System

Abstract

A system is provided.SOLUTION: The system includes a means for receiving a question from a user, a means for analyzing contents of the question by using a natural language processing technology, a means for retrieving related information from a knowledge base and an information source, a means for generating an optimum answer based on a retrieval result, and a means for providing the generated answer 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] Employees face challenges finding the internal information and best practices they need, and lack quick access to information that can help them learn and solve problems. This reduces work efficiency and productivity, making it difficult to make quick decisions and solve problems. [Means for solving the problem]

[0005] A system is provided that includes a means for receiving questions from users, a means for analyzing the content of the questions using natural language processing technology, a means for searching for related information from a knowledge base and information sources, a means for generating an optimal answer based on the search results, and a means for providing the generated answer to the user. Furthermore, a system is constructed that includes a means for receiving best practice information from users, a means for analyzing the received best practice information and adding it to a knowledge base, and a means for providing the best practice information added to the knowledge base in response to questions from other users. The system also includes a means for collecting and regularly updating knowledge within an organization, a means for generating learning content to improve problem-solving capabilities using a machine learning algorithm, and a means for providing personalized learning content based on the user's usage history, thereby enabling employees to quickly and easily access the information they need.

[0006] "Users" are employees or members of an organization who use the system.

[0007] A "question" refers to text data that a user inputs into the system to request information.

[0008] "Natural language processing technology" is a technology that enables computers to understand, analyze, and generate human language.

[0009] A "knowledge base" is a digital database that stores knowledge and information within an organization.

[0010] "Sources" refers to the knowledge base as well as any relevant digital information that exists externally or elsewhere.

[0011] "Related information" is information necessary to provide an appropriate answer to a user's question.

[0012] An "answer" is the system's output in response to a user's question, and is text data generated based on information obtained from a knowledge base or information source.

[0013] "Providing to the user" refers to displaying the generated answer on the user's terminal.

[0014] "Best practice information" is information about methods and procedures that are considered optimal for improving business efficiency and results.

[0015] A "machine learning algorithm" is a mathematical technique for learning patterns from large amounts of data and making predictions and optimizations.

[0016] "Learning content" refers to the learning materials and information employees use for self-study and problem-solving.

[0017] "Usage history" refers to the user's system usage record and access history. [Brief explanation of the drawings]

[0018] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9]1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0019] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0021] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0022] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0023] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0024] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0025] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0026] [First embodiment]

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

[0028] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0029] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0030] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0031] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0032] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0033] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0035] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0036] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0037] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0038] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0039] This invention provides a method for building an in-house knowledge base system that employees can easily access and obtain the information they need. This system exchanges information between a server, terminals, and users, and analyzes, generates, and shares information using natural language processing technology and machine learning algorithms.

[0040] System configuration

[0041] server

[0042] The server plays a central role in the system, receiving questions from users and searching for information from the knowledge base and related information sources. It also uses natural language processing technology to analyze the questions and generate the most appropriate answers. It also has the function of adding best practice information to the knowledge base and updating it regularly. The server provides the generated answers and learning content to the device, and analyzes the user's usage history to provide individually optimized learning content.

[0043] Terminal

[0044] The terminal is a device that the user operates directly, sends input from the user to the server, and displays information provided by the server. The user can enter questions through the terminal and view answers and learning content from the server.

[0045] User

[0046] Users are employees or members of an organization who access the system through terminals. They can enter questions, obtain information, share best practice information, and utilize the learning content provided for self-study.

[0047] Program processing

[0048] Information search function

[0049] When the server receives a user's question, it uses natural language processing technology to analyze the question, then searches for relevant information from a knowledge base and related sources to generate the most appropriate answer, which is then sent to the terminal and displayed to the user.

[0050] Best practice sharing feature

[0051] When users submit best practice information from their devices, the server analyzes the information and adds it to a knowledge base that can be used by other users when they have similar questions.

[0052] Self-learning and problem-solving features

[0053] The server collects and regularly updates knowledge within the organization. It also uses machine learning algorithms to generate learning content that improves users' problem-solving abilities. It provides individually optimized learning content based on the user's usage history.

[0054] Specific examples

[0055] Specific examples of information search

[0056] The user types "What is the latest salary policy?" into the terminal and sends it. The server analyzes the question and searches the knowledge base for information about the latest salary policy. The server generates an answer, "The latest salary policy was updated on January 1, 2023. Click here for details." and sends it to the terminal. The terminal displays this answer to the user.

[0057] Examples of best practice sharing

[0058] A user types "It is important to apologize quickly when handling customer complaints" into a terminal and sends it. The server analyzes the information and adds it to the knowledge base. When another user asks "How do I handle customer complaints?", the server provides relevant best practices. The terminal displays the answer to the user.

[0059] Specific examples of self-study

[0060] The server regularly collects internal materials on new sales strategies and updates the knowledge base. When a user requests, "I want to learn about the latest trends in sales strategies," the server generates the most appropriate learning content and sends it to the device. The device displays "Latest Sales Strategies: The Impact of Digital Marketing and Effective Techniques" to the user. The user can study the content and use it in their work.

[0061] As can be seen, the present invention provides a concrete means for enabling employees to quickly and easily access the information they need, thereby improving productivity and efficiency throughout an organization.

[0062] The processing flow will be explained below.

[0063] Program processing flow

[0064] Information search function

[0065] Step 1:

[0066] The user enters a question in the input field on the terminal and presses the send button.

[0067] Step 2:

[0068] The terminal sends the user's question to the server.

[0069] Step 3:

[0070] The server receives the question and passes it to a natural language processing engine.

[0071] Step 4:

[0072] The server analyzes the question using a natural language processing engine.

[0073] Step 5:

[0074] The server retrieves relevant information from a knowledge base and related sources.

[0075] Step 6:

[0076] The server generates the best answer based on the search results.

[0077] Step 7:

[0078] The server generates a response and sends it to the terminal.

[0079] Step 8:

[0080] The terminal receives the response from the server and displays it to the user.

[0081] Best practice sharing feature

[0082] Step 1:

[0083] The user enters best practice information into the input field on the device and presses the send button.

[0084] Step 2:

[0085] The device sends the best practice information to the server.

[0086] Step 3:

[0087] The server receives the best practice information and passes it to the natural language processing engine.

[0088] Step 4:

[0089] The server analyzes the information using a natural language processing engine.

[0090] Step 5:

[0091] The server adds the analysis results to the knowledge base.

[0092] Step 6:

[0093] A confirmation message is sent to the device to notify that the server has been successfully added to the knowledge base.

[0094] Step 7:

[0095] The terminal receives a confirmation message from the server and displays it to the user.

[0096] Self-learning and problem-solving features

[0097] Step 1:

[0098] The server collects knowledge within the organization and updates it periodically.

[0099] Step 2:

[0100] The server uses machine learning algorithms to generate learning content.

[0101] Step 3:

[0102] A user sends a request from a terminal to access self-learning content.

[0103] Step 4:

[0104] The server analyzes the user's usage history and selects individually optimized learning content based on the user's request.

[0105] Step 5:

[0106] The server transmits the selected learning content to the terminal.

[0107] Step 6:

[0108] The terminal receives the learning content from the server and displays it to the user.

[0109] Information sharing and collection functions

[0110] Step 1:

[0111] The user enters a question in the input field on the terminal and presses the send button.

[0112] Step 2:

[0113] The terminal sends a question to the server.

[0114] Step 3:

[0115] The server receives the question and passes it to a natural language processing engine.

[0116] Step 4:

[0117] The server analyzes the question using a natural language processing engine.

[0118] Step 5:

[0119] The server retrieves relevant information from a knowledge base and related sources.

[0120] Step 6:

[0121] The server generates the best answer based on the search results.

[0122] Step 7:

[0123] Add the server-generated answer to the knowledge base.

[0124] Step 8:

[0125] The server generates a response and sends it to the terminal.

[0126] Step 9:

[0127] The terminal receives the response from the server and displays it to the user.

[0128] The above process flow allows users to quickly and easily access the information they need, improving productivity and efficiency across the organization.

[0129] Example 1

[0130] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0131] In order for employees to perform their work efficiently, they need an internal knowledge base system that allows them to quickly and easily access the information they need. However, conventional systems often lack efficient information search and answer generation, resulting in problems such as users having to wait a long time to find the information they need. Furthermore, because knowledge base updates are performed manually, the information can become outdated. Furthermore, the lack of functionality to provide learning content optimized for individual users makes it difficult to maximize users' problem-solving abilities and work efficiency.

[0132] 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.

[0133] In this invention, the server includes: means for receiving questions from users; means for analyzing the content of the questions using natural language processing technology; means for searching for relevant information from a knowledge base and related information sources; means for generating an optimal answer based on the search results; means for providing the answer in a format easy to understand for the user using a generative AI model; and means for transmitting the generated answer to a terminal and displaying it to the user. This allows users to quickly search for information and obtain the optimal answer. Furthermore, by including means for receiving best practice information from users and adding it to the knowledge base and means for providing best practice information added to the knowledge base in response to questions from other users, the knowledge base information is always up-to-date and rich. Furthermore, by including means for collecting and regularly updating knowledge within an organization, means for generating learning content to improve problem-solving skills using machine learning algorithms, and means for providing personalized learning content based on the user's usage history, it is possible to promote users' self-learning and improve work efficiency.

[0134] The "means for receiving questions from users" is a function by which the server receives questions and information input by users using their terminals.

[0135] "Means for analyzing the content of a question using natural language processing technology" is a function for analyzing the text of a received question using natural language processing technology and understanding its meaning and intent.

[0136] The "means for searching for relevant information from a knowledge base and related information sources" is a function for searching for target information from an internal knowledge base and related information sources based on the content of the analyzed question.

[0137] The "means for generating the most appropriate answer based on the search results" is a function for generating the most appropriate answer for the user based on the searched information.

[0138] "Means of using a generative AI model to provide answers in a format that is easy for users to understand" refers to a function that uses a generative AI model to convert generated answers into a format that is easy for users to understand and provide them.

[0139] The "means for transmitting the generated answer to the terminal and displaying it to the user" is a function for transmitting the answer generated by the server to the terminal so that the user can view the answer on the terminal.

[0140] The "means for receiving best practice information from a user" is a function in which a user inputs best practices that they know via a terminal, and the server receives that information.

[0141] The "means for analyzing received best practice information and adding it to a knowledge base" is a function for analyzing best practice information received from a user and adding it to the organization's knowledge base.

[0142] The "means for providing best practice information added to the knowledge base in response to questions from other users" is a function for searching for and providing best practice information added to the knowledge base in response to questions from other users.

[0143] The "means for collecting and periodically updating knowledge within an organization" is a function for continuously collecting knowledge generated within an organization and periodically updating the knowledge base.

[0144] "Means for generating learning content to improve problem-solving ability using machine learning algorithms" is a function that uses machine learning algorithms to generate learning content to improve a user's problem-solving ability.

[0145] "Means for providing personalized learning content based on a user's usage history" is a function that provides individually optimized learning content based on a user's past usage history.

[0146] This invention provides a method for building an in-house knowledge base system that allows employees to easily access and quickly obtain the information they need. This system exchanges information between a server, terminals, and users, and analyzes, generates, and shares information using natural language processing technology and machine learning algorithms.

[0147] System configuration

[0148] server

[0149] The server plays a central role in the system and receives questions from users. Once a question is received, it is analyzed using natural language processing technology. For this analysis, Google® Cloud Natural Language API can be used as a general analysis tool. After analyzing the question, related information is searched for from a knowledge base and related information sources. Specifically, ElasticSearch® is used for the search to efficiently extract related information.

[0150] The server can utilize a generative AI model (such as OpenAI's GPT-3®) to generate the optimal answer based on the search results. The generated answer may not be understandable to the user as it is, so it is converted into a user-friendly format through the generative AI model. This converted answer is sent to the device and displayed to the user.

[0151] The server also receives best practice information from users, analyzes it, and adds it to the knowledge base, which allows other users to quickly provide appropriate answers when they ask similar questions.

[0152] Furthermore, the server periodically collects knowledge from within the organization and updates the knowledge base. During this process, it generates learning content that improves problem-solving skills using machine learning algorithms (e.g., TENSORFLOW (registered trademark) and PyTorch). This allows users to receive learning content that is individually optimized based on their usage history.

[0153] Terminal

[0154] The terminal is a device that the user directly operates and is responsible for sending input from the user to the server. The user can enter questions or provide best practice information through the terminal. The terminal receives the information provided by the server and displays it to the user.

[0155] User

[0156] Users are employees or members of an organization who access the system through their terminals. They can enter questions to obtain information, share best practice information, and improve their problem-solving skills by utilizing self-learning content provided by the server.

[0157] Specific examples

[0158] A specific example is given below.

[0159] Specific examples of information search

[0160] The user types "What is the latest salary policy?" into the terminal and sends it. The server analyzes the question and searches the knowledge base for information about the latest salary policy. The server generates an answer, "The latest salary policy was updated on January 1, 2023. Click here for details." and sends it to the terminal. The terminal displays this answer to the user.

[0161] Examples of best practice sharing

[0162] A user types "It is important to apologize quickly when handling customer complaints" into a terminal and sends it. The server analyzes the information and adds it to the knowledge base. When another user asks "How do I handle customer complaints?", the server provides relevant best practices. The terminal displays the answer to the user.

[0163] Specific examples of self-study

[0164] The server regularly collects internal materials on new sales strategies and updates the knowledge base. When a user requests, "I want to learn about the latest trends in sales strategies," the server generates the most appropriate learning content and sends it to the device. The device displays "Latest Sales Strategies: The Impact of Digital Marketing and Effective Techniques" to the user. The user can study the content and use it in their work.

[0165] As described above, this invention provides a concrete means for enabling employees to quickly and easily access the information they need, which is expected to improve productivity and efficiency throughout the organization.

[0166] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0167] Step 1: Receiving a question from the user

[0168] User action: The user enters a question into the input interface of the terminal and clicks the send button.

[0169] Terminal operation: The terminal receives user input and sends the input text to the server.

[0170] Input: Text data of the question (e.g., "What is the latest salary policy?").

[0171] Output: Sending the query data to the server.

[0172] Step 2: Parsing the Question

[0173] Server operation: The server analyzes the questions received from the device using natural language processing technology. Specifically, it uses the Google Cloud Natural Language API to tokenize the text, tag it by part of speech, and analyze its meaning.

[0174] Input: Received text data.

[0175] Output: Analysis results (e.g., question intent and key keywords).

[0176] Step 3: Find related information

[0177] Server operation: Based on the analysis results, the server searches for the most relevant information from its internal knowledge base and related sources, using Elasticsearch to efficiently extract the information.

[0178] Input: A search query based on the parsed results.

[0179] Output: Search results (e.g., information about the latest salary policy).

[0180] Step 4: Generate an answer

[0181] Server operation: The server uses a generative AI model (e.g., OpenAI's GPT-3) to generate the optimal answer based on the searched information. The generated answer is then converted into a format that is easy for the user to understand.

[0182] Input: Search result data.

[0183] Output: The generated answer text (e.g., "The latest salary policy was updated on January 1, 2023. Click here for more information.").

[0184] Step 5: Submit and view your responses

[0185] Server operation: The server sends the generated answer to the terminal.

[0186] Terminal operation: The terminal displays the answer received from the server to the user. The user can check the answer on the screen and click on the details link.

[0187] Input: Generated response data.

[0188] Output: Display the answer to the user.

[0189] Step 6: Share best practices

[0190] User action: The user inputs best practice information into the terminal and sends it. For example, the user inputs and sends "It is important to apologize promptly when handling customer complaints."

[0191] Terminal action: The terminal sends this information to the server.

[0192] Server operation: The server analyzes the received information using natural language processing techniques and adds it to the knowledge base.

[0193] Input: Best practice information from users.

[0194] Output: The parsed best practice information is added to the knowledge base.

[0195] Step 7: Share best practices with others

[0196] Server operation: Other users enter similar questions into their terminals and send them to the server, which searches the knowledge base for the most appropriate best practice information, generates an answer, and sends it.

[0197] Terminal operation: The terminal receives the response and displays it to the user.

[0198] Input: Question data from other users.

[0199] Output: Appropriate best practice information is displayed to the user.

[0200] Step 8: Generate and deliver self-learning content

[0201] Server operation: The server periodically collects knowledge within the organization, updates the knowledge base, and generates learning content that improves problem-solving abilities using machine learning algorithms, using TensorFlow and PyTorch.

[0202] Input: Knowledge and user usage history data within the organization.

[0203] Output: Generated learning content (e.g., "Modern Sales Strategies: The Impact and Effective Techniques of Digital Marketing").

[0204] Step 9: Provide learning content to users

[0205] Server operation: The user inputs and sends a self-learning request (e.g., "I want to learn the latest trends in sales strategies") to the device. The server generates optimal learning content and sends it to the device.

[0206] Terminal operation: The terminal receives the learning content and displays it to the user.

[0207] Input: Self-learning request.

[0208] Output: The generated learning content is displayed to the user.

[0209] (Application example 1)

[0210] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0211] There is a need to improve worker efficiency within factories and achieve instant access to real-time information. In conventional systems, information on work procedures and troubleshooting was scattered, making it difficult to access quickly. Furthermore, there was a lack of an appropriate mechanism for responding to sudden on-site troubles and sharing best practices, which led to issues with worker efficiency and safety.

[0212] 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.

[0213] In this invention, the server includes means for receiving questions from users, means for analyzing the content of the questions using natural language processing technology, means for searching for related information from a knowledge base and information sources, means for generating an optimal answer based on the search results, means for providing the generated answer to the user, and means for the user to display work procedures and safety guidelines in real time via a mobile information terminal. This enables workers to quickly obtain the information and guidelines they need on-site, improving work efficiency and ensuring safety.

[0214] "User" refers to the worker or employee who operates the system and obtains information.

[0215] A "question" refers to a natural language inquiry that a user enters into a system seeking information.

[0216] "Natural language processing technology" refers to technology that enables computers to understand, analyze, and generate human language.

[0217] A "knowledge base" refers to a database that stores knowledge and information shared within an organization.

[0218] "Information Source" refers to a data source other than the Knowledge Base that provides relevant information.

[0219] "Answer" refers to information generated by the system in response to a user's question.

[0220] "Personal digital assistant" refers to mobile devices such as smart glasses and mobile phones.

[0221] "Best practice information" refers to information on methods and procedures that are considered to be the most effective and efficient in business operations and tasks.

[0222] "Analysis" refers to the process by which the system understands the questions and information it receives and extracts meaning from them.

[0223] "Related information" refers to information that can be used to find an appropriate answer to a user's question.

[0224] "Troubleshooting information" refers to information on methods and procedures for resolving malfunctions in machinery and equipment.

[0225] The present invention provides a system that enables workers to quickly obtain necessary information at a work site, thereby improving work efficiency and safety. Specific embodiments for carrying out the present invention will be described below.

[0226] System configuration

[0227] server

[0228] The server is the core of the system and has the following functions:

[0229] Natural language processing techniques are used to analyze user questions.

[0230] Search and generate the best answers to your questions from a knowledge base and related sources.

[0231] The results are provided to the user's mobile information terminal.

[0232] Analyze best practice information and add it to your knowledge base.

[0233] Regularly collect and update organizational knowledge.

[0234] Terminal

[0235] Terminals are devices operated directly by field workers and have the following functions:

[0236] Enter a question and send it to the server.

[0237] The response from the server is received and displayed in real time.

[0238] View instructions and troubleshooting information.

[0239] Share best practice information and send it to the server.

[0240] User

[0241] Users are factory workers and employees and have the following features:

[0242] The system is accessed through smart glasses or a mobile phone.

[0243] Check work procedures and safety guidelines in real time.

[0244] Get troubleshooting and maintenance procedures quickly.

[0245] Sharing best practice information will help build a knowledge base across the organization.

[0246] Specific operations and usage examples

[0247] 1. Display of work procedures

[0248] The user inputs a question through the smart glasses, such as "Please tell me the maintenance procedure for this machine." The server analyzes the question using natural language processing technology and searches for relevant information from a knowledge base. The maintenance procedure obtained from the search results is displayed on the user's smart glasses, allowing the work to be carried out efficiently.

[0249] 2. Troubleshooting Support

[0250] When a machine malfunctions, the user inputs a question through the smart glasses, such as "How do I deal with this error message?" The server analyzes the question and provides troubleshooting procedures by searching the knowledge base. This enables a rapid response to malfunctions.

[0251] 3. Sharing best practices

[0252] When a user discovers a new, efficient maintenance technique, they send the information to the server via the smart glasses. The server analyzes the information and adds it to a knowledge base. When other users encounter similar problems, they can share the best practice information added to the knowledge base.

[0253] Hardware and software used

[0254] Hardware:

[0255] Smart glasses (e.g., Google Glass (registered trademark), Vuzix Blade)

[0256] Server (high-performance data processing server)

[0257] software:

[0258] Natural language processing engine (e.g. Google Cloud Natural Language API)

[0259] Machine learning platforms (e.g. TensorFlow, PyTorch)

[0260] Web frameworks for the server (e.g., Flask, Django)

[0261] Prompt Sentence Examples

[0262] The prompt has the following format:

[0263] Towards a natural language processing engine

[0264] input:

[0265] Question: "What are the maintenance procedures for this machine?"

[0266] output:

[0267] Answer: "1. Turn off the power, 2. Remove the screws, 3. Clean the filter."

[0268] As described above, this invention makes it possible to dramatically improve work efficiency and safety by applying a knowledge base system to on-site factory work.

[0269] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0270] Step 1:

[0271] Users access the system using smart glasses or a personal digital assistant and enter their questions.

[0272] Specifically, the user uses the voice input function of the smart glasses to say, "Please tell me the maintenance procedure for this machine."

[0273] Input: User question (voice, text)

[0274] Output: Question data (text format)

[0275] Step 2:

[0276] The terminal transmits the question data received from the user to the server.

[0277] Specifically, the operation is to convert the received voice input into text data and send the text data to the server.

[0278] Input: Question data (text format)

[0279] Output: Send a query to the server

[0280] Step 3:

[0281] The server analyzes the received question using a natural language processing engine.

[0282] Specifically, it uses the Google Cloud Natural Language API to extract the intent and keywords of the question.

[0283] Input: Question data (text format)

[0284] Output: Question analysis results (keywords, intent)

[0285] Step 4:

[0286] The server searches for relevant information from knowledge bases and information sources based on the analysis results.

[0287] Specifically, it queries the database for relevant maintenance procedures and work procedures.

