System

Abstract

To provide a system.SOLUTION: The system includes: means for preparing a digital clone of an employee on the basis of digital data showing at least one of knowledge and experience of the employee; means for recognizing the emotion of the user; means for extracting the recognized emotion of the user and a response to input information of the user from the digital clone by using a generation Al model; and means for providing the response to the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Traditional knowledge management systems have difficulty storing employees' knowledge logs and thoughts and quickly retrieving them when needed. It is also difficult to access other people's knowledge 24 hours a day, 365 days a year while taking privacy issues into consideration. [Means for solving the problem]

[0005] This invention digitizes employees' knowledge logs and thoughts, creating digital clones. It then uses AI to extract information and makes it accessible at any time without memorizing past knowledge. Furthermore, while taking privacy issues into consideration, by asking questions to the clones of others, it makes the knowledge of a wide range of experts available 24 hours a day, 365 days a year. This makes it possible to quickly obtain necessary information and significantly improve work efficiency. [Brief explanation of the drawings]

[0006] [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. 2 is a sequence diagram showing a flow of processing in the data processing system according to the first embodiment of the first form example. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Embodiment 1. [Figure 13] FIG. 10 is a sequence diagram showing a processing flow of a data processing system in a second embodiment of the second form example. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Embodiment Example 2. [Figure 15]FIG. 10 is a sequence diagram showing the flow of processing in a data processing system according to a third embodiment of the third embodiment. [Figure 16] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Embodiment 3. [Figure 17] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the first embodiment of the first form example when an emotion engine is combined. [Figure 18] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Form Example 1 when an emotion engine is combined. [Figure 19] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the second embodiment of the second form example when an emotion engine is combined. [Figure 20] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Form Example 2 when an emotion engine is combined. [Figure 21] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the third embodiment of the third form example when an emotion engine is combined. [Figure 22] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Form Example 3 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0007] 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.

[0008] First, the terms used in the following description will be explained.

[0009] 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, the 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), an APU (Accelerated Processing Unit), or a TPU (TENSOR PROCESSING UNIT (registered trademark)).

[0010] 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.

[0011] 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.

[0012] 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.

[0013] 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."

[0014] [First embodiment]

[0015] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0016] 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.

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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.

[0022] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0023] 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.

[0024] 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.

[0025] 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.

[0026] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0027] "Example 1"

[0028] This invention uses specific algorithms and programs to digitize employees' knowledge, logs, and thoughts, accumulating the knowledge, experience, and thoughts that employees gain through their daily work. Specifically, it collects data from the tools and systems that employees use in their work, analyzes and organizes it, and digitizes it.

[0029] "Example 2"

[0030] Next, as a means of creating a digital clone, an AI model is constructed based on the digitized data to mimic the knowledge, experience, and thinking of the employee. Specifically, an AI model that reproduces the knowledge, experience, and thinking of the employee is created by learning the data.

[0031] "Example 3"

[0032] Furthermore, as a means of extracting information using AI, questions and instructions can be given to the created AI model to obtain the necessary information. Specifically, questions such as "What are some successful examples of past projects?" can be asked of the AI ​​model and the answer obtained.

[0033] "Example 4"

[0034] Additionally, as a means of making past knowledge accessible at any time without having to memorize it, the AI ​​model is linked to a database to obtain the necessary information in real time. Specifically, the AI ​​model searches and extracts past data stored in the database and provides it to the user. "Example 5"

[0035] Finally, multiple AI models can be used simultaneously to access the knowledge of diverse experts 24 hours a day, 365 days a year by asking questions to clones of other people. Specifically, questions can be asked to multiple AI models as needed, and the answers from each model can be compared and integrated to obtain the optimal answer.

[0036] The processing flow of each embodiment will be described below.

[0037] "Example 1"

[0038] Step 1: Collect data from the tools and systems employees use to do their jobs. For example, collect project progress and deliverables from work management tools, and message content from communication tools.

[0039] Step 2: Analyze, organize, and digitize the collected data. Specifically, use text analysis and natural language processing techniques to extract useful information from the data and store it in a database.

[0040] "Example 2"

[0041] Step 1: Build an AI model based on the digitized data. Specifically, use the data as learning data and apply machine learning algorithms to train the AI ​​model.

[0042] Step 2: Use the trained AI model to make predictions and analyze new data. For example, use knowledge learned from past data to make predictions and analyze new business data.

[0043] "Example 3"

[0044] Step 1: Ask questions or give instructions to the created AI model. Specifically, the user's question is given as input to the AI ​​model, and the AI ​​model returns the output to the user.

[0045] Step 2: Evaluate the output from the AI ​​model and update it if necessary. For example, if the output from the AI ​​model does not meet user expectations, use that feedback to retrain the AI ​​model.

[0046] "Example 4"

[0047] Step 1: Link the AI ​​model with the database. Specifically, make it possible for the AI ​​model to obtain the data it needs from the database.

[0048] Step 2: The AI ​​model searches and extracts past data stored in the database and provides it to the user. For example, it searches the database for an answer to a user's question and returns the result to the user.

[0049] "Example 5"

[0050] Step 1: Ask multiple AI models questions as needed. Specifically, provide the user's question as input to multiple AI models, and obtain output from each AI model.

[0051] Step 2: Compare and integrate the outputs from each AI model to obtain the optimal answer. For example, compare the outputs from multiple AI models and return the most reliable output to the user.

[0052] Example 1

[0053] Next, a description will be given of Example 1 of Form 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."

[0054] With conventional systems, it was difficult to efficiently digitize and store the knowledge, experience, and thoughts that employees gained in their daily work. There was also a lack of means to properly analyze and organize the stored information and quickly access it when needed. Furthermore, there were also privacy issues and a lack of adequate means to access the knowledge of a variety of experts 24 hours a day, 365 days a year.

[0055] 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.

[0056] In this invention, the server includes means for digitizing employee knowledge, records, and thoughts, means for creating digital clones, means for extracting information using artificial intelligence, means for making past knowledge accessible at any time without memorizing it, means for collecting data from tools and systems used in business, means for analyzing the collected data, means for organizing the analyzed data, means for storing the organized data in a database, and means for providing the stored data. This allows for efficient digitization of employee knowledge, experience, and thoughts, appropriate analysis and organization of accumulated information, and rapid access when needed. Furthermore, while taking privacy issues into consideration, the knowledge of diverse experts can be accessed 24 hours a day, 365 days a year.

[0057] "Means for digitizing employee knowledge, records, and thoughts" refers to a method for storing the knowledge, experience, and thoughts that employees gain in their daily work in electronic form.

[0058] "Means for creating a digital clone" refers to a method for creating a virtual persona based on an employee's knowledge and experience.

[0059] "Means of extracting information using artificial intelligence" refers to a method of extracting necessary information from accumulated data using AI technology.

[0060] "Means for making past knowledge accessible at any time without having to remember it" refers to a method for making information stored in a database searchable and retrievalable in real time.

[0061] "Means of collecting data from tools and systems used in business" refers to methods for obtaining data from software and hardware that employees use on a daily basis.

[0062] "Means for analyzing collected data" refers to methods for analyzing acquired data and extracting meaningful information.

[0063] "Means for organizing analyzed data" refers to methods for classifying the analysis results into categories, tags, etc., and converting them into a format that is easy to search later.

[0064] "Means for storing organized data in a database" refers to a method for storing classified data in a database.

[0065] A "means for providing stored data" is a method by which information stored in a database can be accessed by a user.

[0066] MODE FOR CARRYING OUT THE INVENTION

[0067] This invention is a system that digitizes the knowledge, records, and thoughts of employees and efficiently stores, analyzes, and provides them. A specific embodiment of this system is described below.

[0068] Data collection

[0069] The server collects data from the tools and systems that employees use in their work. Specifically, it obtains data from communication tools (e.g., Slack (registered trademark), MICROSOFT (registered trademark) TEAMS (registered trademark)), office tools (e.g., MICROSOFT OFFICE (registered trademark), GOOGLE WORKSPACE (registered trademark)), project management tools, etc. The server periodically collects data using the APIs of these tools.

[0070] Data analysis

[0071] The server analyzes the collected data. Specifically, it uses the Python (registered trademark) NLTK library and spaCy to analyze the text data. The server tokenizes the message content and extracts important keywords and phrases. For example, the server extracts the keywords "project," "progress," and "report" from the message "Report on the progress of the project."

[0072] Data organization

[0073] The server organizes the analyzed data. Specifically, it categorizes the extracted keywords and phrases. The server organizes the data by category and converts it into a format that makes it easier to search later. For example, it creates categories such as "project management," "communication," and "reporting," and organizes the keywords that correspond to each category.

[0074] Data storage

[0075] The server saves the organized data in a database. Specifically, it saves the data using a database management system (DBMS) such as MySQL (registered trademark) or PostgreSQL. The server connects to the database and inserts the organized data into the appropriate table. For example, it inserts the data into the "knowledge_logs" table.

[0076] Data provision

[0077] The server provides the digitized data in a format that employees can access. Specifically, the data is made accessible through a web application using Django and Flask. Employees can search and view the data by opening a browser and accessing the Django application. For example, if an employee selects the category "Project Management," relevant messages are displayed.

[0078] Examples of concrete examples and prompts

[0079] As a concrete example, consider a case where employees use Slack to report project progress. The server collects message data through the Slack API and analyzes the message content using the Python NLTK library. The analyzed results are stored in a MySQL database and made accessible to employees through a web application using Django.

[0080] Example prompt sentence:

[0081] "Collect Slack message data, parse it using Python's NLTK library, and store it in a MySQL database. Then, make the data accessible through a web application using Django."

[0082] In this way, the knowledge, experience, and thoughts of employees can be efficiently digitized, the accumulated information can be properly analyzed and organized, and it can be quickly accessed when needed. Furthermore, the knowledge of a wide range of experts can be accessed 24 hours a day, 365 days a year, while taking privacy issues into consideration.

[0083] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0084] Step 1: Data collection

[0085] The server collects data from the tools and systems employees use in their work. Specifically, the server obtains message data using Slack's API. The server calls the "https: / / slack.com / api / conversations.history" endpoint at 9:00 AM every day to collect the previous day's message data. The input is the message data obtained from Slack's API, and the output is the collected message data.

[0086] Step 2: Data analysis

[0087] The server analyzes the collected message data. Specifically, it uses Python's NLTK library to parse the text data. The server tokenizes the messages and extracts important words such as nouns and verbs. For example, the server extracts the keywords "project," "progress," and "report" from the message "Report on the progress of the project." The input is the collected message data, and the output is the extracted keywords.

[0088] Step 3: Data organization

[0089] The server organizes the analyzed data. Specifically, the server classifies the extracted keywords into categories. For example, it creates categories such as "project management," "communication," and "reporting," and organizes the keywords corresponding to each category. The input is the extracted keywords, and the output is the keywords organized by category.

[0090] Step 4: Save Data

[0091] The server saves the organized data in a database. Specifically, the server uses a MySQL database to store the data. The server connects to the database and inserts data into the "knowledge_logs" table. For example, it executes the SQL query "INSERT INTO knowledge_logs (category, keyword, message) VALUES ('Project management', 'Progress', 'Report project progress')". The input is keywords organized by category, and the output is the data stored in the database.

[0092] Step 5: Provide data

[0093] The server provides the digitized data in a format that employees can access. Specifically, the server makes the data accessible through a web application using Django. Employees can search and view the data by opening a browser and accessing the Django application. For example, if an employee selects the category "Project Management," relevant messages are displayed. The input is the data stored in the database, and the output is the data provided through the web application.

[0094] (Application example 1)

[0095] Next, a description will be given of Application Example 1 of Embodiment 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."

[0096] In traditional factories, the knowledge and experience of workers is dependent on the individual, and there is a lack of means to efficiently share and utilize this knowledge. It is also difficult to digitize and store the operations and thoughts of workers, making knowledge transfer and work efficiency an issue. Furthermore, privacy issues and the effort required to organize and analyze data are also major obstacles.

[0097] 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.

[0098] In this invention, the server includes a means for digitizing employees' knowledge, logs, and thoughts, a means for creating digital clones, a means for extracting information using AI, a means for making past knowledge accessible at any time without memorizing it, a means for recording workers' operations and adding them to a log with a timestamp, and a means for analyzing and organizing the recorded data and saving it in a database. This allows workers' knowledge and experience to be efficiently digitized and shared, improving work efficiency and enabling the succession of knowledge.

[0099] "Means for digitizing employee knowledge logs and thoughts" refers to a method for collecting, analyzing, organizing, and digitizing the knowledge, experience, and thoughts that employees gain in their daily work.

[0100] "Methods for creating digital clones" refers to the creation of a virtual entity that digitally mimics the actions and thoughts of employees based on their knowledge logs and thoughts.

[0101] "Methods of extracting information using AI" refers to a method of using artificial intelligence technology to search and extract necessary information from digital clones and accumulated data.

[0102] "Means for making past knowledge accessible at any time without having to remember it" refers to a method for making knowledge and information stored in a database instantly accessible when needed, without relying on a specific storage device.

[0103] The "means for recording the operations of the worker and adding them to the log together with a timestamp" is a method for recording the operations performed by the worker in real time and adding them to the log together with a timestamp indicating the date and time when the operation was performed.

[0104] "Means for analyzing and organizing the recorded data and storing it in a database" refers to a method for analyzing and organizing the recorded operation data and then storing it in a database.

[0105] The system for implementing this invention includes a set of means for digitizing employees' knowledge logs and thoughts, creating digital clones, and extracting information using AI. A specific embodiment of this system is shown below.

[0106] System Configuration

[0107] The system consists of the following main components:

[0108] 1. Data Collection Module: Collects data from the tools and systems employees use to do their jobs.

[0109] 2. Digitization module: Analyzes, organizes, and digitizes the collected data.

[0110] 3. Digital clone generation module: Generates a digital clone based on the digitized data.

[0111] 4. AI information extraction module: Uses AI technology to extract necessary information from digital clones and databases.

[0112] 5. Logging module: Records the worker's actions and adds them to the log with a timestamp.

[0113] 6. Database management module: Analyzes and organizes the recorded data and stores it in a database.

[0114] Hardware and software used

[0115] Hardware: Factory robots, tablets and smartphones used by workers

[0116] Software: Python, JSON library, database management system (e.g., MySQL)

[0117] Data processing and calculation

[0118] 1. Data collection: The data collection module collects data in real time from worker operations and business tools.

[0119] 2. Digitization: The digitization module analyzes, organizes, and digitizes the collected data. For example, when an operator changes a machine's settings, the operation details and settings are recorded.

[0120] 3. Digital Clone Generation: The digital clone generation module generates a digital clone that mimics the employee's behavior and thoughts based on the digitized data.

[0121] 4. AI Information Extraction: The AI ​​information extraction module uses AI technology to search and extract the necessary information from the generated digital clones and databases.

[0122] 5. Logging: The logging module records the worker's actions with a timestamp and adds them to the log.

[0123] 6. Database Management: The database management module analyzes and organizes the recorded data and stores it in a database.

[0124] Specific examples

[0125] For example, if an operator changes the speed setting of a factory robot, the operation details and setting value are recorded in real time and added to the log along with a timestamp. The recorded data is analyzed and organized by the digitization module and stored in a database. Later, the AI ​​information extraction module can be used to search and extract past operation history and setting values.

[0126] Prompt Sentence Examples

[0127] We are developing a system in which a factory robot records the operations of a worker. When a worker changes the robot's speed settings, please create a Python program to record the operation and the setting value. The recorded data needs to be saved in JSON format.

[0128] In this way, the system efficiently digitizes and shares workers' knowledge and experience, improving work efficiency and ensuring knowledge transfer.

[0129] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0130] Step 1:

[0131] Data collection

[0132] The server collects data in real time from the tools and systems that workers use in their work. Specifically, when a worker operates a factory robot, the details of that operation (e.g., changing the speed setting) are obtained from sensors and log files. The input is the worker's operation data, and the output is the collected raw data.

[0133] Step 2:

[0134] Digitalization

[0135] The server analyzes and organizes the collected raw data using a digitization module. Specifically, it analyzes the collected operation data, extracts the operation details and setting values, and converts them into digital format. The input is the collected raw data, and the output is analyzed and organized digital data.

[0136] Step 3:

[0137] Digital Cloning

[0138] The server generates a digital clone based on the digitized data. Specifically, it uses the digitized operational data to generate a virtual entity that mimics the worker's actions and thoughts. The input is the digitized data, and the output is the generated digital clone.

[0139] Step 4:

[0140] AI Information Extraction

[0141] The server uses an AI information extraction module to search and extract the necessary information from the digital clone and database. Specifically, it uses an AI model to search past operation history and setting values ​​and extract the necessary information. The input is the search query, and the output is the extracted information.

[0142] Step 5:

[0143] Logging

[0144] The server records the worker's operations along with a timestamp and adds them to the log. Specifically, it records the operations performed by the worker and the date and time, and adds them to a log file. The input is the worker's operation data and timestamp, and the output is an updated log file.

[0145] Step 6:

[0146] Database Management

[0147] The server analyzes and organizes the recorded data and saves it in a database. Specifically, it analyzes and organizes the data recorded by the log recording module and saves it in a database. The input is the analyzed and organized data, and the output is an updated database.

[0148] In this way, the system efficiently digitizes and shares workers' knowledge and experience, improving work efficiency and ensuring knowledge transfer.

[0149] Example 2

[0150] Next, a description will be given of Example 2 of Form 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."

[0151] In modern companies, it is important to efficiently share and utilize the knowledge and experience of employees. However, employees' knowledge and experience are accumulated individually, making it difficult to share this throughout the organization. There is also a risk that this knowledge and experience will be lost when an employee leaves the company. Furthermore, there is a need for a method to utilize employees' knowledge while taking privacy issues into consideration. To solve these issues, a system is needed that digitizes employees' knowledge and experience and provides it in a form that can be accessed at any time.

[0152] 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. In this invention, the server includes a means for digitizing the knowledge, history, and thoughts of employees, a means for creating a digital clone, a means for extracting information using a machine learning model, a means for making past knowledge accessible at any time without storing it, a means for collecting and preprocessing data, training an AI model, and evaluating and adjusting it, and a means for inputting prompt sentences into the generative AI model to utilize the digital clone function. This makes it possible to efficiently share and utilize the knowledge and experience of employees.

[0153] "Means for digitizing employee knowledge, history, and thinking" refers to a method for collecting and storing employees' work-related knowledge, past work history, and thought processes as digital data.

[0154] "Means for creating a digital clone" refers to a method for building an AI model that mimics the behavior and judgment of employees based on the knowledge, experience, and thoughts of digitized employees.

[0155] "Methods of extracting information using machine learning models" are methods that use trained AI models to provide appropriate information in response to specific questions or requests.

[0156] "Means for making past knowledge accessible at any time without having to remember it" refers to a method that allows information based on past knowledge and experience to be obtained in real time through a digital clone.

[0157] "Means for collecting, preprocessing, training an AI model, and evaluating and adjusting data" refers to a series of processes for collecting the data necessary to create a digital clone, converting that data into a format suitable for an AI model, training the model, evaluating its performance, and adjusting it as necessary.

[0158] "Means for utilizing the functions of a digital clone by inputting prompt sentences into a generative AI model" refers to a method in which a user inputs a specific question or request into an AI model as a prompt sentence, and as a result receives appropriate information or advice from the digital clone.

[0159] MODE FOR CARRYING OUT THE INVENTION

[0160] This invention relates to a system that digitizes the knowledge and experience of employees and creates digital clones. This system includes elements such as a server, a terminal, and a user, and operates as follows.

[0161] 1. Data Collection

[0162] The server collects digitized data such as employee work history, emails, chat logs, documents, etc. This includes retrieving data from databases and cloud storage. Specifically, the server retrieves work history from databases using SQL queries and downloads documents from cloud storage via APIs.

[0163] 2. Data Preprocessing

[0164] The server preprocesses the collected data. Specifically, it cleans the data (removes noise and missing values), normalizes it (transforms it to maintain data consistency), and tokenizes it (divides the text data into words and phrases). The server uses Python's pandas library to create a data frame and imputes missing values. It also tokenizes the text data using the NLTK library.

[0165] 3. Training the AI ​​model

[0166] The server uses the preprocessed data to train an AI model using a machine learning framework such as TENSORFLOW (registered trademark) or PyTorch. The server defines a neural network model using TensorFlow's Keras API and trains the model using the preprocessed data as input.

[0167] 4. Evaluate and adjust the model

[0168] The server evaluates the trained AI model and makes adjustments as necessary. The evaluation checks the model's accuracy using actual work data from employees. The server calculates the model's evaluation indicators (e.g., precision, recall, F1 score) and adjusts the hyperparameters based on the results. The hyperparameter optimization library is used to optimize the hyperparameters.

[0169] 5. Creating a Digital Clone

[0170] The server then uses the final, adjusted AI model to generate a digital clone that can mimic the employee's knowledge, experience, and thinking. The server then stores the trained model and makes it accessible to users via an API.

[0171] 6. Use of digital clones

[0172] Users can use the digital clone's functions by entering prompts from their terminal. Specifically, when creating a proposal for a new project, the user can ask the digital clone for advice. For example, the user can enter the following prompt:

[0173] Example prompt sentence:

[0174] "Can you give me some examples of successful proposals from past projects? Also, can you give me some examples of unsuccessful proposals?"

[0175] The server receives the input prompt and passes it to the generative AI model. The generative AI model extracts the characteristics of successful proposals and the problems of unsuccessful proposals based on past data, and generates an answer. The server returns the generated answer to the user's device. The user can check the answer on their device and use it as a reference when creating a new proposal.

[0176] In this way, it becomes possible to efficiently share and utilize the knowledge and experience of employees.

[0177] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0178] Step 1: Data collection

[0179] The server collects digitized data such as employee work history, emails, chat logs, documents, etc. Specifically, the server retrieves work history from the database using SQL queries and downloads documents from cloud storage via API. The input is the connection information for the database and cloud storage, and the output is the collected raw data.

[0180] Step 2: Data Preprocessing

[0181] The server preprocesses the collected data. Specifically, it cleans the data (removes noise and missing values), normalizes it (transforms it to maintain data consistency), and tokenizes it (divides the text data into words and phrases). The input is the collected raw data, and the output is the preprocessed data. The server creates a data frame using Python's pandas library and completes missing values. It also tokenizes the text data using the NLTK library.

[0182] Step 3: Training the AI ​​model

[0183] The server uses the preprocessed data to train an AI model. Specifically, it uses machine learning frameworks such as TensorFlow and PyTorch. The input is the preprocessed data, and the output is a trained AI model. The server defines a neural network model using TensorFlow's Keras API and trains the model using the preprocessed data as input.

[0184] Step 4: Evaluate and tune the model

[0185] The server evaluates the trained AI model and makes adjustments as necessary. Specifically, it calculates the model's evaluation indicators (e.g., precision, recall, F1 score) and adjusts the hyperparameters based on the results. The input is the trained AI model and evaluation data, and the output is the adjusted AI model. The server optimizes the hyperparameters using a hyperparameter optimization library.

[0186] Step 5: Create a digital clone

[0187] The server generates a digital clone using the final adjusted AI model. The input is the adjusted AI model, and the output is the digital clone. The server stores the trained model and makes it accessible to users through an API.

[0188] Step 6: Use a digital clone

[0189] Users utilize the digital clone's functions by inputting prompts from their device. Specifically, when creating a proposal for a new project, the user can ask the digital clone for advice. The input is the prompt, and the output is the answer from the digital clone. For example, a user might input the following prompt: "Please give me some examples of successful proposals from past projects. Also, please give me some examples of unsuccessful proposals." The server receives the input prompt and passes it to the generative AI model. The generative AI model uses past data to extract the characteristics of successful proposals and the problems of unsuccessful proposals, and generates an answer. The server then sends the generated answer back to the user's device. The user can check the answer on their device and use it as a reference when creating a new proposal.

[0190] (Application example 2)

[0191] Next, a description will be given of Application Example 2 of Form 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."

