System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting information about an enterprise's support system; means for formatting and pre-processing the information; means for having a generative model that learns based on the information; means for accepting a user's question; means for analyzing the question; means for retrieving relevant information; means for generating an answer based on the retrieved information; and means for displaying the generated answer to the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Employees do not fully understand their company's systems and support programs, which results in them being unaware of the support and assistance they are entitled to and not being able to use it. Furthermore, companies are also unable to maximize the effectiveness of their systems because employees do not fully utilize them. To solve this problem, there is a need for technology that enables employees to properly understand and effectively use all of the company's support and assistance programs. [Means for solving the problem]

[0005] The present invention includes a system that collects information about corporate support programs, formats and preprocesses it, and then trains it using a generative model. When a user (employee) inputs a question, the system analyzes the question and searches for information on relevant corporate programs and support programs. The system then generates a detailed answer based on the searched information and displays the answer to the user, thereby quickly and accurately providing the support information the user needs. The system analyzes questions using natural language processing and provides information through an online portal, enabling employees to properly understand and effectively use corporate programs.

[0006] "Corporate support systems" is a general term for various subsidies and support programs that companies provide to their employees, including employee benefits, health management, special leave, and medical expense subsidies.

[0007] "Means of collecting information" refers to the functions and methods for collecting documents and data related to a company's support system.

[0008] "Information formatting and preprocessing means" refers to processes to enable effective learning by the generative model, such as formatting collected information and removing unnecessary data.

[0009] A "generative model" refers to a machine learning model that efficiently learns from large amounts of data and generates answers to questions. Examples include generative AI models (e.g., ChatGPT).

[0010] "Means for accepting questions" refers to the interface or method by which users input questions about a company's support system.

[0011] "Question analysis means" refers to a function that uses natural language processing technology to analyze input questions and understand their intent.

[0012] "Means for searching relevant information" refers to the function of searching for appropriate information from documents and databases related to corporate support programs based on the analyzed questions.

[0013] "Answer generation means" refers to a function for creating a specific answer for the user based on the searched information.

[0014] "Answer display means" refers to a method or interface for displaying the generated answers so that the user can check them.

[0015] "Online Portal" means a website or application that users access via the Internet and that allows them to enter and display information. [Brief explanation of the drawings]

[0016] [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

[0017] 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.

[0018] First, the terms used in the following description will be explained.

[0019] 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).

[0020] 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.

[0021] 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.

[0022] 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.

[0023] 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."

[0024] [First embodiment]

[0025] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0026] 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.

[0027] 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).

[0028] 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.

[0029] 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.

[0030] 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.

[0031] 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.

[0032] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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."

[0037] This system collects and learns information about corporate support systems and provides appropriate answers to user questions. This system is mainly composed of three elements: a server, a terminal, and a user.

[0038] Data collection and learning

[0039] First, the server collects documents from company administrators about the company's support systems. These documents include employee benefit programs, special leave policies, medical expense subsidies, etc. The collected documents are formatted and preprocessed before being input into a generative AI (e.g., ChatGPT). The server uses this generative AI to learn information and prepare it to generate answers to questions.

[0040] Accepting user questions

[0041] Next, users access the Smart Corporate System Navigator through their company's online portal or dedicated application, and enter a specific question in the search box, such as, "What kind of support is available if I get the flu?"

[0042] Question analysis and answer generation

[0043] The device sends the entered question to the server, which uses generative AI to analyze the question and understand its intent. Based on the results of the analysis, the server searches for relevant information in the company's documents. For example, if the question contains the keyword "influenza," the search will return "health management support documents" and the like.

[0044] Based on the retrieved information, the server generates a detailed answer, such as, "If employees catch the flu, they can use special paid leave. Medical expense assistance programs are also available."

[0045] Show Answers

[0046] Finally, the server sends the generated answer to the terminal, which then displays the answer to the user. The user can view the displayed answer, understand the support or assistance they need, and use it appropriately.

[0047] Specific examples

[0048] Example 1: Healthcare support inquiry

[0049] Users access an online portal and type in the question, "What help is available if I have the flu?"

[0050] The terminal sends this question to the server.

[0051] The server uses generative AI to analyze the question and search documents for information related to "influenza."

[0052] The server generates a response that "employees can take special paid leave and also have access to medical expense assistance."

[0053] The terminal displays the generated answer, and the user understands the necessary support information and uses it appropriately.

[0054] Example 2: Benefits Program Inquiry

[0055] The user types the question, "What support is available for childcare leave?"

[0056] The terminal sends this question to the server.

[0057] The server analyzes the query and retrieves the relevant assistance program documentation.

[0058] The server generates a response saying, "If you take child care leave, you will be granted special leave and some medical expenses will be subsidized."

[0059] The terminal displays the generated answer, and the user uses the corresponding support information.

[0060] In this way, the present invention provides a system that efficiently collects and learns information about corporate support systems and provides quick and accurate answers to user questions.

[0061] The processing flow will be explained below.

[0062] Step 1: Data entry and learning phase

[0063] The server receives documents related to support systems from company administrators. Specifically, it collects "support documents related to health management," "personnel system guidebooks," "details of employee benefit programs," etc.

[0064] The server formats the documents it receives and performs data cleaning if necessary, which includes standardizing the document format and removing unnecessary textual information.

[0065] The server feeds the formatted documents into a generative model (e.g., ChatGPT) and trains the model, which then has detailed knowledge of the company's systems.

[0066] Step 2: Accepting questions from users

[0067] Users access the Smart Corporate System Navigator through their company's online portal or application.

[0068] The user enters a specific question into the question entry form on the portal site, for example, "What kind of support is available if I get the flu?"

[0069] The terminal (user's device) sends the entered question data to the server.

[0070] Step 3: Parsing the Question

[0071] The server analyzes the received question data. Using a generative AI model, it understands the intent of the question and extracts keywords and important phrases. For example, "influenza" and "support" are extracted as important keywords.

[0072] Step 4: Find related information

[0073] The server uses the extracted keywords to search for relevant information from related documents, such as "health management support documents" and "personnel system guidebooks."

[0074] The server sorts through the search results and extracts the most relevant parts.

[0075] Step 5: Generate an answer

[0076] The server generates answers to provide to users based on the extracted information. The generation AI summarizes the information and adjusts the writing style to create easy-to-understand sentences.

[0077] For example, it generates a specific answer such as "If employees catch the flu, they can use special paid leave. Medical expense assistance systems are also available."

[0078] Step 6: Submit and view your responses

[0079] The server sends the generated answer to the user's terminal.

[0080] The terminal displays the received response to the user, allowing the user to quickly ascertain details of the assistance or assistance required.

[0081] In this way, by performing specific processing at each step, a system is provided that allows users to efficiently obtain and utilize information about corporate systems and support programs.

[0082] Example 1

[0083] 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."

[0084] Information about corporate support programs is diverse, making it difficult for employees to quickly obtain the information they need. Furthermore, if the information provided is inaccurate, employees may not receive appropriate support, hindering efficient work performance. In addition, there is the problem that technology for accurately analyzing questions entered by users in natural language and generating appropriate answers is still immature. Furthermore, if the generated answers are not displayed clearly to users, user convenience is reduced.

[0085] 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.

[0086] In this invention, the server includes: means for collecting information about corporate support programs; means for formatting and preprocessing the information; means for having a generative model that learns based on the information; means for accepting user questions; means for analyzing the questions; means for searching for relevant information; means for generating answers based on the searched information; means for displaying the generated answers to the user; means for removing unnecessary parts and extracting only important information using a natural language processing tool when formatting and preprocessing the information; means for generating detailed answers to the user questions based on prompt sentences using a generative AI model; and means for displaying the answers generated based on the prompt sentences in a user-friendly format. This makes it possible to quickly and accurately obtain information about corporate support programs, generate appropriate answers to questions entered in natural language, and display the answers in an easy-to-understand format to the user.

[0087] "Means for collecting information" refers to devices and methods for receiving and storing data relating to the company's support system from the company administrator.

[0088] "Information formatting and preprocessing means" refers to devices and methods that use natural language processing tools to convert collected data into an appropriate format and remove unnecessary parts.

[0089] "Means having a learning generative model" refers to a device or method that uses a generative AI model to learn information based on preprocessed data.

[0090] The "means for accepting a user's question" refers to a device or method that provides a function for a user to input a question in natural language and receives the input content.

[0091] The "means for analyzing a question" refers to a device or method that uses natural language processing technology to analyze a question received from a user and understand its intent.

[0092] The "means for retrieving relevant information" refers to a device or method for retrieving relevant information from relevant company documents or databases based on the analyzed query.

[0093] A "means for generating an answer" is a device or method that uses a generative AI model to create a detailed answer based on the searched information.

[0094] The "means for displaying the generated answers to the user" refers to a device or method for displaying the generated answers in a user-friendly format so that the user can view them.

[0095] "Means using natural language processing tools" refers to the techniques and devices used to analyze and format text when preprocessing collected data.

[0096] A "means for generating a detailed answer based on a prompt sentence" is a device or method that uses a generative AI model to input a prompt corresponding to a specific question and generate a detailed answer based on that prompt.

[0097] The "means for displaying in a user-friendly format" refers to a device or method that provides the generated answers to the user in a format that is easy to view and understand.

[0098] This system collects and learns information about corporate support systems and provides appropriate answers to user questions. This system is mainly composed of three elements: a server, a terminal, and a user.

[0099] Data collection and learning

[0100] The server receives documents related to the company's support system from the company administrator. The received documents are saved as text files or PDFs. The saved documents are preprocessed using an NLP tool (e.g., SpaCy, NLTK). Preprocessing refers to removing unnecessary parts and extracting important information. The formatted data from this process is input into a generative AI model (e.g., ChatGPT). The server uses this generative AI model to learn information and prepare to generate answers to user questions.

[0101] Accepting user questions

[0102] Users log in to their company's online portal or dedicated application to access the "Smart Corporate System Navigator." They enter a specific question into the search box, for example, "What kind of support is available if I have the flu?" The device then sends this input to the server. The transmitted data is protected using a secure communication protocol (e.g., HTTPS).

[0103] Question analysis and answer generation

[0104] The server converts the received question into a prompt to be input into the generative AI. Using the generative AI model, the server analyzes the intent of the user's question and extracts related keywords. Based on the results of this analysis, the server searches for relevant information from the company's documents. For example, if the keyword "influenza" is included, support documents related to health management will be searched for. Based on the search results, the generative AI model is used to generate a detailed answer. An example of a specific answer would be: "If employees catch the flu, they can use special paid leave. Medical expense subsidy systems are also available."

[0105] Show Answers

[0106] The server sends the generated answer to the terminal, which displays the answer in a user-friendly format (e.g., a card or pop-up).The user can view the displayed answer, understand the support or assistance they need, and use it appropriately.

[0107] Specific examples

[0108] Example 1: Healthcare support inquiry

[0109] The user accesses the online portal and enters the question, "What support is available if I get the flu?" The device then sends this question to the server. The server uses generative AI to analyze the question and searches documents for information related to "flu." The server then generates the answer, "Employees can use special paid leave and medical expense subsidy systems." The user then understands the necessary support information and uses it appropriately.

[0110] Example 2: Benefits Program Inquiry

[0111] The user inputs the question, "What kind of support is available for child care leave?" The device sends this question to the server. The server analyzes the question and searches for documents on related support programs. The server generates an answer that says, "If you take child care leave, you will be granted special leave and some medical expenses will be subsidized." The device displays the generated answer, allowing the user to use the relevant support information appropriately.

[0112] This system can efficiently collect and learn information about corporate support systems and provide quick and accurate answers to users' questions.

[0113] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0114] Step 1:

[0115] The server receives documents relating to the company's support system from the company administrator.

[0116] Specifically, the document is collected using the file upload function.

[0117] Input: Support program documentation provided by company administrator.

[0118] Output: Saved support system documents.

[0119] Step 2:

[0120] The server saves the received documents in text file or PDF format.

[0121] Specifically, the uploaded file is saved in an appropriate folder or database.

[0122] Input: Uploaded assistance system document.

[0123] Output: Saved text file and PDF.

[0124] Step 3:

[0125] The server preprocesses the stored documents using natural language processing (NLP) tools (e.g., SpaCy, NLTK).

[0126] Specifically, it removes unnecessary parts from the document (e.g., page numbers, headers, footers) and extracts important information.

[0127] Input: Saved text files or PDFs.

[0128] Output: Preprocessed text data.

[0129] Step 4:

[0130] The server formats the preprocessed data and generates prompts to input into a generative AI model (e.g., ChatGPT).

[0131] Specifically, the process creates a prompt statement template based on the preprocessed data.

[0132] Input: Preprocessed text data.

[0133] Output: The prompt sentence to be input to the generative AI model.

[0134] Step 5:

[0135] The server uses a generative AI model to learn information and prepare to generate answers to user questions.

[0136] Specifically, the prompt sentence is input into the generative AI model and the model is trained.

[0137] Input: prompt statement.

[0138] Output: A trained generative AI model.

[0139] Step 6:

[0140] Users log in to the online portal or dedicated application and access the "Smart Corporate System Navigator."

[0141] Specifically, the user authentication and login process are carried out.

[0142] Input: User credentials.

[0143] Output: Authenticated user interface.

[0144] Step 7:

[0145] Users enter a specific question into a search box.

[0146] Specifically, you enter a question in the search box and press the send button.

[0147] Input: The question text provided by the user.

[0148] Output: The question data sent to the terminal.

[0149] Step 8:

[0150] The terminal transmits the entered question to the server.

[0151] Specifically, the input data is transferred to the server using a secure communication protocol (e.g., HTTPS).

[0152] Input: Question data from the user.

[0153] Output: The query data sent to the server.

[0154] Step 9:

[0155] The server converts the received question into a prompt sentence to be input to the generation AI.

[0156] Specifically, the question data is formatted into an appropriate prompt format.

[0157] Input: Question data from the user.

[0158] Output: The prompt statement.

[0159] Step 10:

[0160] The server uses generative AI to analyze the intent of the question and extract relevant keywords.

[0161] Specifically, the prompt sentence is input into the generative AI model, and intent analysis and keyword extraction are performed.

[0162] Input: prompt statement.

[0163] Output: Parsed intent and extracted keywords.

[0164] Step 11:

[0165] The server searches for company documents based on the extracted keywords.

[0166] Specifically, the operation involves searching for the relevant information from a document management system or database.

[0167] Input: Parsed intent and extracted keywords.

[0168] Output: Search results (relevant documents).

[0169] Step 12:

[0170] The server generates detailed answers based on the search results using a generative AI model.

[0171] Specifically, the search results are input into a generative AI model to generate a detailed answer.

[0172] Input: Search results.

[0173] Output: The detailed answer generated.

[0174] Step 13:

[0175] The server sends the generated response to the terminal.

[0176] Specifically, the generated answer is formatted into a user-friendly format and sent to the terminal.

[0177] Input: The generated detailed answer.

[0178] Output: The response data sent to the device.

[0179] Step 14:

[0180] The device displays the received answers in a user-friendly format (e.g., card format or popup).

[0181] Specifically, the response data is read and displayed on the user interface.

[0182] Input: Response data sent to the terminal.

[0183] Output: The answer displayed to the user.

[0184] Step 15:

[0185] The user can view the displayed answers, understand the support and assistance they need, and use it appropriately.

[0186] Specifically, the displayed information is checked and the next action is taken as necessary.

[0187] Input: The answer shown to the user.

[0188] Output: User understanding and use of information.

[0189] (Application example 1)

[0190] 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."

[0191] There is a wide variety of information available about corporate support programs, making it difficult for employees to quickly and accurately identify the program that best suits them. Furthermore, especially in busy environments, such as store employees, who often lack the time to use digital devices, employees tend to neglect using support programs. This leads to a decline in employee utilization of support programs, resulting in the inability to fully realize the benefits of corporate employee benefits.

[0192] 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.

[0193] In this invention, the server includes: means for collecting information on corporate support programs; means for formatting and preprocessing the information; means for having a generative model that learns based on the information; means for accepting user questions; means for analyzing the questions; means for searching for relevant information; means for generating answers based on the searched information; means for displaying the generated answers to the user; a voice input means installed in the smart device; means for converting voice to text; means for transmitting the text questions to the server; and means for displaying the generated answers on the smart device. This allows employees to easily ask questions through voice input and quickly access information on appropriate support programs.

[0194] "Corporate support systems" is a general term for welfare programs, special leave provisions, medical expense subsidies, etc. that companies provide to their employees.

[0195] "Means for collecting information" refers to a method or system for collecting documents related to the support system from company administrators.

[0196] "Information formatting and preprocessing means" refers to methods and systems for converting collected documents into a format that is easy to analyze and for preprocessing the data.

[0197] A "generative model" is a machine learning model that learns from collected and preprocessed information and generates appropriate answers to user questions.

[0198] The "means for accepting a user's question" is an interface for receiving a question input by a user and transmitting it to the system.

[0199] The "means for analyzing questions" refers to natural language processing techniques and algorithms for analyzing received questions and understanding their intent.

[0200] The "means for searching relevant information" is a method or system for searching company documents for information relevant to the analyzed question.

[0201] "Answer generation means" refers to a method or system for generating a specific answer based on the retrieved information.

[0202] The "means for displaying the generated answer to the user" refers to a method or system for displaying the generated answer on the user's device (terminal).

[0203] "Smart devices" is a general term for devices that can connect to the Internet and run various applications, such as smartphones, smart glasses, and head-mounted displays.

[0204] The "voice input means" is a function for inputting the user's voice using a microphone or the like installed on the smart device.

[0205] "Means for converting speech to text" refers to speech recognition technology and algorithms for analyzing input speech and converting it into text data.

[0206] The "means for transmitting a question converted into text to a server" is a communication function for transmitting a user's question converted into text by voice recognition to a server.

[0207] The "means for displaying the answer on the smart device" is a function for receiving the answer generated from the server and displaying it on the display screen of the smart device.

[0208] This system collects and learns information about corporate support programs and provides appropriate answers to user questions. This system is composed of three elements: a server, a terminal, and a user.

[0209] First, the server collects documents from company administrators about the company's support systems, including employee benefit programs, special leave policies, medical expense subsidies, etc. The collected documents undergo formatting and preprocessing, and then the generative AI learns the information and prepares them to generate answers to questions.

[0210] Next, the user accesses the company's smart device and asks a question through voice input, for example, "What kind of support is available if I have the flu?" The microphone installed on the smart device captures this voice and converts it into text using voice recognition technology.

[0211] The device sends the converted text data to the server, which uses generative AI to analyze the question and understand its intent. Based on the results of the analysis, the server searches for relevant information in the company's documents. For example, if a question contains the keyword "influenza," the search will return "health management support documents" and other relevant information.

[0212] Based on the searched information, the server generates a detailed answer. For example, it creates a specific answer such as, "If an employee catches the flu, they can use special paid leave. Medical expense subsidies are also available." The generated answer is sent to the smart device as text data and displayed on the screen. The user can view the displayed answer, understand the support and assistance they need, and use it appropriately.

[0213] Examples:

[0214] A user uses the smart glasses to ask a question out loud: "How much is the employee discount?" The question is converted into text using speech recognition, and the question "How much is the employee discount?" is sent to the server. The server uses a generative AI model to analyze the question and generates the answer: "The employee discount is 10% on all products." This answer is displayed on the smart glasses for the user to see immediately.

[0215] Example prompt sentence:

[0216] Q: What support is available for employees with children?

[0217] Answer: Employees with children are eligible for special parental leave, medical assistance, and assistance with childcare equipment.

[0218] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0219] Step 1:

[0220] The server collects documents related to the company's support system from the company administrator. The collected documents include information such as "employee benefit programs," "special leave regulations," and "medical expense subsidies." These documents are stored on the server and used for subsequent processing.

[0221] Input: Company support program document provided by company administrator

[0222] Output: A database of collected support system documents

[0223] Step 2:

[0224] The server formats and preprocesses the collected documents, removing unnecessary information and using natural language processing (NLP) techniques to structure the text, converting it into a format suitable for input into the generative AI model.

[0225] Input: Collected assistance system documents

[0226] Output: A formatted and preprocessed document

[0227] Step 3:

[0228] The server uses the preprocessed documents to train the generative AI model, specifically learning various information about the business support program and training the model to generate appropriate answers to questions.

[0229] Input: formatted and preprocessed document

[0230] Output: Trained generative AI model

[0231] Step 4:

[0232] A user accesses a smart device and uses voice input to ask a question, for example, "What can I do to help if I get the flu?" The smart device uses a microphone to capture the user's voice.

[0233] Input: User's voice question

[0234] Output: Captured audio data

[0235] Step 5:

[0236] The device uses speech recognition technology to convert the captured speech into text, specifically by using a speech recognition library to analyze the speech data and convert it into the appropriate text.

[0237] Input: Captured audio data

[0238] Output: Texted question

[0239] Step 6:

[0240] The device sends a textual question to the server, which receives the question and prepares it for processing by the generative AI model.

[0241] Input: Texted question

[0242] Output: The question sent to the server

[0243] Step 7:

[0244] The server uses a generative AI model to analyze the question and understand its intent. Specifically, it uses NLP technology to identify keywords and phrases in the question and analyze their meaning.

[0245] Input: Texted question

[0246] Output: Parsed intent (keywords and phrases in the question)

[0247] Step 8:

[0248] The server searches for relevant information from the company's documents based on the analyzed intent. For example, if the keyword "influenza" is included, it searches for support documents related to health care.

[0249] Input: Parsed intent

[0250] Output: Searched support system information

[0251] Step 9:

[0252] The server uses the searched information to generate a specific answer to the question, such as "If employees catch the flu, they can use their paid leave. Medical expense assistance is also available."