[0288] Input: Question analysis results (keywords, intent)

[0289] Output: Related information (maintenance procedures, work procedures)

[0290] Step 5:

[0291] The server generates the best answer based on the search results.

[0292] Specifically, it uses a generative AI model to generate answers to users' questions in text format.

[0293] Input: Related information (maintenance procedures, work procedures)

[0294] Output: Best answer (text format)

[0295] Step 6:

[0296] The server sends the generated response to the terminal.

[0297] Specifically, the generated answer is sent to the user's smart glasses via a network.

[0298] Input: Best answer (text format)

[0299] Output: Sends the answer to the terminal

[0300] Step 7:

[0301] The terminal displays the response received from the server to the user.

[0302] A specific operation is to display maintenance procedures and work procedures on the display of the smart glasses.

[0303] Input: Response from the server (text format)

[0304] Output: Display to the user (maintenance procedures, work procedures)

[0305] Step 8:

[0306] The user proceeds with the work based on the displayed information.

[0307] Specifically, the machine maintenance is carried out according to the displayed maintenance procedure.

[0308] Input: Information to be displayed to users (maintenance procedures, work procedures)

[0309] Output: Actual work progress

[0310] 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.

[0311] This invention provides a method for building an in-house knowledge base system that employees can easily access and obtain the information they need. This system exchanges information between a server, terminals, and users, and analyzes, generates, and shares information using natural language processing technology, machine learning algorithms, and emotion engines.

[0312] System configuration

[0313] server

[0314] The server plays a central role in the system, receiving questions from users and searching for information from a knowledge base and related information sources. It also uses natural language processing technology to analyze the questions and generate optimal answers. It also has the ability to recognize the emotions contained in the user's questions using an emotion engine and adjust the analysis results. It also adds best practice information to the knowledge base and regularly updates it. The server provides the generated answers and learning content to the device, and analyzes the user's usage history to provide individually optimized learning content.

[0315] Terminal

[0316] The terminal is a device that the user operates directly, sends input from the user to the server, and displays information provided by the server. The user can enter questions through the terminal and view answers and learning content from the server.

[0317] User

[0318] Users are employees or members of an organization who access the system through terminals. They can enter questions, obtain information, share best practice information, and utilize the learning content provided for self-study.

[0319] Program processing

[0320] Information search function

[0321] When the server receives a question from a user, it analyzes the question using natural language processing technology. It then uses an emotion engine to recognize the emotion contained in the question and adjust the analysis results. It then searches for relevant information from a knowledge base and related sources to generate the optimal answer. The generated answer is sent to the device and displayed to the user.

[0322] Best practice sharing feature

[0323] When a user submits best practice information from their device, the server analyzes the information and adds it to the knowledge base. This information is then available to other users when they ask similar questions. The information may also be displayed in an appropriate tone depending on the user's emotions.

[0324] Self-learning and problem-solving features

[0325] The server collects knowledge from within the organization and updates it regularly. It also uses machine learning algorithms to generate learning content that improves users' problem-solving abilities. It provides individually optimized learning content based on the user's usage history. An emotion engine also collects information about the user's emotions and uses it to provide future content.

[0326] Specific examples

[0327] Specific examples of information search

[0328] The user types "Please tell me the latest salary policy" into the terminal and sends it. The server analyzes the question and recognizes the user's emotion using an emotion engine. For example, if the server recognizes that the user is feeling anxious, it generates a gentle response saying, "The latest salary policy was updated on January 1, 2023. Click here for details." The terminal displays this response to the user.

[0329] Examples of best practice sharing

[0330] The user types "It is important to apologize quickly when handling customer complaints" into the terminal and sends it. The server analyzes the information, recognizes the emotion of the received information using an emotion engine, and adds it to the knowledge base. When another user asks "How do I handle customer complaints?", the server provides relevant best practices in an appropriate tone. The terminal displays the answer to the user.

[0331] Specific examples of self-study

[0332] When a user requests, "I want to learn about the latest trends in sales strategies," the server generates the most appropriate learning content and sends it to the device in a tone adjusted according to the user's emotions. For example, if the user is feeling very interested, the server will provide the content, "Latest sales strategies: The impact of digital marketing and effective techniques. Click here for more information." The device displays this content, and the user can learn from it and apply it to their work.

[0333] As described above, the present invention provides a concrete means for enabling employees to quickly and easily access the information they need while taking into consideration the user's feelings, thereby improving the productivity and efficiency of the entire organization.

[0334] The processing flow will be explained below.

[0335] Program processing flow

[0336] Information search function

[0337] Step 1:

[0338] The user enters a question in the input field on the terminal and presses the send button.

[0339] Step 2:

[0340] The terminal sends the user's question to the server.

[0341] Step 3:

[0342] The server receives the question and passes it to a natural language processing engine.

[0343] Step 4:

[0344] The server analyzes the question using a natural language processing engine.

[0345] Step 5:

[0346] The server requests the emotion engine to analyze the emotion of the question.

[0347] Step 6:

[0348] The emotion engine analyzes the emotion of the question and returns the emotion information to the server.

[0349] Step 7:

[0350] The server takes the emotion information into consideration and searches for relevant information from a knowledge base and related information sources.

[0351] Step 8:

[0352] The server generates the best answer based on the search results, adjusting the tone according to the user's emotions.

[0353] Step 9:

[0354] The server generates a response and sends it to the terminal.

[0355] Step 10:

[0356] The terminal receives the response from the server and displays it to the user.

[0357] Best practice sharing feature

[0358] Step 1:

[0359] The user enters best practice information into the input field on the device and presses the send button.

[0360] Step 2:

[0361] The device sends the best practice information to the server.

[0362] Step 3:

[0363] The server receives the best practice information and passes it to the natural language processing engine.

[0364] Step 4:

[0365] The server analyzes the best practice information using a natural language processing engine.

[0366] Step 5:

[0367] The server requests the emotion engine to analyze the emotion of the received information.

[0368] Step 6:

[0369] The emotion engine analyzes the emotion of the information and returns the emotion information to the server.

[0370] Step 7:

[0371] The server considers the analysis results and emotional information and adds them to the knowledge base.

[0372] Step 8:

[0373] A confirmation message is sent to the device to notify that the server has been successfully added to the knowledge base.

[0374] Step 9:

[0375] The terminal receives a confirmation message from the server and displays it to the user.

[0376] Self-learning and problem-solving features

[0377] Step 1:

[0378] The server collects knowledge within the organization and updates it periodically.

[0379] Step 2:

[0380] The server uses machine learning algorithms to generate learning content.

[0381] Step 3:

[0382] A user sends a request from a terminal to access self-learning content.

[0383] Step 4:

[0384] The server analyzes the user's usage history and generates individually optimized learning content.

[0385] Step 5:

[0386] The server requests the emotion engine to analyze the user's emotions.

[0387] Step 6:

[0388] The emotion engine returns the analysis results to the server.

[0389] Step 7:

[0390] The server optimizes learning content with an appropriate tone based on emotional information and sends it to the device.

[0391] Step 8:

[0392] The terminal receives the learning content from the server and displays it to the user.

[0393] Information sharing and collection functions

[0394] Step 1:

[0395] The user enters a question in the input field on the terminal and presses the send button.

[0396] Step 2:

[0397] The terminal sends a question to the server.

[0398] Step 3:

[0399] The server receives the question and passes it to a natural language processing engine.

[0400] Step 4:

[0401] The server analyzes the question using a natural language processing engine.

[0402] Step 5:

[0403] The server requests the emotion engine to analyze the emotion of the question.

[0404] Step 6:

[0405] The emotion engine analyzes the emotion of the question and returns the emotion information to the server.

[0406] Step 7:

[0407] The server takes the emotion information into consideration to search for relevant information from a knowledge base and related information sources.

[0408] Step 8:

[0409] The server generates the best answer based on the search results and adds it to the knowledge base.

[0410] Step 9:

[0411] The server generates a response in a tone based on the emotion information and transmits it to the terminal.

[0412] Step 10:

[0413] The terminal receives the response from the server and displays it to the user.

[0414] This allows the system, combined with the emotion engine, to recognize the user's emotions and adjust the tone of its analysis results and responses to provide more appropriate information. The specific processing flow allows users to access the information they need quickly and easily, improving productivity and efficiency across the organization.

[0415] Example 2

[0416] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0417] Conventional in-house knowledge base systems made it difficult for users to access the information they needed accurately and quickly. They also lacked appropriate feedback and learning content that took users' emotions into consideration. This led to problems such as reduced user satisfaction and work efficiency.

[0418] 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.

[0419] In this invention, the server includes means for receiving a question from a user, means for analyzing the content of the question using natural language processing technology, means for recognizing the emotion contained in the question and adjusting the analysis result, means for searching for related information from a knowledge base and information sources, means for generating an optimal answer according to the emotion based on the search results, and means for providing the generated answer to the user. This enables the user to quickly and accurately access the information they need and to receive appropriate feedback according to their emotion.

[0420] A "User" is a person or member of an organization who accesses the system to obtain or provide information.

[0421] A "server" is a central device that receives requests from users and analyzes, searches, generates, and provides information.

[0422] A "terminal" is a device that is directly operated by a user and that communicates with a server to send and receive information.

[0423] The "means for receiving questions" is a mechanism by which the server receives questions from users.

[0424] "Natural language processing technology" is a technology for analyzing questions entered by users, and involves grammatical and semantic analysis.

[0425] "Means for recognizing emotions" refers to technology that analyzes the emotions contained in a user's question and provides appropriate feedback.

[0426] A "knowledge base" is a database that stores past questions, best practices, and related information.

[0427] "Source" means a source of information, including external data or materials other than the Knowledge Base.

[0428] A "search method" is a mechanism for finding relevant information from knowledge bases and information sources.

[0429] "Answer generation means" refers to a technique for creating the most appropriate answer to a user's question based on search results.

[0430] "Best practices" are the optimal methods and techniques for improving business efficiency and results.

[0431] "Learning content" refers to content provided to help users improve their skills and solve problems.

[0432] A "machine learning algorithm" is a technology that analyzes large amounts of data, finds patterns, and makes predictions and classifications.

[0433] This invention provides a concrete means for realizing an in-house knowledge base system. The main components of the system include a server, terminals, and users, each of which plays a specific role. The details are explained below.

[0434] server

[0435] The server plays the central role in the system, receiving questions from users, analyzing them, and generating answers. The server uses the following main software technologies:

[0436] Natural language processing technology: We use "spaCy" and "BERT" to analyze user questions.

[0437] Emotion recognition technology: We use "Affectiva" to analyze the emotions contained in users' questions.

[0438] Knowledge base: Databases such as "MySQL (registered trademark)" and "PostgreSQL" are used, and best practices and related information are stored.

[0439] Machine learning algorithms: "scikit-learn" and "TensorFlow" are used to generate learning content.

[0440] The server utilizes these technologies to generate appropriate answers to user questions, periodically updates the collected knowledge, and provides individually optimized learning content based on the user's usage history and emotional information.

[0441] For example, if a user asks, "What is the latest salary policy?", the server analyzes the question and uses emotion recognition technology to determine whether the user is feeling anxious. Based on the results, the server generates the most appropriate response and tone and sends the following response to the device: "The latest salary policy was updated on January 1, 2023. Click here for details."

[0442] Terminal

[0443] The terminal is a device operated by the user that sends and receives information to and from the server. Specifically, it has the following functions:

[0444] Submit user input: Enter a question or best practice information and submit it to the server.

[0445] Display information: Displays answers and learning content received from the server to the user.

[0446] For example, a user inputs "It is important to apologize promptly when handling customer complaints" into an input field on the terminal and sends it. The terminal transfers this information to the server, receives appropriate feedback from the server again, and displays it.

[0447] User

[0448] The user is the end user of the system and performs the following actions:

[0449] Entering a question: Requesting the required information from the system through the terminal.

[0450] Best practice information sharing: Providing useful information from the terminal to the system.

[0451] Utilizing learning content: Use the provided learning content for self-study and problem-solving.

[0452] For example, if a user types "I want to learn about the latest trends in sales strategies" into their device, the server will generate relevant learning content, adjust the tone according to their emotions, and provide information such as "The latest sales strategies: The impact of digital marketing and effective techniques. Click here for more information." Users can view this information and apply it to their work.

[0453] By combining these functions, it is possible to provide necessary information quickly and accurately while taking into consideration the user's feelings. Also, by providing individually optimized learning content, productivity and efficiency will be improved across the organization.

[0454] Prompt Sentence Examples

[0455] Use spaCy to parse the user question "What is the latest salary policy?" and process the sentiment with Affectiva to generate an answer with the appropriate tone.

[0456] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0457] Information retrieval function processing steps

[0458] Step 1: User enters question

[0459] The user types a question into a field on the device, such as "What's the new vacation policy?"

[0460] Input: User question text

[0461] Output: The question you typed is displayed on the terminal.

[0462] Step 2: The device sends a question to the server

[0463] The device sends the entered question to the server as an API request.

[0464] Input: Question text entered by the user

[0465] Output: API request sent to the server

[0466] Step 3: The server parses the question

[0467] The server analyzes the received question using natural language processing technology (e.g., "spaCy") to identify the main topics, and through this analysis, understands the context and intent of the question.

[0468] Input: Question text sent to the server

[0469] Output: Analysis of the question (e.g. "vacation policy")

[0470] Step 4: The server recognizes the emotion

[0471] The server uses emotion recognition technology (e.g., "Affectiva") to analyze the emotions contained in the questions, and through this recognition, it understands the user's emotional state and adjusts the tone of its answers accordingly.

[0472] Input: Analysis result of question content

[0473] Output: Sentiment analysis result (e.g., "anxiety")

[0474] Step 5: The server searches the knowledge base for information

[0475] The server searches for relevant information from a knowledge base and other sources based on the analysis results and sentiment analysis results.

[0476] Input: Question content analysis results and sentiment analysis results

[0477] Output: Search results (e.g. "new vacation policy details")

[0478] Step 6: Server generates answer

[0479] The server generates a response in an appropriate tone based on the sentiment analysis results, such as "Our latest vacation policy was updated on January 1, 2023. Click here for more information."

[0480] Input: Search results and sentiment analysis results

[0481] Output: Generated answer text

[0482] Step 7: The server sends the answer to the device

[0483] The server sends the generated answer to the terminal as an API response.

[0484] Input: Generated answer text

[0485] Output: API response sent to the device

[0486] Step 8: The device displays the answer to the user

[0487] The terminal displays the received answer to the user, who can then view it to find the answer to his or her question.

[0488] Input: Answer text sent from the server

[0489] Output: Answer displayed on the terminal

[0490] Best Practice Sharing Feature Processing Steps

[0491] Step 1: User inputs best practices

[0492] The user inputs best practice information into an input field on the terminal. For example, the user might input "It is important to apologize promptly when handling customer complaints."

[0493] Input: User best practice information

[0494] Output: The entered best practices are displayed on the terminal.

[0495] Step 2: The device sends the information to the server

[0496] The device sends the entered best practice information to the server as an API request.

[0497] Input: Best practice information entered by the user

[0498] Output: API request sent to the server

[0499] Step 3: The server analyzes the information

[0500] The server analyzes the received best practice information using natural language processing technology (e.g., "BERT") to identify key points.

[0501] Input: Best practice information sent to the server

[0502] Output: Best practice analysis results

[0503] Step 4: The server recognizes the emotion

[0504] The server uses emotion recognition technology to analyze the emotions contained in the best practice information, allowing it to present the information to other users in an appropriate tone.

[0505] Input: Best practice analysis results

[0506] Output: Emotion analysis results

[0507] Step 5: The server adds the information to the knowledge base

[0508] Based on the results of the analysis and sentiment analysis, the server adds new best practice information to the knowledge base, which can be used by other users when they ask similar questions.

[0509] Input: Best practice content analysis results and sentiment analysis results

[0510] Output: Information added to the Knowledge Base

[0511] Step 6: The server provides best practices for other users' questions

[0512] When other users ask similar questions, the server retrieves additional best practice information from the knowledge base and delivers it in the appropriate tone.

[0513] Input: Question from another user

[0514] Output: Best practice information provided

[0515] Self-learning and problem-solving processing steps

[0516] Step 1: User requests learning content

[0517] The user inputs "I want to learn the latest sales strategies" into the terminal and sends a request.

[0518] Input: User's learning content request

[0519] Output: The input request is displayed on the terminal.

[0520] Step 2: The device sends a request to the server

[0521] The device sends this request to the server as an API request.

[0522] Input: Learning content request entered by the user

[0523] Output: API request sent to the server

[0524] Step 3: Server checks the knowledge base

[0525] The server retrieves relevant learning material from a knowledge base.

[0526] Input: Learning content request sent to the server

[0527] Output: Search results (related learning materials)

[0528] Step 4: Server Generates Content

[0529] The server uses machine learning algorithms to generate learning content that best suits the request (for example, "scikit-learn" or "TensorFlow").

[0530] Input: Search results

[0531] Output: Generated learning content

[0532] Step 5: The server adjusts the tone based on the emotion

[0533] The server uses emotion recognition technology to tailor the learning content to a tone that corresponds to the user's emotions.

[0534] Input: Generated learning content and sentiment analysis results

[0535] Output: Tailored learning content

[0536] Step 6: The server sends the learning content to the device

[0537] The server sends the adjusted learning content to the device as an API response.

[0538] Input: Tailored learning content

[0539] Output: API response sent to the device

[0540] Step 7: The device displays the content to the user

[0541] The device displays the received learning content to the user, who can then view it and apply it to their work.

[0542] Input: Learning content sent from the server

[0543] Output: Learning content displayed on the device

[0544] (Application example 2)

[0545] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0546] While conventional in-house knowledge base systems allow users to quickly and easily access the information they need, they lack the ability to provide appropriate answers based on the user's emotions, making it difficult to respond appropriately, especially under stressful situations.In addition, for robots used on factory floors, there was a need for a means to provide optimal information in real time during troubleshooting and maintenance work.

[0547] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving a question from a user, means for analyzing the content of the question using natural language processing technology, means for searching for related information from a knowledge base and an information source, means for generating an optimal answer based on the search results, means for providing the generated answer to the user, emotion analysis means for recognizing the user's emotion and adjusting the analysis result, and means for displaying the generated answer via the robot's output device. This enables the factory robot to provide information in an appropriate tone according to the user's emotion and to support troubleshooting and maintenance work in real time.

[0548] A "user" is an individual or member of an organization who utilizes the system to enter questions and receive answers.

[0549] A "server" is a computer system that receives and analyzes questions from users, searches for relevant information from knowledge bases and information sources, and generates and provides optimal answers.

[0550] A "terminal" is a device that a user directly operates and connects to a server to input questions and receive answers.

[0551] A "knowledge base" is a database that stores knowledge and data collected from various sources.

[0552] "Natural language processing technology" is a technology that allows computers to understand and process human language.

[0553] A "machine learning algorithm" is an algorithm that finds patterns based on large amounts of data and automatically learns and predicts.

[0554] The "emotion engine" is a technology that recognizes emotions from user input and adjusts the tone of the response based on that information.

[0555] A "source" is a location where external or internal data or information exists outside of the knowledge base.

[0556] "Search Results" refers to relevant information retrieved from knowledge bases and information sources based on a user's question.

[0557] "Best practice information" is information that summarizes methods or procedures that are known to be most effective in a particular situation.

[0558] "Learning content" is educational content generated by the server to improve the user's problem-solving ability.

[0559] "Troubleshooting" is the process of identifying and resolving problems or errors in machines or systems.

[0560] "Maintenance work" refers to maintenance work carried out periodically or as needed to ensure stable operation of machines and systems.

[0561] "Sentiment analysis" is the process of detecting emotions from user input data and adapting that information.

[0562] An "output device" is a device for displaying generated information and answers to the user.

[0563] This invention provides an in-house knowledge base system that allows employees to easily access and obtain the information they need. The system receives questions from users, analyzes the questions using natural language processing technology, and searches for information from the knowledge base and related information sources to generate and provide optimal answers. It also includes a function that recognizes user emotions using an emotion engine and adjusts the analysis results. Furthermore, this system is implemented in factory robots to support troubleshooting and maintenance work.

[0564] System configuration and functions

[0565] server

[0566] The server receives questions from users and analyzes the content of the questions using natural language processing technology. Based on the analyzed question, it searches for relevant information from a knowledge base and information sources and generates the optimal answer. In doing so, it analyzes the user's emotions using an emotion engine and adjusts the analysis results. The server also has the function of receiving best practice information from users and adding it to the knowledge base. Furthermore, it uses machine learning algorithms to generate and provide learning content to improve the user's problem-solving ability.

[0567] Terminal

[0568] The terminal is a device that is directly operated by the user, allowing them to input questions, send them to the server, and display the information provided by the server. The terminal also functions as an output device for factory robots, displaying real-time information necessary for troubleshooting and maintenance work.

[0569] User

[0570] Users can use the system to enter questions and obtain the necessary information. For example, when a problem occurs at a factory, users can enter a question through their terminal and obtain the most appropriate answer from the server. Users can also provide their own best practice information to the system and share it with other users.

[0571] Hardware and software used

[0572] This system mainly uses the following hardware and software:

[0573] Hardware: Server equipment, user devices (PCs, tablets, smartphones, etc.), factory robots, output devices (displays)

[0574] Software: Natural language processing techniques (e.g., Hugging Face transformers), machine learning algorithms, sentiment analysis engines

[0575] Data processing and calculation

[0576] The server analyzes the user's question using natural language processing technology, then analyzes the user's emotions using an emotion engine. Based on the analysis results, it searches for relevant information from a knowledge base and information sources to generate the optimal answer. The tone of this answer is adjusted according to the user's emotions and sent to the device.

[0577] Specific examples

[0578] Specific examples of information search

[0579] The user types "What should you do if this machine stops?" into the terminal and sends it. The server analyzes the question and recognizes the user's emotions using an emotion engine. For example, if the server recognizes that the user is feeling anxious, it generates a response in a gentle tone: "If this machine stops, first press the emergency stop button to ensure safety. Then, follow the diagnostic steps below: 1. Check the power supply. 2. Check the connection cable." The terminal displays this response to the user.

[0580] Prompt Sentence Examples

[0581] Here are some examples of specific prompts:

[0582] "What should we do if this machine stops working?"

[0583] "What are your latest sales strategies?"

[0584] How do you handle customer complaints?

[0585] This system allows users to quickly obtain the information they need in a way that takes their emotions into consideration, making it possible to efficiently perform troubleshooting and maintenance work, particularly on factory floors.

[0586] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0587] Step 1:

[0588] The user operates the terminal to input a question and sends it to the server. The input data is a question about the information the user wants to know. For example, a prompt such as "What should we do if this machine stops working?" is input.

