system

Abstract

Provide a system. A means for obtaining authentication information; means for performing user authentication based on the authentication information; A means to input company-specific know-how and schemes, A means for inputting the input data into a generative artificial intelligence model and performing analysis; a means for storing the analysis results in a database; A means for searching know-how information from the database; means for providing the search results to a user; A means of collecting feedback data and using it to retrain the model A system including:

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] Modern companies need to efficiently manage and share the know-how and schemes accumulated within their companies. However, previous systems made storing and searching know-how cumbersome, resulting in a lack of user convenience. They also lacked a mechanism for updating AI models based on feedback. This resulted in a stagnation in the sharing of know-how, hindering productivity improvements. [Means for solving the problem]

[0005] This invention provides a system for storing and searching in-house know-how using a generative artificial intelligence model. The system includes a means for acquiring authentication information and performing user authentication based on that information. It also includes a means for users to input in-house know-how and schemes, a means for inputting the input data into a generative artificial intelligence model for analysis, a means for saving the analysis results in a database, and a means for searching know-how information from the database and providing the search results to the user. The system also includes a means for collecting feedback data and using it to retrain the generative artificial intelligence model, thereby ensuring that the latest information is always available. This allows users to easily access know-how and improve the AI ​​model based on the feedback.

[0006] "Authentication information" is information used to identify a user and verify the user's access privileges.

[0007] A "generative artificial intelligence model" is a machine learning model that has the ability to analyze input data and generate new information and relationships.

[0008] "In-house know-how and schemes" refers to a unique body of knowledge that does not exist in other companies, such as business procedures, best practices, and technical information accumulated within the company.

[0009] "User authentication" refers to the process of verifying whether a user has valid access rights based on the authentication information entered.

[0010] A "database" is an electronic information storage system that organizes information so that it can be efficiently accessed and managed.

[0011] "Feedback" refers to ratings and comments provided by users, and this information is used to improve and adjust the system.

[0012] "Retraining" is the process of retraining an existing artificial intelligence model based on new data and feedback to improve its performance and accuracy.

[0013] A "session ID" is a unique identifier issued by a server to identify the session in which a user is logged in.

[0014] "Natural language processing" is a general term for methods and technologies that allow computers to understand, analyze, and generate human language. [Brief explanation of the drawings]

[0015] [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

[0016] 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.

[0017] First, the terms used in the following description will be explained.

[0018] 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).

[0019] 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.

[0020] 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.

[0021] 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.

[0022] 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."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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."

[0036] The present invention relates to a system for efficiently storing know-how and schemes specific to a company, and facilitating sharing and searching thereof. A specific embodiment of this system will be described below.

[0037] ---

[0038] User Authentication Process

[0039] When a user logs in to a terminal, they enter their company ID and password, which confirms their authority to access the system.

[0040] 1. The device collects the user's ID and password and sends them to the server using the HTTPS protocol.

[0041] 2. The server compares the received authentication information with the database and authenticates the user. If authentication is successful, it generates a session ID and returns it to the terminal.

[0042] 3. The terminal receives the session ID and displays a successful login message to the user.

[0043] Know-how input process

[0044] This is the process by which users input new know-how and schemes into the system, thereby accumulating in-house knowledge.

[0045] 1. The user enters new know-how or scheme (e.g., "Agile development is being adopted for Project A") into a dedicated form on the device.

[0046] 2. The device collects input data and sends it to the server in JSON format.

[0047] 3. The server preprocesses the data, normalizing and structuring the text.

[0048] 4. The server inputs the preprocessed data into a generative AI model for analysis. The analysis results include related existing data and new insights.

[0049] 5. The server saves the analysis results in a database. When saving, the date and time of saving and information about the person who entered the data are also recorded.

[0050] Know-how search process

[0051] It is a process in which users search for existing know-how and schemes.

[0052] 1. A user enters a query (e.g., "Agile development best practices") into a search form on their device.

[0053] 2. The device sends a query in JSON format to the server.

[0054] 3. The server analyzes the received query using natural language processing to extract the intent.

[0055] 4. The server sends a search query to the database to retrieve relevant know-how data.

[0056] 5. The server returns the search results to the device in JSON format.

[0057] 6. The device displays the search results to the user.

[0058] Feedback Providing Process

[0059] This is the process of providing feedback to users on the know-how they have acquired.

[0060] 1. The user inputs feedback from the terminal after applying the know-how (e.g., "This scheme is useful, but specifically, task management is lacking").

[0061] 2. The device sends the feedback content in JSON format to the server.

[0062] 3. The server receives the feedback, analyzes it, and saves it as data for retraining the model.

[0063] 4. The server periodically retrains the generative AI model to improve its accuracy.

[0064] 5. The server applies the improved model to the system and reflects it in the next user search or know-how input.

[0065] ---

[0066] Specific examples

[0067] Example of user authentication

[0068] 1. User "Tanaka" enters ID "tanaka" and password "password123".

[0069] 2. The device sends this to the server.

[0070] 3. The server performs authentication, is successful, and generates a session ID "abc123".

[0071] 4. The terminal saves the session ID and displays "Login successful" to the user.

[0072] Specific examples of know-how input

[0073] 1. User "Tanaka" inputs "Scrum development is recommended for Project B."

[0074] 2. The device sends the data to the server.

[0075] 3. The server inputs the data into a generative artificial intelligence model and stores the analysis results in a database.

[0076] This will create a system that allows users to efficiently input, search, and share in-house know-how.

[0077] The processing flow will be explained below.

[0078] User Authentication Process

[0079] Step 1:

[0080] The terminal prompts the user to enter their company ID and password. For example, the user enters "tanaka" and "password123."

[0081] Step 2:

[0082] The terminal sends the entered authentication information to the server using the HTTPS protocol.

[0083] Step 3:

[0084] The server queries the database for the received authentication information and searches for a match.

[0085] Step 4:

[0086] The server retrieves the search results from the database and, if authentication is successful, generates a new session ID (e.g., "sessionID_abc123").

[0087] Step 5:

[0088] The server returns the authentication result and session ID to the terminal as an HTTPS response.

[0089] Step 6:

[0090] The terminal receives the response from the server, notifies the user that login was successful, and saves the session ID.

[0091] ---

[0092] Know-how input process

[0093] Step 1:

[0094] The user inputs new know-how or schemes into a dedicated form on the terminal. For example, the user inputs "Scrum-based development is recommended for Project B."

[0095] Step 2:

[0096] The device collects the input data and sends it to the server in JSON format.

[0097] Step 3:

[0098] The server preprocesses the data received and normalizes the text (matching case, removing unnecessary whitespace, etc.).

[0099] Step 4:

[0100] The server inputs the preprocessed data into a generative artificial intelligence model and analyzes the know-how data.

[0101] Step 5:

[0102] The server obtains the analysis results and extracts link information with related existing know-how and new know-how.

[0103] Step 6:

[0104] The server saves the generated know-how data in a database, along with the date and time of saving and information about the person who entered the data.

[0105] ---

[0106] Know-how search process

[0107] Step 1:

[0108] A user enters a query into a search form on their device. For example, the user enters "Agile development best practices."

[0109] Step 2:

[0110] The device sends the entered query in JSON format to the server.

[0111] Step 3:

[0112] The server analyzes the received query using natural language processing to understand the intent of the query and extract related keywords.

[0113] Step 4:

[0114] The server sends a search query to the database to retrieve relevant know-how data, and the search results also include a relevance score.

[0115] Step 5:

[0116] The server returns the search results to the device in JSON format.

[0117] Step 6:

[0118] The device receives the response from the server and displays the search results to the user, providing an interface that makes it easy for users to find the information they need.

[0119] ---

[0120] Feedback Providing Process

[0121] Step 1:

[0122] It provides feedback on the know-how acquired by the user. For example, the user may input, "This scheme is easy to understand, but I would like to see a concrete example task."

[0123] Step 2:

[0124] The device collects the feedback and sends it to the server in JSON format.

[0125] Step 3:

[0126] The server receives the feedback and performs text analysis to extract the intent and key points of the feedback.

[0127] Step 4:

[0128] The server stores the analyzed feedback data as retraining data for the generative artificial intelligence model.

[0129] Step 5:

[0130] The server runs a retraining process at regular intervals to improve the performance and accuracy of the model.

[0131] Step 6:

[0132] The server applies the improved model to the system and reflects it in the next user search or know-how input.

[0133] Through these steps, users can efficiently input, search, and provide feedback on their internal know-how.

[0134] Example 1

[0135] 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."

[0136] Previous methods for sharing in-house know-how and schemes mainly involved manual recording and management, which lacked efficiency. It was also difficult to search for know-how or discover new insights, resulting in low utilization of information. Furthermore, there was no established method for collecting feedback and using it to update and improve knowledge. For these reasons, a new system for effectively storing and utilizing in-house knowledge resources was needed.

[0137] 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.

[0138] In this invention, the server includes means for acquiring authentication information, means for authenticating a user based on the authentication information, means for inputting in-house know-how and schemes, means for transmitting the input data to the server in JSON format, means for inputting the input data to a generative artificial intelligence model for analysis, means for saving the analysis results in a database, means for searching the database for know-how information using natural language processing, means for providing the search results to the user in JSON format, and means for collecting feedback data and using it to retrain the model. This enables efficient accumulation, sharing, and search of in-house know-how, and further enables the model to be improved based on feedback to increase the quality of knowledge.

[0139] "Authentication Information" means information required to verify a user's identity, and primarily includes a user ID and password.

[0140] "User authentication" is the process of verifying that a user is a legitimate user using authentication information.

[0141] "Know-how" refers to information that refers to knowledge, techniques, and methodologies related to a specific task or project.

[0142] "Scheme" means the plan, structure or procedure for a particular project or business process.

[0143] "JSON format" is a lightweight data exchange format for structuring data, and is an abbreviation for JavaScript (registered trademark) Object Notation.

[0144] A "generative artificial intelligence model" is a set of machine learning algorithms that generate new information and insights based on input data.

[0145] "Analysis" is the process performed to understand input data and find patterns and relationships.

[0146] A "database" is a system for organizing, efficiently storing, and retrieving information.

[0147] "Natural language processing" is a technical field that uses computers to understand and process human language.

[0148] "Feedback data" refers to information such as ratings, comments, and improvements provided by users regarding the system's output.

[0149] "Model retraining" is the process of updating a generative artificial intelligence model with new data collected to improve its accuracy and performance.

[0150] This invention relates to a system that efficiently accumulates in-house know-how and schemes and makes it easy to share and search them. This system includes a series of processes for collecting, transmitting, analyzing, and storing data between a server, terminals, and users.

[0151] User Authentication Process

[0152] When a user enters their company ID and password into the terminal, the terminal encrypts the authentication information and sends it to the server using the HTTPS protocol. The server compares the authentication information with a database, and if authentication is successful, generates a session ID and returns it to the terminal. This allows the user to access the system securely.

[0153] Know-how input process

[0154] The user inputs new know-how or schemes into a dedicated form on the device. For example, this information might be something like "Scrum development is recommended for Project B." The device then sends the input data in JSON format to the server. The server preprocesses the data and analyzes it using a generative artificial intelligence model. The analysis results are saved in a database, along with the date and time of saving and information about the person who entered the data.

[0155] Know-how search process

[0156] When a user enters a query (e.g., "Agile development best practices") into the device's search form, the device sends the query in JSON format to the server. The server analyzes the query using natural language processing technology and sends the search query to the database. The search results are returned to the device in JSON format and displayed to the user.

[0157] Feedback Providing Process

[0158] After applying the know-how, the user inputs feedback from the device. For example, the user might say, "This scheme is useful, but specifically, task management is lacking." The device then sends the feedback in JSON format to the server. The server analyzes the feedback and saves it as retraining data for the generative AI model. This improves the model's accuracy, which is reflected the next time the user searches or inputs know-how.

[0159] Hardware and software used

[0160] This system uses the following hardware and software:

[0161] Device: A computer or smartphone operated by a user.

[0162] Server: A computer system that processes and stores data.

[0163] Database: A system for storing know-how, schemes, analysis results, and feedback data.

[0164] HTTPS protocol: A communications protocol for secure data transmission.

[0165] Generative artificial intelligence models: Algorithms that analyze input data and generate new insights.

[0166] Specific examples

[0167] Example of user authentication

[0168] The user enters the ID "tanaka" and password "password123" into the terminal, which then encrypts and sends it to the server. The server compares it with the database, and if authentication is successful, it generates a session ID "abc123" and returns it to the terminal. The terminal saves the session ID and displays "Login successful" to the user.

[0169] Specific examples of know-how input

[0170] The user inputs "Scrum development is recommended for Project B," and the device sends the data in JSON format to the server. The server preprocesses the data, inputs it into a generative artificial intelligence model for analysis, and stores the analysis results in a database.

[0171] Specific examples of know-how search

[0172] A user types "Agile development best practices" into a search form, and the device sends the query to the server, which parses the query, searches the database, and returns relevant results to the device, where the user is presented with the search results.

[0173] Example prompt sentence:

[0174] "Project B is recommended to be developed using Scrum. What are some best practices that can be used for other projects?"

[0175] This system enables efficient accumulation, sharing, and searching of internal know-how, allowing for improvements to generative AI models based on feedback.

[0176] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0177] Program processing flow

[0178] User Authentication Process

[0179] Step 1:

[0180] The user enters their company ID and password into the device's login screen.

[0181] Input: Company ID, password

[0182] Output: Credentials sent to the device

[0183] Step 2:

[0184] The terminal encrypts the entered ID and password using the HTTPS protocol and sends it to the server.

[0185] Input: Company ID, password

[0186] Output: Authentication information encrypted using the HTTPS protocol

[0187] Step 3:

[0188] The server checks the received authentication information against a database and authenticates the user.

[0189] Input: Credentials

[0190] Output: Authentication success or failure, session ID

[0191] Step 4:

[0192] If the authentication is successful, the server generates a session ID and returns it to the terminal.

[0193] Input: Authentication success result

[0194] Output: Session ID

[0195] Step 5:

[0196] The terminal receives the session ID and displays a successful login message to the user.

[0197] Input: Session ID

[0198] Output: Login success message, session ID stored on the device

[0199] Know-how input process

[0200] Step 1:

[0201] The user inputs new know-how and schemes into a dedicated form on the terminal.

[0202] Input: Know-how such as "Scrum development is recommended for Project B"

[0203] Output: Input data temporarily saved on the device

[0204] Step 2:

[0205] The terminal sends the input data to the server in JSON format.

[0206] Input: Know-how input data

[0207] Output: JSON formatted data, data sent to server

[0208] Step 3:

[0209] The server preprocesses the data it receives, including normalizing and structuring the text.

[0210] Input: JSON format data

[0211] Output: Preprocessed data

[0212] Step 4:

[0213] The server inputs the preprocessed data into a generative AI model for analysis, which includes extracting relevant data and generating new insights.

[0214] Input: Preprocessed data

[0215] Output: Analysis results

[0216] Step 5:

[0217] The server saves the analysis results in a database, along with the date and time of saving and information about the person who entered the data.

[0218] Input: Analysis results, inputter information

[0219] Output: Know-how data stored in a database

[0220] Know-how search process

[0221] Step 1:

[0222] The user enters a query into a search form on the device.

[0223] Input: A query such as "Agile development best practices"

[0224] Output: Query saved to the device

[0225] Step 2:

[0226] The device sends a query in JSON format to the server.

[0227] Input: Input data for the query

[0228] Output: Query in JSON format, query sent to server

[0229] Step 3:

[0230] The server receives the query, analyzes it using natural language processing, and extracts the intent.

[0231] Input: JSON format query

[0232] Output: Parsed intent

[0233] Step 4:

[0234] The server sends a search query to a database based on the analyzed intent to obtain related know-how data.

[0235] Input: Parsed intent

[0236] Output: Search results

[0237] Step 5:

[0238] The server returns the search results to the device in JSON format.

[0239] Input: Search results

[0240] Output: Search results in JSON format, search results sent to the device

[0241] Step 6:

[0242] The terminal displays the received search results to the user.

[0243] Input: JSON format search results

[0244] Output: Search results displayed for the user to see

[0245] Feedback Providing Process

[0246] Step 1:

[0247] After applying the know-how, the user inputs feedback from the terminal.

[0248] Input: Feedback such as "This scheme is useful, but specifically task management is lacking."

[0249] Output: Feedback data temporarily saved on the device

[0250] Step 2:

[0251] The device sends the feedback content in JSON format to the server.

[0252] Input: Feedback data

[0253] Output: Feedback data in JSON format, data sent to server

[0254] Step 3:

[0255] The server receives the feedback and performs analysis, which includes categorizing and extracting topics from the feedback text.

[0256] Input: Feedback data in JSON format

[0257] Output: Parsed feedback data

[0258] Step 4:

[0259] The server stores the analyzed feedback data as data for retraining the model.

[0260] Input: Parsed feedback data

[0261] Output: Save the data for retraining

[0262] Step 5:

[0263] The server periodically retrains the generative AI model to improve its accuracy. The improved model is applied to the system and reflected the next time a user searches or inputs know-how.

[0264] Input: Retraining data

[0265] Output: Improved model, newly applied model

[0266] Through the above processing steps, this system allows users to efficiently accumulate, search, and share know-how, and can continuously improve the system based on feedback.

[0267] (Application example 1)

[0268] 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."

[0269] To improve the operational efficiency of robots used in factories, it is necessary to efficiently share operational know-how and troubleshooting knowledge, and to search and utilize it in real time.However, conventional systems do not fully integrate knowledge input, search, and feedback, making it difficult to immediately use operational know-how and improve operations based on that feedback.

[0270] 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.

[0271] In this invention, the server includes means for acquiring user authentication information, means for authenticating the user based on the authentication information, means for inputting knowledge, means for inputting the input data to a generative AI model for analysis, means for saving the analysis results in a database, means for searching for knowledge information from the database, means for providing the search results to the user, means for collecting feedback data and using it to retrain the generative AI model, means for providing knowledge information to the robot in the form of voice and text, and means for generating and saving feedback in real time. This allows factory operational know-how to be efficiently input, searched, and shared, and enables optimization of operations through the provision of real-time feedback.

[0272] "User authentication information" is information used to identify a user and verify their authorization to access a system.

[0273] "Knowledge" refers to information about operational know-how and schemes that is input into the system.

[0274] A "generative artificial intelligence model" is an artificial intelligence algorithm that analyzes input data and extracts and generates relevant information.

[0275] A "robot" is an automated mechanical device used in factories.

[0276] "Feedback data" refers to information about evaluations and improvements provided by users to the system.

[0277] A "database" is a data management system that stores analysis results and knowledge information and makes them searchable.

[0278] "Search results" are related knowledge information retrieved based on a query entered by a user.

[0279] "Secure" means that authentication information and data are obtained and transmitted using safe and protected communications methods.

[0280] "Tagging" is a technique for improving searchability by adding metadata to knowledge data analyzed by a generative artificial intelligence model.

[0281] "Relevance" refers to the degree of match or relevance between a search query and knowledge information in a database.

[0282] MODE FOR CARRYING OUT THE INVENTION

[0283] This invention relates to a system for improving the operational efficiency of robots used in factories. This system integrates knowledge input, search, and feedback, enabling the utilization of operational know-how in real time. Specifically, it is realized with the following configuration.

[0284] User Authentication Process

[0285] When a user logs in to a terminal to operate a robot, they enter user authentication information. This authentication information includes a user ID and password. The terminal then securely sends the authentication information to the server, which then compares it with a database to authenticate the user. If successful, the server generates a session ID and returns it to the terminal.

[0286] Examples:

[0287] The user enters the ID "robot_operator" and password "password123" and confirms that the login was successful.

[0288] Knowledge Input Process

[0289] When a user inputs new knowledge information (for example, operational know-how or maintenance procedures), they use a dedicated input form. The terminal collects this data and sends it in JSON format to the server. The server preprocesses the data, analyzes it using a generative artificial intelligence model, and saves the analysis results in a database. At this time, the date and time of saving and information about the person who input the information are also recorded.

[0290] Examples:

[0291] The user inputs a "new maintenance procedure for the robot" and the analysis results are saved in a database.

[0292] Know-how search process

[0293] When a user searches for existing knowledge information, they enter a query into a search form. The device sends the query in JSON format to the server, which analyzes it using natural language processing to extract the user's intent. The server then sends the search query to a database to retrieve related knowledge data. The resulting search results are returned to the device in JSON format and provided to the user.

[0294] Examples:

[0295] A user searches for "maintenance procedures" and retrieves and displays relevant results.

[0296] Feedback Providing Process

[0297] After applying the knowledge information, the user provides feedback on its effectiveness and areas for improvement. The device sends the feedback in JSON format to the server, which receives and analyzes the feedback and stores it as retraining data for the generative AI model. This data is used to periodically retrain the model and improve the accuracy of the system.

[0298] Examples:

[0299] The user provides feedback saying "this step is effective" and the feedback is saved.

[0300] System operating environment

[0301] Hardware: Factory robots, terminals (operation panels)

[0302] Software: authentication protocols (HTTPS), data formats (JSON), generative artificial intelligence models, natural language processing (NLP) engines, database management systems

[0303] Each process is executed with a specific prompt:

[0304] Login:

[0305] "User ID: robot_operator, Password: password123"

[0306] Enter your know-how:

[0307] "New maintenance procedures for robots"

[0308] Search for know-how:

[0309] Maintenance Procedures

[0310] Providing feedback:

[0311] "This procedure is effective."

[0312] In this way, knowledge sharing and feedback can be efficiently managed in the operation of robots within a factory.

[0313] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0314] Step 1:

[0315] To log in to a terminal, a user enters a user ID and password. The entered authentication information is securely sent by the terminal to the server. The server checks the authentication information against a database, and if authentication is successful, generates a session ID and returns it to the terminal.

[0316] Input: User ID, Password

[0317] Output: Session ID (if authentication is successful)

[0318] Step 2:

[0319] The user enters new knowledge information into an input form. The device sends this information in JSON format to the server. The server preprocesses the received data and analyzes it using a generative AI model. The analysis results are stored in a database. The time of entry and information about the person who entered the information are also recorded.

[0320] Input: New knowledge information

[0321] Output: Analysis results, save date and time, and inputter information

[0322] Step 3:

[0323] A user inputs a query to search existing knowledge information. The device sends this query in JSON format to the server. The server uses a natural language processing (NLP) engine to analyze the query and extract the search intent. The server then sends the search query to a database, retrieves related knowledge data, and returns it to the device.

[0324] Input: Search query

[0325] Output: Search results (related knowledge data)

[0326] Step 4:

[0327] The user operates the robot based on the acquired knowledge information and inputs the results and opinions as feedback. The device sends the feedback information in JSON format to the server. The server analyzes the feedback and saves it as retraining data for the generative AI model. This improves the accuracy of the generative AI model.

[0328] Input: Feedback information

[0329] Output: Feedback analysis results and retraining data

[0330] Step 5:

[0331] The server periodically retrains the generative AI model using the collected feedback data, improving the model's accuracy for the next know-how input, search, or feedback.

[0332] Input: Feedback data

[0333] Output: Retrained generative artificial intelligence model

[0334] These processing steps enable effective management of knowledge input, search, sharing, and feedback on robot operations within the factory, and provide real-time feedback to optimize operations.

[0335] 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.

[0336] This invention relates to a system that efficiently accumulates know-how and schemes specific to a company, makes it easy to share and search them, and also recognizes user emotions and uses the results to provide feedback and information. Specific embodiments of this system are described below.

[0337] ---

[0338] User Authentication Process

[0339] When a user logs in to a terminal, they enter their company ID and password, which confirms their authority to access the system.

[0340] 1. The device collects the user's ID and password and sends them to the server using the HTTPS protocol.