[0192] In conventional factories, it was difficult to transfer the knowledge and experience of skilled workers to new workers or robots, making it difficult to automate tasks efficiently. There was also a risk that important work would be delayed if skilled workers retired or went on vacation. This led to problems with reduced productivity and quality in factories. Furthermore, there were issues with privacy and information accessibility.

[0193] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0194] In this invention, the server includes a means for digitizing employees' knowledge logs and thoughts, a means for creating digital clones, a means for extracting information using AI, a means for making past knowledge accessible at any time without memorizing it, a means for learning factory work data, a means for being installed in factory robots, a means for imitating the knowledge and experience of skilled workers, and a means for automatically performing complex tasks. This allows the knowledge and experience of skilled workers to be transferred to robots, enabling efficient automation of work. Furthermore, it improves information accessibility while taking privacy issues into consideration.

[0195] "Means for digitizing employee knowledge logs and thinking" refers to technology for recording and storing employees' knowledge, experiences, and thought processes as digital data.

[0196] "Means for creating digital clones" refers to technology that builds AI models that mimic employees' knowledge, experience, and thinking based on digitized data.

[0197] "Methods of extracting information using AI" refers to technology that uses artificial intelligence to search for and obtain the necessary information from a digital clone.

[0198] "Means for making past knowledge accessible at any time without storing it" refers to technology that allows knowledge and information stored in a digital clone to be accessed instantly when needed without being stored in a specific storage device.

[0199] "Means for learning factory work data" refers to technology that collects data on work performed in factories and trains an AI model using that data.

[0200] "Means to be installed on factory robots" refers to technology for incorporating trained AI models into robots used in factories.

[0201] "Means of imitating the knowledge and experience of skilled workers" refers to technology that uses an AI model to reproduce the advanced skills and experience of skilled workers and implement them in robots.

[0202] "Means for automating complex tasks" refers to technology that uses AI models to enable factory robots to automatically perform complex tasks in the same way as skilled human workers.

[0203] The system for implementing this invention includes a means for digitizing employees' knowledge logs and thoughts and creating digital clones. It also includes a means for extracting information using AI and making it accessible at any time without memorizing past knowledge. This allows factory work data to be learned and installed in factory robots, enabling them to imitate the knowledge and experience of skilled workers and perform complex tasks automatically.

[0204] System Program

[0205] The system's program uses Python to build an AI model using TensorFlow and Keras, and trains it on data from skilled workers. The server loads the data, builds the model, compiles the model, trains the model, and saves the model.

[0206] Processing Description

[0207] The server first loads digitized data on skilled workers' knowledge and experience. This data includes detailed logs of work performed in the factory and sensor data. Next, TensorFlow and Keras are used to build an AI model with LSTM and Dense layers. This model is designed to effectively process time-series data and learn the work patterns of skilled workers.

[0208] After the model is built, the server compiles it using the Adam optimizer and categorical cross-entropy loss function, which improves the model's training efficiency and accuracy. The server then trains the model using the data. Once the model is trained, it is installed on the factory robot.

[0209] Factory robots use these trained models to mimic the knowledge and experience of skilled workers and perform complex tasks automatically, helping maintain factory productivity and quality even when skilled workers are unavailable.

[0210] Specific examples

[0211] As a concrete example, this system can be installed on robots used on factory assembly lines. The robots can then act as digital clones of skilled workers, automatically performing complex assembly tasks, such as installing specific parts or carrying out quality inspections, exactly as skilled workers would.

[0212] Example prompts to input to a generative AI model:

[0213] "Build an AI model that automates factory assembly work based on data from skilled workers. Specifically, create a model that learns from the work data of skilled workers using LSTM and Dense layers, and install it on factory robots."

[0214] In this way, factory robots can utilize the knowledge and experience of skilled workers to perform advanced tasks.

[0215] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0216] Step 1:

[0217] The server reads the digitized data of skilled workers' knowledge and experience. Specifically, it collects detailed logs and sensor data of work performed in the factory and stores this in a database. The input is the factory work data, and the output is the digitized data stored in the database.

[0218] Step 2:

[0219] The server builds an AI model using TensorFlow and Keras. Specifically, it designs a neural network with LSTM and Dense layers and defines the model architecture. The input is the digitized data stored in Step 1, and the output is the built AI model.

[0220] Step 3:

[0221] The server compiles the model using the Adam optimizer and categorical cross-entropy loss function, which improves the model's learning efficiency and accuracy. The input is the AI ​​model built in step 2, and the output is the compiled AI model.

[0222] Step 4:

[0223] The server uses the digitized data to train the AI ​​model. Specifically, it inputs the data into the model and sets the number of epochs and batch size to perform training. The input is the digitized data saved in step 1 and the AI ​​model compiled in step 3, and the output is the trained AI model.

[0224] Step 5:

[0225] The server saves the trained AI model. Specifically, it saves the model as a file so that it can be installed on the factory robot later. The input is the AI ​​model trained in step 4, and the output is the saved AI model file.

[0226] Step 6:

[0227] The server installs the saved AI model on the factory robot. Specifically, it incorporates the model into the robot's control system, allowing the robot to utilize the learned knowledge and experience. The input is the AI ​​model file saved in step 5, and the output is a factory robot with the installed AI model.

[0228] Step 7:

[0229] Factory robots use installed AI models to mimic the knowledge and experience of skilled workers and automatically perform complex tasks, such as assembly and quality inspection. The input is factory work instructions and sensor data, and the output is the results of the work performed.

[0230] Example 3

[0231] Next, a third embodiment of the third embodiment will be described. 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."

[0232] In modern companies, it is important to efficiently utilize the knowledge and experience of employees, but there is a lack of means to achieve this. There is also a need for methods to make past knowledge accessible in real time and to make the knowledge of diverse experts available 24 hours a day, 365 days a year. Furthermore, a system is needed that can simultaneously use multiple generative AI models to obtain information while taking privacy issues into consideration.

[0233] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0234] In this invention, the server includes means for digitizing employee knowledge, records, and thoughts, means for creating digital clones, means for extracting information using a generative AI model, means for making past knowledge accessible at any time without storing it, means for acquiring information by simultaneously using multiple generative AI models, means for acquiring information in real time by linking with a database, and means for considering privacy issues. This makes it possible to efficiently utilize employee knowledge and experience, acquire necessary information in real time, and access the knowledge of a variety of experts 24 hours a day, 365 days a year.

[0235] "Means for digitizing employee knowledge, records, and thoughts" refers to technology for storing and managing employees' knowledge, experiences, and thought processes in digital format.

[0236] "Means for creating a digital clone" refers to technology that uses an employee's knowledge and experience to digitally generate a virtual entity that mimics the employee's thoughts and actions.

[0237] "Means of extracting information using a generative AI model" refers to technology for asking questions or giving instructions to a generative AI model to obtain the necessary information.

[0238] "A means of making past knowledge accessible at any time without having to remember it" is a technology that searches and extracts past data in real time and provides the necessary information instantly.

[0239] "Means of obtaining information by simultaneously using multiple generative AI models" refers to a technology that simultaneously asks questions to multiple generative AI models, compares and integrates their responses, and obtains the optimal answer.

[0240] "Means of obtaining information in real time in collaboration with a database" refers to a technology that searches and extracts data stored in a database in real time, and uses that data to generate answers using a generative AI model.

[0241] "Privacy-conscious measures" are technologies that ensure the protection of personal and confidential information when using digital clones and generative AI models.

[0242] MODE FOR CARRYING OUT THE INVENTION

[0243] This invention is a system that efficiently utilizes the knowledge and experience of employees, acquires necessary information in real time, and utilizes the knowledge of various experts 24 hours a day, 365 days a year. This system is implemented using the following hardware and software.

[0244] Hardware and software used

[0245] Server: Uses a server equipped with a powerful processor and large memory capacity to run generative AI models, access databases, and process user requests.

[0246] Terminal: A device used by a user, such as a computer or smartphone. The terminal receives user input and communicates with the server.

[0247] Generative AI model: An AI model for natural language processing, such as an advanced generative AI model like GPT-3 (registered trademark) or BERT.

[0248] Database: A database for storing historical knowledge and data. Use a SQL database or a NoSQL database.

[0249] Specific operation of the system

[0250] 1. The user enters a question

[0251] The user uses the device's input interface to enter questions or instructions for the generative AI model, such as "What are some successful examples from past projects?"

[0252] 2. The device sends a question to the server

[0253] The terminal sends the question entered by the user to the server using an HTTP request.

[0254] 3. The server passes the question to the AI ​​model

[0255] The server passes the received question to a generative AI model, which searches its internal knowledge base and generates an appropriate answer.

[0256] 4. The server sends the answer back to the device

[0257] The server receives the generated answer and sends it back to the user's device using an HTTP response.

[0258] 5. The device will display the answer

[0259] The terminal displays the response received from the server to the user using a GUI (Graphical User Interface).

[0260] Examples of specific examples and prompts

[0261] Example: A user asks, "What are some success stories from past projects?"

[0262] Example prompt: "Tell me about a success story from a past project."

[0263] This system allows users to efficiently obtain the necessary information and support decision-making. It is also possible to obtain information in real time by linking with a database, or to obtain information by simultaneously using multiple generative AI models. This makes it possible to make the most of employees' knowledge and experience and improve the competitiveness of the company. The flow of the specific processing in Example 3 will be described with reference to FIG. 15.

[0264] Step 1:

[0265] The user enters a question.

[0266] The user uses the device's input interface to input questions or instructions to the generative AI model. For example, they might type, "What are some successful examples of past projects?" The input question is saved as text data on the device.

[0267] Step 2:

[0268] The terminal sends a question to the server.

[0269] The terminal sends the question entered by the user to the server as an HTTP request. The data sent is the user's question text. The server receives this request and analyzes the question text.

[0270] Step 3:

[0271] The server passes the question to the AI ​​model.

[0272] The server passes the received question text to the generative AI model, which searches its internal knowledge base and generates an appropriate answer. Specifically, the AI ​​model uses natural language processing algorithms to analyze the question and extract relevant information. The output is the generated answer text.

[0273] Step 4:

[0274] The server sends a response back to the terminal.

[0275] The server receives the generated answer text and returns it to the user's device as an HTTP response. The returned data is the answer text generated by the AI ​​model. The device receives this response and analyzes the answer text.

[0276] Step 5:

[0277] The device will display the answer.

[0278] The terminal displays the answer text received from the server to the user. A GUI (Graphical User Interface) is used for display. Specifically, the answer text is displayed on the screen so that the user can check it.

[0279] (Application example 3)

[0280] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0281] Modern manufacturing requires rapid and accurate responses to problems that occur during the manufacturing process. However, with conventional systems, it is difficult to acquire past knowledge and success stories in real time, and it is also difficult to integrate solutions from multiple perspectives. Furthermore, there is a lack of systems that allow operators to operate hands-free, making it difficult to achieve efficient manufacturing support.

[0282] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes means for digitizing employee knowledge logs and thoughts, means for creating digital clones, means for extracting information using AI, means for making past knowledge accessible at any time without memorizing it, means for acquiring information in real time about problems that occur during the manufacturing process and providing optimal manufacturing procedures and troubleshooting methods, means for using multiple AI models to provide solutions from different perspectives and select the optimal method, and means for responding to voice questions and instructions and enabling hands-free operation. This enables quick and accurate responses to problems that occur during the manufacturing process and enables efficient manufacturing support.

[0283] "Means for digitizing employee knowledge logs and thoughts" is a technology for recording and storing employees' knowledge, experiences, and ideas in digital form.

[0284] "Methods for creating digital clones" are technologies that generate digital models that mimic the knowledge and thought patterns of a particular person.

[0285] "Means of extracting information using AI" refers to technology that uses artificial intelligence to search for and obtain the necessary information.

[0286] "A means of making past knowledge accessible at any time without having to remember it" is technology that allows past data to be searched and retrieved in real time.

[0287] "A means of obtaining real-time information on problems that occur during the manufacturing process and providing optimal manufacturing procedures and troubleshooting methods" refers to technology that instantly detects problems that occur during manufacturing and provides solutions.

[0288] "A means of using multiple AI models to provide solutions from different perspectives and select the optimal method" is a technology that uses multiple artificial intelligence models simultaneously to compare and integrate solutions from different approaches.

[0289] "Means for responding to voice inquiries and instructions and enabling hands-free operation" refers to technology that uses voice recognition technology to enable users to operate the system without using their hands.

[0290] A system for implementing this invention has the following configuration: a server includes means for digitizing employee knowledge logs and thoughts, means for creating digital clones, means for retrieving information using AI, means for making past knowledge accessible at any time without storing it, means for obtaining information in real time about problems that occur during the manufacturing process and providing optimal manufacturing procedures and troubleshooting methods, means for using multiple AI models to provide solutions from different perspectives and select the optimal method, and means for responding to voice questions and instructions and enabling hands-free operation.

[0291] Program processing explanation

[0292] The server runs the program using Python. It uses SQLite as the database and the OpenAI (registered trademark) API for the AI ​​model. Specifically, it performs the following processes:

[0293] 1. How to digitize employee knowledge logs and thinking:

[0294] The server digitally stores the knowledge and experience entered by employees, which can be entered by text or voice.

[0295] 2. How to create a digital clone:

[0296] The server uses the collected digital data to create a digital clone that mimics the knowledge and thought patterns of a specific employee, allowing that knowledge to be used even when the employee is absent.

[0297] 3. How to use AI to extract information:

[0298] The server receives questions from users and uses the AI ​​model to search for and retrieve the necessary information. For example, if a user asks, "What are some successful examples of past projects?", the AI ​​model will pull relevant information from the database.

[0299] 4. A way to make past knowledge accessible at any time without having to memorize it:

[0300] The server allows users to search and retrieve historical data in real time, allowing them to instantly obtain the information they need.

[0301] 5. A means of obtaining real-time information about problems that occur during the manufacturing process and providing optimal manufacturing procedures and troubleshooting methods:

[0302] The server detects problems that occur on the production line and uses databases and AI models to provide optimal solutions. For example, if you ask, "What caused the production line to stop and what should be done?", it will suggest solutions based on past data and AI models.

[0303] 6. Using multiple AI models to provide solutions from different perspectives and select the best method:

[0304] The server uses multiple AI models simultaneously to compare and integrate solutions from different perspectives, allowing it to select the optimal solution.

[0305] 7. Means of responding to voice prompts and commands, enabling hands-free operation:

[0306] The server utilizes voice recognition technology to allow users to operate the system hands-free, improving efficiency on the manufacturing floor.

[0307] Examples of specific examples and prompts

[0308] For example, if a production line at a factory stops, an operator can ask, "What caused the production line to stop and what are the countermeasures?" The server will search a past database and provide relevant success stories and countermeasures. If the information is not available in the database, the server will query the AI ​​model and provide the optimal solution.

[0309] Example prompt sentence:

[0310] Prompt: "What caused the production line to stop and what was the solution?"

[0311] In this way, the server can provide optimal manufacturing support in real time.

[0312] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0313] Step 1:

[0314] A user inputs a question using a terminal. For example, "What is the cause of the production line stopping and what is the solution?" This input is sent to the server in text format.

[0315] Step 2:

[0316] The server analyzes the received question and extracts relevant keywords using natural language processing technology. For example, keywords such as "production line," "stop," "cause," and "measures" are extracted.

[0317] Step 3:

[0318] The server searches an SQLite database based on the extracted keywords. The database contains data on past success stories and troubleshooting. The relevant data is retrieved as search results.

[0319] Step 4:

[0320] The server provides the user with information retrieved from the database, such as "A similar problem occurred in a past project and the following measures were effective."

[0321] Step 5:

[0322] If the database does not contain relevant information, the server uses OpenAI's API to ask the AI ​​model a question. The question is generated as a prompt and sent to the AI ​​model. For example, a prompt might be generated: "What caused the production line to stop, and what measures can be taken?"

[0323] Step 6:

[0324] The AI ​​model generates an answer based on the received prompt. This answer is returned to the server in text format. For example, the generated answer might be, "The cause of the production line stoppage may be a machine failure or a problem with the power supply. As a countermeasure, it is necessary to inspect the machine and check the power supply."

[0325] Step 7:

[0326] The server provides the answer from the AI ​​model to the user, who can view this information through their device.

[0327] Step 8:

[0328] When the user speaks a question, the device uses speech recognition technology to convert the speech into text, which is then sent to the server for processing from step 2 onwards.

[0329] Step 9:

[0330] The server uses multiple AI models to provide solutions from different perspectives. For example, it asks the same question to different AI models, compares and integrates their answers, and provides this integrated answer to the user.

[0331] Step 10:

[0332] The server collects feedback from users and uses it to improve the system. For example, users can rate the solutions provided, and the server updates the AI ​​model and database based on that evaluation data.

[0333] 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.

[0334] "Example 1"

[0335] One embodiment of the present invention is a system incorporating an emotion engine. This system recognizes a user's emotions and adjusts the digital clone's responses based on those emotions. Specifically, when a user expresses an emotion such as happiness or sadness, the emotion engine recognizes the emotion and provides a response appropriate to that emotion to the digital clone. For example, if the user expresses happiness, the digital clone will provide a positive response that is tailored to the happiness.

[0336] "Example 2"

[0337] The emotion engine also adjusts the questions posed to the other person's clone based on the user's emotions. Specifically, if the user expresses emotions such as anger or frustration, the emotion engine recognizes the emotion and adjusts the questions posed to the other person's clone. For example, if the user expresses anger, the questions posed to the other person's clone are adjusted to provide information or advice to alleviate the anger.

[0338] "Example 3"

[0339] Furthermore, the emotion engine learns changes in the user's emotions and adjusts the digital clone's responses and the questions it asks other clones based on the learning results. Specifically, the emotion engine learns changes in the user's emotions over time and adjusts the digital clone's responses and the questions it asks other clones based on the learning results. For example, if the user always shows anger in a certain situation, the emotion engine learns that pattern and, when a similar situation occurs, adjusts the digital clone's responses and the questions it asks other clones in advance.

[0340] The processing flow of each embodiment will be described below.

[0341] "Example 1"

[0342] Step 1: The user uses the system to input information, which indicates the user's feelings.

[0343] Step 2: The emotion engine analyzes the user's input information and recognizes the user's emotion.

[0344] Step 3: Adjust the digital clone's response based on the emotions recognized by the emotion engine.

[0345] "Example 2"

[0346] Step 1: The user uses the system to input information, which indicates the user's feelings.

[0347] Step 2: The emotion engine analyzes the user's input information and recognizes the user's emotion.

[0348] Step 3: Adjust the questions posed to the other clone based on the emotions recognized by the emotion engine.

[0349] "Example 3"

[0350] Step 1: The user uses the system to input information, which indicates the user's feelings.

[0351] Step 2: The emotion engine analyzes the user's input information and recognizes the user's emotion.

[0352] Step 3: The emotion engine learns the user's emotional changes.

[0353] Step 4: Based on what the emotion engine has learned, adjust the digital clone's responses and the questions it asks other people's clones.

[0354] Example 1

[0355] Next, a description will be given of Example 1 of Form 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."

[0356] In modern companies, it is important for employees to effectively accumulate and share the knowledge, experience, and thoughts they gain in their daily work. However, there is a lack of means to digitize this information and properly manage and utilize it. Furthermore, there is no system that provides appropriate responses based on user emotions, which can lead to a decline in employee motivation and the quality of communication. Furthermore, there is a need for a system that allows employees to access other people's knowledge 24 hours a day, 365 days a year, while taking privacy issues into consideration.

[0357] 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.

[0358] In this invention, the server includes a means for digitizing employees' knowledge logs and thoughts, a means for creating digital clones, a means for extracting information using artificial intelligence, a means for collecting data from business tools, a means for analyzing and organizing the collected data, a means for recognizing users' emotions and adjusting responses based on those emotions, and a means for generating appropriate responses using a generative AI model. This enables employees' knowledge and experience to be effectively digitized and shared. Furthermore, providing appropriate responses based on users' emotions can improve employee motivation and the quality of communication. Furthermore, a system is realized that allows access to other people's knowledge 24 hours a day, 365 days a year, while taking privacy issues into consideration.

[0359] "Means for digitizing employee knowledge logs and thoughts" refers to technologies and methods for digitally recording and storing the knowledge, experience, and thoughts that employees gain in their daily work.

[0360] "Means of creating a digital clone" refers to technologies and methods that create a virtual digital persona based on an employee's knowledge and experience.

[0361] "Means of extracting information using artificial intelligence" refers to techniques and methods for using artificial intelligence technology to search for and extract necessary information from accumulated data.

[0362] "Means of collecting data from business tools" refers to technologies and methods for automatically obtaining data from business tools and systems used by employees.

[0363] "Means for analyzing and organizing collected data" refers to the techniques and methods used to analyze, organize, and digitize collected data.

[0364] "Means for recognizing a user's emotions and adjusting a response based on those emotions" refers to techniques or methods for analyzing a user's emotions and generating a response appropriate to those emotions.

[0365] "Means for generating appropriate responses using a generative AI model" refers to techniques or methods that use a generative AI model to generate appropriate responses based on user input and emotions.

[0366] This invention is a system that digitizes employees' knowledge logs and thoughts and creates digital clones. The system uses artificial intelligence to extract information and make it accessible at any time without memorizing past knowledge. It also has the ability to recognize users' emotions and adjust responses based on those emotions.

[0367] Hardware and software used

[0368] Hardware

[0369] Server: Analyzes and organizes data, creates digital clones, recognizes emotions, and generates responses.

[0370] Terminal: Collects data, accepts user input, and displays responses.

[0371] software

[0372] Data analysis software: Python, R

[0373] Sentiment analysis API: IBM Watson (registered trademark), Microsoft Azure (registered trademark)

[0374] Work tools: Office suite, Slack

[0375] Generative AI model: A specific generative AI model (e.g., GPT-3)

[0376] Data collection and analysis

[0377] The device collects data from the work tools used by employees (e.g., office suites, Slack). For example, it periodically scans spreadsheet files in office suites and Slack message logs to acquire data. The collected data is encrypted and sent to a server using the HTTPS protocol.

[0378] The server analyzes the received data using the Python pandas library. For example, it converts the contents of the spreadsheet file into a data frame and removes unnecessary columns. The analyzed data is then converted into a format that is easy to store in a database (e.g., JSON format).

[0379] Emotion Recognition and Response Generation

[0380] When a user provides input to the system, the device sends that input to the emotion engine. The emotion engine uses emotion analysis APIs from IBM Watson and Microsoft Azure to recognize the user's emotions. For example, if a user enters "The project was a success. I'm very happy," the emotion engine will recognize the emotion of "joy" from that text.

[0381] Based on the emotion data obtained from the emotion engine, the server inputs the prompt sentence "The user reports the success of the project and expresses his joy. Please generate an appropriate response" to the generative AI model. The generative AI model generates the response "Congratulations! That's a great achievement."

[0382] Providing a response

[0383] The terminal displays the response received from the server to the user, who sees a response from the digital clone saying, "Congratulations! That's a great job!"

[0384] Specific examples

[0385] For example, when a user reports the success of a project, the system recognizes the user's joy through its emotion engine, and the digital clone responds, "Congratulations! That's a great achievement."

[0386] Prompt Sentence Examples

[0387] "The project was a success. I'm very happy."

[0388] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0389] Step 1:

[0390] The device collects data from the work tools used by employees (e.g., office suites, Slack). Specifically, the device periodically scans the log data and documents of these tools to obtain the necessary data. For example, a script that automatically collects data is executed at the end of each day. The input is data from the work tools, and the output is the collected data.

[0391] Step 2:

[0392] The terminal encrypts the collected data and sends it to the server using the HTTPS protocol. Specifically, it sends the data via a POST request using the Python requests library. The input is the collected data, and the output is the encrypted data.

[0393] Step 3:

[0394] The server analyzes the received data. Specifically, it uses the Python pandas library to cleanse and classify the data. For example, it converts the contents of a spreadsheet file into a data frame and removes unnecessary columns. The input is the encrypted data, and the output is the analyzed data.