[0589] Step 2:

[0590] The server receives questions sent by users. It analyzes the received questions using natural language processing technology, tokenizing and tagging them to understand their meaning. The input data is the user's question, and the output is the structure of the analyzed question.

[0591] Step 3:

[0592] The server passes the analyzed question to the emotion engine to recognize the user's emotion. The emotion engine analyzes the tone and expression of the question to recognize emotions such as "anxiety," "interest," and "calm." The input data is the analyzed question, and the output is the recognized emotion information.

[0593] Step 4:

[0594] The server searches for relevant information from a knowledge base and related information sources based on the analysis results and emotion information. The knowledge base stores pre-accumulated knowledge, and generates search queries to extract relevant information. The input data are the analysis results and emotion information, and the output is the search results including related information.

[0595] Step 5:

[0596] The server generates the optimal answer based on the search results. In particular, it adjusts the tone of the answer depending on the user's emotions. It uses a generative AI model to construct an answer in natural language that combines the question and search results. The input data is the search results and emotional information, and the output is the optimal answer.

[0597] Step 6:

[0598] The server sends the generated answer to the terminal. The terminal displays the received answer to the user. This allows the user to quickly obtain the information they need in a way that takes their emotions into consideration. The input data is the generated answer, and the output is the answer displayed on the terminal screen.

[0599] Step 7:

[0600] The best practice information provided by the user is sent from the device to the server. The server analyzes the received information, recognizes the emotion information, and adds it to the knowledge base. The input data is the best practice information, and the output is an updated knowledge base.

[0601] Step 8:

[0602] The server uses machine learning algorithms to generate learning content to improve the user's problem-solving ability. The generated learning content is personalized based on the user's usage history and emotional information. The input data is usage history and emotional information, and the output is the generated learning content.

[0603] Specific examples

[0604] For example, if a user enters a prompt such as "What should you do if this machine stops working?", the server analyzes the question and recognizes the user's anxiety using an emotion engine. It then searches the knowledge base for relevant information and generates an answer such as "If this machine stops working, first press the emergency stop button to ensure safety. Then, follow the following diagnostic steps: 1. Check the power supply. 2. Check the connection cables." This answer is then sent to the terminal and displayed. In this way, the user can learn the appropriate measures.

[0605] 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.

[0606] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0607] 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.

[0608] [Second embodiment]

[0609] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0610] 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.

[0611] 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).

[0612] 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.

[0613] 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.

[0614] 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).

[0615] 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.

[0616] 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.

[0617] 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.

[0618] 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.

[0619] 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.

[0620] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0621] This invention provides a method for building an in-house knowledge base system that employees can easily access and obtain the information they need. This system exchanges information between a server, terminals, and users, and analyzes, generates, and shares information using natural language processing technology and machine learning algorithms.

[0622] System configuration

[0623] server

[0624] The server plays a central role in the system, receiving questions from users and searching for information from the knowledge base and related information sources. It also uses natural language processing technology to analyze the questions and generate the most appropriate answers. It also has the function of adding best practice information to the knowledge base and updating it regularly. The server provides the generated answers and learning content to the device, and analyzes the user's usage history to provide individually optimized learning content.

[0625] Terminal

[0626] The terminal is a device that the user operates directly, sends input from the user to the server, and displays information provided by the server. The user can enter questions through the terminal and view answers and learning content from the server.

[0627] User

[0628] Users are employees or members of an organization who access the system through terminals. They can enter questions, obtain information, share best practice information, and utilize the learning content provided for self-study.

[0629] Program processing

[0630] Information search function

[0631] When the server receives a user's question, it uses natural language processing technology to analyze the question, then searches for relevant information from a knowledge base and related sources to generate the most appropriate answer, which is then sent to the terminal and displayed to the user.

[0632] Best practice sharing feature

[0633] When users submit best practice information from their devices, the server analyzes the information and adds it to a knowledge base that can be used by other users when they have similar questions.

[0634] Self-learning and problem-solving features

[0635] The server collects and regularly updates knowledge within the organization. It also uses machine learning algorithms to generate learning content that improves users' problem-solving abilities. It provides individually optimized learning content based on the user's usage history.

[0636] Specific examples

[0637] Specific examples of information search

[0638] The user types "What is the latest salary policy?" into the terminal and sends it. The server analyzes the question and searches the knowledge base for information about the latest salary policy. The server generates an answer, "The latest salary policy was updated on January 1, 2023. Click here for details." and sends it to the terminal. The terminal displays this answer to the user.

[0639] Examples of best practice sharing

[0640] A user types "It is important to apologize quickly when handling customer complaints" into a terminal and sends it. The server analyzes the information and adds it to the knowledge base. When another user asks "How do I handle customer complaints?", the server provides relevant best practices. The terminal displays the answer to the user.

[0641] Specific examples of self-study

[0642] The server regularly collects internal materials on new sales strategies and updates the knowledge base. When a user requests, "I want to learn about the latest trends in sales strategies," the server generates the most appropriate learning content and sends it to the device. The device displays "Latest Sales Strategies: The Impact of Digital Marketing and Effective Techniques" to the user. The user can study the content and use it in their work.

[0643] As can be seen, the present invention provides a concrete means for enabling employees to quickly and easily access the information they need, thereby improving productivity and efficiency throughout an organization.

[0644] The processing flow will be explained below.

[0645] Program processing flow

[0646] Information search function

[0647] Step 1:

[0648] The user enters a question in the input field on the terminal and presses the send button.

[0649] Step 2:

[0650] The terminal sends the user's question to the server.

[0651] Step 3:

[0652] The server receives the question and passes it to a natural language processing engine.

[0653] Step 4:

[0654] The server analyzes the question using a natural language processing engine.

[0655] Step 5:

[0656] The server retrieves relevant information from a knowledge base and related sources.

[0657] Step 6:

[0658] The server generates the best answer based on the search results.

[0659] Step 7:

[0660] The server generates a response and sends it to the terminal.

[0661] Step 8:

[0662] The terminal receives the response from the server and displays it to the user.

[0663] Best practice sharing feature

[0664] Step 1:

[0665] The user enters best practice information into the input field on the device and presses the send button.

[0666] Step 2:

[0667] The device sends the best practice information to the server.

[0668] Step 3:

[0669] The server receives the best practice information and passes it to the natural language processing engine.

[0670] Step 4:

[0671] The server analyzes the information using a natural language processing engine.

[0672] Step 5:

[0673] The server adds the analysis results to the knowledge base.

[0674] Step 6:

[0675] A confirmation message is sent to the device to notify that the server has been successfully added to the knowledge base.

[0676] Step 7:

[0677] The terminal receives a confirmation message from the server and displays it to the user.

[0678] Self-learning and problem-solving features

[0679] Step 1:

[0680] The server collects knowledge within the organization and updates it periodically.

[0681] Step 2:

[0682] The server uses machine learning algorithms to generate learning content.

[0683] Step 3:

[0684] A user sends a request from a terminal to access self-learning content.

[0685] Step 4:

[0686] The server analyzes the user's usage history and selects individually optimized learning content based on the user's request.

[0687] Step 5:

[0688] The server transmits the selected learning content to the terminal.

[0689] Step 6:

[0690] The terminal receives the learning content from the server and displays it to the user.

[0691] Information sharing and collection functions

[0692] Step 1:

[0693] The user enters a question in the input field on the terminal and presses the send button.

[0694] Step 2:

[0695] The terminal sends a question to the server.

[0696] Step 3:

[0697] The server receives the question and passes it to a natural language processing engine.

[0698] Step 4:

[0699] The server analyzes the question using a natural language processing engine.

[0700] Step 5:

[0701] The server retrieves relevant information from a knowledge base and related sources.

[0702] Step 6:

[0703] The server generates the best answer based on the search results.

[0704] Step 7:

[0705] Add the server-generated answer to the knowledge base.

[0706] Step 8:

[0707] The server generates a response and sends it to the terminal.

[0708] Step 9:

[0709] The terminal receives the response from the server and displays it to the user.

[0710] The above process flow allows users to quickly and easily access the information they need, improving productivity and efficiency across the organization.

[0711] Example 1

[0712] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0713] In order for employees to perform their work efficiently, they need an internal knowledge base system that allows them to quickly and easily access the information they need. However, conventional systems often lack efficient information search and answer generation, resulting in problems such as users having to wait a long time to find the information they need. Furthermore, because knowledge base updates are performed manually, the information can become outdated. Furthermore, the lack of functionality to provide learning content optimized for individual users makes it difficult to maximize users' problem-solving abilities and work efficiency.

[0714] 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.

[0715] In this invention, the server includes: means for receiving questions from users; means for analyzing the content of the questions using natural language processing technology; means for searching for relevant information from a knowledge base and related information sources; means for generating an optimal answer based on the search results; means for providing the answer in a format easy to understand for the user using a generative AI model; and means for transmitting the generated answer to a terminal and displaying it to the user. This allows users to quickly search for information and obtain the optimal answer. Furthermore, by including means for receiving best practice information from users and adding it to the knowledge base and means for providing best practice information added to the knowledge base in response to questions from other users, the knowledge base information is always up-to-date and rich. Furthermore, by including means for collecting and regularly updating knowledge within an organization, means for generating learning content to improve problem-solving skills using machine learning algorithms, and means for providing personalized learning content based on the user's usage history, it is possible to promote users' self-learning and improve work efficiency.

[0716] The "means for receiving questions from users" is a function by which the server receives questions and information input by users using their terminals.

[0717] "Means for analyzing the content of a question using natural language processing technology" is a function for analyzing the text of a received question using natural language processing technology and understanding its meaning and intent.

[0718] The "means for searching for relevant information from a knowledge base and related information sources" is a function for searching for target information from an internal knowledge base and related information sources based on the content of the analyzed question.

[0719] The "means for generating the most appropriate answer based on the search results" is a function for generating the most appropriate answer for the user based on the searched information.

[0720] "Means of using a generative AI model to provide answers in a format that is easy for users to understand" refers to a function that uses a generative AI model to convert generated answers into a format that is easy for users to understand and provide them.

[0721] The "means for transmitting the generated answer to the terminal and displaying it to the user" is a function for transmitting the answer generated by the server to the terminal so that the user can view the answer on the terminal.

[0722] The "means for receiving best practice information from a user" is a function in which a user inputs best practices that they know via a terminal, and the server receives that information.

[0723] The "means for analyzing received best practice information and adding it to a knowledge base" is a function for analyzing best practice information received from a user and adding it to the organization's knowledge base.

[0724] The "means for providing best practice information added to the knowledge base in response to questions from other users" is a function for searching for and providing best practice information added to the knowledge base in response to questions from other users.

[0725] The "means for collecting and periodically updating knowledge within an organization" is a function for continuously collecting knowledge generated within an organization and periodically updating the knowledge base.

[0726] "Means for generating learning content to improve problem-solving ability using machine learning algorithms" is a function that uses machine learning algorithms to generate learning content to improve a user's problem-solving ability.

[0727] "Means for providing personalized learning content based on a user's usage history" is a function that provides individually optimized learning content based on a user's past usage history.

[0728] This invention provides a method for building an in-house knowledge base system that allows employees to easily access and quickly obtain the information they need. This system exchanges information between a server, terminals, and users, and analyzes, generates, and shares information using natural language processing technology and machine learning algorithms.

[0729] System configuration

[0730] server

[0731] The server plays a central role in the system and receives questions from users. Once a question is received, it is analyzed using natural language processing technology. For this analysis, Google Cloud Natural Language API can be used as a general analysis tool. After analyzing the question, related information is searched for from the knowledge base and related information sources. Specifically, Elasticsearch is used for the search to efficiently extract related information.

[0732] The server can utilize a generative AI model (such as OpenAI's GPT-3) to generate the optimal answer based on the search results. The generated answer may not be understandable to the user as it is, so it is converted into a user-friendly format through the generative AI model. This converted answer is sent to the device and displayed to the user.

[0733] The server also receives best practice information from users, analyzes it, and adds it to the knowledge base, which allows other users to quickly provide appropriate answers when they ask similar questions.

[0734] Furthermore, the server periodically collects knowledge from within the organization and updates the knowledge base. During this process, it generates learning content that improves problem-solving skills using machine learning algorithms (e.g., TensorFlow and PyTorch). This allows users to receive individually optimized learning content based on their usage history.

[0735] Terminal

[0736] The terminal is a device that the user directly operates and is responsible for sending input from the user to the server. The user can enter questions or provide best practice information through the terminal. The terminal receives the information provided by the server and displays it to the user.

[0737] User

[0738] Users are employees or members of an organization who access the system through their terminals. They can enter questions to obtain information, share best practice information, and improve their problem-solving skills by utilizing self-learning content provided by the server.

[0739] Specific examples

[0740] A specific example is given below.

[0741] Specific examples of information search

[0742] The user types "What is the latest salary policy?" into the terminal and sends it. The server analyzes the question and searches the knowledge base for information about the latest salary policy. The server generates an answer, "The latest salary policy was updated on January 1, 2023. Click here for details." and sends it to the terminal. The terminal displays this answer to the user.

[0743] Examples of best practice sharing

[0744] A user types "It is important to apologize quickly when handling customer complaints" into a terminal and sends it. The server analyzes the information and adds it to the knowledge base. When another user asks "How do I handle customer complaints?", the server provides relevant best practices. The terminal displays the answer to the user.

[0745] Specific examples of self-study

[0746] The server regularly collects internal materials on new sales strategies and updates the knowledge base. When a user requests, "I want to learn about the latest trends in sales strategies," the server generates the most appropriate learning content and sends it to the device. The device displays "Latest Sales Strategies: The Impact of Digital Marketing and Effective Techniques" to the user. The user can study the content and use it in their work.

[0747] As described above, this invention provides a concrete means for enabling employees to quickly and easily access the information they need, which is expected to improve productivity and efficiency throughout the organization.

[0748] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0749] Step 1: Receiving a question from the user

[0750] User action: The user enters a question into the input interface of the terminal and clicks the send button.

[0751] Terminal operation: The terminal receives user input and sends the input text to the server.

[0752] Input: Text data of the question (e.g., "What is the latest salary policy?").

[0753] Output: Sending the query data to the server.

[0754] Step 2: Parsing the Question

[0755] Server operation: The server analyzes the questions received from the device using natural language processing technology. Specifically, it uses the Google Cloud Natural Language API to tokenize the text, tag it by part of speech, and analyze its meaning.

[0756] Input: Received text data.

[0757] Output: Analysis results (e.g., question intent and key keywords).

[0758] Step 3: Find related information

[0759] Server operation: Based on the analysis results, the server searches for the most relevant information from its internal knowledge base and related sources, using Elasticsearch to efficiently extract the information.

[0760] Input: A search query based on the parsed results.

[0761] Output: Search results (e.g., information about the latest salary policy).

[0762] Step 4: Generate an answer

[0763] Server operation: The server uses a generative AI model (e.g., OpenAI's GPT-3) to generate the optimal answer based on the searched information. The generated answer is then converted into a format that is easy for the user to understand.

[0764] Input: Search result data.

[0765] Output: The generated answer text (e.g., "The latest salary policy was updated on January 1, 2023. Click here for more information.").

[0766] Step 5: Submit and view your responses

[0767] Server operation: The server sends the generated answer to the terminal.

[0768] Terminal operation: The terminal displays the answer received from the server to the user. The user can check the answer on the screen and click on the details link.

[0769] Input: Generated response data.

[0770] Output: Display the answer to the user.

[0771] Step 6: Share best practices

[0772] User action: The user inputs best practice information into the terminal and sends it. For example, the user inputs and sends "It is important to apologize promptly when handling customer complaints."

[0773] Terminal action: The terminal sends this information to the server.

[0774] Server operation: The server analyzes the received information using natural language processing techniques and adds it to the knowledge base.

[0775] Input: Best practice information from users.

[0776] Output: The parsed best practice information is added to the knowledge base.

[0777] Step 7: Share best practices with others

[0778] Server operation: Other users enter similar questions into their terminals and send them to the server, which searches the knowledge base for the most appropriate best practice information, generates an answer, and sends it.

[0779] Terminal operation: The terminal receives the response and displays it to the user.

[0780] Input: Question data from other users.

[0781] Output: Appropriate best practice information is displayed to the user.

[0782] Step 8: Generate and deliver self-learning content

[0783] Server operation: The server periodically collects knowledge within the organization, updates the knowledge base, and generates learning content that improves problem-solving abilities using machine learning algorithms, using TensorFlow and PyTorch.

[0784] Input: Knowledge and user usage history data within the organization.

[0785] Output: Generated learning content (e.g., "Modern Sales Strategies: The Impact and Effective Techniques of Digital Marketing").

[0786] Step 9: Provide learning content to users

[0787] Server operation: The user inputs and sends a self-learning request (e.g., "I want to learn the latest trends in sales strategies") to the device. The server generates optimal learning content and sends it to the device.

[0788] Terminal operation: The terminal receives the learning content and displays it to the user.

[0789] Input: Self-learning request.

[0790] Output: The generated learning content is displayed to the user.

[0791] (Application example 1)

[0792] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0793] There is a need to improve worker efficiency within factories and achieve instant access to real-time information. In conventional systems, information on work procedures and troubleshooting was scattered, making it difficult to access quickly. Furthermore, there was a lack of an appropriate mechanism for responding to sudden on-site troubles and sharing best practices, which led to issues with worker efficiency and safety.

[0794] 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.

[0795] In this invention, the server includes means for receiving questions from users, means for analyzing the content of the questions using natural language processing technology, means for searching for related information from a knowledge base and information sources, means for generating an optimal answer based on the search results, means for providing the generated answer to the user, and means for the user to display work procedures and safety guidelines in real time via a mobile information terminal. This enables workers to quickly obtain the information and guidelines they need on-site, improving work efficiency and ensuring safety.

[0796] "User" refers to the worker or employee who operates the system and obtains information.

[0797] A "question" refers to a natural language inquiry that a user enters into a system seeking information.

[0798] "Natural language processing technology" refers to technology that enables computers to understand, analyze, and generate human language.

[0799] A "knowledge base" refers to a database that stores knowledge and information shared within an organization.

[0800] "Information Source" refers to a data source other than the Knowledge Base that provides relevant information.

[0801] "Answer" refers to information generated by the system in response to a user's question.

[0802] "Personal digital assistant" refers to mobile devices such as smart glasses and mobile phones.

[0803] "Best practice information" refers to information on methods and procedures that are considered to be the most effective and efficient in business operations and tasks.

[0804] "Analysis" refers to the process by which the system understands the questions and information it receives and extracts meaning from them.

[0805] "Related information" refers to information that can be used to find an appropriate answer to a user's question.

[0806] "Troubleshooting information" refers to information on methods and procedures for resolving malfunctions in machinery and equipment.

[0807] The present invention provides a system that enables workers to quickly obtain necessary information at a work site, thereby improving work efficiency and safety. Specific embodiments for carrying out the present invention will be described below.

[0808] System configuration

[0809] server

[0810] The server is the core of the system and has the following functions:

[0811] Natural language processing techniques are used to analyze user questions.

[0812] Search and generate the best answers to your questions from a knowledge base and related sources.

[0813] The results are provided to the user's mobile information terminal.

[0814] Analyze best practice information and add it to your knowledge base.

[0815] Regularly collect and update organizational knowledge.

[0816] Terminal

[0817] Terminals are devices operated directly by field workers and have the following functions:

[0818] Enter a question and send it to the server.

[0819] The response from the server is received and displayed in real time.

[0820] View instructions and troubleshooting information.

[0821] Share best practice information and send it to the server.

[0822] User

[0823] Users are factory workers and employees and have the following features:

[0824] The system is accessed through smart glasses or a mobile phone.

[0825] Check work procedures and safety guidelines in real time.

[0826] Get troubleshooting and maintenance procedures quickly.

[0827] Sharing best practice information will help build a knowledge base across the organization.

[0828] Specific operations and usage examples

[0829] 1. Display of work procedures

[0830] The user inputs a question through the smart glasses, such as "Please tell me the maintenance procedure for this machine." The server analyzes the question using natural language processing technology and searches for relevant information from a knowledge base. The maintenance procedure obtained from the search results is displayed on the user's smart glasses, allowing the work to be carried out efficiently.

[0831] 2. Troubleshooting Support

[0832] When a machine malfunctions, the user inputs a question through the smart glasses, such as "How do I deal with this error message?" The server analyzes the question and provides troubleshooting procedures by searching the knowledge base. This enables a rapid response to malfunctions.

[0833] 3. Sharing best practices

[0834] When a user discovers a new, efficient maintenance technique, they send the information to the server via the smart glasses. The server analyzes the information and adds it to a knowledge base. When other users encounter similar problems, they can share the best practice information added to the knowledge base.

[0835] Hardware and software used

[0836] Hardware:

[0837] Smart glasses (e.g. Google Glass, Vuzix Blade)

[0838] Server (high-performance data processing server)

[0839] software:

[0840] Natural language processing engine (e.g. Google Cloud Natural Language API)

[0841] Machine learning platforms (e.g. TensorFlow, PyTorch)

[0842] Web frameworks for the server (e.g., Flask, Django)

[0843] Prompt Sentence Examples

[0844] The prompt has the following format:

[0845] Towards a natural language processing engine

[0846] input:

[0847] Question: "What are the maintenance procedures for this machine?"

[0848] output:

[0849] Answer: "1. Turn off the power, 2. Remove the screws, 3. Clean the filter."

[0850] As described above, this invention makes it possible to dramatically improve work efficiency and safety by applying a knowledge base system to on-site factory work.

[0851] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0852] Step 1:

[0853] Users access the system using smart glasses or a personal digital assistant and enter their questions.

[0854] Specifically, the user uses the voice input function of the smart glasses to say, "Please tell me the maintenance procedure for this machine."

[0855] Input: User question (voice, text)

[0856] Output: Question data (text format)

[0857] Step 2:

[0858] The terminal transmits the question data received from the user to the server.

[0859] Specifically, the operation is to convert the received voice input into text data and send the text data to the server.

[0860] Input: Question data (text format)

[0861] Output: Send a query to the server

[0862] Step 3:

[0863] The server analyzes the received question using a natural language processing engine.

[0864] Specifically, it uses the Google Cloud Natural Language API to extract the intent and keywords of the question.

[0865] Input: Question data (text format)

[0866] Output: Question analysis results (keywords, intent)

[0867] Step 4:

[0868] The server searches for relevant information from knowledge bases and information sources based on the analysis results.

[0869] Specifically, it queries the database for relevant maintenance procedures and work procedures.