[0341] 2. The server compares the received authentication information with the database and authenticates the user. If authentication is successful, it generates a session ID and returns it to the terminal.

[0342] 3. The terminal receives the session ID and displays a successful login message to the user.

[0343] ---

[0344] Know-how input process

[0345] This is the process by which users input new know-how and schemes into the system, thereby accumulating in-house knowledge.

[0346] 1. The user enters new know-how or scheme (e.g., "We recommend development using Scrum for Project B") into a dedicated form on the terminal.

[0347] 2. The device collects input data and sends it to the server in JSON format.

[0348] 3. The server preprocesses the data and normalizes the text (matching case, removing unnecessary whitespace, etc.).

[0349] 4. The server inputs the preprocessed data into a generative artificial intelligence model and analyzes the know-how data.

[0350] 5. The server obtains the analysis results and extracts link information with related existing know-how and new know-how.

[0351] 6. The server saves the generated know-how data in the database, along with the date and time of saving and information about the person who entered the data.

[0352] ---

[0353] Know-how search process

[0354] It is a process in which users search for existing know-how and schemes.

[0355] 1. A user enters a query (e.g., "Agile development best practices") into a search form on their device.

[0356] 2. The device sends the entered query in JSON format to the server.

[0357] 3. The server analyzes the received query using natural language processing and extracts the intent.

[0358] 4. The server sends a search query to the database to retrieve relevant know-how data. The search results also include a relevance score.

[0359] 5. The server returns the search results to the device in JSON format.

[0360] 6. The device receives the response from the server and displays the search results to the user, providing an interface that makes it easy for the user to find the information they need.

[0361] ---

[0362] User emotion recognition process

[0363] A distinctive feature of this invention is the use of an emotion engine that recognizes the user's emotions.

[0364] 1. When a user enters know-how or browses search results on the device, the emotion engine collects the user's facial expressions and voice data.

[0365] 2. The device analyzes the data collected in real time and estimates the user's emotional state (e.g., satisfaction, dissatisfaction, confusion, etc.).

[0366] 3. The device sends the estimated emotion data to the server.

[0367] 4. The server collects and stores this emotional data as feedback data, and also uses it for know-how feedback and retraining processes.

[0368] Feedback Providing Process

[0369] This is a process in which users provide feedback on the know-how they have acquired. It is characterized by the collection of feedback using an emotion engine.

[0370] 1. When a user inputs feedback from a terminal after applying know-how, the emotion engine automatically recognizes the user's emotional state.

[0371] 2. The device sends the feedback content and emotion data in JSON format to the server.

[0372] 3. The server receives the feedback and emotion data, analyzes it, and saves it as data for retraining the model.

[0373] 4. The server runs a retraining process at regular intervals to improve the performance and accuracy of the model.

[0374] 5. The server applies the improved model to the system and reflects it in the next user search or know-how input.

[0375] ---

[0376] Specific examples

[0377] Example of user authentication

[0378] 1. User "Tanaka" enters ID "tanaka" and password "password123".

[0379] 2. The device sends this to the server.

[0380] 3. The server performs authentication, is successful, and generates a session ID "abc123".

[0381] 4. The terminal saves the session ID and displays "Login successful" to the user.

[0382] Specific examples of know-how input

[0383] 1. User "Tanaka" inputs "Scrum development is recommended for Project B."

[0384] 2. The device sends the data to the server.

[0385] 3. The server inputs the data into a generative artificial intelligence model and stores the analysis results in a database.

[0386] Examples of emotion engines

[0387] 1. While the user “Tanaka” is inputting know-how, the emotion engine collects and analyzes the user’s facial expression data.

[0388] 2. The device sends the analysis results to the server and stores them as feedback data.

[0389] 3. The server uses this emotion data to retrain the AI ​​model.

[0390] This allows users to efficiently input, search, and provide feedback on internal know-how. The addition of an emotion engine also enables advanced feedback collection and customized information provision based on emotions.

[0391] The processing flow will be explained below.

[0392] User Authentication Process

[0393] Step 1:

[0394] The terminal prompts the user to enter their company ID and password. For example, the user enters "tanaka" and "password123."

[0395] Step 2:

[0396] The terminal sends the entered authentication information (ID and password) to the server using the HTTPS protocol.

[0397] Step 3:

[0398] The server queries the database for the received authentication information and searches for a match.

[0399] Step 4:

[0400] The server retrieves the search results from the database and, if authentication is successful, generates a new session ID (e.g., "sessionID_abc123").

[0401] Step 5:

[0402] The server returns the authentication result and session ID to the terminal as an HTTPS response.

[0403] Step 6:

[0404] The terminal receives the response from the server, notifies the user that login was successful, and saves the session ID.

[0405] ---

[0406] Know-how input process

[0407] Step 1:

[0408] The user inputs new know-how or schemes into a dedicated form on the terminal. For example, the user inputs "Scrum-based development is recommended for Project B."

[0409] Step 2:

[0410] The terminal sends the entered data to the server in JSON format.

[0411] Step 3:

[0412] The server preprocesses the data received and normalizes the text (matching case, removing unnecessary whitespace, etc.).

[0413] Step 4:

[0414] The server inputs the preprocessed data into a generative artificial intelligence model and analyzes the know-how data.

[0415] Step 5:

[0416] The server obtains the analysis results and extracts link information with related existing know-how and new know-how.

[0417] Step 6:

[0418] The server saves the generated know-how data in a database, along with the date and time of saving and information about the person who entered the data.

[0419] ---

[0420] Know-how search process

[0421] Step 1:

[0422] A user enters a query into a search form on their device. For example, the user enters "Agile development best practices."

[0423] Step 2:

[0424] The device sends the entered query in JSON format to the server.

[0425] Step 3:

[0426] The server analyzes the received query using natural language processing to extract the intent.

[0427] Step 4:

[0428] The server sends a search query to the database to retrieve relevant know-how data, and the search results also include a relevance score.

[0429] Step 5:

[0430] The server returns the search results to the device in JSON format.

[0431] Step 6:

[0432] The device receives the response from the server and displays the search results to the user, providing an interface that makes it easy for users to find the information they need.

[0433] ---

[0434] User emotion recognition process

[0435] Step 1:

[0436] When a user inputs know-how or browses search results on a device, the emotion engine uses the device's camera and microphone to collect the user's facial expressions and voice data.

[0437] Step 2:

[0438] The device analyzes the facial expression and voice data collected in real time, and the emotion engine estimates the user's emotional state (e.g., satisfaction, dissatisfaction, confusion, etc.).

[0439] Step 3:

[0440] The device sends the estimated emotion data in JSON format to the server.

[0441] Step 4:

[0442] The server collects and stores the received emotional data as feedback data, which is also used for know-how feedback and the retraining process of AI models.

[0443] ---

[0444] Feedback Providing Process

[0445] Step 1:

[0446] It provides feedback on the know-how acquired by the user. For example, the user may input, "This scheme is easy to understand, but I would like to see a concrete example task."

[0447] Step 2:

[0448] The device collects the feedback and sends it to the server in JSON format.

[0449] Step 3:

[0450] The server receives the feedback and performs text analysis to extract the intent and key points of the feedback.

[0451] Step 4:

[0452] The server stores the analyzed feedback data as retraining data for the generative artificial intelligence model.

[0453] Step 5:

[0454] The server runs a retraining process at regular intervals to improve the performance and accuracy of the AI ​​model.

[0455] Step 6:

[0456] The server applies the improved AI model to the system and reflects it in the next user search or know-how input.

[0457] ---

[0458] Specific examples

[0459] Example of user authentication

[0460] Step 1:

[0461] User "Tanaka" enters ID "tanaka" and password "password123".

[0462] Step 2:

[0463] The terminal sends this to the server.

[0464] Step 3:

[0465] The server performs authentication, is successful, and generates a session ID "abc123".

[0466] Step 4:

[0467] The terminal stores the session ID and displays "Login successful" to the user.

[0468] Specific examples of know-how input

[0469] Step 1:

[0470] User "Tanaka" inputs "Scrum development is recommended for Project B."

[0471] Step 2:

[0472] The device sends the data to the server.

[0473] Step 3:

[0474] The server inputs the data into a generative artificial intelligence model and stores the analysis results in a database.

[0475] Examples of emotion engines

[0476] Step 1:

[0477] While the user "Tanaka" is inputting know-how, the emotion engine collects and analyzes facial expression data using the device's camera and microphone.

[0478] Step 2:

[0479] The device sends the analysis results to the server and stores them as feedback data.

[0480] Step 3:

[0481] The server uses this emotion data to retrain the AI ​​model.

[0482] This allows users to efficiently input, search, and provide feedback on internal know-how. The addition of an emotion engine also enables advanced feedback collection and customized information provision based on emotions.

[0483] Example 2

[0484] 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."

[0485] Existing knowledge management systems make it difficult to effectively share and search know-how and schemes accumulated within an organization. They also lack a feedback mechanism that takes user sentiment into account, limiting the means to improve system usability and knowledge quality. This poses a challenge: it is difficult to effectively collect useful knowledge held by users and share and utilize it throughout the organization.

[0486] 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.

[0487] In this invention, the server includes means for acquiring authentication information, means for authenticating users, means for inputting organization-specific knowledge and schemes, means for inputting input data into a generative AI model for analysis, means for saving the analysis results in a database, means for searching for knowledge information from the database, means for providing search results to users, means for recognizing the emotional state of users and collecting data on the same, and means for collecting the emotional data as feedback and using it to retrain the model, thereby enabling users to effectively input and search for knowledge and providing feedback based on their emotional state.

[0488] "Authentication information" refers to information such as ID and password required when a user logs in to a system.

[0489] "User authentication" is the process of verifying a user's access rights based on the authentication information entered.

[0490] "Organization-specific knowledge and schemes" refers to information assets such as experience, know-how, business processes, and best practices within the organization.

[0491] A "generative AI model" is a generative artificial intelligence model that analyzes input data and generates new insights and related information.

[0492] "Analysis results" are the information and insights obtained after input data is analyzed by a generative AI model.

[0493] A "database" is a system for systematically storing and managing analysis results and other information.

[0494] A "search query" is a keyword or phrase that a user enters to search for specific information.

[0495] "Emotional state" refers to emotions estimated from the user's facial expressions and voice data, and includes satisfaction, dissatisfaction, confusion, etc.

[0496] "Feedback" refers to the ratings and comments users provide after using the system.

[0497] "Retraining methods" refers to the process of retraining an AI model using collected feedback and sentiment data to improve the model's performance.

[0498] This invention is a system for effectively inputting, searching, and sharing knowledge and schemes accumulated within an organization, and it can also recognize users' emotions and reflect their feedback. Specific embodiments of this system are described below.

[0499] This system mainly consists of three elements: a server, a terminal, and a user.

[0500] User Authentication Process

[0501] When a user logs in to a terminal, they enter their company ID and password to confirm their authorization to access the system. This authentication process is carried out securely using the HTTPS protocol. The server checks the authentication information against a database, and if it matches, it generates a session ID and returns it to the terminal.

[0502] Know-how input process

[0503] When a user inputs new knowledge or schemes, they use a dedicated form on their device. This data is sent in JSON format to the server, which then inputs the data into a generative AI model for analysis. The analysis results are stored in a database and shared across the organization. For example, if a user inputs "We recommend developing using Scrum for Project B," the data is preprocessed on the server and analyzed by a generative AI model (e.g., OpenAI® GPT-3®).

[0504] Know-how search process

[0505] When a user searches for existing knowledge or schemes, they enter a query into a search form on their device. The query is sent in JSON format to the server, which then analyzes the query's intent using natural language processing. Based on the analysis results, the server retrieves relevant information from the database and provides it to the user. For example, if a user searches for "best practices in agile development," the server provides the user with relevant information based on this query.

[0506] User emotion recognition process

[0507] When a user enters knowledge or browses search results, the device's built-in camera and microphone collect facial and voice data in real time, which is then input into the emotion analysis algorithm. The server receives this emotion data, collects it as feedback, and uses it to retrain the AI ​​model. For example, if a user is confused while entering know-how, the emotion engine will identify that situation and send it to the server.

[0508] Feedback Providing Process

[0509] When users provide feedback on the information obtained by the system, their emotional data is collected along with the feedback. The server analyzes the feedback and emotional data and stores it as retraining data to improve the accuracy and performance of the AI ​​model.

[0510] This system allows users to effectively input and search knowledge within an organization and reflects feedback based on emotional data. For example, by inputting important know-how about a project and reusing that know-how in other projects, the efficiency of the entire organization can be improved. Emotion recognition is also expected to improve usability.

[0511] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0512] Step 1:

[0513] The user enters their company ID and password into the terminal.

[0514] (Input): User ID, password

[0515] (Operation): The user enters their company ID and password on the device's login screen.

[0516] (Output): The device collects the ID and password.

[0517] Step 2:

[0518] The device sends the collected authentication information to the server using the HTTPS protocol.

[0519] (Input): User ID, password

[0520] (Operation): The device converts the ID and password into JSON format and sends it as an HTTPS request to the server's authentication API endpoint.

[0521] (Output): The server receives the authentication information.

[0522] Step 3:

[0523] The server compares the received authentication information with a database and authenticates the user.

[0524] (Input): Authentication information (user ID, password)

[0525] (What happens): The server compares the user information in its database with the entered authentication information.

[0526] (Output): Authentication result (success / failure)

[0527] Step 4:

[0528] When authentication is successful, the server generates a session ID and returns it to the terminal.

[0529] (Input): Authentication information (if successful)

[0530] (Operation): The server generates a unique session ID and sends it to the device in JSON format.

[0531] (Output): Session ID

[0532] Step 5:

[0533] The terminal receives the session ID and displays a "Login successful" message to the user.

[0534] (Input): Session ID

[0535] (Action): The device saves the session ID in local storage and displays a login success message.

[0536] (Output): User interface update (login success message)

[0537] Step 6:

[0538] The user inputs new knowledge or scheme into an input form.

[0539] (Input): Know-how and scheme (text data)

[0540] (Operation): The user inputs new knowledge or scheme into a dedicated form on the terminal.

[0541] (Output): The terminal collects input data.

[0542] Step 7:

[0543] The input data collected by the terminal is sent to the server in JSON format.

[0544] (Input): Know-how and scheme (JSON format)

[0545] (Operation): The terminal converts the input data into JSON format and sends it to the server as an HTTPS request.

[0546] (Output): The server receives the data.

[0547] Step 8:

[0548] The server preprocesses the data and inputs it into a generative AI model for analysis.

[0549] (Input): Know-how and scheme (JSON format)

[0550] (Operation): The server normalizes the data (standardizing uppercase and lowercase letters, removing unnecessary spaces) and inputs it into a generative AI model. The generative AI model analyzes the data and generates new knowledge and related information.

[0551] (Output): Analysis results (new knowledge, related information)

[0552] Step 9:

[0553] The server stores the analysis results in a database.

[0554] (Input): Analysis results (new knowledge, related information)

[0555] (Operation): The server adds the analysis results to the database as a new record, and also records the save date and time and the ID of the person who entered the data.

[0556] (Output): Update the database (save new knowledge)

[0557] Step 10:

[0558] A user enters a query into a search form.

[0559] (Input): Search query (text data)

[0560] Action: A user enters a query into a search form on their device.

[0561] (Output): The terminal collects input data.

[0562] Step 11:

[0563] The device sends the entered query in JSON format to the server.

[0564] (Input): Search query (JSON format)

[0565] (Operation): The terminal converts the input data into JSON format and sends it to the server as an HTTPS request.

[0566] (Output): The server receives the data.

[0567] Step 12:

[0568] The server parses the query and performs a search against the database.

[0569] (Input): Search query (JSON format)

[0570] (Operation): The server analyzes the query using a natural language processing algorithm and sends the search query to the database.

[0571] (Output): Search results (related knowledge information)

[0572] Step 13:

[0573] The server returns the search results to the device in JSON format.

[0574] (Input): Search results (JSON format)

[0575] (Operation): The server encodes the search results into JSON format and sends them to the device as an HTTPS request.

[0576] (Output): The device receives the data.

[0577] Step 14:

[0578] Your device will display the search results.

[0579] (Input): Search results (JSON format)

[0580] (Operation): The terminal displays the search results received on the user interface, making it easy for the user to find the information they need.

[0581] (Output): Update the user interface (display search results)

[0582] Step 15:

[0583] While the user is entering know-how or viewing search results, the device's built-in camera and microphone collect facial expressions and voice data.

[0584] (Input): facial expression data, voice data

[0585] (Operation): The device uses its built-in camera and microphone to collect facial and voice data in real time.

[0586] (Output): Collected emotion data

[0587] Step 16:

[0588] The device analyzes the emotional data collected in real time to estimate the user's emotional state.

[0589] (Input): Collected emotion data

[0590] (Action): The device uses an emotion analysis algorithm to estimate the emotional state (e.g., happy, unhappy, confused).

[0591] (Output): Estimated emotional state

[0592] Step 17:

[0593] The device transmits the estimated emotion data to the server.

[0594] (Input): Estimated emotional state

[0595] (Operation): The device converts the emotion data into JSON format and sends it to the server as an HTTPS request.

[0596] (Output): The server receives the emotion data.

[0597] Step 18:

[0598] The server collects the emotion data as feedback data and uses it to retrain the model.

[0599] (Input): Emotion data (JSON format)

[0600] (How it works): The server stores the emotion data in a feedback database and uses it to retrain the AI ​​model.

[0601] (Output): Update the database and save the retraining data.

[0602] Step 19:

[0603] The server runs a retraining process at regular intervals to improve the model's performance.

[0604] (Input): Retraining data

[0605] (Operation): The server periodically runs a retraining process to improve the accuracy and performance of the AI ​​model.

[0606] (Output): Improved AI model

[0607] Step 20:

[0608] The server applies the improved model to the system and reflects it in the next user search or know-how input.

[0609] (Input): Improved AI model

[0610] (Operation): The server deploys the updated AI model to the system and applies it the next time the user searches or types something.

[0611] (Output): System update (application of improved AI model)

[0612] The above are the specific processing steps of the program of this system.

[0613] (Application example 2)

[0614] 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."

[0615] Improving work efficiency and sharing knowledge within factories are important issues, but conventional systems have had difficulty providing real-time support or feedback based on workers' emotions. Furthermore, collecting feedback that reflects emotional states and how to utilize it has also been an unsolved problem.

[0616] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for identifying the emotional state of a user using an emotion recognition engine and collecting and saving that data, means for collecting feedback and providing information using the emotion data, and means for tagging knowledge data analyzed using a generative artificial intelligence model and making it searchable based on relevance. This not only improves work efficiency in the factory but also makes it possible to provide appropriate feedback based on the emotions of workers.

[0617] "Authentication information" refers to information used to verify a user's identity, and is usually comprised of an ID and password.

[0618] "User authentication" refers to the process of verifying whether a user is a person with legitimate authority using the acquired authentication information.

[0619] "Internal knowledge" refers to specialized information and procedures accumulated within an organization.

[0620] A "generative artificial intelligence model" is an artificial intelligence model that has the ability to generate and analyze knowledge and information based on new data.

[0621] A "database" is a system for efficiently storing and managing data.

[0622] "Knowledge information" refers to internal knowledge and procedures stored in a database.

[0623] "Search results" refers to the related knowledge information obtained when searching a database.

[0624] "Feedback data" refers to data collected from users' evaluations and opinions.

[0625] "Model retraining" is the process of using new data to improve the performance of an artificial intelligence model.

[0626] An "emotion recognition engine" is a technology that analyzes a user's facial expressions and voice to identify their emotional state.

[0627] "Emotional state" is information indicating the type and intensity of the user's emotion.

[0628] "Feedback collection" is the process of collecting opinions and sentiment data from users.

[0629] "Information provision" is the process of displaying search results and feedback to users.

[0630] "Tagging" means assigning labels to data that indicate its relevance.

[0631] This invention uses smart glasses and head-mounted displays (HMDs) to build a system that improves work efficiency in factories. The system includes multiple means for user authentication, know-how input, know-how search, emotion recognition, and feedback collection.

[0632] User Authentication

[0633] The user puts on the smart glasses and enters their ID and password. This authentication information is securely sent to the server using the HTTPS protocol. The server compares the received authentication information with a database, and if authentication is successful, it generates a session ID and returns it to the device. The device saves the session ID and displays a message to the user indicating successful login.

[0634] Input of know-how

[0635] When a user wants to input new know-how or procedures, they use a dedicated form on the smart glasses. The device sends the input data in JSON format to the server, which then preprocesses the data and analyzes it using a generative artificial intelligence model. The analysis results are stored in a database. This process efficiently accumulates knowledge and procedures acquired within the factory.

[0636] Search for know-how

[0637] When a user searches for a specific know-how or procedure, they enter a query into the search form on the smart glasses. The device sends the query in JSON format to the server, which uses natural language processing technology to analyze the query's intent. The server then sends the search query to a database to retrieve relevant know-how information. The search results are returned to the device and displayed visually to the user.

[0638] emotion recognition

[0639] When a user enters know-how or browses search results, the emotion engine collects the user's facial expression and voice data. The device analyzes this data in real time and estimates the user's emotional state. The estimated emotion data is sent to the server and stored as feedback data. This emotion data is used to retrain the model and provide feedback on know-how.

[0640] Feedback collection

[0641] When a user provides feedback, the emotion engine automatically recognizes the user's emotional state. The feedback content and emotion data are sent from the device to the server, where they are analyzed and stored as retraining data for the model. A periodic retraining process is performed to improve the model's performance and accuracy.

[0642] Specific examples

[0643] For example, if a user wears smart glasses and types "Scrum development is recommended for Project B," the device sends the data to the server, where it is analyzed by a generative artificial intelligence model. The analysis results are stored in a database and will be displayed as relevant information when other users later search for "best practices for agile development."

[0644] Prompt Sentence Examples

[0645] Analyze the following know-how data for your generative AI model:

[0646] "Scrum development is recommended for Project B."

[0647] This invention will enable efficient accumulation and sharing of knowledge within factories, and will also enable feedback collection and information provision based on user emotions, which is expected to improve worker satisfaction and work efficiency.

[0648] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0649] Step 1:

[0650] The user puts on the smart glasses and enters their ID and password.

[0651] Input: User ID, Password

[0652] Specific operation: The user enters their ID and password into the login form displayed on the smart glasses display. The smart glasses then send this authentication information to the server using the HTTPS protocol.

[0653] Output: Authentication information sent to the server

[0654] Step 2:

[0655] The server receives the authentication information and authenticates the user by checking it against a database.

[0656] Input: Authentication information (ID, password)

[0657] How it works: The server checks the received authentication information against the user information in its database, and if authentication is successful, generates a session ID, which is then sent back to the smart glasses in a secure manner.

[0658] Output: Session ID

[0659] Step 3:

[0660] The terminal stores the session ID and displays a successful login message to the user.

[0661] Input: Session ID

[0662] Specific operation: The smart glasses store the session ID in memory and notify the user that the login was successful by displaying it on the display.

[0663] Output: Login successful message

[0664] Step 4:

[0665] The user enters new know-how and procedures into a special form on the smart glasses.

[0666] Input: Know-how and procedures

[0667] Specific operation: The user inputs new know-how or procedures (e.g., "We recommend developing using Scrum for Project B") into a specific form through the smart glasses' display. The input data is temporarily stored in the smart glasses' internal storage.

[0668] Output: Input data is temporarily saved

[0669] Step 5:

[0670] The terminal sends the entered data to the server in JSON format.

[0671] Input: Input data

[0672] Specific operation: The smart glasses convert the temporarily stored know-how and procedure data into JSON format and send it to the server.

[0673] Output: JSON format data is sent to the server

[0674] Step 6:

[0675] The server preprocesses the data and performs analysis using a generative artificial intelligence model.