[0253] Input: Searched support system information

[0254] Output: The specific answer generated

[0255] Step 10:

[0256] The server sends the generated answer as text data to the terminal, which then displays the received answer on a smart device, such as a smart eyeglasses display.

[0257] Input: Generated specific answer

[0258] Output: Answers displayed on a smart device

[0259] Through the above steps, information about corporate support programs can be provided to users quickly and accurately.

[0260] 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.

[0261] This system combines a system that collects and learns information about corporate support systems, provides appropriate answers to user questions, and an emotion engine that recognizes user emotions. This system is primarily composed of three elements: a server, a terminal, and a user.

[0262] Data collection and learning

[0263] First, the server collects documents about support systems from company administrators. These documents include employee benefit programs, special leave policies, medical expense subsidies, etc. The collected documents are formatted and preprocessed before being input into a generative AI (e.g., ChatGPT). The server uses this generative AI to learn information and prepare to generate answers to user questions.

[0264] Accepting user questions

[0265] Next, the user accesses the "Smart Corporate System Navigator" through the company's online portal or a dedicated application. The user enters a specific question into the question input form on the portal site. For example, the user enters a question such as "What kind of support is available if I get the flu?" The terminal (user's device) then sends the entered question data to the server.

[0266] Emotion recognition by emotion engine

[0267] The device sends the user's facial expressions, voice, and context in real time to the emotion engine, which analyzes this data and recognizes the user's emotional state (e.g., stress, anxiety, relief, etc.). The recognized emotion data is then sent to the server.

[0268] Question analysis and answer generation

[0269] The server analyzes the question using a generative AI model based on the received question data and sentiment data. It understands the intent of the question and extracts keywords and important phrases. For example, "influenza" and "assistance" are extracted as important keywords. The server then searches for information from related documents based on the extracted keywords, organizes the search results, and extracts the most relevant parts.

[0270] The server uses the emotional data to tailor the format and content of the response to suit the user. For example, if the user is feeling anxious, the tone of the response will be gentler and phrased in a way that provides a sense of security. A specific response generated might be, "If you catch the flu, employees can use special paid leave and medical expense assistance. If you have any questions, please contact our support desk."

[0271] Submitting and viewing answers

[0272] The server sends the generated answer to the user's terminal, which displays the received answer to the user, allowing the user to quickly confirm details of the help or assistance required.

[0273] Specific examples

[0274] Example 1: Healthcare support inquiry

[0275] Users access an online portal and type in the question, "What help is available if I have the flu?"

[0276] The device sends emotional data indicating anxiety based on the user's voice and facial expression during input to the emotion engine, which then analyzes the data and recognizes anxiety.

[0277] The terminal transmits the question data and emotion data to the server.

[0278] The server uses generative AI to analyze the question and search documents for information related to "influenza."

[0279] The server takes into account the user's anxious state and generates a gentle response saying, "Employees can take special paid leave and medical expense assistance programs are also available. If you have any questions, please contact our support desk."

[0280] The terminal displays the generated answer, and the user understands the necessary support information and uses the device with peace of mind.

[0281] Example 2: Benefits Program Inquiry

[0282] The user types the question, "What support is available for childcare leave?"

[0283] The terminal sends the user's voice data to the emotion engine, which recognizes that the user is calm.

[0284] The terminal transmits the question data and emotion data to the server.

[0285] The server analyzes the query and retrieves information from the documentation of the relevant assistance program.

[0286] The server generates a clear answer for the calm user: "If you take childcare leave, you will be granted special leave and some medical expenses will be subsidized."

[0287] The terminal displays the generated answer, and the user uses the corresponding support information.

[0288] In this way, the present invention realizes a system that provides more appropriate and personalized answers by combining an emotion engine that recognizes the user's emotions.

[0289] The processing flow will be explained below.

[0290] Step 1: Data entry and learning phase

[0291] The server collects documents about support systems from company administrators, including details of employee benefit programs, personnel system guidebooks, and medical assistance.

[0292] The server formats the collected documents and cleans them of unnecessary information, standardizing the document format and removing noise.

[0293] The server inputs the formatted documents into a generative model (e.g., ChatGPT) and trains the model, which accumulates knowledge about the support system in the generative AI.

[0294] Step 2: Accepting questions from users

[0295] Users access the Smart Corporate System Navigator through their company's online portal or application.

[0296] The user enters a specific question (e.g., "What kind of support is available if I have the flu?") into an input form.

[0297] The terminal transmits the input question data to the server in real time.

[0298] Step 3: Emotion recognition by the emotion engine

[0299] The device sends facial, vocal, and contextual information to the emotion engine as the user inputs, including camera and microphone data.

[0300] The server uses an emotion engine to analyze this data and recognize the user's emotional state (e.g., anxiety, relief, anger, etc.).

[0301] The recognized emotion data is sent to a server for further analysis.

[0302] Step 4: Parsing the Question

[0303] The server uses generative AI to analyze the question entered by the user and extract important keywords and phrases from the question (e.g., "influenza" and "assistance").

[0304] The server searches for the necessary information from related documents based on the extracted keywords.

[0305] Step 5: Generate an answer

[0306] Based on the searched information, the server uses a generative AI model to generate answers for the user.

[0307] The server uses the recognized emotion data to customize the tone and content of the response. For example, if the user is feeling anxious, the response will be changed to a more reassuring tone.

[0308] A specific answer generated is "If an employee catches the flu, they can use special paid leave. They can also use the medical expense subsidy system."

[0309] Step 6: Submit and view your responses

[0310] The server sends the generated answer to the user's terminal.

[0311] The terminal displays the received response to the user, allowing the user to quickly check the details of the system or support they need.

[0312] Examples:

[0313] Example 1: Healthcare support inquiry

[0314] Users access an online portal and type in the question, "What help is available if I have the flu?"

[0315] The device sends the user's question data and emotional data, such as facial expressions and voice, to the emotion engine, which then recognizes anxiety.

[0316] The terminal transmits the question data and emotion data to the server.

[0317] The server parses the query and searches the documents for information related to "influenza."

[0318] The server generates a gentle response to the anxious user, saying, "Employees can take special paid leave and medical expense assistance programs are also available. Please feel free to receive support."

[0319] The terminal displays the generated answer, allowing the user to use appropriate support information with peace of mind.

[0320] Example 2: Benefits Program Inquiry

[0321] The user types the question, "What support is available for childcare leave?"

[0322] The terminal sends the user's voice and facial expression data to the emotion engine, which recognizes that the user is calm.

[0323] The terminal transmits the question data and emotion data to the server.

[0324] The server analyzes the query and retrieves relevant support program information from the document.

[0325] The server generates a clear answer for the calm user: "If you take childcare leave, you will be granted special leave and some medical expenses will be subsidized."

[0326] The terminal displays the generated answer and the user can use the corresponding support information.

[0327] In this way, the present invention realizes a system that provides more appropriate and personalized answers according to the user's emotional state by combining an emotion engine.

[0328] Example 2

[0329] 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."

[0330] Conventional information provision systems for business support programs have the problem that it is difficult for users to quickly and accurately obtain the detailed information they require. As a result, users often experience inconvenience because they are unable to quickly access the support and assistance they need. Furthermore, responses that do not take into account the user's emotional state may cause the user to feel anxious or stressed.

[0331] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting information on corporate support programs, means for formatting and preprocessing the information, means for having a generative AI model that learns based on the information, means for accepting a user's question, means for collecting facial expression, voice, and contextual data when asking a question, means for analyzing the emotional data to recognize the emotional state, means for analyzing the question and the recognized emotional state and generating an answer using the generative AI model, and means for displaying the generated answer to the user. This allows the user to quickly and accurately obtain information about the support programs they need and to receive a personalized answer that takes their emotional state into consideration.

[0332] "Corporate support systems" refers to various systems and programs that companies use to provide support to their employees, such as employee benefits, special leave, and medical expense subsidies.

[0333] "Means of collecting information" refers to the means and methods for obtaining documents and data related to the support system from company managers.

[0334] "Information formatting and preprocessing methods" refers to the means or methods for converting collected documents or data into a format that is easier to analyze and for removing unnecessary characters and formatting.

[0335] A "generative AI model" is an artificial intelligence model trained on large amounts of data, for example, to generate appropriate answers to user questions using natural language processing.

[0336] "Means for accepting user questions" refers to a means or interface for a user to input a specific question into the system and transmit that information to the server.

[0337] "Means for collecting facial, audio, and contextual data" refers to means for collecting data to understand the user's emotional state using a user's device, such as a camera or microphone.

[0338] "Means for analyzing emotional data and recognizing emotional states" refers to means for analyzing and recognizing a user's emotional state based on collected data using a deep learning model or the like.

[0339] "Means for generating an answer" refers to a means for using a generative AI model to analyze the user's question and perceived emotional state and generate an appropriate answer.

[0340] The "means for displaying the answer to the user" refers to a means for displaying the generated answer on the user's device so that the user can easily view the required information.

[0341] "Online portal" refers to a website or dedicated application that provides information about support programs for businesses.

[0342] This invention is a system that collects and learns information about corporate support systems and provides appropriate answers to user questions. This system is mainly composed of three elements: a server, a terminal, and a user.

[0343] Data collection and learning

[0344] The server collects documents related to employee support programs from company administrators. This collection is done using an FTP server or API. The documents include employee benefit programs, special leave policies, medical expense subsidies, etc. The collected documents are saved in a specific directory.

[0345] The server extracts the stored document, removes unnecessary characters and formatting, and formats it into a format that is easy to parse. This formatted data is then fed into a generative AI model (e.g., ChatGPT) to train the model. Once trained, the generative AI model is stored on the server and ready to respond to user queries.

[0346] Accepting user questions

[0347] Users access the company's online portal or dedicated application and enter specific questions (e.g., "What kind of support is available if I get the flu?") into the question input form of the "Smart Corporate System Navigator."

[0348] The terminal (user's device) sends the entered questions to the server in real time, and the communication is secure because it is done via the HTTPS protocol.

[0349] Emotion recognition by emotion engine

[0350] The device uses a camera and microphone to collect facial expressions, voice, and contextual data during user input. It then transmits the collected emotion data to the emotion engine in real time. The emotion engine uses a deep learning model to analyze the collected data and recognize the user's emotional state (e.g., stress, anxiety, relief, etc.). The recognized emotion data is then sent to the server.

[0351] Question analysis and answer generation

[0352] The server uses a generative AI model to analyze the received question data, understand the intent of the question, and extract keywords and key phrases (e.g., "influenza" and "assistance"). The server then searches documents related to relevant assistance programs and systems based on the extracted keywords and organizes the most relevant information.

[0353] The server takes emotion data into account and uses a generative AI model to generate the optimal answer. For example, if a user is anxious, the server might generate a gentle response such as, "Employees can take special paid leave and medical expense assistance programs are also available. If you have any questions, please contact our support desk."

[0354] Submitting and viewing answers

[0355] The server then sends the generated answer to the user's device, again securely via the HTTPS protocol, where it displays the answer to the user, allowing them to quickly view details of the assistance or support they require.

[0356] Specific examples

[0357] Example 1: Healthcare support inquiry

[0358] Users access an online portal and type in the question, "What help is available if I have the flu?"

[0359] The terminal transmits the question data to the server in real time.

[0360] The device transmits the user's facial expression and voice data to the emotion engine, which then recognizes that the user is feeling anxious.

[0361] The terminal transmits the emotion data to the server.

[0362] The server analyzes the question data and emotion data and searches documents for information related to "influenza."

[0363] The server considers the anxious user and generates a gentle response saying, "Employees can take special paid leave and medical expense assistance programs are also available. If you have any questions, please contact our support desk."

[0364] The server sends the generated answer to the user's device, which displays it. The user understands the necessary support information and can use the service with peace of mind.

[0365] Example 2: Benefits Program Inquiry

[0366] The user types the question, "What support is available for childcare leave?"

[0367] The terminal transmits the question data to the server in real time.

[0368] The terminal transmits the user's voice data to the emotion engine, which recognizes that the user is calm.

[0369] The terminal transmits the emotion data to the server.

[0370] The server analyzes the question data and emotion data and searches documents for information related to "child care leave."

[0371] For a calm user, the server generates a clear answer: "If you take childcare leave, you will be granted special leave and some medical expenses will be subsidized."

[0372] The server sends the generated answer to the user's device, which displays it, allowing the user to understand and use the corresponding support information appropriately.

[0373] Prompt Sentence Examples

[0374] "Please tell me about the company's support system if I get the flu. Please speak in a reassuring tone, as users are feeling anxious."

[0375] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0376] Step 1:

[0377] The server collects documents related to the support system from company administrators. This collection is often done via an FTP server or API. The input is the company's document data, and the output is a document file saved in a specific directory.

[0378] Step 2:

[0379] The server extracts the stored document files, removes unnecessary characters and formatting, and formats them into a format that is easy to analyze. Specifically, it uses text analysis tools to remove noise from the data and generate formatted data. The input is the document file read from storage, and the output is formatted data that can be used as training data for the generative AI model.

[0380] Step 3:

[0381] The server inputs the shaped data into a generative AI model for training. A generative AI model (such as ChatGPT) is used for this purpose. The trained generative AI model is stored on the server. The input is the shaped data, and the output is the trained generative AI model.

[0382] Step 4:

[0383] A user accesses a company's online portal or dedicated application and enters a specific question into a question input form. For example, "What kind of support is available if I get the flu?" The input is the user's question text, and the output is data sent from the terminal to the server.

[0384] Step 5:

[0385] The terminal transmits the entered question to the server in real time, securely using the HTTPS protocol. The input is the user's question text, and the output is the data sent to the server.

[0386] Step 6:

[0387] The device uses a camera and microphone to collect facial, voice, and contextual data during user input. This data is sent to the emotion engine in real time for preprocessing. The input is the user's emotion data, and the output is preprocessed emotion data.

[0388] Step 7:

[0389] The emotion engine analyzes the collected emotion data using a deep learning model to recognize the user's emotional state (e.g., stress, anxiety, relief, etc.). The input is the preprocessed emotion data, and the output is the recognized emotional state.

[0390] Step 8:

[0391] The server performs analysis based on the received question data and recognized emotion data. It uses a generative AI model to understand the intent of the question and extract keywords and important phrases (e.g., "influenza" or "assistance"). The input is the question data and emotion data, and the output is the extracted keywords and semantic data.

[0392] Step 9:

[0393] The server searches documents related to related support programs and systems based on the extracted keywords and organizes the most relevant information. The input is the extracted keywords, and the output is organized information data.

[0394] Step 10:

[0395] The server takes into account the emotional data and uses a generative AI model to generate the optimal answer. For example, to an anxious user, it generates a gentle answer such as, "Employees can take special paid leave, and medical expense subsidies are also available. If you have any questions, please contact our support desk." The input is organized information data and emotional data, and the output is the generated answer text.

[0396] Step 11:

[0397] The server sends the generated answer to the user's terminal, also securely using the HTTPS protocol. The input is the generated answer text, and the output is the data sent to the terminal.

[0398] Step 12:

[0399] The terminal displays the received answer to the user, allowing the user to quickly check the details of the assistance or support required. The input is the received answer text, and the output is the display data in a format that the user can view.

[0400] (Application example 2)

[0401] 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."

[0402] In information provision systems for corporate support systems, it is difficult for employees to quickly obtain appropriate information tailored to their own situation. Furthermore, there are currently few systems that provide personalized answers that reflect the emotional state of employees when they ask questions. This issue is particularly important in factories, where providing immediate support information directly leads to an improvement in the working environment.

[0403] 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.

[0404] In this invention, the server includes means for collecting information on corporate support programs, means for formatting and preprocessing the information, means for having a generative model that learns based on the information, means for accepting user questions, means for analyzing the questions, means for searching for relevant information, means for generating answers based on the searched information, means for displaying the generated answers to the user, emotion analysis means for recognizing the user's emotions, and means for adjusting the format and content of the answers based on the emotion data recognized by the emotion analysis means. This makes it possible to instantly provide appropriate support information according to the emotions of employees when they ask questions.

[0405] "Corporate support systems" is a general term for various systems such as employee benefits, special leave, and medical expense subsidies that companies provide to their employees.

[0406] "Means of collecting information" refers to the methods and tools used to collect documents and data related to support systems within the company.

[0407] "Information formatting and preprocessing means" refers to methods and tools used to convert collected information into a format that is easy to analyze.

[0408] A "generative model" refers to a computer program or algorithm that learns from collected information and generates appropriate answers to user questions.

[0409] "Means for accepting questions" refers to an interface or application that allows a user to input questions to the server.

[0410] "Means for analyzing questions" refers to methods and tools that use natural language processing technology to analyze questions from users and extract intent and keywords.

[0411] "Means for retrieving relevant information" refers to methods and tools for retrieving relevant information from a database based on the analyzed question.

[0412] "Answer generation means" refers to a computer program or algorithm that generates an appropriate answer to a user's question based on the retrieved information.

[0413] "Means for displaying to the user" refers to a method or tool for displaying the generated answer on a device so that the user can review it.

[0414] "Emotion analysis means" refers to technologies and algorithms for recognizing and analyzing a user's emotional state.

[0415] "Means for adjusting the format and content of responses based on emotional data" refers to methods or tools for adjusting the tone and content of responses according to the user's emotional state based on emotional data obtained by the emotion analysis means.

[0416] To implement this invention, the system consists of three main components: a server, a terminal, and a user. The server collects information about companies' support programs and performs learning using a generative AI model. The terminal accepts user questions, performs sentiment analysis, and transmits the question and sentiment data to the server. Users access the system through an online portal or a dedicated application.

[0417] server

[0418] The server processes the information as follows:

[0419] Information gathering: Collect documents from company managers regarding support programs, including employee benefits, special leave, and medical assistance.

[0420] Preprocessing and formatting: Format and preprocess the collected documents and feed them into a generative AI model (e.g., OpenAI's GPT-3).

[0421] Learning: Generative AI models are used to learn from collected information.

[0422] Question analysis: Question data sent from the device is analyzed using natural language processing technology to extract important keywords.

[0423] Information retrieval: Search for information from related documents based on the analysis results.

[0424] Answer generation: Using a generative AI model, appropriate answers are generated based on retrieved information and sentiment data.

[0425] Prepare for display: Send the generated answer to the device.

[0426] Terminal

[0427] The terminal works as follows:

[0428] Question reception: Receive questions from users and convert them into data.

[0429] Sentiment analysis: The emotion analysis engine analyzes the entered question as well as emotional data such as the user's facial expressions and voice.

[0430] Data transmission: Question data and emotion data are sent to the server.

[0431] Display Answer: The answer sent from the server is displayed to the user.

[0432] User

[0433] The user uses the system as follows:

[0434] Access: Access an online portal or dedicated application to enter your questions.

[0435] Enter a question: Enter a specific question and wait for the system to answer.

[0436] Specific examples

[0437] Example 1: Healthcare support inquiry

[0438] Question: User types, "I'm sick, can I leave work early?"

[0439] Sentiment analysis: Recognize when a user is feeling anxious.

[0440] Answer generation: The server responds, "You can leave work early by using special paid leave. Please contact our support desk for details."

[0441] Example 2: Querying rest facilities

[0442] Question: User types, "What rest stop is available when I finish work today?"

[0443] Sentiment analysis: Recognize that the user is calm.

[0444] Answer generation: The server responds, "After you finish work, break rooms A and B will be available."

[0445] Prompt Sentence Examples

[0446] Example 1 prompt statement:

[0447] User Question: Can I leave work early because I'm not feeling well?

[0448] User Emotion: Anxiety

[0449] Generate the appropriate answer.

[0450] Example 2 prompt statement:

[0451] User question: What rest areas can I use when I'm done with today's work?

[0452] User Sentiment: Calm

[0453] Generate the appropriate answer.

[0454] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0455] Step 1:

[0456] The server collects documents related to the company's support systems, such as employee benefit programs, special leave policies, and medical expense subsidies provided by company administrators, and stores them in a database.

[0457] Input: Documents about corporate support programs

[0458] Output: The formatted and preprocessed document data

[0459] Step 2:

[0460] The server formats and preprocesses the collected documents, specifically converting the text data format and removing unnecessary data to make it easier for the generative AI model to process.

[0461] Input: Data from collected documents

[0462] Output: Preprocessed data that can be input into a generative AI model

[0463] Step 3:

[0464] The server uses the preprocessed data to train a generative AI model. Specifically, it uses OpenAI's GPT-3 model to learn information and build a knowledge base for generating appropriate answers to user questions.

[0465] Input: Preprocessed data

[0466] Output: Knowledge base based on generative AI models

[0467] Step 4:

[0468] Users enter their questions through a dedicated application or online portal. Specifically, users enter text-based questions, which are then collected as digital data by the device.

[0469] Input: User question (e.g., "I'm not feeling well, can I leave early?")

[0470] Output: Digitized question data

[0471] Step 5:

[0472] The device performs emotion analysis to recognize the user's emotions, specifically by analyzing the user's facial expressions, voice, or other input data to identify their emotional state (e.g., stress, anxiety, calm).

[0473] Input: User facial expressions, voice, and contextual data

[0474] Output: Recognized emotion data (e.g., "anxiety")

[0475] Step 6:

[0476] The terminal transmits the question data and emotion data to the server. Specifically, the terminal transmits the digitized question data and the recognized emotion data to the server as a data packet.

[0477] Input: Question data, emotion data

[0478] Output: Send data to the server

[0479] Step 7:

[0480] The server analyzes the question using a generative AI model based on the question data and sentiment data received, and extracts important keywords. Specifically, it uses natural language processing technology to understand the intent of the question and identify related keywords.

[0481] Input: Question data, emotion data

[0482] Output: Extracted keywords (e.g., "Leave early", "health")

[0483] Step 8:

[0484] The server searches for relevant information from related documents based on the extracted keywords, using a database search algorithm to identify and organize relevant information.