[0870] Input: Question analysis results (keywords, intent)

[0871] Output: Related information (maintenance procedures, work procedures)

[0872] Step 5:

[0873] The server generates the best answer based on the search results.

[0874] Specifically, it uses a generative AI model to generate answers to users' questions in text format.

[0875] Input: Related information (maintenance procedures, work procedures)

[0876] Output: Best answer (text format)

[0877] Step 6:

[0878] The server sends the generated response to the terminal.

[0879] Specifically, the generated answer is sent to the user's smart glasses via a network.

[0880] Input: Best answer (text format)

[0881] Output: Sends the answer to the terminal

[0882] Step 7:

[0883] The terminal displays the response received from the server to the user.

[0884] A specific operation is to display maintenance procedures and work procedures on the display of the smart glasses.

[0885] Input: Response from the server (text format)

[0886] Output: Display to the user (maintenance procedures, work procedures)

[0887] Step 8:

[0888] The user proceeds with the work based on the displayed information.

[0889] Specifically, the machine maintenance is carried out according to the displayed maintenance procedure.

[0890] Input: Information to be displayed to users (maintenance procedures, work procedures)

[0891] Output: Actual work progress

[0892] 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.

[0893] This invention provides a method for building an in-house knowledge base system that employees can easily access and obtain the information they need. This system exchanges information between a server, terminals, and users, and analyzes, generates, and shares information using natural language processing technology, machine learning algorithms, and emotion engines.

[0894] System configuration

[0895] server

[0896] The server plays a central role in the system, receiving questions from users and searching for information from a knowledge base and related information sources. It also uses natural language processing technology to analyze the questions and generate optimal answers. It also has the ability to recognize the emotions contained in the user's questions using an emotion engine and adjust the analysis results. It also adds best practice information to the knowledge base and regularly updates it. The server provides the generated answers and learning content to the device, and analyzes the user's usage history to provide individually optimized learning content.

[0897] Terminal

[0898] The terminal is a device that the user operates directly, sends input from the user to the server, and displays information provided by the server. The user can enter questions through the terminal and view answers and learning content from the server.

[0899] User

[0900] Users are employees or members of an organization who access the system through terminals. They can enter questions, obtain information, share best practice information, and utilize the learning content provided for self-study.

[0901] Program processing

[0902] Information search function

[0903] When the server receives a question from a user, it analyzes the question using natural language processing technology. It then uses an emotion engine to recognize the emotion contained in the question and adjust the analysis results. It then searches for relevant information from a knowledge base and related sources to generate the optimal answer. The generated answer is sent to the device and displayed to the user.

[0904] Best practice sharing feature

[0905] When a user submits best practice information from their device, the server analyzes the information and adds it to the knowledge base. This information is then available to other users when they ask similar questions. The information may also be displayed in an appropriate tone depending on the user's emotions.

[0906] Self-learning and problem-solving features

[0907] The server collects knowledge from within the organization and updates it regularly. It also uses machine learning algorithms to generate learning content that improves users' problem-solving abilities. It provides individually optimized learning content based on the user's usage history. An emotion engine also collects information about the user's emotions and uses it to provide future content.

[0908] Specific examples

[0909] Specific examples of information search

[0910] The user types "Please tell me the latest salary policy" into the terminal and sends it. The server analyzes the question and recognizes the user's emotion using an emotion engine. For example, if the server recognizes that the user is feeling anxious, it generates a gentle response saying, "The latest salary policy was updated on January 1, 2023. Click here for details." The terminal displays this response to the user.

[0911] Examples of best practice sharing

[0912] The user types "It is important to apologize quickly when handling customer complaints" into the terminal and sends it. The server analyzes the information, recognizes the emotion of the received information using an emotion engine, and adds it to the knowledge base. When another user asks "How do I handle customer complaints?", the server provides relevant best practices in an appropriate tone. The terminal displays the answer to the user.

[0913] Specific examples of self-study

[0914] When a user requests, "I want to learn about the latest trends in sales strategies," the server generates the most appropriate learning content and sends it to the device in a tone adjusted according to the user's emotions. For example, if the user is feeling very interested, the server will provide the content, "Latest sales strategies: The impact of digital marketing and effective techniques. Click here for more information." The device displays this content, and the user can learn from it and apply it to their work.

[0915] As described above, the present invention provides a concrete means for enabling employees to quickly and easily access the information they need while taking into consideration the user's feelings, thereby improving the productivity and efficiency of the entire organization.

[0916] The processing flow will be explained below.

[0917] Program processing flow

[0918] Information search function

[0919] Step 1:

[0920] The user enters a question in the input field on the terminal and presses the send button.

[0921] Step 2:

[0922] The terminal sends the user's question to the server.

[0923] Step 3:

[0924] The server receives the question and passes it to a natural language processing engine.

[0925] Step 4:

[0926] The server analyzes the question using a natural language processing engine.

[0927] Step 5:

[0928] The server requests the emotion engine to analyze the emotion of the question.

[0929] Step 6:

[0930] The emotion engine analyzes the emotion of the question and returns the emotion information to the server.

[0931] Step 7:

[0932] The server takes the emotion information into consideration and searches for relevant information from a knowledge base and related information sources.

[0933] Step 8:

[0934] The server generates the best answer based on the search results, adjusting the tone according to the user's emotions.

[0935] Step 9:

[0936] The server generates a response and sends it to the terminal.

[0937] Step 10:

[0938] The terminal receives the response from the server and displays it to the user.

[0939] Best practice sharing feature

[0940] Step 1:

[0941] The user enters best practice information into the input field on the device and presses the send button.

[0942] Step 2:

[0943] The device sends the best practice information to the server.

[0944] Step 3:

[0945] The server receives the best practice information and passes it to the natural language processing engine.

[0946] Step 4:

[0947] The server analyzes the best practice information using a natural language processing engine.

[0948] Step 5:

[0949] The server requests the emotion engine to analyze the emotion of the received information.

[0950] Step 6:

[0951] The emotion engine analyzes the emotion of the information and returns the emotion information to the server.

[0952] Step 7:

[0953] The server considers the analysis results and emotional information and adds them to the knowledge base.

[0954] Step 8:

[0955] A confirmation message is sent to the device to notify that the server has been successfully added to the knowledge base.

[0956] Step 9:

[0957] The terminal receives a confirmation message from the server and displays it to the user.

[0958] Self-learning and problem-solving features

[0959] Step 1:

[0960] The server collects knowledge within the organization and updates it periodically.

[0961] Step 2:

[0962] The server uses machine learning algorithms to generate learning content.

[0963] Step 3:

[0964] A user sends a request from a terminal to access self-learning content.

[0965] Step 4:

[0966] The server analyzes the user's usage history and generates individually optimized learning content.

[0967] Step 5:

[0968] The server requests the emotion engine to analyze the user's emotions.

[0969] Step 6:

[0970] The emotion engine returns the analysis results to the server.

[0971] Step 7:

[0972] The server optimizes learning content with an appropriate tone based on emotional information and sends it to the device.

[0973] Step 8:

[0974] The terminal receives the learning content from the server and displays it to the user.

[0975] Information sharing and collection functions

[0976] Step 1:

[0977] The user enters a question in the input field on the terminal and presses the send button.

[0978] Step 2:

[0979] The terminal sends a question to the server.

[0980] Step 3:

[0981] The server receives the question and passes it to a natural language processing engine.

[0982] Step 4:

[0983] The server analyzes the question using a natural language processing engine.

[0984] Step 5:

[0985] The server requests the emotion engine to analyze the emotion of the question.

[0986] Step 6:

[0987] The emotion engine analyzes the emotion of the question and returns the emotion information to the server.

[0988] Step 7:

[0989] The server takes the emotion information into consideration to search for relevant information from a knowledge base and related information sources.

[0990] Step 8:

[0991] The server generates the best answer based on the search results and adds it to the knowledge base.

[0992] Step 9:

[0993] The server generates a response in a tone based on the emotion information and transmits it to the terminal.

[0994] Step 10:

[0995] The terminal receives the response from the server and displays it to the user.

[0996] This allows the system, combined with the emotion engine, to recognize the user's emotions and adjust the tone of its analysis results and responses to provide more appropriate information. The specific processing flow allows users to access the information they need quickly and easily, improving productivity and efficiency across the organization.

[0997] Example 2

[0998] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0999] Conventional in-house knowledge base systems made it difficult for users to access the information they needed accurately and quickly. They also lacked appropriate feedback and learning content that took users' emotions into consideration. This led to problems such as reduced user satisfaction and work efficiency.

[1000] 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.

[1001] In this invention, the server includes means for receiving a question from a user, means for analyzing the content of the question using natural language processing technology, means for recognizing the emotion contained in the question and adjusting the analysis result, means for searching for related information from a knowledge base and information sources, means for generating an optimal answer according to the emotion based on the search results, and means for providing the generated answer to the user. This enables the user to quickly and accurately access the information they need and to receive appropriate feedback according to their emotion.

[1002] A "User" is a person or member of an organization who accesses the system to obtain or provide information.

[1003] A "server" is a central device that receives requests from users and analyzes, searches, generates, and provides information.

[1004] A "terminal" is a device that is directly operated by a user and that communicates with a server to send and receive information.

[1005] The "means for receiving questions" is a mechanism by which the server receives questions from users.

[1006] "Natural language processing technology" is a technology for analyzing questions entered by users, and involves grammatical and semantic analysis.

[1007] "Means for recognizing emotions" refers to technology that analyzes the emotions contained in a user's question and provides appropriate feedback.

[1008] A "knowledge base" is a database that stores past questions, best practices, and related information.

[1009] "Source" means a source of information, including external data or materials other than the Knowledge Base.

[1010] A "search method" is a mechanism for finding relevant information from knowledge bases and information sources.

[1011] "Answer generation means" refers to a technique for creating the most appropriate answer to a user's question based on search results.

[1012] "Best practices" are the optimal methods and techniques for improving business efficiency and results.

[1013] "Learning content" refers to content provided to help users improve their skills and solve problems.

[1014] A "machine learning algorithm" is a technology that analyzes large amounts of data, finds patterns, and makes predictions and classifications.

[1015] This invention provides a concrete means for realizing an in-house knowledge base system. The main components of the system include a server, terminals, and users, each of which plays a specific role. The details are explained below.

[1016] server

[1017] The server plays the central role in the system, receiving questions from users, analyzing them, and generating answers. The server uses the following main software technologies:

[1018] Natural language processing technology: We use "spaCy" and "BERT" to analyze user questions.

[1019] Emotion recognition technology: We use "Affectiva" to analyze the emotions contained in users' questions.

[1020] Knowledge base: A database such as MySQL or PostgreSQL is used to store best practices and related information.

[1021] Machine learning algorithms: "scikit-learn" and "TensorFlow" are used to generate learning content.

[1022] The server utilizes these technologies to generate appropriate answers to user questions, periodically updates the collected knowledge, and provides individually optimized learning content based on the user's usage history and emotional information.

[1023] For example, if a user asks, "What is the latest salary policy?", the server analyzes the question and uses emotion recognition technology to determine whether the user is feeling anxious. Based on the results, the server generates the most appropriate response and tone and sends the following response to the device: "The latest salary policy was updated on January 1, 2023. Click here for details."

[1024] Terminal

[1025] The terminal is a device operated by the user that sends and receives information to and from the server. Specifically, it has the following functions:

[1026] Submit user input: Enter a question or best practice information and submit it to the server.

[1027] Display information: Displays answers and learning content received from the server to the user.

[1028] For example, a user inputs "It is important to apologize promptly when handling customer complaints" into an input field on the terminal and sends it. The terminal transfers this information to the server, receives appropriate feedback from the server again, and displays it.

[1029] User

[1030] The user is the end user of the system and performs the following actions:

[1031] Entering a question: Requesting the required information from the system through the terminal.

[1032] Best practice information sharing: Providing useful information from the terminal to the system.

[1033] Utilizing learning content: Use the provided learning content for self-study and problem-solving.

[1034] For example, if a user types "I want to learn about the latest trends in sales strategies" into their device, the server will generate relevant learning content, adjust the tone according to their emotions, and provide information such as "The latest sales strategies: The impact of digital marketing and effective techniques. Click here for more information." Users can view this information and apply it to their work.

[1035] By combining these functions, it is possible to provide necessary information quickly and accurately while taking into consideration the user's feelings. Also, by providing individually optimized learning content, productivity and efficiency will be improved across the organization.

[1036] Prompt Sentence Examples

[1037] Use spaCy to parse the user question "What is the latest salary policy?" and process the sentiment with Affectiva to generate an answer with the appropriate tone.

[1038] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1039] Information retrieval function processing steps

[1040] Step 1: User enters question

[1041] The user types a question into a field on the device, such as "What's the new vacation policy?"

[1042] Input: User question text

[1043] Output: The question you typed is displayed on the terminal.

[1044] Step 2: The device sends a question to the server

[1045] The device sends the entered question to the server as an API request.

[1046] Input: Question text entered by the user

[1047] Output: API request sent to the server

[1048] Step 3: The server parses the question

[1049] The server analyzes the received question using natural language processing technology (e.g., "spaCy") to identify the main topics, and through this analysis, understands the context and intent of the question.

[1050] Input: Question text sent to the server

[1051] Output: Analysis of the question (e.g. "vacation policy")

[1052] Step 4: The server recognizes the emotion

[1053] The server uses emotion recognition technology (e.g., "Affectiva") to analyze the emotions contained in the questions, and through this recognition, it understands the user's emotional state and adjusts the tone of its answers accordingly.

[1054] Input: Analysis result of question content

[1055] Output: Sentiment analysis result (e.g., "anxiety")

[1056] Step 5: The server searches the knowledge base for information

[1057] The server searches for relevant information from a knowledge base and other sources based on the analysis results and sentiment analysis results.

[1058] Input: Question content analysis results and sentiment analysis results

[1059] Output: Search results (e.g. "new vacation policy details")

[1060] Step 6: Server generates answer

[1061] The server generates a response in an appropriate tone based on the sentiment analysis results, such as "Our latest vacation policy was updated on January 1, 2023. Click here for more information."

[1062] Input: Search results and sentiment analysis results

[1063] Output: Generated answer text

[1064] Step 7: The server sends the answer to the device

[1065] The server sends the generated answer to the terminal as an API response.

[1066] Input: Generated answer text

[1067] Output: API response sent to the device

[1068] Step 8: The device displays the answer to the user

[1069] The terminal displays the received answer to the user, who can then view it to find the answer to his or her question.

[1070] Input: Answer text sent from the server

[1071] Output: Answer displayed on the terminal

[1072] Best Practice Sharing Feature Processing Steps

[1073] Step 1: User inputs best practices

[1074] The user inputs best practice information into an input field on the terminal. For example, the user might input "It is important to apologize promptly when handling customer complaints."

[1075] Input: User best practice information

[1076] Output: The entered best practices are displayed on the terminal.

[1077] Step 2: The device sends the information to the server

[1078] The device sends the entered best practice information to the server as an API request.

[1079] Input: Best practice information entered by the user

[1080] Output: API request sent to the server

[1081] Step 3: The server analyzes the information

[1082] The server analyzes the received best practice information using natural language processing technology (e.g., "BERT") to identify key points.

[1083] Input: Best practice information sent to the server

[1084] Output: Best practice analysis results

[1085] Step 4: The server recognizes the emotion

[1086] The server uses emotion recognition technology to analyze the emotions contained in the best practice information, allowing it to present the information to other users in an appropriate tone.

[1087] Input: Best practice analysis results

[1088] Output: Emotion analysis results

[1089] Step 5: The server adds the information to the knowledge base

[1090] Based on the results of the analysis and sentiment analysis, the server adds new best practice information to the knowledge base, which can be used by other users when they ask similar questions.

[1091] Input: Best practice content analysis results and sentiment analysis results

[1092] Output: Information added to the Knowledge Base

[1093] Step 6: The server provides best practices for other users' questions

[1094] When other users ask similar questions, the server retrieves additional best practice information from the knowledge base and delivers it in the appropriate tone.

[1095] Input: Question from another user

[1096] Output: Best practice information provided

[1097] Self-learning and problem-solving processing steps

[1098] Step 1: User requests learning content

[1099] The user inputs "I want to learn the latest sales strategies" into the terminal and sends a request.

[1100] Input: User's learning content request

[1101] Output: The input request is displayed on the terminal.

[1102] Step 2: The device sends a request to the server

[1103] The device sends this request to the server as an API request.

[1104] Input: Learning content request entered by the user

[1105] Output: API request sent to the server

[1106] Step 3: Server checks the knowledge base

[1107] The server retrieves relevant learning material from a knowledge base.

[1108] Input: Learning content request sent to the server

[1109] Output: Search results (related learning materials)

[1110] Step 4: Server Generates Content

[1111] The server uses machine learning algorithms to generate learning content that best suits the request (for example, "scikit-learn" or "TensorFlow").

[1112] Input: Search results

[1113] Output: Generated learning content

[1114] Step 5: The server adjusts the tone based on the emotion

[1115] The server uses emotion recognition technology to tailor the learning content to a tone that corresponds to the user's emotions.

[1116] Input: Generated learning content and sentiment analysis results

[1117] Output: Tailored learning content

[1118] Step 6: The server sends the learning content to the device

[1119] The server sends the adjusted learning content to the device as an API response.

[1120] Input: Tailored learning content

[1121] Output: API response sent to the device

[1122] Step 7: The device displays the content to the user

[1123] The device displays the received learning content to the user, who can then view it and apply it to their work.

[1124] Input: Learning content sent from the server

[1125] Output: Learning content displayed on the device

[1126] (Application example 2)

[1127] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1128] While conventional in-house knowledge base systems allow users to quickly and easily access the information they need, they lack the ability to provide appropriate answers based on the user's emotions, making it difficult to respond appropriately, especially under stressful situations.In addition, for robots used on factory floors, there was a need for a means to provide optimal information in real time during troubleshooting and maintenance work.

[1129] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving a question from a user, means for analyzing the content of the question using natural language processing technology, means for searching for related information from a knowledge base and an information source, means for generating an optimal answer based on the search results, means for providing the generated answer to the user, emotion analysis means for recognizing the user's emotion and adjusting the analysis result, and means for displaying the generated answer via the robot's output device. This enables the factory robot to provide information in an appropriate tone according to the user's emotion and to support troubleshooting and maintenance work in real time.

[1130] A "user" is an individual or member of an organization who utilizes the system to enter questions and receive answers.

[1131] A "server" is a computer system that receives and analyzes questions from users, searches for relevant information from knowledge bases and information sources, and generates and provides optimal answers.

[1132] A "terminal" is a device that a user directly operates and connects to a server to input questions and receive answers.

[1133] A "knowledge base" is a database that stores knowledge and data collected from various sources.

[1134] "Natural language processing technology" is a technology that allows computers to understand and process human language.

[1135] A "machine learning algorithm" is an algorithm that finds patterns based on large amounts of data and automatically learns and predicts.

[1136] The "emotion engine" is a technology that recognizes emotions from user input and adjusts the tone of the response based on that information.

[1137] A "source" is a location where external or internal data or information exists outside of the knowledge base.

[1138] "Search Results" refers to relevant information retrieved from knowledge bases and information sources based on a user's question.

[1139] "Best practice information" is information that summarizes methods or procedures that are known to be most effective in a particular situation.

[1140] "Learning content" is educational content generated by the server to improve the user's problem-solving ability.

[1141] "Troubleshooting" is the process of identifying and resolving problems or errors in machines or systems.

[1142] "Maintenance work" refers to maintenance work carried out periodically or as needed to ensure stable operation of machines and systems.

[1143] "Sentiment analysis" is the process of detecting emotions from user input data and adapting that information.

[1144] An "output device" is a device for displaying generated information and answers to the user.

[1145] This invention provides an in-house knowledge base system that allows employees to easily access and obtain the information they need. The system receives questions from users, analyzes the questions using natural language processing technology, and searches for information from the knowledge base and related information sources to generate and provide optimal answers. It also includes a function that recognizes user emotions using an emotion engine and adjusts the analysis results. Furthermore, this system is implemented in factory robots to support troubleshooting and maintenance work.

[1146] System configuration and functions

[1147] server

[1148] The server receives questions from users and analyzes the content of the questions using natural language processing technology. Based on the analyzed question, it searches for relevant information from a knowledge base and information sources and generates the optimal answer. In doing so, it analyzes the user's emotions using an emotion engine and adjusts the analysis results. The server also has the function of receiving best practice information from users and adding it to the knowledge base. Furthermore, it uses machine learning algorithms to generate and provide learning content to improve the user's problem-solving ability.

[1149] Terminal

[1150] The terminal is a device that is directly operated by the user, allowing them to input questions, send them to the server, and display the information provided by the server. The terminal also functions as an output device for factory robots, displaying real-time information necessary for troubleshooting and maintenance work.

[1151] User

[1152] Users can use the system to enter questions and obtain the necessary information. For example, when a problem occurs at a factory, users can enter a question through their terminal and obtain the most appropriate answer from the server. Users can also provide their own best practice information to the system and share it with other users.

[1153] Hardware and software used

[1154] This system mainly uses the following hardware and software:

[1155] Hardware: Server equipment, user devices (PCs, tablets, smartphones, etc.), factory robots, output devices (displays)

[1156] Software: Natural language processing techniques (e.g., Hugging Face transformers), machine learning algorithms, sentiment analysis engines

[1157] Data processing and calculation

[1158] The server analyzes the user's question using natural language processing technology, then analyzes the user's emotions using an emotion engine. Based on the analysis results, it searches for relevant information from a knowledge base and information sources to generate the optimal answer. The tone of this answer is adjusted according to the user's emotions and sent to the device.

[1159] Specific examples

[1160] Specific examples of information search

[1161] The user types "What should you do if this machine stops?" into the terminal and sends it. The server analyzes the question and recognizes the user's emotions using an emotion engine. For example, if the server recognizes that the user is feeling anxious, it generates a response in a gentle tone: "If this machine stops, first press the emergency stop button to ensure safety. Then, follow the diagnostic steps below: 1. Check the power supply. 2. Check the connection cable." The terminal displays this response to the user.

[1162] Prompt Sentence Examples

[1163] Here are some examples of specific prompts:

[1164] "What should we do if this machine stops working?"

[1165] "What are your latest sales strategies?"

[1166] How do you handle customer complaints?

[1167] This system allows users to quickly obtain the information they need in a way that takes their emotions into consideration, making it possible to efficiently perform troubleshooting and maintenance work, particularly on factory floors.

[1168] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1169] Step 1:

[1170] The user operates the terminal to input a question and sends it to the server. The input data is a question about the information the user wants to know. For example, a prompt such as "What should we do if this machine stops working?" is input.