[0676] Input: Input data in JSON format

[0677] How it works: The server preprocesses the received JSON data (e.g., standardizing case, removing unnecessary whitespace, etc.) and then feeds it into a generative artificial intelligence model. The model then analyzes the data and extracts new information and related existing knowledge.

[0678] Output: Analysis results

[0679] Step 7:

[0680] The server stores the analysis results in a database.

[0681] Input: Analysis results

[0682] Specific operation: The server saves the analysis results in a database, and also records the date and time of saving and information about the person who entered the data.

[0683] Output: Analysis results stored in a database

[0684] Step 8:

[0685] A user searches for know-how by entering a query into a search form on the smart glasses.

[0686] Input: Search query

[0687] What it does: A user enters a specific query (e.g., "Agile development best practices") into the search form on the smart glasses. The smart glasses convert the query into JSON format and send it to the server.

[0688] Output: The search query is sent to the server

[0689] Step 9:

[0690] The server analyzes the received query and searches the database for related know-how information.

[0691] Input: Search query

[0692] Specific operation: The server analyzes the received search query using natural language processing technology, sends an appropriate search query to the database, and retrieves related know-how information.

[0693] Output: Search results

[0694] Step 10:

[0695] The server sends the search results back to the device, which then displays them on the smart glasses.

[0696] Input: Search results

[0697] Specific operation: The server converts the search results into JSON format and sends them to the smart glasses, which then display them visually for the user to view.

[0698] Output: Search results displayed to the user

[0699] Step 11:

[0700] The emotion recognition engine collects the user's facial expression and voice data and analyzes their emotional state in real time.

[0701] Input: facial expression data, voice data

[0702] How it works: The smart glasses' built-in cameras and microphones collect the user's facial expressions and voice in real time. The emotion recognition engine analyzes this data and estimates the user's emotional state (e.g., satisfaction, dissatisfaction, confusion, etc.).

[0703] Output: Estimated emotional state

[0704] Step 12:

[0705] The device transmits the estimated emotion data to the server and stores it as feedback data.

[0706] Input: Emotion data

[0707] Specific operation: The emotion recognition engine converts the estimated emotional state into JSON format and sends it to the server. The server then stores the received emotional data in a database as feedback data.

[0708] Output: Emotion data stored in a database

[0709] Step 13:

[0710] The server uses the feedback data to retrain the model.

[0711] Input: Feedback data

[0712] Specific operation: The server periodically collects feedback data and retrains the generative AI model, thereby improving the model's performance and accuracy.

[0713] Output: Retrained generative artificial intelligence model

[0714] Through the above processing steps, the present invention improves work efficiency in a factory and provides appropriate feedback based on the user's emotions.

[0715] 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.

[0716] 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.

[0717] 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.

[0718] [Second embodiment]

[0719] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0720] 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.

[0721] 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).

[0722] 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.

[0723] 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.

[0724] 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).

[0725] 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.

[0726] 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.

[0727] 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.

[0728] 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.

[0729] 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.

[0730] 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."

[0731] The present invention relates to a system for efficiently storing know-how and schemes specific to a company, and facilitating sharing and searching thereof. A specific embodiment of this system will be described below.

[0732] ---

[0733] User Authentication Process

[0734] When a user logs in to a terminal, they enter their company ID and password, which confirms their authority to access the system.

[0735] 1. The device collects the user's ID and password and sends them to the server using the HTTPS protocol.

[0736] 2. The server compares the received authentication information with the database and authenticates the user. If authentication is successful, it generates a session ID and returns it to the terminal.

[0737] 3. The terminal receives the session ID and displays a successful login message to the user.

[0738] Know-how input process

[0739] This is the process by which users input new know-how and schemes into the system, thereby accumulating in-house knowledge.

[0740] 1. The user enters new know-how or scheme (e.g., "Agile development is being adopted for Project A") into a dedicated form on the device.

[0741] 2. The device collects input data and sends it to the server in JSON format.

[0742] 3. The server preprocesses the data, normalizing and structuring the text.

[0743] 4. The server inputs the preprocessed data into a generative AI model for analysis. The analysis results include related existing data and new insights.

[0744] 5. The server saves the analysis results in a database. When saving, the date and time of saving and information about the person who entered the data are also recorded.

[0745] Know-how search process

[0746] It is a process in which users search for existing know-how and schemes.

[0747] 1. A user enters a query (e.g., "Agile development best practices") into a search form on their device.

[0748] 2. The device sends a query in JSON format to the server.

[0749] 3. The server analyzes the received query using natural language processing to extract the intent.

[0750] 4. The server sends a search query to the database to retrieve relevant know-how data.

[0751] 5. The server returns the search results to the device in JSON format.

[0752] 6. The device displays the search results to the user.

[0753] Feedback Providing Process

[0754] This is the process of providing feedback to users on the know-how they have acquired.

[0755] 1. The user inputs feedback from the terminal after applying the know-how (e.g., "This scheme is useful, but specifically, task management is lacking").

[0756] 2. The device sends the feedback content in JSON format to the server.

[0757] 3. The server receives the feedback, analyzes it, and saves it as data for retraining the model.

[0758] 4. The server periodically retrains the generative AI model to improve its accuracy.

[0759] 5. The server applies the improved model to the system and reflects it in the next user search or know-how input.

[0760] ---

[0761] Specific examples

[0762] Example of user authentication

[0763] 1. User "Tanaka" enters ID "tanaka" and password "password123".

[0764] 2. The device sends this to the server.

[0765] 3. The server performs authentication, is successful, and generates a session ID "abc123".

[0766] 4. The terminal saves the session ID and displays "Login successful" to the user.

[0767] Specific examples of know-how input

[0768] 1. User "Tanaka" inputs "Scrum development is recommended for Project B."

[0769] 2. The device sends the data to the server.

[0770] 3. The server inputs the data into a generative artificial intelligence model and stores the analysis results in a database.

[0771] This will create a system that allows users to efficiently input, search, and share in-house know-how.

[0772] The processing flow will be explained below.

[0773] User Authentication Process

[0774] Step 1:

[0775] The terminal prompts the user to enter their company ID and password. For example, the user enters "tanaka" and "password123."

[0776] Step 2:

[0777] The terminal sends the entered authentication information to the server using the HTTPS protocol.

[0778] Step 3:

[0779] The server queries the database for the received authentication information and searches for a match.

[0780] Step 4:

[0781] The server retrieves the search results from the database and, if authentication is successful, generates a new session ID (e.g., "sessionID_abc123").

[0782] Step 5:

[0783] The server returns the authentication result and session ID to the terminal as an HTTPS response.

[0784] Step 6:

[0785] The terminal receives the response from the server, notifies the user that login was successful, and saves the session ID.

[0786] ---

[0787] Know-how input process

[0788] Step 1:

[0789] The user inputs new know-how or schemes into a dedicated form on the terminal. For example, the user inputs "Scrum-based development is recommended for Project B."

[0790] Step 2:

[0791] The device collects the input data and sends it to the server in JSON format.

[0792] Step 3:

[0793] The server preprocesses the data received and normalizes the text (matching case, removing unnecessary whitespace, etc.).

[0794] Step 4:

[0795] The server inputs the preprocessed data into a generative artificial intelligence model and analyzes the know-how data.

[0796] Step 5:

[0797] The server obtains the analysis results and extracts link information with related existing know-how and new know-how.

[0798] Step 6:

[0799] The server saves the generated know-how data in a database, along with the date and time of saving and information about the person who entered the data.

[0800] ---

[0801] Know-how search process

[0802] Step 1:

[0803] A user enters a query into a search form on their device. For example, the user enters "Agile development best practices."

[0804] Step 2:

[0805] The device sends the entered query in JSON format to the server.

[0806] Step 3:

[0807] The server analyzes the received query using natural language processing to understand the intent of the query and extract related keywords.

[0808] Step 4:

[0809] The server sends a search query to the database to retrieve relevant know-how data, and the search results also include a relevance score.

[0810] Step 5:

[0811] The server returns the search results to the device in JSON format.

[0812] Step 6:

[0813] The device receives the response from the server and displays the search results to the user, providing an interface that makes it easy for users to find the information they need.

[0814] ---

[0815] Feedback Providing Process

[0816] Step 1:

[0817] It provides feedback on the know-how acquired by the user. For example, the user may input, "This scheme is easy to understand, but I would like to see a concrete example task."

[0818] Step 2:

[0819] The device collects the feedback and sends it to the server in JSON format.

[0820] Step 3:

[0821] The server receives the feedback and performs text analysis to extract the intent and key points of the feedback.

[0822] Step 4:

[0823] The server stores the analyzed feedback data as retraining data for the generative artificial intelligence model.

[0824] Step 5:

[0825] The server runs a retraining process at regular intervals to improve the performance and accuracy of the model.

[0826] Step 6:

[0827] The server applies the improved model to the system and reflects it in the next user search or know-how input.

[0828] Through these steps, users can efficiently input, search, and provide feedback on their internal know-how.

[0829] Example 1

[0830] 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."

[0831] Previous methods for sharing in-house know-how and schemes mainly involved manual recording and management, which lacked efficiency. It was also difficult to search for know-how or discover new insights, resulting in low utilization of information. Furthermore, there was no established method for collecting feedback and using it to update and improve knowledge. For these reasons, a new system for effectively storing and utilizing in-house knowledge resources was needed.

[0832] 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.

[0833] In this invention, the server includes means for acquiring authentication information, means for authenticating a user based on the authentication information, means for inputting in-house know-how and schemes, means for transmitting the input data to the server in JSON format, means for inputting the input data to a generative artificial intelligence model for analysis, means for saving the analysis results in a database, means for searching the database for know-how information using natural language processing, means for providing the search results to the user in JSON format, and means for collecting feedback data and using it to retrain the model. This enables efficient accumulation, sharing, and search of in-house know-how, and further enables the model to be improved based on feedback to increase the quality of knowledge.

[0834] "Authentication Information" means information required to verify a user's identity, and primarily includes a user ID and password.

[0835] "User authentication" is the process of verifying that a user is a legitimate user using authentication information.

[0836] "Know-how" refers to information that refers to knowledge, techniques, and methodologies related to a specific task or project.

[0837] "Scheme" means the plan, structure or procedure for a particular project or business process.

[0838] "JSON format" is a lightweight data exchange format for structuring data, and is an abbreviation for JavaScript Object Notation.

[0839] A "generative artificial intelligence model" is a set of machine learning algorithms that generate new information and insights based on input data.

[0840] "Analysis" is the process performed to understand input data and find patterns and relationships.

[0841] A "database" is a system for organizing, efficiently storing, and retrieving information.

[0842] "Natural language processing" is a technical field that uses computers to understand and process human language.

[0843] "Feedback data" refers to information such as ratings, comments, and improvements provided by users regarding the system's output.

[0844] "Model retraining" is the process of updating a generative artificial intelligence model with new data collected to improve its accuracy and performance.

[0845] This invention relates to a system that efficiently accumulates in-house know-how and schemes and makes it easy to share and search them. This system includes a series of processes for collecting, transmitting, analyzing, and storing data between a server, terminals, and users.

[0846] User Authentication Process

[0847] When a user enters their company ID and password into the terminal, the terminal encrypts the authentication information and sends it to the server using the HTTPS protocol. The server compares the authentication information with a database, and if authentication is successful, generates a session ID and returns it to the terminal. This allows the user to access the system securely.

[0848] Know-how input process

[0849] The user inputs new know-how or schemes into a dedicated form on the device. For example, this information might be something like "Scrum development is recommended for Project B." The device then sends the input data in JSON format to the server. The server preprocesses the data and analyzes it using a generative artificial intelligence model. The analysis results are saved in a database, along with the date and time of saving and information about the person who entered the data.

[0850] Know-how search process

[0851] When a user enters a query (e.g., "Agile development best practices") into the device's search form, the device sends the query in JSON format to the server. The server analyzes the query using natural language processing technology and sends the search query to the database. The search results are returned to the device in JSON format and displayed to the user.

[0852] Feedback Providing Process

[0853] After applying the know-how, the user inputs feedback from the device. For example, the user might say, "This scheme is useful, but specifically, task management is lacking." The device then sends the feedback in JSON format to the server. The server analyzes the feedback and saves it as retraining data for the generative AI model. This improves the model's accuracy, which is reflected the next time the user searches or inputs know-how.

[0854] Hardware and software used

[0855] This system uses the following hardware and software:

[0856] Device: A computer or smartphone operated by a user.

[0857] Server: A computer system that processes and stores data.

[0858] Database: A system for storing know-how, schemes, analysis results, and feedback data.

[0859] HTTPS protocol: A communications protocol for secure data transmission.

[0860] Generative artificial intelligence models: Algorithms that analyze input data and generate new insights.

[0861] Specific examples

[0862] Example of user authentication

[0863] The user enters the ID "tanaka" and password "password123" into the terminal, which then encrypts and sends it to the server. The server compares it with the database, and if authentication is successful, it generates a session ID "abc123" and returns it to the terminal. The terminal saves the session ID and displays "Login successful" to the user.

[0864] Specific examples of know-how input

[0865] The user inputs "Scrum development is recommended for Project B," and the device sends the data in JSON format to the server. The server preprocesses the data, inputs it into a generative artificial intelligence model for analysis, and stores the analysis results in a database.

[0866] Specific examples of know-how search

[0867] A user types "Agile development best practices" into a search form, and the device sends the query to the server, which parses the query, searches the database, and returns relevant results to the device, where the user is presented with the search results.

[0868] Example prompt sentence:

[0869] "Project B is recommended to be developed using Scrum. What are some best practices that can be used for other projects?"

[0870] This system enables efficient accumulation, sharing, and searching of internal know-how, allowing for improvements to generative AI models based on feedback.

[0871] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0872] Program processing flow

[0873] User Authentication Process

[0874] Step 1:

[0875] The user enters their company ID and password into the device's login screen.

[0876] Input: Company ID, password

[0877] Output: Credentials sent to the device

[0878] Step 2:

[0879] The terminal encrypts the entered ID and password using the HTTPS protocol and sends it to the server.

[0880] Input: Company ID, password

[0881] Output: Authentication information encrypted using the HTTPS protocol

[0882] Step 3:

[0883] The server checks the received authentication information against a database and authenticates the user.

[0884] Input: Credentials

[0885] Output: Authentication success or failure, session ID

[0886] Step 4:

[0887] If the authentication is successful, the server generates a session ID and returns it to the terminal.

[0888] Input: Authentication success result

[0889] Output: Session ID

[0890] Step 5:

[0891] The terminal receives the session ID and displays a successful login message to the user.

[0892] Input: Session ID

[0893] Output: Login success message, session ID stored on the device

[0894] Know-how input process

[0895] Step 1:

[0896] The user inputs new know-how and schemes into a dedicated form on the terminal.

[0897] Input: Know-how such as "Scrum development is recommended for Project B"

[0898] Output: Input data temporarily saved on the device

[0899] Step 2:

[0900] The terminal sends the input data to the server in JSON format.

[0901] Input: Know-how input data

[0902] Output: JSON formatted data, data sent to server

[0903] Step 3:

[0904] The server preprocesses the data it receives, including normalizing and structuring the text.

[0905] Input: JSON format data

[0906] Output: Preprocessed data

[0907] Step 4:

[0908] The server inputs the preprocessed data into a generative AI model for analysis, which includes extracting relevant data and generating new insights.

[0909] Input: Preprocessed data

[0910] Output: Analysis results

[0911] Step 5:

[0912] The server saves the analysis results in a database, along with the date and time of saving and information about the person who entered the data.

[0913] Input: Analysis results, inputter information

[0914] Output: Know-how data stored in a database

[0915] Know-how search process

[0916] Step 1:

[0917] The user enters a query into a search form on the device.

[0918] Input: A query such as "Agile development best practices"

[0919] Output: Query saved to the device

[0920] Step 2:

[0921] The device sends a query in JSON format to the server.

[0922] Input: Input data for the query

[0923] Output: Query in JSON format, query sent to server

[0924] Step 3:

[0925] The server receives the query, analyzes it using natural language processing, and extracts the intent.

[0926] Input: JSON format query

[0927] Output: Parsed intent

[0928] Step 4:

[0929] The server sends a search query to a database based on the analyzed intent to obtain related know-how data.

[0930] Input: Parsed intent

[0931] Output: Search results

[0932] Step 5:

[0933] The server returns the search results to the device in JSON format.

[0934] Input: Search results

[0935] Output: Search results in JSON format, search results sent to the device

[0936] Step 6:

[0937] The terminal displays the received search results to the user.

[0938] Input: JSON format search results

[0939] Output: Search results displayed for the user to see

[0940] Feedback Providing Process

[0941] Step 1:

[0942] After applying the know-how, the user inputs feedback from the terminal.

[0943] Input: Feedback such as "This scheme is useful, but specifically task management is lacking."

[0944] Output: Feedback data temporarily saved on the device

[0945] Step 2:

[0946] The device sends the feedback content in JSON format to the server.

[0947] Input: Feedback data

[0948] Output: Feedback data in JSON format, data sent to server

[0949] Step 3:

[0950] The server receives the feedback and performs analysis, which includes categorizing and extracting topics from the feedback text.

[0951] Input: Feedback data in JSON format

[0952] Output: Parsed feedback data

[0953] Step 4:

[0954] The server stores the analyzed feedback data as data for retraining the model.

[0955] Input: Parsed feedback data

[0956] Output: Save the data for retraining

[0957] Step 5:

[0958] The server periodically retrains the generative AI model to improve its accuracy. The improved model is applied to the system and reflected the next time a user searches or inputs know-how.

[0959] Input: Retraining data

[0960] Output: Improved model, newly applied model

[0961] Through the above processing steps, this system allows users to efficiently accumulate, search, and share know-how, and can continuously improve the system based on feedback.

[0962] (Application example 1)

[0963] 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."

[0964] To improve the operational efficiency of robots used in factories, it is necessary to efficiently share operational know-how and troubleshooting knowledge, and to search and utilize it in real time.However, conventional systems do not fully integrate knowledge input, search, and feedback, making it difficult to immediately use operational know-how and improve operations based on that feedback.

[0965] 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.

[0966] In this invention, the server includes means for acquiring user authentication information, means for authenticating the user based on the authentication information, means for inputting knowledge, means for inputting the input data to a generative AI model for analysis, means for saving the analysis results in a database, means for searching for knowledge information from the database, means for providing the search results to the user, means for collecting feedback data and using it to retrain the generative AI model, means for providing knowledge information to the robot in the form of voice and text, and means for generating and saving feedback in real time. This allows factory operational know-how to be efficiently input, searched, and shared, and enables optimization of operations through the provision of real-time feedback.

[0967] "User authentication information" is information used to identify a user and verify their authorization to access a system.

[0968] "Knowledge" refers to information about operational know-how and schemes that is input into the system.

[0969] A "generative artificial intelligence model" is an artificial intelligence algorithm that analyzes input data and extracts and generates relevant information.

[0970] A "robot" is an automated mechanical device used in factories.

[0971] "Feedback data" refers to information about evaluations and improvements provided by users to the system.

[0972] A "database" is a data management system that stores analysis results and knowledge information and makes them searchable.

[0973] "Search results" are related knowledge information retrieved based on a query entered by a user.

[0974] "Secure" means that authentication information and data are obtained and transmitted using safe and protected communications methods.

[0975] "Tagging" is a technique for improving searchability by adding metadata to knowledge data analyzed by a generative artificial intelligence model.

[0976] "Relevance" refers to the degree of match or relevance between a search query and knowledge information in a database.

[0977] MODE FOR CARRYING OUT THE INVENTION

[0978] This invention relates to a system for improving the operational efficiency of robots used in factories. This system integrates knowledge input, search, and feedback, enabling the utilization of operational know-how in real time. Specifically, it is realized with the following configuration.

[0979] User Authentication Process

[0980] When a user logs in to a terminal to operate a robot, they enter user authentication information. This authentication information includes a user ID and password. The terminal then securely sends the authentication information to the server, which then compares it with a database to authenticate the user. If successful, the server generates a session ID and returns it to the terminal.

[0981] Examples:

[0982] The user enters the ID "robot_operator" and password "password123" and confirms that the login was successful.

[0983] Knowledge Input Process

[0984] When a user inputs new knowledge information (for example, operational know-how or maintenance procedures), they use a dedicated input form. The terminal collects this data and sends it in JSON format to the server. The server preprocesses the data, analyzes it using a generative artificial intelligence model, and saves the analysis results in a database. At this time, the date and time of saving and information about the person who input the information are also recorded.

[0985] Examples:

[0986] The user inputs a "new maintenance procedure for the robot" and the analysis results are saved in a database.

[0987] Know-how search process

[0988] When a user searches for existing knowledge information, they enter a query into a search form. The device sends the query in JSON format to the server, which analyzes it using natural language processing to extract the user's intent. The server then sends the search query to a database to retrieve related knowledge data. The resulting search results are returned to the device in JSON format and provided to the user.

[0989] Examples:

[0990] A user searches for "maintenance procedures" and retrieves and displays relevant results.

[0991] Feedback Providing Process

[0992] After applying the knowledge information, the user provides feedback on its effectiveness and areas for improvement. The device sends the feedback in JSON format to the server, which receives and analyzes the feedback and stores it as retraining data for the generative AI model. This data is used to periodically retrain the model and improve the accuracy of the system.

[0993] Examples:

[0994] The user provides feedback saying "this step is effective" and the feedback is saved.

[0995] System operating environment

[0996] Hardware: Factory robots, terminals (operation panels)

[0997] Software: authentication protocols (HTTPS), data formats (JSON), generative artificial intelligence models, natural language processing (NLP) engines, database management systems

[0998] Each process is executed with a specific prompt:

[0999] Login:

[1000] "User ID: robot_operator, Password: password123"

[1001] Enter your know-how:

[1002] "New maintenance procedures for robots"

[1003] Search for know-how:

[1004] Maintenance Procedures

[1005] Providing feedback:

[1006] "This procedure is effective."

[1007] In this way, knowledge sharing and feedback can be efficiently managed in the operation of robots within a factory.

[1008] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1009] Step 1:

[1010] To log in to a terminal, a user enters a user ID and password. The entered authentication information is securely sent by the terminal to the server. The server checks the authentication information against a database, and if authentication is successful, generates a session ID and returns it to the terminal.

[1011] Input: User ID, Password

[1012] Output: Session ID (if authentication is successful)

[1013] Step 2:

[1014] The user enters new knowledge information into an input form. The device sends this information in JSON format to the server. The server preprocesses the received data and analyzes it using a generative AI model. The analysis results are stored in a database. The time of entry and information about the person who entered the information are also recorded.

[1015] Input: New knowledge information

[1016] Output: Analysis results, save date and time, and inputter information

[1017] Step 3:

[1018] A user inputs a query to search existing knowledge information. The device sends this query in JSON format to the server. The server uses a natural language processing (NLP) engine to analyze the query and extract the search intent. The server then sends the search query to a database, retrieves related knowledge data, and returns it to the device.

[1019] Input: Search query

[1020] Output: Search results (related knowledge data)

[1021] Step 4:

[1022] The user operates the robot based on the acquired knowledge information and inputs the results and opinions as feedback. The device sends the feedback information in JSON format to the server. The server analyzes the feedback and saves it as retraining data for the generative AI model. This improves the accuracy of the generative AI model.

[1023] Input: Feedback information

[1024] Output: Feedback analysis results and retraining data

[1025] Step 5:

[1026] The server periodically retrains the generative AI model using the collected feedback data, improving the model's accuracy for the next know-how input, search, or feedback.

[1027] Input: Feedback data

[1028] Output: Retrained generative artificial intelligence model

[1029] These processing steps enable effective management of knowledge input, search, sharing, and feedback on robot operations within the factory, and provide real-time feedback to optimize operations.