[0395] Step 4:

[0396] The server organizes and digitizes the analyzed data. Specifically, it converts it into a format that is easy to store in a database (e.g., JSON format). For example, it uses the pandas to_json method to convert a data frame into JSON format. The input is the analyzed data, and the output is the digitized data.

[0397] Step 5:

[0398] When a user provides input to the system, the device sends the input to the emotion engine. For example, if a user types "The project was a success. I'm very happy," the text is sent to the emotion engine. The input is the user's text input, and the output is the data sent to the emotion engine.

[0399] Step 6:

[0400] The emotion engine uses IBM Watson and Microsoft Azure sentiment analysis APIs to recognize user emotions. Specifically, it analyzes emotions such as "joy" from text. The input is the data sent to the emotion engine, and the output is the recognized emotion data.

[0401] Step 7:

[0402] The server inputs the emotion data obtained from the emotion engine as a prompt to the generative AI model. Specifically, it inputs "The user reports the success of the project and expresses joy. Please generate an appropriate response." The input is emotion data, and the output is a prompt to the generative AI model.

[0403] Step 8:

[0404] The generative AI model generates an appropriate response based on the prompt sentence, specifically, "Congratulations! That's a great job!" The input is the prompt sentence, and the output is the generated response.

[0405] Step 9:

[0406] The terminal displays the response received from the server to the user, specifically displaying the message "Congratulations! Great job!" on the user's screen. The input is the response generated, and the output is what is displayed to the user.

[0407] (Application example 1)

[0408] Next, a description will be given of Application Example 1 of Embodiment 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."

[0409] In conventional factories, there is a lack of effective means to digitize and share the knowledge and thoughts of workers. Furthermore, there is no system that can recognize workers' emotions and provide appropriate responses accordingly, which leads to problems of reduced work efficiency and quality of communication. Furthermore, there are privacy issues and insufficient means to access the knowledge of diverse experts 24 hours a day, 365 days a year.

[0410] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means. In this invention, the server includes means for digitizing employees' knowledge logs and thoughts, means for creating digital clones, means for extracting information using AI, means for making past knowledge accessible at any time without memorizing it, means for recognizing emotions, means for adjusting the digital clone's responses based on emotions, and means for digitizing factory workers' knowledge logs and thoughts and providing responses according to their emotions. This enables the effective digitization and sharing of workers' knowledge and thoughts, improving work efficiency and the quality of communication. It also makes it possible to access the knowledge of diverse experts 24 hours a day, 365 days a year, while taking privacy issues into consideration.

[0411] "Means for digitizing employee knowledge logs and thoughts" refers to a method for collecting, analyzing, organizing, and digitizing the knowledge, experience, and thoughts that employees gain in their daily work.

[0412] "Methods for creating digital clones" are methods for generating a virtual entity that digitally mimics the actions and thoughts of an employee based on the employee's knowledge log and thoughts.

[0413] "Means of extracting information using AI" refers to a method that uses artificial intelligence technology to search for necessary information from digital clones or knowledge logs and provide it to users.

[0414] "Means for making past knowledge accessible at any time without having to remember it" refers to a method for temporarily storing knowledge accumulated in the past so that it can be accessed immediately when needed.

[0415] "Means for recognizing emotions" refers to technology for analyzing and recognizing emotions from the user's facial expressions, tone of voice, etc.

[0416] The "means for adjusting the response of the digital clone based on emotions" is a method for the digital clone to generate an appropriate response depending on the recognized emotions of the user.

[0417] "A means of digitizing the knowledge, logs, and thoughts of factory workers and providing responses based on their emotions" is a method for digitizing the knowledge and thoughts that factory workers gain in the course of their work and providing appropriate responses based on the workers' emotions.

[0418] As an embodiment of this invention, we will build a system that digitizes the knowledge, logs, and thoughts of factory workers and provides responses according to their emotions. A specific embodiment of this system is shown below.

[0419] System configuration

[0420] The system includes the following major hardware and software components:

[0421] Hardware: Factory robots, cameras, microphones

[0422] Software: EmotionEngine (emotion recognition engine), KnowledgeLog (knowledge log management system), DigitalClone (digital clone response generation system)

[0423] Program processing explanation

[0424] emotion recognition

[0425] The server uses a camera and microphone to capture the worker's facial expressions and tone of voice, and the Emotion Engine analyzes this data to recognize the worker's emotions.

[0426] Knowledge Log Updates

[0427] The server records the newly acquired knowledge and experience of workers in KnowledgeLog, which organizes and digitizes this data.

[0428] Response Generation

[0429] The server uses DigitalClone to generate an appropriate response based on the recognized emotion and the latest knowledge log, and DigitalClone tailors the response to the emotion and provides it to the worker.

[0430] Specific examples

[0431] For example, if a factory worker discovers an improvement in a new manufacturing process and expresses joy, the system might:

[0432] 1. Cameras and microphones capture the facial expressions and voices of workers.

[0433] 2. The Emotion Engine analyzes this data and recognizes that the worker is expressing the emotion of joy.

[0434] 3. KnowledgeLog records new knowledge (new manufacturing process improvements) from workers.

[0435] 4. DigitalClone generates a response based on the perceived sentiment and the latest knowledge log, such as:

[0436] "That's a great discovery! Let's record the improvement immediately and share it with other workers."

[0437] Prompt Sentence Examples

[0438] An example of a prompt sentence to input to the generative AI model is as follows:

[0439] "If a worker discovers an improvement in a new manufacturing process and expresses joy, what response does it generate?"

[0440] In this way, the system effectively digitizes and shares the knowledge and thoughts of workers, improving work efficiency and the quality of communication. It also makes it possible to access the knowledge of a variety of experts 24 hours a day, 365 days a year, while taking privacy issues into consideration.

[0441] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0442] Step 1:

[0443] The server uses a camera and microphone to capture the worker's facial expressions and tone of voice. It receives real-time video and audio data of the worker as input. This data is sent to the Emotion Engine, where it undergoes preprocessing for emotion recognition. Specifically, it extracts facial features from the video data and analyzes the tone and pitch of the voice from the audio data.

[0444] Step 2:

[0445] The server uses the Emotion Engine to recognize the worker's emotions from the captured data. As input, it uses preprocessed facial feature data and voice analysis data. The Emotion Engine analyzes these data and identifies the worker's emotions (e.g., joy, sadness, surprise, etc.). As output, recognized emotion data is generated.

[0446] Step 3:

[0447] The server records the new knowledge and experience gained by workers in KnowledgeLog. As input, it receives text data of the new knowledge and experience reported by workers. KnowledgeLog organizes and digitizes this data. Specifically, it stores the text data in a structured database and converts it into a searchable format. As output, it generates updated knowledge log data.

[0448] Step 4:

[0449] The server uses DigitalClone to generate an appropriate response based on the recognized emotion and the latest knowledge log. It uses emotion data and knowledge log data as input. DigitalClone analyzes these data and generates a response based on the emotion. Specifically, it adjusts the tone of the response based on the emotion data and creates a response with specific content based on the knowledge log data. As output, the generated response message is provided to the worker.

[0450] Step 5:

[0451] The server provides the generated response message to the worker. As input, it receives the response message generated by DigitalClone. Specifically, it sends the response message in text format to the worker's device or provides it audibly using speech synthesis technology. As output, the worker can receive and confirm the response message.

[0452] In this way, the system effectively digitizes and shares the knowledge and thoughts of workers, improving work efficiency and the quality of communication. It also makes it possible to access the knowledge of a variety of experts 24 hours a day, 365 days a year, while taking privacy issues into consideration.

[0453] Example 2

[0454] Next, a description will be given of Example 2 of Form 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."

[0455] With conventional systems, it was difficult to effectively digitize employee knowledge and experience and quickly retrieve the necessary information. Furthermore, there were problems with not being able to provide appropriate responses based on the user's emotions, which did not improve the user experience. Furthermore, there were privacy issues and a lack of means to access the knowledge of diverse experts 24 hours a day, 365 days a year.

[0456] 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.

[0457] In this invention, the server includes a means for digitizing employee knowledge, experience, and thinking, a means for creating a digital clone, a means for training data using a machine learning framework, a means for recognizing user emotions and adjusting question content, a means for generating appropriate responses using a generative AI model, and a means for making past knowledge accessible at any time without storing it. This effectively digitizes employee knowledge and experience and provides appropriate responses based on user emotions. Furthermore, while taking privacy issues into consideration, the knowledge of a variety of experts can be accessed 24 hours a day, 365 days a year.

[0458] "Means for digitizing employee knowledge, experience, and thinking" refers to a method for collecting and storing employees' work-related knowledge, experience, and thought processes as digital data.

[0459] "Means for creating a digital clone" refers to a method for generating an AI model that mimics the behavior and judgment of an employee based on the knowledge, experience, and thoughts of the digitized employee.

[0460] "Methods of training data using machine learning frameworks" refers to methods of training AI models based on collected data using machine learning frameworks such as TensorFlow and PyTorch.

[0461] "Means for recognizing user emotions and adjusting questions" refers to a method for analyzing the user's voice and text input to identify emotions and then changing the questions posed to the AI ​​model according to those emotions.

[0462] "Means for generating appropriate responses using a generative AI model" refers to a method for generating appropriate responses to user questions using a generative AI model (e.g., GPT-3).

[0463] "Means for making past knowledge accessible at any time without memorizing it" refers to a method for making past knowledge accessible immediately when needed without temporarily storing it.

[0464] "Privacy-conscious measures" are methods to prevent personal or confidential information from being leaked when extracting information from a digital clone.

[0465] "A means of accessing the knowledge of diverse experts 24 hours a day, 365 days a year" is a method that allows necessary information to be extracted at any time from multiple digital clones with different specialized knowledge.

[0466] MODE FOR CARRYING OUT THE INVENTION

[0467] This invention is a system for digitizing the knowledge, experience, and thoughts of employees and creating digital clones. A specific embodiment of this system will be described below.

[0468] Digital cloning

[0469] The server first collects data to digitize employees' knowledge, experience, and thoughts. This data includes employees' work history, email and chat logs, project documents, etc. The server automatically collects data from email servers and project management tools.

[0470] The server then preprocesses the collected data, specifically cleaning, normalizing, and tokenizing it. It uses the Python pandas library to impute missing values ​​and removes unnecessary strings using regular expressions.

[0471] The server then uses the preprocessed data to train an AI model using machine learning frameworks such as TensorFlow and PyTorch. The server uses TensorFlow to build a neural network and trains the model using employee data as input.

[0472] Finally, the server evaluates the performance of the trained model and tunes the hyperparameters as needed. It evaluates the precision and recall of the model and finds the optimal hyperparameters using grid search or random search.

[0473] Adjusting question content using an emotion engine

[0474] The device is equipped with an emotion engine to recognize the user's emotions. It analyzes the user's voice and text input to identify emotions. The device converts the user's voice into text using a speech recognition API (e.g., Google® Cloud Speech-to-Text) and analyzes emotions using a natural language processing (NLP) model.

[0475] If the user expresses emotions such as anger or frustration, the device will recognize that emotion and adjust the questions it asks the other clone. For example, if the user types, "This project isn't progressing at all!", the emotion engine will recognize the anger and ask the other clone, "Please give me some specific advice to help the project progress smoothly."

[0476] The server generates an appropriate response from the clone of another person based on the adjusted question. The generated response is sent to the device. The server uses a generative AI model (e.g., GPT-3) to generate an appropriate response to the question and sends it to the device.

[0477] Specific examples

[0478] For example, if an employee has expertise in project management, an AI model can be created that learns from that employee's past project data and communication history to generate appropriate answers to questions about project management.

[0479] Example prompt: "Tell me how to effectively manage project progress."

[0480] This system effectively digitizes the knowledge and experience of employees, enabling them to provide appropriate responses based on user sentiment. It also makes it possible to access the knowledge of a wide range of experts 24 hours a day, 365 days a year, while taking privacy issues into consideration.

[0481] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0482] Step 1:

[0483] Data collection

[0484] The server collects data to digitize employees' knowledge, experience, and thoughts. As input, it receives employee work history, email and chat logs, project documents, etc. The server automatically collects data from email servers and project management tools and stores it in a database. The output is the collected dataset.

[0485] Step 2:

[0486] Data Preprocessing

[0487] The server preprocesses the collected data. It uses the dataset collected in step 1 as input. Specifically, it uses Python's pandas library to fill in missing values ​​in the data and removes unnecessary strings using regular expressions. The output is a preprocessed, clean dataset.

[0488] Step 3:

[0489] Model training

[0490] The server trains an AI model using the preprocessed data. It uses the dataset preprocessed in step 2 as input. Specifically, it uses a machine learning framework such as TensorFlow or PyTorch to build a neural network and train the data. The output is a trained AI model.

[0491] Step 4:

[0492] Model evaluation and tuning

[0493] The server evaluates the performance of the trained model and tunes the hyperparameters as necessary. The input is the AI ​​model trained in step 3 and the evaluation dataset. Specifically, it evaluates the precision and recall of the model and finds the optimal hyperparameters using grid search and random search. The output is an optimized AI model.

[0494] Step 5:

[0495] emotion recognition

[0496] The device launches an emotion engine to recognize the user's emotions. It receives the user's voice and text input as input. Specifically, it converts the user's voice into text using a speech recognition API (e.g., Google Cloud Speech-to-Text) and analyzes the emotion using a natural language processing (NLP) model. The output is the recognized user's emotion data.

[0497] Step 6:

[0498] Adjusting the content of questions

[0499] The device adjusts the questions it asks the other person's clone based on the recognized emotion. The input is the user's emotion data recognized in step 5. Specifically, if the user inputs "This project isn't progressing at all!", the emotion engine recognizes the anger and asks the other person's clone, "Please give me some specific advice to help the project progress smoothly." The output is the adjusted questions.

[0500] Step 7:

[0501] Response Generation

[0502] The server generates an appropriate response from the clone of another person based on the adjusted question. The question adjusted in step 6 is used as input. Specifically, it uses a generative AI model (e.g., GPT-3) to generate an appropriate response to the question and sends it to the device. The output is the generated response.

[0503] (Application example 2)

[0504] Next, a description will be given of Application Example 2 of Form 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."

[0505] Traditional security services struggle to effectively utilize the knowledge and experience of their employees, and require flexible responses based on user sentiment. Furthermore, the lack of a means to provide real-time security advice and support makes it difficult to improve user satisfaction.

[0506] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0507] In this invention, the server includes means for digitizing employee knowledge, logs, and thoughts, means for creating a digital clone, means for extracting information using AI, means for making past knowledge accessible at any time without memorizing it, means for recognizing a user's emotions and adjusting the questions posed to the digital clone based on those emotions, means for providing security advice and support in real time, and means for being installed on a smart device. This makes it possible to effectively utilize the knowledge and experience of employees, respond flexibly to user emotions, and provide security advice and support in real time.

[0508] "Means for digitizing employee knowledge logs and thinking" refers to technology that records and stores employees' knowledge, experiences, and thought patterns as digital data.

[0509] "Means for creating digital clones" refers to technology that generates digital agents that mimic the behavior and thinking of employees based on their digitized knowledge logs and thoughts.

[0510] "Means of extracting information using AI" refers to a technique that uses artificial intelligence technology to search for and obtain the necessary information from a digital clone.

[0511] "Means for making past knowledge accessible at any time without storing it" refers to technology that allows the knowledge and information held by a digital clone to be accessed instantly when needed without being stored in a specific storage device.

[0512] "Means for recognizing the user's emotions and adjusting the questions asked to the digital clone based on those emotions" refers to technology that analyzes the user's emotions from their facial expressions and voice, and changes the information and advice provided to the digital clone based on those emotions.

[0513] "Means for providing security advice and support in real time" refers to technology that provides immediate and appropriate advice and support for security issues faced by users.

[0514] "Means to be installed on smart devices" refers to technology that is installed and operates as an application on portable electronic devices such as smartphones and head-mounted displays.

[0515] A system for carrying out this invention has the following configuration: a server including means for digitizing employees' knowledge logs and thoughts, means for creating digital clones, means for extracting information using AI, means for making past knowledge accessible at any time without storing it, means for recognizing a user's emotions and adjusting questions to be asked of the digital clone based on those emotions, means for providing security advice and support in real time, and means installed on a smart device.

[0516] Hardware and Software Configuration

[0517] 1. Hardware:

[0518] Smartphones (e.g., iPhone, Android devices)

[0519] Head-mounted display (e.g., Microsoft HoloLens (registered trademark))

[0520] 2. Software:

[0521] AI model building: TensorFlow, PyTorch

[0522] Emotion recognition engine: Microsoft Azure Emotion API

[0523] Database: MySQL

[0524] Frontend: React Native, 3D modeling software

[0525] Data processing and calculation

[0526] The server digitizes employees' knowledge logs and thoughts, and creates a digital clone based on these. The digital clone uses an AI model to mimic the employee's behavior and thoughts. When a user accesses the system through a smart device, an emotion recognition engine analyzes the user's emotions from their facial expressions and voice. The analyzed emotional data is input into the AI ​​model, which then provides appropriate information and advice based on the user's emotions.

[0527] Specific examples

[0528] For example, consider a case where a user accesses a security assistant on a smartphone and asks a security question. If the user feels anxious, the emotion recognition engine analyzes the emotion, and the AI ​​model provides specific advice to alleviate the anxiety. The following is an example of a prompt sentence:

[0529] "How should a security assistant respond if a user is angry?"

[0530] "What advice should we offer users if they are feeling anxious?"

[0531] In this way, the server can provide security advice and support according to the user's feelings in real time, which can improve user satisfaction.

[0532] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0533] Step 1:

[0534] The server digitizes employees' knowledge logs and thoughts. Specifically, it collects data on employees' past security response history and knowledge base, and stores this as digital data. The input is employees' knowledge logs and thoughts, and the output is digitized data.

[0535] Step 2:

[0536] The server creates a digital clone based on the digitized data. Specifically, it uses an AI model to generate a digital agent that mimics the employee's behavior and thoughts. The input is the digitized data, and the output is the digital clone.

[0537] Step 3:

[0538] Users access the system through smart devices, specifically smartphones or head-mounted displays, to connect to the security assistant. The input is the user's access request, and the output is the connection to the system.

[0539] Step 4:

[0540] The device sends the user's facial expressions and voice to the emotion recognition engine. Specifically, it captures the user's facial expressions and voice using the smart device's camera and microphone and sends them to the emotion recognition engine. The input is the user's facial expressions and voice data, and the output is data sent to the emotion recognition engine.

[0541] Step 5:

[0542] The server uses an emotion recognition engine to analyze the user's emotions. Specifically, it uses the Microsoft Azure Emotion API to analyze emotions from the user's facial expressions and voice. The input is the user's facial expression and voice data, and the output is the analyzed emotional data.

[0543] Step 6:

[0544] The server adjusts the questions it asks the digital clone based on the analyzed emotional data. Specifically, if the user shows anger or anxiety, it changes the digital clone's responses to provide information or advice to alleviate those emotions. The input is the analyzed emotional data, and the output is the adjusted questions.

[0545] Step 7:

[0546] The server provides real-time security advice and support based on the tailored questions. Specifically, it uses an AI model to generate appropriate advice and information and provide it to the user. The input is the tailored questions, and the output is advice and support provided in real time.

[0547] Step 8:

[0548] The terminal displays the advice and support provided by the server to the user. Specifically, it displays appropriate information to the user through the screen of the smart device or a head-mounted display. The input is the advice and support from the server, and the output is the information displayed to the user.

[0549] Example 3

[0550] Next, a third embodiment of the third embodiment will be described. 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."

[0551] With conventional systems, it was difficult to effectively digitize employee knowledge and thinking and quickly retrieve the necessary information. Furthermore, there was a lack of means to make past knowledge accessible in real time without having to memorize it, so users had to spend a lot of time and effort to obtain the information they needed. Furthermore, there were limited ways to simultaneously access the knowledge of multiple experts, making it difficult for users to obtain information from diverse perspectives.

[0552] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0553] In this invention, the server includes: a means for a user to input a question or instruction; a means for the terminal to transmit the user's input to the server; a means for the server to receive the input and transmit a prompt to an appropriate generative AI model; a means for the generative AI model to generate an answer based on the prompt; a means for the server to receive the answer from the generative AI model and complement the information as needed in cooperation with a database; a means for the server to transmit the final answer to the terminal; and a means for the terminal to display the answer to the user, thereby enabling the user to quickly and efficiently obtain the information they need.

[0554] "Means for digitizing employee knowledge logs and thinking" is a means for recording and storing employees' knowledge, experiences, and thought processes in digital form.

[0555] "Means for creating a digital clone" refers to the creation of a virtual entity that digitally mimics the actions and thoughts of an employee based on that employee's knowledge log and thoughts.

[0556] "Means of extracting information using AI" refers to the use of artificial intelligence technology to search for and retrieve necessary information from digital clones and databases.

[0557] "Means for making past knowledge accessible at any time without having to remember it" refers to means for searching and extracting past data and knowledge in real time, allowing users to access it immediately when needed.

[0558] The "means for the user to input questions or instructions" refers to a means for providing an interface for the user to input questions or instructions to the system.

[0559] The "means by which the terminal transmits user input to the server" refers to a communication means for transmitting questions or instructions input by the user to the server.

[0560] "Means for the server to receive input and send prompt sentences to the appropriate generative AI model" refers to means for the server to analyze input received from the user and send prompt sentences to the appropriate generative AI model.

[0561] "Means by which a generative AI model generates an answer based on a prompt sentence" refers to means by which a generative AI model generates an appropriate answer from pre-trained data based on a prompt sentence.

[0562] "Means for the server to receive answers from the generative AI model and, if necessary, complement the information in cooperation with the database" refers to means for the server to receive answers from the generative AI model and, if necessary, obtain additional information from the database to complement the answers.

[0563] The "means by which the server transmits the final answer to the terminal" refers to the means by which the server transmits the final generated answer to the user's terminal.

[0564] The "means for the terminal to display the answer to the user" refers to the means for the user's terminal to display the answer received from the server to the user.

[0565] This invention is a system that digitizes employees' knowledge logs and thoughts, creates digital clones, and uses AI to extract information. Furthermore, it makes past knowledge accessible at any time without memorizing it, and allows the knowledge of multiple experts to be used simultaneously.

[0566] Hardware and software used

[0567] The system is implemented using the following hardware and software:

[0568] Server: Use a server with a powerful processor and large memory capacity that receives input from the user, sends prompts to the appropriate generative AI model, and processes the generated answers.

[0569] Device: A device that allows a user to enter questions or instructions, such as a smartphone, tablet, or PC.

[0570] Generative AI models: Models built using machine learning libraries such as TensorFlow and PyTorch that generate answers based on prompts.

[0571] Database: Uses SQL or NoSQL databases to store historical data and knowledge. The server interacts with the database to supplement information as needed.

[0572] Data processing and calculation

[0573] The server receives input from the user, analyzes it, and sends a prompt to the appropriate generative AI model. The generative AI model generates an answer to the prompt based on pre-trained data. The generated answer is returned to the server, which complements the information as needed by linking with a database. The final answer is sent from the server to the device, which displays it to the user.

[0574] Specific examples

[0575] Consider a case where a user asks, "What are some examples of success stories from past projects?" The following is an example of a prompt sentence to input this question to a generative AI model:

[0576] Example prompt sentence:

[0577] Users: What are some success stories from past projects?

[0578] The server sends this prompt to the generative AI model, which generates the following answer based on past data:

[0579] Example of an answer from the AI ​​model:

[0580] AI Model: A successful example of a past project is the "New Product Development Project" carried out in 2022. In this project, the entire team worked together to develop a high-quality product in a short period of time, which achieved great success in the market.

[0581] In this way, the user can quickly and efficiently obtain the necessary information through the AI ​​model. The flow of the identification process in the third embodiment will be described with reference to FIG.

[0582] Step 1:

[0583] The user enters a question or instruction.

[0584] The user enters questions or instructions through the device interface, for example, "What are some success stories from past projects?" The entered data is stored in text format on the device.

[0585] Step 2:

[0586] The terminal sends the user's input to the server.