[1171] Step 2:

[1172] The server receives questions sent by users. It analyzes the received questions using natural language processing technology, tokenizing and tagging them to understand their meaning. The input data is the user's question, and the output is the structure of the analyzed question.

[1173] Step 3:

[1174] The server passes the analyzed question to the emotion engine to recognize the user's emotion. The emotion engine analyzes the tone and expression of the question to recognize emotions such as "anxiety," "interest," and "calm." The input data is the analyzed question, and the output is the recognized emotion information.

[1175] Step 4:

[1176] The server searches for relevant information from a knowledge base and related information sources based on the analysis results and emotion information. The knowledge base stores pre-accumulated knowledge, and generates search queries to extract relevant information. The input data are the analysis results and emotion information, and the output is the search results including related information.

[1177] Step 5:

[1178] The server generates the optimal answer based on the search results. In particular, it adjusts the tone of the answer depending on the user's emotions. It uses a generative AI model to construct an answer in natural language that combines the question and search results. The input data is the search results and emotional information, and the output is the optimal answer.

[1179] Step 6:

[1180] The server sends the generated answer to the terminal. The terminal displays the received answer to the user. This allows the user to quickly obtain the information they need in a way that takes their emotions into consideration. The input data is the generated answer, and the output is the answer displayed on the terminal screen.

[1181] Step 7:

[1182] The best practice information provided by the user is sent from the device to the server. The server analyzes the received information, recognizes the emotion information, and adds it to the knowledge base. The input data is the best practice information, and the output is an updated knowledge base.

[1183] Step 8:

[1184] The server uses machine learning algorithms to generate learning content to improve the user's problem-solving ability. The generated learning content is personalized based on the user's usage history and emotional information. The input data is usage history and emotional information, and the output is the generated learning content.

[1185] Specific examples

[1186] For example, if a user enters a prompt such as "What should you do if this machine stops working?", the server analyzes the question and recognizes the user's anxiety using an emotion engine. It then searches the knowledge base for relevant information and generates an answer such as "If this machine stops working, first press the emergency stop button to ensure safety. Then, follow the following diagnostic steps: 1. Check the power supply. 2. Check the connection cables." This answer is then sent to the terminal and displayed. In this way, the user can learn the appropriate measures.

[1187] 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.

[1188] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1189] 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.

[1190] [Third embodiment]

[1191] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1192] 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.

[1193] 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).

[1194] 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.

[1195] 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.

[1196] 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).

[1197] 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.

[1198] 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.

[1199] 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.

[1200] 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.

[1201] 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.

[1202] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1203] This invention provides a method for building an in-house knowledge base system that employees can easily access and obtain the information they need. This system exchanges information between a server, terminals, and users, and analyzes, generates, and shares information using natural language processing technology and machine learning algorithms.

[1204] System configuration

[1205] server

[1206] The server plays a central role in the system, receiving questions from users and searching for information from the knowledge base and related information sources. It also uses natural language processing technology to analyze the questions and generate the most appropriate answers. It also has the function of adding best practice information to the knowledge base and updating it regularly. The server provides the generated answers and learning content to the device, and analyzes the user's usage history to provide individually optimized learning content.

[1207] Terminal

[1208] The terminal is a device that the user operates directly, sends input from the user to the server, and displays information provided by the server. The user can enter questions through the terminal and view answers and learning content from the server.

[1209] User

[1210] Users are employees or members of an organization who access the system through terminals. They can enter questions, obtain information, share best practice information, and utilize the learning content provided for self-study.

[1211] Program processing

[1212] Information search function

[1213] When the server receives a user's question, it uses natural language processing technology to analyze the question, then searches for relevant information from a knowledge base and related sources to generate the most appropriate answer, which is then sent to the terminal and displayed to the user.

[1214] Best practice sharing feature

[1215] When users submit best practice information from their devices, the server analyzes the information and adds it to a knowledge base that can be used by other users when they have similar questions.

[1216] Self-learning and problem-solving features

[1217] The server collects and regularly updates knowledge within the organization. It also uses machine learning algorithms to generate learning content that improves users' problem-solving abilities. It provides individually optimized learning content based on the user's usage history.

[1218] Specific examples

[1219] Specific examples of information search

[1220] The user types "What is the latest salary policy?" into the terminal and sends it. The server analyzes the question and searches the knowledge base for information about the latest salary policy. The server generates an answer, "The latest salary policy was updated on January 1, 2023. Click here for details." and sends it to the terminal. The terminal displays this answer to the user.

[1221] Examples of best practice sharing

[1222] A user types "It is important to apologize quickly when handling customer complaints" into a terminal and sends it. The server analyzes the information and adds it to the knowledge base. When another user asks "How do I handle customer complaints?", the server provides relevant best practices. The terminal displays the answer to the user.

[1223] Specific examples of self-study

[1224] The server regularly collects internal materials on new sales strategies and updates the knowledge base. When a user requests, "I want to learn about the latest trends in sales strategies," the server generates the most appropriate learning content and sends it to the device. The device displays "Latest Sales Strategies: The Impact of Digital Marketing and Effective Techniques" to the user. The user can study the content and use it in their work.

[1225] As can be seen, the present invention provides a concrete means for enabling employees to quickly and easily access the information they need, thereby improving productivity and efficiency throughout an organization.

[1226] The processing flow will be explained below.

[1227] Program processing flow

[1228] Information search function

[1229] Step 1:

[1230] The user enters a question in the input field on the terminal and presses the send button.

[1231] Step 2:

[1232] The terminal sends the user's question to the server.

[1233] Step 3:

[1234] The server receives the question and passes it to a natural language processing engine.

[1235] Step 4:

[1236] The server analyzes the question using a natural language processing engine.

[1237] Step 5:

[1238] The server retrieves relevant information from a knowledge base and related sources.

[1239] Step 6:

[1240] The server generates the best answer based on the search results.

[1241] Step 7:

[1242] The server generates a response and sends it to the terminal.

[1243] Step 8:

[1244] The terminal receives the response from the server and displays it to the user.

[1245] Best practice sharing feature

[1246] Step 1:

[1247] The user enters best practice information into the input field on the device and presses the send button.

[1248] Step 2:

[1249] The device sends the best practice information to the server.

[1250] Step 3:

[1251] The server receives the best practice information and passes it to the natural language processing engine.

[1252] Step 4:

[1253] The server analyzes the information using a natural language processing engine.

[1254] Step 5:

[1255] The server adds the analysis results to the knowledge base.

[1256] Step 6:

[1257] A confirmation message is sent to the device to notify that the server has been successfully added to the knowledge base.

[1258] Step 7:

[1259] The terminal receives a confirmation message from the server and displays it to the user.

[1260] Self-learning and problem-solving features

[1261] Step 1:

[1262] The server collects knowledge within the organization and updates it periodically.

[1263] Step 2:

[1264] The server uses machine learning algorithms to generate learning content.

[1265] Step 3:

[1266] A user sends a request from a terminal to access self-learning content.

[1267] Step 4:

[1268] The server analyzes the user's usage history and selects individually optimized learning content based on the user's request.

[1269] Step 5:

[1270] The server transmits the selected learning content to the terminal.

[1271] Step 6:

[1272] The terminal receives the learning content from the server and displays it to the user.

[1273] Information sharing and collection functions

[1274] Step 1:

[1275] The user enters a question in the input field on the terminal and presses the send button.

[1276] Step 2:

[1277] The terminal sends a question to the server.

[1278] Step 3:

[1279] The server receives the question and passes it to a natural language processing engine.

[1280] Step 4:

[1281] The server analyzes the question using a natural language processing engine.

[1282] Step 5:

[1283] The server retrieves relevant information from a knowledge base and related sources.

[1284] Step 6:

[1285] The server generates the best answer based on the search results.

[1286] Step 7:

[1287] Add the server-generated answer to the knowledge base.

[1288] Step 8:

[1289] The server generates a response and sends it to the terminal.

[1290] Step 9:

[1291] The terminal receives the response from the server and displays it to the user.

[1292] The above process flow allows users to quickly and easily access the information they need, improving productivity and efficiency across the organization.

[1293] Example 1

[1294] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1295] In order for employees to perform their work efficiently, they need an internal knowledge base system that allows them to quickly and easily access the information they need. However, conventional systems often lack efficient information search and answer generation, resulting in problems such as users having to wait a long time to find the information they need. Furthermore, because knowledge base updates are performed manually, the information can become outdated. Furthermore, the lack of functionality to provide learning content optimized for individual users makes it difficult to maximize users' problem-solving abilities and work efficiency.

[1296] 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.

[1297] In this invention, the server includes: means for receiving questions from users; means for analyzing the content of the questions using natural language processing technology; means for searching for relevant information from a knowledge base and related information sources; means for generating an optimal answer based on the search results; means for providing the answer in a format easy to understand for the user using a generative AI model; and means for transmitting the generated answer to a terminal and displaying it to the user. This allows users to quickly search for information and obtain the optimal answer. Furthermore, by including means for receiving best practice information from users and adding it to the knowledge base and means for providing best practice information added to the knowledge base in response to questions from other users, the knowledge base information is always up-to-date and rich. Furthermore, by including means for collecting and regularly updating knowledge within an organization, means for generating learning content to improve problem-solving skills using machine learning algorithms, and means for providing personalized learning content based on the user's usage history, it is possible to promote users' self-learning and improve work efficiency.

[1298] The "means for receiving questions from users" is a function by which the server receives questions and information input by users using their terminals.

[1299] "Means for analyzing the content of a question using natural language processing technology" is a function for analyzing the text of a received question using natural language processing technology and understanding its meaning and intent.

[1300] The "means for searching for relevant information from a knowledge base and related information sources" is a function for searching for target information from an internal knowledge base and related information sources based on the content of the analyzed question.

[1301] The "means for generating the most appropriate answer based on the search results" is a function for generating the most appropriate answer for the user based on the searched information.

[1302] "Means of using a generative AI model to provide answers in a format that is easy for users to understand" refers to a function that uses a generative AI model to convert generated answers into a format that is easy for users to understand and provide them.

[1303] The "means for transmitting the generated answer to the terminal and displaying it to the user" is a function for transmitting the answer generated by the server to the terminal so that the user can view the answer on the terminal.

[1304] The "means for receiving best practice information from a user" is a function in which a user inputs best practices that they know via a terminal, and the server receives that information.

[1305] The "means for analyzing received best practice information and adding it to a knowledge base" is a function for analyzing best practice information received from a user and adding it to the organization's knowledge base.

[1306] The "means for providing best practice information added to the knowledge base in response to questions from other users" is a function for searching for and providing best practice information added to the knowledge base in response to questions from other users.

[1307] The "means for collecting and periodically updating knowledge within an organization" is a function for continuously collecting knowledge generated within an organization and periodically updating the knowledge base.

[1308] "Means for generating learning content to improve problem-solving ability using machine learning algorithms" is a function that uses machine learning algorithms to generate learning content to improve a user's problem-solving ability.

[1309] "Means for providing personalized learning content based on a user's usage history" is a function that provides individually optimized learning content based on a user's past usage history.

[1310] This invention provides a method for building an in-house knowledge base system that allows employees to easily access and quickly obtain the information they need. This system exchanges information between a server, terminals, and users, and analyzes, generates, and shares information using natural language processing technology and machine learning algorithms.

[1311] System configuration

[1312] server

[1313] The server plays a central role in the system and receives questions from users. Once a question is received, it is analyzed using natural language processing technology. For this analysis, Google Cloud Natural Language API can be used as a general analysis tool. After analyzing the question, related information is searched for from the knowledge base and related information sources. Specifically, Elasticsearch is used for the search to efficiently extract related information.

[1314] The server can utilize a generative AI model (such as OpenAI's GPT-3) to generate the optimal answer based on the search results. The generated answer may not be understandable to the user as it is, so it is converted into a user-friendly format through the generative AI model. This converted answer is sent to the device and displayed to the user.

[1315] The server also receives best practice information from users, analyzes it, and adds it to the knowledge base, which allows other users to quickly provide appropriate answers when they ask similar questions.

[1316] Furthermore, the server periodically collects knowledge from within the organization and updates the knowledge base. During this process, it generates learning content that improves problem-solving skills using machine learning algorithms (e.g., TensorFlow and PyTorch). This allows users to receive individually optimized learning content based on their usage history.

[1317] Terminal

[1318] The terminal is a device that the user directly operates and is responsible for sending input from the user to the server. The user can enter questions or provide best practice information through the terminal. The terminal receives the information provided by the server and displays it to the user.

[1319] User

[1320] Users are employees or members of an organization who access the system through their terminals. They can enter questions to obtain information, share best practice information, and improve their problem-solving skills by utilizing self-learning content provided by the server.

[1321] Specific examples

[1322] A specific example is given below.

[1323] Specific examples of information search

[1324] The user types "What is the latest salary policy?" into the terminal and sends it. The server analyzes the question and searches the knowledge base for information about the latest salary policy. The server generates an answer, "The latest salary policy was updated on January 1, 2023. Click here for details." and sends it to the terminal. The terminal displays this answer to the user.

[1325] Examples of best practice sharing

[1326] A user types "It is important to apologize quickly when handling customer complaints" into a terminal and sends it. The server analyzes the information and adds it to the knowledge base. When another user asks "How do I handle customer complaints?", the server provides relevant best practices. The terminal displays the answer to the user.

[1327] Specific examples of self-study

[1328] The server regularly collects internal materials on new sales strategies and updates the knowledge base. When a user requests, "I want to learn about the latest trends in sales strategies," the server generates the most appropriate learning content and sends it to the device. The device displays "Latest Sales Strategies: The Impact of Digital Marketing and Effective Techniques" to the user. The user can study the content and use it in their work.

[1329] As described above, this invention provides a concrete means for enabling employees to quickly and easily access the information they need, which is expected to improve productivity and efficiency throughout the organization.

[1330] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1331] Step 1: Receiving a question from the user

[1332] User action: The user enters a question into the input interface of the terminal and clicks the send button.

[1333] Terminal operation: The terminal receives user input and sends the input text to the server.

[1334] Input: Text data of the question (e.g., "What is the latest salary policy?").

[1335] Output: Sending the query data to the server.

[1336] Step 2: Parsing the Question

[1337] Server operation: The server analyzes the questions received from the device using natural language processing technology. Specifically, it uses the Google Cloud Natural Language API to tokenize the text, tag it by part of speech, and analyze its meaning.

[1338] Input: Received text data.

[1339] Output: Analysis results (e.g., question intent and key keywords).

[1340] Step 3: Find related information

[1341] Server operation: Based on the analysis results, the server searches for the most relevant information from its internal knowledge base and related sources, using Elasticsearch to efficiently extract the information.

[1342] Input: A search query based on the parsed results.

[1343] Output: Search results (e.g., information about the latest salary policy).

[1344] Step 4: Generate an answer

[1345] Server operation: The server uses a generative AI model (e.g., OpenAI's GPT-3) to generate the optimal answer based on the searched information. The generated answer is then converted into a format that is easy for the user to understand.

[1346] Input: Search result data.

[1347] Output: The generated answer text (e.g., "The latest salary policy was updated on January 1, 2023. Click here for more information.").

[1348] Step 5: Submit and view your responses

[1349] Server operation: The server sends the generated answer to the terminal.

[1350] Terminal operation: The terminal displays the answer received from the server to the user. The user can check the answer on the screen and click on the details link.

[1351] Input: Generated response data.

[1352] Output: Display the answer to the user.

[1353] Step 6: Share best practices

[1354] User action: The user inputs best practice information into the terminal and sends it. For example, the user inputs and sends "It is important to apologize promptly when handling customer complaints."

[1355] Terminal action: The terminal sends this information to the server.

[1356] Server operation: The server analyzes the received information using natural language processing techniques and adds it to the knowledge base.

[1357] Input: Best practice information from users.

[1358] Output: The parsed best practice information is added to the knowledge base.

[1359] Step 7: Share best practices with others

[1360] Server operation: Other users enter similar questions into their terminals and send them to the server, which searches the knowledge base for the most appropriate best practice information, generates an answer, and sends it.

[1361] Terminal operation: The terminal receives the response and displays it to the user.

[1362] Input: Question data from other users.

[1363] Output: Appropriate best practice information is displayed to the user.

[1364] Step 8: Generate and deliver self-learning content

[1365] Server operation: The server periodically collects knowledge within the organization, updates the knowledge base, and generates learning content that improves problem-solving abilities using machine learning algorithms, using TensorFlow and PyTorch.

[1366] Input: Knowledge and user usage history data within the organization.

[1367] Output: Generated learning content (e.g., "Modern Sales Strategies: The Impact and Effective Techniques of Digital Marketing").

[1368] Step 9: Provide learning content to users

[1369] Server operation: The user inputs and sends a self-learning request (e.g., "I want to learn the latest trends in sales strategies") to the device. The server generates optimal learning content and sends it to the device.

[1370] Terminal operation: The terminal receives the learning content and displays it to the user.

[1371] Input: Self-learning request.

[1372] Output: The generated learning content is displayed to the user.

[1373] (Application example 1)

[1374] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1375] There is a need to improve worker efficiency within factories and achieve instant access to real-time information. In conventional systems, information on work procedures and troubleshooting was scattered, making it difficult to access quickly. Furthermore, there was a lack of an appropriate mechanism for responding to sudden on-site troubles and sharing best practices, which led to issues with worker efficiency and safety.

[1376] 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.

[1377] In this invention, the server includes means for receiving questions from users, means for analyzing the content of the questions using natural language processing technology, means for searching for related information from a knowledge base and information sources, means for generating an optimal answer based on the search results, means for providing the generated answer to the user, and means for the user to display work procedures and safety guidelines in real time via a mobile information terminal. This enables workers to quickly obtain the information and guidelines they need on-site, improving work efficiency and ensuring safety.

[1378] "User" refers to the worker or employee who operates the system and obtains information.

[1379] A "question" refers to a natural language inquiry that a user enters into a system seeking information.

[1380] "Natural language processing technology" refers to technology that enables computers to understand, analyze, and generate human language.

[1381] A "knowledge base" refers to a database that stores knowledge and information shared within an organization.

[1382] "Information Source" refers to a data source other than the Knowledge Base that provides relevant information.

[1383] "Answer" refers to information generated by the system in response to a user's question.

[1384] "Personal digital assistant" refers to mobile devices such as smart glasses and mobile phones.

[1385] "Best practice information" refers to information on methods and procedures that are considered to be the most effective and efficient in business operations and tasks.

[1386] "Analysis" refers to the process by which the system understands the questions and information it receives and extracts meaning from them.

[1387] "Related information" refers to information that can be used to find an appropriate answer to a user's question.

[1388] "Troubleshooting information" refers to information on methods and procedures for resolving malfunctions in machinery and equipment.

[1389] The present invention provides a system that enables workers to quickly obtain necessary information at a work site, thereby improving work efficiency and safety. Specific embodiments for carrying out the present invention will be described below.

[1390] System configuration

[1391] server

[1392] The server is the core of the system and has the following functions:

[1393] Natural language processing techniques are used to analyze user questions.

[1394] Search and generate the best answers to your questions from a knowledge base and related sources.

[1395] The results are provided to the user's mobile information terminal.

[1396] Analyze best practice information and add it to your knowledge base.

[1397] Regularly collect and update organizational knowledge.

[1398] Terminal

[1399] Terminals are devices operated directly by field workers and have the following functions:

[1400] Enter a question and send it to the server.

[1401] The response from the server is received and displayed in real time.

[1402] View instructions and troubleshooting information.

[1403] Share best practice information and send it to the server.

[1404] User

[1405] Users are factory workers and employees and have the following features:

[1406] The system is accessed through smart glasses or a mobile phone.

[1407] Check work procedures and safety guidelines in real time.

[1408] Get troubleshooting and maintenance procedures quickly.

[1409] Sharing best practice information will help build a knowledge base across the organization.

[1410] Specific operations and usage examples

[1411] 1. Display of work procedures

[1412] The user inputs a question through the smart glasses, such as "Please tell me the maintenance procedure for this machine." The server analyzes the question using natural language processing technology and searches for relevant information from a knowledge base. The maintenance procedure obtained from the search results is displayed on the user's smart glasses, allowing the work to be carried out efficiently.

[1413] 2. Troubleshooting Support

[1414] When a machine malfunctions, the user inputs a question through the smart glasses, such as "How do I deal with this error message?" The server analyzes the question and provides troubleshooting procedures by searching the knowledge base. This enables a rapid response to malfunctions.

[1415] 3. Sharing best practices

[1416] When a user discovers a new, efficient maintenance technique, they send the information to the server via the smart glasses. The server analyzes the information and adds it to a knowledge base. When other users encounter similar problems, they can share the best practice information added to the knowledge base.

[1417] Hardware and software used

[1418] Hardware:

[1419] Smart glasses (e.g. Google Glass, Vuzix Blade)

[1420] Server (high-performance data processing server)

[1421] software:

[1422] Natural language processing engine (e.g. Google Cloud Natural Language API)

[1423] Machine learning platforms (e.g. TensorFlow, PyTorch)

[1424] Web frameworks for the server (e.g., Flask, Django)

[1425] Prompt Sentence Examples

[1426] The prompt has the following format:

[1427] Towards a natural language processing engine

[1428] input:

[1429] Question: "What are the maintenance procedures for this machine?"

[1430] output:

[1431] Answer: "1. Turn off the power, 2. Remove the screws, 3. Clean the filter."

[1432] As described above, this invention makes it possible to dramatically improve work efficiency and safety by applying a knowledge base system to on-site factory work.

[1433] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1434] Step 1:

[1435] Users access the system using smart glasses or a personal digital assistant and enter their questions.

[1436] Specifically, the user uses the voice input function of the smart glasses to say, "Please tell me the maintenance procedure for this machine."

[1437] Input: User question (voice, text)

[1438] Output: Question data (text format)

[1439] Step 2:

[1440] The terminal transmits the question data received from the user to the server.

[1441] Specifically, the operation is to convert the received voice input into text data and send the text data to the server.

[1442] Input: Question data (text format)

[1443] Output: Send a query to the server

[1444] Step 3:

[1445] The server analyzes the received question using a natural language processing engine.

[1446] Specifically, it uses the Google Cloud Natural Language API to extract the intent and keywords of the question.

[1447] Input: Question data (text format)

[1448] Output: Question analysis results (keywords, intent)

[1449] Step 4:

[1450] The server searches for relevant information from knowledge bases and information sources based on the analysis results.

[1451] Specifically, it queries the database for relevant maintenance procedures and work procedures.