[1030] 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.

[1031] This invention relates to a system that efficiently accumulates know-how and schemes specific to a company, makes it easy to share and search them, and also recognizes user emotions and uses the results to provide feedback and information. Specific embodiments of this system are described below.

[1032] ---

[1033] User Authentication Process

[1034] When a user logs in to a terminal, they enter their company ID and password, which confirms their authority to access the system.

[1035] 1. The device collects the user's ID and password and sends them to the server using the HTTPS protocol.

[1036] 2. The server compares the received authentication information with the database and authenticates the user. If authentication is successful, it generates a session ID and returns it to the terminal.

[1037] 3. The terminal receives the session ID and displays a successful login message to the user.

[1038] ---

[1039] Know-how input process

[1040] This is the process by which users input new know-how and schemes into the system, thereby accumulating in-house knowledge.

[1041] 1. The user enters new know-how or scheme (e.g., "We recommend development using Scrum for Project B") into a dedicated form on the terminal.

[1042] 2. The device collects input data and sends it to the server in JSON format.

[1043] 3. The server preprocesses the data and normalizes the text (matching case, removing unnecessary whitespace, etc.).

[1044] 4. The server inputs the preprocessed data into a generative artificial intelligence model and analyzes the know-how data.

[1045] 5. The server obtains the analysis results and extracts link information with related existing know-how and new know-how.

[1046] 6. The server saves the generated know-how data in the database, along with the date and time of saving and information about the person who entered the data.

[1047] ---

[1048] Know-how search process

[1049] It is a process in which users search for existing know-how and schemes.

[1050] 1. A user enters a query (e.g., "Agile development best practices") into a search form on their device.

[1051] 2. The device sends the entered query in JSON format to the server.

[1052] 3. The server analyzes the received query using natural language processing and extracts the intent.

[1053] 4. The server sends a search query to the database to retrieve relevant know-how data. The search results also include a relevance score.

[1054] 5. The server returns the search results to the device in JSON format.

[1055] 6. The device receives the response from the server and displays the search results to the user, providing an interface that makes it easy for the user to find the information they need.

[1056] ---

[1057] User emotion recognition process

[1058] A distinctive feature of this invention is the use of an emotion engine that recognizes the user's emotions.

[1059] 1. When a user enters know-how or browses search results on the device, the emotion engine collects the user's facial expressions and voice data.

[1060] 2. The device analyzes the data collected in real time and estimates the user's emotional state (e.g., satisfaction, dissatisfaction, confusion, etc.).

[1061] 3. The device sends the estimated emotion data to the server.

[1062] 4. The server collects and stores this emotional data as feedback data, and also uses it for know-how feedback and retraining processes.

[1063] Feedback Providing Process

[1064] This is a process in which users provide feedback on the know-how they have acquired. It is characterized by the collection of feedback using an emotion engine.

[1065] 1. When a user inputs feedback from a terminal after applying know-how, the emotion engine automatically recognizes the user's emotional state.

[1066] 2. The device sends the feedback content and emotion data in JSON format to the server.

[1067] 3. The server receives the feedback and emotion data, analyzes it, and saves it as data for retraining the model.

[1068] 4. The server runs a retraining process at regular intervals to improve the performance and accuracy of the model.

[1069] 5. The server applies the improved model to the system and reflects it in the next user search or know-how input.

[1070] ---

[1071] Specific examples

[1072] Example of user authentication

[1073] 1. User "Tanaka" enters ID "tanaka" and password "password123".

[1074] 2. The device sends this to the server.

[1075] 3. The server performs authentication, is successful, and generates a session ID "abc123".

[1076] 4. The terminal saves the session ID and displays "Login successful" to the user.

[1077] Specific examples of know-how input

[1078] 1. User "Tanaka" inputs "Scrum development is recommended for Project B."

[1079] 2. The device sends the data to the server.

[1080] 3. The server inputs the data into a generative artificial intelligence model and stores the analysis results in a database.

[1081] Examples of emotion engines

[1082] 1. While the user “Tanaka” is inputting know-how, the emotion engine collects and analyzes the user’s facial expression data.

[1083] 2. The device sends the analysis results to the server and stores them as feedback data.

[1084] 3. The server uses this emotion data to retrain the AI ​​model.

[1085] This allows users to efficiently input, search, and provide feedback on internal know-how. The addition of an emotion engine also enables advanced feedback collection and customized information provision based on emotions.

[1086] The processing flow will be explained below.

[1087] User Authentication Process

[1088] Step 1:

[1089] The terminal prompts the user to enter their company ID and password. For example, the user enters "tanaka" and "password123."

[1090] Step 2:

[1091] The terminal sends the entered authentication information (ID and password) to the server using the HTTPS protocol.

[1092] Step 3:

[1093] The server queries the database for the received authentication information and searches for a match.

[1094] Step 4:

[1095] The server retrieves the search results from the database and, if authentication is successful, generates a new session ID (e.g., "sessionID_abc123").

[1096] Step 5:

[1097] The server returns the authentication result and session ID to the terminal as an HTTPS response.

[1098] Step 6:

[1099] The terminal receives the response from the server, notifies the user that login was successful, and saves the session ID.

[1100] ---

[1101] Know-how input process

[1102] Step 1:

[1103] The user inputs new know-how or schemes into a dedicated form on the terminal. For example, the user inputs "Scrum-based development is recommended for Project B."

[1104] Step 2:

[1105] The terminal sends the entered data to the server in JSON format.

[1106] Step 3:

[1107] The server preprocesses the data received and normalizes the text (matching case, removing unnecessary whitespace, etc.).

[1108] Step 4:

[1109] The server inputs the preprocessed data into a generative artificial intelligence model and analyzes the know-how data.

[1110] Step 5:

[1111] The server obtains the analysis results and extracts link information with related existing know-how and new know-how.

[1112] Step 6:

[1113] The server saves the generated know-how data in a database, along with the date and time of saving and information about the person who entered the data.

[1114] ---

[1115] Know-how search process

[1116] Step 1:

[1117] A user enters a query into a search form on their device. For example, the user enters "Agile development best practices."

[1118] Step 2:

[1119] The device sends the entered query in JSON format to the server.

[1120] Step 3:

[1121] The server analyzes the received query using natural language processing to extract the intent.

[1122] Step 4:

[1123] The server sends a search query to the database to retrieve relevant know-how data, and the search results also include a relevance score.

[1124] Step 5:

[1125] The server returns the search results to the device in JSON format.

[1126] Step 6:

[1127] The device receives the response from the server and displays the search results to the user, providing an interface that makes it easy for users to find the information they need.

[1128] ---

[1129] User emotion recognition process

[1130] Step 1:

[1131] When a user inputs know-how or browses search results on a device, the emotion engine uses the device's camera and microphone to collect the user's facial expressions and voice data.

[1132] Step 2:

[1133] The device analyzes the facial expression and voice data collected in real time, and the emotion engine estimates the user's emotional state (e.g., satisfaction, dissatisfaction, confusion, etc.).

[1134] Step 3:

[1135] The device sends the estimated emotion data in JSON format to the server.

[1136] Step 4:

[1137] The server collects and stores the received emotional data as feedback data, which is also used for know-how feedback and the retraining process of AI models.

[1138] ---

[1139] Feedback Providing Process

[1140] Step 1:

[1141] It provides feedback on the know-how acquired by the user. For example, the user may input, "This scheme is easy to understand, but I would like to see a concrete example task."

[1142] Step 2:

[1143] The device collects the feedback and sends it to the server in JSON format.

[1144] Step 3:

[1145] The server receives the feedback and performs text analysis to extract the intent and key points of the feedback.

[1146] Step 4:

[1147] The server stores the analyzed feedback data as retraining data for the generative artificial intelligence model.

[1148] Step 5:

[1149] The server runs a retraining process at regular intervals to improve the performance and accuracy of the AI ​​model.

[1150] Step 6:

[1151] The server applies the improved AI model to the system and reflects it in the next user search or know-how input.

[1152] ---

[1153] Specific examples

[1154] Example of user authentication

[1155] Step 1:

[1156] User "Tanaka" enters ID "tanaka" and password "password123".

[1157] Step 2:

[1158] The terminal sends this to the server.

[1159] Step 3:

[1160] The server performs authentication, is successful, and generates a session ID "abc123".

[1161] Step 4:

[1162] The terminal stores the session ID and displays "Login successful" to the user.

[1163] Specific examples of know-how input

[1164] Step 1:

[1165] User "Tanaka" inputs "Scrum development is recommended for Project B."

[1166] Step 2:

[1167] The device sends the data to the server.

[1168] Step 3:

[1169] The server inputs the data into a generative artificial intelligence model and stores the analysis results in a database.

[1170] Examples of emotion engines

[1171] Step 1:

[1172] While the user "Tanaka" is inputting know-how, the emotion engine collects and analyzes facial expression data using the device's camera and microphone.

[1173] Step 2:

[1174] The device sends the analysis results to the server and stores them as feedback data.

[1175] Step 3:

[1176] The server uses this emotion data to retrain the AI ​​model.

[1177] This allows users to efficiently input, search, and provide feedback on internal know-how. The addition of an emotion engine also enables advanced feedback collection and customized information provision based on emotions.

[1178] Example 2

[1179] 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."

[1180] Existing knowledge management systems make it difficult to effectively share and search know-how and schemes accumulated within an organization. They also lack a feedback mechanism that takes user sentiment into account, limiting the means to improve system usability and knowledge quality. This poses a challenge: it is difficult to effectively collect useful knowledge held by users and share and utilize it throughout the organization.

[1181] 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.

[1182] In this invention, the server includes means for acquiring authentication information, means for authenticating users, means for inputting organization-specific knowledge and schemes, means for inputting input data into a generative AI model for analysis, means for saving the analysis results in a database, means for searching for knowledge information from the database, means for providing search results to users, means for recognizing the emotional state of users and collecting data on the same, and means for collecting the emotional data as feedback and using it to retrain the model, thereby enabling users to effectively input and search for knowledge and providing feedback based on their emotional state.

[1183] "Authentication information" refers to information such as ID and password required when a user logs in to a system.

[1184] "User authentication" is the process of verifying a user's access rights based on the authentication information entered.

[1185] "Organization-specific knowledge and schemes" refers to information assets such as experience, know-how, business processes, and best practices within the organization.

[1186] A "generative AI model" is a generative artificial intelligence model that analyzes input data and generates new insights and related information.

[1187] "Analysis results" are the information and insights obtained after input data is analyzed by a generative AI model.

[1188] A "database" is a system for systematically storing and managing analysis results and other information.

[1189] A "search query" is a keyword or phrase that a user enters to search for specific information.

[1190] "Emotional state" refers to emotions estimated from the user's facial expressions and voice data, and includes satisfaction, dissatisfaction, confusion, etc.

[1191] "Feedback" refers to the ratings and comments users provide after using the system.

[1192] "Retraining methods" refers to the process of retraining an AI model using collected feedback and sentiment data to improve the model's performance.

[1193] This invention is a system for effectively inputting, searching, and sharing knowledge and schemes accumulated within an organization, and it can also recognize users' emotions and reflect their feedback. Specific embodiments of this system are described below.

[1194] This system mainly consists of three elements: a server, a terminal, and a user.

[1195] User Authentication Process

[1196] When a user logs in to a terminal, they enter their company ID and password to confirm their authorization to access the system. This authentication process is carried out securely using the HTTPS protocol. The server checks the authentication information against a database, and if it matches, it generates a session ID and returns it to the terminal.

[1197] Know-how input process

[1198] When a user inputs new knowledge or schemes, they use a dedicated form on their device. This data is sent in JSON format to the server, which then inputs it into a generative AI model for analysis. The analysis results are stored in a database and shared across the organization. For example, if a user inputs "We recommend developing using Scrum for Project B," the data is preprocessed on the server and analyzed by a generative AI model (e.g., OpenAI GPT-3).

[1199] Know-how search process

[1200] When a user searches for existing knowledge or schemes, they enter a query into a search form on their device. The query is sent in JSON format to the server, which then analyzes the query's intent using natural language processing. Based on the analysis results, the server retrieves relevant information from the database and provides it to the user. For example, if a user searches for "best practices in agile development," the server provides the user with relevant information based on this query.

[1201] User emotion recognition process

[1202] When a user enters knowledge or browses search results, the device's built-in camera and microphone collect facial and voice data in real time, which is then input into the emotion analysis algorithm. The server receives this emotion data, collects it as feedback, and uses it to retrain the AI ​​model. For example, if a user is confused while entering know-how, the emotion engine will identify that situation and send it to the server.

[1203] Feedback Providing Process

[1204] When users provide feedback on the information obtained by the system, their emotional data is collected along with the feedback. The server analyzes the feedback and emotional data and stores it as retraining data to improve the accuracy and performance of the AI ​​model.

[1205] This system allows users to effectively input and search knowledge within an organization and reflects feedback based on emotional data. For example, by inputting important know-how about a project and reusing that know-how in other projects, the efficiency of the entire organization can be improved. Emotion recognition is also expected to improve usability.

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

[1207] Step 1:

[1208] The user enters their company ID and password into the terminal.

[1209] (Input): User ID, password

[1210] (Operation): The user enters their company ID and password on the device's login screen.

[1211] (Output): The device collects the ID and password.

[1212] Step 2:

[1213] The device sends the collected authentication information to the server using the HTTPS protocol.

[1214] (Input): User ID, password

[1215] (Operation): The device converts the ID and password into JSON format and sends it as an HTTPS request to the server's authentication API endpoint.

[1216] (Output): The server receives the authentication information.

[1217] Step 3:

[1218] The server compares the received authentication information with a database and authenticates the user.

[1219] (Input): Authentication information (user ID, password)

[1220] (What happens): The server compares the user information in its database with the entered authentication information.

[1221] (Output): Authentication result (success / failure)

[1222] Step 4:

[1223] When authentication is successful, the server generates a session ID and returns it to the terminal.

[1224] (Input): Authentication information (if successful)

[1225] (Operation): The server generates a unique session ID and sends it to the device in JSON format.

[1226] (Output): Session ID

[1227] Step 5:

[1228] The terminal receives the session ID and displays a "Login successful" message to the user.

[1229] (Input): Session ID

[1230] (Action): The device saves the session ID in local storage and displays a login success message.

[1231] (Output): User interface update (login success message)

[1232] Step 6:

[1233] The user inputs new knowledge or scheme into an input form.

[1234] (Input): Know-how and scheme (text data)

[1235] (Operation): The user inputs new knowledge or scheme into a dedicated form on the terminal.

[1236] (Output): The terminal collects input data.

[1237] Step 7:

[1238] The input data collected by the terminal is sent to the server in JSON format.

[1239] (Input): Know-how and scheme (JSON format)

[1240] (Operation): The terminal converts the input data into JSON format and sends it to the server as an HTTPS request.

[1241] (Output): The server receives the data.

[1242] Step 8:

[1243] The server preprocesses the data and inputs it into a generative AI model for analysis.

[1244] (Input): Know-how and scheme (JSON format)

[1245] (Operation): The server normalizes the data (standardizing uppercase and lowercase letters, removing unnecessary spaces) and inputs it into a generative AI model. The generative AI model analyzes the data and generates new knowledge and related information.

[1246] (Output): Analysis results (new knowledge, related information)

[1247] Step 9:

[1248] The server stores the analysis results in a database.

[1249] (Input): Analysis results (new knowledge, related information)

[1250] (Operation): The server adds the analysis results to the database as a new record, and also records the save date and time and the ID of the person who entered the data.

[1251] (Output): Update the database (save new knowledge)

[1252] Step 10:

[1253] A user enters a query into a search form.

[1254] (Input): Search query (text data)

[1255] Action: A user enters a query into a search form on their device.

[1256] (Output): The terminal collects input data.

[1257] Step 11:

[1258] The device sends the entered query in JSON format to the server.

[1259] (Input): Search query (JSON format)

[1260] (Operation): The terminal converts the input data into JSON format and sends it to the server as an HTTPS request.

[1261] (Output): The server receives the data.

[1262] Step 12:

[1263] The server parses the query and performs a search against the database.

[1264] (Input): Search query (JSON format)

[1265] (Operation): The server analyzes the query using a natural language processing algorithm and sends the search query to the database.

[1266] (Output): Search results (related knowledge information)

[1267] Step 13:

[1268] The server returns the search results to the device in JSON format.

[1269] (Input): Search results (JSON format)

[1270] (Operation): The server encodes the search results into JSON format and sends them to the device as an HTTPS request.

[1271] (Output): The device receives the data.

[1272] Step 14:

[1273] Your device will display the search results.

[1274] (Input): Search results (JSON format)

[1275] (Operation): The terminal displays the search results received on the user interface, making it easy for the user to find the information they need.

[1276] (Output): Update the user interface (display search results)

[1277] Step 15:

[1278] While the user is entering know-how or viewing search results, the device's built-in camera and microphone collect facial expressions and voice data.

[1279] (Input): facial expression data, voice data

[1280] (Operation): The device uses its built-in camera and microphone to collect facial and voice data in real time.

[1281] (Output): Collected emotion data

[1282] Step 16:

[1283] The device analyzes the emotional data collected in real time to estimate the user's emotional state.

[1284] (Input): Collected emotion data

[1285] (Action): The device uses an emotion analysis algorithm to estimate the emotional state (e.g., happy, unhappy, confused).

[1286] (Output): Estimated emotional state

[1287] Step 17:

[1288] The device transmits the estimated emotion data to the server.

[1289] (Input): Estimated emotional state

[1290] (Operation): The device converts the emotion data into JSON format and sends it to the server as an HTTPS request.

[1291] (Output): The server receives the emotion data.

[1292] Step 18:

[1293] The server collects the emotion data as feedback data and uses it to retrain the model.

[1294] (Input): Emotion data (JSON format)

[1295] (How it works): The server stores the emotion data in a feedback database and uses it to retrain the AI ​​model.

[1296] (Output): Update the database and save the retraining data.

[1297] Step 19:

[1298] The server runs a retraining process at regular intervals to improve the model's performance.

[1299] (Input): Retraining data

[1300] (Operation): The server periodically runs a retraining process to improve the accuracy and performance of the AI ​​model.

[1301] (Output): Improved AI model

[1302] Step 20:

[1303] The server applies the improved model to the system and reflects it in the next user search or know-how input.

[1304] (Input): Improved AI model

[1305] (Operation): The server deploys the updated AI model to the system and applies it the next time the user searches or types something.

[1306] (Output): System update (application of improved AI model)

[1307] The above are the specific processing steps of the program of this system.

[1308] (Application example 2)

[1309] 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."

[1310] Improving work efficiency and sharing knowledge within factories are important issues, but conventional systems have had difficulty providing real-time support or feedback based on workers' emotions. Furthermore, collecting feedback that reflects emotional states and how to utilize it has also been an unsolved problem.

[1311] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for identifying the emotional state of a user using an emotion recognition engine and collecting and saving that data, means for collecting feedback and providing information using the emotion data, and means for tagging knowledge data analyzed using a generative artificial intelligence model and making it searchable based on relevance. This not only improves work efficiency in the factory but also makes it possible to provide appropriate feedback based on the emotions of workers.

[1312] "Authentication information" refers to information used to verify a user's identity, and is usually comprised of an ID and password.

[1313] "User authentication" refers to the process of verifying whether a user is a person with legitimate authority using the acquired authentication information.

[1314] "Internal knowledge" refers to specialized information and procedures accumulated within an organization.

[1315] A "generative artificial intelligence model" is an artificial intelligence model that has the ability to generate and analyze knowledge and information based on new data.

[1316] A "database" is a system for efficiently storing and managing data.

[1317] "Knowledge information" refers to internal knowledge and procedures stored in a database.

[1318] "Search results" refers to the related knowledge information obtained when searching a database.

[1319] "Feedback data" refers to data collected from users' evaluations and opinions.

[1320] "Model retraining" is the process of using new data to improve the performance of an artificial intelligence model.

[1321] An "emotion recognition engine" is a technology that analyzes a user's facial expressions and voice to identify their emotional state.

[1322] "Emotional state" is information indicating the type and intensity of the user's emotion.

[1323] "Feedback collection" is the process of collecting opinions and sentiment data from users.

[1324] "Information provision" is the process of displaying search results and feedback to users.

[1325] "Tagging" means assigning labels to data that indicate its relevance.

[1326] This invention uses smart glasses and head-mounted displays (HMDs) to build a system that improves work efficiency in factories. The system includes multiple means for user authentication, know-how input, know-how search, emotion recognition, and feedback collection.

[1327] User Authentication

[1328] The user puts on the smart glasses and enters their ID and password. This authentication information is securely sent to the server using the HTTPS protocol. The server compares the received authentication information with a database, and if authentication is successful, it generates a session ID and returns it to the device. The device saves the session ID and displays a message to the user indicating successful login.

[1329] Input of know-how

[1330] When a user wants to input new know-how or procedures, they use a dedicated form on the smart glasses. The device sends the input data in JSON format to the server, which then preprocesses the data and analyzes it using a generative artificial intelligence model. The analysis results are stored in a database. This process efficiently accumulates knowledge and procedures acquired within the factory.

[1331] Search for know-how

[1332] When a user searches for a specific know-how or procedure, they enter a query into the search form on the smart glasses. The device sends the query in JSON format to the server, which uses natural language processing technology to analyze the query's intent. The server then sends the search query to a database to retrieve relevant know-how information. The search results are returned to the device and displayed visually to the user.

[1333] emotion recognition

[1334] When a user enters know-how or browses search results, the emotion engine collects the user's facial expression and voice data. The device analyzes this data in real time and estimates the user's emotional state. The estimated emotion data is sent to the server and stored as feedback data. This emotion data is used to retrain the model and provide feedback on know-how.

[1335] Feedback collection

[1336] When a user provides feedback, the emotion engine automatically recognizes the user's emotional state. The feedback content and emotion data are sent from the device to the server, where they are analyzed and stored as retraining data for the model. A periodic retraining process is performed to improve the model's performance and accuracy.

[1337] Specific examples

[1338] For example, if a user wears smart glasses and types "Scrum development is recommended for Project B," the device sends the data to the server, where it is analyzed by a generative artificial intelligence model. The analysis results are stored in a database and will be displayed as relevant information when other users later search for "best practices for agile development."

[1339] Prompt Sentence Examples

[1340] Analyze the following know-how data for your generative AI model:

[1341] "Scrum development is recommended for Project B."

[1342] This invention will enable efficient accumulation and sharing of knowledge within factories, and will also enable feedback collection and information provision based on user emotions, which is expected to improve worker satisfaction and work efficiency.

[1343] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1344] Step 1:

[1345] The user puts on the smart glasses and enters their ID and password.

[1346] Input: User ID, Password

[1347] Specific operation: The user enters their ID and password into the login form displayed on the smart glasses display. The smart glasses then send this authentication information to the server using the HTTPS protocol.

[1348] Output: Authentication information sent to the server

[1349] Step 2:

[1350] The server receives the authentication information and authenticates the user by checking it against a database.

[1351] Input: Authentication information (ID, password)

[1352] How it works: The server checks the received authentication information against the user information in its database, and if authentication is successful, generates a session ID, which is then sent back to the smart glasses in a secure manner.

[1353] Output: Session ID

[1354] Step 3:

[1355] The terminal stores the session ID and displays a successful login message to the user.

[1356] Input: Session ID

[1357] Specific operation: The smart glasses store the session ID in memory and notify the user that the login was successful by displaying it on the display.

[1358] Output: Login successful message

[1359] Step 4:

[1360] The user enters new know-how and procedures into a special form on the smart glasses.