[0485] Input: Extracted keywords

[0486] Output: Relevant information (e.g. special paid leave regulations for early departure)

[0487] Step 9:

[0488] The server generates answers using a generative AI model based on search results and emotional data. Specifically, for users in an anxious state, it creates answers in a gentle tone that include appropriate support information.

[0489] Input: Search results, emotion data

[0490] Output: Generated answer (e.g. "You can leave early using special paid leave. Please contact support for details.")

[0491] Step 10:

[0492] The server sends the generated answer to the terminal, and the terminal displays it to the user, specifically, displays the answer text on the user's device, quickly meeting the user's needs.

[0493] Input: Generated Answer

[0494] Output: Answer displayed on the user's terminal

[0495] 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.

[0496] 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.

[0497] 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.

[0498] [Second embodiment]

[0499] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0500] 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.

[0501] 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).

[0502] 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.

[0503] 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.

[0504] 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).

[0505] 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.

[0506] 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.

[0507] 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.

[0508] 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.

[0509] 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.

[0510] 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."

[0511] This system collects and learns information about corporate support systems and provides appropriate answers to user questions. This system is mainly composed of three elements: a server, a terminal, and a user.

[0512] Data collection and learning

[0513] First, the server collects documents from company administrators about the company's support systems. These documents include employee benefit programs, special leave policies, medical expense subsidies, etc. The collected documents are formatted and preprocessed before being input into a generative AI (e.g., ChatGPT). The server uses this generative AI to learn information and prepare it to generate answers to questions.

[0514] Accepting user questions

[0515] Next, users access the Smart Corporate System Navigator through their company's online portal or dedicated application, and enter a specific question in the search box, such as, "What kind of support is available if I get the flu?"

[0516] Question analysis and answer generation

[0517] The device sends the entered question to the server, which uses generative AI to analyze the question and understand its intent. Based on the results of the analysis, the server searches for relevant information in the company's documents. For example, if the question contains the keyword "influenza," the search will return "health management support documents" and the like.

[0518] Based on the retrieved information, the server generates a detailed answer, such as, "If employees catch the flu, they can use special paid leave. Medical expense assistance programs are also available."

[0519] Show Answers

[0520] Finally, the server sends the generated answer to the terminal, which then displays the answer to the user. The user can view the displayed answer, understand the support or assistance they need, and use it appropriately.

[0521] Specific examples

[0522] Example 1: Healthcare support inquiry

[0523] Users access an online portal and type in the question, "What help is available if I have the flu?"

[0524] The terminal sends this question to the server.

[0525] The server uses generative AI to analyze the question and search documents for information related to "influenza."

[0526] The server generates a response that "employees can take special paid leave and also have access to medical expense assistance."

[0527] The terminal displays the generated answer, and the user understands the necessary support information and uses it appropriately.

[0528] Example 2: Benefits Program Inquiry

[0529] The user types the question, "What support is available for childcare leave?"

[0530] The terminal sends this question to the server.

[0531] The server analyzes the query and retrieves the relevant assistance program documentation.

[0532] The server generates a response saying, "If you take child care leave, you will be granted special leave and some medical expenses will be subsidized."

[0533] The terminal displays the generated answer, and the user uses the corresponding support information.

[0534] In this way, the present invention provides a system that efficiently collects and learns information about corporate support systems and provides quick and accurate answers to user questions.

[0535] The processing flow will be explained below.

[0536] Step 1: Data entry and learning phase

[0537] The server receives documents related to support systems from company administrators. Specifically, it collects "support documents related to health management," "personnel system guidebooks," "details of employee benefit programs," etc.

[0538] The server formats the documents it receives and performs data cleaning if necessary, which includes standardizing the document format and removing unnecessary textual information.

[0539] The server feeds the formatted documents into a generative model (e.g., ChatGPT) and trains the model, which then has detailed knowledge of the company's systems.

[0540] Step 2: Accepting questions from users

[0541] Users access the Smart Corporate System Navigator through their company's online portal or application.

[0542] The user enters a specific question into the question entry form on the portal site, for example, "What kind of support is available if I get the flu?"

[0543] The terminal (user's device) sends the entered question data to the server.

[0544] Step 3: Parsing the Question

[0545] The server analyzes the received question data. Using a generative AI model, it understands the intent of the question and extracts keywords and important phrases. For example, "influenza" and "support" are extracted as important keywords.

[0546] Step 4: Find related information

[0547] The server uses the extracted keywords to search for relevant information from related documents, such as "health management support documents" and "personnel system guidebooks."

[0548] The server sorts through the search results and extracts the most relevant parts.

[0549] Step 5: Generate an answer

[0550] The server generates answers to provide to users based on the extracted information. The generation AI summarizes the information and adjusts the writing style to create easy-to-understand sentences.

[0551] For example, it generates a specific answer such as "If employees catch the flu, they can use special paid leave. Medical expense assistance systems are also available."

[0552] Step 6: Submit and view your responses

[0553] The server sends the generated answer to the user's terminal.

[0554] The terminal displays the received response to the user, allowing the user to quickly ascertain details of the assistance or assistance required.

[0555] In this way, by performing specific processing at each step, a system is provided that allows users to efficiently obtain and utilize information about corporate systems and support programs.

[0556] Example 1

[0557] 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."

[0558] Information about corporate support programs is diverse, making it difficult for employees to quickly obtain the information they need. Furthermore, if the information provided is inaccurate, employees may not receive appropriate support, hindering efficient work performance. In addition, there is the problem that technology for accurately analyzing questions entered by users in natural language and generating appropriate answers is still immature. Furthermore, if the generated answers are not displayed clearly to users, user convenience is reduced.

[0559] 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.

[0560] In this invention, the server includes: means for collecting information about corporate support programs; means for formatting and preprocessing the information; means for having a generative model that learns based on the information; means for accepting user questions; means for analyzing the questions; means for searching for relevant information; means for generating answers based on the searched information; means for displaying the generated answers to the user; means for removing unnecessary parts and extracting only important information using a natural language processing tool when formatting and preprocessing the information; means for generating detailed answers to the user questions based on prompt sentences using a generative AI model; and means for displaying the answers generated based on the prompt sentences in a user-friendly format. This makes it possible to quickly and accurately obtain information about corporate support programs, generate appropriate answers to questions entered in natural language, and display the answers in an easy-to-understand format to the user.

[0561] "Means for collecting information" refers to devices and methods for receiving and storing data relating to the company's support system from the company administrator.

[0562] "Information formatting and preprocessing means" refers to devices and methods that use natural language processing tools to convert collected data into an appropriate format and remove unnecessary parts.

[0563] "Means having a learning generative model" refers to a device or method that uses a generative AI model to learn information based on preprocessed data.

[0564] The "means for accepting a user's question" refers to a device or method that provides a function for a user to input a question in natural language and receives the input content.

[0565] The "means for analyzing a question" refers to a device or method that uses natural language processing technology to analyze a question received from a user and understand its intent.

[0566] The "means for retrieving relevant information" refers to a device or method for retrieving relevant information from relevant company documents or databases based on the analyzed query.

[0567] A "means for generating an answer" is a device or method that uses a generative AI model to create a detailed answer based on the searched information.

[0568] The "means for displaying the generated answers to the user" refers to a device or method for displaying the generated answers in a user-friendly format so that the user can view them.

[0569] "Means using natural language processing tools" refers to the techniques and devices used to analyze and format text when preprocessing collected data.

[0570] A "means for generating a detailed answer based on a prompt sentence" is a device or method that uses a generative AI model to input a prompt corresponding to a specific question and generate a detailed answer based on that prompt.

[0571] The "means for displaying in a user-friendly format" refers to a device or method that provides the generated answers to the user in a format that is easy to view and understand.

[0572] This system collects and learns information about corporate support systems and provides appropriate answers to user questions. This system is mainly composed of three elements: a server, a terminal, and a user.

[0573] Data collection and learning

[0574] The server receives documents related to the company's support system from the company administrator. The received documents are saved as text files or PDFs. The saved documents are preprocessed using an NLP tool (e.g., SpaCy, NLTK). Preprocessing refers to removing unnecessary parts and extracting important information. The formatted data from this process is input into a generative AI model (e.g., ChatGPT). The server uses this generative AI model to learn information and prepare to generate answers to user questions.

[0575] Accepting user questions

[0576] Users log in to their company's online portal or dedicated application to access the "Smart Corporate System Navigator." They enter a specific question into the search box, for example, "What kind of support is available if I have the flu?" The device then sends this input to the server. The transmitted data is protected using a secure communication protocol (e.g., HTTPS).

[0577] Question analysis and answer generation

[0578] The server converts the received question into a prompt to be input into the generative AI. Using the generative AI model, the server analyzes the intent of the user's question and extracts related keywords. Based on the results of this analysis, the server searches for relevant information from the company's documents. For example, if the keyword "influenza" is included, support documents related to health management will be searched for. Based on the search results, the generative AI model is used to generate a detailed answer. An example of a specific answer would be: "If employees catch the flu, they can use special paid leave. Medical expense subsidy systems are also available."

[0579] Show Answers

[0580] The server sends the generated answer to the terminal, which displays the answer in a user-friendly format (e.g., a card or pop-up).The user can view the displayed answer, understand the support or assistance they need, and use it appropriately.

[0581] Specific examples

[0582] Example 1: Healthcare support inquiry

[0583] The user accesses the online portal and enters the question, "What support is available if I get the flu?" The device then sends this question to the server. The server uses generative AI to analyze the question and searches documents for information related to "flu." The server then generates the answer, "Employees can use special paid leave and medical expense subsidy systems." The user then understands the necessary support information and uses it appropriately.

[0584] Example 2: Benefits Program Inquiry

[0585] The user inputs the question, "What kind of support is available for child care leave?" The device sends this question to the server. The server analyzes the question and searches for documents on related support programs. The server generates an answer that says, "If you take child care leave, you will be granted special leave and some medical expenses will be subsidized." The device displays the generated answer, allowing the user to use the relevant support information appropriately.

[0586] This system can efficiently collect and learn information about corporate support systems and provide quick and accurate answers to users' questions.

[0587] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0588] Step 1:

[0589] The server receives documents relating to the company's support system from the company administrator.

[0590] Specifically, the document is collected using the file upload function.

[0591] Input: Support program documentation provided by company administrator.

[0592] Output: Saved support system documents.

[0593] Step 2:

[0594] The server saves the received documents in text file or PDF format.

[0595] Specifically, the uploaded file is saved in an appropriate folder or database.

[0596] Input: Uploaded assistance system document.

[0597] Output: Saved text file and PDF.

[0598] Step 3:

[0599] The server preprocesses the stored documents using natural language processing (NLP) tools (e.g., SpaCy, NLTK).

[0600] Specifically, it removes unnecessary parts from the document (e.g., page numbers, headers, footers) and extracts important information.

[0601] Input: Saved text files or PDFs.

[0602] Output: Preprocessed text data.

[0603] Step 4:

[0604] The server formats the preprocessed data and generates prompts to input into a generative AI model (e.g., ChatGPT).

[0605] Specifically, the process creates a prompt statement template based on the preprocessed data.

[0606] Input: Preprocessed text data.

[0607] Output: The prompt sentence to be input to the generative AI model.

[0608] Step 5:

[0609] The server uses a generative AI model to learn information and prepare to generate answers to user questions.

[0610] Specifically, the prompt sentence is input into the generative AI model and the model is trained.

[0611] Input: prompt statement.

[0612] Output: A trained generative AI model.

[0613] Step 6:

[0614] Users log in to the online portal or dedicated application and access the "Smart Corporate System Navigator."

[0615] Specifically, the user authentication and login process are carried out.

[0616] Input: User credentials.

[0617] Output: Authenticated user interface.

[0618] Step 7:

[0619] Users enter a specific question into a search box.

[0620] Specifically, you enter a question in the search box and press the send button.

[0621] Input: The question text provided by the user.

[0622] Output: The question data sent to the terminal.

[0623] Step 8:

[0624] The terminal transmits the entered question to the server.

[0625] Specifically, the input data is transferred to the server using a secure communication protocol (e.g., HTTPS).

[0626] Input: Question data from the user.

[0627] Output: The query data sent to the server.

[0628] Step 9:

[0629] The server converts the received question into a prompt sentence to be input to the generation AI.

[0630] Specifically, the question data is formatted into an appropriate prompt format.

[0631] Input: Question data from the user.

[0632] Output: The prompt statement.

[0633] Step 10:

[0634] The server uses generative AI to analyze the intent of the question and extract relevant keywords.

[0635] Specifically, the prompt sentence is input into the generative AI model, and intent analysis and keyword extraction are performed.

[0636] Input: prompt statement.

[0637] Output: Parsed intent and extracted keywords.

[0638] Step 11:

[0639] The server searches for company documents based on the extracted keywords.

[0640] Specifically, the operation involves searching for the relevant information from a document management system or database.

[0641] Input: Parsed intent and extracted keywords.

[0642] Output: Search results (relevant documents).

[0643] Step 12:

[0644] The server generates detailed answers based on the search results using a generative AI model.

[0645] Specifically, the search results are input into a generative AI model to generate a detailed answer.

[0646] Input: Search results.

[0647] Output: The detailed answer generated.

[0648] Step 13:

[0649] The server sends the generated response to the terminal.

[0650] Specifically, the generated answer is formatted into a user-friendly format and sent to the terminal.

[0651] Input: The generated detailed answer.

[0652] Output: The response data sent to the device.

[0653] Step 14:

[0654] The device displays the received answers in a user-friendly format (e.g., card format or popup).

[0655] Specifically, the response data is read and displayed on the user interface.

[0656] Input: Response data sent to the terminal.

[0657] Output: The answer displayed to the user.

[0658] Step 15:

[0659] The user can view the displayed answers, understand the support and assistance they need, and use it appropriately.

[0660] Specifically, the displayed information is checked and the next action is taken as necessary.

[0661] Input: The answer shown to the user.

[0662] Output: User understanding and use of information.

[0663] (Application example 1)

[0664] 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."

[0665] There is a wide variety of information available about corporate support programs, making it difficult for employees to quickly and accurately identify the program that best suits them. Furthermore, especially in busy environments, such as store employees, who often lack the time to use digital devices, employees tend to neglect using support programs. This leads to a decline in employee utilization of support programs, resulting in the inability to fully realize the benefits of corporate employee benefits.

[0666] 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.

[0667] In this invention, the server includes: means for collecting information on corporate support programs; means for formatting and preprocessing the information; means for having a generative model that learns based on the information; means for accepting user questions; means for analyzing the questions; means for searching for relevant information; means for generating answers based on the searched information; means for displaying the generated answers to the user; a voice input means installed in the smart device; means for converting voice to text; means for transmitting the text questions to the server; and means for displaying the generated answers on the smart device. This allows employees to easily ask questions through voice input and quickly access information on appropriate support programs.

[0668] "Corporate support systems" is a general term for welfare programs, special leave provisions, medical expense subsidies, etc. that companies provide to their employees.

[0669] "Means for collecting information" refers to a method or system for collecting documents related to the support system from company administrators.

[0670] "Information formatting and preprocessing means" refers to methods and systems for converting collected documents into a format that is easy to analyze and for preprocessing the data.

[0671] A "generative model" is a machine learning model that learns from collected and preprocessed information and generates appropriate answers to user questions.

[0672] The "means for accepting a user's question" is an interface for receiving a question input by a user and transmitting it to the system.

[0673] The "means for analyzing questions" refers to natural language processing techniques and algorithms for analyzing received questions and understanding their intent.

[0674] The "means for searching relevant information" is a method or system for searching company documents for information relevant to the analyzed question.

[0675] "Answer generation means" refers to a method or system for generating a specific answer based on the retrieved information.

[0676] The "means for displaying the generated answer to the user" refers to a method or system for displaying the generated answer on the user's device (terminal).

[0677] "Smart devices" is a general term for devices that can connect to the Internet and run various applications, such as smartphones, smart glasses, and head-mounted displays.

[0678] The "voice input means" is a function for inputting the user's voice using a microphone or the like installed on the smart device.

[0679] "Means for converting speech to text" refers to speech recognition technology and algorithms for analyzing input speech and converting it into text data.

[0680] The "means for transmitting a question converted into text to a server" is a communication function for transmitting a user's question converted into text by voice recognition to a server.

[0681] The "means for displaying the answer on the smart device" is a function for receiving the answer generated from the server and displaying it on the display screen of the smart device.

[0682] This system collects and learns information about corporate support programs and provides appropriate answers to user questions. This system is composed of three elements: a server, a terminal, and a user.

[0683] First, the server collects documents from company administrators about the company's support systems, including employee benefit programs, special leave policies, medical expense subsidies, etc. The collected documents undergo formatting and preprocessing, and then the generative AI learns the information and prepares them to generate answers to questions.

[0684] Next, the user accesses the company's smart device and asks a question through voice input, for example, "What kind of support is available if I have the flu?" The microphone installed on the smart device captures this voice and converts it into text using voice recognition technology.

[0685] The device sends the converted text data to the server, which uses generative AI to analyze the question and understand its intent. Based on the results of the analysis, the server searches for relevant information in the company's documents. For example, if a question contains the keyword "influenza," the search will return "health management support documents" and other relevant information.

[0686] Based on the searched information, the server generates a detailed answer. For example, it creates a specific answer such as, "If an employee catches the flu, they can use special paid leave. Medical expense subsidies are also available." The generated answer is sent to the smart device as text data and displayed on the screen. The user can view the displayed answer, understand the support and assistance they need, and use it appropriately.

[0687] Examples:

[0688] A user uses the smart glasses to ask a question out loud: "How much is the employee discount?" The question is converted into text using speech recognition, and the question "How much is the employee discount?" is sent to the server. The server uses a generative AI model to analyze the question and generates the answer: "The employee discount is 10% on all products." This answer is displayed on the smart glasses for the user to see immediately.

[0689] Example prompt sentence:

[0690] Q: What support is available for employees with children?

[0691] Answer: Employees with children are eligible for special parental leave, medical assistance, and assistance with childcare equipment.

[0692] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0693] Step 1:

[0694] The server collects documents related to the company's support system from the company administrator. The collected documents include information such as "employee benefit programs," "special leave regulations," and "medical expense subsidies." These documents are stored on the server and used for subsequent processing.

[0695] Input: Company support program document provided by company administrator

[0696] Output: A database of collected support system documents

[0697] Step 2:

[0698] The server formats and preprocesses the collected documents, removing unnecessary information and using natural language processing (NLP) techniques to structure the text, converting it into a format suitable for input into the generative AI model.

[0699] Input: Collected assistance system documents

[0700] Output: A formatted and preprocessed document

[0701] Step 3:

[0702] The server uses the preprocessed documents to train the generative AI model, specifically learning various information about the business support program and training the model to generate appropriate answers to questions.

[0703] Input: formatted and preprocessed document

[0704] Output: Trained generative AI model

[0705] Step 4:

[0706] A user accesses a smart device and uses voice input to ask a question, for example, "What can I do to help if I get the flu?" The smart device uses a microphone to capture the user's voice.

[0707] Input: User's voice question

[0708] Output: Captured audio data

[0709] Step 5:

[0710] The device uses speech recognition technology to convert the captured speech into text, specifically by using a speech recognition library to analyze the speech data and convert it into the appropriate text.

[0711] Input: Captured audio data

[0712] Output: Texted question

[0713] Step 6:

[0714] The device sends a textual question to the server, which receives the question and prepares it for processing by the generative AI model.

[0715] Input: Texted question

[0716] Output: The question sent to the server

[0717] Step 7:

[0718] The server uses a generative AI model to analyze the question and understand its intent. Specifically, it uses NLP technology to identify keywords and phrases in the question and analyze their meaning.

[0719] Input: Texted question

[0720] Output: Parsed intent (keywords and phrases in the question)

[0721] Step 8:

[0722] The server searches for relevant information from the company's documents based on the analyzed intent. For example, if the keyword "influenza" is included, it searches for support documents related to health care.

[0723] Input: Parsed intent

[0724] Output: Searched support system information

[0725] Step 9:

[0726] The server uses the searched information to generate a specific answer to the question, such as "If employees catch the flu, they can use their paid leave. Medical expense assistance is also available."

[0727] Input: Searched support system information

[0728] Output: The specific answer generated

[0729] Step 10:

[0730] The server sends the generated answer as text data to the terminal, which then displays the received answer on a smart device, such as a smart eyeglasses display.

[0731] Input: Generated specific answer

[0732] Output: Answers displayed on a smart device

[0733] Through the above steps, information about corporate support programs can be provided to users quickly and accurately.

[0734] 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.

[0735] This system combines a system that collects and learns information about corporate support systems, provides appropriate answers to user questions, and an emotion engine that recognizes user emotions. This system is primarily composed of three elements: a server, a terminal, and a user.

[0736] Data collection and learning

[0737] First, the server collects documents about support systems from company administrators. These documents include employee benefit programs, special leave policies, medical expense subsidies, etc. The collected documents are formatted and preprocessed before being input into a generative AI (e.g., ChatGPT). The server uses this generative AI to learn information and prepare to generate answers to user questions.

[0738] Accepting user questions

[0739] Next, the user accesses the "Smart Corporate System Navigator" through the company's online portal or a dedicated application. The user enters a specific question into the question input form on the portal site. For example, the user enters a question such as "What kind of support is available if I get the flu?" The terminal (user's device) then sends the entered question data to the server.

[0740] Emotion recognition by emotion engine

[0741] The device sends the user's facial expressions, voice, and context in real time to the emotion engine, which analyzes this data and recognizes the user's emotional state (e.g., stress, anxiety, relief, etc.). The recognized emotion data is then sent to the server.

