System

Abstract

A system is provided.SOLUTION: A system comprising: means for periodically collecting mails and documents generated in a company; means for pre-processing the collected mails and documents and extracting business-related information; means for analyzing the pre-processed business-related information, extracting know-how, and storing the know-how in a database; means for receiving a question from a user through a user interface; means for analyzing the received question and searching for related information from the know-how database; means for generating a response to the question using a generative artificial intelligence model based on a search result; and means for returning the generated response to the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] As senior employees approach retirement, companies face the challenge of making it difficult to pass on the wealth of know-how they have accumulated over the years. If a successor has not yet been assigned or if younger employees lack business knowledge, it can be difficult to quickly acquire the necessary know-how. This can result in a lack of smooth transfer of know-how from senior employees, leading to a decline in business efficiency. [Means for solving the problem]

[0005] The present invention provides a means for periodically collecting emails and documents generated within a company and preprocessing them to extract business-related information. This business-related information is analyzed using a text analysis tool, and know-how is extracted and stored in a database. The system also includes a means for accepting questions from users through a user interface, analyzing the accepted questions, and searching for related information from a know-how database. Based on the search results, a generative artificial intelligence model is used to generate a response to the question, and the generated response is returned to the user. This makes it possible to effectively manage knowledge assets within a company and create an environment in which young employees can quickly acquire know-how.

[0006] "Means for periodically collecting emails and documents generated within a company" refers to a method for automatically obtaining new emails and documents from a mail server or file server within a company at specific intervals.

[0007] The "means for preprocessing collected emails and documents to extract business-related information" refers to a method for filtering out spam and unimportant information from the collected emails and documents and extracting useful business-related information.

[0008] "Means for analyzing preprocessed business-related information, extracting know-how, and storing it in a database" refers to a method for analyzing preprocessed information using natural language processing technology and text analysis tools, identifying data that is useful as corporate know-how, and systematically storing this in a database.

[0009] The "means for accepting questions from users through a user interface" refers to a method for providing a dedicated interface through which users can input questions in natural language, and accepting questions through this interface.

[0010] The "means for analyzing a received question and searching for related information from a know-how database" is a method for analyzing the content of a question received from a user and searching for and extracting related information from a know-how database.

[0011] "Means for generating responses to questions using a generative artificial intelligence model based on search results" refers to a method for generating appropriate responses to questions using a generative artificial intelligence model (e.g., ChatGPT (registered trademark)) based on the searched information.

[0012] The "means for returning the generated response to the user" is a technique for providing the generated response to the user through a user interface and returning an answer to the question.

[0013] A "text analysis tool" is a tool that uses natural language processing technology to analyze the content of text data and extract important keywords and phrases.

[0014] A "generative artificial intelligence model" is an artificial intelligence technology that generates natural, human-like responses based on given data and context. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6]FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0016] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0018] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0019] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0020] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0021] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0022] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0023] [First embodiment]

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

[0025] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0026] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0027] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0028] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0029] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0030] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0032] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0033] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0034] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0035] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0036] MODE FOR CARRYING OUT THE INVENTION

[0037] The present invention provides a system for effectively passing on the know-how of senior employees to junior employees. The details of this system are described below.

[0038] Data collection

[0039] First, the server periodically accesses the company's mail server and file server to retrieve new emails and documents. During this process, it connects to the mail server using the IMAP or POP3 protocol to retrieve unread emails, and to the file server using the FTP or SMB protocol to retrieve new documents from specified folders.

[0040] Data Preprocessing

[0041] The server then preprocesses the emails and documents it receives. Specifically, it filters out spam and unimportant information, and analyzes only business-related information. This preprocessing involves using regular expressions and natural language processing techniques to remove unnecessary information.

[0042] Data analysis and know-how extraction

[0043] The server uses text analysis tools to analyze the preprocessed data. For example, it uses techniques such as TF-IDF and Word2Vec to analyze the content of emails and documents and extract important keywords and phrases. Based on the results of this analysis, it identifies know-how data and stores it in a database.

[0044] Accepting and analyzing questions

[0045] A user inputs a question through a dedicated interface (e.g., a chatbot). When the user inputs a question from a terminal, the question is sent from the terminal to a server. The server analyzes the received question using natural language processing tools to identify the intent and keywords of the question.

[0046] Know-how search and response generation

[0047] Based on the analyzed question, the server searches for relevant information from a know-how database. Based on the search results, the server generates a response to the question using a generative artificial intelligence model such as ChatGPT. For example, if a junior employee asks, "Please tell me how to proceed with a new project," the server will search for know-how information about past projects and use that information to generate a specific approach using ChatGPT.

[0048] Providing a response

[0049] The generated response is sent from the server to the user via the terminal, and the user can receive specific and practical answers through the chatbot interface.

[0050] Specific examples

[0051] For example, consider a case where a junior employee types a question such as, "Please tell me your strategy for entering a new market." The device receives this question and sends it to the server. The server analyzes the question and searches for information on relevant market entry strategies from its know-how database. It then uses ChatGPT to generate responses on specific market research methods, competitive analysis, and promotion strategies, and sends them back to the user via the device.

[0052] This system prevents the loss of know-how when senior employees retire and allows younger employees to easily acquire business knowledge, thereby effectively utilizing knowledge assets within the company and improving business efficiency.

[0053] The processing flow will be explained below.

[0054] Step 1:

[0055] The server periodically accesses the company's mail server and file server, using the IMAP or POP3 protocol to retrieve unread emails from the mail server, and the FTP or SMB protocol to connect to the file server and retrieve new documents from specified folders.

[0056] Step 2:

[0057] The server pre-processes emails and documents retrieved by the server, specifically filtering out spam, personal information not related to business, and unnecessary data, using regular expressions and natural language processing techniques.

[0058] Step 3:

[0059] The server analyzes the preprocessed email and document data using text analysis tools, such as TF-IDF (Term Frequency-Inverse Document Frequency) and Word2Vec, to extract important keywords and phrases.

[0060] Step 4:

[0061] Based on the analysis results, the server organizes the company's know-how and stores it in a database as structured data. The know-how information stored also includes related metadata (creation date and time, creator, related projects, etc.).

[0062] Step 5:

[0063] The user inputs a question through a dedicated user interface (e.g., a chatbot). The terminal receives the user's input and sends the question to the server.

[0064] Step 6:

[0065] The server analyzes the received question using natural language processing tools, extracting the intent of the question and key keywords, and providing the analysis results.

[0066] Step 7:

[0067] The server searches for relevant information based on keywords extracted from the know-how database and the intent of the question, and identifies the necessary data based on similar questions from the past and related documents.

[0068] Step 8:

[0069] The server generates a response to the question based on the search results using a generative artificial intelligence model (e.g., ChatGPT). Specifically, it refers to the searched know-how and constructs an appropriate answer in natural language.

[0070] Step 9:

[0071] The server returns the generated response to the user through the user interface, and the user can receive a specific and practical answer through the interface on the terminal.

[0072] Example 1

[0073] 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."

[0074] The transfer of knowledge and know-how within a company has become a major issue due to the retirement or transfer of senior employees. In particular, there is a huge amount of information accumulated in emails and documents, and there is a need for a method to quickly extract the necessary information and efficiently convey it to younger employees. This has led to a demand for a system that can improve work productivity and enable the systematic transfer of knowledge.

[0075] 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.

[0076] In this invention, the server includes means for periodically collecting messages and documents generated within the company, means for preprocessing the collected messages and documents to filter out unnecessary information and extract only business-related information, and means for analyzing the preprocessed business-related information, extracting important keywords and phrases, and storing them in a database. This enables the system to efficiently collect and analyze the know-how of senior employees and provide necessary information to junior employees in a timely manner.

[0077] "Messages" are digital communications such as emails and instant messages sent and received within a company.

[0078] "Documents" are digital or paper text data such as reports, memos, plans, etc. generated within a company.

[0079] "Collection" is the act of periodically acquiring specified digital data from a company's network or file server.

[0080] "Preprocessing" is the process of filtering unnecessary information from collected digital data and extracting only the data necessary for analysis.

[0081] "Filtering" is the process of selecting necessary information from digital data based on certain criteria and removing unnecessary information.

[0082] "Business-related information" refers to digital data such as knowledge, know-how, and procedures necessary for a company to carry out its business.

[0083] "Analysis" is the process of analyzing collected and pre-processed digital data to identify important keywords and phrases.

[0084] A "keyword" is a word or phrase that has particular meaning or significance within digital data.

[0085] A "phrase" is a short sentence consisting of one or more words that express a specific meaning.

[0086] A "database" is an information system that systematically stores and manages collected and analyzed digital data.

[0087] A "user interface" is an interaction means through which a user operates or inputs information into a system.

[0088] A "question" is an inquiry that a user inputs to the system to request specific information.

[0089] "Natural language processing tools" are software and algorithms used to understand and analyze human language.

[0090] A "generative artificial intelligence model" is an artificial intelligence algorithm that can generate new information from previously learned data.

[0091] A "response" is an answer that the system provides to a question from a user.

[0092] MODE FOR CARRYING OUT THE INVENTION

[0093] The present invention relates to a system for effectively transferring knowledge and know-how within a company. In this system, a server collects messages and documents within the company, performs preprocessing and analysis, and generates responses based on the collected data using a generative artificial intelligence model.

[0094] First, the server periodically connects to the company's mail server and file server to collect new messages and documents. It connects to the mail server using the IMAP or POP3 protocol to retrieve unread messages, and connects to the file server using the FTP or SMB protocol to retrieve new documents from specified folders. This collection process continuously collects the company's digital assets.

[0095] The server then pre-processes the collected messages and documents, using regular expressions and natural language processing techniques to filter out spam and other irrelevant information, leaving the pre-processed data with business-relevant information extracted.

[0096] The server then performs text analysis on the preprocessed data. For example, it uses TF-IDF technology to extract important keywords from the data and Word2Vec technology to cluster sentences and phrases. This allows the company's know-how to be systematically stored in a database.

[0097] When a user inputs a question through a dedicated interface (e.g., a chatbot), the device sends the question to a server. The server uses natural language processing tools to analyze the intent and keywords of the question, and then searches a know-how database for relevant information based on that analysis. It then uses a generative artificial intelligence model (e.g., ChatGPT) to generate a specific response to the question. This generated response is then returned to the user via the device.

[0098] For example, if a junior employee inputs a question such as, "What should I pay attention to when planning a new product?", the terminal sends this question to the server, which analyzes the question and then searches for related information in the know-how database. A generative AI model is used to generate a specific response such as, "When planning a new product, it is important to start with market research and then conduct a competitive analysis." This response is then sent back to the user via the terminal.

[0099] This invention allows the know-how of senior employees to be collected and analyzed effectively, and necessary information to be provided to younger employees in a timely manner, thereby contributing to improved work efficiency.

[0100] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0101] Step 1: Data collection

[0102] The server periodically connects to the company's mail server and file server to collect new messages and documents. For input, it uses the IMAP or POP3 protocol for mail servers and the FTP or SMB protocol for file servers. For output, it obtains unread messages and new document data. For example, the server connects to "imap.example.com" using the IMAP protocol to obtain unread emails.

[0103] Step 2: Data Preprocessing

[0104] The server pre-processes the collected messages and documents. It uses the collected message and document data as input and outputs business-relevant information that has been spam filtered and filtered out using regular expressions. For example, the server uses regular expressions to filter out emails containing the word "SPAM" and remove unimportant information.

[0105] Step 3: Data analysis

[0106] The server analyzes the preprocessed business-related information. The preprocessed business-related information is provided as input, and important keywords and phrases are extracted and clustered using TF-IDF and Word2Vec techniques. As output, the extracted important keywords and phrases are stored in a database. For example, the server uses the "TF-IDF" library to identify high-frequency keywords from email content.

[0107] Step 4: Ask a question

[0108] A user inputs a question through a dedicated interface (e.g., a chatbot). The question text from the user is entered into the terminal as input and sent to the server via the terminal as output. For example, a user might input into the terminal, "What points should I pay attention to when planning a new product?"

[0109] Step 5: Parsing the Question

[0110] The server analyzes the received questions using natural language processing tools. The question text sent to the server is used as input, and the intent and keywords of the question are identified as output. For example, the server uses the "spaCy" library to extract key keywords from the question text.

[0111] Step 6: Know-how search and response generation

[0112] The server searches for relevant information from a know-how database based on the analyzed question content and generates a response to the question using a generative artificial intelligence model. The intent of the question and keywords are used as input, and a specific response text is generated as output. For example, the server uses ChatGPT to generate a response such as, "When planning a new product, it is important to start with market research and conduct a competitive analysis."

[0113] Step 7: Providing a response

[0114] The server sends the generated response back to the user via the terminal. The generated response text is used as input and displayed on the user's interface as output. For example, the terminal receives the response text from the server and displays it on the chatbot screen.

[0115] Through these steps, the system can effectively collect and analyze know-how within the company and provide appropriate information to young employees.

[0116] (Application example 1)

[0117] 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."

[0118] The knowledge and experience of senior employees within a company are valuable assets, but the loss of know-how due to retirement and other reasons is an issue. Furthermore, there is a lack of systems that allow younger employees to quickly and effectively acquire that know-how. Factories, in particular, need systems that can provide immediate and appropriate instructions and advice. Unless this issue is resolved, the company's competitiveness may decline.

[0119] 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.

[0120] In this invention, the server includes: means for periodically collecting emails and documents generated within the company; means for preprocessing the collected emails and documents to extract business-related information; means for analyzing the preprocessed business-related information, extracting experience data, and storing the extracted experience data in a database; means for accepting questions from users through a user interface; means for analyzing the accepted questions and searching for related information from an experience database; means for generating responses to the questions using a generative artificial intelligence model based on the search results; means for returning the generated responses to the users; and means for being installed in a robot installed in a factory to support young engineers and operators in sharing the know-how of senior engineers. This makes it possible to effectively pass on the know-how of senior employees to junior employees and to provide immediate and appropriate instructions and advice, particularly on the factory floor.

[0121] "Intra-company" refers to the inside of a particular company or organization, and is the scope related to the operation and business of that organization.

[0122] "Mail" refers to electronic mail, a message containing text and attachments sent and received over the Internet.

[0123] "Documents" refers to any paper or electronic document created, stored, and shared within an organization, including reports, presentations, memoranda, etc.

[0124] "Preprocessing" refers to the process of removing unnecessary information and reformatting the format of raw data to make it easier to analyze.

[0125] "Business-related information" refers to information that is directly related to a company's business operations and decision-making, including project progress reports and performance reports.

[0126] "Analysis" is the process of analyzing data in detail and finding useful information and patterns within it.

[0127] "Experience" refers to the knowledge, skills, and know-how acquired in the past, which are required to carry out specific tasks or work.

[0128] A "database" is a system that stores and manages information in an orderly manner, making it easy to search and use.

[0129] "User interface" refers to the means and screen display through which a user interacts with a system or application.

[0130] A "generative artificial intelligence model" refers to artificial intelligence technology that automatically generates responses and sentences in response to input data or questions, and includes natural language processing technology.

[0131] A "factory" is a facility where products are manufactured or assembled, and where industrial machinery and production lines are installed.

[0132] A "robot" is an automated machine or device designed to assist or replace human tasks.

[0133] "Young engineers" refer to relatively new employees within a company who are honing their expertise and skills in technical fields.

[0134] "Operator" refers to a staff member responsible for operating and managing a machine or system.

[0135] "Senior engineers" refer to technical leaders and veteran employees within a company who have extensive experience and advanced expertise.

[0136] The present invention provides a system for efficiently passing on the know-how of senior employees to junior employees in a company. This system is composed of the following means.

[0137] First, the server periodically collects emails and documents generated within the company. Emails are collected using the IMAP or POP3 protocol to retrieve unread emails, and documents are collected using the FTP or SMB protocol to retrieve new documents from specified folders.

[0138] The server then pre-processes the collected emails and documents to extract business-related information. This pre-processing removes spam and unimportant information, leaving only information relevant to the company's business. This process involves filtering out unnecessary information using regular expressions and natural language processing techniques.

[0139] The preprocessed business-related information is then analyzed by the server to extract experiences. Text analysis tools such as TF-IDF (Term Frequency-Inverse Document Frequency) and Word2Vec are used for the analysis. These tools are used to analyze the content of emails and documents and extract important keywords and phrases. Based on the results of this analysis, the experience data is stored in a database.

[0140] Users input questions through a dedicated interface (for example, through a chatbot or a robot installed in a factory). When a user inputs a question from a terminal, the question is sent from the terminal to a server. The server analyzes the received question using natural language processing tools to identify the intent of the question and important keywords.

[0141] Based on the analyzed question, the server searches for relevant information from an experience database. Based on the search results, the server generates a response to the question using a generative AI model (e.g., ChatGPT). For example, if a junior engineer asks, "Please tell me how to proceed with a new project," the server searches for experience information about past projects and uses the generative AI model to generate a specific approach based on that information.

[0142] The generated responses are sent from the server to the user via the terminal, allowing the user to receive specific and practical answers through the interface. Particularly in factory settings, robots can provide young engineers and operators with appropriate instructions and advice in real time.

[0143] As a concrete example, the prompt sentence is shown below.

[0144] Example prompt:

[0145] question: "How do I proceed with a new project?"

[0146] context: "Search for know-how information from past projects and explain the specific steps you took based on that information. Include market research methods, competitive analysis, and promotion strategies."

[0147] This system allows senior employees to effectively pass on their know-how to younger employees, and makes it possible to provide immediate and appropriate instructions and advice, particularly on factory floors, thereby preserving knowledge assets within the company and improving operational efficiency.

[0148] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0149] Step 1:

[0150] The server periodically accesses the company's mail server and file server to retrieve new emails and documents. The input here is the configuration information for the mail server and file server, and the output is the newly retrieved emails and documents. Specifically, it retrieves unread emails from the mail server using the IMAP or POP3 protocol, and retrieves new documents from the file server using the FTP or SMB protocol.

[0151] Step 2:

[0152] The server preprocesses the emails and documents it receives. The input is the emails and documents received in step 1, and the output is refined data that contains only business-related information. Specifically, regular expressions are used to remove spam and unimportant information, and natural language processing techniques are used to extract business-related information.

[0153] Step 3:

[0154] The server analyzes the preprocessed business-related information, extracts experiences, and stores them in a database. The input is the result data from Step 2, and the output is experience data analyzed using TF-IDF and Word2Vec. Specifically, the TF-IDF vectorizer is used to calculate feature weights, and the Word2Vec model is used to learn the relationships between words in the data, thereby extracting important experience data and storing it in a database.

[0155] Step 4:

[0156] The user inputs a question through a dedicated interface. The input here is the user's question, and the output is the transmission of that question. Specifically, the user's question is received using a chatbot interface or a robot's voice input system.

[0157] Step 5:

[0158] The server analyzes the received question and searches the experience database for relevant information. The input is the user's question, and the output is the analyzed keywords and important phrases of the question, as well as related experience data. Specifically, it uses natural language processing tools to extract the intent and keywords of the question, and then searches the experience database based on them.

[0159] Step 6:

[0160] Based on the search results, the server uses a generative AI model to generate a response to the question. The input is the relevant experience data obtained in step 5, and the output is the generated response text. Specifically, a generative AI model such as ChatGPT is used to generate a specific response based on relevant information. For example, in response to a question such as "Please tell me how to proceed with a new project," the server incorporates experience data from past projects and suggests a specific approach.

[0161] Step 7:

[0162] The server returns the generated response to the user via the terminal. The input is the response generated in step 6, and the output is the response provided to the user. Specifically, the generated response is delivered to the user via the chatbot screen or the robot's voice output. This series of processes allows junior engineers and operators to share the know-how of senior engineers in real time.

[0163] 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.