[1361] Input: Know-how and procedures

[1362] Specific operation: The user inputs new know-how or procedures (e.g., "We recommend developing using Scrum for Project B") into a specific form through the smart glasses' display. The input data is temporarily stored in the smart glasses' internal storage.

[1363] Output: Input data is temporarily saved

[1364] Step 5:

[1365] The terminal sends the entered data to the server in JSON format.

[1366] Input: Input data

[1367] Specific operation: The smart glasses convert the temporarily stored know-how and procedure data into JSON format and send it to the server.

[1368] Output: JSON format data is sent to the server

[1369] Step 6:

[1370] The server preprocesses the data and performs analysis using a generative artificial intelligence model.

[1371] Input: Input data in JSON format

[1372] How it works: The server preprocesses the received JSON data (e.g., standardizing case, removing unnecessary whitespace, etc.) and then feeds it into a generative artificial intelligence model. The model then analyzes the data and extracts new information and related existing knowledge.

[1373] Output: Analysis results

[1374] Step 7:

[1375] The server stores the analysis results in a database.

[1376] Input: Analysis results

[1377] Specific operation: The server saves the analysis results in a database, and also records the date and time of saving and information about the person who entered the data.

[1378] Output: Analysis results stored in a database

[1379] Step 8:

[1380] A user searches for know-how by entering a query into a search form on the smart glasses.

[1381] Input: Search query

[1382] What it does: A user enters a specific query (e.g., "Agile development best practices") into the search form on the smart glasses. The smart glasses convert the query into JSON format and send it to the server.

[1383] Output: The search query is sent to the server

[1384] Step 9:

[1385] The server analyzes the received query and searches the database for related know-how information.

[1386] Input: Search query

[1387] Specific operation: The server analyzes the received search query using natural language processing technology, sends an appropriate search query to the database, and retrieves related know-how information.

[1388] Output: Search results

[1389] Step 10:

[1390] The server sends the search results back to the device, which then displays them on the smart glasses.

[1391] Input: Search results

[1392] Specific operation: The server converts the search results into JSON format and sends them to the smart glasses, which then display them visually for the user to view.

[1393] Output: Search results displayed to the user

[1394] Step 11:

[1395] The emotion recognition engine collects the user's facial expression and voice data and analyzes their emotional state in real time.

[1396] Input: facial expression data, voice data

[1397] How it works: The smart glasses' built-in cameras and microphones collect the user's facial expressions and voice in real time. The emotion recognition engine analyzes this data and estimates the user's emotional state (e.g., satisfaction, dissatisfaction, confusion, etc.).

[1398] Output: Estimated emotional state

[1399] Step 12:

[1400] The device transmits the estimated emotion data to the server and stores it as feedback data.

[1401] Input: Emotion data

[1402] Specific operation: The emotion recognition engine converts the estimated emotional state into JSON format and sends it to the server. The server then stores the received emotional data in a database as feedback data.

[1403] Output: Emotion data stored in a database

[1404] Step 13:

[1405] The server uses the feedback data to retrain the model.

[1406] Input: Feedback data

[1407] Specific operation: The server periodically collects feedback data and retrains the generative AI model, thereby improving the model's performance and accuracy.

[1408] Output: Retrained generative artificial intelligence model

[1409] Through the above processing steps, the present invention improves work efficiency in a factory and provides appropriate feedback based on the user's emotions.

[1410] 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.

[1411] 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.

[1412] 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.

[1413] [Third embodiment]

[1414] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1415] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1416] 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).

[1417] 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.

[1418] 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.

[1419] 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).

[1420] 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.

[1421] 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.

[1422] 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.

[1423] 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.

[1424] 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.

[1425] 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."

[1426] The present invention relates to a system for efficiently storing know-how and schemes specific to a company, and facilitating sharing and searching thereof. A specific embodiment of this system will be described below.

[1427] ---

[1428] User Authentication Process

[1429] When a user logs in to a terminal, they enter their company ID and password, which confirms their authority to access the system.

[1430] 1. The device collects the user's ID and password and sends them to the server using the HTTPS protocol.

[1431] 2. The server compares the received authentication information with the database and authenticates the user. If authentication is successful, it generates a session ID and returns it to the terminal.

[1432] 3. The terminal receives the session ID and displays a successful login message to the user.

[1433] Know-how input process

[1434] This is the process by which users input new know-how and schemes into the system, thereby accumulating in-house knowledge.

[1435] 1. The user enters new know-how or scheme (e.g., "Agile development is being adopted for Project A") into a dedicated form on the device.

[1436] 2. The device collects input data and sends it to the server in JSON format.

[1437] 3. The server preprocesses the data, normalizing and structuring the text.

[1438] 4. The server inputs the preprocessed data into a generative AI model for analysis. The analysis results include related existing data and new insights.

[1439] 5. The server saves the analysis results in a database. When saving, the date and time of saving and information about the person who entered the data are also recorded.

[1440] Know-how search process

[1441] It is a process in which users search for existing know-how and schemes.

[1442] 1. A user enters a query (e.g., "Agile development best practices") into a search form on their device.

[1443] 2. The device sends a query in JSON format to the server.

[1444] 3. The server analyzes the received query using natural language processing to extract the intent.

[1445] 4. The server sends a search query to the database to retrieve relevant know-how data.

[1446] 5. The server returns the search results to the device in JSON format.

[1447] 6. The device displays the search results to the user.

[1448] Feedback Providing Process

[1449] This is the process of providing feedback to users on the know-how they have acquired.

[1450] 1. The user inputs feedback from the terminal after applying the know-how (e.g., "This scheme is useful, but specifically, task management is lacking").

[1451] 2. The device sends the feedback content in JSON format to the server.

[1452] 3. The server receives the feedback, analyzes it, and saves it as data for retraining the model.

[1453] 4. The server periodically retrains the generative AI model to improve its accuracy.

[1454] 5. The server applies the improved model to the system and reflects it in the next user search or know-how input.

[1455] ---

[1456] Specific examples

[1457] Example of user authentication

[1458] 1. User "Tanaka" enters ID "tanaka" and password "password123".

[1459] 2. The device sends this to the server.

[1460] 3. The server performs authentication, is successful, and generates a session ID "abc123".

[1461] 4. The terminal saves the session ID and displays "Login successful" to the user.

[1462] Specific examples of know-how input

[1463] 1. User "Tanaka" inputs "Scrum development is recommended for Project B."

[1464] 2. The device sends the data to the server.

[1465] 3. The server inputs the data into a generative artificial intelligence model and stores the analysis results in a database.

[1466] This will create a system that allows users to efficiently input, search, and share in-house know-how.

[1467] The processing flow will be explained below.

[1468] User Authentication Process

[1469] Step 1:

[1470] The terminal prompts the user to enter their company ID and password. For example, the user enters "tanaka" and "password123."

[1471] Step 2:

[1472] The terminal sends the entered authentication information to the server using the HTTPS protocol.

[1473] Step 3:

[1474] The server queries the database for the received authentication information and searches for a match.

[1475] Step 4:

[1476] The server retrieves the search results from the database and, if authentication is successful, generates a new session ID (e.g., "sessionID_abc123").

[1477] Step 5:

[1478] The server returns the authentication result and session ID to the terminal as an HTTPS response.

[1479] Step 6:

[1480] The terminal receives the response from the server, notifies the user that login was successful, and saves the session ID.

[1481] ---

[1482] Know-how input process

[1483] Step 1:

[1484] The user inputs new know-how or schemes into a dedicated form on the terminal. For example, the user inputs "Scrum-based development is recommended for Project B."

[1485] Step 2:

[1486] The device collects the input data and sends it to the server in JSON format.

[1487] Step 3:

[1488] The server preprocesses the data received and normalizes the text (matching case, removing unnecessary whitespace, etc.).

[1489] Step 4:

[1490] The server inputs the preprocessed data into a generative artificial intelligence model and analyzes the know-how data.

[1491] Step 5:

[1492] The server obtains the analysis results and extracts link information with related existing know-how and new know-how.

[1493] Step 6:

[1494] The server saves the generated know-how data in a database, along with the date and time of saving and information about the person who entered the data.

[1495] ---

[1496] Know-how search process

[1497] Step 1:

[1498] A user enters a query into a search form on their device. For example, the user enters "Agile development best practices."

[1499] Step 2:

[1500] The device sends the entered query in JSON format to the server.

[1501] Step 3:

[1502] The server analyzes the received query using natural language processing to understand the intent of the query and extract related keywords.

[1503] Step 4:

[1504] The server sends a search query to the database to retrieve relevant know-how data, and the search results also include a relevance score.

[1505] Step 5:

[1506] The server returns the search results to the device in JSON format.

[1507] Step 6:

[1508] The device receives the response from the server and displays the search results to the user, providing an interface that makes it easy for users to find the information they need.

[1509] ---

[1510] Feedback Providing Process

[1511] Step 1:

[1512] It provides feedback on the know-how acquired by the user. For example, the user may input, "This scheme is easy to understand, but I would like to see a concrete example task."

[1513] Step 2:

[1514] The device collects the feedback and sends it to the server in JSON format.

[1515] Step 3:

[1516] The server receives the feedback and performs text analysis to extract the intent and key points of the feedback.

[1517] Step 4:

[1518] The server stores the analyzed feedback data as retraining data for the generative artificial intelligence model.

[1519] Step 5:

[1520] The server runs a retraining process at regular intervals to improve the performance and accuracy of the model.

[1521] Step 6:

[1522] The server applies the improved model to the system and reflects it in the next user search or know-how input.

[1523] Through these steps, users can efficiently input, search, and provide feedback on their internal know-how.

[1524] Example 1

[1525] 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."

[1526] Previous methods for sharing in-house know-how and schemes mainly involved manual recording and management, which lacked efficiency. It was also difficult to search for know-how or discover new insights, resulting in low utilization of information. Furthermore, there was no established method for collecting feedback and using it to update and improve knowledge. For these reasons, a new system for effectively storing and utilizing in-house knowledge resources was needed.

[1527] 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.

[1528] In this invention, the server includes means for acquiring authentication information, means for authenticating a user based on the authentication information, means for inputting in-house know-how and schemes, means for transmitting the input data to the server in JSON format, means for inputting the input data to a generative artificial intelligence model for analysis, means for saving the analysis results in a database, means for searching the database for know-how information using natural language processing, means for providing the search results to the user in JSON format, and means for collecting feedback data and using it to retrain the model. This enables efficient accumulation, sharing, and search of in-house know-how, and further enables the model to be improved based on feedback to increase the quality of knowledge.

[1529] "Authentication Information" means information required to verify a user's identity, and primarily includes a user ID and password.

[1530] "User authentication" is the process of verifying that a user is a legitimate user using authentication information.

[1531] "Know-how" refers to information that refers to knowledge, techniques, and methodologies related to a specific task or project.

[1532] "Scheme" means the plan, structure or procedure for a particular project or business process.

[1533] "JSON format" is a lightweight data exchange format for structuring data, and is an abbreviation for JavaScript Object Notation.

[1534] A "generative artificial intelligence model" is a set of machine learning algorithms that generate new information and insights based on input data.

[1535] "Analysis" is the process performed to understand input data and find patterns and relationships.

[1536] A "database" is a system for organizing, efficiently storing, and retrieving information.

[1537] "Natural language processing" is a technical field that uses computers to understand and process human language.

[1538] "Feedback data" refers to information such as ratings, comments, and improvements provided by users regarding the system's output.

[1539] "Model retraining" is the process of updating a generative artificial intelligence model with new data collected to improve its accuracy and performance.

[1540] This invention relates to a system that efficiently accumulates in-house know-how and schemes and makes it easy to share and search them. This system includes a series of processes for collecting, transmitting, analyzing, and storing data between a server, terminals, and users.

[1541] User Authentication Process

[1542] When a user enters their company ID and password into the terminal, the terminal encrypts the authentication information and sends it to the server using the HTTPS protocol. The server compares the authentication information with a database, and if authentication is successful, generates a session ID and returns it to the terminal. This allows the user to access the system securely.

[1543] Know-how input process

[1544] The user inputs new know-how or schemes into a dedicated form on the device. For example, this information might be something like "Scrum development is recommended for Project B." The device then sends the input data in JSON format to the server. The server preprocesses the data and analyzes it using a generative artificial intelligence model. The analysis results are saved in a database, along with the date and time of saving and information about the person who entered the data.

[1545] Know-how search process

[1546] When a user enters a query (e.g., "Agile development best practices") into the device's search form, the device sends the query in JSON format to the server. The server analyzes the query using natural language processing technology and sends the search query to the database. The search results are returned to the device in JSON format and displayed to the user.

[1547] Feedback Providing Process

[1548] After applying the know-how, the user inputs feedback from the device. For example, the user might say, "This scheme is useful, but specifically, task management is lacking." The device then sends the feedback in JSON format to the server. The server analyzes the feedback and saves it as retraining data for the generative AI model. This improves the model's accuracy, which is reflected the next time the user searches or inputs know-how.

[1549] Hardware and software used

[1550] This system uses the following hardware and software:

[1551] Device: A computer or smartphone operated by a user.

[1552] Server: A computer system that processes and stores data.

[1553] Database: A system for storing know-how, schemes, analysis results, and feedback data.

[1554] HTTPS protocol: A communications protocol for secure data transmission.

[1555] Generative artificial intelligence models: Algorithms that analyze input data and generate new insights.

[1556] Specific examples

[1557] Example of user authentication

[1558] The user enters the ID "tanaka" and password "password123" into the terminal, which then encrypts and sends it to the server. The server compares it with the database, and if authentication is successful, it generates a session ID "abc123" and returns it to the terminal. The terminal saves the session ID and displays "Login successful" to the user.

[1559] Specific examples of know-how input

[1560] The user inputs "Scrum development is recommended for Project B," and the device sends the data in JSON format to the server. The server preprocesses the data, inputs it into a generative artificial intelligence model for analysis, and stores the analysis results in a database.

[1561] Specific examples of know-how search

[1562] A user types "Agile development best practices" into a search form, and the device sends the query to the server, which parses the query, searches the database, and returns relevant results to the device, where the user is presented with the search results.

[1563] Example prompt sentence:

[1564] "Project B is recommended to be developed using Scrum. What are some best practices that can be used for other projects?"

[1565] This system enables efficient accumulation, sharing, and searching of internal know-how, allowing for improvements to generative AI models based on feedback.

[1566] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1567] Program processing flow

[1568] User Authentication Process

[1569] Step 1:

[1570] The user enters their company ID and password into the device's login screen.

[1571] Input: Company ID, password

[1572] Output: Credentials sent to the device

[1573] Step 2:

[1574] The terminal encrypts the entered ID and password using the HTTPS protocol and sends it to the server.

[1575] Input: Company ID, password

[1576] Output: Authentication information encrypted using the HTTPS protocol

[1577] Step 3:

[1578] The server checks the received authentication information against a database and authenticates the user.

[1579] Input: Credentials

[1580] Output: Authentication success or failure, session ID

[1581] Step 4:

[1582] If the authentication is successful, the server generates a session ID and returns it to the terminal.

[1583] Input: Authentication success result

[1584] Output: Session ID

[1585] Step 5:

[1586] The terminal receives the session ID and displays a successful login message to the user.

[1587] Input: Session ID

[1588] Output: Login success message, session ID stored on the device

[1589] Know-how input process

[1590] Step 1:

[1591] The user inputs new know-how and schemes into a dedicated form on the terminal.

[1592] Input: Know-how such as "Scrum development is recommended for Project B"

[1593] Output: Input data temporarily saved on the device

[1594] Step 2:

[1595] The terminal sends the input data to the server in JSON format.

[1596] Input: Know-how input data

[1597] Output: JSON formatted data, data sent to server

[1598] Step 3:

[1599] The server preprocesses the data it receives, including normalizing and structuring the text.

[1600] Input: JSON format data

[1601] Output: Preprocessed data

[1602] Step 4:

[1603] The server inputs the preprocessed data into a generative AI model for analysis, which includes extracting relevant data and generating new insights.

[1604] Input: Preprocessed data

[1605] Output: Analysis results

[1606] Step 5:

[1607] The server saves the analysis results in a database, along with the date and time of saving and information about the person who entered the data.

[1608] Input: Analysis results, inputter information

[1609] Output: Know-how data stored in a database

[1610] Know-how search process

[1611] Step 1:

[1612] The user enters a query into a search form on the device.

[1613] Input: A query such as "Agile development best practices"

[1614] Output: Query saved to the device

[1615] Step 2:

[1616] The device sends a query in JSON format to the server.

[1617] Input: Input data for the query

[1618] Output: Query in JSON format, query sent to server

[1619] Step 3:

[1620] The server receives the query, analyzes it using natural language processing, and extracts the intent.

[1621] Input: JSON format query

[1622] Output: Parsed intent

[1623] Step 4:

[1624] The server sends a search query to a database based on the analyzed intent to obtain related know-how data.

[1625] Input: Parsed intent

[1626] Output: Search results

[1627] Step 5:

[1628] The server returns the search results to the device in JSON format.

[1629] Input: Search results

[1630] Output: Search results in JSON format, search results sent to the device

[1631] Step 6:

[1632] The terminal displays the received search results to the user.

[1633] Input: JSON format search results

[1634] Output: Search results displayed for the user to see

[1635] Feedback Providing Process

[1636] Step 1:

[1637] After applying the know-how, the user inputs feedback from the terminal.

[1638] Input: Feedback such as "This scheme is useful, but specifically task management is lacking."

[1639] Output: Feedback data temporarily saved on the device

[1640] Step 2:

[1641] The device sends the feedback content in JSON format to the server.

[1642] Input: Feedback data

[1643] Output: Feedback data in JSON format, data sent to server

[1644] Step 3:

[1645] The server receives the feedback and performs analysis, which includes categorizing and extracting topics from the feedback text.

[1646] Input: Feedback data in JSON format

[1647] Output: Parsed feedback data

[1648] Step 4:

[1649] The server stores the analyzed feedback data as data for retraining the model.

[1650] Input: Parsed feedback data

[1651] Output: Save the data for retraining

[1652] Step 5:

[1653] The server periodically retrains the generative AI model to improve its accuracy. The improved model is applied to the system and reflected the next time a user searches or inputs know-how.

[1654] Input: Retraining data

[1655] Output: Improved model, newly applied model

[1656] Through the above processing steps, this system allows users to efficiently accumulate, search, and share know-how, and can continuously improve the system based on feedback.

[1657] (Application example 1)

[1658] 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."

[1659] To improve the operational efficiency of robots used in factories, it is necessary to efficiently share operational know-how and troubleshooting knowledge, and to search and utilize it in real time.However, conventional systems do not fully integrate knowledge input, search, and feedback, making it difficult to immediately use operational know-how and improve operations based on that feedback.

[1660] 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.

[1661] In this invention, the server includes means for acquiring user authentication information, means for authenticating the user based on the authentication information, means for inputting knowledge, means for inputting the input data to a generative AI model for analysis, means for saving the analysis results in a database, means for searching for knowledge information from the database, means for providing the search results to the user, means for collecting feedback data and using it to retrain the generative AI model, means for providing knowledge information to the robot in the form of voice and text, and means for generating and saving feedback in real time. This allows factory operational know-how to be efficiently input, searched, and shared, and enables optimization of operations through the provision of real-time feedback.

[1662] "User authentication information" is information used to identify a user and verify their authorization to access a system.

[1663] "Knowledge" refers to information about operational know-how and schemes that is input into the system.

[1664] A "generative artificial intelligence model" is an artificial intelligence algorithm that analyzes input data and extracts and generates relevant information.

[1665] A "robot" is an automated mechanical device used in factories.

[1666] "Feedback data" refers to information about evaluations and improvements provided by users to the system.

[1667] A "database" is a data management system that stores analysis results and knowledge information and makes them searchable.

[1668] "Search results" are related knowledge information retrieved based on a query entered by a user.

[1669] "Secure" means that authentication information and data are obtained and transmitted using safe and protected communications methods.

[1670] "Tagging" is a technique for improving searchability by adding metadata to knowledge data analyzed by a generative artificial intelligence model.

[1671] "Relevance" refers to the degree of match or relevance between a search query and knowledge information in a database.

[1672] MODE FOR CARRYING OUT THE INVENTION

[1673] This invention relates to a system for improving the operational efficiency of robots used in factories. This system integrates knowledge input, search, and feedback, enabling the utilization of operational know-how in real time. Specifically, it is realized with the following configuration.

[1674] User Authentication Process

[1675] When a user logs in to a terminal to operate a robot, they enter user authentication information. This authentication information includes a user ID and password. The terminal then securely sends the authentication information to the server, which then compares it with a database to authenticate the user. If successful, the server generates a session ID and returns it to the terminal.

[1676] Examples:

[1677] The user enters the ID "robot_operator" and password "password123" and confirms that the login was successful.

[1678] Knowledge Input Process

[1679] When a user inputs new knowledge information (for example, operational know-how or maintenance procedures), they use a dedicated input form. The terminal collects this data and sends it in JSON format to the server. The server preprocesses the data, analyzes it using a generative artificial intelligence model, and saves the analysis results in a database. At this time, the date and time of saving and information about the person who input the information are also recorded.

[1680] Examples:

[1681] The user inputs a "new maintenance procedure for the robot" and the analysis results are saved in a database.

[1682] Know-how search process

[1683] When a user searches for existing knowledge information, they enter a query into a search form. The device sends the query in JSON format to the server, which analyzes it using natural language processing to extract the user's intent. The server then sends the search query to a database to retrieve related knowledge data. The resulting search results are returned to the device in JSON format and provided to the user.

[1684] Examples:

[1685] A user searches for "maintenance procedures" and retrieves and displays relevant results.

[1686] Feedback Providing Process

[1687] After applying the knowledge information, the user provides feedback on its effectiveness and areas for improvement. The device sends the feedback in JSON format to the server, which receives and analyzes the feedback and stores it as retraining data for the generative AI model. This data is used to periodically retrain the model and improve the accuracy of the system.

[1688] Examples:

[1689] The user provides feedback saying "this step is effective" and the feedback is saved.

[1690] System operating environment

[1691] Hardware: Factory robots, terminals (operation panels)

[1692] Software: authentication protocols (HTTPS), data formats (JSON), generative artificial intelligence models, natural language processing (NLP) engines, database management systems

[1693] Each process is executed with a specific prompt:

[1694] Login:

[1695] "User ID: robot_operator, Password: password123"

[1696] Enter your know-how:

[1697] "New maintenance procedures for robots"

[1698] Search for know-how:

[1699] Maintenance Procedures

[1700] Providing feedback:

[1701] "This procedure is effective."

[1702] In this way, knowledge sharing and feedback can be efficiently managed in the operation of robots within a factory.

[1703] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1704] Step 1:

[1705] To log in to a terminal, a user enters a user ID and password. The entered authentication information is securely sent by the terminal to the server. The server checks the authentication information against a database, and if authentication is successful, generates a session ID and returns it to the terminal.

[1706] Input: User ID, Password

[1707] Output: Session ID (if authentication is successful)

[1708] Step 2:

[1709] The user enters new knowledge information into an input form. The device sends this information in JSON format to the server. The server preprocesses the received data and analyzes it using a generative AI model. The analysis results are stored in a database. The time of entry and information about the person who entered the information are also recorded.

[1710] Input: New knowledge information

[1711] Output: Analysis results, save date and time, and inputter information

[1712] Step 3:

[1713] A user inputs a query to search existing knowledge information. The device sends this query in JSON format to the server. The server uses a natural language processing (NLP) engine to analyze the query and extract the search intent. The server then sends the search query to a database, retrieves related knowledge data, and returns it to the device.

[1714] Input: Search query

[1715] Output: Search results (related knowledge data)

[1716] Step 4:

[1717] The user operates the robot based on the acquired knowledge information and inputs the results and opinions as feedback. The device sends the feedback information in JSON format to the server. The server analyzes the feedback and saves it as retraining data for the generative AI model. This improves the accuracy of the generative AI model.