[0742] Question analysis and answer generation

[0743] The server analyzes the question using a generative AI model based on the received question data and sentiment data. It understands the intent of the question and extracts keywords and important phrases. For example, "influenza" and "assistance" are extracted as important keywords. The server then searches for information from related documents based on the extracted keywords, organizes the search results, and extracts the most relevant parts.

[0744] The server uses the emotional data to tailor the format and content of the response to suit the user. For example, if the user is feeling anxious, the tone of the response will be gentler and phrased in a way that provides a sense of security. A specific response generated might be, "If you catch the flu, employees can use special paid leave and medical expense assistance. If you have any questions, please contact our support desk."

[0745] Submitting and viewing answers

[0746] The server sends the generated answer to the user's terminal, which displays the received answer to the user, allowing the user to quickly confirm details of the help or assistance required.

[0747] Specific examples

[0748] Example 1: Healthcare support inquiry

[0749] Users access an online portal and type in the question, "What help is available if I have the flu?"

[0750] The device sends emotional data indicating anxiety based on the user's voice and facial expression during input to the emotion engine, which then analyzes the data and recognizes anxiety.

[0751] The terminal transmits the question data and emotion data to the server.

[0752] The server uses generative AI to analyze the question and search documents for information related to "influenza."

[0753] The server takes into account the user's anxious state and generates a gentle response saying, "Employees can take special paid leave and medical expense assistance programs are also available. If you have any questions, please contact our support desk."

[0754] The terminal displays the generated answer, and the user understands the necessary support information and uses the device with peace of mind.

[0755] Example 2: Benefits Program Inquiry

[0756] The user types the question, "What support is available for childcare leave?"

[0757] The terminal sends the user's voice data to the emotion engine, which recognizes that the user is calm.

[0758] The terminal transmits the question data and emotion data to the server.

[0759] The server analyzes the query and retrieves information from the documentation of the relevant assistance program.

[0760] The server generates a clear answer for the calm user: "If you take childcare leave, you will be granted special leave and some medical expenses will be subsidized."

[0761] The terminal displays the generated answer, and the user uses the corresponding support information.

[0762] In this way, the present invention realizes a system that provides more appropriate and personalized answers by combining an emotion engine that recognizes the user's emotions.

[0763] The processing flow will be explained below.

[0764] Step 1: Data entry and learning phase

[0765] The server collects documents about support systems from company administrators, including details of employee benefit programs, personnel system guidebooks, and medical assistance.

[0766] The server formats the collected documents and cleans them of unnecessary information, standardizing the document format and removing noise.

[0767] The server inputs the formatted documents into a generative model (e.g., ChatGPT) and trains the model, which accumulates knowledge about the support system in the generative AI.

[0768] Step 2: Accepting questions from users

[0769] Users access the Smart Corporate System Navigator through their company's online portal or application.

[0770] The user enters a specific question (e.g., "What kind of support is available if I have the flu?") into an input form.

[0771] The terminal transmits the input question data to the server in real time.

[0772] Step 3: Emotion recognition by the emotion engine

[0773] The device sends facial, vocal, and contextual information to the emotion engine as the user inputs, including camera and microphone data.

[0774] The server uses an emotion engine to analyze this data and recognize the user's emotional state (e.g., anxiety, relief, anger, etc.).

[0775] The recognized emotion data is sent to a server for further analysis.

[0776] Step 4: Parsing the Question

[0777] The server uses generative AI to analyze the question entered by the user and extract important keywords and phrases from the question (e.g., "influenza" and "assistance").

[0778] The server searches for the necessary information from related documents based on the extracted keywords.

[0779] Step 5: Generate an answer

[0780] Based on the searched information, the server uses a generative AI model to generate answers for the user.

[0781] The server uses the recognized emotion data to customize the tone and content of the response. For example, if the user is feeling anxious, the response will be changed to a more reassuring tone.

[0782] A specific answer generated is "If an employee catches the flu, they can use special paid leave. They can also use the medical expense subsidy system."

[0783] Step 6: Submit and view your responses

[0784] The server sends the generated answer to the user's terminal.

[0785] The terminal displays the received response to the user, allowing the user to quickly check the details of the system or support they need.

[0786] Examples:

[0787] Example 1: Healthcare support inquiry

[0788] Users access an online portal and type in the question, "What help is available if I have the flu?"

[0789] The device sends the user's question data and emotional data, such as facial expressions and voice, to the emotion engine, which then recognizes anxiety.

[0790] The terminal transmits the question data and emotion data to the server.

[0791] The server parses the query and searches the documents for information related to "influenza."

[0792] The server generates a gentle response to the anxious user, saying, "Employees can take special paid leave and medical expense assistance programs are also available. Please feel free to receive support."

[0793] The terminal displays the generated answer, allowing the user to use appropriate support information with peace of mind.

[0794] Example 2: Benefits Program Inquiry

[0795] The user types the question, "What support is available for childcare leave?"

[0796] The terminal sends the user's voice and facial expression data to the emotion engine, which recognizes that the user is calm.

[0797] The terminal transmits the question data and emotion data to the server.

[0798] The server analyzes the query and retrieves relevant support program information from the document.

[0799] The server generates a clear answer for the calm user: "If you take childcare leave, you will be granted special leave and some medical expenses will be subsidized."

[0800] The terminal displays the generated answer and the user can use the corresponding support information.

[0801] In this way, the present invention realizes a system that provides more appropriate and personalized answers according to the user's emotional state by combining an emotion engine.

[0802] Example 2

[0803] 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."

[0804] Conventional information provision systems for business support programs have the problem that it is difficult for users to quickly and accurately obtain the detailed information they require. As a result, users often experience inconvenience because they are unable to quickly access the support and assistance they need. Furthermore, responses that do not take into account the user's emotional state may cause the user to feel anxious or stressed.

[0805] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting information on corporate support programs, means for formatting and preprocessing the information, means for having a generative AI model that learns based on the information, means for accepting a user's question, means for collecting facial expression, voice, and contextual data when asking a question, means for analyzing the emotional data to recognize the emotional state, means for analyzing the question and the recognized emotional state and generating an answer using the generative AI model, and means for displaying the generated answer to the user. This allows the user to quickly and accurately obtain information about the support programs they need and to receive a personalized answer that takes their emotional state into consideration.

[0806] "Corporate support systems" refers to various systems and programs that companies use to provide support to their employees, such as employee benefits, special leave, and medical expense subsidies.

[0807] "Means of collecting information" refers to the means and methods for obtaining documents and data related to the support system from company managers.

[0808] "Information formatting and preprocessing methods" refers to the means or methods for converting collected documents or data into a format that is easier to analyze and for removing unnecessary characters and formatting.

[0809] A "generative AI model" is an artificial intelligence model trained on large amounts of data, for example, to generate appropriate answers to user questions using natural language processing.

[0810] "Means for accepting user questions" refers to a means or interface for a user to input a specific question into the system and transmit that information to the server.

[0811] "Means for collecting facial, audio, and contextual data" refers to means for collecting data to understand the user's emotional state using a user's device, such as a camera or microphone.

[0812] "Means for analyzing emotional data and recognizing emotional states" refers to means for analyzing and recognizing a user's emotional state based on collected data using a deep learning model or the like.

[0813] "Means for generating an answer" refers to a means for using a generative AI model to analyze the user's question and perceived emotional state and generate an appropriate answer.

[0814] The "means for displaying the answer to the user" refers to a means for displaying the generated answer on the user's device so that the user can easily view the required information.

[0815] "Online portal" refers to a website or dedicated application that provides information about support programs for businesses.

[0816] This invention is a system that collects and learns information about corporate support systems and provides appropriate answers to user questions. This system is mainly composed of three elements: a server, a terminal, and a user.

[0817] Data collection and learning

[0818] The server collects documents related to employee support programs from company administrators. This collection is done using an FTP server or API. The documents include employee benefit programs, special leave policies, medical expense subsidies, etc. The collected documents are saved in a specific directory.

[0819] The server extracts the stored document, removes unnecessary characters and formatting, and formats it into a format that is easy to parse. This formatted data is then fed into a generative AI model (e.g., ChatGPT) to train the model. Once trained, the generative AI model is stored on the server and ready to respond to user queries.

[0820] Accepting user questions

[0821] Users access the company's online portal or dedicated application and enter specific questions (e.g., "What kind of support is available if I get the flu?") into the question input form of the "Smart Corporate System Navigator."

[0822] The terminal (user's device) sends the entered questions to the server in real time, and the communication is secure because it is done via the HTTPS protocol.

[0823] Emotion recognition by emotion engine

[0824] The device uses a camera and microphone to collect facial expressions, voice, and contextual data during user input. It then transmits the collected emotion data to the emotion engine in real time. The emotion engine uses a deep learning model to analyze the collected data and recognize the user's emotional state (e.g., stress, anxiety, relief, etc.). The recognized emotion data is then sent to the server.

[0825] Question analysis and answer generation

[0826] The server uses a generative AI model to analyze the received question data, understand the intent of the question, and extract keywords and key phrases (e.g., "influenza" and "assistance"). The server then searches documents related to relevant assistance programs and systems based on the extracted keywords and organizes the most relevant information.

[0827] The server takes emotion data into account and uses a generative AI model to generate the optimal answer. For example, if a user is anxious, the server might generate a gentle response such as, "Employees can take special paid leave and medical expense assistance programs are also available. If you have any questions, please contact our support desk."

[0828] Submitting and viewing answers

[0829] The server then sends the generated answer to the user's device, again securely via the HTTPS protocol, where it displays the answer to the user, allowing them to quickly view details of the assistance or support they require.

[0830] Specific examples

[0831] Example 1: Healthcare support inquiry

[0832] Users access an online portal and type in the question, "What help is available if I have the flu?"

[0833] The terminal transmits the question data to the server in real time.

[0834] The device transmits the user's facial expression and voice data to the emotion engine, which then recognizes that the user is feeling anxious.

[0835] The terminal transmits the emotion data to the server.

[0836] The server analyzes the question data and emotion data and searches documents for information related to "influenza."

[0837] The server considers the anxious user and generates a gentle response saying, "Employees can take special paid leave and medical expense assistance programs are also available. If you have any questions, please contact our support desk."

[0838] The server sends the generated answer to the user's device, which displays it. The user understands the necessary support information and can use the service with peace of mind.

[0839] Example 2: Benefits Program Inquiry

[0840] The user types the question, "What support is available for childcare leave?"

[0841] The terminal transmits the question data to the server in real time.

[0842] The terminal transmits the user's voice data to the emotion engine, which recognizes that the user is calm.

[0843] The terminal transmits the emotion data to the server.

[0844] The server analyzes the question data and emotion data and searches documents for information related to "child care leave."

[0845] For a calm user, the server generates a clear answer: "If you take childcare leave, you will be granted special leave and some medical expenses will be subsidized."

[0846] The server sends the generated answer to the user's device, which displays it, allowing the user to understand and use the corresponding support information appropriately.

[0847] Prompt Sentence Examples

[0848] "Please tell me about the company's support system if I get the flu. Please speak in a reassuring tone, as users are feeling anxious."

[0849] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0850] Step 1:

[0851] The server collects documents related to the support system from company administrators. This collection is often done via an FTP server or API. The input is the company's document data, and the output is a document file saved in a specific directory.

[0852] Step 2:

[0853] The server extracts the stored document files, removes unnecessary characters and formatting, and formats them into a format that is easy to analyze. Specifically, it uses text analysis tools to remove noise from the data and generate formatted data. The input is the document file read from storage, and the output is formatted data that can be used as training data for the generative AI model.

[0854] Step 3:

[0855] The server inputs the shaped data into a generative AI model for training. A generative AI model (such as ChatGPT) is used for this purpose. The trained generative AI model is stored on the server. The input is the shaped data, and the output is the trained generative AI model.

[0856] Step 4:

[0857] A user accesses a company's online portal or dedicated application and enters a specific question into a question input form. For example, "What kind of support is available if I get the flu?" The input is the user's question text, and the output is data sent from the terminal to the server.

[0858] Step 5:

[0859] The terminal transmits the entered question to the server in real time, securely using the HTTPS protocol. The input is the user's question text, and the output is the data sent to the server.

[0860] Step 6:

[0861] The device uses a camera and microphone to collect facial, voice, and contextual data during user input. This data is sent to the emotion engine in real time for preprocessing. The input is the user's emotion data, and the output is preprocessed emotion data.

[0862] Step 7:

[0863] The emotion engine analyzes the collected emotion data using a deep learning model to recognize the user's emotional state (e.g., stress, anxiety, relief, etc.). The input is the preprocessed emotion data, and the output is the recognized emotional state.

[0864] Step 8:

[0865] The server performs analysis based on the received question data and recognized emotion data. It uses a generative AI model to understand the intent of the question and extract keywords and important phrases (e.g., "influenza" or "assistance"). The input is the question data and emotion data, and the output is the extracted keywords and semantic data.

[0866] Step 9:

[0867] The server searches documents related to related support programs and systems based on the extracted keywords and organizes the most relevant information. The input is the extracted keywords, and the output is organized information data.

[0868] Step 10:

[0869] The server takes into account the emotional data and uses a generative AI model to generate the optimal answer. For example, to an anxious user, it generates a gentle answer such as, "Employees can take special paid leave, and medical expense subsidies are also available. If you have any questions, please contact our support desk." The input is organized information data and emotional data, and the output is the generated answer text.

[0870] Step 11:

[0871] The server sends the generated answer to the user's terminal, also securely using the HTTPS protocol. The input is the generated answer text, and the output is the data sent to the terminal.

[0872] Step 12:

[0873] The terminal displays the received answer to the user, allowing the user to quickly check the details of the assistance or support required. The input is the received answer text, and the output is the display data in a format that the user can view.

[0874] (Application example 2)

[0875] 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."

[0876] In information provision systems for corporate support systems, it is difficult for employees to quickly obtain appropriate information tailored to their own situation. Furthermore, there are currently few systems that provide personalized answers that reflect the emotional state of employees when they ask questions. This issue is particularly important in factories, where providing immediate support information directly leads to an improvement in the working environment.

[0877] 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.

[0878] In this invention, the server includes means for collecting information on corporate support programs, means for formatting and preprocessing the information, means for having a generative model that learns based on the information, means for accepting user questions, means for analyzing the questions, means for searching for relevant information, means for generating answers based on the searched information, means for displaying the generated answers to the user, emotion analysis means for recognizing the user's emotions, and means for adjusting the format and content of the answers based on the emotion data recognized by the emotion analysis means. This makes it possible to instantly provide appropriate support information according to the emotions of employees when they ask questions.

[0879] "Corporate support systems" is a general term for various systems such as employee benefits, special leave, and medical expense subsidies that companies provide to their employees.

[0880] "Means of collecting information" refers to the methods and tools used to collect documents and data related to support systems within the company.

[0881] "Information formatting and preprocessing means" refers to methods and tools used to convert collected information into a format that is easy to analyze.

[0882] A "generative model" refers to a computer program or algorithm that learns from collected information and generates appropriate answers to user questions.

[0883] "Means for accepting questions" refers to an interface or application that allows a user to input questions to the server.

[0884] "Means for analyzing questions" refers to methods and tools that use natural language processing technology to analyze questions from users and extract intent and keywords.

[0885] "Means for retrieving relevant information" refers to methods and tools for retrieving relevant information from a database based on the analyzed question.

[0886] "Answer generation means" refers to a computer program or algorithm that generates an appropriate answer to a user's question based on the retrieved information.

[0887] "Means for displaying to the user" refers to a method or tool for displaying the generated answer on a device so that the user can review it.

[0888] "Emotion analysis means" refers to technologies and algorithms for recognizing and analyzing a user's emotional state.

[0889] "Means for adjusting the format and content of responses based on emotional data" refers to methods or tools for adjusting the tone and content of responses according to the user's emotional state based on emotional data obtained by the emotion analysis means.

[0890] To implement this invention, the system consists of three main components: a server, a terminal, and a user. The server collects information about companies' support programs and performs learning using a generative AI model. The terminal accepts user questions, performs sentiment analysis, and transmits the question and sentiment data to the server. Users access the system through an online portal or a dedicated application.

[0891] server

[0892] The server processes the information as follows:

[0893] Information gathering: Collect documents from company managers regarding support programs, including employee benefits, special leave, and medical assistance.

[0894] Preprocessing and formatting: Format and preprocess the collected documents and feed them into a generative AI model (e.g., OpenAI's GPT-3).

[0895] Learning: Generative AI models are used to learn from collected information.

[0896] Question analysis: Question data sent from the device is analyzed using natural language processing technology to extract important keywords.

[0897] Information retrieval: Search for information from related documents based on the analysis results.

[0898] Answer generation: Using a generative AI model, appropriate answers are generated based on retrieved information and sentiment data.

[0899] Prepare for display: Send the generated answer to the device.

[0900] Terminal

[0901] The terminal works as follows:

[0902] Question reception: Receive questions from users and convert them into data.

[0903] Sentiment analysis: The emotion analysis engine analyzes the entered question as well as emotional data such as the user's facial expressions and voice.

[0904] Data transmission: Question data and emotion data are sent to the server.

[0905] Display Answer: The answer sent from the server is displayed to the user.

[0906] User

[0907] The user uses the system as follows:

[0908] Access: Access an online portal or dedicated application to enter your questions.

[0909] Enter a question: Enter a specific question and wait for the system to answer.

[0910] Specific examples

[0911] Example 1: Healthcare support inquiry

[0912] Question: User types, "I'm sick, can I leave work early?"

[0913] Sentiment analysis: Recognize when a user is feeling anxious.

[0914] Answer generation: The server responds, "You can leave work early by using special paid leave. Please contact our support desk for details."

[0915] Example 2: Querying rest facilities

[0916] Question: User types, "What rest stop is available when I finish work today?"

[0917] Sentiment analysis: Recognize that the user is calm.

[0918] Answer generation: The server responds, "After you finish work, break rooms A and B will be available."

[0919] Prompt Sentence Examples

[0920] Example 1 prompt statement:

[0921] User Question: Can I leave work early because I'm not feeling well?

[0922] User Emotion: Anxiety

[0923] Generate the appropriate answer.

[0924] Example 2 prompt statement:

[0925] User question: What rest areas can I use when I'm done with today's work?

[0926] User Sentiment: Calm

[0927] Generate the appropriate answer.

[0928] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0929] Step 1:

[0930] The server collects documents related to the company's support systems, such as employee benefit programs, special leave regulations, and medical expense subsidies provided by company administrators, and stores them in a database.

[0931] Input: Documents about corporate support programs

[0932] Output: The formatted and preprocessed document data

[0933] Step 2:

[0934] The server formats and preprocesses the collected documents, specifically converting the text data format and removing unnecessary data to make it easier for the generative AI model to process.

[0935] Input: Data from collected documents

[0936] Output: Preprocessed data that can be input into a generative AI model

[0937] Step 3:

[0938] The server uses the preprocessed data to train a generative AI model. Specifically, it uses OpenAI's GPT-3 model to learn information and build a knowledge base for generating appropriate answers to user questions.

[0939] Input: Preprocessed data

[0940] Output: Knowledge base based on generative AI models

[0941] Step 4:

[0942] Users enter their questions through a dedicated application or online portal. Specifically, users enter text-based questions, which are then collected as digital data by the device.

[0943] Input: User question (e.g., "I'm not feeling well, can I leave early?")

[0944] Output: Digitized question data

[0945] Step 5:

[0946] The device performs emotion analysis to recognize the user's emotions, specifically by analyzing the user's facial expressions, voice, or other input data to identify their emotional state (e.g., stress, anxiety, calm).

[0947] Input: User facial expressions, voice, and contextual data

[0948] Output: Recognized emotion data (e.g., "anxiety")

[0949] Step 6:

[0950] The terminal transmits the question data and emotion data to the server. Specifically, the terminal transmits the digitized question data and the recognized emotion data to the server as a data packet.

[0951] Input: Question data, emotion data

[0952] Output: Send data to the server

[0953] Step 7:

[0954] The server analyzes the question using a generative AI model based on the question data and sentiment data received, and extracts important keywords. Specifically, it uses natural language processing technology to understand the intent of the question and identify related keywords.

[0955] Input: Question data, emotion data

[0956] Output: Extracted keywords (e.g., "Leave early", "health")

[0957] Step 8:

[0958] The server searches for relevant information from related documents based on the extracted keywords, using a database search algorithm to identify and organize relevant information.

[0959] Input: Extracted keywords

[0960] Output: Relevant information (e.g. special paid leave regulations for early departure)

[0961] Step 9:

[0962] The server generates answers using a generative AI model based on search results and emotional data. Specifically, for users in an anxious state, it creates answers in a gentle tone that include appropriate support information.

[0963] Input: Search results, emotion data

[0964] Output: Generated answer (e.g. "You can leave early using special paid leave. Please contact support for details.")

[0965] Step 10:

[0966] The server sends the generated answer to the terminal, and the terminal displays it to the user, specifically, displays the answer text on the user's device, quickly meeting the user's needs.

[0967] Input: Generated Answer

[0968] Output: Answer displayed on the user's terminal

[0969] 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.

[0970] 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.

[0971] 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.

[0972] [Third embodiment]

[0973] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0974] 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.

[0975] 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).

[0976] 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.

[0977] 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.

[0978] 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).

[0979] 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.

[0980] 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.

[0981] 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.

[0982] 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.

[0983] 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.

[0984] 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."

[0985] This system collects and learns information about corporate support systems and provides appropriate answers to user questions. This system is mainly composed of three elements: a server, a terminal, and a user.

[0986] Data collection and learning

[0987] First, the server collects documents from company administrators about the company's support systems. These documents include employee benefit programs, special leave policies, medical expense subsidies, etc. The collected documents are formatted and preprocessed before being input into a generative AI (e.g., ChatGPT). The server uses this generative AI to learn information and prepare it to generate answers to questions.