[0164] MODE FOR CARRYING OUT THE INVENTION

[0165] The present invention provides a system for effectively passing on the know-how of senior employees to younger employees, and by incorporating a user emotion recognition function, it achieves more personalized responses.

[0166] Data collection

[0167] The server periodically accesses the company's mail and file servers to retrieve new emails and documents, using the IMAP or POP3 protocols for mail servers and the FTP or SMB protocols for file servers.

[0168] Data Preprocessing

[0169] The server pre-processes the emails and documents it receives. During this process, it filters out spam and irrelevant information and extracts only business-related information. Regular expressions and natural language processing techniques are used for pre-processing.

[0170] Data analysis and know-how extraction

[0171] The server then uses text analysis tools to analyze the preprocessed data and extract important keywords and phrases. The extracted information is then used to organize the company's know-how and store it in a database. Metadata (creation date and time, creator, related projects, etc.) is also stored.

[0172] Accepting and analyzing questions

[0173] The user inputs a question through a dedicated user interface (e.g., a chatbot). The device receives the user's question and sends it to a server. The server analyzes the question using natural language processing tools and extracts key keywords and intent.

[0174] Emotion recognition and know-how retrieval

[0175] The server uses an emotion engine to recognize the user's emotional state when analyzing the question. The emotion engine recognizes emotions by analyzing the user's input and other emotional indicators. Based on this emotional information, it searches for related information from a know-how database.

[0176] Response Generation

[0177] The server generates a response to the question using a generative AI model (e.g., ChatGPT) based on the emotion recognition results and know-how search results. By taking the emotion recognition results into account and adjusting the response content, it provides a more personalized answer.

[0178] Providing a response

[0179] The server returns the generated response to the user through the user interface, allowing the user to instantly receive a specific and actionable answer through the interface on their device.

[0180] Specific examples

[0181] For example, consider the case where a junior employee types in, "Please tell me about our strategy for entering a new market." The device receives this question and sends it to the server. The server analyzes the question and uses an emotion engine to recognize the user's emotions, such as nervousness or confusion. Based on this information, it searches a know-how database for information about market entry and uses a generative artificial intelligence model to generate a response regarding specific market research methods, competitive analysis, and promotion strategies. The generated response is then sent back to the user via the device.

[0182] This system enables senior employees to pass on their know-how to junior employees in an efficient and personalized manner. The addition of emotion recognition functionality enables responses that take into account the user's emotional state, resulting in more effective problem-solving.

[0183] The processing flow will be explained below.

[0184] Step 1:

[0185] The server periodically accesses the company's mail server and file server, using the IMAP or POP3 protocol to retrieve unread emails from the mail server, and the FTP or SMB protocol to connect to the file server and retrieve new documents from specified folders.

[0186] Step 2:

[0187] The server pre-processes the emails and documents it receives. Specifically, it filters out spam and irrelevant information and extracts only business-related information. This pre-processing uses regular expressions and natural language processing techniques.

[0188] Step 3:

[0189] The server analyzes the preprocessed email and document data using text analysis tools, such as TF-IDF (Term Frequency-Inverse Document Frequency) and Word2Vec, to extract important keywords and phrases.

[0190] Step 4:

[0191] Based on the analysis results, the server organizes the company's know-how and stores it as structured data in a know-how database. The stored know-how information also includes related metadata (creation date and time, creator, related projects, etc.).

[0192] Step 5:

[0193] The user inputs a question through a dedicated user interface (e.g., a chatbot). The terminal receives the user's input and sends the question to the server.

[0194] Step 6:

[0195] The server analyzes the received question using natural language processing tools, extracting the intent of the question and key keywords, and providing the analysis results.

[0196] Step 7:

[0197] The server uses an emotion engine to analyze the emotion contained in the user's question. The emotion engine uses natural language processing technology and machine learning algorithms to recognize emotions from the user's input. For example, it analyzes whether the user is confused or nervous based on the question.

[0198] Step 8:

[0199] The server searches for relevant information based on keywords extracted from the know-how database and the intent of the question, and identifies the necessary data based on similar questions from the past and related documents.

[0200] Step 9:

[0201] The server generates a response to the question using a generative AI model (e.g., ChatGPT) based on the search results and the user's emotion recognition results. By taking the emotion recognition results into account and adjusting the response content, it generates the optimal answer for the user.

[0202] Step 10:

[0203] The server returns the generated response to the user through the user interface, and the user can receive a specific and practical answer through the interface on the terminal.

[0204] Specific examples

[0205] For example, consider the case where a junior employee types, "Please tell me about our strategy for entering a new market." The device receives this question and sends it to the server. The server analyzes the question and uses an emotion engine to recognize that the user is confused. Based on this information, it searches a know-how database for information about market entry and uses a generative artificial intelligence model to generate a response on specific market research methods, competitive analysis, and promotion strategies. The generated response is then sent back to the user via the device. In this way, a response that takes the user's emotional state into account is provided, enabling more effective problem solving.

[0206] Example 2

[0207] 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."

[0208] Efficiently transferring the know-how of senior employees to junior employees is an important issue for many companies. However, many current systems lack emotion recognition capabilities, respond uniformly, and are unable to provide personalized responses according to the emotional state of each user, making it difficult to achieve effective problem solving.

[0209] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes: means for periodically collecting communications and documents generated within the company; means for preprocessing the collected communications and documents to extract business-related information; means for analyzing the preprocessed business-related information, extracting knowledge, and storing it in a database; means for accepting inquiries from users through a user interface; means for analyzing the accepted inquiries and searching for related information from a knowledge database; means for generating a response to the inquiry using a generation system based on the search results; means for returning the generated response to the user; and means for having an emotion engine that recognizes the user's emotional state and adjusting the response content based on the recognized emotion information. This enables senior employees to pass on their know-how to junior employees in an efficient and personalized manner, and enables responses that take the user's emotional state into consideration, thereby achieving more effective problem solving.

[0210] "Communications" refers to the means of information exchange generated within a company, such as emails and chat messages.

[0211] "Document" refers to written or electronic records of information generated within an enterprise, such as reports, memos, and presentations.

[0212] "Business-related information" refers to information and data related to a company's business activities, including information on project management, risk management, and market research.

[0213] "Knowledge" refers to useful information and know-how extracted from business-related information and used to solve specific business problems.

[0214] A "database" refers to a collection of information that systematically organizes knowledge and makes it possible to store and search it.

[0215] "User interface" refers to the means by which a user interacts with a system, including graphical user interfaces (GUIs) and chatbots.

[0216] A "generative system" refers to a system that automatically generates responses to user inquiries using natural language processing technology or generative AI models.

[0217] An "emotion engine" refers to a function that analyzes a user's input and other emotional indicators to recognize their emotional state.

[0218] "Preprocessing" refers to a series of processes that filter spam and irrelevant information from collected communications and documents and extract only business-related information.

[0219] "Analysis" refers to the process of analyzing pre-processed business-related information in detail and extracting important keywords and phrases.

[0220] The present invention provides a system that effectively transfers the know-how of senior employees to junior employees while incorporating a function for recognizing user emotions to provide more personalized responses. A specific implementation method of this system will be described below.

[0221] First, the server, which is the core of this system, periodically collects communications and documents generated within the company. The server uses IMAP (Internet Message Access Protocol) and POP3 (Post Office Protocol 3) protocols to retrieve emails, and FTP (File Transfer Protocol) and SMB (Server Message Block) protocols to communicate with the file server. This process is performed automatically and periodically using CRON job settings.

[0222] The collected communications and documents are pre-processed by the server. In this pre-processing step, spam emails and irrelevant information are filtered out, and only business-related information is extracted. Regular expressions and natural language processing techniques (e.g., NLTK, SpaCy) are used in this process. For example, only emails containing keywords such as "important notice" or "weekly report" are extracted.

[0223] The preprocessed data is then subjected to detailed analysis using data analysis tools. The server tokenizes the information and performs morphological analysis to extract frequently occurring keywords and phrases. Important phrases, such as "project management" and "risk management," are automatically tagged and this information is stored in a knowledge database. Additionally, metadata such as creation date, creator, and related projects are also stored.

[0224] When a user inputs a question through a dedicated user interface (e.g., a chatbot), the device receives the question and sends it to a server. The server then analyzes the received question using natural language processing technology to extract key keywords and intent. Morphological analysis and parse tree analysis are used to analyze the user's text.

[0225] Furthermore, the server uses an emotion engine (e.g., EmotionAPI) to recognize the user's emotional state. The emotion engine recognizes emotions by analyzing the user's input and facial expression recognition using facial recognition technology. For example, if a user inputs "I'm very troubled," the emotion engine will recognize emotions such as "confusion" and "tension." Based on this recognition result, it searches for related information from a knowledge database.

[0226] Finally, the server generates a response using a generative AI model (e.g., ChatGPT) based on the emotional information and search results. The server inputs a prompt into the model, such as "Please explain your strategy for entering a new market to a nervous junior employee," to generate an appropriate response. The generated response is then sent back to the user via their device.

[0227] Specific examples

[0228] For example, consider the case where a junior employee types, "Please tell me about our strategy for entering a new market." The device receives this question and sends it to the server. The server analyzes the question and uses an emotion engine to recognize the user's emotions, such as nervousness or confusion. Based on this information, it searches a know-how database for information about market entry, and uses a generative artificial intelligence model to generate a response on specific market research methods, competitive analysis, and promotion strategies. The generated response is sent back to the user via the device. For example, it provides a detailed explanation in the form of, "To enter a market, we first need to research the market. Next, we conduct a competitive analysis and develop a promotion strategy. Specifically..."

[0229] Prompt Sentence Examples

[0230] "Please explain to a nervous junior employee the strategy for entering a new market."

[0231] "Explain project management best practices to struggling users."

[0232] "Advise a confused employee on key customer service points."

[0233] The system allows senior employees to share their experience and knowledge with junior employees in an efficient and personalized way, and by taking into account the user's emotional state, it can provide more appropriate and useful advice.

[0234] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0235] Step 1:

[0236] The server accesses the company's mail and file servers to retrieve new communications and documents. As input, the addresses and authentication information for the mail and file servers are required. Specifically, the server retrieves emails using the IMAP or POP3 protocol and downloads documents from the file server using the FTP or SMB protocol. As output, the new communications and documents are stored on the server.

[0237] Step 2:

[0238] The server preprocesses the communications and documents it receives. The emails and documents obtained in step 1 are required as input. Specifically, the server filters out spam and removes irrelevant information, extracting only business-related information. Regular expressions and natural language processing tools (e.g., NLTK, SpaCy) are used in this process. The output is the preprocessed, clean data.

[0239] Step 3:

[0240] The server analyzes the preprocessed data and extracts important keywords and phrases. It requires the clean data from step 2 as input. Specifically, the server uses text analysis tools to tokenize the data and analyze it using morphological analysis and parse trees. The output is the extracted keywords and phrases, along with their associated metadata (creation date, author, related projects, etc.).

[0241] Step 4:

[0242] The server stores the analyzed information in a knowledge database. The inputs required are the keywords and phrases obtained in step 3, as well as metadata. Specifically, the server stores this information in a relational or NoSQL database (e.g., MongoDB, PostgreSQL) and organizes it for efficient searches. The output is an updated knowledge database.

[0243] Step 5:

[0244] A user inputs a question through a dedicated user interface (e.g., a chatbot). The user's question is acquired as input through the user interface. Specifically, this role is played by a chatbot or desktop application that runs on a web browser. The user's question is sent to a server as output.

[0245] Step 6:

[0246] The server analyzes the user's question and extracts key keywords and intent. The input required is the user's question obtained in step 5. Specifically, the question is tokenized using natural language processing techniques and analyzed using morphological analysis and parse tree. The output is the extracted keywords and phrases.

[0247] Step 7:

[0248] The server uses an emotion engine to recognize the user's emotional state. The input required is the analysis result of the question obtained in step 6. Specifically, the emotion engine (e.g., EmotionAPI) is used to analyze the user's input and emotion indicators using facial expression recognition technology. The output is the recognized user's emotional information.

[0249] Step 8:

[0250] The server searches for relevant information from the knowledge database. The inputs required are the keywords and phrases obtained in step 6 and the emotion information obtained in step 7. Specifically, the server executes a search query on the knowledge database to obtain relevant know-how information. The output is the search results.

[0251] Step 9:

[0252] The server generates a response using a generative AI model (e.g., ChatGPT). The inputs required are the search results obtained in step 8 and the emotion information obtained in step 7. Specifically, the server inputs a prompt sentence into the generative AI model to generate an appropriate response. The generated response is obtained as the output.

[0253] Step 10:

[0254] The server returns the generated response to the user through a user interface. The generated response obtained in step 9 is required as input. Specifically, the server displays the response to the user through a chatbot or an on-screen message box. As output, the user receives a specific and actionable answer.

[0255] (Application example 2)

[0256] 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."

[0257] Companies are faced with the need to efficiently process the work-related information contained in the large volume of emails and documents generated within their organizations, and to effectively pass on the know-how of senior employees to junior employees. However, when junior employees ask specific questions, traditional systems struggle to provide personalized responses, and they lack an understanding of the emotional aspects of the situation, resulting in a lack of effective support. This can result in junior employees taking a long time to resolve their issues.

[0258] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0259] In this invention, the server includes means for periodically collecting emails and documents generated within the company, means for preprocessing the collected emails and documents to extract business-related information, means for analyzing the preprocessed business-related information, extracting know-how and storing it in a database, means for accepting questions from users through a user interface, means for analyzing the accepted questions and searching for related information from a know-how database, means for generating responses to the questions using a generative artificial intelligence model based on the search results, means for returning the generated responses to the users, means for recognizing the emotional state of the user when analyzing the questions, and means for adjusting the content of the responses based on the emotional information. This allows senior employees to pass on their know-how to junior employees in an efficient and personalized manner, and the emotion recognition function enables more effective problem solving.

[0260] "Means for regularly collecting emails and documents generated within a company" is a function for automatically obtaining emails and document files created within an organization at specified intervals.

[0261] "Means of preprocessing collected emails and documents to extract business-related information" refers to a function for removing unnecessary data from acquired emails and document files and selecting important business-related information.

[0262] "Means for analyzing preprocessed business-related information, extracting know-how, and storing it in a database" refers to a function for further analyzing preprocessed data and organizing and storing knowledge and experience that are useful for business activities.

[0263] The "means for accepting questions from users through a user interface" is a function that provides an interface for users to send questions to the system.

[0264] The "means for analyzing a received question and searching for related information from a know-how database" is a function for analyzing the content of a question from a user and searching for related information from a saved know-how database.

[0265] "Means for generating responses to questions using a generative artificial intelligence model based on search results" refers to a function that uses an AI model to generate specific responses based on the related information found.

[0266] The "means for returning the generated response to the user" is a function for sending the generated response to the user via the user interface.

[0267] The "means for recognizing the emotional state of the user when analyzing a question" is a function for analyzing the emotions of the user when asking a question and recognizing that state.

[0268] The "means for adjusting the content of a response based on emotional information" is a function for adjusting the content of a response to be generated based on the recognized emotional information.

[0269] The present invention provides a system for passing on the know-how of senior employees to junior employees in a factory, and incorporates an emotion recognition function. Detailed embodiments of the system will be described below.

[0270] Hardware and software used

[0271] 1. Hardware:

[0272] Factory robots: Used to answer questions from junior employees and provide know-how from senior employees.

[0273] Smartphone or tablet device (optional): Can be used by junior employees to enter questions and receive answers.

[0274] 2. Software:

[0275] Natural language processing tools (e.g., NLTK, spaCy): Used for text analysis of emails and documents.

[0276] Emotion recognition engine (e.g., Microsoft® Azure® Cognitive Services): Analyzes and recognizes the user's emotional state.

[0277] Generative AI models (e.g., ChatGPT): Generate answers to questions.

[0278] System Operation

[0279] 1. Data Collection:

[0280] The server uses the IMAP protocol for the company's mail server and the FTP protocol for the file server to periodically retrieve emails and documents generated by senior employees.

[0281] 2. Data Preprocessing:

[0282] The server preprocesses the acquired data using natural language processing tools, filtering out spam and irrelevant information and extracting only business-related information.

[0283] 3. Data analysis and know-how extraction:

[0284] The preprocessed data is analyzed to extract important keywords and phrases, which are then used to organize the company's know-how and store it in a database along with metadata (creation date, creator, related projects, etc.).

[0285] 4. Question acceptance and analysis:

[0286] Users (young employees) input questions through the factory robot's interface. The input questions are analyzed using natural language processing tools to extract key keywords and intent.

[0287] 5. Emotion recognition and know-how retrieval:

[0288] The server uses an emotion recognition engine to recognize the user's emotional state when analyzing the question, and based on this emotional information, searches for related information from a know-how database.

[0289] 6. Response Generation:

[0290] Based on the analyzed question content and emotion recognition results, the server uses a generative artificial intelligence model (ChatGPT) to generate an appropriate response and adjust the response content.

[0291] 7. Providing a Response:

[0292] The generated responses are then sent back to junior employees through the factory robot's interface, allowing users to receive instant, specific, and emotionally sensitive answers.

[0293] Specific examples

[0294] Consider a case where a junior employee asks, "How do I troubleshoot a particular machine?" A factory robot receives this question and sends it to a server. The server analyzes the question and uses an emotion recognition engine to identify the user's emotions, such as urgency. Based on this information, it searches for relevant information in a know-how database and generates a specific troubleshooting method using ChatGPT. The generated response is then sent back to the junior employee through the robot's interface.

[0295] Prompt Sentence Examples

[0296] A junior employee asks:

[0297] How do I troubleshoot a specific machine?

[0298] The server analyzes the questions, recognizes emotions, and generates a response using a generative AI model. This system ensures that senior employees' know-how is passed on in an efficient and personalized way, resulting in more effective problem-solving.

[0299] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0300] Step 1: Data collection

[0301] The server accesses the company's mail server using the IMAP protocol to periodically retrieve emails sent and received by senior employees. It also accesses the file server using the FTP protocol to periodically collect document files. These emails and documents are imported into the system's initial database. The input is the emails and documents generated within the company, and the output is the collected raw data.

[0302] Step 2: Data Preprocessing

[0303] The server preprocesses the collected emails and document files. This includes spam filtering and the removal of irrelevant information. It uses natural language processing tools (such as NLTK and spaCy) to analyze the text content of each email and document and extract business-related information. The input is the collected raw data, and the output is the preprocessed meaningful data. Specific operations include filtering out spam emails and extracting important keywords and phrases.

[0304] Step 3: Data analysis and know-how extraction

[0305] The server further analyzes the preprocessed data. It uses natural language processing tools to detect important keywords and phrases and extracts know-how based on these. The extracted know-how information is stored in a database along with metadata (creation date and time, creator, related projects, etc.). The input is the preprocessed meaningful data, and the output is the know-how information stored in the database. Specifically, it performs morphological analysis to organize and classify highly relevant information.

[0306] Step 4: Receive and analyze questions

[0307] The user inputs a question through the factory robot's interface. The terminal sends the question to the server. The server analyzes the received question using a natural language processing tool and extracts key keywords and intent. The input is the user's question, and the output is the analyzed question content. Specifically, the natural language processing tool is used to perform morphological analysis of the question and understand the intent.

[0308] Step 5: Emotion recognition and know-how retrieval

[0309] The server uses an emotion recognition engine (such as Microsoft Azure Cognitive Services) to recognize the user's emotional state when analyzing the question. Based on this emotional information, it searches for related information from a know-how database. The input is the analyzed question content and the user's emotional information, and the output is the search results. Specifically, it combines natural language processing and emotion analysis to optimize the search query.