[0587] The terminal receives user input and sends it to the server. The communication protocol is HTTP POST request. For example, send data to the endpoint POST / api / query. The input is in text format, and the output is the data to be sent to the server.

[0588] Step 3:

[0589] The server receives the input and sends prompts to the appropriate generative AI model.

[0590] The server analyzes the input received from the device and sends a prompt to the appropriate generative AI model. For example, if the input is "What are some successful examples of past projects?", the server sends that question as a prompt to the generative AI model. The input is a text question, and the output is a prompt to the generative AI model.

[0591] Step 4:

[0592] A generative AI model generates an answer based on the prompt.

[0593] The generative AI model receives a prompt and generates an answer based on pre-trained data. For example, it generates an answer such as "The new product development project implemented in 2022 was successful." The input is the prompt, and the output is the generated answer text.

[0594] Step 5:

[0595] The server receives the answers from the generative AI model and complements the information by interacting with the database as needed.

[0596] The server receives the answer from the generative AI model and complements it with a database if necessary. For example, it retrieves detailed information about a specific project from the database. The input is the answer text from the generative AI model, and the output is the complemented answer text.

[0597] Step 6:

[0598] The server sends the final answer to the device.

[0599] The server generates a final answer and sends it to the device. For example, it sends an answer such as, "The new product development project conducted in 2022 was successful, and the entire team worked together to develop a high-quality product in a short period of time." The input is the completed answer text, and the output is the data sent to the device.

[0600] Step 7:

[0601] The terminal displays the answer to the user.

[0602] The device displays the answer received from the server to the user. The user confirms the answer on the device screen. For example, a smartphone application might display the text "The new product development project carried out in 2022 was successful, with the entire team working together to develop a high-quality product in a short period of time." The input is the answer data from the server, and the output is the text displayed to the user.

[0603] (Application example 3)

[0604] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0605] Modern companies are required to effectively digitize employee knowledge and thinking and quickly retrieve necessary information. However, conventional systems have difficulty retaining past knowledge and retrieving information in real time. Furthermore, they lacked a means to obtain optimal answers from multiple perspectives, and did not adjust responses to take user emotions into account. In particular, they were unable to propose optimal countermeasures in real time for security-related questions, nor did they provide advice to reduce user stress and anxiety. A new system is needed to solve these issues.

[0606] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[0607] In this invention, the server includes means for digitizing employee knowledge logs and thoughts, means for creating digital clones, means for retrieving information using AI, means for making past knowledge accessible at any time without memorizing it, means for simultaneously using multiple AI models to obtain optimal answers, means for learning changes in a user's emotions using an emotion engine and adjusting responses, means for answering security questions in real time and proposing optimal security measures, and means for learning the user's emotions and providing advice to reduce stress and anxiety. This makes it possible to quickly obtain necessary information and obtain optimal answers from multiple perspectives, adjust responses taking into account the user's emotions, propose security measures, and reduce stress and anxiety.

[0608] "Means for digitizing employee knowledge logs and thinking" refers to technology for recording and storing employees' knowledge, experiences, and thought processes in digital form.

[0609] "Means for creating digital clones" refers to technology that generates digital agents that mimic the knowledge and thought patterns of a specific individual.

[0610] "Means of extracting information using AI" refers to technology that uses artificial intelligence to search for and provide the necessary information based on the user's questions and instructions.

[0611] "Means for making past knowledge accessible at any time without having to remember it" refers to technology that temporarily stores past data and makes it accessible in real time when needed.

[0612] "Means of obtaining optimal answers by simultaneously using multiple AI models" refers to a technology that uses multiple different artificial intelligence models simultaneously, compares and integrates their responses, and derives the optimal answer.

[0613] "Means for using an emotion engine to learn changes in a user's emotions and adjust responses" is a technology that analyzes a user's emotional state, learns those changes, and generates appropriate responses.

[0614] "Means for answering security-related questions in real time and proposing optimal security measures" refers to technology that provides immediate answers to security-related questions from users and proposes optimal security measures.

[0615] "Means for learning a user's emotions and providing advice to reduce stress and anxiety" is a technology that learns a user's emotional state and provides specific advice to reduce stress and anxiety.

[0616] A system for implementing this invention includes means for digitizing employee knowledge logs and thoughts and creating digital clones, thereby allowing employees' knowledge and experience to be stored in digital form and quickly accessed when needed.

[0617] The server uses artificial intelligence (AI) to retrieve information, searches for and provides the necessary information based on user questions and instructions, and uses the OpenAI API to generate answers to user questions.

[0618] Furthermore, the server has a means to make past knowledge accessible at any time without storing it, and searches and extracts past data stored in the database in real time, allowing users to always access the latest information.

[0619] The server has the means to simultaneously use multiple AI models to obtain the optimal answer, and compares and integrates answers from different perspectives to provide the optimal answer, allowing users to obtain information from multiple perspectives.

[0620] The system also includes a means for using an emotion engine to learn changes in the user's emotions and adjust responses accordingly. The emotion engine analyzes the user's emotional state, learns those changes, and generates appropriate responses. For example, if a user always shows anger in a certain situation, the emotion engine learns that pattern and, when a similar situation occurs, adjusts the digital clone's responses and the questions it asks other clones in advance.

[0621] It also includes a means to answer security questions in real time and suggest optimal security measures. When a user enters a security question, the server searches the database and uses an AI model to suggest the optimal security measures.

[0622] It also includes a means for learning the user's emotions and providing advice to reduce stress and anxiety. The emotion engine learns the user's emotional state and provides specific advice to reduce stress and anxiety.

[0623] For example, if a user asks the question "How can I improve the security of my home?", the following prompt sentence is input to the generative AI model:

[0624] Security Questions: How can I improve the security of my home?

[0625] answer:

[0626] Based on this prompt, the AI ​​model generates an answer, integrates it with information from the database, and provides it to the user, allowing them to take quick and accurate security measures.

[0627] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0628] Step 1:

[0629] The user enters a security question.

[0630] Input: A user types the question, "How can I improve the security of my home?"

[0631] Output: The user's question is sent to the server in text form.

[0632] Step 2:

[0633] The server receives the user's query and searches the database.

[0634] Input: The user's question text.

[0635] Data Processing: The server connects to an SQLite database and retrieves security advice related to the question.

[0636] Output: The security advice retrieved from the database.

[0637] Step 3:

[0638] The server asks the AI ​​model questions.

[0639] Input: The user's question text.

[0640] Data calculation: The server uses the OpenAI API to send prompts to the generative AI model and generate answers.

[0641] Specific behavior: Send the prompt "Security Question: How can I improve the security of my home?\nAnswer:" to the generative AI model.

[0642] Output: Answer text from the generative AI model.

[0643] Step 4:

[0644] The server combines advice from the database with answers from the AI ​​model.

[0645] Input: Security advice from the database, answer text from the generative AI model.

[0646] Data processing: The server compares both answers and selects the best security measure.

[0647] Output: Text of the consolidated best security measures.

[0648] Step 5:

[0649] The server uses an emotion engine to learn the user's emotions and adjust its response.

[0650] Input: User's emotional state (e.g., anxiety).

[0651] Data Computation: The emotion engine analyzes the user's emotional state and learns its changes.

[0652] Specific behavior: If a user consistently exhibits anxiety in a particular situation, learn that pattern and adjust your response.

[0653] Output: The adjusted response text.

[0654] Step 6:

[0655] The server provides the user with the best security measures and a tailored response.

[0656] Input: Unified best security measure text, tailored response text.

[0657] Output: The optimal security measure and response provided to the user.

[0658] Step 7:

[0659] The user reviews the security measures and responses provided and asks additional questions as needed.

[0660] Input: The security measures and responses provided by the server.

[0661] Output: Additional questions for the user (if necessary).

[0662] 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.

[0663] 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> ) 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.

[0664] Another example of generative AI is Gemini (registered trademark) (Internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[0665] 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.

[0666] [Second embodiment]

[0667] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0668] 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.

[0669] 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).

[0670] 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.

[0671] 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.

[0672] 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).

[0673] 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.

[0674] 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.

[0675] 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.

[0676] 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.

[0677] 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.

[0678] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0679] "Example 1"

[0680] This invention uses specific algorithms and programs to digitize employees' knowledge, logs, and thoughts, accumulating the knowledge, experience, and thoughts that employees gain through their daily work. Specifically, it collects data from the tools and systems that employees use in their work, analyzes and organizes it, and digitizes it.

[0681] "Example 2"

[0682] Next, as a means of creating a digital clone, an AI model is constructed based on the digitized data to mimic the knowledge, experience, and thinking of the employee. Specifically, an AI model that reproduces the knowledge, experience, and thinking of the employee is created by learning the data.

[0683] "Example 3"

[0684] Furthermore, as a means of extracting information using AI, questions and instructions can be given to the created AI model to obtain the necessary information. Specifically, questions such as "What are some successful examples of past projects?" can be asked of the AI ​​model and the answer obtained.

[0685] "Example 4"

[0686] Additionally, as a means of making past knowledge accessible at any time without having to memorize it, the AI ​​model is linked to a database to obtain the necessary information in real time. Specifically, the AI ​​model searches and extracts past data stored in the database and provides it to the user. "Example 5"

[0687] Finally, multiple AI models can be used simultaneously to access the knowledge of a wide range of experts 24 hours a day, 365 days a year, by asking questions to clones of other people. Specifically, questions can be asked to multiple AI models as needed, and the answers from each model can be compared and integrated to obtain the optimal answer.

[0688] The processing flow of each embodiment will be described below.

[0689] "Example 1"

[0690] Step 1: Collect data from the tools and systems employees use to do their work. For example, collect project progress and deliverables from work management tools, and message content from communication tools.

[0691] Step 2: Analyze, organize, and digitize the collected data. Specifically, use text analysis and natural language processing techniques to extract useful information from the data and store it in a database.

[0692] "Example 2"

[0693] Step 1: Build an AI model based on the digitized data. Specifically, use the data as learning data and apply machine learning algorithms to train the AI ​​model.

[0694] Step 2: Use the trained AI model to make predictions and analyze new data. For example, use knowledge learned from past data to make predictions and analyze new business data.

[0695] "Example 3"

[0696] Step 1: Ask questions or give instructions to the created AI model. Specifically, the user's question is given as input to the AI ​​model, and the AI ​​model returns the output to the user.

[0697] Step 2: Evaluate the output from the AI ​​model and update it if necessary. For example, if the output from the AI ​​model does not meet user expectations, use that feedback to retrain the AI ​​model.

[0698] "Example 4"

[0699] Step 1: Link the AI ​​model with the database. Specifically, make it possible for the AI ​​model to obtain the data it needs from the database.

[0700] Step 2: The AI ​​model searches and extracts past data stored in the database and provides it to the user. For example, it searches the database for an answer to a user's question and returns the result to the user.

[0701] "Example 5"

[0702] Step 1: Ask multiple AI models questions as needed. Specifically, provide the user's question as input to multiple AI models, and obtain output from each AI model.

[0703] Step 2: Compare and integrate the outputs from each AI model to obtain the optimal answer. For example, compare the outputs from multiple AI models and return the most reliable output to the user.

[0704] Example 1

[0705] Next, a description will be given of Example 1 of Form 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."

[0706] With conventional systems, it was difficult to efficiently digitize and store the knowledge, experience, and thoughts that employees gained in their daily work. There was also a lack of means to properly analyze and organize the stored information and quickly access it when needed. Furthermore, there were also privacy issues and a lack of adequate means to access the knowledge of a variety of experts 24 hours a day, 365 days a year.

[0707] 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.

[0708] In this invention, the server includes means for digitizing employee knowledge, records, and thoughts, means for creating digital clones, means for extracting information using artificial intelligence, means for making past knowledge accessible at any time without memorizing it, means for collecting data from tools and systems used in business, means for analyzing the collected data, means for organizing the analyzed data, means for storing the organized data in a database, and means for providing the stored data. This allows for efficient digitization of employee knowledge, experience, and thoughts, appropriate analysis and organization of accumulated information, and rapid access when needed. Furthermore, while taking privacy issues into consideration, the knowledge of diverse experts can be accessed 24 hours a day, 365 days a year.

[0709] "Means for digitizing employee knowledge, records, and thoughts" refers to a method for storing the knowledge, experience, and thoughts that employees gain in their daily work in electronic form.

[0710] "Means for creating a digital clone" refers to a method for creating a virtual persona based on an employee's knowledge and experience.

[0711] "Means of extracting information using artificial intelligence" refers to a method of extracting necessary information from accumulated data using AI technology.

[0712] "Means for making past knowledge accessible at any time without having to remember it" refers to a method for making information stored in a database searchable and retrievalable in real time.

[0713] "Means of collecting data from tools and systems used in business" refers to methods for obtaining data from software and hardware that employees use on a daily basis.

[0714] "Means for analyzing collected data" refers to methods for analyzing acquired data and extracting meaningful information.

[0715] "Means for organizing analyzed data" refers to methods for classifying the analysis results into categories, tags, etc., and converting them into a format that is easy to search later.

[0716] "Means for storing organized data in a database" refers to a method for storing classified data in a database.

[0717] A "means for providing stored data" is a method by which information stored in a database can be accessed by a user.

[0718] MODE FOR CARRYING OUT THE INVENTION

[0719] This invention is a system that digitizes the knowledge, records, and thoughts of employees and efficiently stores, analyzes, and provides them. A specific embodiment of this system is described below.

[0720] Data collection

[0721] The server collects data from the tools and systems that employees use in their work. Specifically, it obtains data from communication tools (e.g., Slack, Microsoft Teams), office tools (e.g., office suites, Google Workspace), project management tools, etc. The server periodically collects data using the APIs of these tools.

[0722] Data analysis

[0723] The server analyzes the collected data. Specifically, it uses Python's NLTK library and spaCy to analyze the text data. The server tokenizes the message content and extracts important keywords and phrases. For example, the server extracts the keywords "project," "progress," and "report" from the message "Report on the progress of the project."

[0724] Data organization

[0725] The server organizes the analyzed data. Specifically, it categorizes the extracted keywords and phrases. The server organizes the data by category and converts it into a format that makes it easier to search later. For example, it creates categories such as "project management," "communication," and "reporting," and organizes the keywords that correspond to each category.

[0726] Data storage

[0727] The server saves the organized data in a database. Specifically, it uses a database management system (DBMS) such as MySQL or PostgreSQL to store the data. The server connects to the database and inserts the organized data into the appropriate table. For example, it inserts the data into the "knowledge_logs" table.

[0728] Data provision

[0729] The server provides the digitized data in a format that employees can access. Specifically, the data is made accessible through a web application using Django and Flask. Employees can search and view the data by opening a browser and accessing the Django application. For example, if an employee selects the category "Project Management," relevant messages are displayed.

[0730] Examples of concrete examples and prompts

[0731] As a concrete example, consider a case where employees use Slack to report project progress. The server collects message data through the Slack API and analyzes the message content using the Python NLTK library. The analyzed results are stored in a MySQL database and made accessible to employees through a web application using Django.

[0732] Example prompt sentence:

[0733] "Collect Slack message data, parse it using Python's NLTK library, and store it in a MySQL database. Then, make the data accessible through a web application using Django."

[0734] In this way, the knowledge, experience, and thoughts of employees can be efficiently digitized, the accumulated information can be properly analyzed and organized, and it can be quickly accessed when needed. Furthermore, the knowledge of a wide range of experts can be accessed 24 hours a day, 365 days a year, while taking privacy issues into consideration.

[0735] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0736] Step 1: Data collection

[0737] The server collects data from the tools and systems employees use in their work. Specifically, the server obtains message data using Slack's API. The server calls the "https: / / slack.com / api / conversations.history" endpoint at 9:00 AM every day to collect the previous day's message data. The input is the message data obtained from Slack's API, and the output is the collected message data.

[0738] Step 2: Data analysis

[0739] The server analyzes the collected message data. Specifically, it uses Python's NLTK library to parse the text data. The server tokenizes the messages and extracts important words such as nouns and verbs. For example, the server extracts the keywords "project," "progress," and "report" from the message "Report on the progress of the project." The input is the collected message data, and the output is the extracted keywords.

[0740] Step 3: Data organization

[0741] The server organizes the analyzed data. Specifically, the server classifies the extracted keywords into categories. For example, it creates categories such as "project management," "communication," and "reporting," and organizes the keywords corresponding to each category. The input is the extracted keywords, and the output is the keywords organized by category.

[0742] Step 4: Save Data

[0743] The server saves the organized data in a database. Specifically, the server uses a MySQL database to store the data. The server connects to the database and inserts data into the "knowledge_logs" table. For example, it executes the SQL query "INSERT INTO knowledge_logs (category, keyword, message) VALUES ('Project management', 'Progress', 'Report project progress')". The input is keywords organized by category, and the output is the data stored in the database.

[0744] Step 5: Provide data

[0745] The server provides the digitized data in a format that employees can access. Specifically, the server makes the data accessible through a web application using Django. Employees can search and view the data by opening a browser and accessing the Django application. For example, if an employee selects the category "Project Management," relevant messages are displayed. The input is the data stored in the database, and the output is the data provided through the web application.

[0746] (Application example 1)

[0747] Next, a description will be given of Application Example 1 of Form 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."

[0748] In traditional factories, the knowledge and experience of workers is dependent on the individual, and there is a lack of means to efficiently share and utilize this knowledge. It is also difficult to digitize and store the operations and thoughts of workers, making knowledge transfer and work efficiency an issue. Furthermore, privacy issues and the effort required to organize and analyze data are also major obstacles.

[0749] 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.

[0750] In this invention, the server includes a means for digitizing employees' knowledge, logs, and thoughts, a means for creating digital clones, a means for extracting information using AI, a means for making past knowledge accessible at any time without memorizing it, a means for recording workers' operations and adding them to a log with a timestamp, and a means for analyzing and organizing the recorded data and saving it in a database. This allows workers' knowledge and experience to be efficiently digitized and shared, improving work efficiency and enabling the succession of knowledge.

[0751] "Means for digitizing employee knowledge logs and thoughts" refers to a method for collecting, analyzing, organizing, and digitizing the knowledge, experience, and thoughts that employees gain in their daily work.

[0752] "Methods for creating digital clones" refers to the creation of a virtual entity that digitally mimics the actions and thoughts of employees based on their knowledge logs and thoughts.

[0753] "Methods of extracting information using AI" refers to a method of using artificial intelligence technology to search and extract necessary information from digital clones and accumulated data.

[0754] "Means for making past knowledge accessible at any time without having to remember it" refers to a method for making knowledge and information stored in a database instantly accessible when needed, without relying on a specific storage device.

[0755] The "means for recording the operations of the worker and adding them to the log together with a timestamp" is a method for recording the operations performed by the worker in real time and adding them to the log together with a timestamp indicating the date and time when the operation was performed.

[0756] "Means for analyzing and organizing the recorded data and storing it in a database" refers to a method for analyzing and organizing the recorded operation data and then storing it in a database.

[0757] The system for implementing this invention includes a set of means for digitizing employees' knowledge logs and thoughts, creating digital clones, and extracting information using AI. A specific embodiment of this system is shown below.

[0758] System Configuration

[0759] The system consists of the following main components:

[0760] 1. Data Collection Module: Collects data from the tools and systems employees use to do their jobs.

[0761] 2. Digitization module: Analyzes, organizes, and digitizes the collected data.

[0762] 3. Digital clone generation module: Generates a digital clone based on the digitized data.

[0763] 4. AI information extraction module: Uses AI technology to extract necessary information from digital clones and databases.

[0764] 5. Logging module: Records the worker's actions and adds them to the log with a timestamp.

[0765] 6. Database management module: Analyzes and organizes the recorded data and stores it in a database.

[0766] Hardware and software used

[0767] Hardware: Factory robots, tablets and smartphones used by workers

[0768] Software: Python, JSON library, database management system (e.g., MySQL)

[0769] Data processing and calculation

[0770] 1. Data collection: The data collection module collects data in real time from worker operations and business tools.

[0771] 2. Digitization: The digitization module analyzes, organizes, and digitizes the collected data. For example, when an operator changes a machine's settings, the operation details and settings are recorded.

[0772] 3. Digital Clone Generation: The digital clone generation module generates a digital clone that mimics the employee's behavior and thoughts based on the digitized data.

[0773] 4. AI Information Extraction: The AI ​​information extraction module uses AI technology to search and extract the necessary information from the generated digital clones and databases.

[0774] 5. Logging: The logging module records the worker's actions with a timestamp and adds them to the log.

[0775] 6. Database Management: The database management module analyzes and organizes the recorded data and stores it in a database.

[0776] Specific examples

[0777] For example, if an operator changes the speed setting of a factory robot, the operation details and setting value are recorded in real time and added to the log along with a timestamp. The recorded data is analyzed and organized by the digitization module and stored in a database. Later, the AI ​​information extraction module can be used to search and extract past operation history and setting values.

[0778] Prompt Sentence Examples

[0779] We are developing a system in which a factory robot records the operations of a worker. When a worker changes the robot's speed settings, please create a Python program to record the operation and the setting value. The recorded data needs to be saved in JSON format.

[0780] In this way, the system efficiently digitizes and shares workers' knowledge and experience, improving work efficiency and ensuring knowledge transfer.

[0781] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0782] Step 1:

[0783] Data collection

[0784] The server collects data in real time from the tools and systems that workers use in their work. Specifically, when a worker operates a factory robot, the details of that operation (e.g., changing the speed setting) are obtained from sensors and log files. The input is the worker's operation data, and the output is the collected raw data.

[0785] Step 2:

[0786] Digitalization

[0787] The server analyzes and organizes the collected raw data using a digitization module. Specifically, it analyzes the collected operation data, extracts the operation details and setting values, and converts them into digital format. The input is the collected raw data, and the output is analyzed and organized digital data.

[0788] Step 3:

[0789] Digital Cloning

[0790] The server generates a digital clone based on the digitized data. Specifically, it uses the digitized operational data to generate a virtual entity that mimics the worker's actions and thoughts. The input is the digitized data, and the output is the generated digital clone.

[0791] Step 4:

[0792] AI Information Extraction

[0793] The server uses an AI information extraction module to search and extract the necessary information from the digital clone and database. Specifically, it uses an AI model to search past operation history and setting values ​​and extract the necessary information. The input is the search query, and the output is the extracted information.

[0794] Step 5:

[0795] Logging

[0796] The server records the worker's operations along with a timestamp and adds them to the log. Specifically, it records the operations performed by the worker and the date and time, and adds them to a log file. The input is the worker's operation data and timestamp, and the output is an updated log file.

[0797] Step 6:

[0798] Database Management

[0799] The server analyzes and organizes the recorded data and saves it in a database. Specifically, it analyzes and organizes the data recorded by the log recording module and saves it in a database. The input is the analyzed and organized data, and the output is an updated database.

[0800] In this way, the system efficiently digitizes and shares workers' knowledge and experience, improving work efficiency and ensuring knowledge transfer.

[0801] Example 2

[0802] Next, a description will be given of Example 2 of Form 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."

[0803] In modern companies, it is important to efficiently share and utilize the knowledge and experience of employees. However, employees' knowledge and experience are accumulated individually, making it difficult to share this throughout the organization. There is also a risk that this knowledge and experience will be lost when an employee leaves the company. Furthermore, there is a need for a method to utilize employees' knowledge while taking privacy issues into consideration. To solve these issues, a system is needed that digitizes employees' knowledge and experience and provides it in a form that can be accessed at any time.

[0804] 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. In this invention, the server includes a means for digitizing the knowledge, history, and thoughts of employees, a means for creating a digital clone, a means for extracting information using a machine learning model, a means for making past knowledge accessible at any time without storing it, a means for collecting and preprocessing data, training an AI model, and evaluating and adjusting it, and a means for inputting prompt sentences into the generative AI model to utilize the digital clone function. This makes it possible to efficiently share and utilize the knowledge and experience of employees.

[0805] "Means for digitizing employee knowledge, history, and thinking" refers to a method for collecting and storing employees' work-related knowledge, past work history, and thought processes as digital data.