[1718] Input: Feedback information

[1719] Output: Feedback analysis results and retraining data

[1720] Step 5:

[1721] The server periodically retrains the generative AI model using the collected feedback data, improving the model's accuracy for the next know-how input, search, or feedback.

[1722] Input: Feedback data

[1723] Output: Retrained generative artificial intelligence model

[1724] These processing steps enable effective management of knowledge input, search, sharing, and feedback on robot operations within the factory, and provide real-time feedback to optimize operations.

[1725] 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.

[1726] This invention relates to a system that efficiently accumulates know-how and schemes specific to a company, makes it easy to share and search them, and also recognizes user emotions and uses the results to provide feedback and information. Specific embodiments of this system are described below.

[1727] ---

[1728] User Authentication Process

[1729] When a user logs in to a terminal, they enter their company ID and password, which confirms their authority to access the system.

[1730] 1. The device collects the user's ID and password and sends them to the server using the HTTPS protocol.

[1731] 2. The server compares the received authentication information with the database and authenticates the user. If authentication is successful, it generates a session ID and returns it to the terminal.

[1732] 3. The terminal receives the session ID and displays a successful login message to the user.

[1733] ---

[1734] Know-how input process

[1735] This is the process by which users input new know-how and schemes into the system, thereby accumulating in-house knowledge.

[1736] 1. The user enters new know-how or scheme (e.g., "We recommend development using Scrum for Project B") into a dedicated form on the terminal.

[1737] 2. The device collects input data and sends it to the server in JSON format.

[1738] 3. The server preprocesses the data and normalizes the text (matching case, removing unnecessary whitespace, etc.).

[1739] 4. The server inputs the preprocessed data into a generative artificial intelligence model and analyzes the know-how data.

[1740] 5. The server obtains the analysis results and extracts link information with related existing know-how and new know-how.

[1741] 6. The server saves the generated know-how data in the database, along with the date and time of saving and information about the person who entered the data.

[1742] ---

[1743] Know-how search process

[1744] It is a process in which users search for existing know-how and schemes.

[1745] 1. A user enters a query (e.g., "Agile development best practices") into a search form on their device.

[1746] 2. The device sends the entered query in JSON format to the server.

[1747] 3. The server analyzes the received query using natural language processing and extracts the intent.

[1748] 4. The server sends a search query to the database to retrieve relevant know-how data. The search results also include a relevance score.

[1749] 5. The server returns the search results to the device in JSON format.

[1750] 6. The device receives the response from the server and displays the search results to the user, providing an interface that makes it easy for the user to find the information they need.

[1751] ---

[1752] User emotion recognition process

[1753] A distinctive feature of this invention is the use of an emotion engine that recognizes the user's emotions.

[1754] 1. When a user enters know-how or browses search results on the device, the emotion engine collects the user's facial expressions and voice data.

[1755] 2. The device analyzes the data collected in real time and estimates the user's emotional state (e.g., satisfaction, dissatisfaction, confusion, etc.).

[1756] 3. The device sends the estimated emotion data to the server.

[1757] 4. The server collects and stores this emotional data as feedback data, and also uses it for know-how feedback and retraining processes.

[1758] Feedback Providing Process

[1759] This is a process in which users provide feedback on the know-how they have acquired. It is characterized by the collection of feedback using an emotion engine.

[1760] 1. When a user inputs feedback from a terminal after applying know-how, the emotion engine automatically recognizes the user's emotional state.

[1761] 2. The device sends the feedback content and emotion data in JSON format to the server.

[1762] 3. The server receives the feedback and emotion data, analyzes it, and saves it as data for retraining the model.

[1763] 4. The server runs a retraining process at regular intervals to improve the performance and accuracy of the model.

[1764] 5. The server applies the improved model to the system and reflects it in the next user search or know-how input.

[1765] ---

[1766] Specific examples

[1767] Example of user authentication

[1768] 1. User "Tanaka" enters ID "tanaka" and password "password123".

[1769] 2. The device sends this to the server.

[1770] 3. The server performs authentication, is successful, and generates a session ID "abc123".

[1771] 4. The terminal saves the session ID and displays "Login successful" to the user.

[1772] Specific examples of know-how input

[1773] 1. User "Tanaka" inputs "Scrum development is recommended for Project B."

[1774] 2. The device sends the data to the server.

[1775] 3. The server inputs the data into a generative artificial intelligence model and stores the analysis results in a database.

[1776] Examples of emotion engines

[1777] 1. While the user “Tanaka” is inputting know-how, the emotion engine collects and analyzes the user’s facial expression data.

[1778] 2. The device sends the analysis results to the server and stores them as feedback data.

[1779] 3. The server uses this emotion data to retrain the AI ​​model.

[1780] This allows users to efficiently input, search, and provide feedback on internal know-how. The addition of an emotion engine also enables advanced feedback collection and customized information provision based on emotions.

[1781] The processing flow will be explained below.

[1782] User Authentication Process

[1783] Step 1:

[1784] The terminal prompts the user to enter their company ID and password. For example, the user enters "tanaka" and "password123."

[1785] Step 2:

[1786] The terminal sends the entered authentication information (ID and password) to the server using the HTTPS protocol.

[1787] Step 3:

[1788] The server queries the database for the received authentication information and searches for a match.

[1789] Step 4:

[1790] The server retrieves the search results from the database and, if authentication is successful, generates a new session ID (e.g., "sessionID_abc123").

[1791] Step 5:

[1792] The server returns the authentication result and session ID to the terminal as an HTTPS response.

[1793] Step 6:

[1794] The terminal receives the response from the server, notifies the user that login was successful, and saves the session ID.

[1795] ---

[1796] Know-how input process

[1797] Step 1:

[1798] The user inputs new know-how or schemes into a dedicated form on the terminal. For example, the user inputs "Scrum-based development is recommended for Project B."

[1799] Step 2:

[1800] The terminal sends the entered data to the server in JSON format.

[1801] Step 3:

[1802] The server preprocesses the data received and normalizes the text (matching case, removing unnecessary whitespace, etc.).

[1803] Step 4:

[1804] The server inputs the preprocessed data into a generative artificial intelligence model and analyzes the know-how data.

[1805] Step 5:

[1806] The server obtains the analysis results and extracts link information with related existing know-how and new know-how.

[1807] Step 6:

[1808] The server saves the generated know-how data in a database, along with the date and time of saving and information about the person who entered the data.

[1809] ---

[1810] Know-how search process

[1811] Step 1:

[1812] A user enters a query into a search form on their device. For example, the user enters "Agile development best practices."

[1813] Step 2:

[1814] The device sends the entered query in JSON format to the server.

[1815] Step 3:

[1816] The server analyzes the received query using natural language processing to extract the intent.

[1817] Step 4:

[1818] The server sends a search query to the database to retrieve relevant know-how data, and the search results also include a relevance score.

[1819] Step 5:

[1820] The server returns the search results to the device in JSON format.

[1821] Step 6:

[1822] The device receives the response from the server and displays the search results to the user, providing an interface that makes it easy for users to find the information they need.

[1823] ---

[1824] User emotion recognition process

[1825] Step 1:

[1826] When a user inputs know-how or browses search results on a device, the emotion engine uses the device's camera and microphone to collect the user's facial expressions and voice data.

[1827] Step 2:

[1828] The device analyzes the facial expression and voice data collected in real time, and the emotion engine estimates the user's emotional state (e.g., satisfaction, dissatisfaction, confusion, etc.).

[1829] Step 3:

[1830] The device sends the estimated emotion data in JSON format to the server.

[1831] Step 4:

[1832] The server collects and stores the received emotional data as feedback data, which is also used for know-how feedback and the retraining process of AI models.

[1833] ---

[1834] Feedback Providing Process

[1835] Step 1:

[1836] It provides feedback on the know-how acquired by the user. For example, the user may input, "This scheme is easy to understand, but I would like to see a concrete example task."

[1837] Step 2:

[1838] The device collects the feedback and sends it to the server in JSON format.

[1839] Step 3:

[1840] The server receives the feedback and performs text analysis to extract the intent and key points of the feedback.

[1841] Step 4:

[1842] The server stores the analyzed feedback data as retraining data for the generative artificial intelligence model.

[1843] Step 5:

[1844] The server runs a retraining process at regular intervals to improve the performance and accuracy of the AI ​​model.

[1845] Step 6:

[1846] The server applies the improved AI model to the system and reflects it in the next user search or know-how input.

[1847] ---

[1848] Specific examples

[1849] Example of user authentication

[1850] Step 1:

[1851] User "Tanaka" enters ID "tanaka" and password "password123".

[1852] Step 2:

[1853] The terminal sends this to the server.

[1854] Step 3:

[1855] The server performs authentication, is successful, and generates a session ID "abc123".

[1856] Step 4:

[1857] The terminal stores the session ID and displays "Login successful" to the user.

[1858] Specific examples of know-how input

[1859] Step 1:

[1860] User "Tanaka" inputs "Scrum development is recommended for Project B."

[1861] Step 2:

[1862] The device sends the data to the server.

[1863] Step 3:

[1864] The server inputs the data into a generative artificial intelligence model and stores the analysis results in a database.

[1865] Examples of emotion engines

[1866] Step 1:

[1867] While the user "Tanaka" is inputting know-how, the emotion engine collects and analyzes facial expression data using the device's camera and microphone.

[1868] Step 2:

[1869] The device sends the analysis results to the server and stores them as feedback data.

[1870] Step 3:

[1871] The server uses this emotion data to retrain the AI ​​model.

[1872] This allows users to efficiently input, search, and provide feedback on internal know-how. The addition of an emotion engine also enables advanced feedback collection and customized information provision based on emotions.

[1873] Example 2

[1874] 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."

[1875] Existing knowledge management systems make it difficult to effectively share and search know-how and schemes accumulated within an organization. They also lack a feedback mechanism that takes user sentiment into account, limiting the means to improve system usability and knowledge quality. This poses a challenge: it is difficult to effectively collect useful knowledge held by users and share and utilize it throughout the organization.

[1876] 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.

[1877] In this invention, the server includes means for acquiring authentication information, means for authenticating users, means for inputting organization-specific knowledge and schemes, means for inputting input data into a generative AI model for analysis, means for saving the analysis results in a database, means for searching for knowledge information from the database, means for providing search results to users, means for recognizing the emotional state of users and collecting data on the same, and means for collecting the emotional data as feedback and using it to retrain the model, thereby enabling users to effectively input and search for knowledge and providing feedback based on their emotional state.

[1878] "Authentication information" refers to information such as ID and password required when a user logs in to a system.

[1879] "User authentication" is the process of verifying a user's access rights based on the authentication information entered.

[1880] "Organization-specific knowledge and schemes" refers to information assets such as experience, know-how, business processes, and best practices within the organization.

[1881] A "generative AI model" is a generative artificial intelligence model that analyzes input data and generates new insights and related information.

[1882] "Analysis results" are the information and insights obtained after input data is analyzed by a generative AI model.

[1883] A "database" is a system for systematically storing and managing analysis results and other information.

[1884] A "search query" is a keyword or phrase that a user enters to search for specific information.

[1885] "Emotional state" refers to emotions estimated from the user's facial expressions and voice data, and includes satisfaction, dissatisfaction, confusion, etc.

[1886] "Feedback" refers to the ratings and comments users provide after using the system.

[1887] "Retraining methods" refers to the process of retraining an AI model using collected feedback and sentiment data to improve the model's performance.

[1888] This invention is a system for effectively inputting, searching, and sharing knowledge and schemes accumulated within an organization, and it can also recognize users' emotions and reflect their feedback. Specific embodiments of this system are described below.

[1889] This system mainly consists of three elements: a server, a terminal, and a user.

[1890] User Authentication Process

[1891] When a user logs in to a terminal, they enter their company ID and password to confirm their authorization to access the system. This authentication process is carried out securely using the HTTPS protocol. The server checks the authentication information against a database, and if it matches, it generates a session ID and returns it to the terminal.

[1892] Know-how input process

[1893] When a user inputs new knowledge or schemes, they use a dedicated form on their device. This data is sent in JSON format to the server, which then inputs it into a generative AI model for analysis. The analysis results are stored in a database and shared across the organization. For example, if a user inputs "We recommend developing using Scrum for Project B," the data is preprocessed on the server and analyzed by a generative AI model (e.g., OpenAI GPT-3).

[1894] Know-how search process

[1895] When a user searches for existing knowledge or schemes, they enter a query into a search form on their device. The query is sent in JSON format to the server, which then analyzes the query's intent using natural language processing. Based on the analysis results, the server retrieves relevant information from the database and provides it to the user. For example, if a user searches for "best practices in agile development," the server provides the user with relevant information based on this query.

[1896] User emotion recognition process

[1897] When a user enters knowledge or browses search results, the device's built-in camera and microphone collect facial and voice data in real time, which is then input into the emotion analysis algorithm. The server receives this emotion data, collects it as feedback, and uses it to retrain the AI ​​model. For example, if a user is confused while entering know-how, the emotion engine will identify that situation and send it to the server.

[1898] Feedback Providing Process

[1899] When users provide feedback on the information obtained by the system, their emotional data is collected along with the feedback. The server analyzes the feedback and emotional data and stores it as retraining data to improve the accuracy and performance of the AI ​​model.

[1900] This system allows users to effectively input and search knowledge within an organization and reflects feedback based on emotional data. For example, by inputting important know-how about a project and reusing that know-how in other projects, the efficiency of the entire organization can be improved. Emotion recognition is also expected to improve usability.

[1901] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1902] Step 1:

[1903] The user enters their company ID and password into the terminal.

[1904] (Input): User ID, password

[1905] (Operation): The user enters their company ID and password on the device's login screen.

[1906] (Output): The device collects the ID and password.

[1907] Step 2:

[1908] The device sends the collected authentication information to the server using the HTTPS protocol.

[1909] (Input): User ID, password

[1910] (Operation): The device converts the ID and password into JSON format and sends it as an HTTPS request to the server's authentication API endpoint.

[1911] (Output): The server receives the authentication information.

[1912] Step 3:

[1913] The server compares the received authentication information with a database and authenticates the user.

[1914] (Input): Authentication information (user ID, password)

[1915] (What happens): The server compares the user information in its database with the entered authentication information.

[1916] (Output): Authentication result (success / failure)

[1917] Step 4:

[1918] When authentication is successful, the server generates a session ID and returns it to the terminal.

[1919] (Input): Authentication information (if successful)

[1920] (Operation): The server generates a unique session ID and sends it to the device in JSON format.

[1921] (Output): Session ID

[1922] Step 5:

[1923] The terminal receives the session ID and displays a "Login successful" message to the user.

[1924] (Input): Session ID

[1925] (Action): The device saves the session ID in local storage and displays a login success message.

[1926] (Output): User interface update (login success message)

[1927] Step 6:

[1928] The user inputs new knowledge or scheme into an input form.

[1929] (Input): Know-how and scheme (text data)

[1930] (Operation): The user inputs new knowledge or scheme into a dedicated form on the terminal.

[1931] (Output): The terminal collects input data.

[1932] Step 7:

[1933] The input data collected by the terminal is sent to the server in JSON format.

[1934] (Input): Know-how and scheme (JSON format)

[1935] (Operation): The terminal converts the input data into JSON format and sends it to the server as an HTTPS request.

[1936] (Output): The server receives the data.

[1937] Step 8:

[1938] The server preprocesses the data and inputs it into a generative AI model for analysis.

[1939] (Input): Know-how and scheme (JSON format)

[1940] (Operation): The server normalizes the data (standardizing uppercase and lowercase letters, removing unnecessary spaces) and inputs it into a generative AI model. The generative AI model analyzes the data and generates new knowledge and related information.

[1941] (Output): Analysis results (new knowledge, related information)

[1942] Step 9:

[1943] The server stores the analysis results in a database.

[1944] (Input): Analysis results (new knowledge, related information)

[1945] (Operation): The server adds the analysis results to the database as a new record, and also records the save date and time and the ID of the person who entered the data.

[1946] (Output): Update the database (save new knowledge)

[1947] Step 10:

[1948] A user enters a query into a search form.

[1949] (Input): Search query (text data)

[1950] Action: A user enters a query into a search form on their device.

[1951] (Output): The terminal collects input data.

[1952] Step 11:

[1953] The device sends the entered query in JSON format to the server.

[1954] (Input): Search query (JSON format)

[1955] (Operation): The terminal converts the input data into JSON format and sends it to the server as an HTTPS request.

[1956] (Output): The server receives the data.

[1957] Step 12:

[1958] The server parses the query and performs a search against the database.

[1959] (Input): Search query (JSON format)

[1960] (Operation): The server analyzes the query using a natural language processing algorithm and sends the search query to the database.

[1961] (Output): Search results (related knowledge information)

[1962] Step 13:

[1963] The server returns the search results to the device in JSON format.

[1964] (Input): Search results (JSON format)

[1965] (Operation): The server encodes the search results into JSON format and sends them to the device as an HTTPS request.

[1966] (Output): The device receives the data.

[1967] Step 14:

[1968] Your device will display the search results.

[1969] (Input): Search results (JSON format)

[1970] (Operation): The terminal displays the search results received on the user interface, making it easy for the user to find the information they need.

[1971] (Output): Update the user interface (display search results)

[1972] Step 15:

[1973] While the user is entering know-how or viewing search results, the device's built-in camera and microphone collect facial expressions and voice data.

[1974] (Input): facial expression data, voice data

[1975] (Operation): The device uses its built-in camera and microphone to collect facial and voice data in real time.

[1976] (Output): Collected emotion data

[1977] Step 16:

[1978] The device analyzes the emotional data collected in real time to estimate the user's emotional state.

[1979] (Input): Collected emotion data

[1980] (Action): The device uses an emotion analysis algorithm to estimate the emotional state (e.g., happy, unhappy, confused).

[1981] (Output): Estimated emotional state

[1982] Step 17:

[1983] The device transmits the estimated emotion data to the server.

[1984] (Input): Estimated emotional state

[1985] (Operation): The device converts the emotion data into JSON format and sends it to the server as an HTTPS request.

[1986] (Output): The server receives the emotion data.

[1987] Step 18:

[1988] The server collects the emotion data as feedback data and uses it to retrain the model.

[1989] (Input): Emotion data (JSON format)

[1990] (How it works): The server stores the emotion data in a feedback database and uses it to retrain the AI ​​model.

[1991] (Output): Update the database and save the retraining data.

[1992] Step 19:

[1993] The server runs a retraining process at regular intervals to improve the model's performance.

[1994] (Input): Retraining data

[1995] (Operation): The server periodically runs a retraining process to improve the accuracy and performance of the AI ​​model.

[1996] (Output): Improved AI model

[1997] Step 20:

[1998] The server applies the improved model to the system and reflects it in the next user search or know-how input.

[1999] (Input): Improved AI model

[2000] (Operation): The server deploys the updated AI model to the system and applies it the next time the user searches or types something.

[2001] (Output): System update (application of improved AI model)

[2002] The above are the specific processing steps of the program of this system.

[2003] (Application example 2)

[2004] 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."

[2005] Improving work efficiency and sharing knowledge within factories are important issues, but conventional systems have had difficulty providing real-time support or feedback based on workers' emotions. Furthermore, collecting feedback that reflects emotional states and how to utilize it has also been an unsolved problem.

[2006] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for identifying the emotional state of a user using an emotion recognition engine and collecting and saving that data, means for collecting feedback and providing information using the emotion data, and means for tagging knowledge data analyzed using a generative artificial intelligence model and making it searchable based on relevance. This not only improves work efficiency in the factory but also makes it possible to provide appropriate feedback based on the emotions of workers.

[2007] "Authentication information" refers to information used to verify a user's identity, and is usually comprised of an ID and password.

[2008] "User authentication" refers to the process of verifying whether a user is a person with legitimate authority using the acquired authentication information.

[2009] "Internal knowledge" refers to specialized information and procedures accumulated within an organization.

[2010] A "generative artificial intelligence model" is an artificial intelligence model that has the ability to generate and analyze knowledge and information based on new data.

[2011] A "database" is a system for efficiently storing and managing data.

[2012] "Knowledge information" refers to internal knowledge and procedures stored in a database.

[2013] "Search results" refers to the related knowledge information obtained when searching a database.

[2014] "Feedback data" refers to data collected from users' evaluations and opinions.

[2015] "Model retraining" is the process of using new data to improve the performance of an artificial intelligence model.

[2016] An "emotion recognition engine" is a technology that analyzes a user's facial expressions and voice to identify their emotional state.

[2017] "Emotional state" is information indicating the type and intensity of the user's emotion.

[2018] "Feedback collection" is the process of collecting opinions and sentiment data from users.

[2019] "Information provision" is the process of displaying search results and feedback to users.

[2020] "Tagging" means assigning labels to data that indicate its relevance.

[2021] This invention uses smart glasses and head-mounted displays (HMDs) to build a system that improves work efficiency in factories. The system includes multiple means for user authentication, know-how input, know-how search, emotion recognition, and feedback collection.

[2022] User Authentication

[2023] The user puts on the smart glasses and enters their ID and password. This authentication information is securely sent to the server using the HTTPS protocol. The server compares the received authentication information with a database, and if authentication is successful, it generates a session ID and returns it to the device. The device saves the session ID and displays a message to the user indicating successful login.

[2024] Input of know-how

[2025] When a user wants to input new know-how or procedures, they use a dedicated form on the smart glasses. The device sends the input data in JSON format to the server, which then preprocesses the data and analyzes it using a generative artificial intelligence model. The analysis results are stored in a database. This process efficiently accumulates knowledge and procedures acquired within the factory.

[2026] Search for know-how

[2027] When a user searches for a specific know-how or procedure, they enter a query into the search form on the smart glasses. The device sends the query in JSON format to the server, which uses natural language processing technology to analyze the query's intent. The server then sends the search query to a database to retrieve relevant know-how information. The search results are returned to the device and displayed visually to the user.

[2028] emotion recognition

[2029] When a user enters know-how or browses search results, the emotion engine collects the user's facial expression and voice data. The device analyzes this data in real time and estimates the user's emotional state. The estimated emotion data is sent to the server and stored as feedback data. This emotion data is used to retrain the model and provide feedback on know-how.

[2030] Feedback collection

[2031] When a user provides feedback, the emotion engine automatically recognizes the user's emotional state. The feedback content and emotion data are sent from the device to the server, where they are analyzed and stored as retraining data for the model. A periodic retraining process is performed to improve the model's performance and accuracy.

[2032] Specific examples

[2033] For example, if a user wears smart glasses and types "Scrum development is recommended for Project B," the device sends the data to the server, where it is analyzed by a generative artificial intelligence model. The analysis results are stored in a database and will be displayed as relevant information when other users later search for "best practices for agile development."

[2034] Prompt Sentence Examples

[2035] Analyze the following know-how data for your generative AI model:

[2036] "Scrum development is recommended for Project B."

[2037] This invention will enable efficient accumulation and sharing of knowledge within factories, and will also enable feedback collection and information provision based on user emotions, which is expected to improve worker satisfaction and work efficiency.

[2038] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2039] Step 1:

[2040] The user puts on the smart glasses and enters their ID and password.

[2041] Input: User ID, Password

[2042] Specific operation: The user enters their ID and password into the login form displayed on the smart glasses display. The smart glasses then send this authentication information to the server using the HTTPS protocol.

[2043] Output: Authentication information sent to the server

[2044] Step 2:

[2045] The server receives the authentication information and authenticates the user by checking it against a database.

[2046] Input: Authentication information (ID, password)

[2047] How it works: The server checks the received authentication information against the user information in its database, and if authentication is successful, generates a session ID, which is then sent back to the smart glasses in a secure manner.