[0310] Step 6: Response Generation

[0311] The server generates a response to the question using a generative AI model (ChatGPT) based on the emotion recognition results and know-how search results. At this time, it adjusts the response content taking into account the emotion information. The input is the search results and emotion information, and the output is the generated response. Specifically, it inputs a prompt into the generative AI model to obtain a personalized answer.

[0312] Step 7: Providing a response

[0313] The server sends the generated response back to the user through the factory robot's interface. The user can instantly receive a specific and actionable answer through the interface on their terminal. The input is the generated response, and the output is the answer provided to the user. Specific operations include displaying the response on the robot's display and providing a voice response using a speech synthesis function.

[0314] 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.

[0315] 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 (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.

[0316] 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.

[0317] [Second embodiment]

[0318] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0319] 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.

[0320] 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).

[0321] 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.

[0322] 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.

[0323] 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).

[0324] 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.

[0325] 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.

[0326] 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.

[0327] 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.

[0328] 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.

[0329] 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."

[0330] MODE FOR CARRYING OUT THE INVENTION

[0331] The present invention provides a system for effectively passing on the know-how of senior employees to junior employees. The details of this system are described below.

[0332] Data collection

[0333] First, the server periodically accesses the company's mail server and file server to retrieve new emails and documents. During this process, it connects to the mail server using the IMAP or POP3 protocol to retrieve unread emails, and to the file server using the FTP or SMB protocol to retrieve new documents from specified folders.

[0334] Data Preprocessing

[0335] The server then preprocesses the emails and documents it receives. Specifically, it filters out spam and unimportant information, and analyzes only business-related information. This preprocessing involves using regular expressions and natural language processing techniques to remove unnecessary information.

[0336] Data analysis and know-how extraction

[0337] The server uses text analysis tools to analyze the preprocessed data. For example, it uses techniques such as TF-IDF and Word2Vec to analyze the content of emails and documents and extract important keywords and phrases. Based on the results of this analysis, it identifies know-how data and stores it in a database.

[0338] Accepting and analyzing questions

[0339] A user inputs a question through a dedicated interface (e.g., a chatbot). When the user inputs a question from a terminal, the question is sent from the terminal to a server. The server analyzes the received question using natural language processing tools to identify the intent and keywords of the question.

[0340] Know-how search and response generation

[0341] Based on the analyzed question, the server searches for relevant information from a know-how database. Based on the search results, the server generates a response to the question using a generative artificial intelligence model such as ChatGPT. For example, if a junior employee asks, "Please tell me how to proceed with a new project," the server will search for know-how information about past projects and use that information to generate a specific approach using ChatGPT.

[0342] Providing a response

[0343] The generated response is sent from the server to the user via the terminal, and the user can receive specific and practical answers through the chatbot interface.

[0344] Specific examples

[0345] For example, consider a case where a junior employee types a question such as, "Please tell me your strategy for entering a new market." The device receives this question and sends it to the server. The server analyzes the question and searches for information on relevant market entry strategies from its know-how database. It then uses ChatGPT to generate responses on specific market research methods, competitive analysis, and promotion strategies, and sends them back to the user via the device.

[0346] This system prevents the loss of know-how when senior employees retire and allows younger employees to easily acquire business knowledge, thereby effectively utilizing knowledge assets within the company and improving business efficiency.

[0347] The processing flow will be explained below.

[0348] Step 1:

[0349] The server periodically accesses the company's mail server and file server, using the IMAP or POP3 protocol to retrieve unread emails from the mail server, and the FTP or SMB protocol to connect to the file server and retrieve new documents from specified folders.

[0350] Step 2:

[0351] The server pre-processes emails and documents retrieved by the server, specifically filtering out spam, personal information not related to business, and unnecessary data, using regular expressions and natural language processing techniques.

[0352] Step 3:

[0353] The server analyzes the preprocessed email and document data using text analysis tools, such as TF-IDF (Term Frequency-Inverse Document Frequency) and Word2Vec, to extract important keywords and phrases.

[0354] Step 4:

[0355] Based on the analysis results, the server organizes the company's know-how and stores it in a database as structured data. The know-how information stored also includes related metadata (creation date and time, creator, related projects, etc.).

[0356] Step 5:

[0357] The user inputs a question through a dedicated user interface (e.g., a chatbot). The terminal receives the user's input and sends the question to the server.

[0358] Step 6:

[0359] The server analyzes the received question using natural language processing tools, extracting the intent of the question and key keywords, and providing the analysis results.

[0360] Step 7:

[0361] The server searches for relevant information based on keywords extracted from the know-how database and the intent of the question, and identifies the necessary data based on similar questions from the past and related documents.

[0362] Step 8:

[0363] The server generates a response to the question based on the search results using a generative artificial intelligence model (e.g., ChatGPT). Specifically, it refers to the searched know-how and constructs an appropriate answer in natural language.

[0364] Step 9:

[0365] The server returns the generated response to the user through the user interface, and the user can receive a specific and practical answer through the interface on the terminal.

[0366] Example 1

[0367] 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."

[0368] The transfer of knowledge and know-how within a company has become a major issue due to the retirement or transfer of senior employees. In particular, there is a huge amount of information accumulated in emails and documents, and there is a need for a method to quickly extract the necessary information and efficiently convey it to younger employees. This has led to a demand for a system that can improve work productivity and enable the systematic transfer of knowledge.

[0369] 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.

[0370] In this invention, the server includes means for periodically collecting messages and documents generated within the company, means for preprocessing the collected messages and documents to filter out unnecessary information and extract only business-related information, and means for analyzing the preprocessed business-related information, extracting important keywords and phrases, and storing them in a database. This enables the system to efficiently collect and analyze the know-how of senior employees and provide necessary information to junior employees in a timely manner.

[0371] "Messages" are digital communications such as emails and instant messages sent and received within a company.

[0372] "Documents" are digital or paper text data such as reports, memos, plans, etc. generated within a company.

[0373] "Collection" is the act of periodically acquiring specified digital data from a company's network or file server.

[0374] "Preprocessing" is the process of filtering unnecessary information from collected digital data and extracting only the data necessary for analysis.

[0375] "Filtering" is the process of selecting necessary information from digital data based on certain criteria and removing unnecessary information.

[0376] "Business-related information" refers to digital data such as knowledge, know-how, and procedures necessary for a company to carry out its business.

[0377] "Analysis" is the process of analyzing collected and pre-processed digital data to identify important keywords and phrases.

[0378] A "keyword" is a word or phrase that has particular meaning or significance within digital data.

[0379] A "phrase" is a short sentence consisting of one or more words that express a specific meaning.

[0380] A "database" is an information system that systematically stores and manages collected and analyzed digital data.

[0381] A "user interface" is an interaction means through which a user operates or inputs information into a system.

[0382] A "question" is an inquiry that a user inputs to the system to request specific information.

[0383] "Natural language processing tools" are software and algorithms used to understand and analyze human language.

[0384] A "generative artificial intelligence model" is an artificial intelligence algorithm that can generate new information from previously learned data.

[0385] A "response" is an answer that the system provides to a question from a user.

[0386] MODE FOR CARRYING OUT THE INVENTION

[0387] The present invention relates to a system for effectively transferring knowledge and know-how within a company. In this system, a server collects messages and documents within the company, performs preprocessing and analysis, and generates responses based on the collected data using a generative artificial intelligence model.

[0388] First, the server periodically connects to the company's mail server and file server to collect new messages and documents. It connects to the mail server using the IMAP or POP3 protocol to retrieve unread messages, and connects to the file server using the FTP or SMB protocol to retrieve new documents from specified folders. This collection process continuously collects the company's digital assets.

[0389] The server then pre-processes the collected messages and documents, using regular expressions and natural language processing techniques to filter out spam and other irrelevant information, leaving the pre-processed data with business-relevant information extracted.

[0390] The server then performs text analysis on the preprocessed data. For example, it uses TF-IDF technology to extract important keywords from the data and Word2Vec technology to cluster sentences and phrases. This allows the company's know-how to be systematically stored in a database.

[0391] When a user inputs a question through a dedicated interface (e.g., a chatbot), the device sends the question to a server. The server uses natural language processing tools to analyze the intent and keywords of the question, and then searches a know-how database for relevant information based on that analysis. It then uses a generative artificial intelligence model (e.g., ChatGPT) to generate a specific response to the question. This generated response is then returned to the user via the device.

[0392] For example, if a junior employee inputs a question such as, "What should I pay attention to when planning a new product?", the terminal sends this question to the server, which analyzes the question and then searches for related information in the know-how database. A generative AI model is used to generate a specific response such as, "When planning a new product, it is important to start with market research and then conduct a competitive analysis." This response is then sent back to the user via the terminal.

[0393] This invention allows the know-how of senior employees to be collected and analyzed effectively, and necessary information to be provided to younger employees in a timely manner, thereby contributing to improved work efficiency.

[0394] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0395] Step 1: Data collection

[0396] The server periodically connects to the company's mail server and file server to collect new messages and documents. For input, it uses the IMAP or POP3 protocol for mail servers and the FTP or SMB protocol for file servers. For output, it obtains unread messages and new document data. For example, the server connects to "imap.example.com" using the IMAP protocol to obtain unread emails.

[0397] Step 2: Data Preprocessing

[0398] The server pre-processes the collected messages and documents. It uses the collected message and document data as input and outputs business-relevant information that has been spam filtered and filtered out using regular expressions. For example, the server uses regular expressions to filter out emails containing the word "SPAM" and remove unimportant information.

[0399] Step 3: Data analysis

[0400] The server analyzes the preprocessed business-related information. The preprocessed business-related information is provided as input, and important keywords and phrases are extracted and clustered using TF-IDF and Word2Vec techniques. As output, the extracted important keywords and phrases are stored in a database. For example, the server uses the "TF-IDF" library to identify high-frequency keywords from email content.

[0401] Step 4: Ask a question

[0402] A user inputs a question through a dedicated interface (e.g., a chatbot). The question text from the user is entered into the terminal as input and sent to the server via the terminal as output. For example, a user might input into the terminal, "What points should I pay attention to when planning a new product?"

[0403] Step 5: Parsing the Question

[0404] The server analyzes the received questions using natural language processing tools. The question text sent to the server is used as input, and the intent and keywords of the question are identified as output. For example, the server uses the "spaCy" library to extract key keywords from the question text.

[0405] Step 6: Know-how search and response generation

[0406] The server searches for relevant information from a know-how database based on the analyzed question content and generates a response to the question using a generative artificial intelligence model. The intent of the question and keywords are used as input, and a specific response text is generated as output. For example, the server uses ChatGPT to generate a response such as, "When planning a new product, it is important to start with market research and conduct a competitive analysis."

[0407] Step 7: Providing a response

[0408] The server sends the generated response back to the user via the terminal. The generated response text is used as input and displayed on the user's interface as output. For example, the terminal receives the response text from the server and displays it on the chatbot screen.

[0409] Through these steps, the system can effectively collect and analyze know-how within the company and provide appropriate information to young employees.

[0410] (Application example 1)

[0411] 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."

[0412] The knowledge and experience of senior employees within a company are valuable assets, but the loss of know-how due to retirement and other reasons is an issue. Furthermore, there is a lack of systems that allow younger employees to quickly and effectively acquire that know-how. Factories, in particular, need systems that can provide immediate and appropriate instructions and advice. Unless this issue is resolved, the company's competitiveness may decline.

[0413] 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.

[0414] In this invention, the server includes: means for periodically collecting emails and documents generated within the company; means for preprocessing the collected emails and documents to extract business-related information; means for analyzing the preprocessed business-related information, extracting experience data, and storing the extracted experience data in a database; means for accepting questions from users through a user interface; means for analyzing the accepted questions and searching for related information from an experience database; means for generating responses to the questions using a generative artificial intelligence model based on the search results; means for returning the generated responses to the users; and means for being installed in a robot installed in a factory to support young engineers and operators in sharing the know-how of senior engineers. This makes it possible to effectively pass on the know-how of senior employees to junior employees and to provide immediate and appropriate instructions and advice, particularly on the factory floor.

[0415] "Intra-company" refers to the inside of a particular company or organization, and is the scope related to the operation and business of that organization.

[0416] "Mail" refers to electronic mail, a message containing text and attachments sent and received over the Internet.

[0417] "Documents" refers to any paper or electronic document created, stored, and shared within an organization, including reports, presentations, memoranda, etc.

[0418] "Preprocessing" refers to the process of removing unnecessary information and reformatting the format of raw data to make it easier to analyze.

[0419] "Business-related information" refers to information that is directly related to a company's business operations and decision-making, including project progress reports and performance reports.

[0420] "Analysis" is the process of analyzing data in detail and finding useful information and patterns within it.

[0421] "Experience" refers to the knowledge, skills, and know-how acquired in the past, which are required to carry out specific tasks or work.

[0422] A "database" is a system that stores and manages information in an orderly manner, making it easy to search and use.

[0423] "User interface" refers to the means and screen display through which a user interacts with a system or application.

[0424] A "generative artificial intelligence model" refers to artificial intelligence technology that automatically generates responses and sentences in response to input data or questions, and includes natural language processing technology.

[0425] A "factory" is a facility where products are manufactured or assembled, and where industrial machinery and production lines are installed.

[0426] A "robot" is an automated machine or device designed to assist or replace human tasks.

[0427] "Young engineers" refer to relatively new employees within a company who are honing their expertise and skills in technical fields.

[0428] "Operator" refers to a staff member responsible for operating and managing a machine or system.

[0429] "Senior engineers" refer to technical leaders and veteran employees within a company who have extensive experience and advanced expertise.

[0430] The present invention provides a system for efficiently passing on the know-how of senior employees to junior employees in a company. This system is composed of the following means.

[0431] First, the server periodically collects emails and documents generated within the company. Emails are collected using the IMAP or POP3 protocol to retrieve unread emails, and documents are collected using the FTP or SMB protocol to retrieve new documents from specified folders.

[0432] The server then pre-processes the collected emails and documents to extract business-related information. This pre-processing removes spam and unimportant information, leaving only information relevant to the company's business. This process involves filtering out unnecessary information using regular expressions and natural language processing techniques.

[0433] The preprocessed business-related information is then analyzed by the server to extract experiences. Text analysis tools such as TF-IDF (Term Frequency-Inverse Document Frequency) and Word2Vec are used for the analysis. These tools are used to analyze the content of emails and documents and extract important keywords and phrases. Based on the results of this analysis, the experience data is stored in a database.

[0434] Users input questions through a dedicated interface (for example, through a chatbot or a robot installed in a factory). When a user inputs a question from a terminal, the question is sent from the terminal to a server. The server analyzes the received question using natural language processing tools to identify the intent of the question and important keywords.

[0435] Based on the analyzed question, the server searches for relevant information from an experience database. Based on the search results, the server generates a response to the question using a generative AI model (e.g., ChatGPT). For example, if a junior engineer asks, "Please tell me how to proceed with a new project," the server searches for experience information about past projects and uses the generative AI model to generate a specific approach based on that information.

[0436] The generated responses are sent from the server to the user via the terminal, allowing the user to receive specific and practical answers through the interface. Particularly in factory settings, robots can provide young engineers and operators with appropriate instructions and advice in real time.

[0437] As a concrete example, the prompt sentence is shown below.

[0438] Example prompt:

[0439] question: "How do I proceed with a new project?"

[0440] context: "Search for know-how information from past projects and explain the specific steps you took based on that information. Include market research methods, competitive analysis, and promotion strategies."

[0441] This system allows senior employees to effectively pass on their know-how to younger employees, and makes it possible to provide immediate and appropriate instructions and advice, particularly on factory floors, thereby preserving knowledge assets within the company and improving operational efficiency.

[0442] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0443] Step 1:

[0444] The server periodically accesses the company's mail server and file server to retrieve new emails and documents. The input here is the configuration information for the mail server and file server, and the output is the newly retrieved emails and documents. Specifically, it retrieves unread emails from the mail server using the IMAP or POP3 protocol, and retrieves new documents from the file server using the FTP or SMB protocol.

[0445] Step 2:

[0446] The server preprocesses the emails and documents it receives. The input is the emails and documents received in step 1, and the output is refined data that contains only business-related information. Specifically, regular expressions are used to remove spam and unimportant information, and natural language processing techniques are used to extract business-related information.

[0447] Step 3:

[0448] The server analyzes the preprocessed business-related information, extracts experiences, and stores them in a database. The input is the result data from Step 2, and the output is experience data analyzed using TF-IDF and Word2Vec. Specifically, the TF-IDF vectorizer is used to calculate feature weights, and the Word2Vec model is used to learn the relationships between words in the data, thereby extracting important experience data and storing it in a database.

[0449] Step 4:

[0450] The user inputs a question through a dedicated interface. The input here is the user's question, and the output is the transmission of that question. Specifically, the user's question is received using a chatbot interface or a robot's voice input system.

[0451] Step 5:

[0452] The server analyzes the received question and searches the experience database for relevant information. The input is the user's question, and the output is the analyzed keywords and important phrases of the question, as well as related experience data. Specifically, it uses natural language processing tools to extract the intent and keywords of the question, and then searches the experience database based on them.

[0453] Step 6:

[0454] Based on the search results, the server uses a generative AI model to generate a response to the question. The input is the relevant experience data obtained in step 5, and the output is the generated response text. Specifically, a generative AI model such as ChatGPT is used to generate a specific response based on relevant information. For example, in response to a question such as "Please tell me how to proceed with a new project," the server incorporates experience data from past projects and suggests a specific approach.

[0455] Step 7:

[0456] The server returns the generated response to the user via the terminal. The input is the response generated in step 6, and the output is the response provided to the user. Specifically, the generated response is delivered to the user via the chatbot screen or the robot's voice output. This series of processes allows junior engineers and operators to share the know-how of senior engineers in real time.

[0457] 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.

[0458] MODE FOR CARRYING OUT THE INVENTION

[0459] The present invention provides a system for effectively passing on the know-how of senior employees to younger employees, and by incorporating a user emotion recognition function, it achieves more personalized responses.

[0460] Data collection

[0461] The server periodically accesses the company's mail and file servers to retrieve new emails and documents, using the IMAP or POP3 protocols for mail servers and the FTP or SMB protocols for file servers.

[0462] Data Preprocessing

[0463] The server pre-processes the emails and documents it receives. During this process, it filters out spam and irrelevant information and extracts only business-related information. Regular expressions and natural language processing techniques are used for pre-processing.

[0464] Data analysis and know-how extraction

[0465] The server then uses text analysis tools to analyze the preprocessed data and extract important keywords and phrases. The extracted information is then used to organize the company's know-how and store it in a database. Metadata (creation date and time, creator, related projects, etc.) is also stored.

[0466] Accepting and analyzing questions

[0467] The user inputs a question through a dedicated user interface (e.g., a chatbot). The device receives the user's question and sends it to a server. The server analyzes the question using natural language processing tools and extracts key keywords and intent.

[0468] Emotion recognition and know-how retrieval

[0469] The server uses an emotion engine to recognize the user's emotional state when analyzing the question. The emotion engine recognizes emotions by analyzing the user's input and other emotional indicators. Based on this emotional information, it searches for related information from a know-how database.

[0470] Response Generation

[0471] The server generates a response to the question using a generative AI model (e.g., ChatGPT) based on the emotion recognition results and know-how search results. By taking the emotion recognition results into account and adjusting the response content, it provides a more personalized answer.

[0472] Providing a response

[0473] The server returns the generated response to the user through the user interface, allowing the user to instantly receive a specific and actionable answer through the interface on their device.

[0474] Specific examples

[0475] For example, consider the case where a junior employee types in, "Please tell me about our strategy for entering a new market." The device receives this question and sends it to the server. The server analyzes the question and uses an emotion engine to recognize the user's emotions, such as nervousness or confusion. Based on this information, it searches a know-how database for information about market entry and uses a generative artificial intelligence model to generate a response regarding specific market research methods, competitive analysis, and promotion strategies. The generated response is then sent back to the user via the device.

