprocessed data to match past sample reports with the current event data and extract commonalities and differences. The input is the formatted past sample report data and the current event data, and the output is a list of extracted commonalities and differences. Specifically, it uses text mining techniques to identify common keywords and phrases, as well as new elements, between the past reports and the new data.

[1870] Step 4:

[1871] The server instructs the generative AI to generate a draft report based on the extracted data. The input is a list of commonalities and differences, and the output is a prompt for the generative AI. The prompt is sent to the generative AI model (for example, OpenAI's API) to generate a detailed draft report. An example of a specific prompt is "Common points: {Common points}, Differences: {Differences}. Please generate a customer service response report that addresses the new request effectively and improves the customer satisfaction."

[1872] Step 5:

[1873] The generation AI generates a draft report based on prompts from the server and sends it back to the server. The input is the prompt text passed from the server, and the output is the generated draft report. Specifically, the generation AI automatically creates a detailed report of the customer interaction based on similarities and differences.

[1874] Step 6:

[1875] The user checks and modifies the generated report draft. To do this, they open the draft using a terminal and modify the content as necessary. The input is the generated report draft, and the output is the modified final report. The user checks the technical details and terminology of the report and makes any necessary modifications based on their own expertise.

[1876] Step 7:

[1877] The server uses an emotion engine to analyze the user's emotions when reviewing and revising a report draft. The input is video and audio data of the user's face, and the output is the analysis result of the user's emotional state. Specifically, it uses image analysis libraries such as OpenCV to recognize emotions from the user's facial expressions and tone of voice.

[1878] Step 8:

[1879] Based on the analysis results of the emotion engine, the server instructs the generation AI to provide feedback. The input is the user's emotion analysis results, and the output is feedback on the adjusted report and additional comments. Specifically, the generation AI readjusts the content of the report according to the user's emotions, providing concise explanations and emphasizing important parts to reduce stress. This allows the user to more efficiently finalize the report.

[1880] 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.

[1881] 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.

[1882] 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.

[1883] 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.

[1884] 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.

[1885] 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.

[1886] 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).

[1887] 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.

[1888] 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."

[1889] 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.

[1890] 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).

[1891] 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.

[1892] 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.

[1893] 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.

[1894] 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.

[1895] 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.

[1896] 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.

[1897] 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.

[1898] 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.

[1899] 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.

[1900] 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.

[1901] The following is further disclosed regarding the above embodiment.

[1902] (Claim 1)

[1903] means for receiving past sample reports and current event data from a user;

[1904] A means to preprocess the received data and convert it into a format that is easy to analyze;

[1905] A means of matching past sample reports with current event data to extract similarities and differences;

[1906] A means for instructing the generation AI to generate a draft report based on the extracted information;

[1907] A means for presenting the report draft generated by the generation AI to a user;

[1908] A system including:

[1909] (Claim 2)

[1910] 2. The system according to claim 1, wherein the data preprocessing means deletes unnecessary line breaks and spaces and formats the data.

[1911] (Claim 3)

[1912] The system of claim 1, wherein the draft report generated by the generation AI includes an overview of the problem, the scope of impact, the cause, and countermeasures.

[1913] "Example 1"

[1914] (Claim 1)

[1915] means for receiving past sample reports and current event data from a user;

[1916] A means of converting the received data into a format that is easy to analyze by extracting text and formatting it;

[1917] A means for matching past sample reports with current event data using natural language processing and machine learning algorithms to extract similarities and differences;

[1918] a means for providing prompts to the generative AI model to generate a draft report based on the extracted information;

[1919] a means for presenting the draft report generated by the generative AI model to a user in a format that can be reviewed and modified;

[1920] A system including:

[1921] (Claim 2)

[1922] 2. The system according to claim 1, wherein the data preprocessing means deletes unnecessary line breaks and spaces, formats the data, and converts it into JSON format.

[1923] (Claim 3)

[1924] The system of claim 1, wherein the draft report generated by the generative AI model includes an overview of the problem, its scope of impact, its cause, and countermeasures.

[1925] "Application Example 1"

[1926] (Claim 1)

[1927] means for receiving past sample reports and current event data from a user;

[1928] A means to preprocess the received data and convert it into a format that is easy to analyze;

[1929] A means of matching past sample reports with current event data to extract similarities and differences;

[1930] A means for instructing the generation AI to generate a draft report based on the extracted information;

[1931] A means for presenting the report draft generated by the generation AI to a user;

[1932] A means to monitor the operating status of factory robots in real time and automatically generate a report when an abnormality occurs;

[1933] A system including:

[1934] (Claim 2)

[1935] 2. The system according to claim 1, wherein the data preprocessing means deletes unnecessary line breaks and spaces and formats the data.

[1936] (Claim 3)

[1937] The system of claim 1, wherein the draft report generated by the generation AI includes an overview of the problem, the scope of impact, the cause, and countermeasures.

[1938] "Example 2: Combining Emotion Engines"

[1939] (Claim 1)

[1940] means for receiving past sample reports and current event data from a user;

[1941] A means to preprocess the received data and convert it into a format that is easy to analyze;

[1942] A means of matching past sample reports with current event data to extract similarities and differences;

[1943] A means for instructing the generation AI to generate a draft report based on the extracted information;

[1944] A means for presenting the report draft generated by the generation AI to a user;

[1945] means for accepting user review and revision of the draft report;

[1946] means for recognizing a user's emotion;

[1947] a means of adjusting the content of the draft report based on the perceived sentiment;

[1948] A system including:

[1949] (Claim 2)

[1950] 2. The system according to claim 1, wherein the data preprocessing means deletes unnecessary line breaks and spaces and formats the data.

[1951] (Claim 3)

[1952] The system of claim 1, wherein the draft report generated by the generation AI includes an overview of the problem, the scope of impact, the cause, and countermeasures.

[1953] "Application example 2 when combining emotion engines"

[1954] (Claim 1)

[1955] means for receiving past sample reports and current event data from a user;

[1956] A means to preprocess the received data and convert it into a format that is easy to analyze;

[1957] A means of matching past sample reports with current event data to extract similarities and differences;

[1958] A means for instructing the generation AI to generate a draft report based on the extracted information;

[1959] A means for presenting the report draft generated by the generation AI to a user;

[1960] means including an emotion engine for recognizing an emotion of a user;

[1961] means for providing feedback based on the user's emotions;

[1962] A system including:

[1963] (Claim 2)

[1964] 2. The system according to claim 1, wherein the data preprocessing means deletes unnecessary line breaks and spaces and formats the data.

[1965] (Claim 3)

[1966] The system of claim 1, wherein the draft report generated by the generation AI includes an overview of the problem, the scope of impact, the cause, and countermeasures. [Explanation of symbols]

[1967] 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 receiving past sample reports and current event data from a user; A means to preprocess the received data and convert it into a format that is easy to analyze; A means of matching past sample reports with current event data to extract similarities and differences; A means for instructing the generation AI to generate a draft report based on the extracted information; A means for presenting the report draft generated by the generation AI to a user; A system including:

2. 2. The system according to claim 1, wherein the data preprocessing means deletes unnecessary line breaks and spaces and formats the data.

3. The system according to claim 1, wherein the draft report generated by the generation AI includes items such as an overview of the problem, the scope of its impact, its cause, and countermeasures.

Images