[1452] Input: Question analysis results (keywords, intent)

[1453] Output: Related information (maintenance procedures, work procedures)

[1454] Step 5:

[1455] The server generates the best answer based on the search results.

[1456] Specifically, it uses a generative AI model to generate answers to users' questions in text format.

[1457] Input: Related information (maintenance procedures, work procedures)

[1458] Output: Best answer (text format)

[1459] Step 6:

[1460] The server sends the generated response to the terminal.

[1461] Specifically, the generated answer is sent to the user's smart glasses via a network.

[1462] Input: Best answer (text format)

[1463] Output: Sends the answer to the terminal

[1464] Step 7:

[1465] The terminal displays the response received from the server to the user.

[1466] A specific operation is to display maintenance procedures and work procedures on the display of the smart glasses.

[1467] Input: Response from the server (text format)

[1468] Output: Display to the user (maintenance procedures, work procedures)

[1469] Step 8:

[1470] The user proceeds with the work based on the displayed information.

[1471] Specifically, the machine maintenance is carried out according to the displayed maintenance procedure.

[1472] Input: Information to be displayed to users (maintenance procedures, work procedures)

[1473] Output: Actual work progress

[1474] 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.

[1475] This invention provides a method for building an in-house knowledge base system that employees can easily access and obtain the information they need. This system exchanges information between a server, terminals, and users, and analyzes, generates, and shares information using natural language processing technology, machine learning algorithms, and emotion engines.

[1476] System configuration

[1477] server

[1478] The server plays a central role in the system, receiving questions from users and searching for information from a knowledge base and related information sources. It also uses natural language processing technology to analyze the questions and generate optimal answers. It also has the ability to recognize the emotions contained in the user's questions using an emotion engine and adjust the analysis results. It also adds best practice information to the knowledge base and regularly updates it. The server provides the generated answers and learning content to the device, and analyzes the user's usage history to provide individually optimized learning content.

[1479] Terminal

[1480] The terminal is a device that the user operates directly, sends input from the user to the server, and displays information provided by the server. The user can enter questions through the terminal and view answers and learning content from the server.

[1481] User

[1482] Users are employees or members of an organization who access the system through terminals. They can enter questions, obtain information, share best practice information, and utilize the learning content provided for self-study.

[1483] Program processing

[1484] Information search function

[1485] When the server receives a question from a user, it analyzes the question using natural language processing technology. It then uses an emotion engine to recognize the emotion contained in the question and adjust the analysis results. It then searches for relevant information from a knowledge base and related sources to generate the optimal answer. The generated answer is sent to the device and displayed to the user.

[1486] Best practice sharing feature

[1487] When a user submits best practice information from their device, the server analyzes the information and adds it to the knowledge base. This information is then available to other users when they ask similar questions. The information may also be displayed in an appropriate tone depending on the user's emotions.

[1488] Self-learning and problem-solving features

[1489] The server collects knowledge from within the organization and updates it regularly. It also uses machine learning algorithms to generate learning content that improves users' problem-solving abilities. It provides individually optimized learning content based on the user's usage history. An emotion engine also collects information about the user's emotions and uses it to provide future content.

[1490] Specific examples

[1491] Specific examples of information search

[1492] The user types "Please tell me the latest salary policy" into the terminal and sends it. The server analyzes the question and recognizes the user's emotion using an emotion engine. For example, if the server recognizes that the user is feeling anxious, it generates a gentle response saying, "The latest salary policy was updated on January 1, 2023. Click here for details." The terminal displays this response to the user.

[1493] Examples of best practice sharing

[1494] The user types "It is important to apologize quickly when handling customer complaints" into the terminal and sends it. The server analyzes the information, recognizes the emotion of the received information using an emotion engine, and adds it to the knowledge base. When another user asks "How do I handle customer complaints?", the server provides relevant best practices in an appropriate tone. The terminal displays the answer to the user.

[1495] Specific examples of self-study

[1496] When a user requests, "I want to learn about the latest trends in sales strategies," the server generates the most appropriate learning content and sends it to the device in a tone adjusted according to the user's emotions. For example, if the user is feeling very interested, the server will provide the content, "Latest sales strategies: The impact of digital marketing and effective techniques. Click here for more information." The device displays this content, and the user can learn from it and apply it to their work.

[1497] As described above, the present invention provides a concrete means for enabling employees to quickly and easily access the information they need while taking into consideration the user's feelings, thereby improving the productivity and efficiency of the entire organization.

[1498] The processing flow will be explained below.

[1499] Program processing flow

[1500] Information search function

[1501] Step 1:

[1502] The user enters a question in the input field on the terminal and presses the send button.

[1503] Step 2:

[1504] The terminal sends the user's question to the server.

[1505] Step 3:

[1506] The server receives the question and passes it to a natural language processing engine.

[1507] Step 4:

[1508] The server analyzes the question using a natural language processing engine.

[1509] Step 5:

[1510] The server requests the emotion engine to analyze the emotion of the question.

[1511] Step 6:

[1512] The emotion engine analyzes the emotion of the question and returns the emotion information to the server.

[1513] Step 7:

[1514] The server takes the emotion information into consideration and searches for relevant information from a knowledge base and related information sources.

[1515] Step 8:

[1516] The server generates the best answer based on the search results, adjusting the tone according to the user's emotions.

[1517] Step 9:

[1518] The server generates a response and sends it to the terminal.

[1519] Step 10:

[1520] The terminal receives the response from the server and displays it to the user.

[1521] Best practice sharing feature

[1522] Step 1:

[1523] The user enters best practice information into the input field on the device and presses the send button.

[1524] Step 2:

[1525] The device sends the best practice information to the server.

[1526] Step 3:

[1527] The server receives the best practice information and passes it to the natural language processing engine.

[1528] Step 4:

[1529] The server analyzes the best practice information using a natural language processing engine.

[1530] Step 5:

[1531] The server requests the emotion engine to analyze the emotion of the received information.

[1532] Step 6:

[1533] The emotion engine analyzes the emotion of the information and returns the emotion information to the server.

[1534] Step 7:

[1535] The server considers the analysis results and emotional information and adds them to the knowledge base.

[1536] Step 8:

[1537] A confirmation message is sent to the device to notify that the server has been successfully added to the knowledge base.

[1538] Step 9:

[1539] The terminal receives a confirmation message from the server and displays it to the user.

[1540] Self-learning and problem-solving features

[1541] Step 1:

[1542] The server collects knowledge within the organization and updates it periodically.

[1543] Step 2:

[1544] The server uses machine learning algorithms to generate learning content.

[1545] Step 3:

[1546] A user sends a request from a terminal to access self-learning content.

[1547] Step 4:

[1548] The server analyzes the user's usage history and generates individually optimized learning content.

[1549] Step 5:

[1550] The server requests the emotion engine to analyze the user's emotions.

[1551] Step 6:

[1552] The emotion engine returns the analysis results to the server.

[1553] Step 7:

[1554] The server optimizes learning content with an appropriate tone based on emotional information and sends it to the device.

[1555] Step 8:

[1556] The terminal receives the learning content from the server and displays it to the user.

[1557] Information sharing and collection functions

[1558] Step 1:

[1559] The user enters a question in the input field on the terminal and presses the send button.

[1560] Step 2:

[1561] The terminal sends a question to the server.

[1562] Step 3:

[1563] The server receives the question and passes it to a natural language processing engine.

[1564] Step 4:

[1565] The server analyzes the question using a natural language processing engine.

[1566] Step 5:

[1567] The server requests the emotion engine to analyze the emotion of the question.

[1568] Step 6:

[1569] The emotion engine analyzes the emotion of the question and returns the emotion information to the server.

[1570] Step 7:

[1571] The server takes the emotion information into consideration to search for relevant information from a knowledge base and related information sources.

[1572] Step 8:

[1573] The server generates the best answer based on the search results and adds it to the knowledge base.

[1574] Step 9:

[1575] The server generates a response in a tone based on the emotion information and transmits it to the terminal.

[1576] Step 10:

[1577] The terminal receives the response from the server and displays it to the user.

[1578] This allows the system, combined with the emotion engine, to recognize the user's emotions and adjust the tone of its analysis results and responses to provide more appropriate information. The specific processing flow allows users to access the information they need quickly and easily, improving productivity and efficiency across the organization.

[1579] Example 2

[1580] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1581] Conventional in-house knowledge base systems made it difficult for users to access the information they needed accurately and quickly. They also lacked appropriate feedback and learning content that took users' emotions into consideration. This led to problems such as reduced user satisfaction and work efficiency.

[1582] 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.

[1583] In this invention, the server includes means for receiving a question from a user, means for analyzing the content of the question using natural language processing technology, means for recognizing the emotion contained in the question and adjusting the analysis result, means for searching for related information from a knowledge base and information sources, means for generating an optimal answer according to the emotion based on the search results, and means for providing the generated answer to the user. This enables the user to quickly and accurately access the information they need and to receive appropriate feedback according to their emotion.

[1584] A "User" is a person or member of an organization who accesses the system to obtain or provide information.

[1585] A "server" is a central device that receives requests from users and analyzes, searches, generates, and provides information.

[1586] A "terminal" is a device that is directly operated by a user and that communicates with a server to send and receive information.

[1587] The "means for receiving questions" is a mechanism by which the server receives questions from users.

[1588] "Natural language processing technology" is a technology for analyzing questions entered by users, and involves grammatical and semantic analysis.

[1589] "Means for recognizing emotions" refers to technology that analyzes the emotions contained in a user's question and provides appropriate feedback.

[1590] A "knowledge base" is a database that stores past questions, best practices, and related information.

[1591] "Source" means a source of information, including external data or materials other than the Knowledge Base.

[1592] A "search method" is a mechanism for finding relevant information from knowledge bases and information sources.

[1593] "Answer generation means" refers to a technique for creating the most appropriate answer to a user's question based on search results.

[1594] "Best practices" are the optimal methods and techniques for improving business efficiency and results.

[1595] "Learning content" refers to content provided to help users improve their skills and solve problems.

[1596] A "machine learning algorithm" is a technology that analyzes large amounts of data, finds patterns, and makes predictions and classifications.

[1597] This invention provides a concrete means for realizing an in-house knowledge base system. The main components of the system include a server, terminals, and users, each of which plays a specific role. The details are explained below.

[1598] server

[1599] The server plays the central role in the system, receiving questions from users, analyzing them, and generating answers. The server uses the following main software technologies:

[1600] Natural language processing technology: We use "spaCy" and "BERT" to analyze user questions.

[1601] Emotion recognition technology: We use "Affectiva" to analyze the emotions contained in users' questions.

[1602] Knowledge base: A database such as MySQL or PostgreSQL is used to store best practices and related information.

[1603] Machine learning algorithms: "scikit-learn" and "TensorFlow" are used to generate learning content.

[1604] The server utilizes these technologies to generate appropriate answers to user questions, periodically updates the collected knowledge, and provides individually optimized learning content based on the user's usage history and emotional information.

[1605] For example, if a user asks, "What is the latest salary policy?", the server analyzes the question and uses emotion recognition technology to determine whether the user is feeling anxious. Based on the results, the server generates the most appropriate response and tone and sends the following response to the device: "The latest salary policy was updated on January 1, 2023. Click here for details."

[1606] Terminal

[1607] The terminal is a device operated by the user that sends and receives information to and from the server. Specifically, it has the following functions:

[1608] Submit user input: Enter a question or best practice information and submit it to the server.

[1609] Display information: Displays answers and learning content received from the server to the user.

[1610] For example, a user inputs "It is important to apologize promptly when handling customer complaints" into an input field on the terminal and sends it. The terminal transfers this information to the server, receives appropriate feedback from the server again, and displays it.

[1611] User

[1612] The user is the end user of the system and performs the following actions:

[1613] Entering a question: Requesting the required information from the system through the terminal.

[1614] Best practice information sharing: Providing useful information from the terminal to the system.

[1615] Utilizing learning content: Use the provided learning content for self-study and problem-solving.

[1616] For example, if a user types "I want to learn about the latest trends in sales strategies" into their device, the server will generate relevant learning content, adjust the tone according to their emotions, and provide information such as "The latest sales strategies: The impact of digital marketing and effective techniques. Click here for more information." Users can view this information and apply it to their work.

[1617] By combining these functions, it is possible to provide necessary information quickly and accurately while taking into consideration the user's feelings. Also, by providing individually optimized learning content, productivity and efficiency will be improved across the organization.

[1618] Prompt Sentence Examples

[1619] Use spaCy to parse the user question "What is the latest salary policy?" and process the sentiment with Affectiva to generate an answer with the appropriate tone.

[1620] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1621] Information retrieval function processing steps

[1622] Step 1: User enters question

[1623] The user types a question into a field on the device, such as "What's the new vacation policy?"

[1624] Input: User question text

[1625] Output: The question you typed is displayed on the terminal.

[1626] Step 2: The device sends a question to the server

[1627] The device sends the entered question to the server as an API request.

[1628] Input: Question text entered by the user

[1629] Output: API request sent to the server

[1630] Step 3: The server parses the question

[1631] The server analyzes the received question using natural language processing technology (e.g., "spaCy") to identify the main topics, and through this analysis, understands the context and intent of the question.

[1632] Input: Question text sent to the server

[1633] Output: Analysis of the question (e.g. "vacation policy")

[1634] Step 4: The server recognizes the emotion

[1635] The server uses emotion recognition technology (e.g., "Affectiva") to analyze the emotions contained in the questions, and through this recognition, it understands the user's emotional state and adjusts the tone of its answers accordingly.

[1636] Input: Analysis result of question content

[1637] Output: Sentiment analysis result (e.g., "anxiety")

[1638] Step 5: The server searches the knowledge base for information

[1639] The server searches for relevant information from a knowledge base and other sources based on the analysis results and sentiment analysis results.

[1640] Input: Question content analysis results and sentiment analysis results

[1641] Output: Search results (e.g. "new vacation policy details")

[1642] Step 6: Server generates answer

[1643] The server generates a response in an appropriate tone based on the sentiment analysis results, such as "Our latest vacation policy was updated on January 1, 2023. Click here for more information."

[1644] Input: Search results and sentiment analysis results

[1645] Output: Generated answer text

[1646] Step 7: The server sends the answer to the device

[1647] The server sends the generated answer to the terminal as an API response.

[1648] Input: Generated answer text

[1649] Output: API response sent to the device

[1650] Step 8: The device displays the answer to the user

[1651] The terminal displays the received answer to the user, who can then view it to find the answer to his or her question.

[1652] Input: Answer text sent from the server

[1653] Output: Answer displayed on the terminal

[1654] Best Practice Sharing Feature Processing Steps

[1655] Step 1: User inputs best practices

[1656] The user inputs best practice information into an input field on the terminal. For example, the user might input "It is important to apologize promptly when handling customer complaints."

[1657] Input: User best practice information

[1658] Output: The entered best practices are displayed on the terminal.

[1659] Step 2: The device sends the information to the server

[1660] The device sends the entered best practice information to the server as an API request.

[1661] Input: Best practice information entered by the user

[1662] Output: API request sent to the server

[1663] Step 3: The server analyzes the information

[1664] The server analyzes the received best practice information using natural language processing technology (e.g., "BERT") to identify key points.

[1665] Input: Best practice information sent to the server

[1666] Output: Best practice analysis results

[1667] Step 4: The server recognizes the emotion

[1668] The server uses emotion recognition technology to analyze the emotions contained in the best practice information, allowing it to present the information to other users in an appropriate tone.

[1669] Input: Best practice analysis results

[1670] Output: Emotion analysis results

[1671] Step 5: The server adds the information to the knowledge base

[1672] Based on the results of the analysis and sentiment analysis, the server adds new best practice information to the knowledge base, which can be used by other users when they ask similar questions.

[1673] Input: Best practice content analysis results and sentiment analysis results

[1674] Output: Information added to the Knowledge Base

[1675] Step 6: The server provides best practices for other users' questions

[1676] When other users ask similar questions, the server retrieves additional best practice information from the knowledge base and delivers it in the appropriate tone.

[1677] Input: Question from another user

[1678] Output: Best practice information provided

[1679] Self-learning and problem-solving processing steps

[1680] Step 1: User requests learning content

[1681] The user inputs "I want to learn the latest sales strategies" into the terminal and sends a request.

[1682] Input: User's learning content request

[1683] Output: The input request is displayed on the terminal.

[1684] Step 2: The device sends a request to the server

[1685] The device sends this request to the server as an API request.

[1686] Input: Learning content request entered by the user

[1687] Output: API request sent to the server

[1688] Step 3: Server checks the knowledge base

[1689] The server retrieves relevant learning material from a knowledge base.

[1690] Input: Learning content request sent to the server

[1691] Output: Search results (related learning materials)

[1692] Step 4: Server Generates Content

[1693] The server uses machine learning algorithms to generate learning content that best suits the request (for example, "scikit-learn" or "TensorFlow").

[1694] Input: Search results

[1695] Output: Generated learning content

[1696] Step 5: The server adjusts the tone based on the emotion

[1697] The server uses emotion recognition technology to tailor the learning content to a tone that corresponds to the user's emotions.

[1698] Input: Generated learning content and sentiment analysis results

[1699] Output: Tailored learning content

[1700] Step 6: The server sends the learning content to the device

[1701] The server sends the adjusted learning content to the device as an API response.

[1702] Input: Tailored learning content

[1703] Output: API response sent to the device

[1704] Step 7: The device displays the content to the user

[1705] The device displays the received learning content to the user, who can then view it and apply it to their work.

[1706] Input: Learning content sent from the server

[1707] Output: Learning content displayed on the device

[1708] (Application example 2)

[1709] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1710] While conventional in-house knowledge base systems allow users to quickly and easily access the information they need, they lack the ability to provide appropriate answers based on the user's emotions, making it difficult to respond appropriately, especially under stressful situations.In addition, for robots used on factory floors, there was a need for a means to provide optimal information in real time during troubleshooting and maintenance work.

[1711] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving a question from a user, means for analyzing the content of the question using natural language processing technology, means for searching for related information from a knowledge base and an information source, means for generating an optimal answer based on the search results, means for providing the generated answer to the user, emotion analysis means for recognizing the user's emotion and adjusting the analysis result, and means for displaying the generated answer via the robot's output device. This enables the factory robot to provide information in an appropriate tone according to the user's emotion and to support troubleshooting and maintenance work in real time.

[1712] A "user" is an individual or member of an organization who utilizes the system to enter questions and receive answers.

[1713] A "server" is a computer system that receives and analyzes questions from users, searches for relevant information from knowledge bases and information sources, and generates and provides optimal answers.

[1714] A "terminal" is a device that a user directly operates and connects to a server to input questions and receive answers.

[1715] A "knowledge base" is a database that stores knowledge and data collected from various sources.

[1716] "Natural language processing technology" is a technology that allows computers to understand and process human language.

[1717] A "machine learning algorithm" is an algorithm that finds patterns based on large amounts of data and automatically learns and predicts.

[1718] The "emotion engine" is a technology that recognizes emotions from user input and adjusts the tone of the response based on that information.

[1719] A "source" is a location where external or internal data or information exists outside of the knowledge base.

[1720] "Search Results" refers to relevant information retrieved from knowledge bases and information sources based on a user's question.

[1721] "Best practice information" is information that summarizes methods or procedures that are known to be most effective in a particular situation.

[1722] "Learning content" is educational content generated by the server to improve the user's problem-solving ability.

[1723] "Troubleshooting" is the process of identifying and resolving problems or errors in machines or systems.

[1724] "Maintenance work" refers to maintenance work carried out periodically or as needed to ensure stable operation of machines and systems.

[1725] "Sentiment analysis" is the process of detecting emotions from user input data and adapting that information.

[1726] An "output device" is a device for displaying generated information and answers to the user.

[1727] This invention provides an in-house knowledge base system that allows employees to easily access and obtain the information they need. The system receives questions from users, analyzes the questions using natural language processing technology, and searches for information from the knowledge base and related information sources to generate and provide optimal answers. It also includes a function that recognizes user emotions using an emotion engine and adjusts the analysis results. Furthermore, this system is implemented in factory robots to support troubleshooting and maintenance work.

[1728] System configuration and functions

[1729] server

[1730] The server receives questions from users and analyzes the content of the questions using natural language processing technology. Based on the analyzed question, it searches for relevant information from a knowledge base and information sources and generates the optimal answer. In doing so, it analyzes the user's emotions using an emotion engine and adjusts the analysis results. The server also has the function of receiving best practice information from users and adding it to the knowledge base. Furthermore, it uses machine learning algorithms to generate and provide learning content to improve the user's problem-solving ability.

[1731] Terminal

[1732] The terminal is a device that is directly operated by the user, allowing them to input questions, send them to the server, and display the information provided by the server. The terminal also functions as an output device for factory robots, displaying real-time information necessary for troubleshooting and maintenance work.

[1733] User

[1734] Users can use the system to enter questions and obtain the necessary information. For example, when a problem occurs at a factory, users can enter a question through their terminal and obtain the most appropriate answer from the server. Users can also provide their own best practice information to the system and share it with other users.

[1735] Hardware and software used

[1736] This system mainly uses the following hardware and software:

[1737] Hardware: Server equipment, user devices (PCs, tablets, smartphones, etc.), factory robots, output devices (displays)

[1738] Software: Natural language processing techniques (e.g., Hugging Face transformers), machine learning algorithms, sentiment analysis engines

[1739] Data processing and calculation

[1740] The server analyzes the user's question using natural language processing technology, then analyzes the user's emotions using an emotion engine. Based on the analysis results, it searches for relevant information from a knowledge base and information sources to generate the optimal answer. The tone of this answer is adjusted according to the user's emotions and sent to the device.

[1741] Specific examples

[1742] Specific examples of information search

[1743] The user types "What should you do if this machine stops?" into the terminal and sends it. The server analyzes the question and recognizes the user's emotions using an emotion engine. For example, if the server recognizes that the user is feeling anxious, it generates a response in a gentle tone: "If this machine stops, first press the emergency stop button to ensure safety. Then, follow the diagnostic steps below: 1. Check the power supply. 2. Check the connection cable." The terminal displays this response to the user.

[1744] Prompt Sentence Examples

[1745] Here are some examples of specific prompts:

[1746] "What should we do if this machine stops working?"

[1747] "What are your latest sales strategies?"

[1748] How do you handle customer complaints?

[1749] This system allows users to quickly obtain the information they need in a way that takes their emotions into consideration, making it possible to efficiently perform troubleshooting and maintenance work, particularly on factory floors.

[1750] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1751] Step 1:

[1752] The user operates the terminal to input a question and sends it to the server. The input data is a question about the information the user wants to know. For example, a prompt such as "What should we do if this machine stops working?" is input.