[2048] Output: Session ID

[2049] Step 3:

[2050] The terminal stores the session ID and displays a successful login message to the user.

[2051] Input: Session ID

[2052] Specific operation: The smart glasses store the session ID in memory and notify the user that the login was successful by displaying it on the display.

[2053] Output: Login successful message

[2054] Step 4:

[2055] The user enters new know-how and procedures into a special form on the smart glasses.

[2056] Input: Know-how and procedures

[2057] Specific operation: The user inputs new know-how or procedures (e.g., "We recommend developing using Scrum for Project B") into a specific form through the smart glasses' display. The input data is temporarily stored in the smart glasses' internal storage.

[2058] Output: Input data is temporarily saved

[2059] Step 5:

[2060] The terminal sends the entered data to the server in JSON format.

[2061] Input: Input data

[2062] Specific operation: The smart glasses convert the temporarily stored know-how and procedure data into JSON format and send it to the server.

[2063] Output: JSON format data is sent to the server

[2064] Step 6:

[2065] The server preprocesses the data and performs analysis using a generative artificial intelligence model.

[2066] Input: Input data in JSON format

[2067] How it works: The server preprocesses the received JSON data (e.g., standardizing case, removing unnecessary whitespace, etc.) and then feeds it into a generative artificial intelligence model. The model then analyzes the data and extracts new information and related existing knowledge.

[2068] Output: Analysis results

[2069] Step 7:

[2070] The server stores the analysis results in a database.

[2071] Input: Analysis results

[2072] Specific operation: The server saves the analysis results in a database, and also records the date and time of saving and information about the person who entered the data.

[2073] Output: Analysis results stored in a database

[2074] Step 8:

[2075] A user searches for know-how by entering a query into a search form on the smart glasses.

[2076] Input: Search query

[2077] What it does: A user enters a specific query (e.g., "Agile development best practices") into the search form on the smart glasses. The smart glasses convert the query into JSON format and send it to the server.

[2078] Output: The search query is sent to the server

[2079] Step 9:

[2080] The server analyzes the received query and searches the database for related know-how information.

[2081] Input: Search query

[2082] Specific operation: The server analyzes the received search query using natural language processing technology, sends an appropriate search query to the database, and retrieves related know-how information.

[2083] Output: Search results

[2084] Step 10:

[2085] The server sends the search results back to the device, which then displays them on the smart glasses.

[2086] Input: Search results

[2087] Specific operation: The server converts the search results into JSON format and sends them to the smart glasses, which then display them visually for the user to view.

[2088] Output: Search results displayed to the user

[2089] Step 11:

[2090] The emotion recognition engine collects the user's facial expression and voice data and analyzes their emotional state in real time.

[2091] Input: facial expression data, voice data

[2092] How it works: The smart glasses' built-in cameras and microphones collect the user's facial expressions and voice in real time. The emotion recognition engine analyzes this data and estimates the user's emotional state (e.g., satisfaction, dissatisfaction, confusion, etc.).

[2093] Output: Estimated emotional state

[2094] Step 12:

[2095] The device transmits the estimated emotion data to the server and stores it as feedback data.

[2096] Input: Emotion data

[2097] Specific operation: The emotion recognition engine converts the estimated emotional state into JSON format and sends it to the server. The server then stores the received emotional data in a database as feedback data.

[2098] Output: Emotion data stored in a database

[2099] Step 13:

[2100] The server uses the feedback data to retrain the model.

[2101] Input: Feedback data

[2102] Specific operation: The server periodically collects feedback data and retrains the generative AI model, thereby improving the model's performance and accuracy.

[2103] Output: Retrained generative artificial intelligence model

[2104] Through the above processing steps, the present invention improves work efficiency in a factory and provides appropriate feedback based on the user's emotions.

[2105] 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.

[2106] 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.

[2107] 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.

[2108] [Fourth embodiment]

[2109] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[2110] 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.

[2111] 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).

[2112] 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.

[2113] 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.

[2114] 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).

[2115] 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.

[2116] 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.

[2117] 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.

[2118] 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.

[2119] 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.

[2120] 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.

[2121] 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."

[2122] The present invention relates to a system for efficiently storing know-how and schemes specific to a company, and facilitating sharing and searching thereof. A specific embodiment of this system will be described below.

[2123] ---

[2124] User Authentication Process

[2125] When a user logs in to a terminal, they enter their company ID and password, which confirms their authority to access the system.

[2126] 1. The device collects the user's ID and password and sends them to the server using the HTTPS protocol.

[2127] 2. The server compares the received authentication information with the database and authenticates the user. If authentication is successful, it generates a session ID and returns it to the terminal.

[2128] 3. The terminal receives the session ID and displays a successful login message to the user.

[2129] Know-how input process

[2130] This is the process by which users input new know-how and schemes into the system, thereby accumulating in-house knowledge.

[2131] 1. The user enters new know-how or scheme (e.g., "Agile development is being adopted for Project A") into a dedicated form on the device.

[2132] 2. The device collects input data and sends it to the server in JSON format.

[2133] 3. The server preprocesses the data, normalizing and structuring the text.

[2134] 4. The server inputs the preprocessed data into a generative AI model for analysis. The analysis results include related existing data and new insights.

[2135] 5. The server saves the analysis results in a database. When saving, the date and time of saving and information about the person who entered the data are also recorded.

[2136] Know-how search process

[2137] It is a process in which users search for existing know-how and schemes.

[2138] 1. A user enters a query (e.g., "Agile development best practices") into a search form on their device.

[2139] 2. The device sends a query in JSON format to the server.

[2140] 3. The server analyzes the received query using natural language processing to extract the intent.

[2141] 4. The server sends a search query to the database to retrieve relevant know-how data.

[2142] 5. The server returns the search results to the device in JSON format.

[2143] 6. The device displays the search results to the user.

[2144] Feedback Providing Process

[2145] This is the process of providing feedback to users on the know-how they have acquired.

[2146] 1. The user inputs feedback from the terminal after applying the know-how (e.g., "This scheme is useful, but specifically, task management is lacking").

[2147] 2. The device sends the feedback content in JSON format to the server.

[2148] 3. The server receives the feedback, analyzes it, and saves it as data for retraining the model.

[2149] 4. The server periodically retrains the generative AI model to improve its accuracy.

[2150] 5. The server applies the improved model to the system and reflects it in the next user search or know-how input.

[2151] ---

[2152] Specific examples

[2153] Example of user authentication

[2154] 1. User "Tanaka" enters ID "tanaka" and password "password123".

[2155] 2. The device sends this to the server.

[2156] 3. The server performs authentication, is successful, and generates a session ID "abc123".

[2157] 4. The terminal saves the session ID and displays "Login successful" to the user.

[2158] Specific examples of know-how input

[2159] 1. User "Tanaka" inputs "Scrum development is recommended for Project B."

[2160] 2. The device sends the data to the server.

[2161] 3. The server inputs the data into a generative artificial intelligence model and stores the analysis results in a database.

[2162] This will create a system that allows users to efficiently input, search, and share in-house know-how.

[2163] The processing flow will be explained below.

[2164] User Authentication Process

[2165] Step 1:

[2166] The terminal prompts the user to enter their company ID and password. For example, the user enters "tanaka" and "password123."

[2167] Step 2:

[2168] The terminal sends the entered authentication information to the server using the HTTPS protocol.

[2169] Step 3:

[2170] The server queries the database for the received authentication information and searches for a match.

[2171] Step 4:

[2172] The server retrieves the search results from the database and, if authentication is successful, generates a new session ID (e.g., "sessionID_abc123").

[2173] Step 5:

[2174] The server returns the authentication result and session ID to the terminal as an HTTPS response.

[2175] Step 6:

[2176] The terminal receives the response from the server, notifies the user that login was successful, and saves the session ID.

[2177] ---

[2178] Know-how input process

[2179] Step 1:

[2180] The user inputs new know-how or schemes into a dedicated form on the terminal. For example, the user inputs "Scrum-based development is recommended for Project B."

[2181] Step 2:

[2182] The device collects the input data and sends it to the server in JSON format.

[2183] Step 3:

[2184] The server preprocesses the data received and normalizes the text (matching case, removing unnecessary whitespace, etc.).

[2185] Step 4:

[2186] The server inputs the preprocessed data into a generative artificial intelligence model and analyzes the know-how data.

[2187] Step 5:

[2188] The server obtains the analysis results and extracts link information with related existing know-how and new know-how.

[2189] Step 6:

[2190] The server saves the generated know-how data in a database, along with the date and time of saving and information about the person who entered the data.

[2191] ---

[2192] Know-how search process

[2193] Step 1:

[2194] A user enters a query into a search form on their device. For example, the user enters "Agile development best practices."

[2195] Step 2:

[2196] The device sends the entered query in JSON format to the server.

[2197] Step 3:

[2198] The server analyzes the received query using natural language processing to understand the intent of the query and extract related keywords.

[2199] Step 4:

[2200] The server sends a search query to the database to retrieve relevant know-how data, and the search results also include a relevance score.

[2201] Step 5:

[2202] The server returns the search results to the device in JSON format.

[2203] Step 6:

[2204] The device receives the response from the server and displays the search results to the user, providing an interface that makes it easy for users to find the information they need.

[2205] ---

[2206] Feedback Providing Process

[2207] Step 1:

[2208] It provides feedback on the know-how acquired by the user. For example, the user may input, "This scheme is easy to understand, but I would like to see a concrete example task."

[2209] Step 2:

[2210] The device collects the feedback and sends it to the server in JSON format.

[2211] Step 3:

[2212] The server receives the feedback and performs text analysis to extract the intent and key points of the feedback.

[2213] Step 4:

[2214] The server stores the analyzed feedback data as retraining data for the generative artificial intelligence model.

[2215] Step 5:

[2216] The server runs a retraining process at regular intervals to improve the performance and accuracy of the model.

[2217] Step 6:

[2218] The server applies the improved model to the system and reflects it in the next user search or know-how input.

[2219] Through these steps, users can efficiently input, search, and provide feedback on their internal know-how.

[2220] Example 1

[2221] 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."

[2222] Previous methods for sharing in-house know-how and schemes mainly involved manual recording and management, which lacked efficiency. It was also difficult to search for know-how or discover new insights, resulting in low utilization of information. Furthermore, there was no established method for collecting feedback and using it to update and improve knowledge. For these reasons, a new system for effectively storing and utilizing in-house knowledge resources was needed.

[2223] 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.

[2224] In this invention, the server includes means for acquiring authentication information, means for authenticating a user based on the authentication information, means for inputting in-house know-how and schemes, means for transmitting the input data to the server in JSON format, means for inputting the input data to a generative artificial intelligence model for analysis, means for saving the analysis results in a database, means for searching the database for know-how information using natural language processing, means for providing the search results to the user in JSON format, and means for collecting feedback data and using it to retrain the model. This enables efficient accumulation, sharing, and search of in-house know-how, and further enables the model to be improved based on feedback to increase the quality of knowledge.

[2225] "Authentication Information" means information required to verify a user's identity, and primarily includes a user ID and password.

[2226] "User authentication" is the process of verifying that a user is a legitimate user using authentication information.

[2227] "Know-how" refers to information that refers to knowledge, techniques, and methodologies related to a specific task or project.

[2228] "Scheme" means the plan, structure or procedure for a particular project or business process.

[2229] "JSON format" is a lightweight data exchange format for structuring data, and is an abbreviation for JavaScript Object Notation.

[2230] A "generative artificial intelligence model" is a set of machine learning algorithms that generate new information and insights based on input data.

[2231] "Analysis" is the process performed to understand input data and find patterns and relationships.

[2232] A "database" is a system for organizing, efficiently storing, and retrieving information.

[2233] "Natural language processing" is a technical field that uses computers to understand and process human language.

[2234] "Feedback data" refers to information such as ratings, comments, and improvements provided by users regarding the system's output.

[2235] "Model retraining" is the process of updating a generative artificial intelligence model with new data collected to improve its accuracy and performance.

[2236] This invention relates to a system that efficiently accumulates in-house know-how and schemes and makes it easy to share and search them. This system includes a series of processes for collecting, transmitting, analyzing, and storing data between a server, terminals, and users.

[2237] User Authentication Process

[2238] When a user enters their company ID and password into the terminal, the terminal encrypts the authentication information and sends it to the server using the HTTPS protocol. The server compares the authentication information with a database, and if authentication is successful, generates a session ID and returns it to the terminal. This allows the user to access the system securely.

[2239] Know-how input process

[2240] The user inputs new know-how or schemes into a dedicated form on the device. For example, this information might be something like "Scrum development is recommended for Project B." The device then sends the input data in JSON format to the server. The server preprocesses the data and analyzes it using a generative artificial intelligence model. The analysis results are saved in a database, along with the date and time of saving and information about the person who entered the data.

[2241] Know-how search process

[2242] When a user enters a query (e.g., "Agile development best practices") into the device's search form, the device sends the query in JSON format to the server. The server analyzes the query using natural language processing technology and sends the search query to the database. The search results are returned to the device in JSON format and displayed to the user.

[2243] Feedback Providing Process

[2244] After applying the know-how, the user inputs feedback from the device. For example, the user might say, "This scheme is useful, but specifically, task management is lacking." The device then sends the feedback in JSON format to the server. The server analyzes the feedback and saves it as retraining data for the generative AI model. This improves the model's accuracy, which is reflected the next time the user searches or inputs know-how.

[2245] Hardware and software used

[2246] This system uses the following hardware and software:

[2247] Device: A computer or smartphone operated by a user.

[2248] Server: A computer system that processes and stores data.

[2249] Database: A system for storing know-how, schemes, analysis results, and feedback data.

[2250] HTTPS protocol: A communications protocol for secure data transmission.

[2251] Generative artificial intelligence models: Algorithms that analyze input data and generate new insights.

[2252] Specific examples

[2253] Example of user authentication

[2254] The user enters the ID "tanaka" and password "password123" into the terminal, which then encrypts and sends it to the server. The server compares it with the database, and if authentication is successful, it generates a session ID "abc123" and returns it to the terminal. The terminal saves the session ID and displays "Login successful" to the user.

[2255] Specific examples of know-how input

[2256] The user inputs "Scrum development is recommended for Project B," and the device sends the data in JSON format to the server. The server preprocesses the data, inputs it into a generative artificial intelligence model for analysis, and stores the analysis results in a database.

[2257] Specific examples of know-how search

[2258] A user types "Agile development best practices" into a search form, and the device sends the query to the server, which parses the query, searches the database, and returns relevant results to the device, where the user is presented with the search results.

[2259] Example prompt sentence:

[2260] "Project B is recommended to be developed using Scrum. What are some best practices that can be used for other projects?"

[2261] This system enables efficient accumulation, sharing, and searching of internal know-how, allowing for improvements to generative AI models based on feedback.

[2262] The flow of the identification process in the first embodiment will be described with reference to FIG.

[2263] Program processing flow

[2264] User Authentication Process

[2265] Step 1:

[2266] The user enters their company ID and password into the device's login screen.

[2267] Input: Company ID, password

[2268] Output: Credentials sent to the device

[2269] Step 2:

[2270] The terminal encrypts the entered ID and password using the HTTPS protocol and sends it to the server.

[2271] Input: Company ID, password

[2272] Output: Authentication information encrypted using the HTTPS protocol

[2273] Step 3:

[2274] The server checks the received authentication information against a database and authenticates the user.

[2275] Input: Credentials

[2276] Output: Authentication success or failure, session ID

[2277] Step 4:

[2278] If the authentication is successful, the server generates a session ID and returns it to the terminal.

[2279] Input: Authentication success result

[2280] Output: Session ID

[2281] Step 5:

[2282] The terminal receives the session ID and displays a successful login message to the user.

[2283] Input: Session ID

[2284] Output: Login success message, session ID stored on the device

[2285] Know-how input process

[2286] Step 1:

[2287] The user inputs new know-how and schemes into a dedicated form on the terminal.

[2288] Input: Know-how such as "Scrum development is recommended for Project B"

[2289] Output: Input data temporarily saved on the device

[2290] Step 2:

[2291] The terminal sends the input data to the server in JSON format.

[2292] Input: Know-how input data

[2293] Output: JSON formatted data, data sent to server

[2294] Step 3:

[2295] The server preprocesses the data it receives, including normalizing and structuring the text.

[2296] Input: JSON format data

[2297] Output: Preprocessed data

[2298] Step 4:

[2299] The server inputs the preprocessed data into a generative AI model for analysis, which includes extracting relevant data and generating new insights.

[2300] Input: Preprocessed data

[2301] Output: Analysis results

[2302] Step 5:

[2303] The server saves the analysis results in a database, along with the date and time of saving and information about the person who entered the data.

[2304] Input: Analysis results, inputter information

[2305] Output: Know-how data stored in a database

[2306] Know-how search process

[2307] Step 1:

[2308] The user enters a query into a search form on the device.

[2309] Input: A query such as "Agile development best practices"

[2310] Output: Query saved to the device

[2311] Step 2:

[2312] The device sends a query in JSON format to the server.

[2313] Input: Input data for the query

[2314] Output: Query in JSON format, query sent to server

[2315] Step 3:

[2316] The server receives the query, analyzes it using natural language processing, and extracts the intent.

[2317] Input: JSON format query

[2318] Output: Parsed intent

[2319] Step 4:

[2320] The server sends a search query to a database based on the analyzed intent to obtain related know-how data.

[2321] Input: Parsed intent

[2322] Output: Search results

[2323] Step 5:

[2324] The server returns the search results to the device in JSON format.

[2325] Input: Search results

[2326] Output: Search results in JSON format, search results sent to the device

[2327] Step 6:

[2328] The terminal displays the received search results to the user.

[2329] Input: JSON format search results

[2330] Output: Search results displayed for the user to see

[2331] Feedback Providing Process

[2332] Step 1:

[2333] After applying the know-how, the user inputs feedback from the terminal.

[2334] Input: Feedback such as "This scheme is useful, but specifically task management is lacking."

[2335] Output: Feedback data temporarily saved on the device

[2336] Step 2:

[2337] The device sends the feedback content in JSON format to the server.

[2338] Input: Feedback data

[2339] Output: Feedback data in JSON format, data sent to server

[2340] Step 3:

[2341] The server receives the feedback and performs analysis, which includes categorizing and extracting topics from the feedback text.

[2342] Input: Feedback data in JSON format

[2343] Output: Parsed feedback data

[2344] Step 4:

[2345] The server stores the analyzed feedback data as data for retraining the model.

[2346] Input: Parsed feedback data

[2347] Output: Save the data for retraining

[2348] Step 5:

[2349] The server periodically retrains the generative AI model to improve its accuracy. The improved model is applied to the system and reflected the next time a user searches or inputs know-how.

[2350] Input: Retraining data

[2351] Output: Improved model, newly applied model

[2352] Through the above processing steps, this system allows users to efficiently accumulate, search, and share know-how, and can continuously improve the system based on feedback.

[2353] (Application example 1)

[2354] 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."

[2355] To improve the operational efficiency of robots used in factories, it is necessary to efficiently share operational know-how and troubleshooting knowledge, and to search and utilize it in real time.However, conventional systems do not fully integrate knowledge input, search, and feedback, making it difficult to immediately use operational know-how and improve operations based on that feedback.

[2356] 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.

[2357] In this invention, the server includes means for acquiring user authentication information, means for authenticating the user based on the authentication information, means for inputting knowledge, means for inputting the input data to a generative AI model for analysis, means for saving the analysis results in a database, means for searching for knowledge information from the database, means for providing the search results to the user, means for collecting feedback data and using it to retrain the generative AI model, means for providing knowledge information to the robot in the form of voice and text, and means for generating and saving feedback in real time. This allows factory operational know-how to be efficiently input, searched, and shared, and enables optimization of operations through the provision of real-time feedback.

[2358] "User authentication information" is information used to identify a user and verify their authorization to access a system.

[2359] "Knowledge" refers to information about operational know-how and schemes that is input into the system.

[2360] A "generative artificial intelligence model" is an artificial intelligence algorithm that analyzes input data and extracts and generates relevant information.

[2361] A "robot" is an automated mechanical device used in factories.

[2362] "Feedback data" refers to information about evaluations and improvements provided by users to the system.

[2363] A "database" is a data management system that stores analysis results and knowledge information and makes them searchable.

[2364] "Search results" are related knowledge information retrieved based on a query entered by a user.

[2365] "Secure" means that authentication information and data are obtained and transmitted using safe and protected communications methods.

[2366] "Tagging" is a technique for improving searchability by adding metadata to knowledge data analyzed by a generative artificial intelligence model.

[2367] "Relevance" refers to the degree of match or relevance between a search query and knowledge information in a database.

[2368] MODE FOR CARRYING OUT THE INVENTION

[2369] This invention relates to a system for improving the operational efficiency of robots used in factories. This system integrates knowledge input, search, and feedback, enabling the utilization of operational know-how in real time. Specifically, it is realized with the following configuration.

[2370] User Authentication Process

[2371] When a user logs in to a terminal to operate a robot, they enter user authentication information. This authentication information includes a user ID and password. The terminal then securely sends the authentication information to the server, which then compares it with a database to authenticate the user. If successful, the server generates a session ID and returns it to the terminal.

[2372] Examples:

[2373] The user enters the ID "robot_operator" and password "password123" and confirms that the login was successful.

[2374] Knowledge Input Process

[2375] When a user inputs new knowledge information (for example, operational know-how or maintenance procedures), they use a dedicated input form. The terminal collects this data and sends it in JSON format to the server. The server preprocesses the data, analyzes it using a generative artificial intelligence model, and saves the analysis results in a database. At this time, the date and time of saving and information about the person who input the information are also recorded.

[2376] Examples:

[2377] The user inputs a "new maintenance procedure for the robot" and the analysis results are saved in a database.

[2378] Know-how search process

[2379] When a user searches for existing knowledge information, they enter a query into a search form. The device sends the query in JSON format to the server, which analyzes it using natural language processing to extract the user's intent. The server then sends the search query to a database to retrieve related knowledge data. The resulting search results are returned to the device in JSON format and provided to the user.

[2380] Examples:

[2381] A user searches for "maintenance procedures" and retrieves and displays relevant results.

[2382] Feedback Providing Process

[2383] After applying the knowledge information, the user provides feedback on its effectiveness and areas for improvement. The device sends the feedback in JSON format to the server, which receives and analyzes the feedback and stores it as retraining data for the generative AI model. This data is used to periodically retrain the model and improve the accuracy of the system.

[2384] Examples:

[2385] The user provides feedback saying "this step is effective" and the feedback is saved.

[2386] System operating environment

[2387] Hardware: Factory robots, terminals (operation panels)

[2388] Software: authentication protocols (HTTPS), data formats (JSON), generative artificial intelligence models, natural language processing (NLP) engines, database management systems

[2389] Each process is executed with a specific prompt:

[2390] Login:

[2391] "User ID: robot_operator, Password: password123"

[2392] Enter your know-how:

[2393] "New maintenance procedures for robots"

[2394] Search for know-how:

[2395] Maintenance Procedures

[2396] Providing feedback:

[2397] "This procedure is effective."

[2398] In this way, knowledge sharing and feedback can be efficiently managed in the operation of robots within a factory.

[2399] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[2400] Step 1:

[2401] To log in to a terminal, a user enters a user ID and password. The entered authentication information is securely sent by the terminal to the server. The server checks the authentication information against a database, and if authentication is successful, generates a session ID and returns it to the terminal.

[2402] Input: User ID, Password

[2403] Output: Session ID (if authentication is successful)

[2404] Step 2:

[2405] The user enters new knowledge information into an input form. The device sends this information in JSON format to the server. The server preprocesses the received data and analyzes it using a generative AI model. The analysis results are stored in a database. The time of entry and information about the person who entered the information are also recorded.