[0476] This system enables senior employees to pass on their know-how to junior employees in an efficient and personalized manner. The addition of emotion recognition functionality enables responses that take into account the user's emotional state, resulting in more effective problem-solving.

[0477] The processing flow will be explained below.

[0478] Step 1:

[0479] The server periodically accesses the company's mail server and file server, using the IMAP or POP3 protocol to retrieve unread emails from the mail server, and the FTP or SMB protocol to connect to the file server and retrieve new documents from specified folders.

[0480] Step 2:

[0481] The server pre-processes the emails and documents it receives. Specifically, it filters out spam and irrelevant information and extracts only business-related information. This pre-processing uses regular expressions and natural language processing techniques.

[0482] Step 3:

[0483] The server analyzes the preprocessed email and document data using text analysis tools, such as TF-IDF (Term Frequency-Inverse Document Frequency) and Word2Vec, to extract important keywords and phrases.

[0484] Step 4:

[0485] Based on the analysis results, the server organizes the company's know-how and stores it as structured data in a know-how database. The stored know-how information also includes related metadata (creation date and time, creator, related projects, etc.).

[0486] Step 5:

[0487] The user inputs a question through a dedicated user interface (e.g., a chatbot). The terminal receives the user's input and sends the question to the server.

[0488] Step 6:

[0489] The server analyzes the received question using natural language processing tools, extracting the intent of the question and key keywords, and providing the analysis results.

[0490] Step 7:

[0491] The server uses an emotion engine to analyze the emotion contained in the user's question. The emotion engine uses natural language processing technology and machine learning algorithms to recognize emotions from the user's input. For example, it analyzes whether the user is confused or nervous based on the question.

[0492] Step 8:

[0493] The server searches for relevant information based on keywords extracted from the know-how database and the intent of the question, and identifies the necessary data based on similar questions from the past and related documents.

[0494] Step 9:

[0495] The server generates a response to the question using a generative AI model (e.g., ChatGPT) based on the search results and the user's emotion recognition results. By taking the emotion recognition results into account and adjusting the response content, it generates the optimal answer for the user.

[0496] Step 10:

[0497] The server returns the generated response to the user through the user interface, and the user can receive a specific and practical answer through the interface on the terminal.

[0498] Specific examples

[0499] For example, consider the case where a junior employee types, "Please tell me about our strategy for entering a new market." The device receives this question and sends it to the server. The server analyzes the question and uses an emotion engine to recognize that the user is confused. Based on this information, it searches a know-how database for information about market entry and uses a generative artificial intelligence model to generate a response on specific market research methods, competitive analysis, and promotion strategies. The generated response is then sent back to the user via the device. In this way, a response that takes the user's emotional state into account is provided, enabling more effective problem solving.

[0500] Example 2

[0501] 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."

[0502] Efficiently transferring the know-how of senior employees to junior employees is an important issue for many companies. However, many current systems lack emotion recognition capabilities, respond uniformly, and are unable to provide personalized responses according to the emotional state of each user, making it difficult to achieve effective problem solving.

[0503] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes: means for periodically collecting communications and documents generated within the company; means for preprocessing the collected communications and documents to extract business-related information; means for analyzing the preprocessed business-related information, extracting knowledge, and storing it in a database; means for accepting inquiries from users through a user interface; means for analyzing the accepted inquiries and searching for related information from a knowledge database; means for generating a response to the inquiry using a generation system based on the search results; means for returning the generated response to the user; and means for having an emotion engine that recognizes the user's emotional state and adjusting the response content based on the recognized emotion information. This enables senior employees to pass on their know-how to junior employees in an efficient and personalized manner, and enables responses that take the user's emotional state into consideration, thereby achieving more effective problem solving.

[0504] "Communications" refers to the means of information exchange generated within a company, such as emails and chat messages.

[0505] "Document" refers to written or electronic records of information generated within an enterprise, such as reports, memos, and presentations.

[0506] "Business-related information" refers to information and data related to a company's business activities, including information on project management, risk management, and market research.

[0507] "Knowledge" refers to useful information and know-how extracted from business-related information and used to solve specific business problems.

[0508] A "database" refers to a collection of information that systematically organizes knowledge and makes it possible to store and search it.

[0509] "User interface" refers to the means by which a user interacts with a system, including graphical user interfaces (GUIs) and chatbots.

[0510] A "generative system" refers to a system that automatically generates responses to user inquiries using natural language processing technology or generative AI models.

[0511] An "emotion engine" refers to a function that analyzes a user's input and other emotional indicators to recognize their emotional state.

[0512] "Preprocessing" refers to a series of processes that filter spam and irrelevant information from collected communications and documents and extract only business-related information.

[0513] "Analysis" refers to the process of analyzing pre-processed business-related information in detail and extracting important keywords and phrases.

[0514] The present invention provides a system that effectively transfers the know-how of senior employees to junior employees while incorporating a function for recognizing user emotions to provide more personalized responses. A specific implementation method of this system will be described below.

[0515] First, the server, which is the core of this system, periodically collects communications and documents generated within the company. The server uses IMAP (Internet Message Access Protocol) and POP3 (Post Office Protocol 3) protocols to retrieve emails, and FTP (File Transfer Protocol) and SMB (Server Message Block) protocols to communicate with the file server. This process is performed automatically and periodically using CRON job settings.

[0516] The collected communications and documents are pre-processed by the server. In this pre-processing step, spam emails and irrelevant information are filtered out, and only business-related information is extracted. Regular expressions and natural language processing techniques (e.g., NLTK, SpaCy) are used in this process. For example, only emails containing keywords such as "important notice" or "weekly report" are extracted.

[0517] The preprocessed data is then subjected to detailed analysis using data analysis tools. The server tokenizes the information and performs morphological analysis to extract frequently occurring keywords and phrases. Important phrases, such as "project management" and "risk management," are automatically tagged and this information is stored in a knowledge database. Additionally, metadata such as creation date, creator, and related projects are also stored.

[0518] When a user inputs a question through a dedicated user interface (e.g., a chatbot), the device receives the question and sends it to a server. The server then analyzes the received question using natural language processing technology to extract key keywords and intent. Morphological analysis and parse tree analysis are used to analyze the user's text.

[0519] Furthermore, the server uses an emotion engine (e.g., EmotionAPI) to recognize the user's emotional state. The emotion engine recognizes emotions by analyzing the user's input and facial expression recognition using facial recognition technology. For example, if a user inputs "I'm very troubled," the emotion engine will recognize emotions such as "confusion" and "tension." Based on this recognition result, it searches for related information from a knowledge database.

[0520] Finally, the server generates a response using a generative AI model (e.g., ChatGPT) based on the emotional information and search results. The server inputs a prompt into the model, such as "Please explain your strategy for entering a new market to a nervous junior employee," to generate an appropriate response. The generated response is then sent back to the user via their device.

[0521] Specific examples

[0522] For example, consider the case where a junior employee types, "Please tell me about our strategy for entering a new market." The device receives this question and sends it to the server. The server analyzes the question and uses an emotion engine to recognize the user's emotions, such as nervousness or confusion. Based on this information, it searches a know-how database for information about market entry, and uses a generative artificial intelligence model to generate a response on specific market research methods, competitive analysis, and promotion strategies. The generated response is sent back to the user via the device. For example, it provides a detailed explanation in the form of, "To enter a market, we first need to research the market. Next, we conduct a competitive analysis and develop a promotion strategy. Specifically..."

[0523] Prompt Sentence Examples

[0524] "Please explain to a nervous junior employee the strategy for entering a new market."

[0525] "Explain project management best practices to struggling users."

[0526] "Advise a confused employee on key customer service points."

[0527] The system allows senior employees to share their experience and knowledge with junior employees in an efficient and personalized way, and by taking into account the user's emotional state, it can provide more appropriate and useful advice.

[0528] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0529] Step 1:

[0530] The server accesses the company's mail and file servers to retrieve new communications and documents. As input, the addresses and authentication information for the mail and file servers are required. Specifically, the server retrieves emails using the IMAP or POP3 protocol and downloads documents from the file server using the FTP or SMB protocol. As output, the new communications and documents are stored on the server.

[0531] Step 2:

[0532] The server preprocesses the communications and documents it receives. The emails and documents obtained in step 1 are required as input. Specifically, the server filters out spam and removes irrelevant information, extracting only business-related information. Regular expressions and natural language processing tools (e.g., NLTK, SpaCy) are used in this process. The output is the preprocessed, clean data.

[0533] Step 3:

[0534] The server analyzes the preprocessed data and extracts important keywords and phrases. It requires the clean data from step 2 as input. Specifically, the server uses text analysis tools to tokenize the data and analyze it using morphological analysis and parse trees. The output is the extracted keywords and phrases, along with their associated metadata (creation date, author, related projects, etc.).

[0535] Step 4:

[0536] The server stores the analyzed information in a knowledge database. The inputs required are the keywords and phrases obtained in step 3, as well as metadata. Specifically, the server stores this information in a relational or NoSQL database (e.g., MongoDB, PostgreSQL) and organizes it for efficient searches. The output is an updated knowledge database.

[0537] Step 5:

[0538] A user inputs a question through a dedicated user interface (e.g., a chatbot). The user's question is acquired as input through the user interface. Specifically, this role is played by a chatbot or desktop application that runs on a web browser. The user's question is sent to a server as output.

[0539] Step 6:

[0540] The server analyzes the user's question and extracts key keywords and intent. The input required is the user's question obtained in step 5. Specifically, the question is tokenized using natural language processing techniques and analyzed using morphological analysis and parse tree. The output is the extracted keywords and phrases.

[0541] Step 7:

[0542] The server uses an emotion engine to recognize the user's emotional state. The input required is the analysis result of the question obtained in step 6. Specifically, the emotion engine (e.g., EmotionAPI) is used to analyze the user's input and emotion indicators using facial expression recognition technology. The output is the recognized user's emotional information.

[0543] Step 8:

[0544] The server searches for relevant information from the knowledge database. The inputs required are the keywords and phrases obtained in step 6 and the emotion information obtained in step 7. Specifically, the server executes a search query on the knowledge database to obtain relevant know-how information. The output is the search results.

[0545] Step 9:

[0546] The server generates a response using a generative AI model (e.g., ChatGPT). The inputs required are the search results obtained in step 8 and the emotion information obtained in step 7. Specifically, the server inputs a prompt sentence into the generative AI model to generate an appropriate response. The generated response is obtained as the output.

[0547] Step 10:

[0548] The server returns the generated response to the user through a user interface. The generated response obtained in step 9 is required as input. Specifically, the server displays the response to the user through a chatbot or an on-screen message box. As output, the user receives a specific and actionable answer.

[0549] (Application example 2)

[0550] 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."

[0551] Companies are faced with the need to efficiently process the work-related information contained in the large volume of emails and documents generated within their organizations, and to effectively pass on the know-how of senior employees to junior employees. However, when junior employees ask specific questions, traditional systems struggle to provide personalized responses, and they lack an understanding of the emotional aspects of the situation, resulting in a lack of effective support. This can result in junior employees taking a long time to resolve their issues.

[0552] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0553] In this invention, the server includes means for periodically collecting emails and documents generated within the company, means for preprocessing the collected emails and documents to extract business-related information, means for analyzing the preprocessed business-related information, extracting know-how and storing it in a database, means for accepting questions from users through a user interface, means for analyzing the accepted questions and searching for related information from a know-how database, means for generating responses to the questions using a generative artificial intelligence model based on the search results, means for returning the generated responses to the users, means for recognizing the emotional state of the user when analyzing the questions, and means for adjusting the content of the responses based on the emotional information. This allows senior employees to pass on their know-how to junior employees in an efficient and personalized manner, and the emotion recognition function enables more effective problem solving.

[0554] "Means for regularly collecting emails and documents generated within a company" is a function for automatically obtaining emails and document files created within an organization at specified intervals.

[0555] "Means of preprocessing collected emails and documents to extract business-related information" refers to a function for removing unnecessary data from acquired emails and document files and selecting important business-related information.

[0556] "Means for analyzing preprocessed business-related information, extracting know-how, and storing it in a database" refers to a function for further analyzing preprocessed data and organizing and storing knowledge and experience that are useful for business activities.

[0557] The "means for accepting questions from users through a user interface" is a function that provides an interface for users to send questions to the system.

[0558] The "means for analyzing a received question and searching for related information from a know-how database" is a function for analyzing the content of a question from a user and searching for related information from a saved know-how database.

[0559] "Means for generating responses to questions using a generative artificial intelligence model based on search results" refers to a function that uses an AI model to generate specific responses based on the related information found.

[0560] The "means for returning the generated response to the user" is a function for sending the generated response to the user via the user interface.

[0561] The "means for recognizing the emotional state of the user when analyzing a question" is a function for analyzing the emotions of the user when asking a question and recognizing that state.

[0562] The "means for adjusting the content of a response based on emotional information" is a function for adjusting the content of a response to be generated based on the recognized emotional information.

[0563] The present invention provides a system for passing on the know-how of senior employees to junior employees in a factory, and incorporates an emotion recognition function. Detailed embodiments of the system will be described below.

[0564] Hardware and software used

[0565] 1. Hardware:

[0566] Factory robots: Used to answer questions from junior employees and provide know-how from senior employees.

[0567] Smartphone or tablet device (optional): Can be used by junior employees to enter questions and receive answers.

[0568] 2. Software:

[0569] Natural language processing tools (e.g., NLTK, spaCy): Used for text analysis of emails and documents.

[0570] Emotion recognition engine (e.g., Microsoft Azure Cognitive Services): Analyzes and recognizes the user's emotional state.

[0571] Generative AI models (e.g., ChatGPT): Generate answers to questions.

[0572] System Operation

[0573] 1. Data Collection:

[0574] The server uses the IMAP protocol for the company's mail server and the FTP protocol for the file server to periodically retrieve emails and documents generated by senior employees.

[0575] 2. Data Preprocessing:

[0576] The server preprocesses the acquired data using natural language processing tools, filtering out spam and irrelevant information and extracting only business-related information.

[0577] 3. Data analysis and know-how extraction:

[0578] The preprocessed data is analyzed to extract important keywords and phrases, which are then used to organize the company's know-how and store it in a database along with metadata (creation date, creator, related projects, etc.).

[0579] 4. Question acceptance and analysis:

[0580] Users (young employees) input questions through the factory robot's interface. The input questions are analyzed using natural language processing tools to extract key keywords and intent.

[0581] 5. Emotion recognition and know-how retrieval:

[0582] The server uses an emotion recognition engine to recognize the user's emotional state when analyzing the question, and based on this emotional information, searches for related information from a know-how database.

[0583] 6. Response Generation:

[0584] Based on the analyzed question content and emotion recognition results, the server uses a generative artificial intelligence model (ChatGPT) to generate an appropriate response and adjust the response content.

[0585] 7. Providing a Response:

[0586] The generated responses are then sent back to junior employees through the factory robot's interface, allowing users to receive instant, specific, and emotionally sensitive answers.

[0587] Specific examples

[0588] Consider a case where a junior employee asks, "How do I troubleshoot a particular machine?" A factory robot receives this question and sends it to a server. The server analyzes the question and uses an emotion recognition engine to identify the user's emotions, such as urgency. Based on this information, it searches for relevant information in a know-how database and generates a specific troubleshooting method using ChatGPT. The generated response is then sent back to the junior employee through the robot's interface.

[0589] Prompt Sentence Examples

[0590] A junior employee asks:

[0591] How do I troubleshoot a specific machine?

[0592] The server analyzes the questions, recognizes emotions, and generates a response using a generative AI model. This system ensures that senior employees' know-how is passed on in an efficient and personalized way, resulting in more effective problem-solving.

[0593] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0594] Step 1: Data collection

[0595] The server accesses the company's mail server using the IMAP protocol to periodically retrieve emails sent and received by senior employees. It also accesses the file server using the FTP protocol to periodically collect document files. These emails and documents are imported into the system's initial database. The input is the emails and documents generated within the company, and the output is the collected raw data.

[0596] Step 2: Data Preprocessing

[0597] The server preprocesses the collected emails and document files. This includes spam filtering and the removal of irrelevant information. It uses natural language processing tools (such as NLTK and spaCy) to analyze the text content of each email and document and extract business-related information. The input is the collected raw data, and the output is the preprocessed meaningful data. Specific operations include filtering out spam emails and extracting important keywords and phrases.

[0598] Step 3: Data analysis and know-how extraction

[0599] The server further analyzes the preprocessed data. It uses natural language processing tools to detect important keywords and phrases and extracts know-how based on these. The extracted know-how information is stored in a database along with metadata (creation date and time, creator, related projects, etc.). The input is the preprocessed meaningful data, and the output is the know-how information stored in the database. Specifically, it performs morphological analysis to organize and classify highly relevant information.

[0600] Step 4: Receive and analyze questions

[0601] The user inputs a question through the factory robot's interface. The terminal sends the question to the server. The server analyzes the received question using a natural language processing tool and extracts key keywords and intent. The input is the user's question, and the output is the analyzed question content. Specifically, the natural language processing tool is used to perform morphological analysis of the question and understand the intent.

[0602] Step 5: Emotion recognition and know-how retrieval

[0603] The server uses an emotion recognition engine (such as Microsoft Azure Cognitive Services) to recognize the user's emotional state when analyzing the question. Based on this emotional information, it searches for related information from a know-how database. The input is the analyzed question content and the user's emotional information, and the output is the search results. Specifically, it combines natural language processing and emotion analysis to optimize the search query.

[0604] Step 6: Response Generation

[0605] The server generates a response to the question using a generative AI model (ChatGPT) based on the emotion recognition results and know-how search results. At this time, it adjusts the response content taking into account the emotion information. The input is the search results and emotion information, and the output is the generated response. Specifically, it inputs a prompt into the generative AI model to obtain a personalized answer.

[0606] Step 7: Providing a response

[0607] The server sends the generated response back to the user through the factory robot's interface. The user can instantly receive a specific and actionable answer through the interface on their terminal. The input is the generated response, and the output is the answer provided to the user. Specific operations include displaying the response on the robot's display and providing a voice response using a speech synthesis function.

[0608] 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.

[0609] 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.

[0610] 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.

[0611] [Third embodiment]

[0612] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0613] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0614] 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).

[0615] 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.

[0616] 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.

[0617] 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).

[0618] 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.

[0619] 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.

[0620] 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.

[0621] 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.

[0622] 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.

[0623] 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."

[0624] MODE FOR CARRYING OUT THE INVENTION

[0625] The present invention provides a system for effectively passing on the know-how of senior employees to junior employees. The details of this system are described below.

[0626] Data collection

[0627] First, the server periodically accesses the company's mail server and file server to retrieve new emails and documents. During this process, it connects to the mail server using the IMAP or POP3 protocol to retrieve unread emails, and to the file server using the FTP or SMB protocol to retrieve new documents from specified folders.

[0628] Data Preprocessing

[0629] The server then preprocesses the emails and documents it receives. Specifically, it filters out spam and unimportant information, and analyzes only business-related information. This preprocessing involves using regular expressions and natural language processing techniques to remove unnecessary information.

[0630] Data analysis and know-how extraction

[0631] The server uses text analysis tools to analyze the preprocessed data. For example, it uses techniques such as TF-IDF and Word2Vec to analyze the content of emails and documents and extract important keywords and phrases. Based on the results of this analysis, it identifies know-how data and stores it in a database.

[0632] Accepting and analyzing questions

[0633] A user inputs a question through a dedicated interface (e.g., a chatbot). When the user inputs a question from a terminal, the question is sent from the terminal to a server. The server analyzes the received question using natural language processing tools to identify the intent and keywords of the question.