[1753] Step 2:

[1754] The server receives questions sent by users. It analyzes the received questions using natural language processing technology, tokenizing and tagging them to understand their meaning. The input data is the user's question, and the output is the structure of the analyzed question.

[1755] Step 3:

[1756] The server passes the analyzed question to the emotion engine to recognize the user's emotion. The emotion engine analyzes the tone and expression of the question to recognize emotions such as "anxiety," "interest," and "calm." The input data is the analyzed question, and the output is the recognized emotion information.

[1757] Step 4:

[1758] The server searches for relevant information from a knowledge base and related information sources based on the analysis results and emotion information. The knowledge base stores pre-accumulated knowledge, and generates search queries to extract relevant information. The input data are the analysis results and emotion information, and the output is the search results including related information.

[1759] Step 5:

[1760] The server generates the optimal answer based on the search results. In particular, it adjusts the tone of the answer depending on the user's emotions. It uses a generative AI model to construct an answer in natural language that combines the question and search results. The input data is the search results and emotional information, and the output is the optimal answer.

[1761] Step 6:

[1762] The server sends the generated answer to the terminal. The terminal displays the received answer to the user. This allows the user to quickly obtain the information they need in a way that takes their emotions into consideration. The input data is the generated answer, and the output is the answer displayed on the terminal screen.

[1763] Step 7:

[1764] The best practice information provided by the user is sent from the device to the server. The server analyzes the received information, recognizes the emotion information, and adds it to the knowledge base. The input data is the best practice information, and the output is an updated knowledge base.

[1765] Step 8:

[1766] The server uses machine learning algorithms to generate learning content to improve the user's problem-solving ability. The generated learning content is personalized based on the user's usage history and emotional information. The input data is usage history and emotional information, and the output is the generated learning content.

[1767] Specific examples

[1768] For example, if a user enters a prompt such as "What should you do if this machine stops working?", the server analyzes the question and recognizes the user's anxiety using an emotion engine. It then searches the knowledge base for relevant information and generates an answer such as "If this machine stops working, first press the emergency stop button to ensure safety. Then, follow the following diagnostic steps: 1. Check the power supply. 2. Check the connection cables." This answer is then sent to the terminal and displayed. In this way, the user can learn the appropriate measures.

[1769] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1770] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1771] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1772] [Fourth embodiment]

[1773] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1774] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1775] 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).

[1776] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1777] 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.

[1778] 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).

[1779] 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.

[1780] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1781] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1782] 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.

[1783] 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.

[1784] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1785] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1786] This invention provides a method for building an in-house knowledge base system that employees can easily access and obtain the information they need. This system exchanges information between a server, terminals, and users, and analyzes, generates, and shares information using natural language processing technology and machine learning algorithms.

[1787] System configuration

[1788] server

[1789] The server plays a central role in the system, receiving questions from users and searching for information from the knowledge base and related information sources. It also uses natural language processing technology to analyze the questions and generate the most appropriate answers. It also has the function of adding best practice information to the knowledge base and updating it regularly. The server provides the generated answers and learning content to the device, and analyzes the user's usage history to provide individually optimized learning content.

[1790] Terminal

[1791] The terminal is a device that the user operates directly, sends input from the user to the server, and displays information provided by the server. The user can enter questions through the terminal and view answers and learning content from the server.

[1792] User

[1793] Users are employees or members of an organization who access the system through terminals. They can enter questions, obtain information, share best practice information, and utilize the learning content provided for self-study.

[1794] Program processing

[1795] Information search function

[1796] When the server receives a user's question, it uses natural language processing technology to analyze the question, then searches for relevant information from a knowledge base and related sources to generate the most appropriate answer, which is then sent to the terminal and displayed to the user.

[1797] Best practice sharing feature

[1798] When users submit best practice information from their devices, the server analyzes the information and adds it to a knowledge base that can be used by other users when they have similar questions.

[1799] Self-learning and problem-solving features

[1800] The server collects and regularly updates knowledge within the organization. It also uses machine learning algorithms to generate learning content that improves users' problem-solving abilities. It provides individually optimized learning content based on the user's usage history.

[1801] Specific examples

[1802] Specific examples of information search

[1803] The user types "What is the latest salary policy?" into the terminal and sends it. The server analyzes the question and searches the knowledge base for information about the latest salary policy. The server generates an answer, "The latest salary policy was updated on January 1, 2023. Click here for details." and sends it to the terminal. The terminal displays this answer to the user.

[1804] Examples of best practice sharing

[1805] A user types "It is important to apologize quickly when handling customer complaints" into a terminal and sends it. The server analyzes the information and adds it to the knowledge base. When another user asks "How do I handle customer complaints?", the server provides relevant best practices. The terminal displays the answer to the user.

[1806] Specific examples of self-study

[1807] The server regularly collects internal materials on new sales strategies and updates the knowledge base. When a user requests, "I want to learn about the latest trends in sales strategies," the server generates the most appropriate learning content and sends it to the device. The device displays "Latest Sales Strategies: The Impact of Digital Marketing and Effective Techniques" to the user. The user can study the content and use it in their work.

[1808] As can be seen, the present invention provides a concrete means for enabling employees to quickly and easily access the information they need, thereby improving productivity and efficiency throughout an organization.

[1809] The processing flow will be explained below.

[1810] Program processing flow

[1811] Information search function

[1812] Step 1:

[1813] The user enters a question in the input field on the terminal and presses the send button.

[1814] Step 2:

[1815] The terminal sends the user's question to the server.

[1816] Step 3:

[1817] The server receives the question and passes it to a natural language processing engine.

[1818] Step 4:

[1819] The server analyzes the question using a natural language processing engine.

[1820] Step 5:

[1821] The server retrieves relevant information from a knowledge base and related sources.

[1822] Step 6:

[1823] The server generates the best answer based on the search results.

[1824] Step 7:

[1825] The server generates a response and sends it to the terminal.

[1826] Step 8:

[1827] The terminal receives the response from the server and displays it to the user.

[1828] Best practice sharing feature

[1829] Step 1:

[1830] The user enters best practice information into the input field on the device and presses the send button.

[1831] Step 2:

[1832] The device sends the best practice information to the server.

[1833] Step 3:

[1834] The server receives the best practice information and passes it to the natural language processing engine.

[1835] Step 4:

[1836] The server analyzes the information using a natural language processing engine.

[1837] Step 5:

[1838] The server adds the analysis results to the knowledge base.

[1839] Step 6:

[1840] A confirmation message is sent to the device to notify that the server has been successfully added to the knowledge base.

[1841] Step 7:

[1842] The terminal receives a confirmation message from the server and displays it to the user.

[1843] Self-learning and problem-solving features

[1844] Step 1:

[1845] The server collects knowledge within the organization and updates it periodically.

[1846] Step 2:

[1847] The server uses machine learning algorithms to generate learning content.

[1848] Step 3:

[1849] A user sends a request from a terminal to access self-learning content.

[1850] Step 4:

[1851] The server analyzes the user's usage history and selects individually optimized learning content based on the user's request.

[1852] Step 5:

[1853] The server transmits the selected learning content to the terminal.

[1854] Step 6:

[1855] The terminal receives the learning content from the server and displays it to the user.

[1856] Information sharing and collection functions

[1857] Step 1:

[1858] The user enters a question in the input field on the terminal and presses the send button.

[1859] Step 2:

[1860] The terminal sends a question to the server.

[1861] Step 3:

[1862] The server receives the question and passes it to a natural language processing engine.

[1863] Step 4:

[1864] The server analyzes the question using a natural language processing engine.

[1865] Step 5:

[1866] The server retrieves relevant information from a knowledge base and related sources.

[1867] Step 6:

[1868] The server generates the best answer based on the search results.

[1869] Step 7:

[1870] Add the server-generated answer to the knowledge base.

[1871] Step 8:

[1872] The server generates a response and sends it to the terminal.

[1873] Step 9:

[1874] The terminal receives the response from the server and displays it to the user.

[1875] The above process flow allows users to quickly and easily access the information they need, improving productivity and efficiency across the organization.

[1876] Example 1

[1877] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1878] In order for employees to perform their work efficiently, they need an internal knowledge base system that allows them to quickly and easily access the information they need. However, conventional systems often lack efficient information search and answer generation, resulting in problems such as users having to wait a long time to find the information they need. Furthermore, because knowledge base updates are performed manually, the information can become outdated. Furthermore, the lack of functionality to provide learning content optimized for individual users makes it difficult to maximize users' problem-solving abilities and work efficiency.

[1879] 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.

[1880] In this invention, the server includes: means for receiving questions from users; means for analyzing the content of the questions using natural language processing technology; means for searching for relevant information from a knowledge base and related information sources; means for generating an optimal answer based on the search results; means for providing the answer in a format easy to understand for the user using a generative AI model; and means for transmitting the generated answer to a terminal and displaying it to the user. This allows users to quickly search for information and obtain the optimal answer. Furthermore, by including means for receiving best practice information from users and adding it to the knowledge base and means for providing best practice information added to the knowledge base in response to questions from other users, the knowledge base information is always up-to-date and rich. Furthermore, by including means for collecting and regularly updating knowledge within an organization, means for generating learning content to improve problem-solving skills using machine learning algorithms, and means for providing personalized learning content based on the user's usage history, it is possible to promote users' self-learning and improve work efficiency.

[1881] The "means for receiving questions from users" is a function by which the server receives questions and information input by users using their terminals.

[1882] "Means for analyzing the content of a question using natural language processing technology" is a function for analyzing the text of a received question using natural language processing technology and understanding its meaning and intent.

[1883] The "means for searching for relevant information from a knowledge base and related information sources" is a function for searching for target information from an internal knowledge base and related information sources based on the content of the analyzed question.

[1884] The "means for generating the most appropriate answer based on the search results" is a function for generating the most appropriate answer for the user based on the searched information.

[1885] "Means of using a generative AI model to provide answers in a format that is easy for users to understand" refers to a function that uses a generative AI model to convert generated answers into a format that is easy for users to understand and provide them.

[1886] The "means for transmitting the generated answer to the terminal and displaying it to the user" is a function for transmitting the answer generated by the server to the terminal so that the user can view the answer on the terminal.

[1887] The "means for receiving best practice information from a user" is a function in which a user inputs best practices that they know via a terminal, and the server receives that information.

[1888] The "means for analyzing received best practice information and adding it to a knowledge base" is a function for analyzing best practice information received from a user and adding it to the organization's knowledge base.

[1889] The "means for providing best practice information added to the knowledge base in response to questions from other users" is a function for searching for and providing best practice information added to the knowledge base in response to questions from other users.

[1890] The "means for collecting and periodically updating knowledge within an organization" is a function for continuously collecting knowledge generated within an organization and periodically updating the knowledge base.

[1891] "Means for generating learning content to improve problem-solving ability using machine learning algorithms" is a function that uses machine learning algorithms to generate learning content to improve a user's problem-solving ability.

[1892] "Means for providing personalized learning content based on a user's usage history" is a function that provides individually optimized learning content based on a user's past usage history.

[1893] This invention provides a method for building an in-house knowledge base system that allows employees to easily access and quickly obtain the information they need. This system exchanges information between a server, terminals, and users, and analyzes, generates, and shares information using natural language processing technology and machine learning algorithms.

[1894] System configuration

[1895] server

[1896] The server plays a central role in the system and receives questions from users. Once a question is received, it is analyzed using natural language processing technology. For this analysis, Google Cloud Natural Language API can be used as a general analysis tool. After analyzing the question, related information is searched for from the knowledge base and related information sources. Specifically, Elasticsearch is used for the search to efficiently extract related information.

[1897] The server can utilize a generative AI model (such as OpenAI's GPT-3) to generate the optimal answer based on the search results. The generated answer may not be understandable to the user as it is, so it is converted into a user-friendly format through the generative AI model. This converted answer is sent to the device and displayed to the user.

[1898] The server also receives best practice information from users, analyzes it, and adds it to the knowledge base, which allows other users to quickly provide appropriate answers when they ask similar questions.

[1899] Furthermore, the server periodically collects knowledge from within the organization and updates the knowledge base. During this process, it generates learning content that improves problem-solving skills using machine learning algorithms (e.g., TensorFlow and PyTorch). This allows users to receive individually optimized learning content based on their usage history.

[1900] Terminal

[1901] The terminal is a device that the user directly operates and is responsible for sending input from the user to the server. The user can enter questions or provide best practice information through the terminal. The terminal receives the information provided by the server and displays it to the user.

[1902] User

[1903] Users are employees or members of an organization who access the system through their terminals. They can enter questions to obtain information, share best practice information, and improve their problem-solving skills by utilizing self-learning content provided by the server.

[1904] Specific examples

[1905] A specific example is given below.

[1906] Specific examples of information search

[1907] The user types "What is the latest salary policy?" into the terminal and sends it. The server analyzes the question and searches the knowledge base for information about the latest salary policy. The server generates an answer, "The latest salary policy was updated on January 1, 2023. Click here for details." and sends it to the terminal. The terminal displays this answer to the user.

[1908] Examples of best practice sharing

[1909] A user types "It is important to apologize quickly when handling customer complaints" into a terminal and sends it. The server analyzes the information and adds it to the knowledge base. When another user asks "How do I handle customer complaints?", the server provides relevant best practices. The terminal displays the answer to the user.

[1910] Specific examples of self-study

[1911] The server regularly collects internal materials on new sales strategies and updates the knowledge base. When a user requests, "I want to learn about the latest trends in sales strategies," the server generates the most appropriate learning content and sends it to the device. The device displays "Latest Sales Strategies: The Impact of Digital Marketing and Effective Techniques" to the user. The user can study the content and use it in their work.

[1912] As described above, this invention provides a concrete means for enabling employees to quickly and easily access the information they need, which is expected to improve productivity and efficiency throughout the organization.

[1913] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1914] Step 1: Receiving a question from the user

[1915] User action: The user enters a question into the input interface of the terminal and clicks the send button.

[1916] Terminal operation: The terminal receives user input and sends the input text to the server.

[1917] Input: Text data of the question (e.g., "What is the latest salary policy?").

[1918] Output: Sending the query data to the server.

[1919] Step 2: Parsing the Question

[1920] Server operation: The server analyzes the questions received from the device using natural language processing technology. Specifically, it uses the Google Cloud Natural Language API to tokenize the text, tag it by part of speech, and analyze its meaning.

[1921] Input: Received text data.

[1922] Output: Analysis results (e.g., question intent and key keywords).

[1923] Step 3: Find related information

[1924] Server operation: Based on the analysis results, the server searches for the most relevant information from its internal knowledge base and related sources, using Elasticsearch to efficiently extract the information.

[1925] Input: A search query based on the parsed results.

[1926] Output: Search results (e.g., information about the latest salary policy).

[1927] Step 4: Generate an answer

[1928] Server operation: The server uses a generative AI model (e.g., OpenAI's GPT-3) to generate the optimal answer based on the searched information. The generated answer is then converted into a format that is easy for the user to understand.

[1929] Input: Search result data.

[1930] Output: The generated answer text (e.g., "The latest salary policy was updated on January 1, 2023. Click here for more information.").

[1931] Step 5: Submit and view your responses

[1932] Server operation: The server sends the generated answer to the terminal.

[1933] Terminal operation: The terminal displays the answer received from the server to the user. The user can check the answer on the screen and click on the details link.

[1934] Input: Generated response data.

[1935] Output: Display the answer to the user.

[1936] Step 6: Share best practices

[1937] User action: The user inputs best practice information into the terminal and sends it. For example, the user inputs and sends "It is important to apologize promptly when handling customer complaints."

[1938] Terminal action: The terminal sends this information to the server.

[1939] Server operation: The server analyzes the received information using natural language processing techniques and adds it to the knowledge base.

[1940] Input: Best practice information from users.

[1941] Output: The parsed best practice information is added to the knowledge base.

[1942] Step 7: Share best practices with others

[1943] Server operation: Other users enter similar questions into their terminals and send them to the server, which searches the knowledge base for the most appropriate best practice information, generates an answer, and sends it.

[1944] Terminal operation: The terminal receives the response and displays it to the user.

[1945] Input: Question data from other users.

[1946] Output: Appropriate best practice information is displayed to the user.

[1947] Step 8: Generate and deliver self-learning content

[1948] Server operation: The server periodically collects knowledge within the organization, updates the knowledge base, and generates learning content that improves problem-solving abilities using machine learning algorithms, using TensorFlow and PyTorch.

[1949] Input: Knowledge and user usage history data within the organization.

[1950] Output: Generated learning content (e.g., "Modern Sales Strategies: The Impact and Effective Techniques of Digital Marketing").

[1951] Step 9: Provide learning content to users

[1952] Server operation: The user inputs and sends a self-learning request (e.g., "I want to learn the latest trends in sales strategies") to the device. The server generates optimal learning content and sends it to the device.

[1953] Terminal operation: The terminal receives the learning content and displays it to the user.

[1954] Input: Self-learning request.

[1955] Output: The generated learning content is displayed to the user.

[1956] (Application example 1)

[1957] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1958] There is a need to improve worker efficiency within factories and achieve instant access to real-time information. In conventional systems, information on work procedures and troubleshooting was scattered, making it difficult to access quickly. Furthermore, there was a lack of an appropriate mechanism for responding to sudden on-site troubles and sharing best practices, which led to issues with worker efficiency and safety.

[1959] 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.

[1960] In this invention, the server includes means for receiving questions from users, means for analyzing the content of the questions using natural language processing technology, means for searching for related information from a knowledge base and information sources, means for generating an optimal answer based on the search results, means for providing the generated answer to the user, and means for the user to display work procedures and safety guidelines in real time via a mobile information terminal. This enables workers to quickly obtain the information and guidelines they need on-site, improving work efficiency and ensuring safety.

[1961] "User" refers to the worker or employee who operates the system and obtains information.

[1962] A "question" refers to a natural language inquiry that a user enters into a system seeking information.

[1963] "Natural language processing technology" refers to technology that enables computers to understand, analyze, and generate human language.

[1964] A "knowledge base" refers to a database that stores knowledge and information shared within an organization.

[1965] "Information Source" refers to a data source other than the Knowledge Base that provides relevant information.

[1966] "Answer" refers to information generated by the system in response to a user's question.

[1967] "Personal digital assistant" refers to mobile devices such as smart glasses and mobile phones.

[1968] "Best practice information" refers to information on methods and procedures that are considered to be the most effective and efficient in business operations and tasks.

[1969] "Analysis" refers to the process by which the system understands the questions and information it receives and extracts meaning from them.

[1970] "Related information" refers to information that can be used to find an appropriate answer to a user's question.

[1971] "Troubleshooting information" refers to information on methods and procedures for resolving malfunctions in machinery and equipment.

[1972] The present invention provides a system that enables workers to quickly obtain necessary information at a work site, thereby improving work efficiency and safety. Specific embodiments for carrying out the present invention will be described below.

[1973] System configuration

[1974] server

[1975] The server is the core of the system and has the following functions:

[1976] Natural language processing techniques are used to analyze user questions.

[1977] Search and generate the best answers to your questions from a knowledge base and related sources.

[1978] The results are provided to the user's mobile information terminal.

[1979] Analyze best practice information and add it to your knowledge base.

[1980] Regularly collect and update organizational knowledge.

[1981] Terminal

[1982] Terminals are devices operated directly by field workers and have the following functions:

[1983] Enter a question and send it to the server.

[1984] The response from the server is received and displayed in real time.

[1985] View instructions and troubleshooting information.

[1986] Share best practice information and send it to the server.

[1987] User

[1988] Users are factory workers and employees and have the following features:

[1989] The system is accessed through smart glasses or a mobile phone.

[1990] Check work procedures and safety guidelines in real time.

[1991] Get troubleshooting and maintenance procedures quickly.

[1992] Sharing best practice information will help build a knowledge base across the organization.

[1993] Specific operations and usage examples

[1994] 1. Display of work procedures

[1995] The user inputs a question through the smart glasses, such as "Please tell me the maintenance procedure for this machine." The server analyzes the question using natural language processing technology and searches for relevant information from a knowledge base. The maintenance procedure obtained from the search results is displayed on the user's smart glasses, allowing the work to be carried out efficiently.

[1996] 2. Troubleshooting Support

[1997] When a machine malfunctions, the user inputs a question through the smart glasses, such as "How do I deal with this error message?" The server analyzes the question and provides troubleshooting procedures by searching the knowledge base. This enables a rapid response to malfunctions.

[1998] 3. Sharing best practices

[1999] When a user discovers a new, efficient maintenance technique, they send the information to the server via the smart glasses. The server analyzes the information and adds it to a knowledge base. When other users encounter similar problems, they can share the best practice information added to the knowledge base.

[2000] Hardware and software used

[2001] Hardware:

[2002] Smart glasses (e.g. Google Glass, Vuzix Blade)

[2003] Server (high-performance data processing server)

[2004] software:

[2005] Natural language processing engine (e.g. Google Cloud Natural Language API)

[2006] Machine learning platforms (e.g. TensorFlow, PyTorch)

[2007] Web frameworks for the server (e.g., Flask, Django)

[2008] Prompt Sentence Examples

[2009] The prompt has the following format:

[2010] Towards a natural language processing engine

[2011] input:

[2012] Question: "What are the maintenance procedures for this machine?"

[2013] output:

[2014] Answer: "1. Turn off the power, 2. Remove the screws, 3. Clean the filter."

[2015] As described above, this invention makes it possible to dramatically improve work efficiency and safety by applying a knowledge base system to on-site factory work.

[2016] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[2017] Step 1:

[2018] Users access the system using smart glasses or a personal digital assistant and enter their questions.

[2019] Specifically, the user uses the voice input function of the smart glasses to say, "Please tell me the maintenance procedure for this machine."

[2020] Input: User question (voice, text)

[2021] Output: Question data (text format)

[2022] Step 2:

[2023] The terminal transmits the question data received from the user to the server.

[2024] Specifically, the operation is to convert the received voice input into text data and send the text data to the server.

[2025] Input: Question data (text format)

[2026] Output: Send a query to the server

[2027] Step 3:

[2028] The server analyzes the received question using a natural language processing engine.

[2029] Specifically, it uses the Google Cloud Natural Language API to extract the intent and keywords of the question.

[2030] Input: Question data (text format)

[2031] Output: Question analysis results (keywords, intent)

[2032] Step 4:

[2033] The server searches for relevant information from knowledge bases and information sources based on the analysis results.

[2034] Specifically, it queries the database for relevant maintenance procedures and work procedures.