[0988] Accepting user questions

[0989] Next, users access the Smart Corporate System Navigator through their company's online portal or dedicated application, and enter a specific question in the search box, such as, "What kind of support is available if I get the flu?"

[0990] Question analysis and answer generation

[0991] The device sends the entered question to the server, which uses generative AI to analyze the question and understand its intent. Based on the results of the analysis, the server searches for relevant information in the company's documents. For example, if the question contains the keyword "influenza," the search will return "health management support documents" and the like.

[0992] Based on the retrieved information, the server generates a detailed answer, such as, "If employees catch the flu, they can use special paid leave. Medical expense assistance programs are also available."

[0993] Show Answers

[0994] Finally, the server sends the generated answer to the terminal, which then displays the answer to the user. The user can view the displayed answer, understand the support or assistance they need, and use it appropriately.

[0995] Specific examples

[0996] Example 1: Healthcare support inquiry

[0997] Users access an online portal and type in the question, "What help is available if I have the flu?"

[0998] The terminal sends this question to the server.

[0999] The server uses generative AI to analyze the question and search documents for information related to "influenza."

[1000] The server generates a response that "employees can take special paid leave and also have access to medical expense assistance."

[1001] The terminal displays the generated answer, and the user understands the necessary support information and uses it appropriately.

[1002] Example 2: Benefits Program Inquiry

[1003] The user types the question, "What support is available for childcare leave?"

[1004] The terminal sends this question to the server.

[1005] The server analyzes the query and retrieves the relevant assistance program documentation.

[1006] The server generates a response saying, "If you take child care leave, you will be granted special leave and some medical expenses will be subsidized."

[1007] The terminal displays the generated answer, and the user uses the corresponding support information.

[1008] In this way, the present invention provides a system that efficiently collects and learns information about corporate support systems and provides quick and accurate answers to user questions.

[1009] The processing flow will be explained below.

[1010] Step 1: Data entry and learning phase

[1011] The server receives documents related to support systems from company administrators. Specifically, it collects "support documents related to health management," "personnel system guidebooks," "details of employee benefit programs," etc.

[1012] The server formats the documents it receives and performs data cleaning if necessary, which includes standardizing the document format and removing unnecessary textual information.

[1013] The server feeds the formatted documents into a generative model (e.g., ChatGPT) and trains the model, which then has detailed knowledge of the company's systems.

[1014] Step 2: Accepting questions from users

[1015] Users access the Smart Corporate System Navigator through their company's online portal or application.

[1016] The user enters a specific question into the question entry form on the portal site, for example, "What kind of support is available if I get the flu?"

[1017] The terminal (user's device) sends the entered question data to the server.

[1018] Step 3: Parsing the Question

[1019] The server analyzes the received question data. Using a generative AI model, it understands the intent of the question and extracts keywords and important phrases. For example, "influenza" and "support" are extracted as important keywords.

[1020] Step 4: Find related information

[1021] The server uses the extracted keywords to search for relevant information from related documents, such as "health management support documents" and "personnel system guidebooks."

[1022] The server sorts through the search results and extracts the most relevant parts.

[1023] Step 5: Generate an answer

[1024] The server generates answers to provide to users based on the extracted information. The generation AI summarizes the information and adjusts the writing style to create easy-to-understand sentences.

[1025] For example, it generates a specific answer such as "If employees catch the flu, they can use special paid leave. Medical expense assistance systems are also available."

[1026] Step 6: Submit and view your responses

[1027] The server sends the generated answer to the user's terminal.

[1028] The terminal displays the received response to the user, allowing the user to quickly ascertain details of the assistance or assistance required.

[1029] In this way, by performing specific processing at each step, a system is provided that allows users to efficiently obtain and utilize information about corporate systems and support programs.

[1030] Example 1

[1031] 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."

[1032] Information about corporate support programs is diverse, making it difficult for employees to quickly obtain the information they need. Furthermore, if the information provided is inaccurate, employees may not receive appropriate support, hindering efficient work performance. In addition, there is the problem that technology for accurately analyzing questions entered by users in natural language and generating appropriate answers is still immature. Furthermore, if the generated answers are not displayed clearly to users, user convenience is reduced.

[1033] 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.

[1034] In this invention, the server includes: means for collecting information about corporate support programs; means for formatting and preprocessing the information; means for having a generative model that learns based on the information; means for accepting user questions; means for analyzing the questions; means for searching for relevant information; means for generating answers based on the searched information; means for displaying the generated answers to the user; means for removing unnecessary parts and extracting only important information using a natural language processing tool when formatting and preprocessing the information; means for generating detailed answers to the user questions based on prompt sentences using a generative AI model; and means for displaying the answers generated based on the prompt sentences in a user-friendly format. This makes it possible to quickly and accurately obtain information about corporate support programs, generate appropriate answers to questions entered in natural language, and display the answers in an easy-to-understand format to the user.

[1035] "Means for collecting information" refers to devices and methods for receiving and storing data relating to the company's support system from the company administrator.

[1036] "Information formatting and preprocessing means" refers to devices and methods that use natural language processing tools to convert collected data into an appropriate format and remove unnecessary parts.

[1037] "Means having a learning generative model" refers to a device or method that uses a generative AI model to learn information based on preprocessed data.

[1038] The "means for accepting a user's question" refers to a device or method that provides a function for a user to input a question in natural language and receives the input content.

[1039] The "means for analyzing a question" refers to a device or method that uses natural language processing technology to analyze a question received from a user and understand its intent.

[1040] The "means for retrieving relevant information" refers to a device or method for retrieving relevant information from relevant company documents or databases based on the analyzed query.

[1041] A "means for generating an answer" is a device or method that uses a generative AI model to create a detailed answer based on the searched information.

[1042] The "means for displaying the generated answers to the user" refers to a device or method for displaying the generated answers in a user-friendly format so that the user can view them.

[1043] "Means using natural language processing tools" refers to the techniques and devices used to analyze and format text when preprocessing collected data.

[1044] A "means for generating a detailed answer based on a prompt sentence" is a device or method that uses a generative AI model to input a prompt corresponding to a specific question and generate a detailed answer based on that prompt.

[1045] The "means for displaying in a user-friendly format" refers to a device or method that provides the generated answers to the user in a format that is easy to view and understand.

[1046] This system collects and learns information about corporate support systems and provides appropriate answers to user questions. This system is mainly composed of three elements: a server, a terminal, and a user.

[1047] Data collection and learning

[1048] The server receives documents related to the company's support system from the company administrator. The received documents are saved as text files or PDFs. The saved documents are preprocessed using an NLP tool (e.g., SpaCy, NLTK). Preprocessing refers to removing unnecessary parts and extracting important information. The formatted data from this process is input into a generative AI model (e.g., ChatGPT). The server uses this generative AI model to learn information and prepare to generate answers to user questions.

[1049] Accepting user questions

[1050] Users log in to their company's online portal or dedicated application to access the "Smart Corporate System Navigator." They enter a specific question into the search box, for example, "What kind of support is available if I have the flu?" The device then sends this input to the server. The transmitted data is protected using a secure communication protocol (e.g., HTTPS).

[1051] Question analysis and answer generation

[1052] The server converts the received question into a prompt to be input into the generative AI. Using the generative AI model, the server analyzes the intent of the user's question and extracts related keywords. Based on the results of this analysis, the server searches for relevant information from the company's documents. For example, if the keyword "influenza" is included, support documents related to health management will be searched for. Based on the search results, the generative AI model is used to generate a detailed answer. An example of a specific answer would be: "If employees catch the flu, they can use special paid leave. Medical expense subsidy systems are also available."

[1053] Show Answers

[1054] The server sends the generated answer to the terminal, which displays the answer in a user-friendly format (e.g., a card or pop-up).The user can view the displayed answer, understand the support or assistance they need, and use it appropriately.

[1055] Specific examples

[1056] Example 1: Healthcare support inquiry

[1057] The user accesses the online portal and enters the question, "What support is available if I get the flu?" The device then sends this question to the server. The server uses generative AI to analyze the question and searches documents for information related to "flu." The server then generates the answer, "Employees can use special paid leave and medical expense subsidy systems." The user then understands the necessary support information and uses it appropriately.

[1058] Example 2: Benefits Program Inquiry

[1059] The user inputs the question, "What kind of support is available for child care leave?" The device sends this question to the server. The server analyzes the question and searches for documents on related support programs. The server generates an answer that says, "If you take child care leave, you will be granted special leave and some medical expenses will be subsidized." The device displays the generated answer, allowing the user to use the relevant support information appropriately.

[1060] This system can efficiently collect and learn information about corporate support systems and provide quick and accurate answers to users' questions.

[1061] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1062] Step 1:

[1063] The server receives documents relating to the company's support system from the company administrator.

[1064] Specifically, the document is collected using the file upload function.

[1065] Input: Support program documentation provided by company administrator.

[1066] Output: Saved support system documents.

[1067] Step 2:

[1068] The server saves the received documents in text file or PDF format.

[1069] Specifically, the uploaded file is saved in an appropriate folder or database.

[1070] Input: Uploaded assistance system document.

[1071] Output: Saved text file and PDF.

[1072] Step 3:

[1073] The server preprocesses the stored documents using natural language processing (NLP) tools (e.g., SpaCy, NLTK).

[1074] Specifically, it removes unnecessary parts from the document (e.g., page numbers, headers, footers) and extracts important information.

[1075] Input: Saved text files or PDFs.

[1076] Output: Preprocessed text data.

[1077] Step 4:

[1078] The server formats the preprocessed data and generates prompts to input into a generative AI model (e.g., ChatGPT).

[1079] Specifically, the process creates a prompt statement template based on the preprocessed data.

[1080] Input: Preprocessed text data.

[1081] Output: The prompt sentence to be input to the generative AI model.

[1082] Step 5:

[1083] The server uses a generative AI model to learn information and prepare to generate answers to user questions.

[1084] Specifically, the prompt sentence is input into the generative AI model and the model is trained.

[1085] Input: prompt statement.

[1086] Output: A trained generative AI model.

[1087] Step 6:

[1088] Users log in to the online portal or dedicated application and access the "Smart Corporate System Navigator."

[1089] Specifically, the user authentication and login process are carried out.

[1090] Input: User credentials.

[1091] Output: Authenticated user interface.

[1092] Step 7:

[1093] Users enter a specific question into a search box.

[1094] Specifically, you enter a question in the search box and press the send button.

[1095] Input: The question text provided by the user.

[1096] Output: The question data sent to the terminal.

[1097] Step 8:

[1098] The terminal transmits the entered question to the server.

[1099] Specifically, the input data is transferred to the server using a secure communication protocol (e.g., HTTPS).

[1100] Input: Question data from the user.

[1101] Output: The query data sent to the server.

[1102] Step 9:

[1103] The server converts the received question into a prompt sentence to be input to the generation AI.

[1104] Specifically, the question data is formatted into an appropriate prompt format.

[1105] Input: Question data from the user.

[1106] Output: The prompt statement.

[1107] Step 10:

[1108] The server uses generative AI to analyze the intent of the question and extract relevant keywords.

[1109] Specifically, the prompt sentence is input into the generative AI model, and intent analysis and keyword extraction are performed.

[1110] Input: prompt statement.

[1111] Output: Parsed intent and extracted keywords.

[1112] Step 11:

[1113] The server searches for company documents based on the extracted keywords.

[1114] Specifically, the operation involves searching for the relevant information from a document management system or database.

[1115] Input: Parsed intent and extracted keywords.

[1116] Output: Search results (relevant documents).

[1117] Step 12:

[1118] The server generates detailed answers based on the search results using a generative AI model.

[1119] Specifically, the search results are input into a generative AI model to generate a detailed answer.

[1120] Input: Search results.

[1121] Output: The detailed answer generated.

[1122] Step 13:

[1123] The server sends the generated response to the terminal.

[1124] Specifically, the generated answer is formatted into a user-friendly format and sent to the terminal.

[1125] Input: The generated detailed answer.

[1126] Output: The response data sent to the device.

[1127] Step 14:

[1128] The device displays the received answers in a user-friendly format (e.g., card format or popup).

[1129] Specifically, the response data is read and displayed on the user interface.

[1130] Input: Response data sent to the terminal.

[1131] Output: The answer displayed to the user.

[1132] Step 15:

[1133] The user can view the displayed answers, understand the support and assistance they need, and use it appropriately.

[1134] Specifically, the displayed information is checked and the next action is taken as necessary.

[1135] Input: The answer shown to the user.

[1136] Output: User understanding and use of information.

[1137] (Application example 1)

[1138] 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."

[1139] There is a wide variety of information available about corporate support programs, making it difficult for employees to quickly and accurately identify the program that best suits them. Furthermore, especially in busy environments, such as store employees, who often lack the time to use digital devices, employees tend to neglect using support programs. This leads to a decline in employee utilization of support programs, resulting in the inability to fully realize the benefits of corporate employee benefits.

[1140] 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.

[1141] In this invention, the server includes: means for collecting information on corporate support programs; means for formatting and preprocessing the information; means for having a generative model that learns based on the information; means for accepting user questions; means for analyzing the questions; means for searching for relevant information; means for generating answers based on the searched information; means for displaying the generated answers to the user; a voice input means installed in the smart device; means for converting voice to text; means for transmitting the text questions to the server; and means for displaying the generated answers on the smart device. This allows employees to easily ask questions through voice input and quickly access information on appropriate support programs.

[1142] "Corporate support systems" is a general term for welfare programs, special leave provisions, medical expense subsidies, etc. that companies provide to their employees.

[1143] "Means for collecting information" refers to a method or system for collecting documents related to the support system from company administrators.

[1144] "Information formatting and preprocessing means" refers to methods and systems for converting collected documents into a format that is easy to analyze and for preprocessing the data.

[1145] A "generative model" is a machine learning model that learns from collected and preprocessed information and generates appropriate answers to user questions.

[1146] The "means for accepting a user's question" is an interface for receiving a question input by a user and transmitting it to the system.

[1147] The "means for analyzing questions" refers to natural language processing techniques and algorithms for analyzing received questions and understanding their intent.

[1148] The "means for searching relevant information" is a method or system for searching company documents for information relevant to the analyzed question.

[1149] "Answer generation means" refers to a method or system for generating a specific answer based on the retrieved information.

[1150] The "means for displaying the generated answer to the user" refers to a method or system for displaying the generated answer on the user's device (terminal).

[1151] "Smart devices" is a general term for devices that can connect to the Internet and run various applications, such as smartphones, smart glasses, and head-mounted displays.

[1152] The "voice input means" is a function for inputting the user's voice using a microphone or the like installed on the smart device.

[1153] "Means for converting speech to text" refers to speech recognition technology and algorithms for analyzing input speech and converting it into text data.

[1154] The "means for transmitting a question converted into text to a server" is a communication function for transmitting a user's question converted into text by voice recognition to a server.

[1155] The "means for displaying the answer on the smart device" is a function for receiving the answer generated from the server and displaying it on the display screen of the smart device.

[1156] This system collects and learns information about corporate support programs and provides appropriate answers to user questions. This system is composed of three elements: a server, a terminal, and a user.

[1157] First, the server collects documents from company administrators about the company's support systems, including employee benefit programs, special leave policies, medical expense subsidies, etc. The collected documents undergo formatting and preprocessing, and then the generative AI learns the information and prepares them to generate answers to questions.

[1158] Next, the user accesses the company's smart device and asks a question through voice input, for example, "What kind of support is available if I have the flu?" The microphone installed on the smart device captures this voice and converts it into text using voice recognition technology.

[1159] The device sends the converted text data to the server, which uses generative AI to analyze the question and understand its intent. Based on the results of the analysis, the server searches for relevant information in the company's documents. For example, if a question contains the keyword "influenza," the search will return "health management support documents" and other relevant information.

[1160] Based on the searched information, the server generates a detailed answer. For example, it creates a specific answer such as, "If an employee catches the flu, they can use special paid leave. Medical expense subsidies are also available." The generated answer is sent to the smart device as text data and displayed on the screen. The user can view the displayed answer, understand the support and assistance they need, and use it appropriately.

[1161] Examples:

[1162] A user uses the smart glasses to ask a question out loud: "How much is the employee discount?" The question is converted into text using speech recognition, and the question "How much is the employee discount?" is sent to the server. The server uses a generative AI model to analyze the question and generates the answer: "The employee discount is 10% on all products." This answer is displayed on the smart glasses for the user to see immediately.

[1163] Example prompt sentence:

[1164] Q: What support is available for employees with children?

[1165] Answer: Employees with children are eligible for special parental leave, medical assistance, and assistance with childcare equipment.

[1166] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1167] Step 1:

[1168] The server collects documents related to the company's support system from the company administrator. The collected documents include information such as "employee benefit programs," "special leave regulations," and "medical expense subsidies." These documents are stored on the server and used for subsequent processing.

[1169] Input: Company support program document provided by company administrator

[1170] Output: A database of collected support system documents

[1171] Step 2:

[1172] The server formats and preprocesses the collected documents, removing unnecessary information and using natural language processing (NLP) techniques to structure the text, converting it into a format suitable for input into the generative AI model.

[1173] Input: Collected assistance system documents

[1174] Output: A formatted and preprocessed document

[1175] Step 3:

[1176] The server uses the preprocessed documents to train the generative AI model, specifically learning various information about the business support program and training the model to generate appropriate answers to questions.

[1177] Input: formatted and preprocessed document

[1178] Output: Trained generative AI model

[1179] Step 4:

[1180] A user accesses a smart device and uses voice input to ask a question, for example, "What can I do to help if I get the flu?" The smart device uses a microphone to capture the user's voice.

[1181] Input: User's voice question

[1182] Output: Captured audio data

[1183] Step 5:

[1184] The device uses speech recognition technology to convert the captured speech into text, specifically by using a speech recognition library to analyze the speech data and convert it into the appropriate text.

[1185] Input: Captured audio data

[1186] Output: Texted question

[1187] Step 6:

[1188] The device sends a textual question to the server, which receives the question and prepares it for processing by the generative AI model.

[1189] Input: Texted question

[1190] Output: The question sent to the server

[1191] Step 7:

[1192] The server uses a generative AI model to analyze the question and understand its intent. Specifically, it uses NLP technology to identify keywords and phrases in the question and analyze their meaning.

[1193] Input: Texted question

[1194] Output: Parsed intent (keywords and phrases in the question)

[1195] Step 8:

[1196] The server searches for relevant information from the company's documents based on the analyzed intent. For example, if the keyword "influenza" is included, it searches for support documents related to health care.

[1197] Input: Parsed intent

[1198] Output: Searched support system information

[1199] Step 9:

[1200] The server uses the searched information to generate a specific answer to the question, such as "If employees catch the flu, they can use their paid leave. Medical expense assistance is also available."

[1201] Input: Searched support system information

[1202] Output: The specific answer generated

[1203] Step 10:

[1204] The server sends the generated answer as text data to the terminal, which then displays the received answer on a smart device, such as a smart eyeglasses display.

[1205] Input: Generated specific answer

[1206] Output: Answers displayed on a smart device

[1207] Through the above steps, information about corporate support programs can be provided to users quickly and accurately.

[1208] 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.

[1209] This system combines a system that collects and learns information about corporate support systems, provides appropriate answers to user questions, and an emotion engine that recognizes user emotions. This system is primarily composed of three elements: a server, a terminal, and a user.

[1210] Data collection and learning

[1211] First, the server collects documents about support systems from company administrators. These documents include employee benefit programs, special leave policies, medical expense subsidies, etc. The collected documents are formatted and preprocessed before being input into a generative AI (e.g., ChatGPT). The server uses this generative AI to learn information and prepare to generate answers to user questions.

[1212] Accepting user questions

[1213] Next, the user accesses the "Smart Corporate System Navigator" through the company's online portal or a dedicated application. The user enters a specific question into the question input form on the portal site. For example, the user enters a question such as "What kind of support is available if I get the flu?" The terminal (user's device) then sends the entered question data to the server.

[1214] Emotion recognition by emotion engine

[1215] The device sends the user's facial expressions, voice, and context in real time to the emotion engine, which analyzes this data and recognizes the user's emotional state (e.g., stress, anxiety, relief, etc.). The recognized emotion data is then sent to the server.

[1216] Question analysis and answer generation

[1217] The server analyzes the question using a generative AI model based on the received question data and sentiment data. It understands the intent of the question and extracts keywords and important phrases. For example, "influenza" and "assistance" are extracted as important keywords. The server then searches for information from related documents based on the extracted keywords, organizes the search results, and extracts the most relevant parts.

[1218] The server uses the emotional data to tailor the format and content of the response to suit the user. For example, if the user is feeling anxious, the tone of the response will be gentler and phrased in a way that provides a sense of security. A specific response generated might be, "If you catch the flu, employees can use special paid leave and medical expense assistance. If you have any questions, please contact our support desk."

[1219] Submitting and viewing answers

[1220] The server sends the generated answer to the user's terminal, which displays the received answer to the user, allowing the user to quickly confirm details of the help or assistance required.

[1221] Specific examples

[1222] Example 1: Healthcare support inquiry

[1223] Users access an online portal and type in the question, "What help is available if I have the flu?"

[1224] The device sends emotional data indicating anxiety based on the user's voice and facial expression during input to the emotion engine, which then analyzes the data and recognizes anxiety.

[1225] The terminal transmits the question data and emotion data to the server.

[1226] The server uses generative AI to analyze the question and search documents for information related to "influenza."

[1227] The server takes into account the user's anxious state and generates a gentle response saying, "Employees can take special paid leave and medical expense assistance programs are also available. If you have any questions, please contact our support desk."

[1228] The terminal displays the generated answer, and the user understands the necessary support information and uses the device with peace of mind.