[0634] Know-how search and response generation

[0635] Based on the analyzed question, the server searches for relevant information from a know-how database. Based on the search results, the server generates a response to the question using a generative artificial intelligence model such as ChatGPT. For example, if a junior employee asks, "Please tell me how to proceed with a new project," the server will search for know-how information about past projects and use that information to generate a specific approach using ChatGPT.

[0636] Providing a response

[0637] The generated response is sent from the server to the user via the terminal, and the user can receive specific and practical answers through the chatbot interface.

[0638] Specific examples

[0639] For example, consider a case where a junior employee types a question such as, "Please tell me your strategy for entering a new market." The device receives this question and sends it to the server. The server analyzes the question and searches for information on relevant market entry strategies from its know-how database. It then uses ChatGPT to generate responses on specific market research methods, competitive analysis, and promotion strategies, and sends them back to the user via the device.

[0640] This system prevents the loss of know-how when senior employees retire and allows younger employees to easily acquire business knowledge, thereby effectively utilizing knowledge assets within the company and improving business efficiency.

[0641] The processing flow will be explained below.

[0642] Step 1:

[0643] The server periodically accesses the company's mail server and file server, using the IMAP or POP3 protocol to retrieve unread emails from the mail server, and the FTP or SMB protocol to connect to the file server and retrieve new documents from specified folders.

[0644] Step 2:

[0645] The server pre-processes emails and documents retrieved by the server, specifically filtering out spam, personal information not related to business, and unnecessary data, using regular expressions and natural language processing techniques.

[0646] Step 3:

[0647] The server analyzes the preprocessed email and document data using text analysis tools, such as TF-IDF (Term Frequency-Inverse Document Frequency) and Word2Vec, to extract important keywords and phrases.

[0648] Step 4:

[0649] Based on the analysis results, the server organizes the company's know-how and stores it in a database as structured data. The know-how information stored also includes related metadata (creation date and time, creator, related projects, etc.).

[0650] Step 5:

[0651] The user inputs a question through a dedicated user interface (e.g., a chatbot). The terminal receives the user's input and sends the question to the server.

[0652] Step 6:

[0653] The server analyzes the received question using natural language processing tools, extracting the intent of the question and key keywords, and providing the analysis results.

[0654] Step 7:

[0655] The server searches for relevant information based on keywords extracted from the know-how database and the intent of the question, and identifies the necessary data based on similar questions from the past and related documents.

[0656] Step 8:

[0657] The server generates a response to the question based on the search results using a generative artificial intelligence model (e.g., ChatGPT). Specifically, it refers to the searched know-how and constructs an appropriate answer in natural language.

[0658] Step 9:

[0659] The server returns the generated response to the user through the user interface, and the user can receive a specific and practical answer through the interface on the terminal.

[0660] Example 1

[0661] 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."

[0662] The transfer of knowledge and know-how within a company has become a major issue due to the retirement or transfer of senior employees. In particular, there is a huge amount of information accumulated in emails and documents, and there is a need for a method to quickly extract the necessary information and efficiently convey it to younger employees. This has led to a demand for a system that can improve work productivity and enable the systematic transfer of knowledge.

[0663] 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.

[0664] In this invention, the server includes means for periodically collecting messages and documents generated within the company, means for preprocessing the collected messages and documents to filter out unnecessary information and extract only business-related information, and means for analyzing the preprocessed business-related information, extracting important keywords and phrases, and storing them in a database. This enables the system to efficiently collect and analyze the know-how of senior employees and provide necessary information to junior employees in a timely manner.

[0665] "Messages" are digital communications such as emails and instant messages sent and received within a company.

[0666] "Documents" are digital or paper text data such as reports, memos, plans, etc. generated within a company.

[0667] "Collection" is the act of periodically acquiring specified digital data from a company's network or file server.

[0668] "Preprocessing" is the process of filtering unnecessary information from collected digital data and extracting only the data necessary for analysis.

[0669] "Filtering" is the process of selecting necessary information from digital data based on certain criteria and removing unnecessary information.

[0670] "Business-related information" refers to digital data such as knowledge, know-how, and procedures necessary for a company to carry out its business.

[0671] "Analysis" is the process of analyzing collected and pre-processed digital data to identify important keywords and phrases.

[0672] A "keyword" is a word or phrase that has particular meaning or significance within digital data.

[0673] A "phrase" is a short sentence consisting of one or more words that express a specific meaning.

[0674] A "database" is an information system that systematically stores and manages collected and analyzed digital data.

[0675] A "user interface" is an interaction means through which a user operates or inputs information into a system.

[0676] A "question" is an inquiry that a user inputs to the system to request specific information.

[0677] "Natural language processing tools" are software and algorithms used to understand and analyze human language.

[0678] A "generative artificial intelligence model" is an artificial intelligence algorithm that can generate new information from previously learned data.

[0679] A "response" is an answer that the system provides to a question from a user.

[0680] MODE FOR CARRYING OUT THE INVENTION

[0681] The present invention relates to a system for effectively transferring knowledge and know-how within a company. In this system, a server collects messages and documents within the company, performs preprocessing and analysis, and generates responses based on the collected data using a generative artificial intelligence model.

[0682] First, the server periodically connects to the company's mail server and file server to collect new messages and documents. It connects to the mail server using the IMAP or POP3 protocol to retrieve unread messages, and connects to the file server using the FTP or SMB protocol to retrieve new documents from specified folders. This collection process continuously collects the company's digital assets.

[0683] The server then pre-processes the collected messages and documents, using regular expressions and natural language processing techniques to filter out spam and other irrelevant information, leaving the pre-processed data with business-relevant information extracted.

[0684] The server then performs text analysis on the preprocessed data. For example, it uses TF-IDF technology to extract important keywords from the data and Word2Vec technology to cluster sentences and phrases. This allows the company's know-how to be systematically stored in a database.

[0685] When a user inputs a question through a dedicated interface (e.g., a chatbot), the device sends the question to a server. The server uses natural language processing tools to analyze the intent and keywords of the question, and then searches a know-how database for relevant information based on that analysis. It then uses a generative artificial intelligence model (e.g., ChatGPT) to generate a specific response to the question. This generated response is then returned to the user via the device.

[0686] For example, if a junior employee inputs a question such as, "What should I pay attention to when planning a new product?", the terminal sends this question to the server, which analyzes the question and then searches for related information in the know-how database. A generative AI model is used to generate a specific response such as, "When planning a new product, it is important to start with market research and then conduct a competitive analysis." This response is then sent back to the user via the terminal.

[0687] This invention allows the know-how of senior employees to be collected and analyzed effectively, and necessary information to be provided to younger employees in a timely manner, thereby contributing to improved work efficiency.

[0688] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0689] Step 1: Data collection

[0690] The server periodically connects to the company's mail server and file server to collect new messages and documents. For input, it uses the IMAP or POP3 protocol for mail servers and the FTP or SMB protocol for file servers. For output, it obtains unread messages and new document data. For example, the server connects to "imap.example.com" using the IMAP protocol to obtain unread emails.

[0691] Step 2: Data Preprocessing

[0692] The server pre-processes the collected messages and documents. It uses the collected message and document data as input and outputs business-relevant information that has been spam filtered and filtered out using regular expressions. For example, the server uses regular expressions to filter out emails containing the word "SPAM" and remove unimportant information.

[0693] Step 3: Data analysis

[0694] The server analyzes the preprocessed business-related information. The preprocessed business-related information is provided as input, and important keywords and phrases are extracted and clustered using TF-IDF and Word2Vec techniques. As output, the extracted important keywords and phrases are stored in a database. For example, the server uses the "TF-IDF" library to identify high-frequency keywords from email content.

[0695] Step 4: Ask a question

[0696] A user inputs a question through a dedicated interface (e.g., a chatbot). The question text from the user is entered into the terminal as input and sent to the server via the terminal as output. For example, a user might input into the terminal, "What points should I pay attention to when planning a new product?"

[0697] Step 5: Parsing the Question

[0698] The server analyzes the received questions using natural language processing tools. The question text sent to the server is used as input, and the intent and keywords of the question are identified as output. For example, the server uses the "spaCy" library to extract key keywords from the question text.

[0699] Step 6: Know-how search and response generation

[0700] The server searches for relevant information from a know-how database based on the analyzed question content and generates a response to the question using a generative artificial intelligence model. The intent of the question and keywords are used as input, and a specific response text is generated as output. For example, the server uses ChatGPT to generate a response such as, "When planning a new product, it is important to start with market research and conduct a competitive analysis."

[0701] Step 7: Providing a response

[0702] The server sends the generated response back to the user via the terminal. The generated response text is used as input and displayed on the user's interface as output. For example, the terminal receives the response text from the server and displays it on the chatbot screen.

[0703] Through these steps, the system can effectively collect and analyze know-how within the company and provide appropriate information to young employees.

[0704] (Application example 1)

[0705] 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."

[0706] The knowledge and experience of senior employees within a company are valuable assets, but the loss of know-how due to retirement and other reasons is an issue. Furthermore, there is a lack of systems that allow younger employees to quickly and effectively acquire that know-how. Factories, in particular, need systems that can provide immediate and appropriate instructions and advice. Unless this issue is resolved, the company's competitiveness may decline.

[0707] 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.

[0708] In this invention, the server includes: means for periodically collecting emails and documents generated within the company; means for preprocessing the collected emails and documents to extract business-related information; means for analyzing the preprocessed business-related information, extracting experience data, and storing the extracted experience data in a database; means for accepting questions from users through a user interface; means for analyzing the accepted questions and searching for related information from an experience database; means for generating responses to the questions using a generative artificial intelligence model based on the search results; means for returning the generated responses to the users; and means for being installed in a robot installed in a factory to support young engineers and operators in sharing the know-how of senior engineers. This makes it possible to effectively pass on the know-how of senior employees to junior employees and to provide immediate and appropriate instructions and advice, particularly on the factory floor.

[0709] "Intra-company" refers to the inside of a particular company or organization, and is the scope related to the operation and business of that organization.

[0710] "Mail" refers to electronic mail, a message containing text and attachments sent and received over the Internet.

[0711] "Documents" refers to any paper or electronic document created, stored, and shared within an organization, including reports, presentations, memoranda, etc.

[0712] "Preprocessing" refers to the process of removing unnecessary information and reformatting the format of raw data to make it easier to analyze.

[0713] "Business-related information" refers to information that is directly related to a company's business operations and decision-making, including project progress reports and performance reports.

[0714] "Analysis" is the process of analyzing data in detail and finding useful information and patterns within it.

[0715] "Experience" refers to the knowledge, skills, and know-how acquired in the past, which are required to carry out specific tasks or work.

[0716] A "database" is a system that stores and manages information in an orderly manner, making it easy to search and use.

[0717] "User interface" refers to the means and screen display through which a user interacts with a system or application.

[0718] A "generative artificial intelligence model" refers to artificial intelligence technology that automatically generates responses and sentences in response to input data or questions, and includes natural language processing technology.

[0719] A "factory" is a facility where products are manufactured or assembled, and where industrial machinery and production lines are installed.

[0720] A "robot" is an automated machine or device designed to assist or replace human tasks.

[0721] "Young engineers" refer to relatively new employees within a company who are honing their expertise and skills in technical fields.

[0722] "Operator" refers to a staff member responsible for operating and managing a machine or system.

[0723] "Senior engineers" refer to technical leaders and veteran employees within a company who have extensive experience and advanced expertise.

[0724] The present invention provides a system for efficiently passing on the know-how of senior employees to junior employees in a company. This system is composed of the following means.

[0725] First, the server periodically collects emails and documents generated within the company. Emails are collected using the IMAP or POP3 protocol to retrieve unread emails, and documents are collected using the FTP or SMB protocol to retrieve new documents from specified folders.

[0726] The server then pre-processes the collected emails and documents to extract business-related information. This pre-processing removes spam and unimportant information, leaving only information relevant to the company's business. This process involves filtering out unnecessary information using regular expressions and natural language processing techniques.

[0727] The preprocessed business-related information is then analyzed by the server to extract experiences. Text analysis tools such as TF-IDF (Term Frequency-Inverse Document Frequency) and Word2Vec are used for the analysis. These tools are used to analyze the content of emails and documents and extract important keywords and phrases. Based on the results of this analysis, the experience data is stored in a database.

[0728] Users input questions through a dedicated interface (for example, through a chatbot or a robot installed in a factory). When a user inputs a question from a terminal, the question is sent from the terminal to a server. The server analyzes the received question using natural language processing tools to identify the intent of the question and important keywords.

[0729] Based on the analyzed question, the server searches for relevant information from an experience database. Based on the search results, the server generates a response to the question using a generative AI model (e.g., ChatGPT). For example, if a junior engineer asks, "Please tell me how to proceed with a new project," the server searches for experience information about past projects and uses the generative AI model to generate a specific approach based on that information.

[0730] The generated responses are sent from the server to the user via the terminal, allowing the user to receive specific and practical answers through the interface. Particularly in factory settings, robots can provide young engineers and operators with appropriate instructions and advice in real time.

[0731] As a concrete example, the prompt sentence is shown below.

[0732] Example prompt:

[0733] question: "How do I proceed with a new project?"

[0734] context: "Search for know-how information from past projects and explain the specific steps you took based on that information. Include market research methods, competitive analysis, and promotion strategies."

[0735] This system allows senior employees to effectively pass on their know-how to younger employees, and makes it possible to provide immediate and appropriate instructions and advice, particularly on factory floors, thereby preserving knowledge assets within the company and improving operational efficiency.

[0736] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0737] Step 1:

[0738] The server periodically accesses the company's mail server and file server to retrieve new emails and documents. The input here is the configuration information for the mail server and file server, and the output is the newly retrieved emails and documents. Specifically, it retrieves unread emails from the mail server using the IMAP or POP3 protocol, and retrieves new documents from the file server using the FTP or SMB protocol.

[0739] Step 2:

[0740] The server preprocesses the emails and documents it receives. The input is the emails and documents received in step 1, and the output is refined data that contains only business-related information. Specifically, regular expressions are used to remove spam and unimportant information, and natural language processing techniques are used to extract business-related information.

[0741] Step 3:

[0742] The server analyzes the preprocessed business-related information, extracts experiences, and stores them in a database. The input is the result data from Step 2, and the output is experience data analyzed using TF-IDF and Word2Vec. Specifically, the TF-IDF vectorizer is used to calculate feature weights, and the Word2Vec model is used to learn the relationships between words in the data, thereby extracting important experience data and storing it in a database.

[0743] Step 4:

[0744] The user inputs a question through a dedicated interface. The input here is the user's question, and the output is the transmission of that question. Specifically, the user's question is received using a chatbot interface or a robot's voice input system.

[0745] Step 5:

[0746] The server analyzes the received question and searches the experience database for relevant information. The input is the user's question, and the output is the analyzed keywords and important phrases of the question, as well as related experience data. Specifically, it uses natural language processing tools to extract the intent and keywords of the question, and then searches the experience database based on them.

[0747] Step 6:

[0748] Based on the search results, the server uses a generative AI model to generate a response to the question. The input is the relevant experience data obtained in step 5, and the output is the generated response text. Specifically, a generative AI model such as ChatGPT is used to generate a specific response based on relevant information. For example, in response to a question such as "Please tell me how to proceed with a new project," the server incorporates experience data from past projects and suggests a specific approach.

[0749] Step 7:

[0750] The server returns the generated response to the user via the terminal. The input is the response generated in step 6, and the output is the response provided to the user. Specifically, the generated response is delivered to the user via the chatbot screen or the robot's voice output. This series of processes allows junior engineers and operators to share the know-how of senior engineers in real time.

[0751] 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.

[0752] MODE FOR CARRYING OUT THE INVENTION

[0753] The present invention provides a system for effectively passing on the know-how of senior employees to younger employees, and by incorporating a user emotion recognition function, it achieves more personalized responses.

[0754] Data collection

[0755] The server periodically accesses the company's mail and file servers to retrieve new emails and documents, using the IMAP or POP3 protocols for mail servers and the FTP or SMB protocols for file servers.

[0756] Data Preprocessing

[0757] The server pre-processes the emails and documents it receives. During this process, it filters out spam and irrelevant information and extracts only business-related information. Regular expressions and natural language processing techniques are used for pre-processing.

[0758] Data analysis and know-how extraction

[0759] The server then uses text analysis tools to analyze the preprocessed data and extract important keywords and phrases. The extracted information is then used to organize the company's know-how and store it in a database. Metadata (creation date and time, creator, related projects, etc.) is also stored.

[0760] Accepting and analyzing questions

[0761] The user inputs a question through a dedicated user interface (e.g., a chatbot). The device receives the user's question and sends it to a server. The server analyzes the question using natural language processing tools and extracts key keywords and intent.

[0762] Emotion recognition and know-how retrieval

[0763] The server uses an emotion engine to recognize the user's emotional state when analyzing the question. The emotion engine recognizes emotions by analyzing the user's input and other emotional indicators. Based on this emotional information, it searches for related information from a know-how database.

[0764] Response Generation

[0765] The server generates a response to the question using a generative AI model (e.g., ChatGPT) based on the emotion recognition results and know-how search results. By taking the emotion recognition results into account and adjusting the response content, it provides a more personalized answer.

[0766] Providing a response

[0767] The server returns the generated response to the user through the user interface, allowing the user to instantly receive a specific and actionable answer through the interface on their device.

[0768] Specific examples

[0769] For example, consider the case where a junior employee types in, "Please tell me about our strategy for entering a new market." The device receives this question and sends it to the server. The server analyzes the question and uses an emotion engine to recognize the user's emotions, such as nervousness or confusion. Based on this information, it searches a know-how database for information about market entry and uses a generative artificial intelligence model to generate a response regarding specific market research methods, competitive analysis, and promotion strategies. The generated response is then sent back to the user via the device.

[0770] This system enables senior employees to pass on their know-how to junior employees in an efficient and personalized manner. The addition of emotion recognition functionality enables responses that take into account the user's emotional state, resulting in more effective problem-solving.

[0771] The processing flow will be explained below.

[0772] Step 1:

[0773] The server periodically accesses the company's mail server and file server, using the IMAP or POP3 protocol to retrieve unread emails from the mail server, and the FTP or SMB protocol to connect to the file server and retrieve new documents from specified folders.

[0774] Step 2:

[0775] The server pre-processes the emails and documents it receives. Specifically, it filters out spam and irrelevant information and extracts only business-related information. This pre-processing uses regular expressions and natural language processing techniques.

[0776] Step 3:

[0777] The server analyzes the preprocessed email and document data using text analysis tools, such as TF-IDF (Term Frequency-Inverse Document Frequency) and Word2Vec, to extract important keywords and phrases.

[0778] Step 4:

[0779] Based on the analysis results, the server organizes the company's know-how and stores it as structured data in a know-how database. The stored know-how information also includes related metadata (creation date and time, creator, related projects, etc.).

[0780] Step 5:

[0781] The user inputs a question through a dedicated user interface (e.g., a chatbot). The terminal receives the user's input and sends the question to the server.

[0782] Step 6:

[0783] The server analyzes the received question using natural language processing tools, extracting the intent of the question and key keywords, and providing the analysis results.

[0784] Step 7:

[0785] The server uses an emotion engine to analyze the emotion contained in the user's question. The emotion engine uses natural language processing technology and machine learning algorithms to recognize emotions from the user's input. For example, it analyzes whether the user is confused or nervous based on the question.

[0786] Step 8:

[0787] The server searches for relevant information based on keywords extracted from the know-how database and the intent of the question, and identifies the necessary data based on similar questions from the past and related documents.

[0788] Step 9:

[0789] The server generates a response to the question using a generative AI model (e.g., ChatGPT) based on the search results and the user's emotion recognition results. By taking the emotion recognition results into account and adjusting the response content, it generates the optimal answer for the user.