[0806] "Means for creating a digital clone" refers to a method for building an AI model that mimics the behavior and judgment of employees based on the knowledge, experience, and thoughts of digitized employees.

[0807] "Methods of extracting information using machine learning models" are methods that use trained AI models to provide appropriate information in response to specific questions or requests.

[0808] "Means for making past knowledge accessible at any time without having to remember it" refers to a method that allows information based on past knowledge and experience to be obtained in real time through a digital clone.

[0809] "Means for collecting, preprocessing, training an AI model, and evaluating and adjusting data" refers to a series of processes for collecting the data necessary to create a digital clone, converting that data into a format suitable for an AI model, training the model, evaluating its performance, and adjusting it as necessary.

[0810] "Means for utilizing the functions of a digital clone by inputting prompt sentences into a generative AI model" refers to a method in which a user inputs a specific question or request into an AI model as a prompt sentence, and as a result receives appropriate information or advice from the digital clone.

[0811] MODE FOR CARRYING OUT THE INVENTION

[0812] This invention relates to a system that digitizes the knowledge and experience of employees and creates digital clones. This system includes elements such as a server, a terminal, and a user, and operates as follows.

[0813] 1. Data Collection

[0814] The server collects digitized data such as employee work history, emails, chat logs, documents, etc. This includes retrieving data from databases and cloud storage. Specifically, the server retrieves work history from databases using SQL queries and downloads documents from cloud storage via APIs.

[0815] 2. Data Preprocessing

[0816] The server preprocesses the collected data. Specifically, it cleans the data (removes noise and missing values), normalizes it (transforms it to maintain data consistency), and tokenizes it (divides the text data into words and phrases). The server uses Python's pandas library to create a data frame and imputes missing values. It also tokenizes the text data using the NLTK library.

[0817] 3. Training the AI ​​model

[0818] The server uses the preprocessed data to train an AI model using a machine learning framework such as TensorFlow or PyTorch. The server defines a neural network model using TensorFlow's Keras API and trains the model using the preprocessed data as input.

[0819] 4. Evaluate and adjust the model

[0820] The server evaluates the trained AI model and makes adjustments as necessary. The evaluation involves checking the model's accuracy using actual work data from employees. The server calculates the model's evaluation indicators (e.g., precision, recall, F1 score) and adjusts the hyperparameters based on the results. The hyperparameter optimization library is used to optimize the hyperparameters.

[0821] 5. Creating a Digital Clone

[0822] The server then uses the final, adjusted AI model to generate a digital clone that can mimic the employee's knowledge, experience, and thinking. The server then stores the trained model and makes it accessible to users via an API.

[0823] 6. Use of digital clones

[0824] Users can use the digital clone's functions by entering prompts from their terminal. Specifically, when creating a proposal for a new project, the user can ask the digital clone for advice. For example, the user can enter the following prompt:

[0825] Example prompt sentence:

[0826] "Can you give me some examples of successful proposals from past projects? Also, can you give me some examples of unsuccessful proposals?"

[0827] The server receives the input prompt and passes it to the generative AI model. The generative AI model extracts the characteristics of successful proposals and the problems of unsuccessful proposals based on past data, and generates an answer. The server returns the generated answer to the user's device. The user can check the answer on their device and use it as a reference when creating a new proposal.

[0828] In this way, it becomes possible to efficiently share and utilize the knowledge and experience of employees.

[0829] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0830] Step 1: Data collection

[0831] The server collects digitized data such as employee work history, emails, chat logs, documents, etc. Specifically, the server retrieves work history from the database using SQL queries and downloads documents from cloud storage via API. The input is the connection information for the database and cloud storage, and the output is the collected raw data.

[0832] Step 2: Data Preprocessing

[0833] The server preprocesses the collected data. Specifically, it cleans the data (removes noise and missing values), normalizes it (transforms it to maintain data consistency), and tokenizes it (divides the text data into words and phrases). The input is the collected raw data, and the output is the preprocessed data. The server creates a data frame using Python's pandas library and completes missing values. It also tokenizes the text data using the NLTK library.

[0834] Step 3: Training the AI ​​model

[0835] The server uses the preprocessed data to train an AI model. Specifically, it uses machine learning frameworks such as TensorFlow and PyTorch. The input is the preprocessed data, and the output is a trained AI model. The server defines a neural network model using TensorFlow's Keras API and trains the model using the preprocessed data as input.

[0836] Step 4: Evaluate and tune the model

[0837] The server evaluates the trained AI model and makes adjustments as necessary. Specifically, it calculates the model's evaluation indicators (e.g., precision, recall, F1 score) and adjusts the hyperparameters based on the results. The input is the trained AI model and evaluation data, and the output is the adjusted AI model. The server optimizes the hyperparameters using a hyperparameter optimization library.

[0838] Step 5: Create a digital clone

[0839] The server generates a digital clone using the final adjusted AI model. The input is the adjusted AI model, and the output is the digital clone. The server stores the trained model and makes it accessible to users through an API.

[0840] Step 6: Use a digital clone

[0841] Users utilize the digital clone's functions by inputting prompts from their device. Specifically, when creating a proposal for a new project, the user can ask the digital clone for advice. The input is the prompt, and the output is the answer from the digital clone. For example, a user might input the following prompt: "Please give me some examples of successful proposals from past projects. Also, please give me some examples of unsuccessful proposals." The server receives the input prompt and passes it to the generative AI model. The generative AI model uses past data to extract the characteristics of successful proposals and the problems of unsuccessful proposals, and generates an answer. The server then sends the generated answer back to the user's device. The user can check the answer on their device and use it as a reference when creating a new proposal.

[0842] (Application example 2)

[0843] Next, a description will be given of Application Example 2 of Form 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."

[0844] In conventional factories, it was difficult to transfer the knowledge and experience of skilled workers to new workers or robots, making it difficult to automate tasks efficiently. There was also a risk that important work would be delayed if skilled workers retired or went on vacation. This led to problems with reduced productivity and quality in factories. Furthermore, there were issues with privacy and information accessibility.

[0845] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0846] In this invention, the server includes a means for digitizing employees' knowledge logs and thoughts, a means for creating digital clones, a means for extracting information using AI, a means for making past knowledge accessible at any time without memorizing it, a means for learning factory work data, a means for being installed in factory robots, a means for imitating the knowledge and experience of skilled workers, and a means for automatically performing complex tasks. This allows the knowledge and experience of skilled workers to be transferred to robots, enabling efficient automation of work. Furthermore, it improves information accessibility while taking privacy issues into consideration.

[0847] "Means for digitizing employee knowledge logs and thinking" refers to technology for recording and storing employees' knowledge, experiences, and thought processes as digital data.

[0848] The "means of creating a digital clone" refers to a technology that builds an AI model that mimics an employee's knowledge, experience, and thinking based on digitized data.

[0849] "Methods of extracting information using AI" refers to technology that uses artificial intelligence to search for and obtain the necessary information from a digital clone.

[0850] "Means for making past knowledge accessible at any time without storing it" refers to technology that allows knowledge and information stored in a digital clone to be accessed instantly when needed without being stored in a specific storage device.

[0851] "Means for learning factory work data" refers to technology that collects data on work performed in factories and trains an AI model using that data.

[0852] "Means to be installed on factory robots" refers to technology for incorporating trained AI models into robots used in factories.

[0853] "Means of imitating the knowledge and experience of skilled workers" refers to technology that uses an AI model to reproduce the advanced skills and experience of skilled workers and implement them in robots.

[0854] "Means for automating complex tasks" refers to technology that uses AI models to enable factory robots to automatically perform complex tasks in the same way as skilled human workers.

[0855] The system for implementing this invention includes a means for digitizing employees' knowledge logs and thoughts and creating digital clones. It also includes a means for extracting information using AI and making it accessible at any time without memorizing past knowledge. This allows factory work data to be learned and installed in factory robots, enabling them to imitate the knowledge and experience of skilled workers and perform complex tasks automatically.

[0856] System Program

[0857] The system's program uses Python to build an AI model using TensorFlow and Keras, and trains it on data from skilled workers. The server loads the data, builds the model, compiles the model, trains the model, and saves the model.

[0858] Processing Description

[0859] The server first loads digitized data on skilled workers' knowledge and experience. This data includes detailed logs of work performed in the factory and sensor data. Next, TensorFlow and Keras are used to build an AI model with LSTM and Dense layers. This model is designed to effectively process time-series data and learn the work patterns of skilled workers.

[0860] After the model is built, the server compiles it using the Adam optimizer and categorical cross-entropy loss function, which improves the model's training efficiency and accuracy. The server then trains the model using the data. Once the model is trained, it is installed on the factory robot.

[0861] Factory robots use these trained models to mimic the knowledge and experience of skilled workers and perform complex tasks automatically, helping maintain factory productivity and quality even when skilled workers are unavailable.

[0862] Specific examples

[0863] As a concrete example, this system can be installed on robots used on factory assembly lines. The robots can then act as digital clones of skilled workers, automatically performing complex assembly tasks, such as installing specific parts or carrying out quality inspections, exactly as skilled workers would.

[0864] Example prompts to input to the generative AI model:

[0865] "Build an AI model that automates factory assembly work based on data from skilled workers. Specifically, create a model that learns from the work data of skilled workers using LSTM and Dense layers, and install it on factory robots."

[0866] In this way, factory robots can utilize the knowledge and experience of skilled workers to perform advanced tasks.

[0867] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0868] Step 1:

[0869] The server reads the digitized data of skilled workers' knowledge and experience. Specifically, it collects detailed logs and sensor data of work performed in the factory and stores this in a database. The input is the factory work data, and the output is the digitized data stored in the database.

[0870] Step 2:

[0871] The server builds an AI model using TensorFlow and Keras. Specifically, it designs a neural network with LSTM and Dense layers and defines the model architecture. The input is the digitized data stored in Step 1, and the output is the built AI model.

[0872] Step 3:

[0873] The server compiles the model using the Adam optimizer and categorical cross-entropy loss function, which improves the model's learning efficiency and accuracy. The input is the AI ​​model built in step 2, and the output is the compiled AI model.

[0874] Step 4:

[0875] The server uses the digitized data to train the AI ​​model. Specifically, it inputs the data into the model and sets the number of epochs and batch size to perform training. The input is the digitized data saved in step 1 and the AI ​​model compiled in step 3, and the output is the trained AI model.

[0876] Step 5:

[0877] The server saves the trained AI model. Specifically, it saves the model as a file so that it can be installed on the factory robot later. The input is the AI ​​model trained in step 4, and the output is the saved AI model file.

[0878] Step 6:

[0879] The server installs the saved AI model on the factory robot. Specifically, it incorporates the model into the robot's control system, allowing the robot to utilize the learned knowledge and experience. The input is the AI ​​model file saved in step 5, and the output is a factory robot with the installed AI model.

[0880] Step 7:

[0881] Factory robots use installed AI models to mimic the knowledge and experience of skilled workers and automatically perform complex tasks, such as assembly work and quality inspections. The input is factory work instructions and sensor data, and the output is the results of the work performed.

[0882] Example 3

[0883] Next, a description will be given of Example 3 of Form Example 3. 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."

[0884] In modern companies, it is important to efficiently utilize the knowledge and experience of employees, but there is a lack of means to achieve this. There is also a need for methods to make past knowledge accessible in real time and to make the knowledge of diverse experts available 24 hours a day, 365 days a year. Furthermore, a system is needed that can simultaneously use multiple generative AI models to obtain information while taking privacy issues into consideration.

[0885] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0886] In this invention, the server includes means for digitizing employee knowledge, records, and thoughts, means for creating digital clones, means for extracting information using a generative AI model, means for making past knowledge accessible at any time without storing it, means for acquiring information by simultaneously using multiple generative AI models, means for acquiring information in real time by linking with a database, and means for considering privacy issues. This makes it possible to efficiently utilize employee knowledge and experience, acquire necessary information in real time, and access the knowledge of a variety of experts 24 hours a day, 365 days a year.

[0887] "Means for digitizing employee knowledge, records, and thoughts" refers to technology for storing and managing employees' knowledge, experiences, and thought processes in digital format.

[0888] "Means for creating a digital clone" refers to technology that uses an employee's knowledge and experience to digitally generate a virtual entity that mimics the employee's thoughts and actions.

[0889] "Means of extracting information using a generative AI model" refers to technology for asking questions or giving instructions to a generative AI model to obtain the necessary information.

[0890] "A means of making past knowledge accessible at any time without having to remember it" is a technology that searches and extracts past data in real time and provides the necessary information instantly.

[0891] "Means of obtaining information by simultaneously using multiple generative AI models" refers to a technology that simultaneously asks questions to multiple generative AI models, compares and integrates their responses, and obtains the optimal answer.

[0892] "Means of obtaining information in real time in collaboration with a database" refers to a technology that searches and extracts data stored in a database in real time, and uses that data to generate answers using a generative AI model.

[0893] "Privacy-conscious measures" are technologies that ensure the protection of personal and confidential information when using digital clones and generative AI models.

[0894] MODE FOR CARRYING OUT THE INVENTION

[0895] This invention is a system that efficiently utilizes the knowledge and experience of employees, acquires necessary information in real time, and utilizes the knowledge of various experts 24 hours a day, 365 days a year. This system is implemented using the following hardware and software.

[0896] Hardware and software used

[0897] Server: Uses a server equipped with a powerful processor and large memory capacity to run generative AI models, access databases, and process user requests.

[0898] Terminal: A device used by a user, such as a computer or smartphone. The terminal receives user input and communicates with the server.

[0899] Generative AI models: AI models for natural language processing, such as advanced generative AI models like GPT-3 and BERT.

[0900] Database: A database for storing historical knowledge and data. Use a SQL database or a NoSQL database.

[0901] Specific operation of the system

[0902] 1. The user enters a question

[0903] The user uses the device's input interface to enter questions or instructions for the generative AI model, such as "What are some successful examples from past projects?"

[0904] 2. The device sends a question to the server

[0905] The terminal sends the question entered by the user to the server using an HTTP request.

[0906] 3. The server passes the question to the AI ​​model

[0907] The server passes the received question to a generative AI model, which searches its internal knowledge base and generates an appropriate answer.

[0908] 4. The server sends the answer back to the device

[0909] The server receives the generated answer and sends it back to the user's device using an HTTP response.

[0910] 5. The device will display the answer

[0911] The terminal displays the response received from the server to the user using a GUI (Graphical User Interface).

[0912] Examples of concrete examples and prompts

[0913] Example: A user asks, "What are some success stories from past projects?"

[0914] Example prompt: "Tell me about a success story from a past project."

[0915] This system allows users to efficiently obtain the necessary information and support decision-making. It is also possible to obtain information in real time by linking with a database, or to obtain information by simultaneously using multiple generative AI models. This makes it possible to make the most of employees' knowledge and experience and improve the competitiveness of the company. The flow of the specific processing in Example 3 will be described with reference to FIG. 15.

[0916] Step 1:

[0917] The user enters a question.

[0918] The user uses the device's input interface to input questions or instructions to the generative AI model. For example, they might type, "What are some successful examples of past projects?" The input question is saved as text data on the device.

[0919] Step 2:

[0920] The terminal sends a question to the server.

[0921] The terminal sends the question entered by the user to the server as an HTTP request. The data sent is the user's question text. The server receives this request and analyzes the question text.

[0922] Step 3:

[0923] The server passes the question to the AI ​​model.

[0924] The server passes the received question text to the generative AI model, which searches its internal knowledge base and generates an appropriate answer. Specifically, the AI ​​model uses natural language processing algorithms to analyze the question and extract relevant information. The output is the generated answer text.

[0925] Step 4:

[0926] The server sends a response back to the terminal.

[0927] The server receives the generated answer text and returns it to the user's device as an HTTP response. The returned data is the answer text generated by the AI ​​model. The device receives this response and analyzes the answer text.

[0928] Step 5:

[0929] The device will display the answer.

[0930] The terminal displays the answer text received from the server to the user. A GUI (Graphical User Interface) is used for display. Specifically, the answer text is displayed on the screen so that the user can check it.

[0931] (Application example 3)

[0932] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0933] Modern manufacturing requires rapid and accurate responses to problems that occur during the manufacturing process. However, with conventional systems, it is difficult to acquire past knowledge and success stories in real time, and it is also difficult to integrate solutions from multiple perspectives. Furthermore, there is a lack of systems that allow operators to operate hands-free, making it difficult to achieve efficient manufacturing support.

[0934] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes means for digitizing employee knowledge logs and thoughts, means for creating digital clones, means for extracting information using AI, means for making past knowledge accessible at any time without memorizing it, means for acquiring information in real time about problems that occur during the manufacturing process and providing optimal manufacturing procedures and troubleshooting methods, means for using multiple AI models to provide solutions from different perspectives and select the optimal method, and means for responding to voice questions and instructions and enabling hands-free operation. This enables quick and accurate responses to problems that occur during the manufacturing process and enables efficient manufacturing support.

[0935] "Means for digitizing employee knowledge logs and thoughts" refers to technology for recording and storing employees' knowledge, experiences, and ideas in digital form.

[0936] "Methods for creating digital clones" are technologies that generate digital models that mimic the knowledge and thought patterns of a particular person.

[0937] "Means of extracting information using AI" refers to technology that uses artificial intelligence to search for and obtain the necessary information.

[0938] "A means of making past knowledge accessible at any time without having to remember it" is technology that allows past data to be searched and retrieved in real time.

[0939] "A means of obtaining real-time information on problems that occur during the manufacturing process and providing optimal manufacturing procedures and troubleshooting methods" refers to technology that instantly detects problems that occur during manufacturing and provides solutions.

[0940] "A means of using multiple AI models to provide solutions from different perspectives and select the optimal method" is a technology that uses multiple artificial intelligence models simultaneously to compare and integrate solutions from different approaches.

[0941] "Means for responding to voice inquiries and instructions and enabling hands-free operation" refers to technology that uses voice recognition technology to enable users to operate the system without using their hands.

[0942] A system for implementing this invention has the following configuration: a server includes means for digitizing employee knowledge logs and thoughts, means for creating digital clones, means for retrieving information using AI, means for making past knowledge accessible at any time without storing it, means for obtaining information in real time about problems that occur during the manufacturing process and providing optimal manufacturing procedures and troubleshooting methods, means for using multiple AI models to provide solutions from different perspectives and select the optimal method, and means for responding to voice questions and instructions and enabling hands-free operation.

[0943] Program processing explanation

[0944] The server runs the program using Python. It uses SQLite as the database and the OpenAI API for the AI ​​model. Specifically, it performs the following processes:

[0945] 1. How to digitize employee knowledge logs and thinking:

[0946] The server digitally stores the knowledge and experience entered by employees, which can be entered by text or voice.

[0947] 2. How to create a digital clone:

[0948] The server uses the collected digital data to create a digital clone that mimics the knowledge and thought patterns of a specific employee, allowing that knowledge to be used even when the employee is absent.

[0949] 3. How to use AI to extract information:

[0950] The server receives questions from users and uses the AI ​​model to search for and retrieve the necessary information. For example, if a user asks, "What are some successful examples of past projects?", the AI ​​model will pull relevant information from the database.

[0951] 4. A way to make past knowledge accessible at any time without having to memorize it:

[0952] The server allows users to search and retrieve historical data in real time, allowing them to instantly obtain the information they need.

[0953] 5. A means of obtaining real-time information about problems that occur during the manufacturing process and providing optimal manufacturing procedures and troubleshooting methods:

[0954] The server detects problems that occur on the production line and uses databases and AI models to provide optimal solutions. For example, if you ask, "What caused the production line to stop and what should be done?", it will suggest solutions based on past data and AI models.

[0955] 6. Using multiple AI models to provide solutions from different perspectives and select the best method:

[0956] The server uses multiple AI models simultaneously to compare and integrate solutions from different perspectives, allowing it to select the optimal solution.

[0957] 7. Means of responding to voice prompts and commands, enabling hands-free operation:

[0958] The server utilizes voice recognition technology to allow users to operate the system hands-free, improving efficiency on the manufacturing floor.

[0959] Examples of concrete examples and prompts

[0960] For example, if a production line at a factory stops, an operator can ask, "What caused the production line to stop and what was the solution?" The server will search a past database and provide relevant success stories and solutions. If the database does not contain the information, it will ask an AI model and provide the optimal solution.

[0961] Example prompt sentence:

[0962] Prompt: "What caused the production line to stop and what was the solution?"

[0963] In this way, the server can provide optimal manufacturing support in real time.

[0964] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0965] Step 1:

[0966] A user inputs a question using a terminal. For example, "What is the cause of the production line stopping and what is the solution?" This input is sent to the server in text format.

[0967] Step 2:

[0968] The server analyzes the received question and extracts relevant keywords using natural language processing technology. For example, keywords such as "production line," "stop," "cause," and "measures" are extracted.

[0969] Step 3:

[0970] The server searches an SQLite database based on the extracted keywords. The database contains data on past success stories and troubleshooting. The relevant data is retrieved as search results.

[0971] Step 4:

[0972] The server provides the user with information retrieved from the database, such as "A similar problem occurred in a past project and the following measures were effective."

[0973] Step 5:

[0974] If the database does not contain relevant information, the server uses OpenAI's API to ask the AI ​​model a question. The question is generated as a prompt and sent to the AI ​​model. For example, a prompt might be generated: "What caused the production line to stop, and what measures can be taken?"

[0975] Step 6:

[0976] The AI ​​model generates an answer based on the received prompt. This answer is returned to the server in text format. For example, the generated answer might be, "The cause of the production line stoppage may be a machine failure or a problem with the power supply. As a countermeasure, it is necessary to inspect the machine and check the power supply."

[0977] Step 7:

[0978] The server provides the answer from the AI ​​model to the user, who can view this information through their device.

[0979] Step 8:

[0980] When a user asks a question by voice, the device uses speech recognition technology to convert the speech into text, which is then sent to the server for processing from step 2 onwards.

[0981] Step 9:

[0982] The server uses multiple AI models to provide solutions from different perspectives. For example, it asks the same question to different AI models, compares and integrates their answers, and provides this integrated answer to the user.

[0983] Step 10:

[0984] The server collects feedback from users and uses it to improve the system. For example, users can rate the solutions provided, and the server updates the AI ​​model and database based on that evaluation data.

[0985] 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.

[0986] "Example 1"

[0987] One embodiment of the present invention is a system incorporating an emotion engine. This system recognizes a user's emotions and adjusts the digital clone's responses based on those emotions. Specifically, when a user expresses an emotion such as happiness or sadness, the emotion engine recognizes the emotion and provides a response appropriate to that emotion to the digital clone. For example, if the user expresses happiness, the digital clone will provide a positive response that is tailored to the happiness.

[0988] "Example 2"

[0989] The emotion engine also adjusts the questions posed to the other person's clone based on the user's emotions. Specifically, if the user expresses emotions such as anger or frustration, the emotion engine recognizes the emotion and adjusts the questions posed to the other person's clone. For example, if the user expresses anger, the questions posed to the other person's clone are adjusted to provide information or advice to alleviate the anger.

[0990] "Example 3"

[0991] Furthermore, the emotion engine learns changes in the user's emotions and adjusts the digital clone's responses and the questions it asks other clones based on the learning results. Specifically, the emotion engine learns changes in the user's emotions over time and adjusts the digital clone's responses and the questions it asks other clones based on the learning results. For example, if the user always shows anger in a certain situation, the emotion engine learns that pattern and, when a similar situation occurs, adjusts the digital clone's responses and the questions it asks other clones in advance.

[0992] The processing flow of each embodiment will be described below.

[0993] "Example 1"

[0994] Step 1: The user uses the system to input information, which indicates the user's feelings.

[0995] Step 2: The emotion engine analyzes the user's input information and recognizes the user's emotion.

[0996] Step 3: Adjust the digital clone's response based on the emotions recognized by the emotion engine.

[0997] "Example 2"

[0998] Step 1: The user uses the system to input information, which indicates the user's feelings.

[0999] Step 2: The emotion engine analyzes the user's input information and recognizes the user's emotion.

[1000] Step 3: Adjust the questions posed to the other clone based on the emotions recognized by the emotion engine.