[1229] Example 2: Benefits Program Inquiry

[1230] The user types the question, "What support is available for childcare leave?"

[1231] The terminal sends the user's voice data to the emotion engine, which recognizes that the user is calm.

[1232] The terminal transmits the question data and emotion data to the server.

[1233] The server analyzes the query and retrieves information from the documentation of the relevant assistance program.

[1234] The server generates a clear answer for the calm user: "If you take childcare leave, you will be granted special leave and some medical expenses will be subsidized."

[1235] The terminal displays the generated answer, and the user uses the corresponding support information.

[1236] In this way, the present invention realizes a system that provides more appropriate and personalized answers by combining an emotion engine that recognizes the user's emotions.

[1237] The processing flow will be explained below.

[1238] Step 1: Data entry and learning phase

[1239] The server collects documents about support systems from company administrators, including details of employee benefit programs, personnel system guidebooks, and medical assistance.

[1240] The server formats the collected documents and cleans them of unnecessary information, standardizing the document format and removing noise.

[1241] The server inputs the formatted documents into a generative model (e.g., ChatGPT) and trains the model, which accumulates knowledge about the support system in the generative AI.

[1242] Step 2: Accepting questions from users

[1243] Users access the Smart Corporate System Navigator through their company's online portal or application.

[1244] The user enters a specific question (e.g., "What kind of support is available if I have the flu?") into an input form.

[1245] The terminal transmits the input question data to the server in real time.

[1246] Step 3: Emotion recognition by the emotion engine

[1247] The device sends facial, vocal, and contextual information to the emotion engine as the user inputs, including camera and microphone data.

[1248] The server uses an emotion engine to analyze this data and recognize the user's emotional state (e.g., anxiety, relief, anger, etc.).

[1249] The recognized emotion data is sent to a server for further analysis.

[1250] Step 4: Parsing the Question

[1251] The server uses generative AI to analyze the question entered by the user and extract important keywords and phrases from the question (e.g., "influenza" and "assistance").

[1252] The server searches for the necessary information from related documents based on the extracted keywords.

[1253] Step 5: Generate an answer

[1254] Based on the searched information, the server uses a generative AI model to generate answers for the user.

[1255] The server uses the recognized emotion data to customize the tone and content of the response. For example, if the user is feeling anxious, the response will be changed to a more reassuring tone.

[1256] A specific answer generated is "If an employee catches the flu, they can use special paid leave. They can also use the medical expense subsidy system."

[1257] Step 6: Submit and view your responses

[1258] The server sends the generated answer to the user's terminal.

[1259] The terminal displays the received response to the user, allowing the user to quickly check the details of the system or support they need.

[1260] Examples:

[1261] Example 1: Healthcare support inquiry

[1262] Users access an online portal and type in the question, "What help is available if I have the flu?"

[1263] The device sends the user's question data and emotional data, such as facial expressions and voice, to the emotion engine, which then recognizes anxiety.

[1264] The terminal transmits the question data and emotion data to the server.

[1265] The server parses the query and searches the documents for information related to "influenza."

[1266] The server generates a gentle response to the anxious user, saying, "Employees can take special paid leave and medical expense assistance programs are also available. Please feel free to receive support."

[1267] The terminal displays the generated answer, allowing the user to use appropriate support information with peace of mind.

[1268] Example 2: Benefits Program Inquiry

[1269] The user types the question, "What support is available for childcare leave?"

[1270] The terminal sends the user's voice and facial expression data to the emotion engine, which recognizes that the user is calm.

[1271] The terminal transmits the question data and emotion data to the server.

[1272] The server analyzes the query and retrieves relevant support program information from the document.

[1273] The server generates a clear answer for the calm user: "If you take childcare leave, you will be granted special leave and some medical expenses will be subsidized."

[1274] The terminal displays the generated answer and the user can use the corresponding support information.

[1275] In this way, the present invention realizes a system that provides more appropriate and personalized answers according to the user's emotional state by combining an emotion engine.

[1276] Example 2

[1277] 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."

[1278] Conventional information provision systems for business support programs have the problem that it is difficult for users to quickly and accurately obtain the detailed information they require. As a result, users often experience inconvenience because they are unable to quickly access the support and assistance they need. Furthermore, responses that do not take into account the user's emotional state may cause the user to feel anxious or stressed.

[1279] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting information on corporate support programs, means for formatting and preprocessing the information, means for having a generative AI model that learns based on the information, means for accepting a user's question, means for collecting facial expression, voice, and contextual data when asking a question, means for analyzing the emotional data to recognize the emotional state, means for analyzing the question and the recognized emotional state and generating an answer using the generative AI model, and means for displaying the generated answer to the user. This allows the user to quickly and accurately obtain information about the support programs they need and to receive a personalized answer that takes their emotional state into consideration.

[1280] "Corporate support systems" refers to various systems and programs that companies use to provide support to their employees, such as employee benefits, special leave, and medical expense subsidies.

[1281] "Means of collecting information" refers to the means and methods for obtaining documents and data related to the support system from company managers.

[1282] "Information formatting and preprocessing methods" refers to the means or methods for converting collected documents or data into a format that is easier to analyze and for removing unnecessary characters and formatting.

[1283] A "generative AI model" is an artificial intelligence model trained on large amounts of data, for example, to generate appropriate answers to user questions using natural language processing.

[1284] "Means for accepting user questions" refers to a means or interface for a user to input a specific question into the system and transmit that information to the server.

[1285] "Means for collecting facial, audio, and contextual data" refers to means for collecting data to understand the user's emotional state using a user's device, such as a camera or microphone.

[1286] "Means for analyzing emotional data and recognizing emotional states" refers to means for analyzing and recognizing a user's emotional state based on collected data using a deep learning model or the like.

[1287] "Means for generating an answer" refers to a means for using a generative AI model to analyze the user's question and perceived emotional state and generate an appropriate answer.

[1288] The "means for displaying the answer to the user" refers to a means for displaying the generated answer on the user's device so that the user can easily view the required information.

[1289] "Online portal" refers to a website or dedicated application that provides information about support programs for businesses.

[1290] This invention is a system that collects and learns information about corporate support systems and provides appropriate answers to user questions. This system is mainly composed of three elements: a server, a terminal, and a user.

[1291] Data collection and learning

[1292] The server collects documents related to employee support programs from company administrators. This collection is done using an FTP server or API. The documents include employee benefit programs, special leave policies, medical expense subsidies, etc. The collected documents are saved in a specific directory.

[1293] The server extracts the stored document, removes unnecessary characters and formatting, and formats it into a format that is easy to parse. This formatted data is then fed into a generative AI model (e.g., ChatGPT) to train the model. Once trained, the generative AI model is stored on the server and ready to respond to user queries.

[1294] Accepting user questions

[1295] Users access the company's online portal or dedicated application and enter specific questions (e.g., "What kind of support is available if I get the flu?") into the question input form of the "Smart Corporate System Navigator."

[1296] The terminal (user's device) sends the entered questions to the server in real time, and the communication is secure because it is done via the HTTPS protocol.

[1297] Emotion recognition by emotion engine

[1298] The device uses a camera and microphone to collect facial expressions, voice, and contextual data during user input. It then transmits the collected emotion data to the emotion engine in real time. The emotion engine uses a deep learning model to analyze the collected data and recognize the user's emotional state (e.g., stress, anxiety, relief, etc.). The recognized emotion data is then sent to the server.

[1299] Question analysis and answer generation

[1300] The server uses a generative AI model to analyze the received question data, understand the intent of the question, and extract keywords and key phrases (e.g., "influenza" and "assistance"). The server then searches documents related to relevant assistance programs and systems based on the extracted keywords and organizes the most relevant information.

[1301] The server takes emotion data into account and uses a generative AI model to generate the optimal answer. For example, if a user is anxious, the server might generate a gentle response such as, "Employees can take special paid leave and medical expense assistance programs are also available. If you have any questions, please contact our support desk."

[1302] Submitting and viewing answers

[1303] The server then sends the generated answer to the user's device, again securely via the HTTPS protocol, where it displays the answer to the user, allowing them to quickly view details of the assistance or support they require.

[1304] Specific examples

[1305] Example 1: Healthcare support inquiry

[1306] Users access an online portal and type in the question, "What help is available if I have the flu?"

[1307] The terminal transmits the question data to the server in real time.

[1308] The device transmits the user's facial expression and voice data to the emotion engine, which then recognizes that the user is feeling anxious.

[1309] The terminal transmits the emotion data to the server.

[1310] The server analyzes the question data and emotion data and searches documents for information related to "influenza."

[1311] The server considers the anxious user and generates a gentle response saying, "Employees can take special paid leave and medical expense assistance programs are also available. If you have any questions, please contact our support desk."

[1312] The server sends the generated answer to the user's device, which displays it. The user understands the necessary support information and can use the service with peace of mind.

[1313] Example 2: Benefits Program Inquiry

[1314] The user types the question, "What support is available for childcare leave?"

[1315] The terminal transmits the question data to the server in real time.

[1316] The terminal transmits the user's voice data to the emotion engine, which recognizes that the user is calm.

[1317] The terminal transmits the emotion data to the server.

[1318] The server analyzes the question data and emotion data and searches documents for information related to "child care leave."

[1319] For a calm user, the server generates a clear answer: "If you take childcare leave, you will be granted special leave and some medical expenses will be subsidized."

[1320] The server sends the generated answer to the user's device, which displays it, allowing the user to understand and use the corresponding support information appropriately.

[1321] Prompt Sentence Examples

[1322] "Please tell me about the company's support system if I get the flu. Please speak in a reassuring tone, as users are feeling anxious."

[1323] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1324] Step 1:

[1325] The server collects documents related to the support system from company administrators. This collection is often done via an FTP server or API. The input is the company's document data, and the output is a document file saved in a specific directory.

[1326] Step 2:

[1327] The server extracts the stored document files, removes unnecessary characters and formatting, and formats them into a format that is easy to analyze. Specifically, it uses text analysis tools to remove noise from the data and generate formatted data. The input is the document file read from storage, and the output is formatted data that can be used as training data for the generative AI model.

[1328] Step 3:

[1329] The server inputs the shaped data into a generative AI model for training. A generative AI model (such as ChatGPT) is used for this purpose. The trained generative AI model is stored on the server. The input is the shaped data, and the output is the trained generative AI model.

[1330] Step 4:

[1331] A user accesses a company's online portal or dedicated application and enters a specific question into a question input form. For example, "What kind of support is available if I get the flu?" The input is the user's question text, and the output is data sent from the terminal to the server.

[1332] Step 5:

[1333] The terminal transmits the entered question to the server in real time, securely using the HTTPS protocol. The input is the user's question text, and the output is the data sent to the server.

[1334] Step 6:

[1335] The device uses a camera and microphone to collect facial, voice, and contextual data during user input. This data is sent to the emotion engine in real time for preprocessing. The input is the user's emotion data, and the output is preprocessed emotion data.

[1336] Step 7:

[1337] The emotion engine analyzes the collected emotion data using a deep learning model to recognize the user's emotional state (e.g., stress, anxiety, relief, etc.). The input is the preprocessed emotion data, and the output is the recognized emotional state.

[1338] Step 8:

[1339] The server performs analysis based on the received question data and recognized emotion data. It uses a generative AI model to understand the intent of the question and extract keywords and important phrases (e.g., "influenza" or "assistance"). The input is the question data and emotion data, and the output is the extracted keywords and semantic data.

[1340] Step 9:

[1341] The server searches documents related to related support programs and systems based on the extracted keywords and organizes the most relevant information. The input is the extracted keywords, and the output is organized information data.

[1342] Step 10:

[1343] The server takes into account the emotional data and uses a generative AI model to generate the optimal answer. For example, to an anxious user, it generates a gentle answer such as, "Employees can take special paid leave, and medical expense subsidies are also available. If you have any questions, please contact our support desk." The input is organized information data and emotional data, and the output is the generated answer text.

[1344] Step 11:

[1345] The server sends the generated answer to the user's terminal, also securely using the HTTPS protocol. The input is the generated answer text, and the output is the data sent to the terminal.

[1346] Step 12:

[1347] The terminal displays the received answer to the user, allowing the user to quickly check the details of the assistance or support required. The input is the received answer text, and the output is the display data in a format that the user can view.

[1348] (Application example 2)

[1349] 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."

[1350] In information provision systems for corporate support systems, it is difficult for employees to quickly obtain appropriate information tailored to their own situation. Furthermore, there are currently few systems that provide personalized answers that reflect the emotional state of employees when they ask questions. This issue is particularly important in factories, where providing immediate support information directly leads to an improvement in the working environment.

[1351] 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.

[1352] In this invention, the server includes means for collecting information on corporate support programs, means for formatting and preprocessing the information, means for having a generative model that learns based on the information, means for accepting user questions, means for analyzing the questions, means for searching for relevant information, means for generating answers based on the searched information, means for displaying the generated answers to the user, emotion analysis means for recognizing the user's emotions, and means for adjusting the format and content of the answers based on the emotion data recognized by the emotion analysis means. This makes it possible to instantly provide appropriate support information according to the emotions of employees when they ask questions.

[1353] "Corporate support systems" is a general term for various systems such as employee benefits, special leave, and medical expense subsidies that companies provide to their employees.

[1354] "Means of collecting information" refers to the methods and tools used to collect documents and data related to support systems within the company.

[1355] "Information formatting and preprocessing means" refers to methods and tools used to convert collected information into a format that is easy to analyze.

[1356] A "generative model" refers to a computer program or algorithm that learns from collected information and generates appropriate answers to user questions.

[1357] "Means for accepting questions" refers to an interface or application that allows a user to input questions to the server.

[1358] "Means for analyzing questions" refers to methods and tools that use natural language processing technology to analyze questions from users and extract intent and keywords.

[1359] "Means for retrieving relevant information" refers to methods and tools for retrieving relevant information from a database based on the analyzed question.

[1360] "Answer generation means" refers to a computer program or algorithm that generates an appropriate answer to a user's question based on the retrieved information.

[1361] "Means for displaying to the user" refers to a method or tool for displaying the generated answer on a device so that the user can review it.

[1362] "Emotion analysis means" refers to technologies and algorithms for recognizing and analyzing a user's emotional state.

[1363] "Means for adjusting the format and content of responses based on emotional data" refers to methods or tools for adjusting the tone and content of responses according to the user's emotional state based on emotional data obtained by the emotion analysis means.

[1364] To implement this invention, the system consists of three main components: a server, a terminal, and a user. The server collects information about companies' support programs and performs learning using a generative AI model. The terminal accepts user questions, performs sentiment analysis, and transmits the question and sentiment data to the server. Users access the system through an online portal or a dedicated application.

[1365] server

[1366] The server processes the information as follows:

[1367] Information gathering: Collect documents from company managers regarding support programs, including employee benefits, special leave, and medical assistance.

[1368] Preprocessing and formatting: Format and preprocess the collected documents and feed them into a generative AI model (e.g., OpenAI's GPT-3).

[1369] Learning: Generative AI models are used to learn from collected information.

[1370] Question analysis: Question data sent from the device is analyzed using natural language processing technology to extract important keywords.

[1371] Information retrieval: Search for information from related documents based on the analysis results.

[1372] Answer generation: Using a generative AI model, appropriate answers are generated based on retrieved information and sentiment data.

[1373] Prepare for display: Send the generated answer to the device.

[1374] Terminal

[1375] The terminal works as follows:

[1376] Question reception: Receive questions from users and convert them into data.

[1377] Sentiment analysis: The emotion analysis engine analyzes the entered question as well as emotional data such as the user's facial expressions and voice.

[1378] Data transmission: Question data and emotion data are sent to the server.

[1379] Display Answer: The answer sent from the server is displayed to the user.

[1380] User

[1381] The user uses the system as follows:

[1382] Access: Access an online portal or dedicated application to enter your questions.

[1383] Enter a question: Enter a specific question and wait for the system to answer.

[1384] Specific examples

[1385] Example 1: Healthcare support inquiry

[1386] Question: User types, "I'm sick, can I leave work early?"

[1387] Sentiment analysis: Recognize when a user is feeling anxious.

[1388] Answer generation: The server responds, "You can leave work early by using special paid leave. Please contact our support desk for details."

[1389] Example 2: Querying rest facilities

[1390] Question: User types, "What rest stop is available when I finish work today?"

[1391] Sentiment analysis: Recognize that the user is calm.

[1392] Answer generation: The server responds, "After you finish work, break rooms A and B will be available."

[1393] Prompt Sentence Examples

[1394] Example 1 prompt statement:

[1395] User Question: Can I leave work early because I'm not feeling well?

[1396] User Emotion: Anxiety

[1397] Generate the appropriate answer.

[1398] Example 2 prompt statement:

[1399] User question: What rest areas can I use when I'm done with today's work?

[1400] User Sentiment: Calm

[1401] Generate the appropriate answer.

[1402] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1403] Step 1:

[1404] The server collects documents related to the company's support systems, such as employee benefit programs, special leave policies, and medical expense subsidies provided by company administrators, and stores them in a database.

[1405] Input: Documents about corporate support programs

[1406] Output: The formatted and preprocessed document data

[1407] Step 2:

[1408] The server formats and preprocesses the collected documents, specifically converting the text data format and removing unnecessary data to make it easier for the generative AI model to process.

[1409] Input: Data from collected documents

[1410] Output: Preprocessed data that can be input into a generative AI model

[1411] Step 3:

[1412] The server uses the preprocessed data to train a generative AI model. Specifically, it uses OpenAI's GPT-3 model to learn information and build a knowledge base for generating appropriate answers to user questions.

[1413] Input: Preprocessed data

[1414] Output: Knowledge base based on generative AI models

[1415] Step 4:

[1416] Users enter their questions through a dedicated application or online portal. Specifically, users enter text-based questions, which are then collected as digital data by the device.

[1417] Input: User question (e.g., "I'm not feeling well, can I leave early?")

[1418] Output: Digitized question data

[1419] Step 5:

[1420] The device performs emotion analysis to recognize the user's emotions, specifically by analyzing the user's facial expressions, voice, or other input data to identify their emotional state (e.g., stress, anxiety, calm).

[1421] Input: User facial expressions, voice, and contextual data

[1422] Output: Recognized emotion data (e.g., "anxiety")

[1423] Step 6:

[1424] The terminal transmits the question data and emotion data to the server. Specifically, the terminal transmits the digitized question data and the recognized emotion data to the server as a data packet.

[1425] Input: Question data, emotion data

[1426] Output: Send data to the server

[1427] Step 7:

[1428] The server analyzes the question using a generative AI model based on the question data and sentiment data received, and extracts important keywords. Specifically, it uses natural language processing technology to understand the intent of the question and identify related keywords.

[1429] Input: Question data, emotion data

[1430] Output: Extracted keywords (e.g., "Leave early", "health")

[1431] Step 8:

[1432] The server searches for relevant information from related documents based on the extracted keywords, using a database search algorithm to identify and organize relevant information.

[1433] Input: Extracted keywords

[1434] Output: Relevant information (e.g. special paid leave regulations for early departure)

[1435] Step 9:

[1436] The server generates answers using a generative AI model based on search results and emotional data. Specifically, for users in an anxious state, it creates answers in a gentle tone that include appropriate support information.

[1437] Input: Search results, emotion data

[1438] Output: Generated answer (e.g. "You can leave early using special paid leave. Please contact support for details.")

[1439] Step 10:

[1440] The server sends the generated answer to the terminal, and the terminal displays it to the user, specifically, displays the answer text on the user's device, quickly meeting the user's needs.

[1441] Input: Generated Answer

[1442] Output: Answer displayed on the user's terminal

[1443] 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.

[1444] 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.

[1445] 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.

[1446] [Fourth embodiment]

[1447] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1448] 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.

[1449] 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).

[1450] 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.

[1451] 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.

[1452] 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).

[1453] 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.

[1454] 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.

[1455] 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.

[1456] 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.

[1457] 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.

[1458] 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.

[1459] 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."

[1460] This system collects and learns information about corporate support systems and provides appropriate answers to user questions. This system is mainly composed of three elements: a server, a terminal, and a user.

[1461] Data collection and learning

[1462] First, the server collects documents from company administrators about the company's support systems. These documents include employee benefit programs, special leave policies, medical expense subsidies, etc. The collected documents are formatted and preprocessed before being input into a generative AI (e.g., ChatGPT). The server uses this generative AI to learn information and prepare it to generate answers to questions.

[1463] Accepting user questions

[1464] Next, users access the Smart Corporate System Navigator through their company's online portal or dedicated application, and enter a specific question in the search box, such as, "What kind of support is available if I get the flu?"

[1465] Question analysis and answer generation

[1466] The device sends the entered question to the server, which uses generative AI to analyze the question and understand its intent. Based on the results of the analysis, the server searches for relevant information in the company's documents. For example, if the question contains the keyword "influenza," the search will return "health management support documents" and the like.

[1467] Based on the retrieved information, the server generates a detailed answer, such as, "If employees catch the flu, they can use special paid leave. Medical expense assistance programs are also available."

[1468] Show Answers

[1469] Finally, the server sends the generated answer to the terminal, which then displays the answer to the user. The user can view the displayed answer, understand the support or assistance they need, and use it appropriately.

[1470] Specific examples

[1471] Example 1: Healthcare support inquiry

[1472] Users access an online portal and type in the question, "What help is available if I have the flu?"

[1473] The terminal sends this question to the server.

[1474] The server uses generative AI to analyze the question and search documents for information related to "influenza."

[1475] The server generates a response that "employees can take special paid leave and also have access to medical expense assistance."

[1476] The terminal displays the generated answer, and the user understands the necessary support information and uses it appropriately.

[1477] Example 2: Benefits Program Inquiry

[1478] The user types the question, "What support is available for childcare leave?"

[1479] The terminal sends this question to the server.