[0790] Step 10:

[0791] The server returns the generated response to the user through the user interface, and the user can receive a specific and practical answer through the interface on the terminal.

[0792] Specific examples

[0793] For example, consider the case where a junior employee types, "Please tell me about our strategy for entering a new market." The device receives this question and sends it to the server. The server analyzes the question and uses an emotion engine to recognize that the user is confused. Based on this information, it searches a know-how database for information about market entry and uses a generative artificial intelligence model to generate a response on specific market research methods, competitive analysis, and promotion strategies. The generated response is then sent back to the user via the device. In this way, a response that takes the user's emotional state into account is provided, enabling more effective problem solving.

[0794] Example 2

[0795] 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."

[0796] Efficiently transferring the know-how of senior employees to junior employees is an important issue for many companies. However, many current systems lack emotion recognition capabilities, respond uniformly, and are unable to provide personalized responses according to the emotional state of each user, making it difficult to achieve effective problem solving.

[0797] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes: means for periodically collecting communications and documents generated within the company; means for preprocessing the collected communications and documents to extract business-related information; means for analyzing the preprocessed business-related information, extracting knowledge, and storing it in a database; means for accepting inquiries from users through a user interface; means for analyzing the accepted inquiries and searching for related information from a knowledge database; means for generating a response to the inquiry using a generation system based on the search results; means for returning the generated response to the user; and means for having an emotion engine that recognizes the user's emotional state and adjusting the response content based on the recognized emotion information. This enables senior employees to pass on their know-how to junior employees in an efficient and personalized manner, and enables responses that take the user's emotional state into consideration, thereby achieving more effective problem solving.

[0798] "Communications" refers to the means of information exchange generated within a company, such as emails and chat messages.

[0799] "Document" refers to written or electronic records of information generated within an enterprise, such as reports, memos, and presentations.

[0800] "Business-related information" refers to information and data related to a company's business activities, including information on project management, risk management, and market research.

[0801] "Knowledge" refers to useful information and know-how extracted from business-related information and used to solve specific business problems.

[0802] A "database" refers to a collection of information that systematically organizes knowledge and makes it possible to store and search it.

[0803] "User interface" refers to the means by which a user interacts with a system, including graphical user interfaces (GUIs) and chatbots.

[0804] A "generative system" refers to a system that automatically generates responses to user inquiries using natural language processing technology or generative AI models.

[0805] An "emotion engine" refers to a function that analyzes a user's input and other emotional indicators to recognize their emotional state.

[0806] "Preprocessing" refers to a series of processes that filter spam and irrelevant information from collected communications and documents and extract only business-related information.

[0807] "Analysis" refers to the process of analyzing pre-processed business-related information in detail and extracting important keywords and phrases.

[0808] The present invention provides a system that effectively transfers the know-how of senior employees to junior employees while incorporating a function for recognizing user emotions to provide more personalized responses. A specific implementation method of this system will be described below.

[0809] First, the server, which is the core of this system, periodically collects communications and documents generated within the company. The server uses IMAP (Internet Message Access Protocol) and POP3 (Post Office Protocol 3) protocols to retrieve emails, and FTP (File Transfer Protocol) and SMB (Server Message Block) protocols to communicate with the file server. This process is performed automatically and periodically using CRON job settings.

[0810] The collected communications and documents are pre-processed by the server. In this pre-processing step, spam emails and irrelevant information are filtered out, and only business-related information is extracted. Regular expressions and natural language processing techniques (e.g., NLTK, SpaCy) are used in this process. For example, only emails containing keywords such as "important notice" or "weekly report" are extracted.

[0811] The preprocessed data is then subjected to detailed analysis using data analysis tools. The server tokenizes the information and performs morphological analysis to extract frequently occurring keywords and phrases. Important phrases, such as "project management" and "risk management," are automatically tagged and this information is stored in a knowledge database. Additionally, metadata such as creation date, creator, and related projects are also stored.

[0812] When a user inputs a question through a dedicated user interface (e.g., a chatbot), the device receives the question and sends it to a server. The server then analyzes the received question using natural language processing technology to extract key keywords and intent. Morphological analysis and parse tree analysis are used to analyze the user's text.

[0813] Furthermore, the server uses an emotion engine (e.g., EmotionAPI) to recognize the user's emotional state. The emotion engine recognizes emotions by analyzing the user's input and facial expression recognition using facial recognition technology. For example, if a user inputs "I'm very troubled," the emotion engine will recognize emotions such as "confusion" and "tension." Based on this recognition result, it searches for related information from a knowledge database.

[0814] Finally, the server generates a response using a generative AI model (e.g., ChatGPT) based on the emotional information and search results. The server inputs a prompt into the model, such as "Please explain your strategy for entering a new market to a nervous junior employee," to generate an appropriate response. The generated response is then sent back to the user via their device.

[0815] Specific examples

[0816] For example, consider the case where a junior employee types, "Please tell me about our strategy for entering a new market." The device receives this question and sends it to the server. The server analyzes the question and uses an emotion engine to recognize the user's emotions, such as nervousness or confusion. Based on this information, it searches a know-how database for information about market entry, and uses a generative artificial intelligence model to generate a response on specific market research methods, competitive analysis, and promotion strategies. The generated response is sent back to the user via the device. For example, it provides a detailed explanation in the form of, "To enter a market, we first need to research the market. Next, we conduct a competitive analysis and develop a promotion strategy. Specifically..."

[0817] Prompt Sentence Examples

[0818] "Please explain to a nervous junior employee the strategy for entering a new market."

[0819] "Explain project management best practices to struggling users."

[0820] "Advise a confused employee on key customer service points."

[0821] The system allows senior employees to share their experience and knowledge with junior employees in an efficient and personalized way, and by taking into account the user's emotional state, it can provide more appropriate and useful advice.

[0822] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0823] Step 1:

[0824] The server accesses the company's mail and file servers to retrieve new communications and documents. As input, the addresses and authentication information for the mail and file servers are required. Specifically, the server retrieves emails using the IMAP or POP3 protocol and downloads documents from the file server using the FTP or SMB protocol. As output, the new communications and documents are stored on the server.

[0825] Step 2:

[0826] The server preprocesses the communications and documents it receives. The emails and documents obtained in step 1 are required as input. Specifically, the server filters out spam and removes irrelevant information, extracting only business-related information. Regular expressions and natural language processing tools (e.g., NLTK, SpaCy) are used in this process. The output is the preprocessed, clean data.

[0827] Step 3:

[0828] The server analyzes the preprocessed data and extracts important keywords and phrases. It requires the clean data from step 2 as input. Specifically, the server uses text analysis tools to tokenize the data and analyze it using morphological analysis and parse trees. The output is the extracted keywords and phrases, along with their associated metadata (creation date, author, related projects, etc.).

[0829] Step 4:

[0830] The server stores the analyzed information in a knowledge database. The inputs required are the keywords and phrases obtained in step 3, as well as metadata. Specifically, the server stores this information in a relational or NoSQL database (e.g., MongoDB, PostgreSQL) and organizes it for efficient searches. The output is an updated knowledge database.

[0831] Step 5:

[0832] A user inputs a question through a dedicated user interface (e.g., a chatbot). The user's question is acquired as input through the user interface. Specifically, this role is played by a chatbot or desktop application that runs on a web browser. The user's question is sent to a server as output.

[0833] Step 6:

[0834] The server analyzes the user's question and extracts key keywords and intent. The input required is the user's question obtained in step 5. Specifically, the question is tokenized using natural language processing techniques and analyzed using morphological analysis and parse tree. The output is the extracted keywords and phrases.

[0835] Step 7:

[0836] The server uses an emotion engine to recognize the user's emotional state. The input required is the analysis result of the question obtained in step 6. Specifically, the emotion engine (e.g., EmotionAPI) is used to analyze the user's input and emotion indicators using facial expression recognition technology. The output is the recognized user's emotional information.

[0837] Step 8:

[0838] The server searches for relevant information from the knowledge database. The inputs required are the keywords and phrases obtained in step 6 and the emotion information obtained in step 7. Specifically, the server executes a search query on the knowledge database to obtain relevant know-how information. The output is the search results.

[0839] Step 9:

[0840] The server generates a response using a generative AI model (e.g., ChatGPT). The inputs required are the search results obtained in step 8 and the emotion information obtained in step 7. Specifically, the server inputs a prompt sentence into the generative AI model to generate an appropriate response. The generated response is obtained as the output.

[0841] Step 10:

[0842] The server returns the generated response to the user through a user interface. The generated response obtained in step 9 is required as input. Specifically, the server displays the response to the user through a chatbot or an on-screen message box. As output, the user receives a specific and actionable answer.

[0843] (Application example 2)

[0844] 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."

[0845] Companies are faced with the need to efficiently process the work-related information contained in the large volume of emails and documents generated within their organizations, and to effectively pass on the know-how of senior employees to junior employees. However, when junior employees ask specific questions, traditional systems struggle to provide personalized responses, and they lack an understanding of the emotional aspects of the situation, resulting in a lack of effective support. This can result in junior employees taking a long time to resolve their issues.

[0846] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0847] In this invention, the server includes means for periodically collecting emails and documents generated within the company, means for preprocessing the collected emails and documents to extract business-related information, means for analyzing the preprocessed business-related information, extracting know-how and storing it in a database, means for accepting questions from users through a user interface, means for analyzing the accepted questions and searching for related information from a know-how database, means for generating responses to the questions using a generative artificial intelligence model based on the search results, means for returning the generated responses to the users, means for recognizing the emotional state of the user when analyzing the questions, and means for adjusting the content of the responses based on the emotional information. This allows senior employees to pass on their know-how to junior employees in an efficient and personalized manner, and the emotion recognition function enables more effective problem solving.

[0848] "Means for regularly collecting emails and documents generated within a company" is a function for automatically obtaining emails and document files created within an organization at specified intervals.

[0849] "Means of preprocessing collected emails and documents to extract business-related information" refers to a function for removing unnecessary data from acquired emails and document files and selecting important business-related information.

[0850] "Means for analyzing preprocessed business-related information, extracting know-how, and storing it in a database" refers to a function for further analyzing preprocessed data and organizing and storing knowledge and experience that are useful for business activities.

[0851] The "means for accepting questions from users through a user interface" is a function that provides an interface for users to send questions to the system.

[0852] The "means for analyzing a received question and searching for related information from a know-how database" is a function for analyzing the content of a question from a user and searching for related information from a saved know-how database.

[0853] "Means for generating responses to questions using a generative artificial intelligence model based on search results" refers to a function that uses an AI model to generate specific responses based on the related information found.

[0854] The "means for returning the generated response to the user" is a function for sending the generated response to the user via the user interface.

[0855] The "means for recognizing the emotional state of the user when analyzing a question" is a function for analyzing the emotions of the user when asking a question and recognizing that state.

[0856] The "means for adjusting the content of a response based on emotional information" is a function for adjusting the content of a response to be generated based on the recognized emotional information.

[0857] The present invention provides a system for passing on the know-how of senior employees to junior employees in a factory, and incorporates an emotion recognition function. Detailed embodiments of the system will be described below.

[0858] Hardware and software used

[0859] 1. Hardware:

[0860] Factory robots: Used to answer questions from junior employees and provide know-how from senior employees.

[0861] Smartphone or tablet device (optional): Can be used by junior employees to enter questions and receive answers.

[0862] 2. Software:

[0863] Natural language processing tools (e.g., NLTK, spaCy): Used for text analysis of emails and documents.

[0864] Emotion recognition engine (e.g., Microsoft Azure Cognitive Services): Analyzes and recognizes the user's emotional state.

[0865] Generative AI models (e.g., ChatGPT): Generate answers to questions.

[0866] System Operation

[0867] 1. Data Collection:

[0868] The server uses the IMAP protocol for the company's mail server and the FTP protocol for the file server to periodically retrieve emails and documents generated by senior employees.

[0869] 2. Data Preprocessing:

[0870] The server preprocesses the acquired data using natural language processing tools, filtering out spam and irrelevant information and extracting only business-related information.

[0871] 3. Data analysis and know-how extraction:

[0872] The preprocessed data is analyzed to extract important keywords and phrases, which are then used to organize the company's know-how and store it in a database along with metadata (creation date, creator, related projects, etc.).

[0873] 4. Question acceptance and analysis:

[0874] Users (young employees) input questions through the factory robot's interface. The input questions are analyzed using natural language processing tools to extract key keywords and intent.

[0875] 5. Emotion recognition and know-how retrieval:

[0876] The server uses an emotion recognition engine to recognize the user's emotional state when analyzing the question, and based on this emotional information, searches for related information from a know-how database.

[0877] 6. Response Generation:

[0878] Based on the analyzed question content and emotion recognition results, the server uses a generative artificial intelligence model (ChatGPT) to generate an appropriate response and adjust the response content.

[0879] 7. Providing a Response:

[0880] The generated responses are then sent back to junior employees through the factory robot's interface, allowing users to receive instant, specific, and emotionally sensitive answers.

[0881] Specific examples

[0882] Consider a case where a junior employee asks, "How do I troubleshoot a particular machine?" A factory robot receives this question and sends it to a server. The server analyzes the question and uses an emotion recognition engine to identify the user's emotions, such as urgency. Based on this information, it searches for relevant information in a know-how database and generates a specific troubleshooting method using ChatGPT. The generated response is then sent back to the junior employee through the robot's interface.

[0883] Prompt Sentence Examples

[0884] A junior employee asks:

[0885] How do I troubleshoot a specific machine?

[0886] The server analyzes the questions, recognizes emotions, and generates a response using a generative AI model. This system ensures that senior employees' know-how is passed on in an efficient and personalized way, resulting in more effective problem-solving.

[0887] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0888] Step 1: Data collection

[0889] The server accesses the company's mail server using the IMAP protocol to periodically retrieve emails sent and received by senior employees. It also accesses the file server using the FTP protocol to periodically collect document files. These emails and documents are imported into the system's initial database. The input is the emails and documents generated within the company, and the output is the collected raw data.

[0890] Step 2: Data Preprocessing

[0891] The server preprocesses the collected emails and document files. This includes spam filtering and the removal of irrelevant information. It uses natural language processing tools (such as NLTK and spaCy) to analyze the text content of each email and document and extract business-related information. The input is the collected raw data, and the output is the preprocessed meaningful data. Specific operations include filtering out spam emails and extracting important keywords and phrases.

[0892] Step 3: Data analysis and know-how extraction

[0893] The server further analyzes the preprocessed data. It uses natural language processing tools to detect important keywords and phrases and extracts know-how based on these. The extracted know-how information is stored in a database along with metadata (creation date and time, creator, related projects, etc.). The input is the preprocessed meaningful data, and the output is the know-how information stored in the database. Specifically, it performs morphological analysis to organize and classify highly relevant information.

[0894] Step 4: Receive and analyze questions

[0895] The user inputs a question through the factory robot's interface. The terminal sends the question to the server. The server analyzes the received question using a natural language processing tool and extracts key keywords and intent. The input is the user's question, and the output is the analyzed question content. Specifically, the natural language processing tool is used to perform morphological analysis of the question and understand the intent.

[0896] Step 5: Emotion recognition and know-how retrieval

[0897] The server uses an emotion recognition engine (such as Microsoft Azure Cognitive Services) to recognize the user's emotional state when analyzing the question. Based on this emotional information, it searches for related information from a know-how database. The input is the analyzed question content and the user's emotional information, and the output is the search results. Specifically, it combines natural language processing and emotion analysis to optimize the search query.

[0898] Step 6: Response Generation

[0899] The server generates a response to the question using a generative AI model (ChatGPT) based on the emotion recognition results and know-how search results. At this time, it adjusts the response content taking into account the emotion information. The input is the search results and emotion information, and the output is the generated response. Specifically, it inputs a prompt into the generative AI model to obtain a personalized answer.

[0900] Step 7: Providing a response

[0901] The server sends the generated response back to the user through the factory robot's interface. The user can instantly receive a specific and actionable answer through the interface on their terminal. The input is the generated response, and the output is the answer provided to the user. Specific operations include displaying the response on the robot's display and providing a voice response using a speech synthesis function.

[0902] 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.

[0903] 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.

[0904] 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.

[0905] [Fourth embodiment]

[0906] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0907] 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.

[0908] 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).

[0909] 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.

[0910] 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.

[0911] 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).

[0912] 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.

[0913] 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.

[0914] 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.

[0915] 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.

[0916] 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.

[0917] 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.

[0918] 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."

[0919] MODE FOR CARRYING OUT THE INVENTION

[0920] The present invention provides a system for effectively passing on the know-how of senior employees to junior employees. The details of this system are described below.

[0921] Data collection

[0922] First, the server periodically accesses the company's mail server and file server to retrieve new emails and documents. During this process, it connects to the mail server using the IMAP or POP3 protocol to retrieve unread emails, and to the file server using the FTP or SMB protocol to retrieve new documents from specified folders.

[0923] Data Preprocessing

[0924] The server then preprocesses the emails and documents it receives. Specifically, it filters out spam and unimportant information, and analyzes only business-related information. This preprocessing involves using regular expressions and natural language processing techniques to remove unnecessary information.

[0925] Data analysis and know-how extraction

[0926] The server uses text analysis tools to analyze the preprocessed data. For example, it uses techniques such as TF-IDF and Word2Vec to analyze the content of emails and documents and extract important keywords and phrases. Based on the results of this analysis, it identifies know-how data and stores it in a database.

[0927] Accepting and analyzing questions

[0928] A user inputs a question through a dedicated interface (e.g., a chatbot). When the user inputs a question from a terminal, the question is sent from the terminal to a server. The server analyzes the received question using natural language processing tools to identify the intent and keywords of the question.

[0929] Know-how search and response generation

[0930] Based on the analyzed question, the server searches for relevant information from a know-how database. Based on the search results, the server generates a response to the question using a generative artificial intelligence model such as ChatGPT. For example, if a junior employee asks, "Please tell me how to proceed with a new project," the server will search for know-how information about past projects and use that information to generate a specific approach using ChatGPT.

[0931] Providing a response

[0932] The generated response is sent from the server to the user via the terminal, and the user can receive specific and practical answers through the chatbot interface.

[0933] Specific examples

[0934] For example, consider a case where a junior employee types a question such as, "Please tell me your strategy for entering a new market." The device receives this question and sends it to the server. The server analyzes the question and searches for information on relevant market entry strategies from its know-how database. It then uses ChatGPT to generate responses on specific market research methods, competitive analysis, and promotion strategies, and sends them back to the user via the device.

[0935] This system prevents the loss of know-how when senior employees retire and allows younger employees to easily acquire business knowledge, thereby effectively utilizing knowledge assets within the company and improving business efficiency.

[0936] The processing flow will be explained below.

[0937] Step 1:

[0938] The server periodically accesses the company's mail server and file server, using the IMAP or POP3 protocol to retrieve unread emails from the mail server, and the FTP or SMB protocol to connect to the file server and retrieve new documents from specified folders.

[0939] Step 2:

[0940] The server pre-processes emails and documents retrieved by the server, specifically filtering out spam, personal information not related to business, and unnecessary data, using regular expressions and natural language processing techniques.

[0941] Step 3:

[0942] The server analyzes the preprocessed email and document data using text analysis tools, such as TF-IDF (Term Frequency-Inverse Document Frequency) and Word2Vec, to extract important keywords and phrases.

[0943] Step 4:

[0944] Based on the analysis results, the server organizes the company's know-how and stores it in a database as structured data. The know-how information stored also includes related metadata (creation date and time, creator, related projects, etc.).

[0945] Step 5:

[0946] The user inputs a question through a dedicated user interface (e.g., a chatbot). The terminal receives the user's input and sends the question to the server.