[1001] "Example 3"

[1002] Step 1: The user uses the system to input information, which indicates the user's feelings.

[1003] Step 2: The emotion engine analyzes the user's input information and recognizes the user's emotion.

[1004] Step 3: The emotion engine learns the user's emotional changes.

[1005] Step 4: Based on what the emotion engine has learned, adjust the digital clone's responses and the questions it asks other people's clones.

[1006] Example 1

[1007] Next, a description will be given of Example 1 of Form 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."

[1008] In modern companies, it is important for employees to effectively accumulate and share the knowledge, experience, and thoughts they gain in their daily work. However, there is a lack of means to digitize this information and properly manage and utilize it. Furthermore, there is no system that provides appropriate responses based on user emotions, which can lead to a decline in employee motivation and the quality of communication. Furthermore, there is a need for a system that allows employees to access other people's knowledge 24 hours a day, 365 days a year, while taking privacy issues into consideration.

[1009] 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.

[1010] In this invention, the server includes a means for digitizing employees' knowledge logs and thoughts, a means for creating digital clones, a means for extracting information using artificial intelligence, a means for collecting data from business tools, a means for analyzing and organizing the collected data, a means for recognizing users' emotions and adjusting responses based on those emotions, and a means for generating appropriate responses using a generative AI model. This enables employees' knowledge and experience to be effectively digitized and shared. Furthermore, providing appropriate responses based on users' emotions can improve employee motivation and the quality of communication. Furthermore, a system is realized that allows access to other people's knowledge 24 hours a day, 365 days a year, while taking privacy issues into consideration.

[1011] "Means for digitizing employee knowledge logs and thoughts" refers to technologies and methods for digitally recording and storing the knowledge, experience, and thoughts that employees gain in their daily work.

[1012] "Means of creating a digital clone" refers to technologies and methods that create a virtual digital persona based on an employee's knowledge and experience.

[1013] "Means of extracting information using artificial intelligence" refers to techniques and methods for using artificial intelligence technology to search for and extract necessary information from accumulated data.

[1014] "Means of collecting data from business tools" refers to technologies and methods for automatically obtaining data from business tools and systems used by employees.

[1015] "Means for analyzing and organizing collected data" refers to the techniques and methods used to analyze, organize, and digitize collected data.

[1016] "Means for recognizing a user's emotions and adjusting a response based on those emotions" refers to techniques or methods for analyzing a user's emotions and generating a response appropriate to those emotions.

[1017] "Means for generating appropriate responses using a generative AI model" refers to techniques or methods that use a generative AI model to generate appropriate responses based on user input and emotions.

[1018] This invention is a system that digitizes employees' knowledge logs and thoughts and creates digital clones. The system uses artificial intelligence to extract information and make it accessible at any time without memorizing past knowledge. It also has the ability to recognize users' emotions and adjust responses based on those emotions.

[1019] Hardware and software used

[1020] Hardware

[1021] Server: Analyzes and organizes data, creates digital clones, recognizes emotions, and generates responses.

[1022] Terminal: Collects data, accepts user input, and displays responses.

[1023] software

[1024] Data analysis software: Python, R

[1025] Sentiment analysis API: IBM Watson, Microsoft Azure

[1026] Work tools: Office suite, Slack

[1027] Generative AI model: A specific generative AI model (e.g., GPT-3)

[1028] Data collection and analysis

[1029] The device collects data from the work tools used by employees (e.g., office suites, Slack). For example, it periodically scans spreadsheet files in office suites and Slack message logs to acquire data. The collected data is encrypted and sent to a server using the HTTPS protocol.

[1030] The server analyzes the received data using the Python pandas library. For example, it converts the contents of the spreadsheet file into a data frame and removes unnecessary columns. The analyzed data is then converted into a format that is easy to store in a database (e.g., JSON format).

[1031] Emotion Recognition and Response Generation

[1032] When a user provides input to the system, the device sends that input to the emotion engine. The emotion engine uses emotion analysis APIs from IBM Watson and Microsoft Azure to recognize the user's emotions. For example, if a user enters "The project was a success. I'm very happy," the emotion engine will recognize the emotion of "joy" from that text.

[1033] Based on the emotion data obtained from the emotion engine, the server inputs the prompt sentence "The user reports the success of the project and expresses his joy. Please generate an appropriate response" to the generative AI model. The generative AI model generates the response "Congratulations! That's a great achievement."

[1034] Providing a response

[1035] The terminal displays the response received from the server to the user, who sees a response from the digital clone saying, "Congratulations! That's a great job!"

[1036] Specific examples

[1037] For example, when a user reports the success of a project, the system recognizes the user's joy through its emotion engine, and the digital clone responds, "Congratulations! That's a great achievement."

[1038] Prompt Sentence Examples

[1039] "The project was a success. I'm very happy."

[1040] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1041] Step 1:

[1042] The device collects data from the work tools used by employees (e.g., office suites, Slack). Specifically, the device periodically scans the log data and documents of these tools to obtain the necessary data. For example, a script that automatically collects data is executed at the end of each day. The input is data from the work tools, and the output is the collected data.

[1043] Step 2:

[1044] The terminal encrypts the collected data and sends it to the server using the HTTPS protocol. Specifically, it sends the data via a POST request using the Python requests library. The input is the collected data, and the output is the encrypted data.

[1045] Step 3:

[1046] The server analyzes the received data. Specifically, it uses the Python pandas library to cleanse and classify the data. For example, it converts the contents of a spreadsheet file into a data frame and removes unnecessary columns. The input is the encrypted data, and the output is the analyzed data.

[1047] Step 4:

[1048] The server organizes and digitizes the analyzed data. Specifically, it converts it into a format that is easy to store in a database (e.g., JSON format). For example, it uses the pandas to_json method to convert a data frame into JSON format. The input is the analyzed data, and the output is the digitized data.

[1049] Step 5:

[1050] When a user provides input to the system, the device sends the input to the emotion engine. For example, if a user types "The project was a success. I'm very happy," the text is sent to the emotion engine. The input is the user's text input, and the output is the data sent to the emotion engine.

[1051] Step 6:

[1052] The emotion engine uses IBM Watson and Microsoft Azure sentiment analysis APIs to recognize user emotions. Specifically, it analyzes emotions such as "joy" from text. The input is the data sent to the emotion engine, and the output is the recognized emotion data.

[1053] Step 7:

[1054] The server inputs the emotion data obtained from the emotion engine as a prompt to the generative AI model. Specifically, it inputs "The user reports the success of the project and expresses joy. Please generate an appropriate response." The input is emotion data, and the output is a prompt to the generative AI model.

[1055] Step 8:

[1056] The generative AI model generates an appropriate response based on the prompt sentence, specifically, "Congratulations! That's a great job!" The input is the prompt sentence, and the output is the generated response.

[1057] Step 9:

[1058] The terminal displays the response received from the server to the user, specifically displaying the message "Congratulations! Great job!" on the user's screen. The input is the response generated, and the output is what is displayed to the user.

[1059] (Application example 1)

[1060] Next, a description will be given of Application Example 1 of Form 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."

[1061] In conventional factories, there is a lack of effective means to digitize and share the knowledge and thoughts of workers. Furthermore, there is no system that can recognize workers' emotions and provide appropriate responses accordingly, which leads to problems of reduced work efficiency and quality of communication. Furthermore, there are privacy issues and insufficient means to access the knowledge of diverse experts 24 hours a day, 365 days a year.

[1062] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means. In this invention, the server includes means for digitizing employees' knowledge logs and thoughts, means for creating digital clones, means for extracting information using AI, means for making past knowledge accessible at any time without memorizing it, means for recognizing emotions, means for adjusting the digital clone's responses based on emotions, and means for digitizing factory workers' knowledge logs and thoughts and providing responses according to their emotions. This enables the effective digitization and sharing of workers' knowledge and thoughts, improving work efficiency and the quality of communication. It also makes it possible to access the knowledge of diverse experts 24 hours a day, 365 days a year, while taking privacy issues into consideration.

[1063] "Means for digitizing employee knowledge logs and thoughts" refers to a method for collecting, analyzing, organizing, and digitizing the knowledge, experience, and thoughts that employees gain in their daily work.

[1064] "Methods for creating digital clones" are methods for generating a virtual entity that digitally mimics the actions and thoughts of an employee based on the employee's knowledge log and thoughts.

[1065] "Means of extracting information using AI" refers to a method that uses artificial intelligence technology to search for necessary information from digital clones or knowledge logs and provide it to users.

[1066] "Means for making past knowledge accessible at any time without having to remember it" refers to a method for temporarily storing knowledge accumulated in the past so that it can be accessed immediately when needed.

[1067] "Means for recognizing emotions" refers to technology for analyzing and recognizing emotions from the user's facial expressions, tone of voice, etc.

[1068] The "means for adjusting the response of the digital clone based on emotions" is a method for the digital clone to generate an appropriate response depending on the recognized emotions of the user.

[1069] "A means of digitizing the knowledge, logs, and thoughts of factory workers and providing responses based on their emotions" is a method for digitizing the knowledge and thoughts that factory workers gain in the course of their work and providing appropriate responses based on the workers' emotions.

[1070] As an embodiment of this invention, we will build a system that digitizes the knowledge, logs, and thoughts of factory workers and provides responses according to their emotions. A specific embodiment of this system is shown below.

[1071] System configuration

[1072] The system includes the following major hardware and software components:

[1073] Hardware: Factory robots, cameras, microphones

[1074] Software: EmotionEngine (emotion recognition engine), KnowledgeLog (knowledge log management system), DigitalClone (digital clone response generation system)

[1075] Program processing explanation

[1076] emotion recognition

[1077] The server uses a camera and microphone to capture the worker's facial expressions and tone of voice, and the Emotion Engine analyzes this data to recognize the worker's emotions.

[1078] Knowledge Log Updates

[1079] The server records the newly acquired knowledge and experience of workers in KnowledgeLog, which organizes and digitizes this data.

[1080] Response Generation

[1081] The server uses DigitalClone to generate an appropriate response based on the recognized emotion and the latest knowledge log, and DigitalClone tailors the response to the emotion and provides it to the worker.

[1082] Specific examples

[1083] For example, if a factory worker discovers an improvement in a new manufacturing process and expresses joy, the system might:

[1084] 1. Cameras and microphones capture the facial expressions and voices of workers.

[1085] 2. The Emotion Engine analyzes this data and recognizes that the worker is expressing the emotion of joy.

[1086] 3. KnowledgeLog records new knowledge (new manufacturing process improvements) from workers.

[1087] 4. DigitalClone generates a response based on the perceived sentiment and the latest knowledge log, such as:

[1088] "That's a great discovery! Let's record the improvement immediately and share it with other workers."

[1089] Prompt Sentence Examples

[1090] An example of a prompt sentence to input to the generative AI model is as follows:

[1091] "If a worker discovers an improvement in a new manufacturing process and expresses joy, what response does it generate?"

[1092] In this way, the system effectively digitizes and shares the knowledge and thoughts of workers, improving work efficiency and the quality of communication. It also makes it possible to access the knowledge of a variety of experts 24 hours a day, 365 days a year, while taking privacy issues into consideration.

[1093] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1094] Step 1:

[1095] The server uses a camera and microphone to capture the worker's facial expressions and tone of voice. It receives real-time video and audio data of the worker as input. This data is sent to the Emotion Engine, where it undergoes preprocessing for emotion recognition. Specifically, it extracts facial features from the video data and analyzes the tone and pitch of the voice from the audio data.

[1096] Step 2:

[1097] The server uses the Emotion Engine to recognize the worker's emotions from the captured data. As input, it uses preprocessed facial feature data and voice analysis data. The Emotion Engine analyzes these data and identifies the worker's emotions (e.g., joy, sadness, surprise, etc.). As output, recognized emotion data is generated.

[1098] Step 3:

[1099] The server records the new knowledge and experience gained by workers in KnowledgeLog. As input, it receives text data of the new knowledge and experience reported by workers. KnowledgeLog organizes and digitizes this data. Specifically, it stores the text data in a structured database and converts it into a searchable format. As output, it generates updated knowledge log data.

[1100] Step 4:

[1101] The server uses DigitalClone to generate an appropriate response based on the recognized emotion and the latest knowledge log. It uses emotion data and knowledge log data as input. DigitalClone analyzes these data and generates a response based on the emotion. Specifically, it adjusts the tone of the response based on the emotion data and creates a response with specific content based on the knowledge log data. As output, the generated response message is provided to the worker.

[1102] Step 5:

[1103] The server provides the generated response message to the worker. As input, it receives the response message generated by DigitalClone. Specifically, it sends the response message in text format to the worker's device or provides it audibly using speech synthesis technology. As output, the worker can receive and confirm the response message.

[1104] In this way, the system effectively digitizes and shares the knowledge and thoughts of workers, improving work efficiency and the quality of communication. It also makes it possible to access the knowledge of a variety of experts 24 hours a day, 365 days a year, while taking privacy issues into consideration.

[1105] Example 2

[1106] Next, a description will be given of Example 2 of Form 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."

[1107] With conventional systems, it was difficult to effectively digitize employee knowledge and experience and quickly retrieve the necessary information. Furthermore, there were problems with not being able to provide appropriate responses based on the user's emotions, which did not improve the user experience. Furthermore, there were privacy issues and a lack of means to access the knowledge of diverse experts 24 hours a day, 365 days a year.

[1108] 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.

[1109] In this invention, the server includes a means for digitizing employee knowledge, experience, and thinking, a means for creating a digital clone, a means for training data using a machine learning framework, a means for recognizing user emotions and adjusting question content, a means for generating appropriate responses using a generative AI model, and a means for making past knowledge accessible at any time without storing it. This makes it possible to effectively digitize employee knowledge and experience and provide appropriate responses based on user emotions. Furthermore, while taking privacy issues into consideration, the knowledge of a variety of experts can be accessed 24 hours a day, 365 days a year.

[1110] "Means for digitizing employee knowledge, experience, and thinking" refers to a method for collecting and storing employees' work-related knowledge, experience, and thought processes as digital data.

[1111] "Means for creating a digital clone" refers to a method for generating an AI model that mimics the behavior and judgment of an employee based on the knowledge, experience, and thoughts of the digitized employee.

[1112] "Methods of training data using machine learning frameworks" refers to a method of training an AI model based on collected data using machine learning frameworks such as TensorFlow and PyTorch.

[1113] "Means for recognizing user emotions and adjusting questions" refers to a method for analyzing the user's voice and text input to identify emotions and then changing the questions posed to the AI ​​model according to those emotions.

[1114] "Means for generating appropriate responses using a generative AI model" refers to a method for generating appropriate responses to user questions using a generative AI model (e.g., GPT-3).

[1115] "Means for making past knowledge accessible at any time without memorizing it" refers to a method for making past knowledge accessible immediately when needed without temporarily storing it.

[1116] "Privacy-conscious measures" are methods to prevent personal or confidential information from being leaked when extracting information from a digital clone.

[1117] "A means of accessing the knowledge of diverse experts 24 hours a day, 365 days a year" is a method that allows necessary information to be extracted at any time from multiple digital clones with different specialized knowledge.

[1118] MODE FOR CARRYING OUT THE INVENTION

[1119] This invention is a system for digitizing the knowledge, experience, and thoughts of employees and creating digital clones. A specific embodiment of this system will be described below.

[1120] Digital cloning

[1121] The server first collects data to digitize employees' knowledge, experience, and thoughts. This data includes employees' work history, email and chat logs, project documents, etc. The server automatically collects data from email servers and project management tools.

[1122] The server then preprocesses the collected data, specifically cleaning, normalizing, and tokenizing it. It uses the Python pandas library to impute missing values ​​and removes unnecessary strings using regular expressions.

[1123] The server then uses the preprocessed data to train an AI model using machine learning frameworks such as TensorFlow and PyTorch. The server uses TensorFlow to build a neural network and trains the model using employee data as input.

[1124] Finally, the server evaluates the performance of the trained model and tunes the hyperparameters as needed. It evaluates the precision and recall of the model and finds the optimal hyperparameters using grid search or random search.

[1125] Adjusting question content using an emotion engine

[1126] The device is equipped with an emotion engine to recognize the user's emotions. It analyzes the user's voice and text input to identify emotions. The device converts the user's voice into text using a speech recognition API (e.g., Google Cloud Speech-to-Text) and analyzes emotions using a natural language processing (NLP) model.

[1127] If the user expresses emotions such as anger or frustration, the device will recognize that emotion and adjust the questions it asks the other person's clone. For example, if the user types, "This project isn't progressing at all!", the emotion engine will recognize the anger and ask the other person's clone, "Please give me some specific advice to help the project progress smoothly."

[1128] The server generates an appropriate response from the clone of another person based on the adjusted question. The generated response is sent to the device. The server uses a generative AI model (e.g., GPT-3) to generate an appropriate response to the question and sends it to the device.

[1129] Specific examples

[1130] For example, if an employee has expertise in project management, an AI model can be created that learns from that employee's past project data and communication history to generate appropriate answers to questions about project management.

[1131] Example prompt: "Tell me how to effectively manage project progress."

[1132] This system effectively digitizes the knowledge and experience of employees, enabling them to provide appropriate responses based on user sentiment. It also makes it possible to access the knowledge of a wide range of experts 24 hours a day, 365 days a year, while taking privacy issues into consideration.

[1133] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1134] Step 1:

[1135] Data collection

[1136] The server collects data to digitize employees' knowledge, experience, and thoughts. As input, it receives employee work history, email and chat logs, project documents, etc. The server automatically collects data from email servers and project management tools and stores it in a database. The output is the collected dataset.

[1137] Step 2:

[1138] Data Preprocessing

[1139] The server preprocesses the collected data. It uses the dataset collected in step 1 as input. Specifically, it uses Python's pandas library to fill in missing values ​​in the data and removes unnecessary strings using regular expressions. The output is a preprocessed, clean dataset.

[1140] Step 3:

[1141] Model training

[1142] The server trains an AI model using the preprocessed data. It uses the dataset preprocessed in step 2 as input. Specifically, it uses a machine learning framework such as TensorFlow or PyTorch to build a neural network and train the data. The output is a trained AI model.

[1143] Step 4:

[1144] Model evaluation and tuning

[1145] The server evaluates the performance of the trained model and tunes the hyperparameters as necessary. The input is the AI ​​model trained in step 3 and the evaluation dataset. Specifically, it evaluates the precision and recall of the model and finds the optimal hyperparameters using grid search and random search. The output is an optimized AI model.

[1146] Step 5:

[1147] emotion recognition

[1148] The device launches an emotion engine to recognize the user's emotions. It receives the user's voice and text input as input. Specifically, it converts the user's voice into text using a speech recognition API (e.g., Google Cloud Speech-to-Text) and analyzes the emotion using a natural language processing (NLP) model. The output is the recognized user's emotion data.

[1149] Step 6:

[1150] Adjusting the content of questions

[1151] The device adjusts the questions it asks the other person's clone based on the recognized emotion. The input is the user's emotion data recognized in step 5. Specifically, if the user inputs "This project isn't progressing at all!", the emotion engine recognizes the anger and asks the other person's clone, "Please give me some specific advice to help the project progress smoothly." The output is the adjusted questions.

[1152] Step 7:

[1153] Response Generation

[1154] The server generates an appropriate response from the clone of another person based on the adjusted question. The question adjusted in step 6 is used as input. Specifically, it uses a generative AI model (e.g., GPT-3) to generate an appropriate response to the question and sends it to the device. The output is the generated response.

[1155] (Application example 2)

[1156] Next, a description will be given of Application Example 2 of Form 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."

[1157] Traditional security services struggle to effectively utilize the knowledge and experience of their employees, and require flexible responses based on user sentiment. Furthermore, the lack of a means to provide real-time security advice and support makes it difficult to improve user satisfaction.

[1158] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1159] In this invention, the server includes means for digitizing employee knowledge, logs, and thoughts, means for creating a digital clone, means for extracting information using AI, means for making past knowledge accessible at any time without memorizing it, means for recognizing a user's emotions and adjusting the questions posed to the digital clone based on those emotions, means for providing security advice and support in real time, and means for being installed on a smart device. This makes it possible to effectively utilize the knowledge and experience of employees, respond flexibly to user emotions, and provide security advice and support in real time.

[1160] "Means for digitizing employee knowledge logs and thinking" refers to technology that records and stores employees' knowledge, experiences, and thought patterns as digital data.

[1161] "Means for creating digital clones" refers to technology that generates digital agents that mimic the behavior and thinking of employees based on their digitized knowledge logs and thoughts.

[1162] "Means of extracting information using AI" refers to a technique that uses artificial intelligence technology to search for and obtain the necessary information from a digital clone.

[1163] "Means for making past knowledge accessible at any time without storing it" refers to technology that allows the knowledge and information held by a digital clone to be accessed instantly when needed without being stored in a specific storage device.

[1164] "Means for recognizing the user's emotions and adjusting the questions asked to the digital clone based on those emotions" refers to technology that analyzes the user's emotions from their facial expressions and voice, and changes the information and advice provided to the digital clone based on those emotions.

[1165] "Means for providing security advice and support in real time" refers to technology that provides immediate and appropriate advice and support for security issues faced by users.

[1166] "Means to be installed on smart devices" refers to technology that is installed and operates as an application on portable electronic devices such as smartphones and head-mounted displays.

[1167] A system for carrying out this invention has the following configuration: a server including means for digitizing employees' knowledge logs and thoughts, means for creating digital clones, means for retrieving information using AI, means for making past knowledge accessible at any time without storing it, means for recognizing a user's emotions and adjusting questions to be asked of the digital clone based on those emotions, means for providing security advice and support in real time, and means installed on a smart device.

[1168] Hardware and Software Configuration

[1169] 1. Hardware:

[1170] Smartphones (e.g. iPhone, Android devices)

[1171] Head-mounted displays (e.g. Microsoft HoloLens)

[1172] 2. Software:

[1173] AI model building: TensorFlow, PyTorch

[1174] Emotion recognition engine: Microsoft Azure Emotion API

[1175] Database: MySQL

[1176] Frontend: React Native, 3D modeling software

[1177] Data processing and calculation

[1178] The server digitizes employees' knowledge logs and thoughts, and creates a digital clone based on these. The digital clone uses an AI model to mimic the employee's behavior and thoughts. When a user accesses the system through a smart device, an emotion recognition engine analyzes the user's emotions from their facial expressions and voice. The analyzed emotional data is input into the AI ​​model, which then provides appropriate information and advice based on the user's emotions.

[1179] Specific examples

[1180] For example, consider a case where a user accesses a security assistant on a smartphone and asks a security question. If the user feels anxious, the emotion recognition engine analyzes the emotion, and the AI ​​model provides specific advice to alleviate the anxiety. The following is an example of a prompt sentence:

[1181] "How should a security assistant respond if a user is angry?"

[1182] "What advice should we offer users if they are feeling anxious?"

[1183] In this way, the server can provide security advice and support according to the user's feelings in real time, which can improve user satisfaction.

[1184] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1185] Step 1:

[1186] The server digitizes employees' knowledge logs and thoughts. Specifically, it collects data on employees' past security response history and knowledge base, and stores this as digital data. The input is employees' knowledge logs and thoughts, and the output is digitized data.

[1187] Step 2:

[1188] The server creates a digital clone based on the digitized data. Specifically, it uses an AI model to generate a digital agent that mimics the employee's behavior and thoughts. The input is the digitized data, and the output is the digital clone.

[1189] Step 3:

[1190] Users access the system through smart devices, specifically smartphones or head-mounted displays, to connect to the security assistant. The input is the user's access request, and the output is the connection to the system.

[1191] Step 4:

[1192] The device sends the user's facial expressions and voice to the emotion recognition engine. Specifically, it captures the user's facial expressions and voice using the smart device's camera and microphone and sends them to the emotion recognition engine. The input is the user's facial expressions and voice data, and the output is data sent to the emotion recognition engine.

[1193] Step 5:

[1194] The server uses an emotion recognition engine to analyze the user's emotions. Specifically, it uses the Microsoft Azure Emotion API to analyze emotions from the user's facial expressions and voice. The input is the user's facial expression and voice data, and the output is the analyzed emotional data.