[1480] The server analyzes the query and retrieves the relevant assistance program documentation.

[1481] The server generates a response saying, "If you take child care leave, you will be granted special leave and some medical expenses will be subsidized."

[1482] The terminal displays the generated answer, and the user uses the corresponding support information.

[1483] In this way, the present invention provides a system that efficiently collects and learns information about corporate support systems and provides quick and accurate answers to user questions.

[1484] The processing flow will be explained below.

[1485] Step 1: Data entry and learning phase

[1486] The server receives documents related to support systems from company administrators. Specifically, it collects "support documents related to health management," "personnel system guidebooks," "details of employee benefit programs," etc.

[1487] The server formats the documents it receives and performs data cleaning if necessary, which includes standardizing the document format and removing unnecessary textual information.

[1488] The server feeds the formatted documents into a generative model (e.g., ChatGPT) and trains the model, which then has detailed knowledge of the company's systems.

[1489] Step 2: Accepting questions from users

[1490] Users access the Smart Corporate System Navigator through their company's online portal or application.

[1491] The user enters a specific question into the question entry form on the portal site, for example, "What kind of support is available if I get the flu?"

[1492] The terminal (user's device) sends the entered question data to the server.

[1493] Step 3: Parsing the Question

[1494] The server analyzes the received question data. Using a generative AI model, it understands the intent of the question and extracts keywords and important phrases. For example, "influenza" and "support" are extracted as important keywords.

[1495] Step 4: Find related information

[1496] The server uses the extracted keywords to search for relevant information from related documents, such as "health management support documents" and "personnel system guidebooks."

[1497] The server sorts through the search results and extracts the most relevant parts.

[1498] Step 5: Generate an answer

[1499] The server generates answers to provide to users based on the extracted information. The generation AI summarizes the information and adjusts the writing style to create easy-to-understand sentences.

[1500] For example, it generates a specific answer such as "If employees catch the flu, they can use special paid leave. Medical expense assistance systems are also available."

[1501] Step 6: Submit and view your responses

[1502] The server sends the generated answer to the user's terminal.

[1503] The terminal displays the received response to the user, allowing the user to quickly ascertain details of the assistance or assistance required.

[1504] In this way, by performing specific processing at each step, a system is provided that allows users to efficiently obtain and utilize information about corporate systems and support programs.

[1505] Example 1

[1506] 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."

[1507] Information about corporate support programs is diverse, making it difficult for employees to quickly obtain the information they need. Furthermore, if the information provided is inaccurate, employees may not receive appropriate support, hindering efficient work performance. In addition, there is the problem that technology for accurately analyzing questions entered by users in natural language and generating appropriate answers is still immature. Furthermore, if the generated answers are not displayed clearly to users, user convenience is reduced.

[1508] 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.

[1509] In this invention, the server includes: means for collecting information about corporate support programs; means for formatting and preprocessing the information; means for having a generative model that learns based on the information; means for accepting user questions; means for analyzing the questions; means for searching for relevant information; means for generating answers based on the searched information; means for displaying the generated answers to the user; means for removing unnecessary parts and extracting only important information using a natural language processing tool when formatting and preprocessing the information; means for generating detailed answers to the user questions based on prompt sentences using a generative AI model; and means for displaying the answers generated based on the prompt sentences in a user-friendly format. This makes it possible to quickly and accurately obtain information about corporate support programs, generate appropriate answers to questions entered in natural language, and display the answers in an easy-to-understand format to the user.

[1510] "Means for collecting information" refers to devices and methods for receiving and storing data relating to the company's support system from the company administrator.

[1511] "Information formatting and preprocessing means" refers to devices and methods that use natural language processing tools to convert collected data into an appropriate format and remove unnecessary parts.

[1512] "Means having a learning generative model" refers to a device or method that uses a generative AI model to learn information based on preprocessed data.

[1513] The "means for accepting a user's question" refers to a device or method that provides a function for a user to input a question in natural language and receives the input content.

[1514] The "means for analyzing a question" refers to a device or method that uses natural language processing technology to analyze a question received from a user and understand its intent.

[1515] The "means for retrieving relevant information" refers to a device or method for retrieving relevant information from relevant company documents or databases based on the analyzed query.

[1516] A "means for generating an answer" is a device or method that uses a generative AI model to create a detailed answer based on the searched information.

[1517] The "means for displaying the generated answers to the user" refers to a device or method for displaying the generated answers in a user-friendly format so that the user can view them.

[1518] "Means using natural language processing tools" refers to the techniques and devices used to analyze and format text when preprocessing collected data.

[1519] A "means for generating a detailed answer based on a prompt sentence" is a device or method that uses a generative AI model to input a prompt corresponding to a specific question and generate a detailed answer based on that prompt.

[1520] The "means for displaying in a user-friendly format" refers to a device or method that provides the generated answers to the user in a format that is easy to view and understand.

[1521] This system collects and learns information about corporate support systems and provides appropriate answers to user questions. This system is mainly composed of three elements: a server, a terminal, and a user.

[1522] Data collection and learning

[1523] The server receives documents related to the company's support system from the company administrator. The received documents are saved as text files or PDFs. The saved documents are preprocessed using an NLP tool (e.g., SpaCy, NLTK). Preprocessing refers to removing unnecessary parts and extracting important information. The formatted data from this process is input into a generative AI model (e.g., ChatGPT). The server uses this generative AI model to learn information and prepare to generate answers to user questions.

[1524] Accepting user questions

[1525] Users log in to their company's online portal or dedicated application to access the "Smart Corporate System Navigator." They enter a specific question into the search box, for example, "What kind of support is available if I have the flu?" The device then sends this input to the server. The transmitted data is protected using a secure communication protocol (e.g., HTTPS).

[1526] Question analysis and answer generation

[1527] The server converts the received question into a prompt to be input into the generative AI. Using the generative AI model, the server analyzes the intent of the user's question and extracts related keywords. Based on the results of this analysis, the server searches for relevant information from the company's documents. For example, if the keyword "influenza" is included, support documents related to health management will be searched for. Based on the search results, the generative AI model is used to generate a detailed answer. An example of a specific answer would be: "If employees catch the flu, they can use special paid leave. Medical expense subsidy systems are also available."

[1528] Show Answers

[1529] The server sends the generated answer to the terminal, which displays the answer in a user-friendly format (e.g., a card or pop-up).The user can view the displayed answer, understand the support or assistance they need, and use it appropriately.

[1530] Specific examples

[1531] Example 1: Healthcare support inquiry

[1532] The user accesses the online portal and enters the question, "What support is available if I get the flu?" The device then sends this question to the server. The server uses generative AI to analyze the question and searches documents for information related to "flu." The server then generates the answer, "Employees can use special paid leave and medical expense subsidy systems." The user then understands the necessary support information and uses it appropriately.

[1533] Example 2: Benefits Program Inquiry

[1534] The user inputs the question, "What kind of support is available for child care leave?" The device sends this question to the server. The server analyzes the question and searches for documents on related support programs. The server generates an answer that says, "If you take child care leave, you will be granted special leave and some medical expenses will be subsidized." The device displays the generated answer, allowing the user to use the relevant support information appropriately.

[1535] This system can efficiently collect and learn information about corporate support systems and provide quick and accurate answers to users' questions.

[1536] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1537] Step 1:

[1538] The server receives documents relating to the company's support system from the company administrator.

[1539] Specifically, the document is collected using the file upload function.

[1540] Input: Support program documentation provided by company administrator.

[1541] Output: Saved support system documents.

[1542] Step 2:

[1543] The server saves the received documents in text file or PDF format.

[1544] Specifically, the uploaded file is saved in an appropriate folder or database.

[1545] Input: Uploaded assistance system document.

[1546] Output: Saved text file and PDF.

[1547] Step 3:

[1548] The server preprocesses the stored documents using natural language processing (NLP) tools (e.g., SpaCy, NLTK).

[1549] Specifically, it removes unnecessary parts from the document (e.g., page numbers, headers, footers) and extracts important information.

[1550] Input: Saved text files or PDFs.

[1551] Output: Preprocessed text data.

[1552] Step 4:

[1553] The server formats the preprocessed data and generates prompts to input into a generative AI model (e.g., ChatGPT).

[1554] Specifically, the process creates a prompt statement template based on the preprocessed data.

[1555] Input: Preprocessed text data.

[1556] Output: The prompt sentence to be input to the generative AI model.

[1557] Step 5:

[1558] The server uses a generative AI model to learn information and prepare to generate answers to user questions.

[1559] Specifically, the prompt sentence is input into the generative AI model and the model is trained.

[1560] Input: prompt statement.

[1561] Output: A trained generative AI model.

[1562] Step 6:

[1563] Users log in to the online portal or dedicated application and access the "Smart Corporate System Navigator."

[1564] Specifically, the user authentication and login process are carried out.

[1565] Input: User credentials.

[1566] Output: Authenticated user interface.

[1567] Step 7:

[1568] Users enter a specific question into a search box.

[1569] Specifically, you enter a question in the search box and press the send button.

[1570] Input: The question text provided by the user.

[1571] Output: The question data sent to the terminal.

[1572] Step 8:

[1573] The terminal transmits the entered question to the server.

[1574] Specifically, the input data is transferred to the server using a secure communication protocol (e.g., HTTPS).

[1575] Input: Question data from the user.

[1576] Output: The query data sent to the server.

[1577] Step 9:

[1578] The server converts the received question into a prompt sentence to be input to the generation AI.

[1579] Specifically, the question data is formatted into an appropriate prompt format.

[1580] Input: Question data from the user.

[1581] Output: The prompt statement.

[1582] Step 10:

[1583] The server uses generative AI to analyze the intent of the question and extract relevant keywords.

[1584] Specifically, the prompt sentence is input into the generative AI model, and intent analysis and keyword extraction are performed.

[1585] Input: prompt statement.

[1586] Output: Parsed intent and extracted keywords.

[1587] Step 11:

[1588] The server searches for company documents based on the extracted keywords.

[1589] Specifically, the operation involves searching for the relevant information from a document management system or database.

[1590] Input: Parsed intent and extracted keywords.

[1591] Output: Search results (relevant documents).

[1592] Step 12:

[1593] The server generates detailed answers based on the search results using a generative AI model.

[1594] Specifically, the search results are input into a generative AI model to generate a detailed answer.

[1595] Input: Search results.

[1596] Output: The detailed answer generated.

[1597] Step 13:

[1598] The server sends the generated response to the terminal.

[1599] Specifically, the generated answer is formatted into a user-friendly format and sent to the terminal.

[1600] Input: The generated detailed answer.

[1601] Output: The response data sent to the device.

[1602] Step 14:

[1603] The device displays the received answers in a user-friendly format (e.g., card format or popup).

[1604] Specifically, the response data is read and displayed on the user interface.

[1605] Input: Response data sent to the terminal.

[1606] Output: The answer displayed to the user.

[1607] Step 15:

[1608] The user can view the displayed answers, understand the support and assistance they need, and use it appropriately.

[1609] Specifically, the displayed information is checked and the next action is taken as necessary.

[1610] Input: The answer shown to the user.

[1611] Output: User understanding and use of information.

[1612] (Application example 1)

[1613] 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."

[1614] There is a wide variety of information available about corporate support programs, making it difficult for employees to quickly and accurately identify the program that best suits them. Furthermore, especially in busy environments, such as store employees, who often lack the time to use digital devices, employees tend to neglect using support programs. This leads to a decline in employee utilization of support programs, resulting in the inability to fully realize the benefits of corporate employee benefits.

[1615] 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.

[1616] In this invention, the server includes: means for collecting information on corporate support programs; means for formatting and preprocessing the information; means for having a generative model that learns based on the information; means for accepting user questions; means for analyzing the questions; means for searching for relevant information; means for generating answers based on the searched information; means for displaying the generated answers to the user; a voice input means installed in the smart device; means for converting voice to text; means for transmitting the text questions to the server; and means for displaying the generated answers on the smart device. This allows employees to easily ask questions through voice input and quickly access information on appropriate support programs.

[1617] "Corporate support systems" is a general term for welfare programs, special leave provisions, medical expense subsidies, etc. that companies provide to their employees.

[1618] "Means for collecting information" refers to a method or system for collecting documents related to the support system from company administrators.

[1619] "Information formatting and preprocessing means" refers to methods and systems for converting collected documents into a format that is easy to analyze and for preprocessing the data.

[1620] A "generative model" is a machine learning model that learns from collected and preprocessed information and generates appropriate answers to user questions.

[1621] The "means for accepting a user's question" is an interface for receiving a question input by a user and transmitting it to the system.

[1622] The "means for analyzing questions" refers to natural language processing techniques and algorithms for analyzing received questions and understanding their intent.

[1623] The "means for searching relevant information" is a method or system for searching company documents for information relevant to the analyzed question.

[1624] "Answer generation means" refers to a method or system for generating a specific answer based on the retrieved information.

[1625] The "means for displaying the generated answer to the user" refers to a method or system for displaying the generated answer on the user's device (terminal).

[1626] "Smart devices" is a general term for devices that can connect to the Internet and run various applications, such as smartphones, smart glasses, and head-mounted displays.

[1627] The "voice input means" is a function for inputting the user's voice using a microphone or the like installed on the smart device.

[1628] "Means for converting speech to text" refers to speech recognition technology and algorithms for analyzing input speech and converting it into text data.

[1629] The "means for transmitting a question converted into text to a server" is a communication function for transmitting a user's question converted into text by voice recognition to a server.

[1630] The "means for displaying the answer on the smart device" is a function for receiving the answer generated from the server and displaying it on the display screen of the smart device.

[1631] This system collects and learns information about corporate support programs and provides appropriate answers to user questions. This system is composed of three elements: a server, a terminal, and a user.

[1632] First, the server collects documents from company administrators about the company's support systems, including employee benefit programs, special leave policies, medical expense subsidies, etc. The collected documents undergo formatting and preprocessing, and then the generative AI learns the information and prepares them to generate answers to questions.

[1633] Next, the user accesses the company's smart device and asks a question through voice input, for example, "What kind of support is available if I have the flu?" The microphone installed on the smart device captures this voice and converts it into text using voice recognition technology.

[1634] The device sends the converted text data to the server, which uses generative AI to analyze the question and understand its intent. Based on the results of the analysis, the server searches for relevant information in the company's documents. For example, if a question contains the keyword "influenza," the search will return "health management support documents" and other relevant information.

[1635] Based on the searched information, the server generates a detailed answer. For example, it creates a specific answer such as, "If an employee catches the flu, they can use special paid leave. Medical expense subsidies are also available." The generated answer is sent to the smart device as text data and displayed on the screen. The user can view the displayed answer, understand the support and assistance they need, and use it appropriately.

[1636] Examples:

[1637] A user uses the smart glasses to ask a question out loud: "How much is the employee discount?" The question is converted into text using speech recognition, and the question "How much is the employee discount?" is sent to the server. The server uses a generative AI model to analyze the question and generates the answer: "The employee discount is 10% on all products." This answer is displayed on the smart glasses for the user to see immediately.

[1638] Example prompt sentence:

[1639] Q: What support is available for employees with children?

[1640] Answer: Employees with children are eligible for special parental leave, medical assistance, and assistance with childcare equipment.

[1641] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1642] Step 1:

[1643] The server collects documents related to the company's support system from the company administrator. The collected documents include information such as "employee benefit programs," "special leave regulations," and "medical expense subsidies." These documents are stored on the server and used for subsequent processing.

[1644] Input: Company support program document provided by company administrator

[1645] Output: A database of collected support system documents

[1646] Step 2:

[1647] The server formats and preprocesses the collected documents, removing unnecessary information and using natural language processing (NLP) techniques to structure the text, converting it into a format suitable for input into the generative AI model.

[1648] Input: Collected assistance system documents

[1649] Output: A formatted and preprocessed document

[1650] Step 3:

[1651] The server uses the preprocessed documents to train the generative AI model, specifically learning various information about the business support program and training the model to generate appropriate answers to questions.

[1652] Input: formatted and preprocessed document

[1653] Output: Trained generative AI model

[1654] Step 4:

[1655] A user accesses a smart device and uses voice input to ask a question, for example, "What can I do to help if I get the flu?" The smart device uses a microphone to capture the user's voice.

[1656] Input: User's voice question

[1657] Output: Captured audio data

[1658] Step 5:

[1659] The device uses speech recognition technology to convert the captured speech into text, specifically by using a speech recognition library to analyze the speech data and convert it into the appropriate text.

[1660] Input: Captured audio data

[1661] Output: Texted question

[1662] Step 6:

[1663] The device sends a textual question to the server, which receives the question and prepares it for processing by the generative AI model.

[1664] Input: Texted question

[1665] Output: The question sent to the server

[1666] Step 7:

[1667] The server uses a generative AI model to analyze the question and understand its intent. Specifically, it uses NLP technology to identify keywords and phrases in the question and analyze their meaning.

[1668] Input: Texted question

[1669] Output: Parsed intent (keywords and phrases in the question)

[1670] Step 8:

[1671] The server searches for relevant information from the company's documents based on the analyzed intent. For example, if the keyword "influenza" is included, it searches for support documents related to health care.

[1672] Input: Parsed intent

[1673] Output: Searched support system information

[1674] Step 9:

[1675] The server uses the searched information to generate a specific answer to the question, such as "If employees catch the flu, they can use their paid leave. Medical expense assistance is also available."

[1676] Input: Searched support system information

[1677] Output: The specific answer generated

[1678] Step 10:

[1679] The server sends the generated answer as text data to the terminal, which then displays the received answer on a smart device, such as a smart eyeglasses display.

[1680] Input: Generated specific answer

[1681] Output: Answers displayed on a smart device

[1682] Through the above steps, information about corporate support programs can be provided to users quickly and accurately.

[1683] 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.

[1684] This system combines a system that collects and learns information about corporate support systems, provides appropriate answers to user questions, and an emotion engine that recognizes user emotions. This system is primarily composed of three elements: a server, a terminal, and a user.

[1685] Data collection and learning

[1686] First, the server collects documents about support systems from company administrators. These documents include employee benefit programs, special leave policies, medical expense subsidies, etc. The collected documents are formatted and preprocessed before being input into a generative AI (e.g., ChatGPT). The server uses this generative AI to learn information and prepare to generate answers to user questions.

[1687] Accepting user questions

[1688] Next, the user accesses the "Smart Corporate System Navigator" through the company's online portal or a dedicated application. The user enters a specific question into the question input form on the portal site. For example, the user enters a question such as "What kind of support is available if I get the flu?" The terminal (user's device) then sends the entered question data to the server.

[1689] Emotion recognition by emotion engine

[1690] The device sends the user's facial expressions, voice, and context in real time to the emotion engine, which analyzes this data and recognizes the user's emotional state (e.g., stress, anxiety, relief, etc.). The recognized emotion data is then sent to the server.

[1691] Question analysis and answer generation

[1692] The server analyzes the question using a generative AI model based on the received question data and sentiment data. It understands the intent of the question and extracts keywords and important phrases. For example, "influenza" and "assistance" are extracted as important keywords. The server then searches for information from related documents based on the extracted keywords, organizes the search results, and extracts the most relevant parts.

[1693] The server uses the emotional data to tailor the format and content of the response to suit the user. For example, if the user is feeling anxious, the tone of the response will be gentler and phrased in a way that provides a sense of security. A specific response generated might be, "If you catch the flu, employees can use special paid leave and medical expense assistance. If you have any questions, please contact our support desk."

[1694] Submitting and viewing answers

[1695] The server sends the generated answer to the user's terminal, which displays the received answer to the user, allowing the user to quickly confirm details of the help or assistance required.

[1696] Specific examples

[1697] Example 1: Healthcare support inquiry

[1698] Users access an online portal and type in the question, "What help is available if I have the flu?"

[1699] The device sends emotional data indicating anxiety based on the user's voice and facial expression during input to the emotion engine, which then analyzes the data and recognizes anxiety.

[1700] The terminal transmits the question data and emotion data to the server.

[1701] The server uses generative AI to analyze the question and search documents for information related to "influenza."

[1702] The server takes into account the user's anxious state and generates a gentle response saying, "Employees can take special paid leave and medical expense assistance programs are also available. If you have any questions, please contact our support desk."

[1703] The terminal displays the generated answer, and the user understands the necessary support information and uses the device with peace of mind.

[1704] Example 2: Benefits Program Inquiry

[1705] The user types the question, "What support is available for childcare leave?"

[1706] The terminal sends the user's voice data to the emotion engine, which recognizes that the user is calm.

[1707] The terminal transmits the question data and emotion data to the server.

[1708] The server analyzes the query and retrieves information from the documentation of the relevant assistance program.

[1709] The server generates a clear answer for the calm user: "If you take childcare leave, you will be granted special leave and some medical expenses will be subsidized."

[1710] The terminal displays the generated answer, and the user uses the corresponding support information.

[1711] In this way, the present invention realizes a system that provides more appropriate and personalized answers by combining an emotion engine that recognizes the user's emotions.

[1712] The processing flow will be explained below.

[1713] Step 1: Data entry and learning phase

[1714] The server collects documents about support systems from company administrators, including details of employee benefit programs, personnel system guidebooks, and medical assistance.

[1715] The server formats the collected documents and cleans them of unnecessary information, standardizing the document format and removing noise.

[1716] The server inputs the formatted documents into a generative model (e.g., ChatGPT) and trains the model, which accumulates knowledge about the support system in the generative AI.

[1717] Step 2: Accepting questions from users

[1718] Users access the Smart Corporate System Navigator through their company's online portal or application.

[1719] The user enters a specific question (e.g., "What kind of support is available if I have the flu?") into an input form.

[1720] The terminal transmits the input question data to the server in real time.