[0947] Step 6:

[0948] The server analyzes the received question using natural language processing tools, extracting the intent of the question and key keywords, and providing the analysis results.

[0949] Step 7:

[0950] The server searches for relevant information based on keywords extracted from the know-how database and the intent of the question, and identifies the necessary data based on similar questions from the past and related documents.

[0951] Step 8:

[0952] The server generates a response to the question based on the search results using a generative artificial intelligence model (e.g., ChatGPT). Specifically, it refers to the searched know-how and constructs an appropriate answer in natural language.

[0953] Step 9:

[0954] The server returns the generated response to the user through the user interface, and the user can receive a specific and practical answer through the interface on the terminal.

[0955] Example 1

[0956] 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."

[0957] The transfer of knowledge and know-how within a company has become a major issue due to the retirement or transfer of senior employees. In particular, there is a huge amount of information accumulated in emails and documents, and there is a need for a method to quickly extract the necessary information and efficiently convey it to younger employees. This has led to a demand for a system that can improve work productivity and enable the systematic transfer of knowledge.

[0958] 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.

[0959] In this invention, the server includes means for periodically collecting messages and documents generated within the company, means for preprocessing the collected messages and documents to filter out unnecessary information and extract only business-related information, and means for analyzing the preprocessed business-related information, extracting important keywords and phrases, and storing them in a database. This enables the system to efficiently collect and analyze the know-how of senior employees and provide necessary information to junior employees in a timely manner.

[0960] "Messages" are digital communications such as emails and instant messages sent and received within a company.

[0961] "Documents" are digital or paper text data such as reports, memos, plans, etc. generated within a company.

[0962] "Collection" is the act of periodically acquiring specified digital data from a company's network or file server.

[0963] "Preprocessing" is the process of filtering unnecessary information from collected digital data and extracting only the data necessary for analysis.

[0964] "Filtering" is the process of selecting necessary information from digital data based on certain criteria and removing unnecessary information.

[0965] "Business-related information" refers to digital data such as knowledge, know-how, and procedures necessary for a company to carry out its business.

[0966] "Analysis" is the process of analyzing collected and pre-processed digital data to identify important keywords and phrases.

[0967] A "keyword" is a word or phrase that has particular meaning or significance within digital data.

[0968] A "phrase" is a short sentence consisting of one or more words that express a specific meaning.

[0969] A "database" is an information system that systematically stores and manages collected and analyzed digital data.

[0970] A "user interface" is an interaction means through which a user operates or inputs information into a system.

[0971] A "question" is an inquiry that a user inputs to the system to request specific information.

[0972] "Natural language processing tools" are software and algorithms used to understand and analyze human language.

[0973] A "generative artificial intelligence model" is an artificial intelligence algorithm that can generate new information from previously learned data.

[0974] A "response" is an answer that the system provides to a question from a user.

[0975] MODE FOR CARRYING OUT THE INVENTION

[0976] The present invention relates to a system for effectively transferring knowledge and know-how within a company. In this system, a server collects messages and documents within the company, performs preprocessing and analysis, and generates responses based on the collected data using a generative artificial intelligence model.

[0977] First, the server periodically connects to the company's mail server and file server to collect new messages and documents. It connects to the mail server using the IMAP or POP3 protocol to retrieve unread messages, and connects to the file server using the FTP or SMB protocol to retrieve new documents from specified folders. This collection process continuously collects the company's digital assets.

[0978] The server then pre-processes the collected messages and documents, using regular expressions and natural language processing techniques to filter out spam and other irrelevant information, leaving the pre-processed data with business-relevant information extracted.

[0979] The server then performs text analysis on the preprocessed data. For example, it uses TF-IDF technology to extract important keywords from the data and Word2Vec technology to cluster sentences and phrases. This allows the company's know-how to be systematically stored in a database.

[0980] When a user inputs a question through a dedicated interface (e.g., a chatbot), the device sends the question to a server. The server uses natural language processing tools to analyze the intent and keywords of the question, and then searches a know-how database for relevant information based on that analysis. It then uses a generative artificial intelligence model (e.g., ChatGPT) to generate a specific response to the question. This generated response is then returned to the user via the device.

[0981] For example, if a junior employee inputs a question such as, "What should I pay attention to when planning a new product?", the terminal sends this question to the server, which analyzes the question and then searches for related information in the know-how database. A generative AI model is used to generate a specific response such as, "When planning a new product, it is important to start with market research and then conduct a competitive analysis." This response is then sent back to the user via the terminal.

[0982] This invention allows the know-how of senior employees to be collected and analyzed effectively, and necessary information to be provided to younger employees in a timely manner, thereby contributing to improved work efficiency.

[0983] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0984] Step 1: Data collection

[0985] The server periodically connects to the company's mail server and file server to collect new messages and documents. For input, it uses the IMAP or POP3 protocol for mail servers and the FTP or SMB protocol for file servers. For output, it obtains unread messages and new document data. For example, the server connects to "imap.example.com" using the IMAP protocol to obtain unread emails.

[0986] Step 2: Data Preprocessing

[0987] The server pre-processes the collected messages and documents. It uses the collected message and document data as input and outputs business-relevant information that has been spam filtered and filtered out using regular expressions. For example, the server uses regular expressions to filter out emails containing the word "SPAM" and remove unimportant information.

[0988] Step 3: Data analysis

[0989] The server analyzes the preprocessed business-related information. The preprocessed business-related information is provided as input, and important keywords and phrases are extracted and clustered using TF-IDF and Word2Vec techniques. As output, the extracted important keywords and phrases are stored in a database. For example, the server uses the "TF-IDF" library to identify high-frequency keywords from email content.

[0990] Step 4: Ask a question

[0991] A user inputs a question through a dedicated interface (e.g., a chatbot). The question text from the user is entered into the terminal as input and sent to the server via the terminal as output. For example, a user might input into the terminal, "What points should I pay attention to when planning a new product?"

[0992] Step 5: Parsing the Question

[0993] The server analyzes the received questions using natural language processing tools. The question text sent to the server is used as input, and the intent and keywords of the question are identified as output. For example, the server uses the "spaCy" library to extract key keywords from the question text.

[0994] Step 6: Know-how search and response generation

[0995] The server searches for relevant information from a know-how database based on the analyzed question content and generates a response to the question using a generative artificial intelligence model. The intent of the question and keywords are used as input, and a specific response text is generated as output. For example, the server uses ChatGPT to generate a response such as, "When planning a new product, it is important to start with market research and conduct a competitive analysis."

[0996] Step 7: Providing a response

[0997] The server sends the generated response back to the user via the terminal. The generated response text is used as input and displayed on the user's interface as output. For example, the terminal receives the response text from the server and displays it on the chatbot screen.

[0998] Through these steps, the system can effectively collect and analyze know-how within the company and provide appropriate information to young employees.

[0999] (Application example 1)

[1000] 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."

[1001] The knowledge and experience of senior employees within a company are valuable assets, but the loss of know-how due to retirement and other reasons is an issue. Furthermore, there is a lack of systems that allow younger employees to quickly and effectively acquire that know-how. Factories, in particular, need systems that can provide immediate and appropriate instructions and advice. Unless this issue is resolved, the company's competitiveness may decline.

[1002] 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.

[1003] In this invention, the server includes: means for periodically collecting emails and documents generated within the company; means for preprocessing the collected emails and documents to extract business-related information; means for analyzing the preprocessed business-related information, extracting experience data, and storing the extracted experience data in a database; means for accepting questions from users through a user interface; means for analyzing the accepted questions and searching for related information from an experience database; means for generating responses to the questions using a generative artificial intelligence model based on the search results; means for returning the generated responses to the users; and means for being installed in a robot installed in a factory to support young engineers and operators in sharing the know-how of senior engineers. This makes it possible to effectively pass on the know-how of senior employees to junior employees and to provide immediate and appropriate instructions and advice, particularly on the factory floor.

[1004] "Intra-company" refers to the inside of a particular company or organization, and is the scope related to the operation and business of that organization.

[1005] "Mail" refers to electronic mail, a message containing text and attachments sent and received over the Internet.

[1006] "Documents" refers to any paper or electronic document created, stored, and shared within an organization, including reports, presentations, memoranda, etc.

[1007] "Preprocessing" refers to the process of removing unnecessary information and reformatting the format of raw data to make it easier to analyze.

[1008] "Business-related information" refers to information that is directly related to a company's business operations and decision-making, including project progress reports and performance reports.

[1009] "Analysis" is the process of analyzing data in detail and finding useful information and patterns within it.

[1010] "Experience" refers to the knowledge, skills, and know-how acquired in the past, which are required to carry out specific tasks or work.

[1011] A "database" is a system that stores and manages information in an orderly manner, making it easy to search and use.

[1012] "User interface" refers to the means and screen display through which a user interacts with a system or application.

[1013] A "generative artificial intelligence model" refers to artificial intelligence technology that automatically generates responses and sentences in response to input data or questions, and includes natural language processing technology.

[1014] A "factory" is a facility where products are manufactured or assembled, and where industrial machinery and production lines are installed.

[1015] A "robot" is an automated machine or device designed to assist or replace human tasks.

[1016] "Young engineers" refer to relatively new employees within a company who are honing their expertise and skills in technical fields.

[1017] "Operator" refers to a staff member responsible for operating and managing a machine or system.

[1018] "Senior engineers" refer to technical leaders and veteran employees within a company who have extensive experience and advanced expertise.

[1019] The present invention provides a system for efficiently passing on the know-how of senior employees to junior employees in a company. This system is composed of the following means.

[1020] First, the server periodically collects emails and documents generated within the company. Emails are collected using the IMAP or POP3 protocol to retrieve unread emails, and documents are collected using the FTP or SMB protocol to retrieve new documents from specified folders.

[1021] The server then pre-processes the collected emails and documents to extract business-related information. This pre-processing removes spam and unimportant information, leaving only information relevant to the company's business. This process involves filtering out unnecessary information using regular expressions and natural language processing techniques.

[1022] The preprocessed business-related information is then analyzed by the server to extract experiences. Text analysis tools such as TF-IDF (Term Frequency-Inverse Document Frequency) and Word2Vec are used for the analysis. These tools are used to analyze the content of emails and documents and extract important keywords and phrases. Based on the results of this analysis, the experience data is stored in a database.

[1023] Users input questions through a dedicated interface (for example, through a chatbot or a robot installed in a factory). When a user inputs a question from a terminal, the question is sent from the terminal to a server. The server analyzes the received question using natural language processing tools to identify the intent of the question and important keywords.

[1024] Based on the analyzed question, the server searches for relevant information from an experience database. Based on the search results, the server generates a response to the question using a generative AI model (e.g., ChatGPT). For example, if a junior engineer asks, "Please tell me how to proceed with a new project," the server searches for experience information about past projects and uses the generative AI model to generate a specific approach based on that information.

[1025] The generated responses are sent from the server to the user via the terminal, allowing the user to receive specific and practical answers through the interface. Particularly in factory settings, robots can provide young engineers and operators with appropriate instructions and advice in real time.

[1026] As a concrete example, the prompt sentence is shown below.

[1027] Example prompt:

[1028] question: "How do I proceed with a new project?"

[1029] context: "Search for know-how information from past projects and explain the specific steps you took based on that information. Include market research methods, competitive analysis, and promotion strategies."

[1030] This system allows senior employees to effectively pass on their know-how to younger employees, and makes it possible to provide immediate and appropriate instructions and advice, particularly on factory floors, thereby preserving knowledge assets within the company and improving operational efficiency.

[1031] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1032] Step 1:

[1033] The server periodically accesses the company's mail server and file server to retrieve new emails and documents. The input here is the configuration information for the mail server and file server, and the output is the newly retrieved emails and documents. Specifically, it retrieves unread emails from the mail server using the IMAP or POP3 protocol, and retrieves new documents from the file server using the FTP or SMB protocol.

[1034] Step 2:

[1035] The server preprocesses the emails and documents it receives. The input is the emails and documents received in step 1, and the output is refined data that contains only business-related information. Specifically, regular expressions are used to remove spam and unimportant information, and natural language processing techniques are used to extract business-related information.

[1036] Step 3:

[1037] The server analyzes the preprocessed business-related information, extracts experiences, and stores them in a database. The input is the result data from Step 2, and the output is experience data analyzed using TF-IDF and Word2Vec. Specifically, the TF-IDF vectorizer is used to calculate feature weights, and the Word2Vec model is used to learn the relationships between words in the data, thereby extracting important experience data and storing it in a database.

[1038] Step 4:

[1039] The user inputs a question through a dedicated interface. The input here is the user's question, and the output is the transmission of that question. Specifically, the user's question is received using a chatbot interface or a robot's voice input system.

[1040] Step 5:

[1041] The server analyzes the received question and searches the experience database for relevant information. The input is the user's question, and the output is the analyzed keywords and important phrases of the question, as well as related experience data. Specifically, it uses natural language processing tools to extract the intent and keywords of the question, and then searches the experience database based on them.

[1042] Step 6:

[1043] Based on the search results, the server uses a generative AI model to generate a response to the question. The input is the relevant experience data obtained in step 5, and the output is the generated response text. Specifically, a generative AI model such as ChatGPT is used to generate a specific response based on relevant information. For example, in response to a question such as "Please tell me how to proceed with a new project," the server incorporates experience data from past projects and suggests a specific approach.

[1044] Step 7:

[1045] The server returns the generated response to the user via the terminal. The input is the response generated in step 6, and the output is the response provided to the user. Specifically, the generated response is delivered to the user via the chatbot screen or the robot's voice output. This series of processes allows junior engineers and operators to share the know-how of senior engineers in real time.

[1046] 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.

[1047] MODE FOR CARRYING OUT THE INVENTION

[1048] The present invention provides a system for effectively passing on the know-how of senior employees to younger employees, and by incorporating a user emotion recognition function, it achieves more personalized responses.

[1049] Data collection

[1050] The server periodically accesses the company's mail and file servers to retrieve new emails and documents, using the IMAP or POP3 protocols for mail servers and the FTP or SMB protocols for file servers.

[1051] Data Preprocessing

[1052] The server pre-processes the emails and documents it receives. During this process, it filters out spam and irrelevant information and extracts only business-related information. Regular expressions and natural language processing techniques are used for pre-processing.

[1053] Data analysis and know-how extraction

[1054] The server then uses text analysis tools to analyze the preprocessed data and extract important keywords and phrases. The extracted information is then used to organize the company's know-how and store it in a database. Metadata (creation date and time, creator, related projects, etc.) is also stored.

[1055] Accepting and analyzing questions

[1056] The user inputs a question through a dedicated user interface (e.g., a chatbot). The device receives the user's question and sends it to a server. The server analyzes the question using natural language processing tools and extracts key keywords and intent.

[1057] Emotion recognition and know-how retrieval

[1058] The server uses an emotion engine to recognize the user's emotional state when analyzing the question. The emotion engine recognizes emotions by analyzing the user's input and other emotional indicators. Based on this emotional information, it searches for related information from a know-how database.

[1059] Response Generation

[1060] The server generates a response to the question using a generative AI model (e.g., ChatGPT) based on the emotion recognition results and know-how search results. By taking the emotion recognition results into account and adjusting the response content, it provides a more personalized answer.

[1061] Providing a response

[1062] The server returns the generated response to the user through the user interface, allowing the user to instantly receive a specific and actionable answer through the interface on their device.

[1063] Specific examples

[1064] For example, consider the case where a junior employee types in, "Please tell me about our strategy for entering a new market." The device receives this question and sends it to the server. The server analyzes the question and uses an emotion engine to recognize the user's emotions, such as nervousness or confusion. Based on this information, it searches a know-how database for information about market entry and uses a generative artificial intelligence model to generate a response regarding specific market research methods, competitive analysis, and promotion strategies. The generated response is then sent back to the user via the device.

[1065] This system enables senior employees to pass on their know-how to junior employees in an efficient and personalized manner. The addition of emotion recognition functionality enables responses that take into account the user's emotional state, resulting in more effective problem-solving.

[1066] The processing flow will be explained below.

[1067] Step 1:

[1068] The server periodically accesses the company's mail server and file server, using the IMAP or POP3 protocol to retrieve unread emails from the mail server, and the FTP or SMB protocol to connect to the file server and retrieve new documents from specified folders.

[1069] Step 2:

[1070] The server pre-processes the emails and documents it receives. Specifically, it filters out spam and irrelevant information and extracts only business-related information. This pre-processing uses regular expressions and natural language processing techniques.

[1071] Step 3:

[1072] The server analyzes the preprocessed email and document data using text analysis tools, such as TF-IDF (Term Frequency-Inverse Document Frequency) and Word2Vec, to extract important keywords and phrases.

[1073] Step 4:

[1074] Based on the analysis results, the server organizes the company's know-how and stores it as structured data in a know-how database. The stored know-how information also includes related metadata (creation date and time, creator, related projects, etc.).

[1075] Step 5:

[1076] The user inputs a question through a dedicated user interface (e.g., a chatbot). The terminal receives the user's input and sends the question to the server.

[1077] Step 6:

[1078] The server analyzes the received question using natural language processing tools, extracting the intent of the question and key keywords, and providing the analysis results.

[1079] Step 7:

[1080] The server uses an emotion engine to analyze the emotion contained in the user's question. The emotion engine uses natural language processing technology and machine learning algorithms to recognize emotions from the user's input. For example, it analyzes whether the user is confused or nervous based on the question.

[1081] Step 8:

[1082] The server searches for relevant information based on keywords extracted from the know-how database and the intent of the question, and identifies the necessary data based on similar questions from the past and related documents.

[1083] Step 9:

[1084] The server generates a response to the question using a generative AI model (e.g., ChatGPT) based on the search results and the user's emotion recognition results. By taking the emotion recognition results into account and adjusting the response content, it generates the optimal answer for the user.

[1085] Step 10:

[1086] The server returns the generated response to the user through the user interface, and the user can receive a specific and practical answer through the interface on the terminal.

[1087] Specific examples

[1088] For example, consider the case where a junior employee types, "Please tell me about our strategy for entering a new market." The device receives this question and sends it to the server. The server analyzes the question and uses an emotion engine to recognize that the user is confused. Based on this information, it searches a know-how database for information about market entry and uses a generative artificial intelligence model to generate a response on specific market research methods, competitive analysis, and promotion strategies. The generated response is then sent back to the user via the device. In this way, a response that takes the user's emotional state into account is provided, enabling more effective problem solving.

[1089] Example 2

[1090] 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."

[1091] Efficiently transferring the know-how of senior employees to junior employees is an important issue for many companies. However, many current systems lack emotion recognition capabilities, respond uniformly, and are unable to provide personalized responses according to the emotional state of each user, making it difficult to achieve effective problem solving.

[1092] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes: means for periodically collecting communications and documents generated within the company; means for preprocessing the collected communications and documents to extract business-related information; means for analyzing the preprocessed business-related information, extracting knowledge, and storing it in a database; means for accepting inquiries from users through a user interface; means for analyzing the accepted inquiries and searching for related information from a knowledge database; means for generating a response to the inquiry using a generation system based on the search results; means for returning the generated response to the user; and means for having an emotion engine that recognizes the user's emotional state and adjusting the response content based on the recognized emotion information. This enables senior employees to pass on their know-how to junior employees in an efficient and personalized manner, and enables responses that take the user's emotional state into consideration, thereby achieving more effective problem solving.

[1093] "Communications" refers to the means of information exchange generated within a company, such as emails and chat messages.

[1094] "Document" refers to written or electronic records of information generated within an enterprise, such as reports, memos, and presentations.

[1095] "Business-related information" refers to information and data related to a company's business activities, including information on project management, risk management, and market research.

[1096] "Knowledge" refers to useful information and know-how extracted from business-related information and used to solve specific business problems.