[1195] Step 6:

[1196] The server adjusts the questions it asks the digital clone based on the analyzed emotional data. Specifically, if the user shows anger or anxiety, it changes the digital clone's responses to provide information or advice to alleviate those emotions. The input is the analyzed emotional data, and the output is the adjusted questions.

[1197] Step 7:

[1198] The server provides real-time security advice and support based on the tailored questions. Specifically, it uses an AI model to generate appropriate advice and information and provide it to the user. The input is the tailored questions, and the output is advice and support provided in real time.

[1199] Step 8:

[1200] The terminal displays the advice and support provided by the server to the user. Specifically, it displays appropriate information to the user through the screen of the smart device or a head-mounted display. The input is the advice and support from the server, and the output is the information displayed to the user.

[1201] Example 3

[1202] Next, a description will be given of Example 3 of Form Example 3. 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."

[1203] With conventional systems, it was difficult to effectively digitize employee knowledge and thinking and quickly retrieve the necessary information. Furthermore, there was a lack of means to make past knowledge accessible in real time without having to memorize it, so users had to spend a lot of time and effort to obtain the information they needed. Furthermore, there were limited ways to simultaneously access the knowledge of multiple experts, making it difficult for users to obtain information from diverse perspectives.

[1204] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1205] In this invention, the server includes: a means for a user to input a question or instruction; a means for the terminal to send the user's input to the server; a means for the server to receive the input and send a prompt to an appropriate generative AI model; a means for the generative AI model to generate an answer based on the prompt; a means for the server to receive the answer from the generative AI model and complement the information as needed in cooperation with a database; a means for the server to send the final answer to the terminal; and a means for the terminal to display the answer to the user, thereby enabling the user to quickly and efficiently obtain the information they need.

[1206] "Means for digitizing employee knowledge logs and thinking" is a means for recording and storing employees' knowledge, experiences, and thought processes in digital form.

[1207] "Means for creating a digital clone" refers to the creation of a virtual entity that digitally mimics the actions and thoughts of an employee based on that employee's knowledge log and thoughts.

[1208] "Means of extracting information using AI" refers to the use of artificial intelligence technology to search for and retrieve necessary information from digital clones and databases.

[1209] "Means for making past knowledge accessible at any time without having to remember it" refers to means for searching and extracting past data and knowledge in real time, allowing users to access it immediately when needed.

[1210] The "means for the user to input questions or instructions" refers to a means for providing an interface for the user to input questions or instructions to the system.

[1211] The "means by which the terminal transmits user input to the server" refers to a communication means for transmitting questions or instructions input by the user to the server.

[1212] "Means for the server to receive input and send prompt sentences to the appropriate generative AI model" refers to means for the server to analyze input received from the user and send prompt sentences to the appropriate generative AI model.

[1213] "Means by which a generative AI model generates an answer based on a prompt sentence" refers to means by which a generative AI model generates an appropriate answer from pre-trained data based on a prompt sentence.

[1214] "Means for the server to receive answers from the generative AI model and, if necessary, complement the information in cooperation with the database" refers to means for the server to receive answers from the generative AI model and, if necessary, obtain additional information from the database to complement the answers.

[1215] The "means by which the server transmits the final answer to the terminal" refers to the means by which the server transmits the final generated answer to the user's terminal.

[1216] The "means for the terminal to display the answer to the user" refers to the means for the user's terminal to display the answer received from the server to the user.

[1217] This invention is a system that digitizes employees' knowledge logs and thoughts, creates digital clones, and uses AI to extract information. Furthermore, it makes past knowledge accessible at any time without memorizing it, and allows the knowledge of multiple experts to be used simultaneously.

[1218] Hardware and software used

[1219] The system is implemented using the following hardware and software:

[1220] Server: Use a server with a powerful processor and large memory capacity that receives input from the user, sends prompts to the appropriate generative AI model, and processes the generated answers.

[1221] Device: A device that allows a user to enter questions or instructions, such as a smartphone, tablet, or PC.

[1222] Generative AI models: Models built using machine learning libraries such as TensorFlow and PyTorch that generate answers based on prompts.

[1223] Database: Uses SQL or NoSQL databases to store historical data and knowledge. The server interacts with the database to supplement information as needed.

[1224] Data processing and calculation

[1225] The server receives input from the user, analyzes it, and sends a prompt to the appropriate generative AI model. The generative AI model generates an answer to the prompt based on pre-trained data. The generated answer is returned to the server, which complements the information as needed by linking with a database. The final answer is sent from the server to the device, which displays it to the user.

[1226] Specific examples

[1227] Consider a case where a user asks, "What are some examples of success stories from past projects?" The following is an example of a prompt sentence to input this question to a generative AI model:

[1228] Example prompt sentence:

[1229] Users: What are some success stories from past projects?

[1230] The server sends this prompt to the generative AI model, which generates the following answer based on past data:

[1231] Example of an answer from the AI ​​model:

[1232] AI Model: A successful example of a past project is the "New Product Development Project" carried out in 2022. In this project, the entire team worked together to develop a high-quality product in a short period of time, which achieved great success in the market.

[1233] In this way, the user can quickly and efficiently obtain the necessary information through the AI ​​model. The flow of the identification process in the third embodiment will be described with reference to FIG.

[1234] Step 1:

[1235] The user enters a question or instruction.

[1236] The user enters questions or instructions through the device interface, for example, "What are some success stories from past projects?" The entered data is stored in text format on the device.

[1237] Step 2:

[1238] The terminal sends the user's input to the server.

[1239] The terminal receives user input and sends it to the server. The communication protocol is HTTP POST request. For example, send data to the endpoint POST / api / query. The input is in text format, and the output is the data to be sent to the server.

[1240] Step 3:

[1241] The server receives the input and sends prompts to the appropriate generative AI model.

[1242] The server analyzes the input received from the device and sends a prompt to the appropriate generative AI model. For example, if the input is "What are some successful examples of past projects?", the server sends that question as a prompt to the generative AI model. The input is a text question, and the output is a prompt to the generative AI model.

[1243] Step 4:

[1244] A generative AI model generates an answer based on the prompt.

[1245] The generative AI model receives a prompt and generates an answer based on pre-trained data. For example, it generates an answer such as "The new product development project implemented in 2022 was successful." The input is the prompt, and the output is the generated answer text.

[1246] Step 5:

[1247] The server receives the answers from the generative AI model and complements the information by interacting with the database as needed.

[1248] The server receives the answer from the generative AI model and complements it with a database if necessary. For example, it retrieves detailed information about a specific project from the database. The input is the answer text from the generative AI model, and the output is the complemented answer text.

[1249] Step 6:

[1250] The server sends the final answer to the device.

[1251] The server generates a final answer and sends it to the device. For example, it sends an answer such as, "The new product development project conducted in 2022 was successful, and the entire team worked together to develop a high-quality product in a short period of time." The input is the completed answer text, and the output is the data sent to the device.

[1252] Step 7:

[1253] The terminal displays the answer to the user.

[1254] The device displays the answer received from the server to the user. The user confirms the answer on the device screen. For example, a smartphone application might display the text "The new product development project carried out in 2022 was successful, with the entire team working together to develop a high-quality product in a short period of time." The input is the answer data from the server, and the output is the text displayed to the user.

[1255] (Application example 3)

[1256] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[1257] Modern companies are required to effectively digitize employee knowledge and thinking and quickly retrieve necessary information. However, conventional systems have difficulty retaining past knowledge and retrieving information in real time. Furthermore, they lacked a means to obtain optimal answers from multiple perspectives, and did not adjust responses to take user emotions into account. In particular, they were unable to propose optimal countermeasures in real time for security-related questions, nor did they provide advice to reduce user stress and anxiety. A new system is needed to solve these issues.

[1258] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[1259] In this invention, the server includes means for digitizing employee knowledge logs and thoughts, means for creating digital clones, means for retrieving information using AI, means for making past knowledge accessible at any time without memorizing it, means for simultaneously using multiple AI models to obtain optimal answers, means for learning changes in a user's emotions using an emotion engine and adjusting responses, means for answering security questions in real time and proposing optimal security measures, and means for learning the user's emotions and providing advice to reduce stress and anxiety. This makes it possible to quickly obtain necessary information and obtain optimal answers from multiple perspectives, and to adjust responses taking into account the user's emotions, propose security measures, and reduce stress and anxiety.

[1260] "Means for digitizing employee knowledge logs and thinking" refers to technology for recording and storing employees' knowledge, experiences, and thought processes in digital form.

[1261] "Means for creating digital clones" refers to technology that generates digital agents that mimic the knowledge and thought patterns of a specific individual.

[1262] "Means of extracting information using AI" refers to technology that uses artificial intelligence to search for and provide the necessary information based on the user's questions and instructions.

[1263] "Means for making past knowledge accessible at any time without having to remember it" refers to technology that temporarily stores past data and makes it accessible in real time when needed.

[1264] "Means of obtaining optimal answers by simultaneously using multiple AI models" refers to a technology that uses multiple different artificial intelligence models simultaneously, compares and integrates their responses, and derives the optimal answer.

[1265] "Means for using an emotion engine to learn changes in a user's emotions and adjust responses" is a technology that analyzes a user's emotional state, learns those changes, and generates appropriate responses.

[1266] "Means for answering security-related questions in real time and proposing optimal security measures" refers to technology that provides immediate answers to security-related questions from users and proposes optimal security measures.

[1267] "Means for learning a user's emotions and providing advice to reduce stress and anxiety" is a technology that learns a user's emotional state and provides specific advice to reduce stress and anxiety.

[1268] A system for implementing this invention includes means for digitizing employee knowledge logs and thoughts and creating digital clones, thereby allowing employees' knowledge and experience to be stored in digital form and quickly accessed when needed.

[1269] The server uses artificial intelligence (AI) to retrieve information, searches for and provides the necessary information based on user questions and instructions, and uses the OpenAI API to generate answers to user questions.

[1270] Furthermore, the server has a means to make past knowledge accessible at any time without storing it, and searches and extracts past data stored in the database in real time, allowing users to always access the latest information.

[1271] The server has the means to simultaneously use multiple AI models to obtain the optimal answer, and compares and integrates answers from different perspectives to provide the optimal answer, allowing users to obtain information from multiple perspectives.

[1272] The system also includes a means for using an emotion engine to learn changes in the user's emotions and adjust responses accordingly. The emotion engine analyzes the user's emotional state, learns those changes, and generates appropriate responses. For example, if a user always shows anger in a certain situation, the emotion engine learns that pattern and, when a similar situation occurs, adjusts the digital clone's responses and the questions it asks other clones in advance.

[1273] It also includes a means to answer security questions in real time and suggest the best security measures. When a user enters a security question, the server searches the database and uses an AI model to suggest the best security measures.

[1274] It also includes a means for learning the user's emotions and providing advice to reduce stress and anxiety. The emotion engine learns the user's emotional state and provides specific advice to reduce stress and anxiety.

[1275] For example, if a user asks the question "How can I improve the security of my home?", the following prompt sentence is input to the generative AI model:

[1276] Security Questions: How can I improve the security of my home?

[1277] answer:

[1278] Based on this prompt, the AI ​​model generates an answer, integrates it with information from the database, and provides it to the user, allowing them to take quick and accurate security measures.

[1279] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1280] Step 1:

[1281] The user enters a security question.

[1282] Input: A user types the question, "How can I improve the security of my home?"

[1283] Output: The user's question is sent to the server in text form.

[1284] Step 2:

[1285] The server receives the user's query and searches the database.

[1286] Input: The user's question text.

[1287] Data Processing: The server connects to an SQLite database and retrieves security advice related to the question.

[1288] Output: The security advice retrieved from the database.

[1289] Step 3:

[1290] The server asks the AI ​​model questions.

[1291] Input: The user's question text.

[1292] Data calculation: The server uses the OpenAI API to send prompts to the generative AI model and generate answers.

[1293] Specific behavior: Send the prompt "Security Question: How can I improve the security of my home?\nAnswer:" to the generative AI model.

[1294] Output: Answer text from the generative AI model.

[1295] Step 4:

[1296] The server combines advice from the database with answers from the AI ​​model.

[1297] Input: Security advice from the database, answer text from the generative AI model.

[1298] Data processing: The server compares both answers and selects the best security measure.

[1299] Output: Text of the consolidated best security measures.

[1300] Step 5:

[1301] The server uses an emotion engine to learn the user's emotions and adjust its response.

[1302] Input: User's emotional state (e.g., anxiety).

[1303] Data Computation: The emotion engine analyzes the user's emotional state and learns its changes.

[1304] Specific behavior: If a user consistently exhibits anxiety in a particular situation, learn that pattern and adjust your response.

[1305] Output: The adjusted response text.

[1306] Step 6:

[1307] The server provides the user with the best security measures and a tailored response.

[1308] Input: Unified best security measure text, tailored response text.

[1309] Output: The optimal security measure and response provided to the user.

[1310] Step 7:

[1311] The user reviews the security measures and responses provided and asks additional questions as needed.

[1312] Input: The security measures and responses provided by the server.

[1313] Output: Additional questions for the user (if necessary).

[1314] 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.

[1315] 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> ) 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.

[1316] Another example of generative AI is Gemini (internet search engine). <url: https: gemini.google.com ?hl="ja">) are listed.

[1317] 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.

[1318] [Third embodiment]

[1319] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1320] 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.

[1321] 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).

[1322] 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.

[1323] 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.

[1324] 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).

[1325] 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.

[1326] 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.

[1327] 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.

[1328] 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.

[1329] 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.

[1330] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1331] "Example 1"

[1332] This invention uses specific algorithms and programs to digitize employees' knowledge, logs, and thoughts, accumulating the knowledge, experience, and thoughts that employees gain through their daily work. Specifically, it collects data from the tools and systems that employees use in their work, analyzes and organizes it, and digitizes it.

[1333] "Example 2"

[1334] Next, as a means of creating a digital clone, an AI model is constructed based on the digitized data to mimic the knowledge, experience, and thinking of the employee. Specifically, an AI model that reproduces the knowledge, experience, and thinking of the employee is created by learning the data.

[1335] "Example 3"

[1336] Furthermore, as a means of extracting information using AI, questions and instructions can be given to the created AI model to obtain the necessary information. Specifically, questions such as "What are some successful examples of past projects?" can be asked of the AI ​​model and the answer obtained.

[1337] "Example 4"

[1338] Additionally, as a means of making past knowledge accessible at any time without having to memorize it, the AI ​​model is linked to a database to obtain the necessary information in real time. Specifically, the AI ​​model searches and extracts past data stored in the database and provides it to the user. "Example 5"

[1339] Finally, multiple AI models can be used simultaneously to access the knowledge of diverse experts 24 hours a day, 365 days a year by asking questions to clones of other people. Specifically, questions can be asked to multiple AI models as needed, and the answers from each model can be compared and integrated to obtain the optimal answer.

[1340] The processing flow of each embodiment will be described below.

[1341] "Example 1"

[1342] Step 1: Collect data from the tools and systems employees use to do their jobs. For example, collect project progress and deliverables from work management tools, and message content from communication tools.

[1343] Step 2: Analyze, organize, and digitize the collected data. Specifically, text analysis and natural language processing techniques are used to extract useful information from the data and store it in a database.

[1344] "Example 2"

[1345] Step 1: Build an AI model based on the digitized data. Specifically, use the data as learning data and apply machine learning algorithms to train the AI ​​model.

[1346] Step 2: Use the trained AI model to make predictions and analyze new data. For example, use knowledge learned from past data to make predictions and analyze new business data.

[1347] "Example 3"

[1348] Step 1: Ask questions or give instructions to the created AI model. Specifically, the user's question is given as input to the AI ​​model, and the AI ​​model returns the output to the user.

[1349] Step 2: Evaluate the output from the AI ​​model and update the AI ​​model if necessary. For example, if the output from the AI ​​model does not meet user expectations, use that feedback to retrain the AI ​​model.

[1350] "Example 4"

[1351] Step 1: Link the AI ​​model with the database. Specifically, make it possible for the AI ​​model to obtain the data it needs from the database.

[1352] Step 2: The AI ​​model searches and extracts past data stored in the database and provides it to the user. For example, it searches the database for an answer to a user's question and returns the result to the user.

[1353] "Example 5"

[1354] Step 1: Ask multiple AI models as needed. Specifically, provide the user's question as input to multiple AI models, and obtain output from each AI model.

[1355] Step 2: Compare and integrate the outputs from each AI model to obtain the optimal answer. For example, compare the outputs from multiple AI models and return the most reliable output to the user.

[1356] Example 1

[1357] Next, a description will be given of Example 1 of Form 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."

[1358] With conventional systems, it was difficult to efficiently digitize and store the knowledge, experience, and thoughts that employees gained in their daily work. There was also a lack of means to properly analyze and organize the stored information and quickly access it when needed. Furthermore, there were also privacy issues and a lack of adequate means to access the knowledge of a variety of experts 24 hours a day, 365 days a year.

[1359] 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.

[1360] In this invention, the server includes means for digitizing employee knowledge, records, and thoughts, means for creating digital clones, means for extracting information using artificial intelligence, means for making past knowledge accessible at any time without memorizing it, means for collecting data from tools and systems used in business, means for analyzing the collected data, means for organizing the analyzed data, means for storing the organized data in a database, and means for providing the stored data. This allows for efficient digitization of employee knowledge, experience, and thoughts, appropriate analysis and organization of accumulated information, and rapid access when needed. Furthermore, while taking privacy issues into consideration, the knowledge of diverse experts can be accessed 24 hours a day, 365 days a year.

[1361] "Means for digitizing employee knowledge, records, and thoughts" refers to a method for storing the knowledge, experience, and thoughts that employees gain in their daily work in electronic form.

[1362] "Means for creating a digital clone" refers to a method for creating a virtual persona based on an employee's knowledge and experience.

[1363] "Means of extracting information using artificial intelligence" refers to a method of extracting necessary information from accumulated data using AI technology.

[1364] "Means for making past knowledge accessible at any time without having to remember it" refers to a method for making information stored in a database searchable and retrievalable in real time.

[1365] "Means of collecting data from tools and systems used in business" refers to methods for obtaining data from software and hardware that employees use on a daily basis.

[1366] "Means for analyzing collected data" refers to methods for analyzing acquired data and extracting meaningful information.

[1367] "Means for organizing analyzed data" refers to methods for classifying the analysis results into categories, tags, etc., and converting them into a format that is easy to search later.

[1368] "Means for storing organized data in a database" refers to a method for storing classified data in a database.

[1369] A "means for providing stored data" is a method by which information stored in a database can be accessed by a user.

[1370] MODE FOR CARRYING OUT THE INVENTION

[1371] This invention is a system that digitizes the knowledge, records, and thoughts of employees and efficiently stores, analyzes, and provides them. A specific embodiment of this system is described below.

[1372] Data collection

[1373] The server collects data from the tools and systems that employees use in their work. Specifically, it obtains data from communication tools (e.g., Slack, Microsoft Teams), office tools (e.g., office suites, Google Workspace), project management tools, etc. The server periodically collects data using the APIs of these tools.

[1374] Data analysis

[1375] The server analyzes the collected data. Specifically, it uses Python's NLTK library and spaCy to analyze the text data. The server tokenizes the message content and extracts important keywords and phrases. For example, the server extracts the keywords "project," "progress," and "report" from the message "Report on the progress of the project."

[1376] Data organization

[1377] The server organizes the analyzed data. Specifically, it categorizes the extracted keywords and phrases. The server organizes the data by category and converts it into a format that makes it easier to search later. For example, it creates categories such as "project management," "communication," and "reporting," and organizes the keywords that correspond to each category.

[1378] Data storage

[1379] The server saves the organized data in a database. Specifically, it uses a database management system (DBMS) such as MySQL or PostgreSQL to store the data. The server connects to the database and inserts the organized data into the appropriate table. For example, it inserts the data into the "knowledge_logs" table.

[1380] Data provision

[1381] The server provides the digitized data in a format that employees can access. Specifically, the data is made accessible through a web application using Django and Flask. Employees can search and view the data by opening a browser and accessing the Django application. For example, if an employee selects the category "Project Management," relevant messages are displayed.

[1382] Examples of specific examples and prompts

[1383] As a concrete example, consider a case where employees use Slack to report project progress. The server collects message data through the Slack API and analyzes the message content using the Python NLTK library. The analyzed results are stored in a MySQL database and made accessible to employees through a web application using Django.

[1384] Example prompt sentence:

[1385] "Collect Slack message data, parse it using Python's NLTK library, and store it in a MySQL database. Then, make the data accessible through a web application using Django."

[1386] In this way, the knowledge, experience, and thoughts of employees can be efficiently digitized, the accumulated information can be properly analyzed and organized, and it can be quickly accessed when needed. Furthermore, the knowledge of a wide range of experts can be accessed 24 hours a day, 365 days a year, while taking privacy issues into consideration.

[1387] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1388] Step 1: Data collection

[1389] The server collects data from the tools and systems employees use in their work. Specifically, the server obtains message data using Slack's API. The server calls the "https: / / slack.com / api / conversations.history" endpoint at 9:00 AM every day to collect the previous day's message data. The input is the message data obtained from Slack's API, and the output is the collected message data.

[1390] Step 2: Data analysis

[1391] The server analyzes the collected message data. Specifically, it uses Python's NLTK library to parse the text data. The server tokenizes the messages and extracts important words such as nouns and verbs. For example, the server extracts the keywords "project," "progress," and "report" from the message "Report on the progress of the project." The input is the collected message data, and the output is the extracted keywords.

[1392] Step 3: Data organization

[1393] The server organizes the analyzed data. Specifically, the server classifies the extracted keywords into categories. For example, it creates categories such as "project management," "communication," and "reporting," and organizes the keywords corresponding to each category. The input is the extracted keywords, and the output is the keywords organized by category.

[1394] Step 4: Save Data

[1395] The server saves the organized data in a database. Specifically, the server uses a MySQL database to store the data. The server connects to the database and inserts data into the "knowledge_logs" table. For example, it executes the SQL query "INSERT INTO knowledge_logs (category, keyword, message) VALUES ('Project management', 'Progress', 'Report project progress')". The input is keywords organized by category, and the output is the data stored in the database.

[1396] Step 5: Provide data

[1397] The server provides the digitized data in a format that employees can access. Specifically, the server makes the data accessible through a web application using Django. Employees can search and view the data by opening a browser and accessing the Django application. For example, if an employee selects the category "Project Management," relevant messages are displayed. The input is the data stored in the database, and the output is the data provided through the web application.

[1398] (Application example 1)

[1399] Next, a description will be given of Application Example 1 of Form 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."

[1400] In traditional factories, the knowledge and experience of workers is dependent on the individual, and there is a lack of means to efficiently share and utilize this knowledge. It is also difficult to digitize and store the operations and thoughts of workers, making knowledge transfer and work efficiency an issue. Furthermore, privacy issues and the effort required to organize and analyze data are also major obstacles.

[1401] 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.

[1402] In this invention, the server includes a means for digitizing employees' knowledge, logs, and thoughts, a means for creating digital clones, a means for extracting information using AI, a means for making past knowledge accessible at any time without memorizing it, a means for recording workers' operations and adding them to a log with a timestamp, and a means for analyzing and organizing the recorded data and saving it in a database. This allows workers' knowledge and experience to be efficiently digitized and shared, improving work efficiency and enabling the succession of knowledge.

[1403] "Means for digitizing employee knowledge logs and thoughts" refers to a method for collecting, analyzing, organizing, and digitizing the knowledge, experience, and thoughts that employees gain in their daily work.

[1404] "Methods for creating digital clones" refers to the creation of a virtual entity that digitally mimics the actions and thoughts of employees based on their knowledge logs and thoughts.

[1405] "Methods of extracting information using AI" refers to a method of using artificial intelligence technology to search and extract necessary information from digital clones and accumulated data.

[1406] "Means for making past knowledge accessible at any time without having to remember it" refers to a method for making knowledge and information stored in a database instantly accessible when needed, without relying on a specific storage device.