[1721] Step 3: Emotion recognition by the emotion engine

[1722] The device sends facial, vocal, and contextual information to the emotion engine as the user inputs, including camera and microphone data.

[1723] The server uses an emotion engine to analyze this data and recognize the user's emotional state (e.g., anxiety, relief, anger, etc.).

[1724] The recognized emotion data is sent to a server for further analysis.

[1725] Step 4: Parsing the Question

[1726] The server uses generative AI to analyze the question entered by the user and extract important keywords and phrases from the question (e.g., "influenza" and "assistance").

[1727] The server searches for the necessary information from related documents based on the extracted keywords.

[1728] Step 5: Generate an answer

[1729] Based on the searched information, the server uses a generative AI model to generate answers for the user.

[1730] The server uses the recognized emotion data to customize the tone and content of the response. For example, if the user is feeling anxious, the response will be changed to a more reassuring tone.

[1731] A specific answer generated is "If an employee catches the flu, they can use special paid leave. They can also use the medical expense subsidy system."

[1732] Step 6: Submit and view your responses

[1733] The server sends the generated answer to the user's terminal.

[1734] The terminal displays the received response to the user, allowing the user to quickly check the details of the system or support they need.

[1735] Examples:

[1736] Example 1: Healthcare support inquiry

[1737] Users access an online portal and type in the question, "What help is available if I have the flu?"

[1738] The device sends the user's question data and emotional data, such as facial expressions and voice, to the emotion engine, which then recognizes anxiety.

[1739] The terminal transmits the question data and emotion data to the server.

[1740] The server parses the query and searches the documents for information related to "influenza."

[1741] The server generates a gentle response to the anxious user, saying, "Employees can take special paid leave and medical expense assistance programs are also available. Please feel free to receive support."

[1742] The terminal displays the generated answer, allowing the user to use appropriate support information with peace of mind.

[1743] Example 2: Benefits Program Inquiry

[1744] The user types the question, "What support is available for childcare leave?"

[1745] The terminal sends the user's voice and facial expression data to the emotion engine, which recognizes that the user is calm.

[1746] The terminal transmits the question data and emotion data to the server.

[1747] The server analyzes the query and retrieves relevant support program information from the document.

[1748] The server generates a clear answer for the calm user: "If you take childcare leave, you will be granted special leave and some medical expenses will be subsidized."

[1749] The terminal displays the generated answer and the user can use the corresponding support information.

[1750] In this way, the present invention realizes a system that provides more appropriate and personalized answers according to the user's emotional state by combining an emotion engine.

[1751] Example 2

[1752] 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."

[1753] Conventional information provision systems for business support programs have the problem that it is difficult for users to quickly and accurately obtain the detailed information they require. As a result, users often experience inconvenience because they are unable to quickly access the support and assistance they need. Furthermore, responses that do not take into account the user's emotional state may cause the user to feel anxious or stressed.

[1754] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting information on corporate support programs, means for formatting and preprocessing the information, means for having a generative AI model that learns based on the information, means for accepting a user's question, means for collecting facial expression, voice, and contextual data when asking a question, means for analyzing the emotional data to recognize the emotional state, means for analyzing the question and the recognized emotional state and generating an answer using the generative AI model, and means for displaying the generated answer to the user. This allows the user to quickly and accurately obtain information about the support programs they need and to receive a personalized answer that takes their emotional state into consideration.

[1755] "Corporate support systems" refers to various systems and programs that companies use to provide support to their employees, such as employee benefits, special leave, and medical expense subsidies.

[1756] "Means of collecting information" refers to the means and methods for obtaining documents and data related to the support system from company managers.

[1757] "Information formatting and preprocessing methods" refers to the means or methods for converting collected documents or data into a format that is easier to analyze and for removing unnecessary characters and formatting.

[1758] A "generative AI model" is an artificial intelligence model trained on large amounts of data, for example, to generate appropriate answers to user questions using natural language processing.

[1759] "Means for accepting user questions" refers to a means or interface for a user to input a specific question into the system and transmit that information to the server.

[1760] "Means for collecting facial, audio, and contextual data" refers to means for collecting data to understand the user's emotional state using a user's device, such as a camera or microphone.

[1761] "Means for analyzing emotional data and recognizing emotional states" refers to means for analyzing and recognizing a user's emotional state based on collected data using a deep learning model or the like.

[1762] "Means for generating an answer" refers to a means for using a generative AI model to analyze the user's question and perceived emotional state and generate an appropriate answer.

[1763] The "means for displaying the answer to the user" refers to a means for displaying the generated answer on the user's device so that the user can easily view the required information.

[1764] "Online portal" refers to a website or dedicated application that provides information about support programs for businesses.

[1765] This invention is a system that collects and learns information about corporate support systems and provides appropriate answers to user questions. This system is mainly composed of three elements: a server, a terminal, and a user.

[1766] Data collection and learning

[1767] The server collects documents related to employee support programs from company administrators. This collection is done using an FTP server or API. The documents include employee benefit programs, special leave policies, medical expense subsidies, etc. The collected documents are saved in a specific directory.

[1768] The server extracts the stored document, removes unnecessary characters and formatting, and formats it into a format that is easy to parse. This formatted data is then fed into a generative AI model (e.g., ChatGPT) to train the model. Once trained, the generative AI model is stored on the server and ready to respond to user queries.

[1769] Accepting user questions

[1770] Users access the company's online portal or dedicated application and enter specific questions (e.g., "What kind of support is available if I get the flu?") into the question input form of the "Smart Corporate System Navigator."

[1771] The terminal (user's device) sends the entered questions to the server in real time, and the communication is secure because it is done via the HTTPS protocol.

[1772] Emotion recognition by emotion engine

[1773] The device uses a camera and microphone to collect facial expressions, voice, and contextual data during user input. It then transmits the collected emotion data to the emotion engine in real time. The emotion engine uses a deep learning model to analyze the collected data and recognize the user's emotional state (e.g., stress, anxiety, relief, etc.). The recognized emotion data is then sent to the server.

[1774] Question analysis and answer generation

[1775] The server uses a generative AI model to analyze the received question data, understand the intent of the question, and extract keywords and key phrases (e.g., "influenza" and "assistance"). The server then searches documents related to relevant assistance programs and systems based on the extracted keywords and organizes the most relevant information.

[1776] The server takes emotion data into account and uses a generative AI model to generate the optimal answer. For example, if a user is anxious, the server might generate a gentle response such as, "Employees can take special paid leave and medical expense assistance programs are also available. If you have any questions, please contact our support desk."

[1777] Submitting and viewing answers

[1778] The server then sends the generated answer to the user's device, again securely via the HTTPS protocol, where it displays the answer to the user, allowing them to quickly view details of the assistance or support they require.

[1779] Specific examples

[1780] Example 1: Healthcare support inquiry

[1781] Users access an online portal and type in the question, "What help is available if I have the flu?"

[1782] The terminal transmits the question data to the server in real time.

[1783] The device transmits the user's facial expression and voice data to the emotion engine, which then recognizes that the user is feeling anxious.

[1784] The terminal transmits the emotion data to the server.

[1785] The server analyzes the question data and emotion data and searches documents for information related to "influenza."

[1786] The server considers the anxious user and generates a gentle response saying, "Employees can take special paid leave and medical expense assistance programs are also available. If you have any questions, please contact our support desk."

[1787] The server sends the generated answer to the user's device, which displays it. The user understands the necessary support information and can use the service with peace of mind.

[1788] Example 2: Benefits Program Inquiry

[1789] The user types the question, "What support is available for childcare leave?"

[1790] The terminal transmits the question data to the server in real time.

[1791] The terminal transmits the user's voice data to the emotion engine, which recognizes that the user is calm.

[1792] The terminal transmits the emotion data to the server.

[1793] The server analyzes the question data and emotion data and searches documents for information related to "child care leave."

[1794] For a calm user, the server generates a clear answer: "If you take childcare leave, you will be granted special leave and some medical expenses will be subsidized."

[1795] The server sends the generated answer to the user's device, which displays it, allowing the user to understand and use the corresponding support information appropriately.

[1796] Prompt Sentence Examples

[1797] "Please tell me about the company's support system if I get the flu. Please speak in a reassuring tone, as users are feeling anxious."

[1798] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1799] Step 1:

[1800] The server collects documents related to the support system from company administrators. This collection is often done via an FTP server or API. The input is the company's document data, and the output is a document file saved in a specific directory.

[1801] Step 2:

[1802] The server extracts the stored document files, removes unnecessary characters and formatting, and formats them into a format that is easy to analyze. Specifically, it uses text analysis tools to remove noise from the data and generate formatted data. The input is the document file read from storage, and the output is formatted data that can be used as training data for the generative AI model.

[1803] Step 3:

[1804] The server inputs the shaped data into a generative AI model for training. A generative AI model (such as ChatGPT) is used for this purpose. The trained generative AI model is stored on the server. The input is the shaped data, and the output is the trained generative AI model.

[1805] Step 4:

[1806] A user accesses a company's online portal or dedicated application and enters a specific question into a question input form. For example, "What kind of support is available if I get the flu?" The input is the user's question text, and the output is data sent from the terminal to the server.

[1807] Step 5:

[1808] The terminal transmits the entered question to the server in real time, securely using the HTTPS protocol. The input is the user's question text, and the output is the data sent to the server.

[1809] Step 6:

[1810] The device uses a camera and microphone to collect facial, voice, and contextual data during user input. This data is sent to the emotion engine in real time for preprocessing. The input is the user's emotion data, and the output is preprocessed emotion data.

[1811] Step 7:

[1812] The emotion engine analyzes the collected emotion data using a deep learning model to recognize the user's emotional state (e.g., stress, anxiety, relief, etc.). The input is the preprocessed emotion data, and the output is the recognized emotional state.

[1813] Step 8:

[1814] The server performs analysis based on the received question data and recognized emotion data. It uses a generative AI model to understand the intent of the question and extract keywords and important phrases (e.g., "influenza" or "assistance"). The input is the question data and emotion data, and the output is the extracted keywords and semantic data.

[1815] Step 9:

[1816] The server searches documents related to related support programs and systems based on the extracted keywords and organizes the most relevant information. The input is the extracted keywords, and the output is organized information data.

[1817] Step 10:

[1818] The server takes into account the emotional data and uses a generative AI model to generate the optimal answer. For example, to an anxious user, it generates a gentle answer such as, "Employees can take special paid leave, and medical expense subsidies are also available. If you have any questions, please contact our support desk." The input is organized information data and emotional data, and the output is the generated answer text.

[1819] Step 11:

[1820] The server sends the generated answer to the user's terminal, also securely using the HTTPS protocol. The input is the generated answer text, and the output is the data sent to the terminal.

[1821] Step 12:

[1822] The terminal displays the received answer to the user, allowing the user to quickly check the details of the assistance or support required. The input is the received answer text, and the output is the display data in a format that the user can view.

[1823] (Application example 2)

[1824] 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."

[1825] In information provision systems for corporate support systems, it is difficult for employees to quickly obtain appropriate information tailored to their own situation. Furthermore, there are currently few systems that provide personalized answers that reflect the emotional state of employees when they ask questions. This issue is particularly important in factories, where providing immediate support information directly leads to an improvement in the working environment.

[1826] 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.

[1827] In this invention, the server includes means for collecting information on corporate support programs, means for formatting and preprocessing the information, means for having a generative model that learns based on the information, means for accepting user questions, means for analyzing the questions, means for searching for relevant information, means for generating answers based on the searched information, means for displaying the generated answers to the user, emotion analysis means for recognizing the user's emotions, and means for adjusting the format and content of the answers based on the emotion data recognized by the emotion analysis means. This makes it possible to instantly provide appropriate support information according to the emotions of employees when they ask questions.

[1828] "Corporate support systems" is a general term for various systems such as employee benefits, special leave, and medical expense subsidies that companies provide to their employees.

[1829] "Means of collecting information" refers to the methods and tools used to collect documents and data related to support systems within the company.

[1830] "Information formatting and preprocessing means" refers to methods and tools used to convert collected information into a format that is easy to analyze.

[1831] A "generative model" refers to a computer program or algorithm that learns from collected information and generates appropriate answers to user questions.

[1832] "Means for accepting questions" refers to an interface or application that allows a user to input questions to the server.

[1833] "Means for analyzing questions" refers to methods and tools that use natural language processing technology to analyze questions from users and extract intent and keywords.

[1834] "Means for retrieving relevant information" refers to methods and tools for retrieving relevant information from a database based on the analyzed question.

[1835] "Answer generation means" refers to a computer program or algorithm that generates an appropriate answer to a user's question based on the retrieved information.

[1836] "Means for displaying to the user" refers to a method or tool for displaying the generated answer on a device so that the user can review it.

[1837] "Emotion analysis means" refers to technologies and algorithms for recognizing and analyzing a user's emotional state.

[1838] "Means for adjusting the format and content of responses based on emotional data" refers to methods or tools for adjusting the tone and content of responses according to the user's emotional state based on emotional data obtained by the emotion analysis means.

[1839] To implement this invention, the system consists of three main components: a server, a terminal, and a user. The server collects information about companies' support programs and performs learning using a generative AI model. The terminal accepts user questions, performs sentiment analysis, and transmits the question and sentiment data to the server. Users access the system through an online portal or a dedicated application.

[1840] server

[1841] The server processes the information as follows:

[1842] Information gathering: Collect documents from company managers regarding support programs, including employee benefits, special leave, and medical assistance.

[1843] Preprocessing and formatting: Format and preprocess the collected documents and feed them into a generative AI model (e.g., OpenAI's GPT-3).

[1844] Learning: Generative AI models are used to learn from collected information.

[1845] Question analysis: Question data sent from the device is analyzed using natural language processing technology to extract important keywords.

[1846] Information retrieval: Search for information from related documents based on the analysis results.

[1847] Answer generation: Using a generative AI model, appropriate answers are generated based on retrieved information and sentiment data.

[1848] Prepare for display: Send the generated answer to the device.

[1849] Terminal

[1850] The terminal works as follows:

[1851] Question reception: Receive questions from users and convert them into data.

[1852] Sentiment analysis: The emotion analysis engine analyzes the entered question as well as emotional data such as the user's facial expressions and voice.

[1853] Data transmission: Question data and emotion data are sent to the server.

[1854] Display Answer: The answer sent from the server is displayed to the user.

[1855] User

[1856] The user uses the system as follows:

[1857] Access: Access an online portal or dedicated application to enter your questions.

[1858] Enter a question: Enter a specific question and wait for the system to answer.

[1859] Specific examples

[1860] Example 1: Healthcare support inquiry

[1861] Question: User types, "I'm sick, can I leave work early?"

[1862] Sentiment analysis: Recognize when a user is feeling anxious.

[1863] Answer generation: The server responds, "You can leave work early by using special paid leave. Please contact our support desk for details."

[1864] Example 2: Querying rest facilities

[1865] Question: User types, "What rest stop is available when I finish work today?"

[1866] Sentiment analysis: Recognize that the user is calm.

[1867] Answer generation: The server responds, "After you finish work, break rooms A and B will be available."

[1868] Prompt Sentence Examples

[1869] Example 1 prompt statement:

[1870] User Question: Can I leave work early because I'm not feeling well?

[1871] User Emotion: Anxiety

[1872] Generate the appropriate answer.

[1873] Example 2 prompt statement:

[1874] User question: What rest areas can I use when I'm done with today's work?

[1875] User Sentiment: Calm

[1876] Generate the appropriate answer.

[1877] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1878] Step 1:

[1879] The server collects documents related to the company's support systems, such as employee benefit programs, special leave regulations, and medical expense subsidies provided by company administrators, and stores them in a database.

[1880] Input: Documents about corporate support programs

[1881] Output: The formatted and preprocessed document data

[1882] Step 2:

[1883] The server formats and preprocesses the collected documents, specifically converting the text data format and removing unnecessary data to make it easier for the generative AI model to process.

[1884] Input: Data from collected documents

[1885] Output: Preprocessed data that can be input into a generative AI model

[1886] Step 3:

[1887] The server uses the preprocessed data to train a generative AI model. Specifically, it uses OpenAI's GPT-3 model to learn information and build a knowledge base for generating appropriate answers to user questions.

[1888] Input: Preprocessed data

[1889] Output: Knowledge base based on generative AI models

[1890] Step 4:

[1891] Users enter their questions through a dedicated application or online portal. Specifically, users enter text-based questions, which are then collected as digital data by the device.

[1892] Input: User question (e.g., "I'm not feeling well, can I leave early?")

[1893] Output: Digitized question data

[1894] Step 5:

[1895] The device performs emotion analysis to recognize the user's emotions, specifically by analyzing the user's facial expressions, voice, or other input data to identify their emotional state (e.g., stress, anxiety, calm).

[1896] Input: User facial expressions, voice, and contextual data

[1897] Output: Recognized emotion data (e.g., "anxiety")

[1898] Step 6:

[1899] The terminal transmits the question data and emotion data to the server. Specifically, the terminal transmits the digitized question data and the recognized emotion data to the server as a data packet.

[1900] Input: Question data, emotion data

[1901] Output: Send data to the server

[1902] Step 7:

[1903] The server analyzes the question using a generative AI model based on the question data and sentiment data received, and extracts important keywords. Specifically, it uses natural language processing technology to understand the intent of the question and identify related keywords.

[1904] Input: Question data, emotion data

[1905] Output: Extracted keywords (e.g., "Leave early", "health")

[1906] Step 8:

[1907] The server searches for relevant information from related documents based on the extracted keywords, using a database search algorithm to identify and organize relevant information.

[1908] Input: Extracted keywords

[1909] Output: Relevant information (e.g. special paid leave regulations for early departure)

[1910] Step 9:

[1911] The server generates answers using a generative AI model based on search results and emotional data. Specifically, for users in an anxious state, it creates answers in a gentle tone that include appropriate support information.

[1912] Input: Search results, emotion data

[1913] Output: Generated answer (e.g. "You can leave early using special paid leave. Please contact support for details.")

[1914] Step 10:

[1915] The server sends the generated answer to the terminal, and the terminal displays it to the user, specifically, displays the answer text on the user's device, quickly meeting the user's needs.

[1916] Input: Generated Answer

[1917] Output: Answer displayed on the user's terminal

[1918] 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.

[1919] 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.

[1920] 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.

[1921] 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.

[1922] 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.

[1923] 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.

[1924] 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).

[1925] 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.

[1926] 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."

[1927] 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.

[1928] 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).

[1929] 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.

[1930] 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.

[1931] 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.

[1932] 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.

[1933] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1934] 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.

[1935] 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.

[1936] 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.

[1937] 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.

[1938] 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.

[1939] The following is further disclosed regarding the above embodiment.

[1940] (Claim 1)

[1941] a means of collecting information on corporate support systems;

[1942] means for formatting and preprocessing said information;

[1943] means for having a generative model that learns based on the information;

[1944] means for accepting user questions;

[1945] means for analyzing said query;

[1946] a means of searching for such information;

[1947] a means for generating an answer based on the retrieved information;

[1948] means for displaying the generated answer to a user;

[1949] A system including:

[1950] (Claim 2)

[1951] 2. The system of claim 1, wherein the means for analyzing the question performs the analysis using natural language processing.

[1952] (Claim 3)

[1953] 2. The system of claim 1, wherein information about corporate support programs is provided through an online portal.

[1954] "Example 1"

[1955] (Claim 1)

[1956] a means of collecting information on corporate support systems;

[1957] means for formatting and preprocessing said information;

[1958] means for having a generative model that learns based on the information;

[1959] means for accepting user questions;

[1960] means for analyzing said query;

[1961] a means of searching for such information;

[1962] a means for generating an answer based on the retrieved information;

[1963] means for displaying the generated answer to a user;

[1964] A system including:

[1965] (Claim 2)

[1966] 2. The system of claim 1, wherein the means for analyzing the question performs the analysis using natural language processing.

[1967] (Claim 3)

[1968] 2. The system of claim 1, wherein information about corporate support programs is provided through an online portal.

[1969] (Claim 4)

[1970] 10. The system of claim 1, further comprising means for formatting and preprocessing the information using natural language processing tools to remove unnecessary parts and extract only important information.

[1971] (Claim 5)

[1972] 10. The system of claim 1, further comprising: means for generating a detailed answer to the user's question based on a prompt sentence using a generative AI model.

[1973] (Claim 6)

[1974] 10. The system of claim 1, further comprising means for displaying the answers generated based on the prompt sentence in a user-friendly format.

[1975] "Application Example 1"

[1976] (Claim 1)

[1977] a means of collecting information on corporate support systems;

[1978] means for formatting and preprocessing said information;

[1979] means for having a generative model that learns based on the information;

[1980] means for accepting user questions;

[1981] means for analyzing said query;

[1982] a means of searching for such information;

[1983] a means for generating an answer based on the retrieved information;

[1984] means for displaying the generated answer to a user;

[1985] A voice input means installed in the smart device;

[1986] a means for converting speech to text;

[1987] means for transmitting the textual question to a server;

[1988] means for displaying the generated answer on a smart device;

[1989] A system including:

[1990] (Claim 2)

[1991] 2. The system of claim 1, wherein the means for analyzing the question performs the analysis using natural language processing.

[1992] (Claim 3)

[1993] 2. The system of claim 1, wherein information about corporate support programs is provided through an online portal.

[1994] "Example 2: Combining Emotion Engines"

[1995] ...

Claims

1. a means of collecting information on corporate support systems; means for formatting and preprocessing said information; means for having a generative model that learns based on the information; means for accepting user questions; means for analyzing said query; a means of searching for such information; a means for generating an answer based on the retrieved information; means for displaying the generated answer to a user; A system including:

2. 2. The system of claim 1, wherein the means for analyzing the question performs the analysis using natural language processing.

3. 2. The system of claim 1, wherein information about corporate support programs is provided through an online portal.

Images