[1097] A "database" refers to a collection of information that systematically organizes knowledge and makes it possible to store and search it.

[1098] "User interface" refers to the means by which a user interacts with a system, including graphical user interfaces (GUIs) and chatbots.

[1099] A "generative system" refers to a system that automatically generates responses to user inquiries using natural language processing technology or generative AI models.

[1100] An "emotion engine" refers to a function that analyzes a user's input and other emotional indicators to recognize their emotional state.

[1101] "Preprocessing" refers to a series of processes that filter spam and irrelevant information from collected communications and documents and extract only business-related information.

[1102] "Analysis" refers to the process of analyzing pre-processed business-related information in detail and extracting important keywords and phrases.

[1103] The present invention provides a system that effectively transfers the know-how of senior employees to junior employees while incorporating a function for recognizing user emotions to provide more personalized responses. A specific implementation method of this system will be described below.

[1104] First, the server, which is the core of this system, periodically collects communications and documents generated within the company. The server uses IMAP (Internet Message Access Protocol) and POP3 (Post Office Protocol 3) protocols to retrieve emails, and FTP (File Transfer Protocol) and SMB (Server Message Block) protocols to communicate with the file server. This process is performed automatically and periodically using CRON job settings.

[1105] The collected communications and documents are pre-processed by the server. In this pre-processing step, spam emails and irrelevant information are filtered out, and only business-related information is extracted. Regular expressions and natural language processing techniques (e.g., NLTK, SpaCy) are used in this process. For example, only emails containing keywords such as "important notice" or "weekly report" are extracted.

[1106] The preprocessed data is then subjected to detailed analysis using data analysis tools. The server tokenizes the information and performs morphological analysis to extract frequently occurring keywords and phrases. Important phrases, such as "project management" and "risk management," are automatically tagged and this information is stored in a knowledge database. Additionally, metadata such as creation date, creator, and related projects are also stored.

[1107] When a user inputs a question through a dedicated user interface (e.g., a chatbot), the device receives the question and sends it to a server. The server then analyzes the received question using natural language processing technology to extract key keywords and intent. Morphological analysis and parse tree analysis are used to analyze the user's text.

[1108] Furthermore, the server uses an emotion engine (e.g., EmotionAPI) to recognize the user's emotional state. The emotion engine recognizes emotions by analyzing the user's input and facial expression recognition using facial recognition technology. For example, if a user inputs "I'm very troubled," the emotion engine will recognize emotions such as "confusion" and "tension." Based on this recognition result, it searches for related information from a knowledge database.

[1109] Finally, the server generates a response using a generative AI model (e.g., ChatGPT) based on the emotional information and search results. The server inputs a prompt into the model, such as "Please explain your strategy for entering a new market to a nervous junior employee," to generate an appropriate response. The generated response is then sent back to the user via their device.

[1110] Specific examples

[1111] For example, consider the case where a junior employee types, "Please tell me about our strategy for entering a new market." The device receives this question and sends it to the server. The server analyzes the question and uses an emotion engine to recognize the user's emotions, such as nervousness or confusion. Based on this information, it searches a know-how database for information about market entry, and uses a generative artificial intelligence model to generate a response on specific market research methods, competitive analysis, and promotion strategies. The generated response is sent back to the user via the device. For example, it provides a detailed explanation in the form of, "To enter a market, we first need to research the market. Next, we conduct a competitive analysis and develop a promotion strategy. Specifically..."

[1112] Prompt Sentence Examples

[1113] "Please explain to a nervous junior employee the strategy for entering a new market."

[1114] "Explain project management best practices to struggling users."

[1115] "Advise a confused employee on key customer service points."

[1116] The system allows senior employees to share their experience and knowledge with junior employees in an efficient and personalized way, and by taking into account the user's emotional state, it can provide more appropriate and useful advice.

[1117] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1118] Step 1:

[1119] The server accesses the company's mail and file servers to retrieve new communications and documents. As input, the addresses and authentication information for the mail and file servers are required. Specifically, the server retrieves emails using the IMAP or POP3 protocol and downloads documents from the file server using the FTP or SMB protocol. As output, the new communications and documents are stored on the server.

[1120] Step 2:

[1121] The server preprocesses the communications and documents it receives. The emails and documents obtained in step 1 are required as input. Specifically, the server filters out spam and removes irrelevant information, extracting only business-related information. Regular expressions and natural language processing tools (e.g., NLTK, SpaCy) are used in this process. The output is the preprocessed, clean data.

[1122] Step 3:

[1123] The server analyzes the preprocessed data and extracts important keywords and phrases. It requires the clean data from step 2 as input. Specifically, the server uses text analysis tools to tokenize the data and analyze it using morphological analysis and parse trees. The output is the extracted keywords and phrases, along with their associated metadata (creation date, author, related projects, etc.).

[1124] Step 4:

[1125] The server stores the analyzed information in a knowledge database. The inputs required are the keywords and phrases obtained in step 3, as well as metadata. Specifically, the server stores this information in a relational or NoSQL database (e.g., MongoDB, PostgreSQL) and organizes it for efficient searches. The output is an updated knowledge database.

[1126] Step 5:

[1127] A user inputs a question through a dedicated user interface (e.g., a chatbot). The user's question is acquired as input through the user interface. Specifically, this role is played by a chatbot or desktop application that runs on a web browser. The user's question is sent to a server as output.

[1128] Step 6:

[1129] The server analyzes the user's question and extracts key keywords and intent. The input required is the user's question obtained in step 5. Specifically, the question is tokenized using natural language processing techniques and analyzed using morphological analysis and parse tree. The output is the extracted keywords and phrases.

[1130] Step 7:

[1131] The server uses an emotion engine to recognize the user's emotional state. The input required is the analysis result of the question obtained in step 6. Specifically, the emotion engine (e.g., EmotionAPI) is used to analyze the user's input and emotion indicators using facial expression recognition technology. The output is the recognized user's emotional information.

[1132] Step 8:

[1133] The server searches for relevant information from the knowledge database. The inputs required are the keywords and phrases obtained in step 6 and the emotion information obtained in step 7. Specifically, the server executes a search query on the knowledge database to obtain relevant know-how information. The output is the search results.

[1134] Step 9:

[1135] The server generates a response using a generative AI model (e.g., ChatGPT). The inputs required are the search results obtained in step 8 and the emotion information obtained in step 7. Specifically, the server inputs a prompt sentence into the generative AI model to generate an appropriate response. The generated response is obtained as the output.

[1136] Step 10:

[1137] The server returns the generated response to the user through a user interface. The generated response obtained in step 9 is required as input. Specifically, the server displays the response to the user through a chatbot or an on-screen message box. As output, the user receives a specific and actionable answer.

[1138] (Application example 2)

[1139] 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."

[1140] Companies are faced with the need to efficiently process the work-related information contained in the large volume of emails and documents generated within their organizations, and to effectively pass on the know-how of senior employees to junior employees. However, when junior employees ask specific questions, traditional systems struggle to provide personalized responses, and they lack an understanding of the emotional aspects of the situation, resulting in a lack of effective support. This can result in junior employees taking a long time to resolve their issues.

[1141] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1142] In this invention, the server includes means for periodically collecting emails and documents generated within the company, means for preprocessing the collected emails and documents to extract business-related information, means for analyzing the preprocessed business-related information, extracting know-how and storing it in a database, means for accepting questions from users through a user interface, means for analyzing the accepted questions and searching for related information from a know-how database, means for generating responses to the questions using a generative artificial intelligence model based on the search results, means for returning the generated responses to the users, means for recognizing the emotional state of the user when analyzing the questions, and means for adjusting the content of the responses based on the emotional information. This allows senior employees to pass on their know-how to junior employees in an efficient and personalized manner, and the emotion recognition function enables more effective problem solving.

[1143] "Means for regularly collecting emails and documents generated within a company" is a function for automatically obtaining emails and document files created within an organization at specified intervals.

[1144] "Means of preprocessing collected emails and documents to extract business-related information" refers to a function for removing unnecessary data from acquired emails and document files and selecting important business-related information.

[1145] "Means for analyzing preprocessed business-related information, extracting know-how, and storing it in a database" refers to a function for further analyzing preprocessed data and organizing and storing knowledge and experience that are useful for business activities.

[1146] The "means for accepting questions from users through a user interface" is a function that provides an interface for users to send questions to the system.

[1147] The "means for analyzing a received question and searching for related information from a know-how database" is a function for analyzing the content of a question from a user and searching for related information from a saved know-how database.

[1148] "Means for generating responses to questions using a generative artificial intelligence model based on search results" refers to a function that uses an AI model to generate specific responses based on the related information found.

[1149] The "means for returning the generated response to the user" is a function for sending the generated response to the user via the user interface.

[1150] The "means for recognizing the emotional state of the user when analyzing a question" is a function for analyzing the emotions of the user when asking a question and recognizing that state.

[1151] The "means for adjusting the content of a response based on emotional information" is a function for adjusting the content of a response to be generated based on the recognized emotional information.

[1152] The present invention provides a system for passing on the know-how of senior employees to junior employees in a factory, and incorporates an emotion recognition function. Detailed embodiments of the system will be described below.

[1153] Hardware and software used

[1154] 1. Hardware:

[1155] Factory robots: Used to answer questions from junior employees and provide know-how from senior employees.

[1156] Smartphone or tablet device (optional): Can be used by junior employees to enter questions and receive answers.

[1157] 2. Software:

[1158] Natural language processing tools (e.g., NLTK, spaCy): Used for text analysis of emails and documents.

[1159] Emotion recognition engine (e.g., Microsoft Azure Cognitive Services): Analyzes and recognizes the user's emotional state.

[1160] Generative AI models (e.g., ChatGPT): Generate answers to questions.

[1161] System Operation

[1162] 1. Data Collection:

[1163] The server uses the IMAP protocol for the company's mail server and the FTP protocol for the file server to periodically retrieve emails and documents generated by senior employees.

[1164] 2. Data Preprocessing:

[1165] The server preprocesses the acquired data using natural language processing tools, filtering out spam and irrelevant information and extracting only business-related information.

[1166] 3. Data analysis and know-how extraction:

[1167] The preprocessed data is analyzed to extract important keywords and phrases, which are then used to organize the company's know-how and store it in a database along with metadata (creation date, creator, related projects, etc.).

[1168] 4. Question acceptance and analysis:

[1169] Users (young employees) input questions through the factory robot's interface. The input questions are analyzed using natural language processing tools to extract key keywords and intent.

[1170] 5. Emotion recognition and know-how retrieval:

[1171] The server uses an emotion recognition engine to recognize the user's emotional state when analyzing the question, and based on this emotional information, searches for related information from a know-how database.

[1172] 6. Response Generation:

[1173] Based on the analyzed question content and emotion recognition results, the server uses a generative artificial intelligence model (ChatGPT) to generate an appropriate response and adjust the response content.

[1174] 7. Providing a Response:

[1175] The generated responses are then sent back to junior employees through the factory robot's interface, allowing users to receive instant, specific, and emotionally sensitive answers.

[1176] Specific examples

[1177] Consider a case where a junior employee asks, "How do I troubleshoot a particular machine?" A factory robot receives this question and sends it to a server. The server analyzes the question and uses an emotion recognition engine to identify the user's emotions, such as urgency. Based on this information, it searches for relevant information in a know-how database and generates a specific troubleshooting method using ChatGPT. The generated response is then sent back to the junior employee through the robot's interface.

[1178] Prompt Sentence Examples

[1179] A junior employee asks:

[1180] How do I troubleshoot a specific machine?

[1181] The server analyzes the questions, recognizes emotions, and generates a response using a generative AI model. This system ensures that senior employees' know-how is passed on in an efficient and personalized way, resulting in more effective problem-solving.

[1182] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1183] Step 1: Data collection

[1184] The server accesses the company's mail server using the IMAP protocol to periodically retrieve emails sent and received by senior employees. It also accesses the file server using the FTP protocol to periodically collect document files. These emails and documents are imported into the system's initial database. The input is the emails and documents generated within the company, and the output is the collected raw data.

[1185] Step 2: Data Preprocessing

[1186] The server preprocesses the collected emails and document files. This includes spam filtering and the removal of irrelevant information. It uses natural language processing tools (such as NLTK and spaCy) to analyze the text content of each email and document and extract business-related information. The input is the collected raw data, and the output is the preprocessed meaningful data. Specific operations include filtering out spam emails and extracting important keywords and phrases.

[1187] Step 3: Data analysis and know-how extraction

[1188] The server further analyzes the preprocessed data. It uses natural language processing tools to detect important keywords and phrases and extracts know-how based on these. The extracted know-how information is stored in a database along with metadata (creation date and time, creator, related projects, etc.). The input is the preprocessed meaningful data, and the output is the know-how information stored in the database. Specifically, it performs morphological analysis to organize and classify highly relevant information.

[1189] Step 4: Receive and analyze questions

[1190] The user inputs a question through the factory robot's interface. The terminal sends the question to the server. The server analyzes the received question using a natural language processing tool and extracts key keywords and intent. The input is the user's question, and the output is the analyzed question content. Specifically, the natural language processing tool is used to perform morphological analysis of the question and understand the intent.

[1191] Step 5: Emotion recognition and know-how retrieval

[1192] The server uses an emotion recognition engine (such as Microsoft Azure Cognitive Services) to recognize the user's emotional state when analyzing the question. Based on this emotional information, it searches for related information from a know-how database. The input is the analyzed question content and the user's emotional information, and the output is the search results. Specifically, it combines natural language processing and emotion analysis to optimize the search query.

[1193] Step 6: Response Generation

[1194] The server generates a response to the question using a generative AI model (ChatGPT) based on the emotion recognition results and know-how search results. At this time, it adjusts the response content taking into account the emotion information. The input is the search results and emotion information, and the output is the generated response. Specifically, it inputs a prompt into the generative AI model to obtain a personalized answer.

[1195] Step 7: Providing a response

[1196] The server sends the generated response back to the user through the factory robot's interface. The user can instantly receive a specific and actionable answer through the interface on their terminal. The input is the generated response, and the output is the answer provided to the user. Specific operations include displaying the response on the robot's display and providing a voice response using a speech synthesis function.

[1197] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1198] 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.

[1199] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1200] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1201] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1202] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1203] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1204] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1205] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1206] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1207] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1208] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1209] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1210] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1211] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1212] The hardware resource for executing a specific process can be any of the following types of processors: A processor, for example, is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1213] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1214] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1215] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1216] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1217] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1218] The following is further disclosed regarding the above embodiment.

[1219] (Claim 1)

[1220] [Means of regularly collecting emails and documents generated within the company;

[1221] [Means of pre-processing collected emails and documents to extract business-related information;

[1222] [Means for analyzing the preprocessed business-related information, extracting know-how and storing it in a database;

[1223] [means for accepting questions from a user through a user interface;

[1224] [Means for analyzing received questions and searching for related information from a know-how database;

[1225] [Means for generating a response to the question based on the search results using a generative artificial intelligence model; and

[1226] [means for returning the generated response to the user; and

[1227] A system including:

[1228] (Claim 2)

[1229] [The system of claim 1, wherein the preprocessed business-related information is analyzed using a text analysis tool.

[1230] (Claim 3)

[1231] [The system of claim 1, wherein the generative artificial intelligence model learns know-how information extracted from past emails and documents.

[1232] "Example 1"

[1233] (Claim 1)

[1234] [Means for periodically collecting messages and documents generated within the enterprise;

[1235] [Means for pre-processing the collected messages and documents to filter unnecessary information and extract only business-related information;

[1236] [Means for analyzing the preprocessed business-related information, extracting important keywords and phrases, and storing them in a database;

[1237] [means for accepting questions from users through a user interface;

[1238] [Means of analyzing received questions using natural language processing tools to identify the intent and keywords of the questions,

[1239] [means for searching for relevant information from a know-how database based on the analyzed question and generating a response to the question using a generative artificial intelligence model;

[1240] [means for returning the generated response to the user; and

[1241] A system including:

[1242] (Claim 2)

[1243] [The system of claim 1, wherein the preprocessed business-related information is analyzed using text analysis techniques.

[1244] (Claim 3)

[1245] The system of claim 1, wherein the generative artificial intelligence model learns know-how information extracted from past messages and documents.

[1246] "Application Example 1"

[1247] New Claims

[1248] (Claim 1)

[1249] [Means of regularly collecting emails and documents generated within the company;

[1250] [Means for preprocessing collected emails and documents to extract business-related information;

[1251] [Means for analyzing the preprocessed business-related information, extracting experiences, and storing them in a database;

[1252] [means for accepting questions from users through a user interface;

[1253] [Means for analyzing received questions and searching for relevant information from an experience database;

[1254] [means for generating a response to the question using a generative artificial intelligence model based on the search results; and

[1255] [means for returning the generated response to the user; and

[1256] [It will be installed on robots installed in factories, and will be a means to help junior engineers and operators share the know-how of senior engineers.

[1257] A system including:

[1258] (Claim 2)

[1259] [The system of claim 1, wherein the preprocessed business-related information is analyzed using a text analysis tool.

[1260] (Claim 3)

[1261] [The system of claim 1, wherein the generative artificial intelligence model learns empirical information extracted from past emails and documents.

[1262] "Example 2: Combining Emotion Engines"

[1263] (Claim 1)

[1264] [Means of regular collection of correspondence and documents generated within the enterprise;

[1265] [Means for pre-processing the collected communications and documents to extract business-related information;

[1266] [Means for analyzing the preprocessed business-related information, extracting knowledge, and storing it in a database;

[1267] [Means for accepting inquiries from users through a user interface;

[1268] [Means for analyzing received inquiries and searching for relevant information from a knowledge database;

[1269] [means for generating a response to the query using a generation system based on the search results;

[1270] [means for returning the generated response to the user; and

[1271] [having an emotion engine that recognizes the user's emotional state and means for adjusting response content based on the recognized emotion information;

[1272] A system including:

[1273] (Claim 2)

[1274] [The system according to claim 1, wherein the preprocessed business-related information is analyzed using natural language processing technology.

[1275] (Claim 3)

[1276] The system of claim 1, wherein the generation system learns knowledge information extracted from past communications and documents.

[1277] "Application example 2 when combining emotion engines"

[1278] (Claim 1)

[1279] [Means of regularly collecting emails and documents generated within the company;

[1280] [Means of pre-processing collected emails and documents to extract business-related information;

[1281] [Means for analyzing the preprocessed business-related information, extracting know-how and storing it in a database;

[1282] [means for accepting questions from a user through a user interface;

[1283] [Means for analyzing received questions and searching for related information from a know-how database;

[1284] [Means for generating a response to the question based on the search results using a generative artificial intelligence model; and

[1285] [means for returning the generated response to the user; and

[1286] [Means for recognizing the emotional state of the user when parsing the question;

[1287] [Means for adjusting response content based on emotional information;

[1288] A system including:

[1289] (Claim 2)

[1290] [The system of claim 1, wherein the preprocessed business-related information is analyzed using a text analysis tool.

[1291] (Claim 3)

[1292] [The system of claim 1, wherein the generative artificial intelligence model learns know-how information extracted from past emails and documents. [Explanation of symbols]

[1293] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of regularly collecting emails and documents generated within the company; A means of pre-processing collected emails and documents to extract business-related information; A means for analyzing the preprocessed business-related information, extracting know-how, and storing it in a database; means for accepting questions from a user through a user interface; A means for analyzing received questions and searching for related information from a know-how database; a means for generating a response to the question based on the search results using a generative artificial intelligence model; and means for returning the generated response to the user; A system including:

2. The system of claim 1 , wherein the preprocessed business-related information is analyzed using a text analysis tool.

3. 10. The system of claim 1, wherein the generative artificial intelligence model learns know-how information extracted from past emails and documents.

Images