[2406] Input: New knowledge information

[2407] Output: Analysis results, save date and time, and inputter information

[2408] Step 3:

[2409] A user inputs a query to search existing knowledge information. The device sends this query in JSON format to the server. The server uses a natural language processing (NLP) engine to analyze the query and extract the search intent. The server then sends the search query to a database, retrieves related knowledge data, and returns it to the device.

[2410] Input: Search query

[2411] Output: Search results (related knowledge data)

[2412] Step 4:

[2413] The user operates the robot based on the acquired knowledge information and inputs the results and opinions as feedback. The device sends the feedback information in JSON format to the server. The server analyzes the feedback and saves it as retraining data for the generative AI model. This improves the accuracy of the generative AI model.

[2414] Input: Feedback information

[2415] Output: Feedback analysis results and retraining data

[2416] Step 5:

[2417] The server periodically retrains the generative AI model using the collected feedback data, improving the model's accuracy for the next know-how input, search, or feedback.

[2418] Input: Feedback data

[2419] Output: Retrained generative artificial intelligence model

[2420] These processing steps enable effective management of knowledge input, search, sharing, and feedback on robot operations within the factory, and provide real-time feedback to optimize operations.

[2421] 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.

[2422] This invention relates to a system that efficiently accumulates know-how and schemes specific to a company, makes it easy to share and search them, and also recognizes user emotions and uses the results to provide feedback and information. Specific embodiments of this system are described below.

[2423] ---

[2424] User Authentication Process

[2425] When a user logs in to a terminal, they enter their company ID and password, which confirms their authority to access the system.

[2426] 1. The device collects the user's ID and password and sends them to the server using the HTTPS protocol.

[2427] 2. The server compares the received authentication information with the database and authenticates the user. If authentication is successful, it generates a session ID and returns it to the terminal.

[2428] 3. The terminal receives the session ID and displays a successful login message to the user.

[2429] ---

[2430] Know-how input process

[2431] This is the process by which users input new know-how and schemes into the system, thereby accumulating in-house knowledge.

[2432] 1. The user enters new know-how or scheme (e.g., "We recommend development using Scrum for Project B") into a dedicated form on the terminal.

[2433] 2. The device collects input data and sends it to the server in JSON format.

[2434] 3. The server preprocesses the data and normalizes the text (matching case, removing unnecessary whitespace, etc.).

[2435] 4. The server inputs the preprocessed data into a generative artificial intelligence model and analyzes the know-how data.

[2436] 5. The server obtains the analysis results and extracts link information with related existing know-how and new know-how.

[2437] 6. The server saves the generated know-how data in the database, along with the date and time of saving and information about the person who entered the data.

[2438] ---

[2439] Know-how search process

[2440] It is a process in which users search for existing know-how and schemes.

[2441] 1. A user enters a query (e.g., "Agile development best practices") into a search form on their device.

[2442] 2. The device sends the entered query in JSON format to the server.

[2443] 3. The server analyzes the received query using natural language processing and extracts the intent.

[2444] 4. The server sends a search query to the database to retrieve relevant know-how data. The search results also include a relevance score.

[2445] 5. The server returns the search results to the device in JSON format.

[2446] 6. The device receives the response from the server and displays the search results to the user, providing an interface that makes it easy for the user to find the information they need.

[2447] ---

[2448] User emotion recognition process

[2449] A distinctive feature of this invention is the use of an emotion engine that recognizes the user's emotions.

[2450] 1. When a user enters know-how or browses search results on the device, the emotion engine collects the user's facial expressions and voice data.

[2451] 2. The device analyzes the data collected in real time and estimates the user's emotional state (e.g., satisfaction, dissatisfaction, confusion, etc.).

[2452] 3. The device sends the estimated emotion data to the server.

[2453] 4. The server collects and stores this emotional data as feedback data, and also uses it for know-how feedback and retraining processes.

[2454] Feedback Providing Process

[2455] This is a process in which users provide feedback on the know-how they have acquired. It is characterized by the collection of feedback using an emotion engine.

[2456] 1. When a user inputs feedback from a terminal after applying know-how, the emotion engine automatically recognizes the user's emotional state.

[2457] 2. The device sends the feedback content and emotion data in JSON format to the server.

[2458] 3. The server receives the feedback and emotion data, analyzes it, and saves it as data for retraining the model.

[2459] 4. The server runs a retraining process at regular intervals to improve the performance and accuracy of the model.

[2460] 5. The server applies the improved model to the system and reflects it in the next user search or know-how input.

[2461] ---

[2462] Specific examples

[2463] Example of user authentication

[2464] 1. User "Tanaka" enters ID "tanaka" and password "password123".

[2465] 2. The device sends this to the server.

[2466] 3. The server performs authentication, is successful, and generates a session ID "abc123".

[2467] 4. The terminal saves the session ID and displays "Login successful" to the user.

[2468] Specific examples of know-how input

[2469] 1. User "Tanaka" inputs "Scrum development is recommended for Project B."

[2470] 2. The device sends the data to the server.

[2471] 3. The server inputs the data into a generative artificial intelligence model and stores the analysis results in a database.

[2472] Examples of emotion engines

[2473] 1. While the user “Tanaka” is inputting know-how, the emotion engine collects and analyzes the user’s facial expression data.

[2474] 2. The device sends the analysis results to the server and stores them as feedback data.

[2475] 3. The server uses this emotion data to retrain the AI ​​model.

[2476] This allows users to efficiently input, search, and provide feedback on internal know-how. The addition of an emotion engine also enables advanced feedback collection and customized information provision based on emotions.

[2477] The processing flow will be explained below.

[2478] User Authentication Process

[2479] Step 1:

[2480] The terminal prompts the user to enter their company ID and password. For example, the user enters "tanaka" and "password123."

[2481] Step 2:

[2482] The terminal sends the entered authentication information (ID and password) to the server using the HTTPS protocol.

[2483] Step 3:

[2484] The server queries the database for the received authentication information and searches for a match.

[2485] Step 4:

[2486] The server retrieves the search results from the database and, if authentication is successful, generates a new session ID (e.g., "sessionID_abc123").

[2487] Step 5:

[2488] The server returns the authentication result and session ID to the terminal as an HTTPS response.

[2489] Step 6:

[2490] The terminal receives the response from the server, notifies the user that login was successful, and saves the session ID.

[2491] ---

[2492] Know-how input process

[2493] Step 1:

[2494] The user inputs new know-how or schemes into a dedicated form on the terminal. For example, the user inputs "Scrum-based development is recommended for Project B."

[2495] Step 2:

[2496] The terminal sends the entered data to the server in JSON format.

[2497] Step 3:

[2498] The server preprocesses the data received and normalizes the text (matching case, removing unnecessary whitespace, etc.).

[2499] Step 4:

[2500] The server inputs the preprocessed data into a generative artificial intelligence model and analyzes the know-how data.

[2501] Step 5:

[2502] The server obtains the analysis results and extracts link information with related existing know-how and new know-how.

[2503] Step 6:

[2504] The server saves the generated know-how data in a database, along with the date and time of saving and information about the person who entered the data.

[2505] ---

[2506] Know-how search process

[2507] Step 1:

[2508] A user enters a query into a search form on their device. For example, the user enters "Agile development best practices."

[2509] Step 2:

[2510] The device sends the entered query in JSON format to the server.

[2511] Step 3:

[2512] The server analyzes the received query using natural language processing to extract the intent.

[2513] Step 4:

[2514] The server sends a search query to the database to retrieve relevant know-how data, and the search results also include a relevance score.

[2515] Step 5:

[2516] The server returns the search results to the device in JSON format.

[2517] Step 6:

[2518] The device receives the response from the server and displays the search results to the user, providing an interface that makes it easy for users to find the information they need.

[2519] ---

[2520] User emotion recognition process

[2521] Step 1:

[2522] When a user inputs know-how or browses search results on a device, the emotion engine uses the device's camera and microphone to collect the user's facial expressions and voice data.

[2523] Step 2:

[2524] The device analyzes the facial expression and voice data collected in real time, and the emotion engine estimates the user's emotional state (e.g., satisfaction, dissatisfaction, confusion, etc.).

[2525] Step 3:

[2526] The device sends the estimated emotion data in JSON format to the server.

[2527] Step 4:

[2528] The server collects and stores the received emotional data as feedback data, which is also used for know-how feedback and the retraining process of AI models.

[2529] ---

[2530] Feedback Providing Process

[2531] Step 1:

[2532] It provides feedback on the know-how acquired by the user. For example, the user may input, "This scheme is easy to understand, but I would like to see a concrete example task."

[2533] Step 2:

[2534] The device collects the feedback and sends it to the server in JSON format.

[2535] Step 3:

[2536] The server receives the feedback and performs text analysis to extract the intent and key points of the feedback.

[2537] Step 4:

[2538] The server stores the analyzed feedback data as retraining data for the generative artificial intelligence model.

[2539] Step 5:

[2540] The server runs a retraining process at regular intervals to improve the performance and accuracy of the AI ​​model.

[2541] Step 6:

[2542] The server applies the improved AI model to the system and reflects it in the next user search or know-how input.

[2543] ---

[2544] Specific examples

[2545] Example of user authentication

[2546] Step 1:

[2547] User "Tanaka" enters ID "tanaka" and password "password123".

[2548] Step 2:

[2549] The terminal sends this to the server.

[2550] Step 3:

[2551] The server performs authentication, is successful, and generates a session ID "abc123".

[2552] Step 4:

[2553] The terminal stores the session ID and displays "Login successful" to the user.

[2554] Specific examples of know-how input

[2555] Step 1:

[2556] User "Tanaka" inputs "Scrum development is recommended for Project B."

[2557] Step 2:

[2558] The device sends the data to the server.

[2559] Step 3:

[2560] The server inputs the data into a generative artificial intelligence model and stores the analysis results in a database.

[2561] Examples of emotion engines

[2562] Step 1:

[2563] While the user "Tanaka" is inputting know-how, the emotion engine collects and analyzes facial expression data using the device's camera and microphone.

[2564] Step 2:

[2565] The device sends the analysis results to the server and stores them as feedback data.

[2566] Step 3:

[2567] The server uses this emotion data to retrain the AI ​​model.

[2568] This allows users to efficiently input, search, and provide feedback on internal know-how. The addition of an emotion engine also enables advanced feedback collection and customized information provision based on emotions.

[2569] Example 2

[2570] 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."

[2571] Existing knowledge management systems make it difficult to effectively share and search know-how and schemes accumulated within an organization. They also lack a feedback mechanism that takes user sentiment into account, limiting the means to improve system usability and knowledge quality. This poses a challenge: it is difficult to effectively collect useful knowledge held by users and share and utilize it throughout the organization.

[2572] 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.

[2573] In this invention, the server includes means for acquiring authentication information, means for authenticating users, means for inputting organization-specific knowledge and schemes, means for inputting input data into a generative AI model for analysis, means for saving the analysis results in a database, means for searching for knowledge information from the database, means for providing search results to users, means for recognizing the emotional state of users and collecting data on the same, and means for collecting the emotional data as feedback and using it to retrain the model, thereby enabling users to effectively input and search for knowledge and providing feedback based on their emotional state.

[2574] "Authentication information" refers to information such as ID and password required when a user logs in to a system.

[2575] "User authentication" is the process of verifying a user's access rights based on the authentication information entered.

[2576] "Organization-specific knowledge and schemes" refers to information assets such as experience, know-how, business processes, and best practices within the organization.

[2577] A "generative AI model" is a generative artificial intelligence model that analyzes input data and generates new insights and related information.

[2578] "Analysis results" are the information and insights obtained after input data is analyzed by a generative AI model.

[2579] A "database" is a system for systematically storing and managing analysis results and other information.

[2580] A "search query" is a keyword or phrase that a user enters to search for specific information.

[2581] "Emotional state" refers to emotions estimated from the user's facial expressions and voice data, and includes satisfaction, dissatisfaction, confusion, etc.

[2582] "Feedback" refers to the ratings and comments users provide after using the system.

[2583] "Retraining methods" refers to the process of retraining an AI model using collected feedback and sentiment data to improve the model's performance.

[2584] This invention is a system for effectively inputting, searching, and sharing knowledge and schemes accumulated within an organization, and it can also recognize users' emotions and reflect their feedback. Specific embodiments of this system are described below.

[2585] This system mainly consists of three elements: a server, a terminal, and a user.

[2586] User Authentication Process

[2587] When a user logs in to a terminal, they enter their company ID and password to confirm their authorization to access the system. This authentication process is carried out securely using the HTTPS protocol. The server checks the authentication information against a database, and if it matches, it generates a session ID and returns it to the terminal.

[2588] Know-how input process

[2589] When a user inputs new knowledge or schemes, they use a dedicated form on their device. This data is sent in JSON format to the server, which then inputs it into a generative AI model for analysis. The analysis results are stored in a database and shared across the organization. For example, if a user inputs "We recommend developing using Scrum for Project B," the data is preprocessed on the server and analyzed by a generative AI model (e.g., OpenAI GPT-3).

[2590] Know-how search process

[2591] When a user searches for existing knowledge or schemes, they enter a query into a search form on their device. The query is sent in JSON format to the server, which then analyzes the query's intent using natural language processing. Based on the analysis results, the server retrieves relevant information from the database and provides it to the user. For example, if a user searches for "best practices in agile development," the server provides the user with relevant information based on this query.

[2592] User emotion recognition process

[2593] When a user enters knowledge or browses search results, the device's built-in camera and microphone collect facial and voice data in real time, which is then input into the emotion analysis algorithm. The server receives this emotion data, collects it as feedback, and uses it to retrain the AI ​​model. For example, if a user is confused while entering know-how, the emotion engine will identify that situation and send it to the server.

[2594] Feedback Providing Process

[2595] When users provide feedback on the information obtained by the system, their emotional data is collected along with the feedback. The server analyzes the feedback and emotional data and stores it as retraining data to improve the accuracy and performance of the AI ​​model.

[2596] This system allows users to effectively input and search knowledge within an organization and reflects feedback based on emotional data. For example, by inputting important know-how about a project and reusing that know-how in other projects, the efficiency of the entire organization can be improved. Emotion recognition is also expected to improve usability.

[2597] The flow of the identification process in the second embodiment will be described with reference to FIG.

[2598] Step 1:

[2599] The user enters their company ID and password into the terminal.

[2600] (Input): User ID, password

[2601] (Operation): The user enters their company ID and password on the device's login screen.

[2602] (Output): The device collects the ID and password.

[2603] Step 2:

[2604] The device sends the collected authentication information to the server using the HTTPS protocol.

[2605] (Input): User ID, password

[2606] (Operation): The device converts the ID and password into JSON format and sends it as an HTTPS request to the server's authentication API endpoint.

[2607] (Output): The server receives the authentication information.

[2608] Step 3:

[2609] The server compares the received authentication information with a database and authenticates the user.

[2610] (Input): Authentication information (user ID, password)

[2611] (What happens): The server compares the user information in its database with the entered authentication information.

[2612] (Output): Authentication result (success / failure)

[2613] Step 4:

[2614] When authentication is successful, the server generates a session ID and returns it to the terminal.

[2615] (Input): Authentication information (if successful)

[2616] (Operation): The server generates a unique session ID and sends it to the device in JSON format.

[2617] (Output): Session ID

[2618] Step 5:

[2619] The terminal receives the session ID and displays a "Login successful" message to the user.

[2620] (Input): Session ID

[2621] (Action): The device saves the session ID in local storage and displays a login success message.

[2622] (Output): User interface update (login success message)

[2623] Step 6:

[2624] The user inputs new knowledge or scheme into an input form.

[2625] (Input): Know-how and scheme (text data)

[2626] (Operation): The user inputs new knowledge or scheme into a dedicated form on the terminal.

[2627] (Output): The terminal collects input data.

[2628] Step 7:

[2629] The input data collected by the terminal is sent to the server in JSON format.

[2630] (Input): Know-how and scheme (JSON format)

[2631] (Operation): The terminal converts the input data into JSON format and sends it to the server as an HTTPS request.

[2632] (Output): The server receives the data.

[2633] Step 8:

[2634] The server preprocesses the data and inputs it into a generative AI model for analysis.

[2635] (Input): Know-how and scheme (JSON format)

[2636] (Operation): The server normalizes the data (standardizing uppercase and lowercase letters, removing unnecessary spaces) and inputs it into a generative AI model. The generative AI model analyzes the data and generates new knowledge and related information.

[2637] (Output): Analysis results (new knowledge, related information)

[2638] Step 9:

[2639] The server stores the analysis results in a database.

[2640] (Input): Analysis results (new knowledge, related information)

[2641] (Operation): The server adds the analysis results to the database as a new record, and also records the save date and time and the ID of the person who entered the data.

[2642] (Output): Update the database (save new knowledge)

[2643] Step 10:

[2644] A user enters a query into a search form.

[2645] (Input): Search query (text data)

[2646] Action: A user enters a query into a search form on their device.

[2647] (Output): The terminal collects input data.

[2648] Step 11:

[2649] The device sends the entered query in JSON format to the server.

[2650] (Input): Search query (JSON format)

[2651] (Operation): The terminal converts the input data into JSON format and sends it to the server as an HTTPS request.

[2652] (Output): The server receives the data.

[2653] Step 12:

[2654] The server parses the query and performs a search against the database.

[2655] (Input): Search query (JSON format)

[2656] (Operation): The server analyzes the query using a natural language processing algorithm and sends the search query to the database.

[2657] (Output): Search results (related knowledge information)

[2658] Step 13:

[2659] The server returns the search results to the device in JSON format.

[2660] (Input): Search results (JSON format)

[2661] (Operation): The server encodes the search results into JSON format and sends them to the device as an HTTPS request.

[2662] (Output): The device receives the data.

[2663] Step 14:

[2664] Your device will display the search results.

[2665] (Input): Search results (JSON format)

[2666] (Operation): The terminal displays the search results received on the user interface, making it easy for the user to find the information they need.

[2667] (Output): Update the user interface (display search results)

[2668] Step 15:

[2669] While the user is entering know-how or viewing search results, the device's built-in camera and microphone collect facial expressions and voice data.

[2670] (Input): facial expression data, voice data

[2671] (Operation): The device uses its built-in camera and microphone to collect facial and voice data in real time.

[2672] (Output): Collected emotion data

[2673] Step 16:

[2674] The device analyzes the emotional data collected in real time to estimate the user's emotional state.

[2675] (Input): Collected emotion data

[2676] (Action): The device uses an emotion analysis algorithm to estimate the emotional state (e.g., happy, unhappy, confused).

[2677] (Output): Estimated emotional state

[2678] Step 17:

[2679] The device transmits the estimated emotion data to the server.

[2680] (Input): Estimated emotional state

[2681] (Operation): The device converts the emotion data into JSON format and sends it to the server as an HTTPS request.

[2682] (Output): The server receives the emotion data.

[2683] Step 18:

[2684] The server collects the emotion data as feedback data and uses it to retrain the model.

[2685] (Input): Emotion data (JSON format)

[2686] (How it works): The server stores the emotion data in a feedback database and uses it to retrain the AI ​​model.

[2687] (Output): Update the database and save the retraining data.

[2688] Step 19:

[2689] The server runs a retraining process at regular intervals to improve the model's performance.

[2690] (Input): Retraining data

[2691] (Operation): The server periodically runs a retraining process to improve the accuracy and performance of the AI ​​model.

[2692] (Output): Improved AI model

[2693] Step 20:

[2694] The server applies the improved model to the system and reflects it in the next user search or know-how input.

[2695] (Input): Improved AI model

[2696] (Operation): The server deploys the updated AI model to the system and applies it the next time the user searches or types something.

[2697] (Output): System update (application of improved AI model)

[2698] The above are the specific processing steps of the program of this system.

[2699] (Application example 2)

[2700] 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."

[2701] Improving work efficiency and sharing knowledge within factories are important issues, but conventional systems have had difficulty providing real-time support or feedback based on workers' emotions. Furthermore, collecting feedback that reflects emotional states and how to utilize it has also been an unsolved problem.

[2702] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for identifying the emotional state of a user using an emotion recognition engine and collecting and saving that data, means for collecting feedback and providing information using the emotion data, and means for tagging knowledge data analyzed using a generative artificial intelligence model and making it searchable based on relevance. This not only improves work efficiency in the factory but also makes it possible to provide appropriate feedback based on the emotions of workers.

[2703] "Authentication information" refers to information used to verify a user's identity, and is usually comprised of an ID and password.

[2704] "User authentication" refers to the process of verifying whether a user is a person with legitimate authority using the acquired authentication information.

[2705] "Internal knowledge" refers to specialized information and procedures accumulated within an organization.

[2706] A "generative artificial intelligence model" is an artificial intelligence model that has the ability to generate and analyze knowledge and information based on new data.

[2707] A "database" is a system for efficiently storing and managing data.

[2708] "Knowledge information" refers to internal knowledge and procedures stored in a database.

[2709] "Search results" refers to the related knowledge information obtained when searching a database.

[2710] "Feedback data" refers to data collected from users' evaluations and opinions.

[2711] "Model retraining" is the process of using new data to improve the performance of an artificial intelligence model.

[2712] An "emotion recognition engine" is a technology that analyzes a user's facial expressions and voice to identify their emotional state.

[2713] "Emotional state" is information indicating the type and intensity of the user's emotion.

[2714] "Feedback collection" is the process of collecting opinions and sentiment data from users.

[2715] "Information provision" is the process of displaying search results and feedback to users.

[2716] "Tagging" means assigning labels to data that indicate its relevance.

[2717] This invention uses smart glasses and head-mounted displays (HMDs) to build a system that improves work efficiency in factories. The system includes multiple means for user authentication, know-how input, know-how search, emotion recognition, and feedback collection.

[2718] User Authentication

[2719] The user puts on the smart glasses and enters their ID and password. This authentication information is securely sent to the server using the HTTPS protocol. The server compares the received authentication information with a database, and if authentication is successful, it generates a session ID and returns it to the device. The device saves the session ID and displays a message to the user indicating successful login.

[2720] Input of know-how

[2721] When a user wants to input new know-how or procedures, they use a dedicated form on the smart glasses. The device sends the input data in JSON format to the server, which then preprocesses the data and analyzes it using a generative artificial intelligence model. The analysis results are stored in a database. This process efficiently accumulates knowledge and procedures acquired within the factory.

[2722] Search for know-how

[2723] When a user searches for a specific know-how or procedure, they enter a query into the search form on the smart glasses. The device sends the query in JSON format to the server, which uses natural language processing technology to analyze the query's intent. The server then sends the search query to a database to retrieve relevant know-how information. The search results are returned to the device and displayed visually to the user.

[2724] emotion recognition

[2725] When a user enters know-how or browses search results, the emotion engine collects the user's facial expression and voice data. The device analyzes this data in real time and estimates the user's emotional state. The estimated emotion data is sent to the server and stored as feedback data. This emotion data is used to retrain the model and provide feedback on know-how.

[2726] Feedback collection

[2727] When a user provides feedback, the emotion engine automatically recognizes the user's emotional state. The feedback content and emotion data are sent from the device to the server, where they are analyzed and stored as retraining data for the model. A periodic retraining...

Claims

1. a means for obtaining authentication information; means for performing user authentication based on the authentication information; A means to input company-specific know-how and schemes, A means for inputting the input data into a generative artificial intelligence model and performing analysis; a means for storing the analysis results in a database; A means for searching know-how information from the database; means for providing the search results to a user; A means of collecting feedback data and using it to retrain the model A system including:

2. The system of claim 1 further comprising means for securely obtaining and transmitting authentication information.

3. 2. The system of claim 1, further comprising means for tagging the know-how data analyzed using the generative artificial intelligence model to make it searchable based on relevance.

Images