[1407] The "means for recording the operations of the worker and adding them to the log together with a timestamp" is a method for recording the operations performed by the worker in real time and adding them to the log together with a timestamp indicating the date and time when the operation was performed.

[1408] "Means for analyzing and organizing the recorded data and saving it in a database" refers to a method for analyzing and organizing the recorded operation data and saving it in a database.

[1409] The system for implementing this invention includes a set of means for digitizing employees' knowledge logs and thoughts, creating digital clones, and extracting information using AI. A specific embodiment of this system is shown below.

[1410] System Configuration

[1411] The system consists of the following main components:

[1412] 1. Data Collection Module: Collects data from the tools and systems employees use to do their jobs.

[1413] 2. Digitization module: Analyzes, organizes, and digitizes the collected data.

[1414] 3. Digital clone generation module: Generates a digital clone based on the digitized data.

[1415] 4. AI information extraction module: Uses AI technology to extract necessary information from digital clones and databases.

[1416] 5. Logging module: Records the worker's actions and adds them to the log with a timestamp.

[1417] 6. Database management module: Analyzes and organizes the recorded data and stores it in a database.

[1418] Hardware and software used

[1419] Hardware: Factory robots, tablets and smartphones used by workers

[1420] Software: Python, JSON library, database management system (e.g., MySQL)

[1421] Data processing and calculation

[1422] 1. Data collection: The data collection module collects data in real time from worker operations and business tools.

[1423] 2. Digitization: The digitization module analyzes, organizes, and digitizes the collected data. For example, when an operator changes a machine's settings, the operation details and settings are recorded.

[1424] 3. Digital Clone Generation: The digital clone generation module generates a digital clone that mimics the employee's behavior and thoughts based on the digitized data.

[1425] 4. AI Information Extraction: The AI ​​information extraction module uses AI technology to search and extract the necessary information from the generated digital clones and databases.

[1426] 5. Logging: The logging module records the worker's actions with a timestamp and adds them to the log.

[1427] 6. Database Management: The database management module analyzes and organizes the recorded data and stores it in a database.

[1428] Specific examples

[1429] For example, if an operator changes the speed setting of a factory robot, the operation details and setting value are recorded in real time and added to the log along with a timestamp. The recorded data is analyzed and organized by the digitization module and stored in a database. Later, the AI ​​information extraction module can be used to search and extract past operation history and setting values.

[1430] Prompt Sentence Examples

[1431] We are developing a system in which a factory robot records the operations of workers. When a worker changes the robot's speed settings, please create a Python program that records the operation and the setting value. The recorded data needs to be saved in JSON format.

[1432] In this way, the system efficiently digitizes and shares workers' knowledge and experience, improving work efficiency and ensuring knowledge transfer.

[1433] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1434] Step 1:

[1435] Data collection

[1436] The server collects data in real time from the tools and systems that workers use in their work. Specifically, when a worker operates a factory robot, the details of that operation (e.g., changing the speed setting) are obtained from sensors and log files. The input is the worker's operation data, and the output is the collected raw data.

[1437] Step 2:

[1438] Digitalization

[1439] The server analyzes and organizes the collected raw data using a digitization module. Specifically, it analyzes the collected operation data, extracts the operation details and setting values, and converts them into digital format. The input is the collected raw data, and the output is analyzed and organized digital data.

[1440] Step 3:

[1441] Digital Cloning

[1442] The server generates a digital clone based on the digitized data. Specifically, it uses the digitized operational data to generate a virtual entity that mimics the worker's actions and thoughts. The input is the digitized data, and the output is the generated digital clone.

[1443] Step 4:

[1444] AI Information Extraction

[1445] The server uses an AI information extraction module to search and extract the necessary information from the digital clone and database. Specifically, it uses an AI model to search past operation history and setting values ​​and extract the necessary information. The input is the search query, and the output is the extracted information.

[1446] Step 5:

[1447] Logging

[1448] The server records the worker's operations along with a timestamp and adds them to the log. Specifically, it records the operations performed by the worker and the date and time, and adds them to a log file. The input is the worker's operation data and timestamp, and the output is an updated log file.

[1449] Step 6:

[1450] Database Management

[1451] The server analyzes and organizes the recorded data and saves it in a database. Specifically, it analyzes and organizes the data recorded by the log recording module and saves it in a database. The input is the analyzed and organized data, and the output is an updated database.

[1452] In this way, the system efficiently digitizes and shares workers' knowledge and experience, improving work efficiency and ensuring knowledge transfer.

[1453] Example 2

[1454] Next, a description will be given of Example 2 of Form 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."

[1455] In modern companies, it is important to efficiently share and utilize the knowledge and experience of employees. However, employees' knowledge and experience are accumulated individually, making it difficult to share this throughout the organization. There is also a risk that this knowledge and experience will be lost when an employee leaves the company. Furthermore, there is a need for a method to utilize employees' knowledge while taking privacy issues into consideration. To solve these issues, a system is needed that digitizes employees' knowledge and experience and provides it in a form that can be accessed at any time.

[1456] 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. In this invention, the server includes a means for digitizing the knowledge, history, and thoughts of employees, a means for creating a digital clone, a means for extracting information using a machine learning model, a means for making past knowledge accessible at any time without storing it, a means for collecting and preprocessing data, training an AI model, and evaluating and adjusting it, and a means for inputting prompt sentences into the generative AI model to utilize the digital clone function. This makes it possible to efficiently share and utilize the knowledge and experience of employees.

[1457] "Means for digitizing employee knowledge, history, and thinking" refers to a method for collecting and storing employees' work-related knowledge, past work history, and thought processes as digital data.

[1458] "Means for creating a digital clone" refers to a method for building an AI model that mimics the behavior and judgment of employees based on the knowledge, experience, and thoughts of digitized employees.

[1459] "Methods of extracting information using machine learning models" are methods that use trained AI models to provide appropriate information in response to specific questions or requests.

[1460] "Means for making past knowledge accessible at any time without having to remember it" refers to a method that allows information based on past knowledge and experience to be obtained in real time through a digital clone.

[1461] "Means for collecting, preprocessing, training an AI model, and evaluating and adjusting data" refers to a series of processes for collecting the data necessary to create a digital clone, converting that data into a format suitable for an AI model, training the model, evaluating its performance, and adjusting it as necessary.

[1462] "Means for utilizing the functions of a digital clone by inputting prompt sentences into a generative AI model" refers to a method in which a user inputs a specific question or request into an AI model as a prompt sentence, and as a result receives appropriate information or advice from the digital clone.

[1463] MODE FOR CARRYING OUT THE INVENTION

[1464] This invention relates to a system that digitizes the knowledge and experience of employees and creates digital clones. This system includes elements such as a server, a terminal, and a user, and operates as follows.

[1465] 1. Data Collection

[1466] The server collects digitized data such as employee work history, emails, chat logs, documents, etc. This includes retrieving data from databases and cloud storage. Specifically, the server retrieves work history from databases using SQL queries and downloads documents from cloud storage via APIs.

[1467] 2. Data Preprocessing

[1468] The server preprocesses the collected data. Specifically, it cleans the data (removes noise and missing values), normalizes it (transforms it to maintain data consistency), and tokenizes it (divides the text data into words and phrases). The server uses Python's pandas library to create a data frame and imputes missing values. It also tokenizes the text data using the NLTK library.

[1469] 3. Training the AI ​​model

[1470] The server uses the preprocessed data to train an AI model using a machine learning framework such as TensorFlow or PyTorch. The server defines a neural network model using TensorFlow's Keras API and trains the model using the preprocessed data as input.

[1471] 4. Evaluate and adjust the model

[1472] The server evaluates the trained AI model and makes adjustments as necessary. The evaluation involves checking the model's accuracy using actual work data from employees. The server calculates the model's evaluation indicators (e.g., precision, recall, F1 score) and adjusts the hyperparameters based on the results. The hyperparameter optimization library is used to optimize the hyperparameters.

[1473] 5. Creating a Digital Clone

[1474] The server then uses the final, adjusted AI model to generate a digital clone, which can mimic the employee's knowledge, experience, and thinking. The server then stores the trained model and makes it accessible to users via an API.

[1475] 6. Use of digital clones

[1476] Users can use the digital clone's functions by entering prompts from their terminal. Specifically, when creating a proposal for a new project, the user can ask the digital clone for advice. For example, the user can enter the following prompt:

[1477] Example prompt sentence:

[1478] "Can you give me some examples of successful proposals from past projects? Also, can you give me some examples of unsuccessful proposals?"

[1479] The server receives the input prompt and passes it to the generative AI model. The generative AI model extracts the characteristics of successful proposals and the problems of unsuccessful proposals based on past data, and generates an answer. The server returns the generated answer to the user's device. The user can check the answer on their device and use it as a reference when creating a new proposal.

[1480] In this way, it becomes possible to efficiently share and utilize the knowledge and experience of employees.

[1481] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1482] Step 1: Data collection

[1483] The server collects digitized data such as employee work history, emails, chat logs, documents, etc. Specifically, the server retrieves work history from the database using SQL queries and downloads documents from cloud storage via API. The input is the connection information for the database and cloud storage, and the output is the collected raw data.

[1484] Step 2: Data Preprocessing

[1485] The server preprocesses the collected data. Specifically, it cleans the data (removes noise and missing values), normalizes it (transforms it to maintain data consistency), and tokenizes it (divides the text data into words and phrases). The input is the collected raw data, and the output is the preprocessed data. The server creates a data frame using Python's pandas library and completes missing values. It also tokenizes the text data using the NLTK library.

[1486] Step 3: Training the AI ​​model

[1487] The server uses the preprocessed data to train an AI model. Specifically, it uses machine learning frameworks such as TensorFlow and PyTorch. The input is the preprocessed data, and the output is a trained AI model. The server defines a neural network model using TensorFlow's Keras API and trains the model using the preprocessed data as input.

[1488] Step 4: Evaluate and tune the model

[1489] The server evaluates the trained AI model and makes adjustments as necessary. Specifically, it calculates the model's evaluation indicators (e.g., precision, recall, F1 score) and adjusts the hyperparameters based on the results. The input is the trained AI model and evaluation data, and the output is the adjusted AI model. The server optimizes the hyperparameters using a hyperparameter optimization library.

[1490] Step 5: Create a digital clone

[1491] The server generates a digital clone using the final adjusted AI model. The input is the adjusted AI model, and the output is the digital clone. The server stores the trained model and makes it accessible to users through an API.

[1492] Step 6: Use a digital clone

[1493] Users utilize the digital clone's functions by inputting prompts from their device. Specifically, when creating a proposal for a new project, the user can ask the digital clone for advice. The input is the prompt, and the output is the answer from the digital clone. For example, a user might input the following prompt: "Please give me some examples of successful proposals from past projects. Also, please give me some examples of unsuccessful proposals." The server receives the input prompt and passes it to the generative AI model. The generative AI model uses past data to extract the characteristics of successful proposals and the problems of unsuccessful proposals, and generates an answer. The server then sends the generated answer back to the user's device. The user can check the answer on their device and use it as a reference when creating a new proposal.

[1494] (Application example 2)

[1495] Next, a description will be given of Application Example 2 of Form 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."

[1496] In conventional factories, it was difficult to transfer the knowledge and experience of skilled workers to new workers or robots, making it difficult to automate tasks efficiently. There was also a risk that important work would be delayed if skilled workers retired or went on vacation. This led to problems with reduced productivity and quality in factories. Furthermore, there were issues with privacy and information accessibility.

[1497] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1498] In this invention, the server includes a means for digitizing employees' knowledge logs and thoughts, a means for creating digital clones, a means for extracting information using AI, a means for making past knowledge accessible at any time without memorizing it, a means for learning factory work data, a means for being installed in factory robots, a means for imitating the knowledge and experience of skilled workers, and a means for automatically performing complex tasks. This allows the knowledge and experience of skilled workers to be transferred to robots, enabling efficient automation of work. Furthermore, it improves information accessibility while taking privacy issues into consideration.

[1499] "Means for digitizing employee knowledge logs and thinking" refers to technology for recording and storing employees' knowledge, experiences, and thought processes as digital data.

[1500] "Means for creating digital clones" refers to technology that builds AI models that mimic employees' knowledge, experience, and thinking based on digitized data.

[1501] "Methods of extracting information using AI" refers to technology that uses artificial intelligence to search for and obtain the necessary information from a digital clone.

[1502] "Means for making past knowledge accessible at any time without storing it" refers to technology that allows knowledge and information stored in a digital clone to be accessed instantly when needed without being stored in a specific storage device.

[1503] "Means for learning factory work data" refers to technology that collects data on work performed in factories and trains an AI model using that data.

[1504] "Means to be installed on factory robots" refers to technology for incorporating trained AI models into robots used in factories.

[1505] "Means of imitating the knowledge and experience of skilled workers" refers to technology that uses an AI model to reproduce the advanced skills and experience of skilled workers and implement them in robots.

[1506] "Means for automating complex tasks" refers to technology that uses AI models to enable factory robots to automatically perform complex tasks in the same way as skilled human workers.

[1507] The system for implementing this invention includes a means for digitizing employees' knowledge logs and thoughts and creating digital clones. It also includes a means for extracting information using AI and making it accessible at any time without memorizing past knowledge. This allows factory work data to be learned and installed in factory robots, enabling them to imitate the knowledge and experience of skilled workers and perform complex tasks automatically.

[1508] System Program

[1509] The system's program uses Python to build an AI model using TensorFlow and Keras, and trains it on data from skilled workers. The server loads the data, builds the model, compiles the model, trains the model, and saves the model.

[1510] Processing Description

[1511] The server first loads digitized data on skilled workers' knowledge and experience. This data includes detailed logs of work performed in the factory and sensor data. Next, TensorFlow and Keras are used to build an AI model with LSTM and Dense layers. This model is designed to effectively process time-series data and learn the work patterns of skilled workers.

[1512] After the model is built, the server compiles it using the Adam optimizer and categorical cross-entropy loss function, which improves the model's training efficiency and accuracy. The server then trains the model using the data. Once the model is trained, it is installed on the factory robot.

[1513] Factory robots use these trained models to mimic the knowledge and experience of skilled workers and perform complex tasks automatically, helping maintain factory productivity and quality even when skilled workers are unavailable.

[1514] Specific examples

[1515] As a concrete example, this system can be installed on robots used on factory assembly lines. The robots can then act as digital clones of skilled workers, automatically performing complex assembly tasks, such as installing specific parts or carrying out quality inspections, exactly as skilled workers would.

[1516] Example prompts to input to a generative AI model:

[1517] "Build an AI model that automates factory assembly work based on data from skilled workers. Specifically, create a model that learns from the work data of skilled workers using LSTM and Dense layers, and install it on factory robots."

[1518] In this way, factory robots can utilize the knowledge and experience of skilled workers to perform advanced tasks.

[1519] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1520] Step 1:

[1521] The server reads the digitized data of skilled workers' knowledge and experience. Specifically, it collects detailed logs and sensor data of work performed in the factory and stores this in a database. The input is the factory work data, and the output is the digitized data stored in the database.

[1522] Step 2:

[1523] The server builds an AI model using TensorFlow and Keras. Specifically, it designs a neural network with LSTM and Dense layers and defines the model architecture. The input is the digitized data stored in Step 1, and the output is the built AI model.

[1524] Step 3:

[1525] The server compiles the model using the Adam optimizer and categorical cross-entropy loss function, which improves the model's learning efficiency and accuracy. The input is the AI ​​model built in step 2, and the output is the compiled AI model.

[1526] Step 4:

[1527] The server uses the digitized data to train the AI ​​model. Specifically, it inputs the data into the model and sets the number of epochs and batch size to perform training. The input is the digitized data saved in step 1 and the AI ​​model compiled in step 3, and the output is the trained AI model.

[1528] Step 5:

[1529] The server saves the trained AI model. Specifically, it saves the model as a file so that it can be installed on the factory robot later. The input is the AI ​​model trained in step 4, and the output is the saved AI model file.

[1530] Step 6:

[1531] The server installs the saved AI model on the factory robot. Specifically, it incorporates the model into the robot's control system, allowing the robot to utilize the learned knowledge and experience. The input is the AI ​​model file saved in step 5, and the output is a factory robot with the installed AI model.

[1532] Step 7:

[1533] Factory robots use installed AI models to mimic the knowledge and experience of skilled workers and automatically perform complex tasks, such as assembly work and quality inspections. The input is factory work instructions and sensor data, and the output is the results of the work performed.

[1534] Example 3

[1535] Next, a third embodiment of the third embodiment will be described. 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."

[1536] In modern companies, it is important to efficiently utilize the knowledge and experience of employees, but there is a lack of means to achieve this. There is also a need for methods to make past knowledge accessible in real time and to make the knowledge of diverse experts available 24 hours a day, 365 days a year. Furthermore, a system is needed that can simultaneously use multiple generative AI models to obtain information while taking privacy issues into consideration.

[1537] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1538] In this invention, the server includes means for digitizing employee knowledge, records, and thoughts, means for creating digital clones, means for extracting information using a generative AI model, means for making past knowledge accessible at any time without storing it, means for acquiring information by simultaneously using multiple generative AI models, means for acquiring information in real time by linking with a database, and means for considering privacy issues. This makes it possible to efficiently utilize employee knowledge and experience, acquire necessary information in real time, and access the knowledge of a variety of experts 24 hours a day, 365 days a year.

[1539] "Means for digitizing employee knowledge, records, and thoughts" refers to technology for storing and managing employees' knowledge, experiences, and thought processes in digital format.

[1540] "Means for creating a digital clone" refers to technology that uses an employee's knowledge and experience to digitally generate a virtual entity that mimics the employee's thoughts and actions.

[1541] "Means of extracting information using a generative AI model" refers to technology for asking questions or giving instructions to a generative AI model to obtain the necessary information.

[1542] "A means of making past knowledge accessible at any time without having to remember it" is a technology that searches and extracts past data in real time and provides the necessary information instantly.

[1543] "Means of obtaining information by simultaneously using multiple generative AI models" refers to a technology that simultaneously asks questions to multiple generative AI models, compares and integrates their responses, and obtains the optimal answer.

[1544] "Means of obtaining information in real time in collaboration with a database" refers to a technology that searches and extracts data stored in a database in real time, and uses that data to generate answers using a generative AI model.

[1545] "Privacy-conscious measures" are technologies that ensure the protection of personal and confidential information when using digital clones and generative AI models.

[1546] MODE FOR CARRYING OUT THE INVENTION

[1547] This invention is a system that efficiently utilizes the knowledge and experience of employees, acquires necessary information in real time, and utilizes the knowledge of various experts 24 hours a day, 365 days a year. This system is implemented using the following hardware and software.

[1548] Hardware and software used

[1549] Server: Uses a server equipped with a powerful processor and large memory capacity to run generative AI models, access databases, and process user requests.

[1550] Terminal: A device used by a user, such as a computer or smartphone. The terminal receives user input and communicates with the server.

[1551] Generative AI models: AI models for natural language processing, such as advanced generative AI models like GPT-3 and BERT.

[1552] Database: A database for storing historical knowledge and data. Use a SQL database or a NoSQL database.

[1553] Specific operation of the system

[1554] 1. The user enters a question

[1555] The user uses the device's input interface to enter questions or instructions for the generative AI model, such as "What are some successful examples from past projects?"

[1556] 2. The device sends a question to the server

[1557] The terminal sends the question entered by the user to the server using an HTTP request.

[1558] 3. The server passes the question to the AI ​​model

[1559] The server passes the received question to a generative AI model, which searches its internal knowledge base and generates an appropriate answer.

[1560] 4. The server sends the answer back to the device

[1561] The server receives the generated answer and sends it back to the user's device using an HTTP response.

[1562] 5. The device will display the answer

[1563] The terminal displays the response received from the server to the user using a GUI (Graphical User Interface).

[1564] Examples of specific examples and prompts

[1565] Example: A user asks, "What are some success stories from past projects?"

[1566] Example prompt: "Tell me about a success story from a past project."

[1567] This system allows users to efficiently obtain the necessary information and support decision-making. It is also possible to obtain information in real time by linking with a database, or to obtain information by simultaneously using multiple generative AI models. This makes it possible to make the most of employees' knowledge and experience and improve the competitiveness of the company. The flow of the specific processing in Example 3 will be described with reference to FIG. 15.

[1568] Step 1:

[1569] The user enters a question.

[1570] The user uses the device's input interface to input questions or instructions to the generative AI model. For example, they might type, "What are some successful examples of past projects?" The input question is saved as text data on the device.

[1571] Step 2:

[1572] The terminal sends a question to the server.

[1573] The terminal sends the question entered by the user to the server as an HTTP request. The data sent is the user's question text. The server receives this request and analyzes the question text.

[1574] Step 3:

[1575] The server passes the question to the AI ​​model.

[1576] The server passes the received question text to the generative AI model, which searches its internal knowledge base and generates an appropriate answer. Specifically, the AI ​​model uses natural language processing algorithms to analyze the question and extract relevant information. The output is the generated answer text.

[1577] Step 4:

[1578] The server sends a response back to the terminal.

[1579] The server receives the generated answer text and returns it to the user's device as an HTTP response. The returned data is the answer text generated by the AI ​​model. The device receives this response and analyzes the answer text.

[1580] Step 5:

[1581] The device will display the answer.

[1582] The terminal displays the answer text received from the server to the user. A GUI (Graphical User Interface) is used for display. Specifically, the answer text is displayed on the screen so that the user can check it.

[1583] (Application example 3)

[1584] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[1585] Modern manufacturing requires rapid and accurate responses to problems that occur during the manufacturing process. However, with conventional systems, it is difficult to acquire past knowledge and success stories in real time, and it is also difficult to integrate solutions from multiple perspectives. Furthermore, there is a lack of systems that allow operators to operate hands-free, making it difficult to achieve efficient manufacturing support.

[1586] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes means for digitizing employee knowledge logs and thoughts, means for creating digital clones, means for extracting information using AI, means for making past knowledge accessible at any time without memorizing it, means for acquiring information in real time about problems that occur during the manufacturing process and providing optimal manufacturing procedures and troubleshooting methods, means for using multiple AI models to provide solutions from different perspectives and select the optimal method, and means for responding to voice questions and instructions and enabling hands-free operation. This enables quick and accurate responses to problems that occur during the manufacturing process and enables efficient manufacturing support.

[1587] "Means for digitizing employee knowledge logs and thoughts" refers to technology for recording and storing employees' knowledge, experiences, and ideas in digital form.

[1588] "Methods for creating digital clones" are technologies that generate digital models that mimic the knowledge and thought patterns of a particular person.

[1589] "Means of extracting information using AI" refers to technology that uses artificial intelligence to search for and obtain the necessary information.

[1590] "A means of making past knowledge accessible at any time without having to remember it" is technology that allows past data to be searched and retrieved in real time.

[1591] "A means of obtaining real-time information on problems that occur during the manufacturing process and providing optimal manufacturing procedures and troubleshooting methods" refers to technology that instantly detects problems that occur during manufacturing and provides solutions.

[1592] "A means of using multiple AI models to provide solutions from different perspectives and select the optimal method" is a technology that uses multiple artificial intelligence models simultaneously to compare and integrate solutions from different approaches.

[1593] "Means for responding to voice inquiries and instructions and enabling hands-free operation" refers to technology that uses voice recognition technology to enable users to operate the system without using their hands.

[1594] A system for implementing this invention has the following configuration: a server includes means for digitizing employee knowledge logs and thoughts, means for creating digital clones, means for retrieving information using AI, means for making past knowledge accessible at any...

Claims

1. means for creating a digital clone of an employee based on digital data representing at least one of the employee's knowledge and experience; A means for recognizing a user's emotion; A means for extracting a response to the user's input information from the digital clone based on the recognized emotion of the user using a generative AI model; means for providing said response to said user; A system including:

2. means for considering privacy concerns of said employee in extracting said digital clone; The system of claim 1 .

3. The extracting means obtains the response by inputting a prompt sentence corresponding to the user's emotion into the generative AI model. The system of claim 1 .

Images