[2035] Input: Question analysis results (keywords, intent)

[2036] Output: Related information (maintenance procedures, work procedures)

[2037] Step 5:

[2038] The server generates the best answer based on the search results.

[2039] Specifically, it uses a generative AI model to generate answers to users' questions in text format.

[2040] Input: Related information (maintenance procedures, work procedures)

[2041] Output: Best answer (text format)

[2042] Step 6:

[2043] The server sends the generated response to the terminal.

[2044] Specifically, the generated answer is sent to the user's smart glasses via a network.

[2045] Input: Best answer (text format)

[2046] Output: Sends the answer to the terminal

[2047] Step 7:

[2048] The terminal displays the response received from the server to the user.

[2049] A specific operation is to display maintenance procedures and work procedures on the display of the smart glasses.

[2050] Input: Response from the server (text format)

[2051] Output: Display to the user (maintenance procedures, work procedures)

[2052] Step 8:

[2053] The user proceeds with the work based on the displayed information.

[2054] Specifically, the machine maintenance is carried out according to the displayed maintenance procedure.

[2055] Input: Information to be displayed to users (maintenance procedures, work procedures)

[2056] Output: Actual work progress

[2057] 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.

[2058] This invention provides a method for building an in-house knowledge base system that employees can easily access and obtain the information they need. This system exchanges information between a server, terminals, and users, and analyzes, generates, and shares information using natural language processing technology, machine learning algorithms, and emotion engines.

[2059] System configuration

[2060] server

[2061] The server plays a central role in the system, receiving questions from users and searching for information from a knowledge base and related information sources. It also uses natural language processing technology to analyze the questions and generate optimal answers. It also has the ability to recognize the emotions contained in the user's questions using an emotion engine and adjust the analysis results. It also adds best practice information to the knowledge base and regularly updates it. The server provides the generated answers and learning content to the device, and analyzes the user's usage history to provide individually optimized learning content.

[2062] Terminal

[2063] The terminal is a device that the user operates directly, sends input from the user to the server, and displays information provided by the server. The user can enter questions through the terminal and view answers and learning content from the server.

[2064] User

[2065] Users are employees or members of an organization who access the system through terminals. They can enter questions, obtain information, share best practice information, and utilize the learning content provided for self-study.

[2066] Program processing

[2067] Information search function

[2068] When the server receives a question from a user, it analyzes the question using natural language processing technology. It then uses an emotion engine to recognize the emotion contained in the question and adjust the analysis results. It then searches for relevant information from a knowledge base and related sources to generate the optimal answer. The generated answer is sent to the device and displayed to the user.

[2069] Best practice sharing feature

[2070] When a user submits best practice information from their device, the server analyzes the information and adds it to the knowledge base. This information is then available to other users when they ask similar questions. The information may also be displayed in an appropriate tone depending on the user's emotions.

[2071] Self-learning and problem-solving features

[2072] The server collects knowledge from within the organization and updates it regularly. It also uses machine learning algorithms to generate learning content that improves users' problem-solving abilities. It provides individually optimized learning content based on the user's usage history. An emotion engine also collects information about the user's emotions and uses it to provide future content.

[2073] Specific examples

[2074] Specific examples of information search

[2075] The user types "Please tell me the latest salary policy" into the terminal and sends it. The server analyzes the question and recognizes the user's emotion using an emotion engine. For example, if the server recognizes that the user is feeling anxious, it generates a gentle response saying, "The latest salary policy was updated on January 1, 2023. Click here for details." The terminal displays this response to the user.

[2076] Examples of best practice sharing

[2077] The user types "It is important to apologize quickly when handling customer complaints" into the terminal and sends it. The server analyzes the information, recognizes the emotion of the received information using an emotion engine, and adds it to the knowledge base. When another user asks "How do I handle customer complaints?", the server provides relevant best practices in an appropriate tone. The terminal displays the answer to the user.

[2078] Specific examples of self-study

[2079] When a user requests, "I want to learn about the latest trends in sales strategies," the server generates the most appropriate learning content and sends it to the device in a tone adjusted according to the user's emotions. For example, if the user is feeling very interested, the server will provide the content, "Latest sales strategies: The impact of digital marketing and effective techniques. Click here for more information." The device displays this content, and the user can learn from it and apply it to their work.

[2080] As described above, the present invention provides a concrete means for enabling employees to quickly and easily access the information they need while taking into consideration the user's feelings, thereby improving the productivity and efficiency of the entire organization.

[2081] The processing flow will be explained below.

[2082] Program processing flow

[2083] Information search function

[2084] Step 1:

[2085] The user enters a question in the input field on the terminal and presses the send button.

[2086] Step 2:

[2087] The terminal sends the user's question to the server.

[2088] Step 3:

[2089] The server receives the question and passes it to a natural language processing engine.

[2090] Step 4:

[2091] The server analyzes the question using a natural language processing engine.

[2092] Step 5:

[2093] The server requests the emotion engine to analyze the emotion of the question.

[2094] Step 6:

[2095] The emotion engine analyzes the emotion of the question and returns the emotion information to the server.

[2096] Step 7:

[2097] The server takes the emotion information into consideration and searches for relevant information from a knowledge base and related information sources.

[2098] Step 8:

[2099] The server generates the best answer based on the search results, adjusting the tone according to the user's emotions.

[2100] Step 9:

[2101] The server generates a response and sends it to the terminal.

[2102] Step 10:

[2103] The terminal receives the response from the server and displays it to the user.

[2104] Best practice sharing feature

[2105] Step 1:

[2106] The user enters best practice information into the input field on the device and presses the send button.

[2107] Step 2:

[2108] The device sends the best practice information to the server.

[2109] Step 3:

[2110] The server receives the best practice information and passes it to the natural language processing engine.

[2111] Step 4:

[2112] The server analyzes the best practice information using a natural language processing engine.

[2113] Step 5:

[2114] The server requests the emotion engine to analyze the emotion of the received information.

[2115] Step 6:

[2116] The emotion engine analyzes the emotion of the information and returns the emotion information to the server.

[2117] Step 7:

[2118] The server considers the analysis results and emotional information and adds them to the knowledge base.

[2119] Step 8:

[2120] A confirmation message is sent to the device to notify that the server has been successfully added to the knowledge base.

[2121] Step 9:

[2122] The terminal receives a confirmation message from the server and displays it to the user.

[2123] Self-learning and problem-solving features

[2124] Step 1:

[2125] The server collects knowledge within the organization and updates it periodically.

[2126] Step 2:

[2127] The server uses machine learning algorithms to generate learning content.

[2128] Step 3:

[2129] A user sends a request from a terminal to access self-learning content.

[2130] Step 4:

[2131] The server analyzes the user's usage history and generates individually optimized learning content.

[2132] Step 5:

[2133] The server requests the emotion engine to analyze the user's emotions.

[2134] Step 6:

[2135] The emotion engine returns the analysis results to the server.

[2136] Step 7:

[2137] The server optimizes learning content with an appropriate tone based on emotional information and sends it to the device.

[2138] Step 8:

[2139] The terminal receives the learning content from the server and displays it to the user.

[2140] Information sharing and collection functions

[2141] Step 1:

[2142] The user enters a question in the input field on the terminal and presses the send button.

[2143] Step 2:

[2144] The terminal sends a question to the server.

[2145] Step 3:

[2146] The server receives the question and passes it to a natural language processing engine.

[2147] Step 4:

[2148] The server analyzes the question using a natural language processing engine.

[2149] Step 5:

[2150] The server requests the emotion engine to analyze the emotion of the question.

[2151] Step 6:

[2152] The emotion engine analyzes the emotion of the question and returns the emotion information to the server.

[2153] Step 7:

[2154] The server takes the emotion information into consideration to search for relevant information from a knowledge base and related information sources.

[2155] Step 8:

[2156] The server generates the best answer based on the search results and adds it to the knowledge base.

[2157] Step 9:

[2158] The server generates a response in a tone based on the emotion information and transmits it to the terminal.

[2159] Step 10:

[2160] The terminal receives the response from the server and displays it to the user.

[2161] This allows the system, combined with the emotion engine, to recognize the user's emotions and adjust the tone of its analysis results and responses to provide more appropriate information. The specific processing flow allows users to access the information they need quickly and easily, improving productivity and efficiency across the organization.

[2162] Example 2

[2163] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[2164] Conventional in-house knowledge base systems made it difficult for users to access the information they needed accurately and quickly. They also lacked appropriate feedback and learning content that took users' emotions into consideration. This led to problems such as reduced user satisfaction and work efficiency.

[2165] 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.

[2166] In this invention, the server includes means for receiving a question from a user, means for analyzing the content of the question using natural language processing technology, means for recognizing the emotion contained in the question and adjusting the analysis result, means for searching for related information from a knowledge base and information sources, means for generating an optimal answer according to the emotion based on the search results, and means for providing the generated answer to the user. This enables the user to quickly and accurately access the information they need and to receive appropriate feedback according to their emotion.

[2167] A "User" is a person or member of an organization who accesses the system to obtain or provide information.

[2168] A "server" is a central device that receives requests from users and analyzes, searches, generates, and provides information.

[2169] A "terminal" is a device that is directly operated by a user and that communicates with a server to send and receive information.

[2170] The "means for receiving questions" is a mechanism by which the server receives questions from users.

[2171] "Natural language processing technology" is a technology for analyzing questions entered by users, and involves grammatical and semantic analysis.

[2172] "Means for recognizing emotions" refers to technology that analyzes the emotions contained in a user's question and provides appropriate feedback.

[2173] A "knowledge base" is a database that stores past questions, best practices, and related information.

[2174] "Source" means a source of information, including external data or materials other than the Knowledge Base.

[2175] A "search method" is a mechanism for finding relevant information from knowledge bases and information sources.

[2176] "Answer generation means" refers to a technique for creating the most appropriate answer to a user's question based on search results.

[2177] "Best practices" are the optimal methods and techniques for improving business efficiency and results.

[2178] "Learning content" refers to content provided to help users improve their skills and solve problems.

[2179] A "machine learning algorithm" is a technology that analyzes large amounts of data, finds patterns, and makes predictions and classifications.

[2180] This invention provides a concrete means for realizing an in-house knowledge base system. The main components of the system include a server, terminals, and users, each of which plays a specific role. The details are explained below.

[2181] server

[2182] The server plays the central role in the system, receiving questions from users, analyzing them, and generating answers. The server uses the following main software technologies:

[2183] Natural language processing technology: We use "spaCy" and "BERT" to analyze user questions.

[2184] Emotion recognition technology: We use "Affectiva" to analyze the emotions contained in users' questions.

[2185] Knowledge base: A database such as MySQL or PostgreSQL is used to store best practices and related information.

[2186] Machine learning algorithms: "scikit-learn" and "TensorFlow" are used to generate learning content.

[2187] The server utilizes these technologies to generate appropriate answers to user questions, periodically updates the collected knowledge, and provides individually optimized learning content based on the user's usage history and emotional information.

[2188] For example, if a user asks, "What is the latest salary policy?", the server analyzes the question and uses emotion recognition technology to determine whether the user is feeling anxious. Based on the results, the server generates the most appropriate response and tone and sends the following response to the device: "The latest salary policy was updated on January 1, 2023. Click here for details."

[2189] Terminal

[2190] The terminal is a device operated by the user that sends and receives information to and from the server. Specifically, it has the following functions:

[2191] Submit user input: Enter a question or best practice information and submit it to the server.

[2192] Display information: Displays answers and learning content received from the server to the user.

[2193] For example, a user inputs "It is important to apologize promptly when handling customer complaints" into an input field on the terminal and sends it. The terminal transfers this information to the server, receives appropriate feedback from the server again, and displays it.

[2194] User

[2195] The user is the end user of the system and performs the following actions:

[2196] Entering a question: Requesting the required information from the system through the terminal.

[2197] Best practice information sharing: Providing useful information from the terminal to the system.

[2198] Utilizing learning content: Use the provided learning content for self-study and problem-solving.

[2199] For example, if a user types "I want to learn about the latest trends in sales strategies" into their device, the server will generate relevant learning content, adjust the tone according to their emotions, and provide information such as "The latest sales strategies: The impact of digital marketing and effective techniques. Click here for more information." Users can view this information and apply it to their work.

[2200] By combining these functions, it is possible to provide necessary information quickly and accurately while taking into consideration the user's feelings. Also, by providing individually optimized learning content, productivity and efficiency will be improved across the organization.

[2201] Prompt Sentence Examples

[2202] Use spaCy to parse the user question "What is the latest salary policy?" and process the sentiment with Affectiva to generate an answer with the appropriate tone.

[2203] The flow of the identification process in the second embodiment will be described with reference to FIG.

[2204] Information retrieval function processing steps

[2205] Step 1: User enters question

[2206] The user types a question into a field on the device, such as "What's the new vacation policy?"

[2207] Input: User question text

[2208] Output: The question you typed is displayed on the terminal.

[2209] Step 2: The device sends a question to the server

[2210] The device sends the entered question to the server as an API request.

[2211] Input: Question text entered by the user

[2212] Output: API request sent to the server

[2213] Step 3: The server parses the question

[2214] The server analyzes the received question using natural language processing technology (e.g., "spaCy") to identify the main topics, and through this analysis, understands the context and intent of the question.

[2215] Input: Question text sent to the server

[2216] Output: Analysis of the question (e.g. "vacation policy")

[2217] Step 4: The server recognizes the emotion

[2218] The server uses emotion recognition technology (e.g., "Affectiva") to analyze the emotions contained in the questions, and through this recognition, it understands the user's emotional state and adjusts the tone of its answers accordingly.

[2219] Input: Analysis result of question content

[2220] Output: Sentiment analysis result (e.g., "anxiety")

[2221] Step 5: The server searches the knowledge base for information

[2222] The server searches for relevant information from a knowledge base and other sources based on the analysis results and sentiment analysis results.

[2223] Input: Question content analysis results and sentiment analysis results

[2224] Output: Search results (e.g. "new vacation policy details")

[2225] Step 6: Server generates answer

[2226] The server generates a response in an appropriate tone based on the sentiment analysis results, such as "Our latest vacation policy was updated on January 1, 2023. Click here for more information."

[2227] Input: Search results and sentiment analysis results

[2228] Output: Generated answer text

[2229] Step 7: The server sends the answer to the device

[2230] The server sends the generated answer to the terminal as an API response.

[2231] Input: Generated answer text

[2232] Output: API response sent to the device

[2233] Step 8: The device displays the answer to the user

[2234] The terminal displays the received answer to the user, who can then view it to find the answer to his or her question.

[2235] Input: Answer text sent from the server

[2236] Output: Answer displayed on the terminal

[2237] Best Practice Sharing Feature Processing Steps

[2238] Step 1: User inputs best practices

[2239] The user inputs best practice information into an input field on the terminal. For example, the user might input "It is important to apologize promptly when handling customer complaints."

[2240] Input: User best practice information

[2241] Output: The entered best practices are displayed on the terminal.

[2242] Step 2: The device sends the information to the server

[2243] The device sends the entered best practice information to the server as an API request.

[2244] Input: Best practice information entered by the user

[2245] Output: API request sent to the server

[2246] Step 3: The server analyzes the information

[2247] The server analyzes the received best practice information using natural language processing technology (e.g., "BERT") to identify key points.

[2248] Input: Best practice information sent to the server

[2249] Output: Best practice analysis results

[2250] Step 4: The server recognizes the emotion

[2251] The server uses emotion recognition technology to analyze the emotions contained in the best practice information, allowing it to present the information to other users in an appropriate tone.

[2252] Input: Best practice analysis results

[2253] Output: Emotion analysis results

[2254] Step 5: The server adds the information to the knowledge base

[2255] Based on the results of the analysis and sentiment analysis, the server adds new best practice information to the knowledge base, which can be used by other users when they ask similar questions.

[2256] Input: Best practice content analysis results and sentiment analysis results

[2257] Output: Information added to the Knowledge Base

[2258] Step 6: The server provides best practices for other users' questions

[2259] When other users ask similar questions, the server retrieves additional best practice information from the knowledge base and delivers it in the appropriate tone.

[2260] Input: Question from another user

[2261] Output: Best practice information provided

[2262] Self-learning and problem-solving processing steps

[2263] Step 1: User requests learning content

[2264] The user inputs "I want to learn the latest sales strategies" into the terminal and sends a request.

[2265] Input: User's learning content request

[2266] Output: The input request is displayed on the terminal.

[2267] Step 2: The device sends a request to the server

[2268] The device sends this request to the server as an API request.

[2269] Input: Learning content request entered by the user

[2270] Output: API request sent to the server

[2271] Step 3: Server checks the knowledge base

[2272] The server retrieves relevant learning material from a knowledge base.

[2273] Input: Learning content request sent to the server

[2274] Output: Search results (related learning materials)

[2275] Step 4: Server Generates Content

[2276] The server uses machine learning algorithms to generate learning content that best suits the request (for example, "scikit-learn" or "TensorFlow").

[2277] Input: Search results

[2278] Output: Generated learning content

[2279] Step 5: The server adjusts the tone based on the emotion

[2280] The server uses emotion recognition technology to tailor the learning content to a tone that corresponds to the user's emotions.

[2281] Input: Generated learning content and sentiment analysis results

[2282] Output: Tailored learning content

[2283] Step 6: The server sends the learning content to the device

[2284] The server sends the adjusted learning content to the device as an API response.

[2285] Input: Tailored learning content

[2286] Output: API response sent to the device

[2287] Step 7: The device displays the content to the user

[2288] The device displays the received learning content to the user, who can then view it and apply it to their work.

[2289] Input: Learning content sent from the server

[2290] Output: Learning content displayed on the device

[2291] (Application example 2)

[2292] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[2293] While conventional in-house knowledge base systems allow users to quickly and easily access the information they need, they lack the ability to provide appropriate answers based on the user's emotions, making it difficult to respond appropriately, especially under stressful situations.In addition, for robots used on factory floors, there was a need for a means to provide optimal information in real time during troubleshooting and maintenance work.

[2294] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving a question from a user, means for analyzing the content of the question using natural language processing technology, means for searching for related information from a knowledge base and an information source, means for generating an optimal answer based on the search results, means for providing the generated answer to the user, emotion analysis means for recognizing the user's emotion and adjusting the analysis result, and means for displaying the generated answer via the robot's output device. This enables the factory robot to provide information in an appropriate tone according to the user's emotion and to support troubleshooting and maintenance work in real time.

[2295] A "user" is an individual or member of an organization who utilizes the system to enter questions and receive answers.

[2296] A "server" is a computer system that receives and analyzes questions from users, searches for relevant information from knowledge bases and information sources, and generates and provides optimal answers.

[2297] A "terminal" is a device that a user directly operates and connects to a server to input questions and receive answers.

[2298] A "knowledge base" is a database that stores knowledge and data collected from various sources.

[2299] "Natural language processing technology" is a technology that allows computers to understand and process human language.

[2300] A "machine learning algorithm" is an algorithm that finds patterns based on large amounts of data and automatically learns and predicts.

[2301] The "emotion engine" is a technology that recognizes emotions from user input and adjusts the tone of the response based on that information.

[2302] A "source" is a location where external or internal data or information exists outside of the knowledge base.

[2303] "Search Results" refers to relevant information retrieved from knowledge bases and information sources based on a user's question.

[2304] "Best practice information" is information that summarizes methods or procedures that are known to be most effective in a particular situation.

[2305] "Learning content" is educational content generated by the server to improve the user's problem-solving ability.

[2306] "Troubleshooting" is the process of identifying and resolving problems or errors in machines or systems.

[2307] "Maintenance work" refers to maintenance work carried out periodically or as needed to ensure stable operation of machines and systems.

[2308] "Sentiment analysis" is the process of detecting emotions from user input data and adapting that information.

[2309] An "output device" is a device for displaying generated information and answers to the user.

[2310] This invention provides an in-house knowledge base system that allows employees to easily access and obtain the information they need. The system receives questions from users, analyzes the questions using natural language processing technology, and searches for information from the knowledge base and related information sources to generate and provide optimal answers. It also includes a function that recognizes user emotions using an emotion engine and adjusts the analysis results. Furthermore, this system is implemented in factory robots to support troubleshooting and maintenance work.

[2311] System configuration and functions

[2312] server

[2313] The server receives questions from users and analyzes the content of the questions using natural language processing technology. Based on the analyzed question, it searches for relevant information from a knowledge base and information sources and generates the optimal answer. In doing so, it analyzes the user's emotions using an emotion engine and adjusts the analysis results. The server also has the function of receiving best practice information from users and adding it to the knowledge base. Furthermore, it uses machine learning algorithms to generate and provide learning content to improve the user's problem-solving ability.

[2314] Terminal

[2315] The terminal is a device that is directly operated by the user, allowing them to input questions, send them to the server, and display the information provided by the server. The terminal also functions as an output device for factory robots, displaying real-time information necessary for troubleshooting and maintenance work.

[2316] User

[2317] Users can use the system to enter questions and obtain the necessary information. For example, when a problem occurs at a factory, users can enter a question through their terminal and obtain the most appropriate answer from the server. Users can also provide their own best practice information to the system and share it with other users.

[2318] Hardware and software used

[2319] This system mainly uses the following hardware and software:

[2320] Hardware: Server equipment, user devices (PCs, tablets, smartphones, etc.), factory robots, output devices (displays)

[2321] Software: Natural language processing techniques (e.g., Hugging Face transformers), machine learning algorithms, sentiment analysis engines

[2322] Data processing and calculation

[2323] The server analyzes the user's question using natural language processing technology, then analyzes the user's emotions using an emotion engine. Based on the analysis results, it searches for relevant information from a knowledge base and information sources to generate the optimal answer. The tone of this answer is adjusted according to the user's emotions and sent to the device.

[2324] Specific examples

[2325] Specific examples of information search

[2326] The user types "What should you do if this machine stops?" into the terminal and sends it. The server analyzes the question and recognizes the user's emotions using an emotion engine. For example, if the server recognizes that the user is feeling anxious, it generates a response in a gentle tone: "If this machine stops,...

Claims

1. means for receiving a query from a user; A means for analyzing the content of the question using natural language processing technology; a means of searching for relevant information from knowledge bases and sources; A means of generating optimal answers based on search results; means for providing the generated answer to the user; A system including:

2. means for receiving best practice information from a user; a means of analyzing and adding received best practice information to a knowledge base; Includes a means to provide best practice information that is added to the knowledge base in response to other users' questions; The system of claim 1 .

3. A means of collecting and regularly updating knowledge within the organization; A means for generating learning content for improving problem-solving ability using a machine learning algorithm; including a means for providing personalized learning content based on a user's usage history; The system of claim 1 .

Images